Article

President Clinton's initiative on race has come back, time and time again, to the nagging question of affirmative action, and for good reason. While most fair-minded Americans are supportive of strong antidiscrimination laws -- against bias in employment, housing, and education -- Americans are deeply divided over whether racial preferences are an appropriate means of remedying past discrimination and fostering integration in higher education and the workforce. For most of the quarter century experiment with racial preferences, the labor movement has been as deeply divided over affirmative action as the rest of the country. The time has come for a new compromise on the affirmative action question -- preferences based on class or socioeconomic status rather than race or gender. This class-based alternative is more appealing morally, politically, and legally; and because it is rooted in the bedrock reality of class unfairness, it ought to be especially appealing to the labor movement.

The affirmative action question fascinates, divides, and tears at Americans because neither side of the debate can be easily dismissed. On the one hand, liberals are clearly right to say that the history of discrimination in this country has left a legacy of racial inequality that needs to be addressed proactively -- beyond simply outlawing future discrimination. If anyone visits the affluent suburbs in this country, and comes across African Americans or Latinos, the chances are quite good that they are there not as residents but as garbage collectors, gardeners, or child care providers. Such is not a colorblind society.

On the other hand, conservatives make a powerful case that we need to be very, very cautious before employing racial preferences when the long term goal is a colorblind society. The casualness with which many advocates of affirmative action now embrace using skin color to decide who should get ahead in society is astonishing. Some advocates say counting race in college admissions is no different than the consideration of other nonacademic factors, such as musical or athletic ability. But these advocates ignore the fundamental reason racial discrimination is morally repugnant in the first place. As a rule, people should be judged by factors within their control. An athlete or violin player can develop her skills through hard work and diligence; skin color is an accident of birth. The color-consciousness of today's left is particularly disturbing as the basis for racial preferences has shifted: where once preferences were justified as a temporary remedy for past discrimination, today preferences are normally supported by the idea that diversity is an overriding and self-justifying goal, and will be forevermore. This vision of unending color-consciousness is highly unappealing to most Americans.

Finally, both liberals and conservatives are noting with increasing frequency the fundamental unfairness of class-blind racial preferences. Since racial affirmative action counts skin color but ignores economic circumstances, it necessarily prefers the son of a black doctor over the son of a white garbage collector. With the growth of the black middle class in recent years, the likelihood that race and class will collide in this manner has increased dramatically. When programs promote such patent unfairness, they lose their moral force. These programs begin to look like any other reflection of interest group politics and political power, and cease to claim the mantle of promoting genuine equal opportunity.

The vote on Proposition 209, the California Civil Rights Initiative to ban race and gender preferences, underlines American ambivalence over affirmative action. Californians did vote to abolish such preferences in the public sphere, underlining concerns about the use of race in deciding who will enjoy the American Dream. But the margin was fairly narrow -- 54-46% -- which suggests that many Californians were unwilling to simply go "cold turkey."

More recently, a New York Times/CBS News poll found that Americans polled nationally favor affirmative efforts to help disadvantaged minorities, but they reject the use of racial preferences to achieve racial diversity. The poll, published in December 1997, found that 25% of Americans would "do away with" affirmative action preferences, 24% would "leave them as they are" and 43% would "change them."

Is it possible to come up with a compromise on this issue, one that respects the best arguments on both sides of the affirmative action debate? Two possible compromises have emerged to "mend" rather than "end" affirmative action.

The first possibility -- the position embraced by President Clinton -- is to say yes to affirmative action preferences, but no to racial quotas. This compromise has been called the "Bakke straddle," named after Justice Lewis Powell's opinion in the famous Bakke case from the 1970s. In that Supreme Court decision, four justices said the University of California at Davis Medical School's system of quotas was illegal; four justices said the program was legal; and Justice Powell, casting the deciding vote, tried to split the difference: quotas are illegal, he said, but using race as "one factor" in admissions is permissible. U.C. Davis could weight the race of minority candidates in their favor, but it could not set aside a fixed number of seats for people of color.

At first, the Bakke straddle was highly criticized by both sides of the affirmative action debate. In the Bakke case itself, the liberal justices were most critical: what is the difference, they asked, between setting aside 16 of 100 seats for minority candidates and using "race as one factor" in admissions with the committee having a pretty good idea of what kind of racial makeup they would like to end up with at the end of the day? Nevertheless, the Bakke straddle has proven a useful rhetorical device, and whenever President Clinton is asked a question about affirmative action, the first thing out of his mouth is that he is opposed to quotas.

But fundamentally the Bakke straddle remains a highly unstable compromise. If the use of skin color is seen as fundamentally unfair, it matters not whether the use of race is a small or large factor. Americans would never tolerate the idea that racial discrimination against people of color is permissible, so long as race is just "one factor" used against a candidate. And in California, and elsewhere, when the issue is posed not as quotas, but as preferences, Americans make clear their opposition to such practices. The courts have become increasingly hostile to the Bakke straddle, and in the 1996 Hopwood v. Texas case, the 5th Circuit ruled that preferences are just as illegal as quotas. The Supreme Court declined to take the Hopwood case, but cases involving the University of Michigan and the University of Washington may eventually end up in the Supreme Court, and there is a good chance that the Bakke straddle will not survive.

The second compromise to the affirmative action issue is to provide preferences on the basis of class or socioeconomic status rather than race or color. The roots of the argument lie not in Justice Powell's decision in Bakke, but rather in the writings of Dr. Martin Luther King, Jr.

In his 1964 book, Why We Can't Wait, King entertained the argument over affirmative action directly. King wrote (as advocates of affirmative action note) that compensation was due: "the nation must not only radically readjust its attitude toward the Negro in the compelling present, but must incorporate in its planning some compensatory consideration for the handicaps he has inherited from the past." At the same time, King had fought too hard and too long for the goal of colorblindness -- treating people based on the content of their character not the color of their skin -- to casually embrace racial preferences as the optimal means for compensation.

Instead of advocating a new Bill of Rights for the Negro exclusively, King advocated a Bill of Rights for the Disadvantaged. He wrote, "While Negroes form the vast majority of

America

's disadvantaged, there are millions of white poor who would also benefit from such a bill." In advocating a colorblind program for the disadvantaged, King had two key insights.

First, King saw that a class-based approach would implicitly compensate the victims of discrimination. Because of past discrimination, blacks were disproportionately poor, and would disproportionately benefit from any class-based program. In this way, the program would implicitly take account of history. But at the same time, King saw that fairness demanded the inclusion of poor whites. Discrimination was not the only cause of unequal opportunity, he knew; deprivation also blocked opportunity.

Inequality of opportunity in this country runs far deeper than race and gender discrimination. Sociological studies over the last several decades have demonstrated the degree to which children born into poor families find the deck stacked against them. A child born into the top 10% by income is 27 times as likely as a child born into the bottom 10% to rise as an adult to earn in the top 10% bracket. Poor children -- regardless of race -- are more likely to grow up in violent neighborhoods, attend inferior schools, and are much less likely to reach their full natural potential. King knew that racial discrimination was but one element of societal unfairness, and he wrote: "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."

Secondly, as a matter of politics, King knew that in the matter of preferential treatment, emphasizing class over race was imperative. Toward the end of his life, King was trying to build a multiracial coalition of disadvantaged people culminating in the Poor People's Campaign of 1968. He was assassinated in Memphis as he was helping sanitation workers in a multiracial effort for economic justice. King knew that racial preferences could split that coalition asunder. He wrote: "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)."

But if King worried about the divisive effect of a policy of racial preference, division was seen as a plus for Richard Nixon, who, in 1969, instituted the first major federal affirmative action program. Nixon offered the Philadelphia Plan, an effort to impose quotas on the construction industry and trade unions in Philadelphia, in part as a way of dividing King's progressive coalition. Nixon aide John Ehrlichman recalled that Nixon delighted in the plan's ability to construct "a political dilemma for the labor union leaders and civil rights groups...Before long, the AFL-CIO and the NAACP were locked in combat over one of the passionate issues of the day."

The great civil rights leader Bayard Rustin, addressing the AFL-CIO in 1969, said the conflict between labor and civil rights groups over the Philadelphia Plan was "a source of tremendous satisfaction to the powerful enemies of the labor movement." The Plan's intent, Rustin said, was to "deliberately throw black and white workers at each other's throats." Years later, Rustin would write: "Any preferential approach postulated on racial, ethnic, religious, or sexual lines will only disrupt a multicultural society and lead to a backlash. However, special treatment can be provided to those who have been exploited or denied opportunities if solutions are predicated on class lines, precisely because all religious, ethnic, and racial groups have a depressed class who would benefit."

Using race and gender-based affirmative action as a model, one can easily envision a system of preferences in education, entry-level employment, and government contracting based on economic status. The basic idea is to define preferences in a way that take into account the economic obstacles that poor and working class youth face. If a youth has faced a number of economic disadvantages -- and done fairly well despite the odds -- the chances are that her long-run potential is greater than a colleague who has a similar record of accomplishment having faced none of the same obstacles.

In the college admissions arena, preferences could be given to students who are disadvantaged by what sociologists say are the three major determinants of life chances: parental income, education, and occupation. That is, students from families with low incomes, whose parents did not attend college, and whose parents work in low status occupations should be given a break.

Today, Bard College provides a leg up to students from low-income families by requiring a notarized copy of IRS form 1040 or letters from a social service agency verifying income. Existing federal guidelines are available to adjust income for family size or even geographic region to reflect the varying value of a given income. Given the importance of a parent's educational level to a student's chances of educational success, Berkeley provides preferences to students whose parents do not have a four year college degree. And numerous academic studies dating back 80 years have devised sophisticated and consistent ways to rank parental occupations.

In addition, a university might consider factoring in such elements as a family's net worth, the quality of secondary education, neighborhood influences, and family structure. Net worth provides a longer term picture of a family's economic position that goes beyond the snapshot provided by annual income. Counting net worth also uncovers one of the significant differences between blacks and whites in

America

: even among whites and African Americans of the same income group, blacks tend to have lower net worths, in part because they are less likely to have inherited money from their parents. The quality of secondary education can be measured by objective standards such as median high school test scores or graduation rates. If a particular student does quite well even though he attends an underachieving high school, that student probably has a great deal to offer in the long term. Neighborhood influences can be measured by such factors as the poverty or unemployment rate in the applicant's census tract. This factor reflects strong sociological evidence that it is a disadvantage to be born into a poor family, but it is doubly disadvantageous to be born into a poor family living in concentrated poverty. This factor will, in the aggregate, tend to benefit people of color, who are generally more likely to live in concentrated poverty than whites. Finally, a school could consider whether the applicant was raised in a single parent family. Numerous studies show that even after controlling for income, growing up without two parents presents an independent obstacle to success.

Many of these factors -- such as income and net worth -- are already known to college admissions officers through financial aid offices. Parental education levels is normally supplied as well. And in recent years schools like the University of California and Los Angeles Law School have succeeded in devising workable class-based affirmative action programs which go beyond the information normally available to financial aid and admissions officers.

Because large numbers of low income youth are not in a position to pursue college, similar preferences could be provided in entry-level employment to students entering the workforce directly from high school. The federal government currently provides such hiring preferences to low income workers who build new public housing. Likewise, federal contractors could be required to provide entry level opportunities to disadvantaged youth as a condition for receiving federal funds, much as such contractors are now required to take account of race in hiring. Class-based preferences in promotions down the line make less sense in part because individuals are more responsible for their economic position as they grow older, and in part because highly skilled jobs are less amenable to flexibility in hiring.

In government contracting, preferences can be provided to small firms (irrespective of the owner's race) which agree to employ individuals from high poverty census tracts. In May 1996, for example, President Clinton issued an executive order directing the Commerce Department to provide preferences to government contractors employing significant numbers of residents from census tracts with poverty rates of 20% or more. In New York City, Mayor Ed Koch instituted a similar program in the 1980s based on economic status rather than race or gender. The program set aside contracts for small firms that did at least 25% of their business in depressed areas or employed economically disadvantaged workers as 25% of their workforce.

All these types of affirmative action will disproportionately benefit people of color, but will do so on the unifying basis of class rather than the dividing basis of race. (For those seeking more detail, I outline how class-based affirmative action might work in chapter 5 of The Remedy: Class, Race, and Affirmative Action.)

It should be emphasized that class-based preferences should supplement, rather than supplant, strict enforcement of civil rights laws. Such laws include not only those aimed at eliminating intentional discrimination (the Civil Rights Act of 1964) but also those aimed at hiring practices which result in a statistical racial imbalance and cannot be justified on nonracial grounds (the Civil Rights Act of 1991). These are powerful tools and should be backed up with increased funding for government enforcement.

What is the likelihood that a class-based affirmative action will be put into place? The chances increase every day. The attacks on affirmative action have led both liberals and conservatives to begin talking seriously about class-based alternatives.

Racial preferences are under heavy attack from the most democratic institution (the ballot initiative) and the least democratic (the judiciary). Passage of Proposition 209 in California has spawned similar efforts to cut back on affirmative action in 1998 in Ohio, Colorado, Florida, and Washington. More severe threats come from the courts. Proponents of affirmative action dodged a bullet when civil rights groups settled the case of Piscataway v. Taxman, but new cases are making their way up to the High Court, and new restrictions are likely.

The legal and political attacks have raised the salience of the class-based alternative as the left begins contingency planning, and the right searches for something positive to say. Among conservatives, Jack Kemp argues affirmative action "should be done on the basis of income and socioeconomic status such as an enterprise zone instead of just giving a tax break to someone based on color." And Newt Gingrich argues racial preferences should be replaced with special help for people "who come out of poor neighborhoods, who come out of poor backgrounds, who go to school in poor counties." These statements, underreported by the mainstream press, constitute a not insignificant development, considering the inattention conservatives have historically given to issues of class inequality.

Among Democrats, President Clinton has flirted on an off with class-based affirmative action. In March 1995, Clinton made headlines by declaring, that instead of emphasizing racial preferences "I want us to emphasize need-based programs where we can because they work better and have a bigger impact and generate broader support." Clinton quickly backed off that statement after Jesse Jackson threatened to run for president, but in late 1997, Clinton returned to the concept of need-based preferences. In the Akron, Ohio town hall meeting, Clinton said "the real issue" is whether a "class-based affirmative opportunity agenda" should supplement or replace race-based affirmative action. "I think that's the question," he stressed; "that's the nub of the affirmative action debate." The president continued: "politically and substantively you'll help more people and build more unity by having an economic basis for social policy now."

Public support for class-based affirmative action continues to grow. In the December 1997 New York Times/CBS poll, Americans supported preferences for people from poor families by 53%-37%. When asked, "Is it a good idea to select a person from a poor family over one from a middle-class or rich family if they are equally qualified," Americans favored the idea by 2:1 (56%-27%).

What is particularly interesting is the attention younger people have given to the idea of class-based affirmative action. The younger generation is generally more racially tolerant, and for that reason sees less need for racial preferences in coming years. During the debate over affirmative action in California, the Berkeley student newspaper came out in favor of paying "special attention to economic status" rather than to race or gender. When students at Berkeley find common cause with Newt Gingrich, it is time to take note.

In particular, it is time for labor to take note. In the debate over race vs. class-based affirmative action, what Clinton calls the "nub" of the affirmative action debate, labor has been astonishingly quiet. There can be little question where the interests and values of labor historically lay: in multiracial approaches which benefit working class people generally. Where racial preferences have spawned division, class-based affirmative action would unite various races, and remind the ranks of labor who their true friends are.

Big business has turned out to be a great champion of affirmative action, which in itself should give labor reason for pause. Under the current legal regime, business can use preferences to achieve racial balance as a way of fending off suits from people of color, secure in the knowledge that "reverse discrimination" suits from whites will rarely prevail. If affirmative action policies are altered, business will be equally subject to lawsuits for unfair treatment from all sides. Because business does not want that kind of accountability, it fought an anti-preference initiative in Houston, Texas in 1997, and it put the bulk of the money up to prevent the Supreme Court from hearing the arguments in the Piscataway affirmative action layoff case.

With the notable exception of the late Albert Shanker, labor has done very little to support the class-based alternative to racial preferences. In another era, labor would surely have been at the forefront: policies which polarize workers by race, labor knew, always give management an advantage.

Class-based affirmative action tries to address the historic wrongs highlighted by proponents of racial preferences but uses means not offensive to the majority of Americans who oppose racial preferences. More importantly, class-based preferences are firmly rooted in an American sense of fairness -- giving a break to those who need it -- in a way that unites rather than divides. This is a natural for the American labor movement. It will be a lasting shame if the Republicans manage to figure all of this out first.