The Volokh Conspiracy

Mostly law professors | Sometimes contrarian | Often libertarian | Always independent

Affirmative Action Without Racial Preferences

In my forthcoming book, Classified, I suggest limiting preferences in higher education to descendants of enslaved Americans and residents of Indian reservations.


The Supreme Court, as most readers surely know by know, has decided to hear appeals to two cases challenging racial and ethnic preferences in higher education. Assuming the Court is disinclined to allow the use of overt racial and ethnic preferences, is it possible that some version of affirmative action that takes ancestral "background" into account may be salvaged?

In my forthcoming book, Classified: The Untold Story of Racial Classification in America, I suggest that the answer is yes, at least with regard to most African Americans and some Native Americans. (The book is not about affirmative action, but obviously a book on racial classifications is going to address that issue.)

The book describes how the familiar categories universities use to sort students by race and ethnicity–Asian American, Black/African American, Hawaiian or Pacific Islander, Hispanic/Latino, Native American, and White–came to be. To make a long story short, they were invented by the Office of Management and Budget in the 1970s to regularize statistics-keeping and reporting within the federal government. While "white" and "black" were familiar categories, almost no one considered themselves or anyone else to be Hispanic, "Latino" or "Asian" before 1970 [as opposed to Mexican, Cuban, Chinese, Japanese, etc.] and it was by no means inevitable that white ethnic groups like Cajuns, Italians, Poles, and Jews would be classified as generic whites.

The classifications the government came up with were never intended to be proxies for "diversity" in higher education or elsewhere, and they explicitly came with the caveat that the "classifications should not be interpreted as being scientific or anthropological in nature." OMB warned that the categories also should not be "viewed as determinants of eligibility for participation in any Federal program," such as affirmative action programs.

Nevertheless, because universities had to use these categories in reporting admissions statistics to the Department of Education, they almost immediately became affirmative action proxy categories. In the book, I first address the use of these categories in Minority Business Enterprise programs:

Businesses owned by African American descendants of slaves (ADOS) were the original primary intended beneficiaries of minority business enterprise (MBE) preferences. Nevertheless, members of all minority groups became equally eligible for these preferences….
Most MBE preferences now go to businesses owned by members of official minority groups who are not descendants of enslaved Americans. The ADOS population is dwarfed demographically by the combined population of Hispanics, Asian Americans, Native Americans, and black immigrants from Africa and the Caribbean and their descendants. The non-ADOS groups not only outnumber black Americans but on average have more of the economic, educational, and social capital needed to obtain government contracts.

Under current rules and norms, anyone with partial Asian or Hispanic ancestry going back at least to one's grandparents and perhaps indefinitely can claim membership in those groups. Americans of mixed ancestry are generally willing to shift their self-identified racial or ethnic status to whatever currently benefits them….Within a generation or two, a large majority of Americans will be eligible for MBE preferences. If almost everyone is eligible for affirmative action preferences, they cease being meaningful. Limiting MBE preferences to fewer people may be the only way the preferences can be saved.

All this suggests that to the extent MBE preferences continue, the government should limit them primarily to the original intended beneficiaries, ADOS. Members of recognized Indian tribes who live on and perhaps very close to reservations, a much smaller demographic, should also be included. Such a limitation would have several advantages. First, ADOS and residents of Indian reservations are the two American groups whose ancestors suffered the most by far from state and private violence, oppression, and exclusion, with continuing reverberations today….

Finally, government-granted preferences to people based on their racial or ethnic category raise constitutional, ethical, and practical concerns. But neither descent from American slaves nor membership in an Indian tribe and residence on an Indian reservation is a racial category, as such [see Morton v. Mancari]. Black Americans born in Africa would no longer qualify for MBE preferences, nor would a Los Angeles resident who has one Native American great-grandparent from whom he inherited tribal membership.

I then turn to racial preferences in higher education:

The only purpose for which the Supreme Court permits university-level affirmative action is to enhance the "diversity" of a school's student body for the benefit of all concerned…. Yet the way colleges go about achieving racial and ethnic diversity makes little sense if diversity per se is the objective, as opposed to using diversity as a subterfuge while pursuing other objectives.

First, many elite schools try to match their percentage of minority students from various groups with their respective percentages of the applicant pool or other demographic baseline. Approximately one-half of one percent of the American population identifies as Native American, compared to 18 percent as Hispanic. In an entering class of, say, one thousand, the one hundred and eightieth Hispanic student surely does not make the class more ethnically diverse than would the sixth Native American.

Moreover, universities often give little or no consideration to the fact that members of official minority groups "may have no interest whatsoever in the culture popularly associated with the group…." Meanwhile, the relevant official minority categories are themselves internally ethnically diverse, often radically so…. [Meanwhile, a] Yemeni Muslim student may add significant religious, ethnic, and cultural diversity to a campus. For campus affirmative action purposes, however, admissions offices classify her as just another non-Hispanic white student. The same is true of an Egyptian Copt, a Hungarian Roma, a Bosnian refugee, a Scandinavian Laplander, a Siberian Tatar, a Bobover Hasid, and their descendants."

Those who qualify for the African American category also are not culturally uniform [including everyone from an African immigrant with one white parent to descendants of American slaves]….
The Native American category is also extremely internally diverse [and fraudulent claims of Native American status are common]….

The best way forward for schools truly interested in attracting a diverse group of students would be to cease relying on crude government-imposed racial and ethnic classifications as a proxy for genuine diversity. As in the MBE context, affirmative action preferences, if pursued, should be limited to African American descendants of slaves and members of American Indian tribes who live on reservations. The goal of such preferences would not be diversity, but the righting historical injustices that have modern reverberations, and helping to bring marginalized groups into the American mainstream.

There is a risk, however, the Supreme Court would hold that the ADOS and Indian reservation resident categories are proxies for racial classifications and therefore presumptively unconstitutional.