This post originally appeared on SCOTUSblog, as part of a symposium on Fisher v. University of Texas at Austin, the challenge to the university’s use of affirmative action in its undergraduate admissions process. 

The Supreme Court’s affirmative action decisions have been suffused with hypocrisy. Justice Ruth Bader Ginsburg called them out, with barely more gentle phrasing, in her lone dissent to the seven-to-one majority opinion the first time Fisher v. University of Texas at Austin (2013) was before the Court. “Only an ostrich,” she observed, “could regard the supposedly neutral alternatives as race unconscious,” and only a (contorted) legal mind “could conclude that an admissions plan designed to produce racial diversity is not race conscious.”

The “diversity” standard in college admissions has gained great popularity because advocates of race-based affirmative action, stymied by the Court since Regents of the University of California v. Bakke, latched onto it as an alternative that could satisfy strict scrutiny. Many proponents have since persuaded themselves that diversity is, after all, a better approach than race-based affirmative action and that if the Court had not required it, we would have had to invent it. Yet while diversity in college classes is certainly an important educational and social goal, its elevation nonetheless dodges the nation’s racial legacy and avoids our constitutional and moral obligation to remedy the effects of centuries of slavery and legally sanctioned segregation. Without acknowledging we were doing so, we have engaged in a legal sleight of hand, substituting enriching the educational experience for remedying past injustice in designing affirmative action policy.

Underlying all this has been the Court majority’s conviction, most recently in Fisher I, that university officials have not identified specific Fourteenth Amendment violations for which their policies are a remedy, and therefore their consideration of race injects, without constitutional justification, a discriminatory racial consideration into the admissions process. The paucity of African Americans at the University of Texas reflects no de jure exclusion, the Fisher I majority believed, but only de factosocial inequality for which there is no race-conscious constitutional remedy. Therefore, including racial diversity in a scheme of skill-based, interest-based, or economic diversity is suspect, requiring very strict scrutiny. Indeed, the conditions set by the Fisher I majority opinion suggest a scrutiny that is strict in theory but fatal in fact. (I discuss the Fisher cases here only as they relate to the treatment of African Americans in affirmative action plans, not to that of other national or ethnic minorities or of disadvantaged economic groups; each has a different history and status, requires different opportunities to succeed, and raises different social policy and constitutional concerns).

Unlike several of his Court colleagues, Justice Anthony Kennedy, who wrote the majority opinion inFisher I, regarded the purportedly de facto racial imbalance in the Texas student body to be a serious social (though not unconstitutional) problem, and so he ordered Texas officials to develop an ostrich-friendly plan that might, by the way, increase the number of African-American students without specifically attempting to do so. Such an assignment is nearly impossible, which is why it would not be surprising if, in Fisher II, the Court concludes that Texas has failed to employ sufficiently race-neutral techniques to increase racial diversity on campus.

The university does employ one superficially race-neutral plan that succeeds in enrolling some African Americans, allocating three-fourths of its first-year undergraduate places to students who are ranked in the top of their high school graduating classes. Although frequently called the “Ten Percent Plan,” about seven percent of high school graduates are nowadays needed to fill three-fourths of entering places and are admitted automatically based on high school rank. The plan exploits the fact that Texas high schools are highly segregated—purportedly de facto, of course, because Texas no longer has laws on the books specifically requiring students to attend racially homogenous high schools. Rather, the neighborhoods in which high schools are located are racially homogenous, not because of an existing racial zoning law but mostly because of a century of nationwide federal, state, and local policies that were explicitly designed to segregate the races.

Under this plan, if you attend an all-black high school in Texas and do well, you are guaranteed admission to the university, increasing its black enrollment. But African-American under-representation nonetheless persists. The University of Texas reports (Table 4.1) that the “Ten Percent Plan” results in black enrollment of only three percent of all first-year students, and (Table 4.2) disproportionately enrolls African Americans from the poorest families, because these are most likely to live in the most segregated neighborhoods and attend the most segregated high schools. Middle-class African Americans, who are generally better prepared for college, remain even more underrepresented as a result of the Ten Percent Plan—although this is somewhat attenuated by the reality that middle-class African Americans, in Texas and elsewhere, frequently live in segregated neighborhoods that are poorer than those in which similarly middle-class whites reside.

The university admits the other one-fourth of its class by examining each application individually, with diversity in mind. It considers not only test scores and grade-point average, but also whether a student excels in the debate club, or in the high school orchestra, or on a lacrosse team, or whether the student has had to overcome difficult challenges, and so on. If these devices still fail to yield a racially diversified entering class, admissions officers can put a light thumb on the scale for applicants who are African American. As a result of this process, Table 4.1 of the Texas report shows that the 2014 entering undergraduate class was four percent black—three percent from the Ten Percent Plan and one percent from holistic review—while African Americans represent about fifteen percent of all college-age youth in Texas. Table 4.2 shows that students admitted in this holistic review process are more likely to be middle-class than those admitted as “Ten Percenters.” Fisher IIdoes not challenge the Ten Percent Plan but only the slight thumb on the scale for African Americans in the holistic review process, a thumb that yields, at most, one percent of the entering class that is black—at most, because some African Americans would succeed in holistic review regardless of their race because they are excellent debaters, violinists, or lacrosse stars.

Supporters of race-neutral alternatives to achieve racial diversity claim that the thumb on the scale for African Americans is unnecessary. Sufficient race-neutral racial diversity, they say, can be achieved if only the university tries harder. Because African-American families have less wealth than white families, admissions officers should favor students of all racial and ethnic backgrounds from low-wealth families, and this will incidentally result in more admitted black students. Because college-going rates for African Americans have historically been lower than white rates, admissions officers should favor students of all racial and ethnic backgrounds who are first in their families to attend college, and this too will incidentally increase African-American enrollment. Because African Americans suffer more discrimination than whites in American society, admissions officers should favor students of all racial and ethnic backgrounds who have had to overcome “great personal challenges,” and this too will incidentally increase African-American enrollment. Such efforts to proxy race without mentioning race should pass the scrutiny of ostriches. And although no race-neutral advocates have actually proposed this, I have a suggestion that can simplify the process considerably: Texas should give preference to applicants, regardless of race or ethnicity, who are descended from American slaves. This will be more effective than any other method in increasing African-American enrollment, and is fully race-neutral.

Such shenanigans (although less extreme than my suggestion) could result from Fisher II’s likely repudiation of Texas’s present policy of holistic applicant review. But a surprise may be in store.

Opponents of considering race in the Texas holistic review plan probably need the vote of Justice Kennedy for victory. Getting it has been a reasonable expectation, because Justice Kennedy has consistently voted to reject the claims of African Americans in civil rights cases, including those involving voting rights, affirmative action, or employment discrimination—until this past June. Then, endorsing the arguments of civil rights advocates, he wrote a majority opinion holding that the Fair Housing Act’s prohibition on racial discrimination includes a prohibition on race-neutral actions that needlessly exacerbate racial segregation, even if a racially discriminatory motive cannot be proven. In support of this conclusion, Justice Kennedy said that the “vestiges of de jure residential segregation by race remain today, intertwined with the country’s economic and social life.” In the twentieth century, he added, “various practices were followed, sometimes with governmental support, to encourage and maintain the separation of the races.”

That’s as far as he went. He did not say that the vestiges of de jure segregation, intertwined with the country’s economic and social life, justify race-conscious remedies, only that the Fair Housing Act can prohibit racially disparate effects. He underplayed the systematic governmental policies that enforced the “various practices” of racial segregation. But he did say that the Fair Housing Act “must play an important part in avoiding [a] grim prophecy that ‘[o]ur Nation is moving toward two societies, one black, one white—separate and unequal’” and he acknowledged “the Fair Housing Act’s continuing role in moving the Nation toward a more integrated society.” This was not race-neutral language.

Justice Kennedy cited an amicus brief of “housing scholars” (I was one) that documented many of these systematic twentieth-century governmental policies that violated the constitutional prohibition of de jure segregation. Consider just one, and its ongoing impact on the challenges facing university admissions:

From the 1930s through the 1950s, the Federal Housing Administration guaranteed production loans to developers for construction of suburban subdivisions, on explicit condition that these projects be barred to otherwise-qualified black families. Nationwide, federal mortgage guarantees subsidized working- and lower-middle-class white families (many headed by returned war veterans) to flee urban neighborhoods for suburban communities whose racial exclusivity was government-enforced. Black veterans and other working families were openly denied such support, not only by prejudiced real estate agents, but by racially explicit public policy.

In the late 1940s and early 1950s, these suburban homes (for example, in places like Levittown outside New York City, Lakewood outside Los Angeles, and hundreds of other communities in between) typically sold for about $125,000 (in today’s dollars) and were easily affordable to working class families. With GI-Bill mortgages (no down payment required), suburban home monthly carrying charges were often less than rent that white families had previously paid in overcrowded urban apartments, the kinds of places where black families had to remain.

Today, these same Levittown or Lakewood homes sell for $500,000 or more. The white families who settled in these places gained, over the next few generations, some $400,000 in equity appreciation. Black families, denied this opportunity by de jure segregation policy, realized no such gains. African-American median family income today is about sixty percent of white median family income, but African-American median household wealth is only five percent of white median household wealth (including home equity). Families that gain home equity appreciation over several generations have many opportunities to convert that equity into more liquid wealth that can be used, for example, to pay for children’s (or grandchildren’s) college, or perhaps to support parents in retirement, relieving children of the financial pressure to do so. For white and black families with similar middle class incomes in 2007, the wealth of black families not attributable to present home equity was only twenty-two percent of that of white families.

Differences in family wealth thus have enduring effects on college attendance. These differences today are, to a great extent, vestiges of explicit twentieth-century federal segregation policy that violated the Fifth and Thirteenth Amendments to the Constitution. College admissions officers, ostrich-like, can favor students of all races and ethnicities whose families have little wealth for whatever reason. Such favoritism, however, is an inefficient means of remedying de jure segregation: it will round up many non-black students from low-wealth families. That is a worthy social policy goal, but is no substitute for remedying a history of state-sponsored racial injustice.

Will Justice Ginsburg find a new ally on affirmative action issues in the author of the Court’s fair housing/disparate impact decision, ensuring that the University of Texas can continue to make very minor efforts to consciously integrate its student body? It is not terribly likely—Justice Kennedy stopped well short of considering the implications of his recommendation to build “a more integrated society.” But if he remembers that a more integrated society is not merely a more diverse one, a surprising result could emerge from Fisher II as well.