When Justice Sandra Day O'Connor's Supreme Court decision three years ago upheld the use of race in admissions, most college officials were jubilant. Absent the use of some preferences, colleges said, competitive institutions would experience a sharp drop in black and Latino enrollments.
Amid the campus celebrations, relatively little attention was paid to a part of the O'Connor ruling that stated that the use of preferences must eventually stop -- she suggested 25 years (from 2003) as an outer limit. "Enshrining a permanent justification for racial preferences would offend this fundamental equal protection principle. We see no reason to exempt race-conscious admissions programs from the requirement that all governmental use of race must have a logical end point," she wrote.
To the extent college officials talked about that part of her decision, many expressed hope that there was enough time so that rising income levels for black and Latino families (which correlate with availability of good schools, and with standardized test scores) would bring students to parity in that time. A study released Friday -- as part of a new book by the College Board -- suggests that view may be a pipe dream.
Two economists at Princeton University and one at the University of Virginia collaborated on an analysis that examined the rates at which black incomes are growing and are projected to grow. Then they looked at the patterns in which increases in black income levels translate into test score increases of the sort needed for admission to top colleges. The scholars' data show that even 25 years after the O'Connor ruling, and assuming significant gains in black income, the abandonment of affirmative action in admissions would lead to a huge drop in black enrollments at top colleges. The gains in black incomes and academic performance just aren't fast enough to meet O'Connor's deadline.
For their analysis, the researchers looked at four groups of competitive colleges (by degree of competitiveness) and examined the black share of the black and white enrollment for each group today, and what it would be with race-neutral admissions policies. (For a variety of reasons, the research looked only at black and white students). For each group, the share of black students would drop substantially under a race-neutral system. A particular challenge, the researchers note, is the large gap in black and white scores, on average, on the SAT. Under the most recent data released by the College Board, the white average was higher than the black average by more than 100 points on mathematics and just under 100 on the other two sections.
Then the researchers looked at models based on data on the rates at which the black-white income gap is narrowing, and did calculations based on the well established relationship between family income and SAT scores. While the researchers acknowledged that no one can be certain of the way income gaps will change in the years ahead, they used figures suggesting that in each generation, the gap in black and white average incomes will be halved. Then the researchers examined data showing a gradual closing of the gap on between black and white students on certain educational tests, and assumed that gap would continue to close.
Even with all of those assumptions, however, the models developed suggest that under race-neutral admissions policies that could be required, competitive colleges of all types would see significant decreases in black enrollments. For their research, the researchers ran projections of a "composite" based on data from various colleges.
Black Share of Black and White Enrollment at Various Types of Colleges
|Most Selective Colleges||Highly Selective Colleges||Moderately Selective Colleges||Public Universities|
|Colleges used in composite||Harvard, Princeton, Yale||Penn, Swarthmore, Williams||Barnard, Bowdoin, Middlebury, Oberlin, Pomona, Wellesley||University of Virginia, Penn State University|
|Actual 2000 enrollment||17%||14%||9%||12%|
|2000 enrollment -- with race-blind admissions||5%||6%||5%||5%|
|2025 -- with race-blind admissions and narrowing of black-white income gap||7%||8%||6%||6%|
|2025 -- with race-blind admissions and narrowing of black-white income gap and testing gap||12%||12%||9%||9%|
What do these figures suggest? Sarah Turner, associate professor of education and economics at Virginia and one of the authors, said that she and her colleagues didn't want to imply huge precision in their predictions. She called the project "an exercise." But she said it was important for educators who care about having a diverse student body to realize that gradual shifts in black incomes are unlikely to produce the kind of impact that would be needed to do away with affirmative action and preserve the black enrollment levels that colleges want to maintain.
"Income changes alone aren't going to close the gap, even with generous assumptions," she said.
Turner's co-authors are Alan B. Krueger, professor of economics and public policy at Princeton, and Jesse Rothstein, assistant professor of economics and public affairs at Princeton. The version of their work that was released Friday appeared in a new College Board collection of essays, College Access: Opportunity or Privilege?
A longer version of the study is forthcoming in American Law and Economics Review.