Will We Still Need Affirmative Action in 2025?

The last major Supreme Court decision dealing with affirmative action, 2003's Grutter v. Bollinger, was notable for two reasons. First, and most obviously, it set new ground rules for colleges that wanted to consider race in its admissions process. Second, it contained an incredibly bold, and seemingly tossed off, prediction. Writing for the majority, Justice Sandra Day O'Connor declared:

It has been 25 years since Justice Powell first approved the use of race to further an interest in student body diversity in the context of public higher education. Since that time, the number of minority applicants with high grades and test scores has indeed increased. We expect that 25 years from now, the use of racial preferences will no longer be necessary to further the interest approved today.

Yep. Educational inequality was apparently on a 25-year egg timer.

Was there any chance she was right? Are we really approaching a point where campuses won't need to change their admissions standards for the sake of diversity? Given that today's court may well be getting ready to put new limits on affirmative action policies, the answer to that question seems freshly relevant.

Thankfully, economists Alan Krueger (now of the Obama administration), Jesse Rothstein, and Sarah Turner decided to tackle the issue in 2005.

Their prognosis? It didn't look good.

The trio started out by looking at the distribution of SAT scores for African American and white students in the year 2000, shown below. In general, African Americans did about 200 points worse on the test than their white counterparts. For them to continue getting into college at the same rates without affirmative action, they'd have to narrow that big gulf within a generation.

The authors considered a few ways that might happen. African Americans might close the income gap somewhat with white families, which would presumably help bring up their SAT scores. African American teens might start doing better in high school -- as measured by their test scores on the NAEP. Or, perhaps most implausibly, U.S. public schools might completely integrate, which could boost minority test scores.

Using some very generous assumptions about how all these trends might play out over the next decades, Krueger, Rothstein, and Turner modeled what would happen to African American SAT scores, shown in the graph below. The red line are scores in 2000. Everything else are the predictions.

There isn't a lot of good news in this graph. If income were to rise and nothing else happened, test scores would barely budge. As a result, they projected that black enrollment at selective colleges would fall by 58 percent. If incomes went up and schools integrated more thoroughly (again, not a likely scenario) it would do a bit more to close the gap. Income growth and improved NAEP scores, on the other hand, would do quite a lot, erasing much of the differences on the SAT, and making up for most of the losses due to the end of affirmative action.

Here's why those findings are more depressing today than in 2005. All of the paper's conclusions are essentially a fancy way of saying that if African American students suddenly got richer and started doing better in class compared to white students, they'd be more likely to get into selective colleges without the assistance of affirmative action. In the time since, the income gap hasn't improved. Meanwhile, African American high school test scores are still stagnating compared to whites, just as they have been since the early 1990s. O'Connor's prediction was already tenuous at best. None of the progress we'd need to see for it to come true has been happening.