The Untapped Pool of Low-Income Strivers

The lead article in Sunday’s New York Times featured important research that could help shape the future of equity debates and affirmative action in higher education.

The article, “Better Colleges Failing to Lure Poorer Strivers,” by David Leonhardt, cites a significant study conducted by Stanford University’s Caroline Hoxby and Harvard University’s Christopher Avery, which finds that while more than three-quarters of wealthy high-achieving students attend selective four-year colleges, only about one-third of high-achieving low-income students do so.

Looking at students whose grades and test scores put them in the top 4 percent of the high-school Class of 2008, Hoxby and Avery found that 34 percent of those from the lowest income quartile of households attended one of the nation’s most selective 238 colleges compared with 78 percent of those from the richest quartile. Leonhardt writes, “The findings underscore that elite public and private colleges, despite a stated desire to recruit an economically diverse group of students, have largely failed to do so.”

The research comes at a particularly important time, as Leonhardt notes, because between now and late June, the U.S. Supreme Court, in the case of Fisher v. University of Texas at Austin, may well hand down a decision constraining the ability of universities to use race in student admissions. If the Supreme Court does so, the experience of states that have banned affirmative action by voter referendum suggests that colleges will shift the focus of recruitment and admissions from race to class.

In the heated debates over race-based versus class-based affirmative action, the Hoxby and Avery study undermines three lines of argument advanced by universities.

* Claim: We already aggressively recruit low-income students of all races. Many universities claim that they are not in a position to shift from race-based to class-based recruitment and preferences because they already give considerable weight to socioeconomic disadvantage. But studies by Princeton’s William Bowen and colleagues and by Georgetown University’s Anthony Carnevale and Stephen J. Rose find that universities give significant weight in admissions to underrepresented minority students but virtually no consideration to socioeconomic disadvantage. Now Hoxby and Avery’s research suggests colleges don’t even recruit low-income students in the top 4 percent nationally, two-thirds of whom don’t attend any of the nation’s top 200-plus colleges.

* Claim: We would love to admit more low-income students, but they can’t do the work. Related to the argument that universities already actively recruit and admit low-income students is the contention that universities can’t do more because admitting additional disadvantaged students would set them up for academic failure. Hoxby and Avery do not directly test this hypothesis, but they do find that there is a large untapped pool of extremely high-scoring low-income students who are not being recruited.

This finding is in the same vein as Carnevale and Rose’s earlier conclusion that a class-based affirmative-action program could boost the proportion of students at the nation’s most selective 146 colleges and universities coming from the bottom socioeconomic half of the population from 10 percent to 38 percent, with graduation rates remaining the same as they are today. The Hoxby and Avery findings are also consistent with research from the University of Texas, where a much more economically diverse student body was recruited from the top 10 percent of high-school graduates statewide and performed quite well academically.

* Claim: An admissions system based on class won’t benefit many black and Latino students. Setting aside their professed interested in socioeconomic diversity, university officials further claim that class-based affirmative action is a poor way of recruiting racially and ethnically diverse student bodies given that colleges would have to admit many low-income white and Asian students along the way.

On the surface, the Hoxby and Avery analysis would seem to lend support to this view: 69.4 percent of low-income very high achievers are white and 15.2 percent are Asian. By contrast, 5.7 percent are black and 7.6 percent are Hispanic.

But when one digs deeper, it becomes clear that the racial dividend of class-based affirmative action could be considerable if race were barred from consideration. To begin with, Hoxby and Avery find that black students are almost four times as likely to be found among the high-achieving low-income cohort than among high-achieving students generally (5.7 percent versus 1.5 percent). Likewise, Hispanics are 1.6 times as likely to be found among high-achieving low-income students as the general high-achieving population (7.6 percent versus 4.7 percent).

Moreover, Hoxby and Avery’s study focuses primarily on income as a measure of economic disadvantage. But there is ample evidence that black and Latino students face, on average, additional disadvantages when compared with whites of similar income—disadvantages that can be counted in a class-based affirmative-action program. For example, research finds that black and Hispanic families with incomes in excess of $75,000 live in neighborhoods with higher poverty rates than white families earning less than $40,000. Likewise, median income of black households is 60 percent that of white households, and the median income of Hispanic households is 70 percent that of white households.

However, scholars have found that black median household net worth is just 5 percent of white household median net worth, and Hispanic median net worth is just 6 percent of white median net worth. Incorporating neighborhood poverty and wealth into a calculus of disadvantage would disproportionately benefit black and Latino students.

Finally, Hoxby and Avery limit themselves to the narrowest slice of the student population—the top 4 percent by grades and test scores. Even the very most selective colleges and universities routinely go deeper into the applicant pool than the top 4 percent, and the argument for doing so in the case of economically disadvantaged students is particularly strong given that these individuals have had to overcome odds in order to achieve their academic success. Moving beyond the thin cohort of students with the highest grades and test scores could boost racial diversity further. This may explain why, in a study of 10 leading universities that replaced the use of race in admissions with race-neutral mechanisms such as class-based affirmative action, Halley Potter and I found that in seven cases, institutions were able to meet or exceed the proportion of black and Latino students.

One would hope that the new research by Caroline Hoxby and Christopher Avery would spur top universities and colleges to do more to recruit high-achieving low-income students because it is just and fair to do so. But as Leonhardt suggests, a U.S. Supreme Court decision this spring limiting the ability of universities to employ race may provide the impetus necessary for institutions to finally get serious about class.

Richard Kahlenberg is a senior fellow at the Century Foundation.

Return to Top