The Supreme Court ended race-conscious admissions in 2023. Whatever you think of that decision, it is the law. The question now is what comes next, and I want to argue that socioeconomic-based preferences — giving a meaningful boost to first-generation college students, students from low-income households, students from under-resourced schools — are both the right policy on the merits and the most effective path to the diversity goals that race-conscious admissions was trying to achieve.
The empirical case: Raj Chetty's work at Opportunity Insights, published 2023, found that at elite universities, students from families in the top 1% of income were more than twice as likely to be admitted as students with identical test scores from middle-income families. Legacy preferences, donor preferences, and recruited athlete slots — which disproportionately favor wealthy applicants — account for a substantial fraction of that gap.
A socioeconomic-based admissions preference would directly address that gap. And — this is the empirical finding that proponents of race-conscious admissions need to reckon with — because race and economic disadvantage are substantially correlated in the US, a robust socioeconomic preference produces significant racial diversity. The Century Foundation has modeled this for multiple institutions. The diversity outcomes are not identical to race-conscious admissions, but they are substantial.
I am not arguing that race does not matter. I am arguing that in a post-SFFA legal landscape, class-based preferences are the most effective available mechanism for achieving what diversity advocates actually care about — meaningful representation of students from historically excluded communities.
The Chetty data is real and the wealth bias in admissions is a scandal that deserves its own debate. I agree with everything my opponent said about legacy preferences and donor slots and athlete recruitment being corrupt distortions of meritocracy. But those are separate from the race question and I want to keep them separate.
Here is why race still matters in ways that socioeconomic preferences do not capture.
First: the lived experience of being Black in America is not reducible to income. A Black student from a middle-class family in suburban Atlanta has navigated racial bias in ways that a white student from the same income bracket has not. They have different experiences of policing, of teacher expectation bias, of the soft discrimination of social networks. Income does not measure those experiences.
Second: the diversity interest that elite universities argued for in Grutter v. Bollinger (2003) was specifically about the educational benefit of racial diversity — the exposure to different perspectives, the correction of in-group assumptions, the preparation for a diverse workforce and society. Socioeconomic diversity does not produce the same educational benefit. A diverse-income classroom that is racially homogeneous is educationally different from a racially diverse classroom.
Third: the Century Foundation modeling my opponent cited has been contested. The most rigorous analyses, including work by Sigal Alon and Marta Tienda, find that race-neutral alternatives produce significantly less racial diversity than race-conscious admissions at selective institutions. The "socioeconomic as a proxy for race" argument empirically underperforms the direct approach.
The Alon and Tienda research is important and I should have flagged the contested nature of the modeling in my opening. You are right that the evidence on how well socioeconomic preferences replicate racial diversity outcomes is mixed. The Century Foundation modeling is more optimistic than some other analyses.
But I want to push back on the experiential argument. You are saying a middle-class Black student's experience of racial bias does not show up in income data and should be independently credited in admissions. That is true as far as it goes. But the counterargument is that the specific experiences you are describing — bias in policing, teacher expectations, social networks — are most severe and most consequential for Black students in low-income, under-resourced communities. The middle-class Black student from suburban Atlanta, while not immune to racial bias, has access to social capital, legal resources, and institutional navigation capacity that mitigates many of those effects.
I am not saying racial bias against middle-class Black students is not real. I am saying the most compelling justice argument for preferential admissions is for students from communities where those effects compound most severely — and that is substantially captured by socioeconomic measures.
Your framing — that the justice argument is strongest for students where effects compound most severely, and that is substantially captured by socioeconomic measures — would be more convincing if it were true that socioeconomic measures actually captured it well. But we know from research on health outcomes, criminal justice outcomes, and educational outcomes that race has independent effects on life outcomes controlling for income. Raj Chetty's own work — which you cited — finds that Black men with middle-class parents have lower earnings than white men with the same parents at the same income level. Race predicts outcomes independently of class. That is the empirical fact.
The deeper problem with "class as proxy for race": it asks Black students to qualify for consideration through a different and additionally burdensome pathway — prove you are also poor — in a way that white students, Asian students, and other groups do not have to. That is not neutral. That is a system that says your race is not a relevant characteristic but your income is, which maps poorly onto the actual structure of disadvantage in American society.
The Chetty finding about Black men's earnings controlling for parental income is real and important and I am not going to wave it away. Race has independent effects on outcomes. That is empirically established.
But here is the practical constraint I keep coming back to: race-conscious admissions is no longer legal at the vast majority of institutions following SFFA. The question is not "race-conscious admissions vs. class-based preferences" as an academic exercise. It is "class-based preferences vs. nothing" or "class-based preferences vs. outright race-neutral meritocracy."
In that choice — which is the actual political choice available — a robust socioeconomic preference is substantially better for Black and Latino students than pure meritocracy as currently constituted. Even if it is imperfect as a proxy. Even if it misses some of what race-conscious admissions captured. The alternative is not a better system. The alternative is Chetty's data showing elite university admissions tilted toward families in the top 1%.
The practical constraint argument is the strongest version of your case and I want to acknowledge it directly. You are right that the legal landscape has changed and that arguing for race-conscious admissions in the current environment is arguing for something that cannot be implemented.
Where I want to push back: the framing "class-based preferences or nothing" assumes that the legal battle is over. It is not. The SFFA decision left open some avenues — universities can consider how an applicant's race affected their personal experiences and personal development, even if they cannot use race as a direct criterion. That is a narrow opening but it is an opening. Schools like MIT and Yale are trying to work through it.
And I want to say something about the political economy of conceding the race question entirely: if the progressive side of this debate fully embraces class-based admissions as the substitute, it becomes much harder to make the case for race-specific remedies in other contexts — employment, housing, lending. "We have moved past race to class" becomes the argument against the next policy that addresses racial disparities directly. The concession has costs beyond higher education.
The political economy argument my opponent raised at the end is the one I find most troubling and I want to be honest about that. If embracing class-based admissions makes it harder to argue for race-specific remedies in other contexts, that is a real cost that goes beyond higher education.
But I keep coming back to this: the students who most need access to elite higher education — the students whose outcomes would be most changed by admission to a selective institution — are disproportionately students in low-income, resource-deprived communities where race and class compound. A robust socioeconomic preference serves those students, under current law, in a way that race-conscious admissions no longer can.
I am not asking us to stop talking about race. I am asking us to use the tools that are available and effective within the legal constraints that now exist. Class-based preferences are the most powerful available tool. Using them is not conceding that race does not matter. It is choosing to fight on terrain where we can win.
Race still matters. Not as a slogan, but as a structural fact about American life that income does not fully capture and that socioeconomic metrics do not fully measure. The history of deliberate exclusion — redlining, restrictive covenants, discriminatory lending, school funding by property tax — produced wealth gaps that will persist for generations and that require race-conscious remedies to address directly.
The practical constraint my opponent raises is real. Race-conscious admissions is legally unavailable at most institutions right now. Class-based preferences are better than nothing. I am not arguing they should be abolished.
What I am arguing is that "better than nothing" is not the same as "adequate." And that the concession from "race matters specifically" to "class matters and race is just correlated" is a concession that has costs beyond the admissions office. We should be clear-eyed about those costs even as we pursue what is practically available.