The Equality of Opportunity Project gathers amazing data on the incomes of college graduates and their families. The New York Times provides an overview and this summary (pdf) of Williams data. Day 1 of 5 on this topic.

The entire discussion around socio-economic diversity at elite colleges is about to change, all because of this new data set, produced by Stanford Professor Raj Chetty and colleagues. But, if you have been reading EphBlog for the last 10+ years, little of this will be news to you. From The New York Times:

Students at elite colleges are even richer than experts realized, according to a new study based on millions of anonymous tax filings and tuition records.

At 38 colleges in America, including five in the Ivy League – Dartmouth, Princeton, Yale, Penn and Brown – more students came from the top 1 percent of the income scale than from the entire bottom 60 percent.

The Times should consult better experts. We have always pointed out that Williams, like all elite schools, is a bastion of privilege, that the student body is, and always has been, dominated by the very wealthy. Recall this discussion from two years ago. I get into trouble when I argue that this is largely inevitable — very smart people are both likely to be rich and blessed with smart children, because of both nature and nuture — and not necessarily a problem. See this ten (!) part series from three years ago for background.

The key data can be summarized in one table:


If you find this surprising, then you haven’t been paying attention. Or you have naively believed some of the drivel from Williams! Recall the news release about early decision for the class of 2021 from December:

[N]early 20 percent of Early Decision admits come from low-income families.

Before reading further, ask yourself, “What is a reasonable definition of the term ‘low-income’ when used in a press release?”

If you are an idiot — or merely one of the “experts” that The Times likes to interview — you probably take this at face value. Why would it be surprising that 20% of Williams students are from low income families? (Yes, I realize that this is just the early decision pool and that the Chetty data does not cover the class of 2021, but those factors don’t matter.) The answer, of course, is that Williams is being about as truthful as Trump’s press secretary when he estimates inauguration attendance. Mary Detloff kindly provided this clarification: at Williams, a “low-income” family is one with less than $85,000 in annual income.

I bet that not a single one of our readers picked a number that high as a fair definition of “low-income.” A much more reasonable definition of low income would be the bottom 20% of the distribution.


By that measure — which is probably what the vast majority of (naively trusting!) applicants and alumni had in mind when they read the College’s news release — only 5.3% of Williams students are low income, not “nearly 20%.”

I have always known (and shown!) that Williams is a place of privilege, a bastion of the economic elite. And that is OK! The elite have to send their children to college somewhere. My great annoyance has always been the College’s tendency to obfuscate this central reality, to pretend otherwise, to twist the meaning of phrases like “low income” in order to mislead. The EOP data makes those sorts of lies much less tenable. Hooray!

Any commentary on the specific values in that table? Richer colleges like Williams/Amherst/Swarthmore/Pomona have higher percentages from below the 60th percentile, not because the people who run those colleges are any more committed to socio-economic diversity than the people who run other schools, but because their endowments are so large that they can afford the extra-spending on financial aid. You really think that Will Dudley ’89 (new president of Washington and Lee) loves non-rich people less than Adam Falk? Hah! But some of the differences might have interesting explanations. Why would, for example, Swarthmore have less than half as many students from the 1% as Williams?

Print  •  Email