For the Wealthiest Colleges, How Many Low-Income Students Are Enough?

Ted Soqui, Corbis via Getty Images

Are the wealthiest colleges educating enough low-income students? It’s a question that’s been debated for years. Back in 2008, Sen. Charles E. Grassley, an Iowa Republican, argued that given their affluence and the tax breaks they enjoy, the richest colleges should do more. Concerns about the rich colleges pulling their weight died down in the wake of the recession, but have recently been bubbling up again.

A couple of years ago, The New York Times created an index to measure “the most economically diverse top colleges,” based in part on the percentage of their freshmen who are low-income. A recent podcast from the best-selling author Malcolm Gladwell suggested — controversially — that Bowdoin College could enroll more low-income students if it scaled back the caliber of food in its dining hall. A new paper from the Education Trust describes the progress wealthy colleges could make in enrolling and supporting low-income students if they increased their endowment spending even slightly.

Policy makers are interested, too. A bipartisan bill that two senators plan to introduce this fall would take money from wealthy colleges that don’t serve enough low-income students and redirect it to less-wealthy colleges that provide more access. And the richest colleges received new congressional scrutiny of their endowments earlier this year.

So what does the low-income student population look like at wealthy colleges these days? The Chronicle examined the enrollment of federal Pell Grant recipients — a measure of low-income status; most recipients have family incomes below $40,000 — at the 25 colleges with the largest endowments per full-time-equivalent undergraduate in 2014-15. Almost three-quarters of those colleges have seen at least some growth in the share of students receiving Pell Grants over the last five years. Still, most of the 25 colleges have less than 20 percent of their students receiving Pell Grants, and a couple have less than half of that percentage. Does that seem like the right level of representation?

Colleges in this group tend to offer generous financial aid to the low-income students they do enroll, they just don’t enroll all that many. Some of them argue that it’s better to take a smaller group of low-income students, but meet all of their financial need — even if that means factoring students’ ability to pay into the admissions process for some or all groups of students.

This is where the conversation usually turns to the big limitations: money and the selective admissions process. Yes, educating more low-income students is expensive. To do it, highly selective colleges don’t just have to spend more on financial aid, but also must forgo the tuition a more-affluent student would have paid to be there. And whatever their intellect and potential, it’s hard for low-income students to compete against applicants whose families have used their significant resources to give them every advantage for the last 18 years.

Still, we’re talking about the very richest colleges, with endowments of at least three-quarters of a million dollars per undergraduate. Surely financial constraints are not the sole driver behind all of their decisions. And whatever its virtues, the selective admissions process was not brought down from the mountain by Moses. So let’s set business as usual aside for a moment and at least investigate the question: What share of an elite college’s students should be poor?

Aiming Low

To the extent that the colleges ask it themselves, their answers seem to reference two main yardsticks: What the budget will allow, and what their peers are doing.

Peer comparisons, though, can let colleges off the hook. A college will win praise for doing a bit better than similar institutions. Most criticism is saved for those who fall at the very bottom of the pack. (When it comes to percent of students on Pell, that has long been Washington University in St. Louis, which is working to double its Pell percentage from 6 percent to 13 percent.)

When colleges really care about something, they want to be the best at it, says Dorothy A. Brown, a professor of law at Emory University who studies race and class. Just look at how they strive to top the rankings. But when it comes to diversity — racial or socioeconomic — that’s not the case, she says. “As long as they’re in the range of their peers,” Ms. Brown says, “then everything is good.” This reveals a lack of commitment. “Somehow mediocrity was fine,” she says, “when we’re talking about diversity.”

Despite the interest in tracking how the wealthy colleges are doing on this measure, individual colleges only rarely set public goals for what their Pell percentages should be. Such goals can be tricky, says Catharine Bond Hill, president of Vassar College. The number of any particular sort of student, she points out, depends not just on how many the college admits but how many accept those admissions offers.

Vassar has not set an explicit Pell goal, but has gotten a lot of attention for doing better than its peers on the measure (Vassar is not one of the 25 richest colleges in this analysis; its endowment per undergraduate is $397,161). Still, a few colleges have set public goals. One is Washington University, with its 13 percent.

Another is Pomona College, which committed to raising its Pell percentage from 17 percent to 20 percent or higher at a 2014 White House summit. That’s a goal Pomona has been able to meet, says Seth Allen, the college’s vice president and dean of admissions and financial aid.

The college’s president led the charge to enroll more Pell recipients, Mr. Allen says, and set the specific goal with several considerations in mind. For one, Pomona felt obligated to do more given its vast resources, he says. Among its peers, 20 percent Pell would put it near the top of the group. And in a way, Pomona didn’t have far to go. It has a program to bring in undocumented students, who are often needy but can’t get federal financial aid. In other words, the college enrolls some low-income students who aren’t reflected in the Pell numbers (international students with financial need are similarly excluded).

And then there’s Willamette University, in Oregon, whose endowment per undergraduate of $115,098 puts it well below our rich colleges group. In 2015, the university’s board set a group of goals around affordability and access. One of them was for Willamette to “match or lead” two peer groups, one regional and one aspirational, in the share of U.S. students who receive Pell Grants and the share who are first generation. Other goals relate to enrolling students of color, having comparable graduation rates for different groups of students, and keeping student debt low.

At the time the goals were set, Willamette already was ahead of both peer groups on Pell, at 21 percent of all undergraduates. But the idea is to meet all the goals at once, and taken together they are “quite ambitious,” says Sandy Rowe, the trustee who chairs the subcommittee that developed them. It’s not enough for Willamette to bring in more low-income students, she says; it also must get them to graduation, with a reasonable debt load.

As her committee worked to pin down the meanings of “access” and “affordability,” it learned that higher ed often uses such terms in a vague and haphazard way, Ms. Rowe says. That meant Willamette had to come up with its own definitions. “We have not accepted a mush, national, unclear definition of access and affordability,” she says.

In an Ideal World …

So it’s rare for colleges to set goals around their shares of students on Pell Grants, and even where they exist they seem to be designed with peer colleges in mind. How else might a college determine whether it’s pulling its weight when it comes to educating low-income students?

This is not a new question. A 2006 paper co-written by Ms. Hill set out to answer it. In an earlier study, the authors had found that 10 percent of students at colleges in the Consortium on Financing Higher Education, a group of highly-selective private colleges, came from the bottom 40 percent of the income spectrum. To check if that figure seemed reasonable, the second paper looked at national data on test scores and family income. The pool of potentially admissible students depends on the test-score threshold, but the researchers concluded that even if the colleges set it at a combined 1420 on the SAT — a pretty high bar — the group of colleges could increase their percentage of Pell students from 10 to 13 percent.

More recent research from Caroline Hoxby, a professor of economics at Stanford University, and her co-authors indicated the existence of a significant untapped pool of high-achieving, low-income students who did not apply to selective colleges, and developed a low-cost way to change their application behavior.

Ms. Hill still finds the method outlined in her paper to be a logical way of determining how many low-income students selective colleges could take. If anything, there are more Pell-eligible students with high test scores than the paper suggests, she says, because the paper used income quintiles, and considered those in the bottom two to be low-income, while Pell eligibility extends into the third.

Research has shown that there are more low-income students out there who meet selective admissions criteria, but those criteria are also quite narrow, says Awilda Rodriguez, an assistant professor in the Center for the Study of Higher and Postsecondary Education at the University of Michigan at Ann Arbor. And the bar for getting in can rise over time — like some other graduates of the most selective colleges, Ms. Rodriguez wonders if she would still get into her alma mater, Princeton University, if she applied today.

At least one other attempt has been made to pinpoint what share of students at elite colleges should be low-income. A 2014 paper from the Education Trust suggested that at least 17 percent of colleges’ full-time freshmen ought to be Pell eligible, a threshold it arrived at by cutting off the bottom 5 percent of four-year colleges on this measure. That benchmark, of course, is based on what other colleges are doing, but it considers all four-year colleges and suggests that any one could do better than the lowest performers are now.

There are still other ways to think through the question. “I’d say in an ideal world, Pell enrollment would be representative to the Pell universe,” says Kim Cook, executive director of the National College Access Network. In other words, the percentage of undergraduates who get Pell nationally — 35 percent — should be reflected in the percentage at each college.

Ms. Rodriguez says her immediate reaction is that “maybe it should reflect the general population of Pell recipients going to four-year colleges.” That figure is actually almost the same, 34 percent.

Suggesting that elite colleges could take as many low-income students as four-year colleges do over all will probably raise some eyebrows. After all, that would be a monumental change in their enrollment. But there’s an even loftier way to look at this.

As a group, the families of college students are better off than the overall population — rich people are overrepresented in college, poor people underrepresented. But what if the elite colleges sought to match the income distribution of the country?, asks John Gudvangen, assistant vice chancellor of enrollment and director of financial aid at the University of Denver. Colleges use the national or regional population as a gauge when it comes to the racial composition of their student bodies, he says. Why not do the same for income? (He knows this sounds idealistic.)

Of course, there’s no particular reason that the percentage of students receiving Pell must be the same at each of the rich colleges. If anything, Ms. Hill says, the very richest should do a little more — they could afford to take more poor students without having to stop doing anything else.

And the elite colleges could make more progress if they worked together. “Clearly this would be much easier for all of us to do,” Ms. Hill says, “if all of us were doing it.”

Beckie Supiano writes about college affordability, the job market for new graduates, and professional schools, among other things. Follow her on Twitter @becksup, or drop her a line at beckie.supiano@chronicle.com.

Dan Bauman is a reporter who investigates and writes about all things data in higher education. Tweet him at @danbauman77 or email him at dan.bauman@chronicle.com.