Expanding access is not a priority for top schools – Study

Elite American colleges and universities are not using increases in their large endowments to facilitate access to these schools by the nation’s poorest or students of colour, says a just-released study, a finding that contradicts what these universities and colleges self-reported on questionnaires completed for the National Association of College and University Business Officers.

In 2018, elite schools self-reported that they allocated 49% of income from the growth of their endowments to financial aid – when the actual figure is closer to 10%, says a new study by University of California, Santa Cruz economics professor George Bulman, conducted for the non-partisan, Cambridge, Massachusetts-based National Bureau of Economic Research.

Colleges and universities “with larger endowments do not increase enrolment overall to expand access to elite education and do not increase the fraction of low-income students or students of colour they serve”.

“Specifically, colleges and universities whose endowments grew relative to their peer institutions due to high investment returns over the last 20 years enrolled fewer students who were eligible for federal Pell Grants [made to the nation’s poorest students] and a smaller percentage of black and Hispanic students,” writes Bulman, the author of The Effect of College and University Endowments on Financial Aid, Admissions, and Student Composition.

Study sample

Bulman studied 140 elite liberal arts colleges and 60 private research universities, 70% and 30%, respectively, of the wealthiest schools in each of these categories as defined by US News and World Report’s rankings.

Together, these institutions educate 1.1 million students, approximately 5.5% of the total number of students in America’s four-year colleges and universities. One quarter of these institutions accepted fewer than 20% of applicants. Yale University in New Haven, in 2022, for example, accepted 4.46% of applicants, its lowest rate ever.

Twenty-five percent of the students in these elite schools received Pell Grants (US$6,500 per year). Hispanics make up some 11% of these schools’ student bodies, while 8% are black. Nationwide, 21.7% of college and university students are Hispanic and 12% are black.

Each of the institutions Bulman studied has an endowment of at least US$20,000 per student. Today, the nation’s largest endowment, of US$53 billion, is held by Harvard University, while a small college such as Haverford College (outside Philadelphia, Pennsylvania) had an endowment of US$641 million.

From 2003-18, the period Bulman studied, the average annual investment returns of endowments for the top 10 liberal arts colleges varied between 5.2% and 10.2%, and between 7.2% and 10.8% for the top 10 research universities. Typically, these schools will spend between 4% and 5% of the growth of their endowment portfolios each year (calculated on either a three- or five-year rolling average).

Spending investment income did not, however, increase access to these campuses.

One way to increase access, Bulman told University World News, would be to use the monies provided by the increased endowment revenue to increase the absolute number of students enrolled in the school.

“We used to have 2,000 students,” he says, “now we are going to have 2,500 students because we can afford to hire more faculty and staff, and build more buildings, and offer more financial aid. We can maintain our quality while expanding.”

Not only does his detailed statistical analysis find no evidence for this, “if anything, there is a small downward pressure on overall enrolment”. While the “figure is a statistically insignificant negative number, it tells us that basically there is no overall growth in enrolment in these elite schools,” Bulman says.

Bulman found that on average these colleges and universities do increase financial aid awards by approximately US$3,000 when their endowments double, a small percentage of the sticker price (tuition, room and board) of US$74,000 at Harvard or US$60,952 at Bowdoin College, in the picturesque town of Brunswick, Maine.

The increase in per-student aid also represents only 10% of the overall increase in institutional spending with respect to endowment growth. The universities and colleges did not, however, reduce their sticker prices. The modest increase in financial aid does not result, Bulman writes, in a statistically significant reduction in estimated net cost per student.

Implications for racial diversity

A quick check of the websites of America’s elite colleges and universities will show numerous statements about their commitment to equity and diversity in their student bodies.

Bulman’s analysis shows, however, that there is an inverse relationship between an increase in funds available for financial aid from growing endowments, and admission to these elite schools by the nation’s poorest students, those who are eligible for Pell Grants.

On average, he found that as endowment income increases and even as individual grants to Pell-eligible students increase, their percent of Pell-eligible students receiving grants fell. On average, a 100% increase in an endowment reduces the fraction of Pell-eligible students attending the university by 5%.

“It’s a small negative number,” he told University World News, “but it’s significant. It shows that these schools with increasing wealth are not increasing access to low-income students.”

The decrease in grants to Pell-eligible students also means that despite what the schools say on their websites, these schools are not increasing access for black and Hispanic students. This can be determined because black and Hispanic students are over-represented in the Pell-eligible group, as compared to whites and Asians.

Accordingly, rather than increasing their enrolment of black and Hispanic students, who are historically under-served by these institutions, their enrolment in elite colleges and universities has declined in response to endowment growth. On average, a 100% increase in endowment reduces the fraction of black and Hispanic students attending the institution by 5%.

“The reduction in enrolment of underrepresented students is statistically significant and meaningful in magnitude relative to the baseline level of diversity at these institutions in 2003. This suggests that institutions do not use endowment wealth to expand access to a more racially diverse student population and is consistent with the reduced fraction of students receiving federal aid [Pell Grants],” he writes.

Rewarded by rankings

While the reduction in the number of black and Hispanic students relative to whites and Asians runs counter to the national push to increase minorities in higher education, somewhat perversely, increased spending and selectivity improves the position of these colleges and universities in league tables like those of US News and World Report.

“The improvement in ranking is partially mechanical,” says Bulman. “The universities are spending more, which feeds into the formula that produces the rankings. Additionally, spending, especially on individual financial aid packages, causes what’s called their ‘yield’ to go up: a ‘yield’ is the percentage of students who enrol at an institution out of the number of accepted applicants.

“This allows institutions to raise their admissions standards, resulting in the admission of fewer low-income students, but the schools have become, in terms of the formula, more selective. This then causes the institutions to go up in the rankings.

“So, it’s cyclical. You spend more and you become more selective, your ranking goes up; you become even more selective, and your ranking goes up further.

“This isn’t visible at the top of the rankings – where Harvard, Princeton or Swarthmore are –because there is no room to go up. But if you are the 60th ranked elite liberal arts college and now, all of a sudden, your investments have done very well and you have the opportunity to become more selective, that is good for them but not, my study shows, for expanding access.”

Political pressure

Bulman’s study is unlikely to relieve the pressure colleges and universities with large endowments are feeling from both the political left and right. Last week, US President Joe Biden announced that the federal government would forgive the first US$10,000 of debt held by people earning less than US$125,000 (US$250,000 for a married couple) per year and US$20,000 for Pell-eligible students who took out student loans (as the vast majority did).

While criticising the president’s plan, a number of Republicans also took aim at schools with large endowments, demanding that they use them to either lower tuition or provide greater financial aid to poor students.

These elite “endowment institutions”, as Bulman calls them, are criticised by the left for not spending enough on low-income students and students of colour.

“And they are under pressure from the right – which generally dislikes these institutions it considers ‘liberal’ – because they see the endowments as something they can potentially go after, as was done in the Tax Cuts and Jobs Act of 2017, which applied a 1.4% tax on endowment returns at institutions where the amount per full-time student is US$500,000 or more.”

A lesson in basic accounting

“Where does your analysis leave the claim by colleges and universities that they are spending 50% of their endowment income on student aid?” I asked toward the end of our interview.

Bulman answered by referencing what’s learned in Accounting 101: money is fungible.

“Self-reported expenditures from endowment revenue are unlikely to account for the fungible nature of revenue sources and do not capture the causal impact of changes in spending. This study documents the effects of endowment spending, where revenue from endowments is fungible with revenue from other sources (such as tuition payments).

“Ultimately, whether endowment spending increases spending for students and to extend elite education to under-served populations is an empirical question, not an accounting question.”

He then added: “This empirical study shows that expanding access, both in terms of numbers and in terms of diversity of students, does not appear to be a high priority for these institutions.”