A Fragile Network

I have been a tremendous fan of Robert Brown’s thinking as it pertains to the structure and funding of American research universities, given the critical role they play in supporting the competitiveness of the United States in a wide variety of fields and industries. His thoughtful approach to analyzing the various elements that comprise “success” in a modern research university in “Don’t Rank Research Universities—Compare Them” (Issues, Winter 2025) gives us something to dwell on beyond the staid rankings from US News and World Report and Times Higher Education. While those rankings may provide some indication of the pool that comprises the nation’s elite research universities, it does little to unpack the components that make public and private institutions excel and how interdependent so many of these elements are to each other. For example, endowments play a major role in supporting merit and need-based aid, which in turn drives graduate and undergraduate enrollment growth, at a relatively low total cost and hence with lower levels of student debt.

As a “student” of the US higher education system throughout my career with firms that have significant exposure to or focus on this sector, I have been able to witness firsthand how research universities have developed strategies for funding and expanding sponsored research. Contrary to public perception, it is a myth that these endeavors are sources of significant margin for most if not all institutions. Brown’s analysis (as he points out in the conclusion) shines a bright light on the fragility of this network of public and private universities that are the lifeblood of innovation in this country.

Recent events are already putting significant additional stress on this system, such as the proposed elimination of negotiated indirect cost rates in favor of a flat 15% for research grants from the National Institutes of Health, as well as discussed caps on graduate student loans and the potential for sharp increases in the taxes on certain university endowments. The data Brown presents should provide industry leaders, higher education boards of trustees, and political and policy officials with a clear-eyed perspective on just what underpins these institutions from a funding perspective, and how destructive some of the proposed changes would be to this relatively modest number of national research universities. As Brown points out, the higher education sector is facing an enrollment cliff, with near-term declines in the number of high school graduates in nearly every geographic region except the South. If much of the proposed policy, regulatory, and legislative agenda is enacted, nearly every revenue stream that Brown examines (e.g., endowments, graduate and undergraduate enrollment, and research indirect and direct costs) will be under some form of stress, resulting in a deleterious impact on the United States’ position possessing the higher education system that is the envy of the world.

Cole Clark

Robert A. Brown provides an interesting statistical evaluation of the factors characterizing American research universities. In his analysis, two principal components are highly correlated with institutional traits, derived from consideration of multiple variables, including six-year graduation rates and research expenditures. The findings reveal two distinct clusters: public universities with large undergraduate enrollments, lower tuition fees, and high proportions of Pell Grant recipients; and private universities with elevated peer assessment scores, low acceptance rates, and high graduation rates.

The results are not surprising, but this analysis offers an essential quantitative framework to advance discussions about the sustainability of the higher education system. The ecosystem of research universities integrates education, research, and service—joint missions that are underappreciated by the general public. For over a century, the synergy of research and education has defined the American research university—and the system remains the envy of the world. Yet the question, “Which one is best?” is not answerable.

While this framework enables universities to situate themselves within their sector, it addresses only part of the issue. All universities serve various stakeholders, including undergraduate and graduate students, parents, federal funding agencies, donors, patients, and corporations.

Stakeholder engagement is influenced by perceived returns on investment, which vary by individual tastes and resources. For instance, prospective undergraduate students may weigh cost, geographic location, and institutional prestige when deciding, often favoring elite private universities if financial constraints are not a concern. Such decisionmaking underscores the optimization or satisficing behaviors of stakeholders. Ranking is done within some notion of sector.

Crucially, Brown’s analysis provides only limited insight into the role of research in determining university clusters or rankings. Research-related factors, such as major scientific awards, collaborations with national laboratories, technology transfers, industry partnerships, and broader scientific impact, are either conflated with or overshadowed by educational metrics such as graduation and acceptance rates. These research dimensions are integral to differentiating the missions of institutions along different vectors of excellence—compare, for example, the University of Michigan to Williams College. A deeper exploration of research-specific characteristics and their implications for the conceptualization of the US research university ecosystem is warranted.

The need for our institutions to focus on being a better match to the needs of our society becomes more acute each day, even before the recent actions of the US administration. Every industry in the world is dramatically different than it was in 1960, with one exception—higher education.

Change driven by data and analysis would be preferable to change imposed by others.

Eric W. Kaler

J. Michael Oakes

Robert Brown’s essay reads like the scientific article it is and reflects the kind of thorough analysis expected from such a scholar-researcher. The fact that he was also a president of a large private research university adds both to his understanding of the need for such a study and his desire to offer more than a criticism but an alternative to the current (flawed) rankings system.

Using Brown’s “sports team” analogy, I will say, for the most part, he hit some home runs. But he also had a few “swing and a miss” moments.

His home runs include highlighting all that is wrong with these rankings from an analytical (statistical modeling) standpoint. Another is his alternative analytic framework, which uses a technique known as principal component analysis to capture 18 features that may influence the rankings of (specifically) research universities, both (and separately) public and private. He also wisely suggests two categories of private institutions (larger and smaller). And his recommendation that US graduate diplomas awarded to top international students “should have a green card stapled to them” clears the outfield wall.

His whiffs came in the section on endowments, specifically the differences (Brown believes to exist) between public and private institutions. First strike: Private research universities are no longer all that different from large public research universities in terms of awarding aid. And public universities’ efforts to keep in-state costs relatively low often are offset by their objective of attracting more out-of-state students who pay higher tuition. The argument that public universities’ missions of access and workforce development (something all universities should have) “by holding down in-state tuition while increasing student numbers to boost revenue” is spurious. Brown’s assertion that institutions without large endowments are forced to raise tuition and fees, or enroll more students, also falls flat.

Instead, I observe, most have turned to philanthropy to continue to keep costs affordable for students and their families and to invest in capital projects that relieve the institution from having to bear those costs entirely on their own. Yes, some universities have used enrollment growth to generate resources, but this is not the norm. Demographics, market realities, competition, and several other factors will require most institutions to look for other revenue streams. The elite privates, while perhaps best positioned to capture increased enrollment market share, often are land-locked and competing for space with their research, clinical, innovation, and entrepreneurship missions.

Most of us hope rankings are dead or at least gone the way of the batter’s box rule. What Brown has given us is better than a replacement ranking system. His analytic framework allows us to examine and better understand meaningful differences between universities, differences that are statistically significant and relevant.

Brown closes with two very important questions, more important today than even a few months ago: Can the United States accept the consequences if the number of research universities dramatically shrinks because fewer can afford to keep pace? And can the United States remain a global leader in innovation if its research universities don’t thrive?

We can replace the last part of each question with “if the federal government dramatically decreases its commitment to and investment in scientific research.” Brown offers this closing swing: “At such a juncture, it is time to take a careful look at the health of the enterprise as a whole to envision its future.” I couldn’t agree more. Indeed, that future is at risk today. Batter up.

David V. Rosowsky

Robert Brown offers important insight into the crazy world of university rankings. Here, a few key findings warrant comment.

Brown suggests correctly that there are too many cases of the same type of university structures—and correlatively, too few departures into lesser known or original organizational territory. We have the traditional private elite colleges and universities; we have those built on the Clark Kerr design of the California system; we have the great public universities such as Michigan, Wisconsin, and North Carolina. But during the period of ascendency of the US research university, especially after World War II, there have been few design changes that complement or perhaps replace the older models. We should have deeper discussions of the redesign of Arizona State University under Michael Crow’s leadership if we are to identify possible new forms that may become the prevailing model as knowledge grows and we become dependent on more multidisciplinary research. Brown’s data set does not allow him to address that question, but it is a significant one. The growth of knowledge tends to shape university structures when it is not fettered by tradition or by impermeable boundaries between university functions and fiscal constraints.

Brown finds and presents in an interesting way the strong association between endowment levels and the prestige of universities. This is not new. The first measurement of university “quality,” done around 1910, found the same thing. But interestingly, Johns Hopkins, perhaps America’s first true research university, began with virtually no endowment. When that first analysis was done, it omitted Hopkins—and those who worked on the study simply decided to throw Hopkins back into the top tier. If Brown had run a simple regression analysis that used “reputational standing” as a dependent variable and regressed it on endowment, he might have found that endowment levels explain most of the variance of reputations. Add to endowment the amount of federal research dollars, controlled for size and whether the university has a medical school, and you will have most of the predicators of so-called quality. A network analysis of universities and colleges also might produce some interesting comparisons—clusters of places that students might want to investigate as they search for the right place for them.

The United States would do well to adopt quickly Brown’s brilliant suggestion to automatically offer a green card to newly minted PhD graduates from abroad. So much of our excellence has depended on the talent of youngsters from other nations, including those escaping totalitarian rule for a better home. If we are not careful, we in fact may become a net loser of talent if President Trump’s proposed policies affecting universities are adopted during the current apparent flirtation with authoritarianism.

One more point. Even people who have spent their entire adult lives at research universities or colleges don’t really have a full knowledge of the teaching and research that goes on at other schools. Yet when called upon to assist in the rankings process, one task is to “evaluate” the reputations of scores of institutions. With limited knowledge, they often end up giving scores based on past reputation, thus making the measure that has traditionally been given the greatest weight in the rankings essentially tautological.

If we are to continue to have rankings, perhaps they ought to be rethought—from the beginning—with a different set of objectives and drawing on a different set of data. I think this is what Brown is proposing in his call for comparing universities rather than ranking them.

Jonathan R. Cole