Higher Ed on Autopilot

A new book projects a bleak future for US higher education.

Higher Ed on Autopilot

Demographics and the Demand for Higher Education by Nathan D. Grawe. Johns Hopkins University Press. 192 pages.

WHAT IF UNIVERSITY EXECUTIVES manage to keep the status quo? Does higher ed stay on an even keel — steadily less loved and less funded, but still generating good volumes of decent credentials for people who will then mostly earn enough money to pay off their loans? The stoic demographer Nathan D. Grawe has written a book with a surprising answer. He takes no policy position, and I can picture him cringing at my impatient shorthanding of the current consensus. But his analysis rings an alarm that will sound very loud even to those with their fingers in their ears.

The consensus among today’s senior academic managers is that the most progressive ideas for universities must be defeated. Large-scale student debt relief? This college president opposes it. Free college at four-year institutions? This president opposes it. Divest the endowment from fossil fuel companies? This president opposes it. Eliminate special admissions for wealthy students? This president opposes it. On the other hand, hide a donation from an influential convicted pedophile? This president allows it. Reject a public university partnership with a for-profit diploma mill? This president created it. But make a trillion-dollar reinvestment in public universities? Where’s the president who has come out in favor of it?

In taking these stands, the presidents are setting themselves against public opinion. When a recent poll asked people if they support proposals to make “two and four year public and tribal colleges and universities tuition-free and debt-free and erase the roughly $1.6 trillion in student loan debt currently owed by close to 45 million Americans,” the answer was yes, 58 to 42 percent. This majority is important because the price tag of $2.2 trillion over 10 years was included in the question, and the policy won anyway. Seventy-two percent of Democrats favored the idea (which is good news for the presidential campaigns of its proposers, Bernie Sanders and Elizabeth Warren); so did 58 percent of independents. In another poll, four-fifths of the public want federal and state governments to spend more money on higher education. College presidents are in danger of repeating their predecessors’ opposition to the 1944 GI Bill that built the modern public university system, such as University of Chicago’s president Robert Maynard Hutchins warning about “educational hoboes” who would burden colleges by living and studying at government expense.

Still, major reinvestment won’t happen if it faces opposition from higher-ed leadership en masse. Given the plans of Sanders, Warren, and others, the country might naturally ask the college presidents: if you say we can’t have zero tuition, or at least much lower tuition, or can’t end student debt, or can’t dial way back on private fundraising, or can’t greatly increase public funding, what can we do?

Higher ed’s Plan A is “you can have what you have now.” Meanwhile, its leaders will wait for free college and debt relief to die a political death while supporting more accountability metrics, simpler financial aid forms, higher income-eligibility thresholds, and calls for the next Secretary of Education to go back to prosecuting predatory for-profits. We are to have a set of dull yet virtuous tweaks, which add up, sort of. We would naturally assume that things at least won’t get worse — that graduation rates, debt levels, and racial imbalances would stay pretty much as they are now, which is mediocre by international standards, very unequal by race and income, yet mediocre and unequal in a settled, familiar way.

Enter Nathan Grawe, who has done the work to show us how wrong we are. He has built a Higher Education Demand Index (HEDI) to model the next 10 years, based on existing patterns. He looks at trends by region, by race and ethnicity, and by type of university, among other factors, so that he can avoid the banality of averages that wash out specific, intensive effects. He assumes that current patterns will stay unchanged, which is reasonable given the entrenched policy complacency within the sector itself. This is what he projects:

The US population will continue to shift from snowbelt to sunbelt — from Northeast and Midwest to West, Southwest, and South. The populations in the faster-growing areas will also tend to have a higher birth rate than the national average. The shift won’t in itself be a big deal for overall college completion — the students who don’t go to college in Massachusetts would in theory be graduating in Arizona, so there would be no overall loss.

But there are complications. “The nation’s total fertility rate has plummeted by more than 12 percent since 2007. And so, beginning in 2026 the number of native-born children reaching college age will begin a rapid decline.” Birth rates hit bottom in 2013, but there has been no meaningful recovery.

In addition, the highest population growth “is concentrated in populations that aren’t likely to attend a regional four-year school.” This is in part because the chances of a high school student finishing college are shockingly dependent on having parents that went to college, and regions with high rates of immigration or poverty are likely to have high shares of what would be first-generation college students — except they are less likely to attend. Polls at least in California suggest that Latinx parents have very high expectations for their children’s college attendance (closer to Asian Americans than to the lower white levels), but translating these desires into completion is easier said than done.

A third issue is that colleges aren’t going to get another major boost to enrollment from women entering the workforce or from “knowledge economy” demand for college graduates (driven by “skill-based technical change”). The former boost was a one-time event, though it will continue on the margins; the latter seems to have stalled and may be going into reverse, given the offshoring of so much medium- and even high-skill work — not to mention expanding automation.

Based on the current autopilot, Grawe’s model yields the following. The two-year colleges that both parties present as the hope of the working classes will experience “dramatically reduced enrollments” — down 15 percent over the decade in some places. In Grawe’s graveside deadpan, “the HEDI anticipates that enrollments among-non-Hispanic whites and non-Hispanic blacks will plummet by 20 and 30 percent, respectively, while the number of Hispanic students is expected to grow by 10 percent.”

Regional four-year colleges should benefit from the decline of two-year enrollments, shouldn’t they? Actually, no. After some mid-2020s growth, birth-death contraction will set in, and four-year numbers will also drop everywhere except in the Pacific region, where over 10 years they will grow by less than one percent.

Grawe’s model forecasts widespread under-enrollment that will produce fiscal crises at many colleges in many regions at once. We should expect systemic stress and turmoil, which would include cascading college closures of the kind threatened at Hampshire College and already occurring elsewhere. Closures will be worse in the Snow Belt/Rust Belt but not confined there. Many endangered colleges will struggle to stay open by cutting payrolls, which is likely to mean staff and faculty layoffs. Colleges may try to avoid layoffs with further adjunctification of the teaching force — or may combine layoffs with adjunctification, sometimes with the same professors (“We have to deny you tenure, but we’ll be happy rehire you on a three-year contract”). Grawe projects a loss of 8,000 faculty positions at two-year colleges and a further 9,000 positions lost at regional four-year schools.

It’s hard to imagine this fading higher-ed sector having the power to uplift or transform society. It will keep cutting educational spending to keep some money in the bank. Enrollment and revenue losses will particularly affect the poorer colleges that provide the most social mobility. State government’s core interest, job readiness, will also be hit: “[U]nless something changes, we should not expect to meet goals for a more highly trained workforce.” This will tempt state government to cut again.

What about racial justice, or at least increased equity? Most people look to education to shrink racial gaps of wealth and income while offering basic social integration and exchange, but higher ed’s record is mixed at best. We might assume that “rising minority populations in the country” will produce proportional increases in college attendance. But in Grawe’s model, this doesn’t happen. Although the Latinx share of the population will grow by five percent, their college share will remain flat overall, with the exception of a three-percent increase at colleges ranked in the top 50 (generally the most selective). The status quo will thus yield no meaningful gain in what is probably higher ed’s most important social contribution: cross-racial equality.

How about overall completion? The most straightforward non-racial social benefits are getting low-income and first-generation students to finish a bachelor’s degree. Students from the bottom quartile by income graduate at about one-seventh the rate of those from the top quartile; a half-century after landmark civil rights and anti-poverty programs, this is a national disgrace. (Unfortunately, Grawe uses attendance rather than completion rates and unusual income categories, which makes his analysis rosier than reality is likely to be.) On policy autopilot, “in all but the elite institutions, the share of first-generation students (defined as students not living with a parent who holds a bachelor’s degree) will fall by more than five percentage points” — more than the decline in 18-year-olds overall. The attendance gap won’t shrink between the top 13 percent by income and everyone else but in fact will grow a little.

Autopilot plus the smaller youth cohort isn’t a falling tide that lowers all boats. Boats will float and sink according to familiar patterns, only more so. Students with college-educated parents and high-income students will know where to apply and what to do to get in, as they do now. But they will be chased by elite colleges with relatively more available slots than they have now. Fewer of these students who can pay full tuition will go to regional public colleges, worsening both their finances and their academic levels. Educational gaps among graduates will increase: students with college-educated parents go to regional colleges at twice the rate of first-generation students, and at six times the rate at “national and elite” schools. In short, Grawe writes, “Overall attendance rates will lag behind projections for our economy’s needs, and large gaps across groups will persist or even grow.” All we need to do to get this less educated and more unequal world is to stay the course.

Just when the reader thinks we’ve doomed ourselves to decline, Grawe reveals how we could cut these gaps. He experiments with his model in two ways: first, he changes it so that racial difference has, by fiat, half the effect it currently does — every group increases participation to close half its gap with Asian Americans; second, he closes half the gap in the effects of income on college attendance. These are big changes, much bigger than anything being suggested by higher education officials.

If all racial groups close half their attendance lag with Asian Americans, “the fraction of students attending a four-year institution would rise to just more than 50 percent — a 10-point increase from the current rate of 41 percent.” Cutting the effects of income in half would mean that “the gap between high- and low-income students in the year 2028 falls by approximately one-quarter, from 38 percentage points to 29 points.” Something similar happens when the model tries to shrink racial gaps by halving differences in two factors that affect racial outcomes — family incomes and parental education.

These are good gains, and yet big gaps remain. There’s a kind of boomerang effect: a huge effort gets the United States of 2028 back to where it was in 2018. It’s worth reading this slowly:

Under the status quo, the four-year college-going gap between Asian Americans and Hispanics is expected to grow from 29 percentage points to 37 points in the next 15 years. If Hispanics, who earn lower incomes on average, attended college at a higher rate due to a reduction in income effects, the resulting gap would be reduced by one-fifth, to 29 points, eliminating (though not reversing) this negative trend. Differences between Asians and non-Hispanic blacks would also be reduced from 35 percentage points under the status quo to 32 points in the counterfactual, a 10 percent reduction.

These racial gaps are very big and slated to get bigger without Grawe’s experimental interventions; even with them, the reductions are small. Why doesn’t major gap-closing in the causes of racial disparities do more to shrink racial disparities? Because the gaps are big, the drivers strong, the system wrenched, the structure twisted, the damage deep.

What about parental education? Here we go again:

Under the status quo, these gaps are expected to hold steady, with children from families with no parental bachelor’s almost 50 percentage points less likely to attend college throughout the next 15 years. A policy that halved income effects would reduce this gap to 42 points.

The size of these gaps boggles the mind. Continuing the iron grip of the parents’ educational level over that of their children, decade after decade, is a kind of social madness. Because the gap-widening momentum is so strong, the counter-move corrects it very little.

Toward the end, the hope of the policy experiments dies away. Grawe concludes like this:

Results of the thought experiments underscore the limitations of policy. For example, despite the large interventions imagined, neither hypothetical policy reduced the expected growth in race/ethnicity gaps. […] Aggressively targeting income can mitigate the problem — but three-quarters of today’s gap would remain even after an aggressive policy intervention. […] [D]emographic changes already in motion foreshadow race/ethnicity, family income, and parental education attendance gaps so large that even aggressive policy interventions are unlikely to eliminate them in the mid-term future.

It’s not really “demographic changes” driving all this, but rather social history and public policy — that is, structural racism and the market pricing of educational goods. It’s the policy autopilot that lets your parents’ education largely decide yours — not demographics. The same is true of Latinx college attendance: lower attendance rates directly contradict stated parental expectations in Latinx communities but follow from social conditions that politics refuses to fix.

Grawe very nicely shows the depth of the hole we’ve dug. Again, even when he cuts inequalities in half, we’re still stuck in our hole. It is dug by the backhoe of the current business model, with its high tuition and forced-debt aid, its rich-college/poor-college forking paths, its pushing of risk onto students, and its defensive contentment with the system.

But there’s a missing final chapter in the book. It would model what happens when income gaps go to zero. This would start with free college. It would also involve, per Sanders and Warren, debt relief. It would require the use of student grant funding to cover room and board and other expenses rather than covering tuition. The revenues public colleges now get from tuition would come from the state and federal governments, since their citizens receive direct benefits from a much larger population holding high-quality degrees.

How much would this close attendance gaps — and completion gaps, which are more important? I hope Grawe at some point uses his Higher Education Demand Index to find out. The result will not be the end of disparity. Racial gaps will survive the removal of income-based differences. But it will reduce disparity to an extent that higher ed will appear to serve the entire society again, by increasing equity rather than undermining it.

Policy can’t control parental educational background. Policy can reduce racial inequality, but it is wired into our culture, geography, and infrastructure and will change more slowly. On the other hand, the effects of income gaps could be changed in one year following the 2020 election. Free college, coupled with debt forgiveness and attendance grants, would come closer to delivering educational justice than any policy in US history. The power of Grawe’s book is that he shows the hard landing of the current autopilot, and the futility of half measures.


Christopher Newfield is a professor of English at the University of California, Santa Barbara. He is the author of The Great Mistake: How We Wrecked Public Universities and How We Can Fix Them (2016), among other books.

LARB Contributor

Christopher Newfield is distinguished professor emeritus of English at the University of California, Santa Barbara and director of research at the Independent Social Research Foundation, London. He is the author of The Great Mistake: How We Wrecked Public Universities and How We Can Fix Them (2016), among other books.


LARB Staff Recommendations

Did you know LARB is a reader-supported nonprofit?

LARB publishes daily without a paywall as part of our mission to make rigorous, incisive, and engaging writing on every aspect of literature, culture, and the arts freely accessible to the public. Help us continue this work with your tax-deductible donation today!