Polarization is an ROI Problem

Why liberal arts education is infrastructure—and why we are rating it wrong

A staggering seventy percent of Americans now believe higher education in the U.S. is headed in the wrong direction. Confidence in universities has nearly halved in a decade. Affordability has been the loudest complaint, but it is not the most cited one. In a 2024 Gallup poll, forty-one percent of Americans pointed to ideological concerns, criticizing colleges for being “too liberal,” “indoctrinating” students, and not letting them think for themselves. Another thirty-seven percent pointed to relevance, saying degrees do not teach useful skills. Only twenty-eight percent cited cost. In other words, the perception problem is not primarily a price problem but a purpose one.

The deeper issue is that American universities quietly abandoned the product that once made them indispensable: the capacity to produce citizens who are not only confident in their viewpoints but can also think across boundaries, tolerate disagreement, change their minds when presented with compelling evidence, and coordinate with people who see the world differently. That capacity is not a soft skill. It is economic infrastructure. And we have been steadily degrading it as our socioeconomy increasingly needs it.

Every time a regulatory negotiation collapses into tribal signaling, every time a consumer brand loses a massive chunk of market over a tweet, every time a leadership team fails to have an honest conversation about strategy because the room has silently sorted itself into ideological camps—those are coordination failures. And coordination failures have price tags that will likely be more pronounced in the new age of AI.

Trust is the operating system that makes markets, institutions, and democracies run. When trust erodes, transaction costs rise: contracts get more stringent, regulators get more adversarial, talent pipelines fracture along cultural lines, and deliberation within organizations—the thing that boards, legislatures, and founding teams all depend on—grinds toward paralysis. By deliberation, I mean the capacity of a group of people to weigh competing views, change their minds in response to arguments, and arrive at decisions to which they can collectively commit. That is not the same as everyone agreeing on everything, but it is a basic operating skill of any body—corporate, political, or institutional—that must make hard decisions under uncertainty and with imperfect information.

A liberal arts education is one of the few interventions that builds this capacity at scale. Not because reading Thucydides makes you a nicer person but because disciplines like history, philosophy, political science, and literature train a specific set of cognitive muscles: the ability to hold competing frameworks simultaneously, distinguish a strong argument from a loud one, and engage with ideas you find uncomfortable without treating disagreement as betrayal. These are the skills of deliberation. And deliberation is the technology on which democracies and markets run.

American universities were not designed as vocational schools. Harvard’s motto is Veritas—“Truth.” The University of Chicago’s is Crescat scientia; vita excolatur—“Let knowledge grow from more to more; and so be human life enriched.” Ohio State's is Disciplina in civitatem—“Education for citizenship.” Similar themes embody virtually all American universities, and none of them point toward financial outcomes or ROI. The founding logic was civic: produce people capable of self-governance, and the economy will follow. For most of the twentieth century, that bet paid off spectacularly.

But somewhere along the way, universities and their parent-student constituencies began optimizing for different metrics. Rankings rewarded job placements and salary outcomes over civic outcomes. STEM departments expanded, and the humanities entered a sustained decline: bachelor’s degrees in humanities fields have fallen by almost one-third nationally since 2012. At Harvard, only seven percent of freshmen in 2022 planned to major in the humanities, down from nearly thirty percent in the 1970s.

The irony is sharp: even in terms of the narrow terms on which universities now compete, the STEM-only bet does not work particularly well. Census data shows that only twenty-eight percent of STEM graduates actually work in STEM occupations, with the other seventy-two percent relying on the transferable skills—critical thinking, communication, ethical reasoning, adaptability—a liberal education is designed to build.

That said, one may ask: if seventy-two percent of STEM graduates are already deploying liberal arts skills in their careers, does that mean formal liberal arts education is unnecessary? Are STEM programs quietly teaching these skills anyway? I don’t think so. What the data reveals, in my view, is a massive inefficiency: we are training people in narrow technical domains, and then the majority of them spend their careers drawing on capabilities their education never deliberately cultivated. They pick up those skills despite their training, not because of it, through life experience; professional necessity; and the slow, expensive school of trial and error. A liberal education does not produce different outcomes by accident. It produces them by design. It’s the difference between a trained sommelier and someone who got good at wine by drinking a lot of it. One got there with fewer hangovers.

Meanwhile, the liberal arts programs that survived have developed their own problems. Intellectual pluralism—the whole point of the humanities—has weakened. Campuses that once modeled how to argue well increasingly model how to avoid argument altogether. When the liberal arts stop practicing what they preach, they lose credibility to defend their own existence.

The result is a higher education system that produces neither the technical specialists it promises nor the independent-minded citizens it was built to create. Consider the recent free speech climate on elite campuses. The 2024 Foundation for Individual Rights and Expression (FIRE) College Free Speech Rankings survey of more than 55,000 students found that twenty-seven percent consider it acceptable in some circumstances to use violence to stop a campus speech, up from twenty percent the year before. Forty-five percent said it is acceptable to block other students from attending a speech. A quarter said they self-censor more than when they started college. No wonder the public is walking away.

Why do I, as an international student, care about American higher education? Sure, America has its problems—every country does. But America has also produced a significant chunk of the innovations on which the modern world depends and, perhaps more importantly, an indispensable share of the culture that has inspired modern aspiration. What is American culture? Many things, but one undeniable theme that keeps the world’s wealthiest and poorest alike flocking to these shores is the independence of thought and the daring to dream. Universities have been the cornerstone of this.

More importantly, emerging economies—my home continent of Africa included—under immense pressure to demonstrate immediate economic returns from education, have overwhelmingly embraced a STEM-first model. Humanities and social sciences are being marginalized precisely when the continent needs them most.

Africa has the world’s youngest population—an exciting opportunity but also a worrisome risk if the status quo persists. By 2050, one in four humans will be African. The continent is navigating simultaneous transitions—demographic, democratic, technological, and economic—that demand exactly the kind of civic imagination and institutional creativity that narrow technical training cannot provide.

This is not a niche view. The philosopher Martha Nussbaum, in Not for Profit: Why Democracy Needs the Humanities, has warned that the worldwide retreat from liberal education is producing generations of technically competent graduates who cannot sustain democratic citizenship. Her argument is especially urgent to Africa. The continent’s core development challenge is not a shortage of resources or even ideas but the absence of a coherent worldview—a way of seeing that creates the preconditions for transformation. That worldview does not emerge from engineering curricula. It emerges from rigorous engagement with history, philosophy, culture, and political economy—the disciplines Africa’s universities are currently overlooking. 

Diagnosis is the easy part. The harder question is what reform looks like in practice. Two potential pathways—one institutional and one financial—strike me as the most tractable place to start.

First, the liberal arts need to be reformed before they can be scaled. The critique from the right—which some liberal voices have also raised—is that wokism on campuses, for all it has gotten right, has threatened honest curiosity and discussion. And it did not emerge in a vacuum. It filled the space left behind when the humanities stopped teaching the substantive intellectual traditions—the actual ideas, debates, and frameworks—that give students something to argue about and through. When universities abandoned their role as custodians of a shared intellectual inheritance, identity became the remaining source of meaning and authority. Wokism became the default substitute for the rigorous intellectual tradition that higher education exists to provide. Asking for more of what liberal arts departments currently produce is not the answer. Asking them to rediscover what they once were is.

Professor Noah Feldman, who co-chaired Harvard’s Institutional Voice Working Group in 2024, made an interesting argument: universities should stop taking official positions on political matters outside their core function. Harvard accepted it. Chicago has operated under its 1967 Kalven Report along similar lines for decades. Vanderbilt, Stanford, and others are moving in the same direction. The non-obvious insight here is that institutional neutrality is not the opposite of civic formation—it is a precondition for it. When the institution itself takes political stances, students learn to read the room rather than form independent judgment. They learn the approved answer. Neutrality at the institutional level creates the pressure and space for pluralism at the individual level.

Second, and more ambitiously, we need to change how ROI is measured. Right now, the federal measurement architecture—the College Scorecard, Gainful Employment rules, and the accreditation standards that gate federal aid—conditions recognition and funding on economic outcomes: median earnings, loan repayment, debt burden. The measurement architecture has become the mechanism by which civic purpose gets squeezed out. You cannot exhort an institution back toward civic formation while its survival depends on optimizing a different metric.

The fix is to add civic KPIs to the existing framework: measures of deliberative capacity (can graduates articulate the best version of an opposing view?), civic participation (jury service, public service uptake, voting, local government involvement), pluralism of alumni outcomes (are graduates distributed across the political spectrum or concentrated in narrow bands?), and longitudinal trust and tolerance indicators tracked against cohorts (do graduates trust institutions more or less than when they arrived, and are they more or less willing to engage with people who disagree with them?). Many of these are already measured by the General Social Survey, Pew, and FIRE. The question is whether to make them load-bearing.

This is less an expansion of state power than a correction to a measurement system that is currently lopsided. And unusually for a higher education proposal, it has a coherent cross-ideological case: the left should support it because it would stop punishing humanities departments on narrow earnings grounds. The right should support it because it would create accountability for ideological monocultures. A pilot approach—voluntary reporting at a handful of institutions, refined over time, then scaled through accreditation and funding—is the most defensible path.

The deeper principle is this: universities did not drift toward narrow economic optimization because administrators decided they hated civics. They drifted because rankings, funding flows, parent expectations, and political pressure all pointed toward earnings. Fix the metrics and you create the incentive structure for the reform.

In business school, we are trained to see misallocations of capital and correct them. This is one of them. The world’s education systems are massively underinvesting in the cognitive infrastructure that makes coordination, trust, and adaptive leadership possible—and overinvesting in narrow specialization that depreciates faster than ever.

The question for us is not whether liberal arts education matters. It is whether we want to lead in a world where every disagreement becomes a culture war, or help rebuild the skills of deliberation that make disagreement productive.

The fix is not nostalgia for some golden age of the humanities. It is recognizing that liberal education is infrastructure—as critical to a functioning society as roads, courts, or broadband—and investing accordingly. When we measure education ROI only by starting salary, we miss the returns that actually compound: lower friction in collective decision-making, higher institutional trust, and the next generation of leaders who can make disagreement productive rather than destructive.

Joris Cyizere (MBA ‘26) grew up in Kigali, Rwanda. He holds a BS in Industrial Engineering from Northwestern University. Prior to HBS, Joris worked in technology consulting at PwC and as a Business Analyst at McKinsey in Chicago and Nairobi before joining the World Economic Forum's Centre for the Fourth Industrial Revolution in Kigali, where he advised the Rwandan government on emerging technology strategy and policy.