The Piece of Paper That Costs $200,000 and Proves Nothing
A Nobel Prize-winning theory suggests your degree signals ability rather than building it. What does the evidence reveal about what college actually delivers?
The Signal vs. The Skill
In 2001, Michael Spence won the Nobel Prize for proving that a college degree might be worth nothing — that it's a signal, not a skill. The university sector ignored him. They were the ones charging for the signal. His theory, first published in 1973, proposed a possibility that threatens to unravel the entire justification for higher education: what if colleges don't make people more productive? What if they simply identify who already is?
The implications are staggering. Americans now owe over $1.7 trillion in student debt. The average private university education costs approximately $230,000 for four years. Yet if Spence is right, the economic value of that investment has nothing to do with learning — it's about proving you were capable enough to survive the selection process. The credential is a expensive filter, not a value-add.
Here's where two economic theories collide. Human capital theory, championed by Gary Becker, argues education genuinely increases productivity. Signaling theory argues education merely reveals pre-existing ability. Both predict that college graduates earn more. But only one can be right about why. And the difference determines whether the modern university is an engine of human advancement — or history's most elaborate sorting machine.
The Rival Theories: What Education Actually Does
Human Capital Theory: Education as Transformation
Gary Becker's human capital theory, developed in the 1960s, treats education as an investment in productive capacity. When you learn calculus, critical thinking, or biochemistry, you become genuinely more valuable to employers. The causal mechanism is straightforward: knowledge and skills transferred through education translate directly into workplace productivity.
Under this framework, the college wage premium — the gap between what graduates and non-graduates earn — reflects real value creation. A graduate earns more because they can do more. Education is not consumption; it's capital accumulation. The returns justify the costs because the learning itself generates economic value.
Becker's theory dominated policy thinking for decades and still underpins most defenses of higher education spending. If human capital theory is correct, then expanding college access is an investment in national productivity, and student debt is merely front-loading lifetime earnings.
Signaling Theory: Education as Revelation
Michael Spence's signaling theory, published in his seminal 1973 paper "Job Market Signaling," proposed a radical alternative. In markets with information asymmetry — where employers cannot directly observe worker productivity — workers need credible ways to communicate their ability. Education, Spence argued, serves this function.
The logic is elegant. Completing a degree requires intelligence, conscientiousness, and the ability to defer gratification. These traits correlate strongly with workplace productivity. Crucially, these same traits make degree completion easier for high-ability individuals. The credential becomes a reliable signal because it's costlier for low-ability workers to obtain.
[!INSIGHT] The distinguishing feature of signaling is that the content of education is irrelevant. A degree in 14th-century French poetry signals ability just as effectively as one in computer science — because the signal is in the completion, not the curriculum.
Under pure signaling, the college wage premium reflects selection, not treatment. Graduates earn more because people who graduate were always going to earn more. The education added nothing; it merely revealed.
Where the Theories Diverge: Empirical Battlegrounds
The Dropout Test
If human capital theory is correct, partial education should yield partial returns — two years of college should provide roughly half the benefit of four years, because you've acquired half the skills. If signaling dominates, the returns should cluster around completion. The degree is binary: you have it or you don't.
The evidence is damning. Studies consistently show that college dropouts earn wages nearly identical to high school graduates who never enrolled, while completers earn substantially more. The return is concentrated almost entirely at the finish line — exactly what signaling predicts and human capital struggles to explain.
The Content Irrelevance Problem
“"I have never once used anything I learned in college in my job.”
If education builds human capital, then curriculum content should matter enormously. A computer science graduate should outperform an English literature graduate in programming roles. The evidence suggests far more overlap than human capital theory would predict. Employers routinely hire graduates from unrelated fields, suggesting they value the signal over the specific training.
A 2023 survey found that 52% of recent college graduates work in jobs that don't require their specific degree. If the content mattered, we would expect far more sorting by major into relevant occupations.
The Sheepskin Effect
The "sheepskin effect" — named after the traditional diploma material — refers to the disproportionate wage increase that occurs specifically at degree completion. The final semester of college appears to generate more wage returns than all previous semesters combined. This makes sense if the value lies in the credential, not the accumulated learning.
[!NOTE] Research by Clark and Martorell (2014) found that for students near the degree-completion threshold, the sheepskin effect explains roughly 40% of the college wage premium. The effect persists even controlling for actual skill levels.
Natural Experiments: When Access Expands
Perhaps the strongest tests come from policy changes that expanded college access without changing student ability. If human capital theory holds, more education should mean higher productivity. If signaling dominates, expanded access should dilute the signal without creating proportional value.
Several studies have examined cohorts affected by compulsory schooling reforms and found minimal effects on productivity, despite substantially more education. When everyone has a degree, the signal loses meaning — but productivity doesn't proportionally increase.
The Hybrid Reality: It's Both — But Not Equally
The evidence doesn't support pure signaling or pure human capital. The truth lies somewhere in between — but the proportions matter enormously for policy, pricing, and personal decisions.
For certain fields, human capital clearly dominates. No one wants a surgeon who simply signaled their way through medical school without acquiring genuine expertise. Engineering, nursing, and computer science show stronger content-based returns. The skills are specific, measurable, and directly applied.
For many degrees, however, signaling appears to account for a substantial portion — estimates range from 30% to 50% or higher — of the wage premium. The liberal arts, general business degrees, and many social sciences function primarily as sorting mechanisms.
[!INSIGHT] The uncomfortable truth: elite universities may deliver the least actual skill-building but command the highest prices precisely because their signal is most exclusive. Harvard's value lies not in superior teaching, but in superior selection.
Implications: The $200,000 Question
If signaling explains a significant portion of higher education's returns, several uncomfortable conclusions follow.
First, the system imposes massive collective costs for individual benefits. If education primarily sorts workers rather than improving them, the $1.7 trillion in student debt represents not an investment in human capacity but a tax on proving yourself — a cost that could be eliminated with better information systems.
Second, expanding college access without changing the sorting function may simply credential inflation. As more people obtain degrees, employers demand higher credentials for the same jobs. The signal degrades; everyone must pay more to stand out.
Third, alternative credentialing systems — apprenticeships, competency-based assessments, direct skills verification — could theoretically achieve the same matching function at a fraction of the cost. The university's monopoly on signaling, not its educational superiority, may be its primary source of economic power.
“"The social function of the university is to certify intelligence, not to create it.”
Conclusion: What Your Degree Actually Proves
The debate between signaling and human capital is not merely academic — it cuts to the heart of whether the modern university justifies its existence. The evidence suggests a hybrid model where signaling accounts for a substantial minority of education's returns, and for many degrees, the majority.
Sources: Spence, M. (1973). "Job Market Signaling." Quarterly Journal of Economics. Becker, G. (1964). "Human Capital." Clark, D. & Martorell, P. (2014). "The Signaling Value of a High School Diploma." Journal of Political Economy. The Federal Reserve (2024). Student Loan Debt Statistics. College Board (2024). Trends in College Pricing.
