Why a Degree Became a Shortcut—and Why the Shortcut Is Failing
This chapter diagnoses a specific weakness in credential-based hiring: a degree can be a convenient signal without being a reliable description of capability. The focus is how degree inflation spread, what evidence says about schooling and actual skills, and why the resulting mismatch matters to employers as well as workers. The distinction is useful because it separates a degree requirement that reflects a genuine legal or technical need from one that survives mainly because it is an easy way to reduce uncertainty.
How the degree became a general-purpose filter
A hiring manager rarely has perfect information about an applicant. Before someone joins the organisation, the manager may not know how that person will solve unfamiliar problems, communicate under pressure or learn the specific work. A credential offers a shortcut through that uncertainty. It is visible, familiar and relatively easy to sort in a form or applicant-tracking system. The employer does not have to claim that a degree is the skill itself. It only has to treat the degree as evidence that the applicant is more likely to possess qualities the job needs.
That logic can be reasonable in some settings. A course of study may build knowledge, practice and persistence. A professional qualification may be a legal condition of work. But a proxy becomes a problem when it travels beyond the situations in which it is informative. Instead of asking what a role requires and how an applicant can demonstrate it, the organisation begins to treat the credential as a general-purpose answer. The filter then persists even when the relationship between the credential and the work is weak.
The growth of higher education made that extension easier. The Burning Glass Institute and Harvard Business School report describes a rise in the share of Americans who had graduated from college from 8% in 1960 to 38% in 2022. The occupational distribution changed much more slowly. In other words, the workforce acquired more degrees faster than the work itself was reorganised around degree-dependent tasks. Employers began adding degree requirements to roles that had not necessarily changed underneath the posting.
This is the report’s meaning of degree inflation. It is not simply that more people went to college, nor that college lost all value. It is that a qualification originally used for some roles became attached to more roles because it was available as a convenient sorting device. Once enough employers used it, the requirement could also become self-reinforcing. Hiring managers saw degrees in comparable postings, candidates learned that degrees were expected, and the credential acquired a social legitimacy that obscured the original question: what must a successful employee actually be able to do?
The filter solves one organisational problem while creating another. It reduces the number of applications that need immediate attention, but it also removes candidates before the employer can inspect their relevant experience, informal learning or practical ability. That trade-off is easy to miss because the cost is distributed. The employer sees a shorter list and perhaps a faster process. The excluded worker sees a closed door. The organisation may never know which capable applicants were screened out, so the lost opportunity does not appear in its hiring data as a failed decision.
The effect can be especially sharp for jobs that once served as stepping stones. The HBS report argues that workers without degrees can be shut out of roles that historically offered upward mobility, while graduates may end up in jobs that their parents could have obtained without the time and expense of college. A requirement can therefore misallocate opportunity in both directions. It may deny a non-degree worker a chance to build the experience that later jobs ask for, while encouraging a degree-holder to compete for work whose tasks do not require that credential.
The central diagnostic distinction is between a proxy and a measure. A proxy is an indirect clue that may correlate with a desired outcome. A measure attempts to observe the relevant capability more directly. A degree can be a useful proxy for some kinds of knowledge or preparation, but it does not identify the full skill bundle required by every job. A work sample, structured interview or practical assessment may provide more direct evidence for a specific task, although each method can also be poorly designed. The choice is not between perfect measurement and no measurement. It is between acknowledging what the signal actually tells the employer and pretending it answers more than it does.
The same distinction applies to work experience. “Three years in a similar role” sounds more specific than “has a degree,” but the label can hide what the person actually did. Experience may reflect organisation-specific systems, routines or relationships that do not transfer to the new job. It may also contain valuable task-relevant knowledge. The important question is not whether experience exists, but which capabilities it developed and whether those capabilities are relevant to the new role.
The diagram separates two paths that hiring systems often collapse. Formal education and work or informal learning can both contribute to underlying capabilities. A credential filter, however, selects on the visible proxy rather than on the capabilities needed for the role. When the filter is badly aligned with the work, the result can be an under-skilled hire, an over-skilled worker in a poorly matched role, or a capable person who never enters consideration. The diagram does not imply that education and experience are irrelevant. It shows why they should be interpreted as routes to capability, not treated as capability in the abstract.
A further complication is that the proxy can acquire symbolic force. Degrees may signal not only knowledge but belonging, legitimacy and the kind of person an organisation expects to employ. The OECD notes that these cultural meanings help preserve credential-based practices even when employers are interested in skills. A manager who has succeeded through a familiar educational route may also find that route easier to recognise in another candidate. What looks like a neutral standard can therefore combine convenience with inherited expectations.
This helps explain why degree inflation cannot be diagnosed as a simple mistake by an individual recruiter. It is a system of incentives and habits. Recruiters need a process that can handle large applicant volumes. Hiring managers carry responsibility for an expensive vacancy and may be judged more harshly for a bad hire than for a missed candidate. Internal norms make one kind of evidence feel safer than another. Unless the organisation creates a credible way to assess capability, removing the formal requirement may leave all of those conditions intact.
What the proxy misses—and why that matters
The first limit of the degree proxy is variation. The OECD reports that people with the same amount of schooling can have substantially different levels of assessed numeracy. In the cited analysis of the Survey of Adult Skills, years of schooling account for about one-fifth of the differences in numeracy scores, expressed as an R² of 0.20. That is a positive relationship, not a negligible one. It is also far from a complete account of the variation. The statistic says that education is informative about one measured skill, while leaving much of the difference unexplained. It does not say that numeracy scores alone determine job performance, or that more schooling has no economic value.
This is the kind of result that a hiring policy can easily misuse. If education and skill are correlated, an employer may reasonably use education as one piece of evidence when direct assessment is costly. But correlation does not make the credential a substitute for the capability. The more a role depends on a particular technical, interpersonal or adaptive skill, the more important it becomes to know whether that skill is present rather than infer it from an undifferentiated qualification.
The mismatch can run in both directions. A person can be formally under-qualified yet possess enough skill for the work, or even more skill than the job uses. Another person can hold the expected qualification and still lack a capability the role requires. The OECD reports that even among workers whose qualifications formally match their job level, around one-third still exhibit some form of skill mismatch. The category “qualified” therefore cannot be read as “fully capable,” just as “under-qualified” cannot be read as “incapable.”
The same logic weakens years of experience as a universal filter. The OECD cites research finding that general cognitive ability tests and work-sample tests had higher average validity for predicting job performance than years of similar-role experience or years of education across the recruitment methods studied. It also cites a later meta-analysis concluding that pre-hire work experience is generally a poor predictor of future performance. These findings do not make a work history unimportant. They suggest that the predictive part of experience is the task-relevant knowledge and skill it produced, not the passage of time or the prestige of the previous employer by itself.
When the proxy misses capability, the consequence is not confined to a candidate’s sense of fairness. It affects the organisation’s allocation of people to work. An employer can report a shortage of talent while rejecting workers who could perform the role with a short period of training. A worker can occupy a job that does not use their abilities while another vacancy remains difficult to fill. The OECD describes this as a matching problem: employers and workers have incomplete information about one another, and coarse credentials make it harder to align the requirements of a job with the capabilities of a person.
Better matching is not merely a matter of filling vacancies faster. If a person’s skills are used more fully, the role may provide greater autonomy, motivation and job satisfaction. The OECD reports that workers whose skills are well matched to their jobs are systematically more likely to report being satisfied at work, with gaps exceeding ten percentage points in some countries in its adjusted analysis. The direction of this relationship supports the importance of fit, but it does not prove that changing a hiring filter alone will create satisfaction. Working conditions, pay and management also matter. The point is that a hiring system that obscures capability can contribute to a poor match before those later factors are even considered.
There is also an opportunity and mobility problem. The HBS report estimates that about 62% of Americans lack a bachelor’s degree. A credential-first system consequently makes access to many roles depend on a path that most adults have not taken. The OECD adds that adults from families without tertiary education remain more likely to lack a degree even when they have high numeracy proficiency. Formal credentials can therefore transmit differences in social background into access to high-productivity work, including for people whose measured skills are strong.
A skills-first approach could weaken that link by accepting evidence from work, informal learning, practical assessments and targeted training. But “could” is important. Recognition does not happen automatically when a requirement disappears. Someone still has to decide which experiences count, how competence will be assessed and whether the evidence will be trusted. If those decisions are left to unstructured judgment, the organisation may replace one exclusion mechanism with another.
The diagnosis is thus provisional in a precise way. The evidence shows that degrees and years of experience are imperfect proxies, that qualifications do not map cleanly onto actual skill, and that these gaps can produce mismatch and unequal access. It does not show that education is irrelevant, that every non-degree candidate is ready for every role, or that a skills test will remove uncertainty. It identifies a reason to improve the evidence used in hiring, not a reason to pretend uncertainty has disappeared.
The most defensible boundary is necessity. Where a licence, safety rule or body of specialised knowledge makes formal education essential, the qualification should not be discarded for the sake of a slogan. Where the degree requirement is primarily an inherited preference, the employer has stronger grounds to examine the underlying skills and accept more than one route to them. That decision requires a view of the work, not a blanket position for or against college.
This diagnosis also clarifies what the next piece of evidence must do. It is not enough to show that degree requirements are historically inflated or statistically imperfect. We need to know whether employers that remove the filter actually hire a different mix of people. If the answer is only a small change, the remaining explanation will not be found in the credential’s history alone. It will be found in what happens after the job posting—inside screening, comparison, assessment and managerial choice.