Non-degree credentials are rapidly multiplying as alternatives to traditional higher education, but in a largely unregulated market. This column draws on nearly 40 million US worker resumes to analyse the returns to non-degree credentials. The findings suggest that they can deliver substantial wage gains if they are job-relevant, especially for workers without college degrees and those early in their careers. However, these are also the workers who are least likely to hold such credentials. Changing this requires not just more credentials, but better information, lower barriers, and quality assurance that protects workers from low-value programmes.
Non-degree credentials – badges, certificates, certifications, licences, and microcredentials – have been multiplying at breakneck speed as alternatives to traditional higher education. By recent counts, over 1.5 million non-degree credentials now exist in the US, with more than one-quarter of American adults holding at least one (Cronen et al. 2017, Credential Engine 2025). Their promise is compelling: accessible, targeted pathways to skill development that sidestep the escalating costs of traditional education while meeting urgent demands for workforce reskilling. But hard evidence on whether these credentials actually pay off remains surprisingly scarce.
In new research (Levy Yeyati et al. 2025), my co-authors and I provide the first comprehensive analysis of the returns to non-degree credentials across all major credential types, drawing on 37.7 million US worker resumes and using machine learning to identify and standardise 54.3 million credentials. What we find tells a nuanced story. Non-degree credentials can deliver substantial wage gains, but those returns hinge critically on three factors: whether the credential is relevant to the worker’s job, what type of credential it is, and who holds it.
The bottom line? Rigorous, job-relevant non-degree credentials can meaningfully narrow earning gaps for workers without college degrees and those early in their careers. But without better quality signals and information, the credential boom risks leaving vulnerable workers stuck with low-value investments.
The non-degree credentials landscape: Concentrated where tech meets professionalisation
Credentials aren’t distributed evenly across the labour market. As Figure 1 shows, they cluster heavily in knowledge-intensive occupations: 23.4% of computer and mathematical workers hold non-degree credentials, along with 16.9% in business and finance, and 16.4% in architecture and engineering – all more than triple the 7.2% overall average. Meanwhile, fewer than 2% of workers in agriculture or cleaning occupations list credentials on their resumes.
Figure 1 Variation in prevalence of non-degree credentials by occupation
This concentration reflects both longstanding professional certification traditions and rapid tech skill churn. The harder question is: can non-degree credentials reach workers in other sectors, or will they remain concentrated among the already-advantaged?
Job relevance is what separates signal from noise
Our core finding is straightforward but consequential: the labour market rewards credentials that match what workers actually do, not credentials in general. We measure job relevance by comparing how common each credential is within specific occupations versus across all workers. A JavaScript certification, for instance, appears 10.66 times more frequently among software developers than in the overall workforce, while project management credentials are 3.82 times more common among management analysts.
When we control for education, experience, demographics, and location, we find that a worker’s first job-relevant credential is associated with a 3.8% wage premium – more than double the 1.8% premium for a first credential that is unrelated to the worker’s occupation. The gap widens even more for accumulation: each additional job-relevant credential brings about 1.0% more in wages, while piling up irrelevant credentials shows no returns at all, and sometimes even negative associations.
This pattern suggests non-degree credentials work through two channels: any credential provides some baseline signal of motivation and trainability, but job-relevant credentials also convey actual productivity gains from aligned skills. The complete absence of returns to irrelevant accumulation drives home a critical point: in an increasingly crowded credential marketplace, quality and fit matter far more than quantity.
The workers who need credentials most benefit most, but face the highest barriers
The wage boost from non-degree credentials varies dramatically depending on who’s earning them, in ways that point toward both promise and frustration (Figure 2). Workers without bachelor’s degrees see a 6.8% premium from their first job-relevant credential – nearly double the 3.4% gain for college graduates. Early-career workers (below median experience) get a 6.1% bump from their first relevant credential versus just 2.8% for experienced workers.
Figure 2 Returns to non-degree credentials are significantly higher for non-college and early-career workers
Returns to accumulation follow the same pattern. Each additional relevant credential yields 1.5% higher wages for non-degree holders compared to 1.0% for college graduates. For early-career workers, the marginal return is 2.3% per credential, while for experienced workers it essentially disappears.
These differentials make sense. Non-degree credentials seem to function as partial substitutes for formal degrees among non-college workers, and as early-career signals before workers have built up track records. For workers lacking traditional credentials or extensive experience, a relevant certification can open doors that would otherwise stay closed.
But here’s the frustrating paradox: the workers who benefit most are the least likely to have credentials. As Figure 3 shows, non-degree-credentials prevalence rises steadily with education level. Workers with a master’s degree are nearly twice as likely to hold credentials (27.7%) as those with only a high school diploma (14.4%), the opposite of what we would expect if non-degree credentials were genuinely democratising access to good jobs.
Figure 3 Prevalence of non-degree credentials by education level
This ‘credential paradox’ likely reflects multiple barriers facing non-degree workers, including financial constraints, less information about which credentials actually pay off, and credential programmes often designed with more advantaged learners in mind (Bitar et al. 2024). College graduates, meanwhile, more often get credential costs covered by employers, tap professional networks for insider knowledge, and face lower risks when investing in credentialing.
Certifications build skills; badges signal quality – and the difference matters
Not all credentials work the same way. When we break down returns by credential type, distinct patterns emerge that reveal different underlying mechanisms.
Certifications – typically requiring proctored exams, regular renewal, and third-party validation – show patterns consistent with genuine skill-building. A first job-relevant certification is associated with a 4.1% wage premium, with each additional relevant certification adding another 2.2%. Critically, these returns depend entirely on job relevance: irrelevant certifications show essentially no initial premium and negative returns to accumulation. For non-college workers specifically, a first relevant certification delivers a 7.1% wage premium (our largest single coefficient), with each additional one adding 3.5%. This looks like human capital accumulation: each certification adds verifiable skills that stack productively.
Badges and certificates work completely differently. Badges show a 5.3% premium even when job-irrelevant, but negative returns (-3.0%) to relevant accumulation. Certificates deliver small premiums for both relevant and irrelevant possession (0.8% and 1.4%, respectively) but negative returns to accumulating either type. This fits a pure signalling story: earning your first badge or certificate separates you from workers with no credentials at all, but additional ones do not add new information and may even signal scattered focus or poor judgement about which credentials to pursue.
Microcredentials show the strongest initial returns: a first relevant microcredential associates with a 4.5% premium, with accumulation adding 3.4% per credential. But they display an interesting twist: they complement formal education rather than substituting for it. Additional relevant microcredentials are associated with 4.2 percentage points higher returns among college graduates than non-graduates, suggesting they work best as specialised skill updates for workers with strong educational foundations.
The fact that certifications reward relevant accumulation while badges and certificates do not – despite similar prevalence on resumes – suggests workers and employers need clearer distinctions between rigorous validation and lighter-touch credentials.
Making credential markets work: What’s missing and what’s needed
Our findings point to real potential for well-designed non-degree credentials to create alternative pathways to economic mobility, especially for workers the current system leaves behind. But several gaps stand between potential and reality.
First, quality signals are nearly nonexistent. Workers cannot easily tell rigorous certifications from low-stakes participation badges. The fact that irrelevant credential accumulation shows zero or negative returns – and that badge accumulation can hurt wages for some groups – reveals the risk of poor-quality investments. Better information infrastructure is critical. Recent work on skills-based hiring frameworks (Levy Yeyati and Seyal 2025) highlights how combining structured skill taxonomies with adaptive ontologies could help – but only if credential providers map their offerings to these frameworks and make the connections transparent.
Second, occupation-specific guidance is lacking. Our relevance weighting shows that the same credential can be extremely valuable in one occupation and worthless in another. Workers need clear, accessible information about which credentials pay off in which jobs. This is not just about better websites – it requires systematic infrastructure linking credential registries to occupational outcomes.
Third, quality assurance mechanisms are minimal. We documented 35% growth in non-degree credentials prevalence in a single year, with badges and microcredentials growing fastest. This proliferation is happening in a largely unregulated market. Without mechanisms to verify quality and prevent exploitation, information-poor workers face a serious risk of predatory programmes.
These gaps suggest concrete policy priorities. Public subsidies for credential acquisition could be justified on equity grounds; our results show the biggest gains accrue to non-college and early-career workers. But blanket subsidies would be wasteful; quality standards and targeting are essential. Investment in credential registries, outcome tracking, and transparent quality assurance would help workers navigate the marketplace. The differentiated returns across credential types suggest training providers should recognise that the market rewards investments in rigorous validation, ongoing assessment, and renewal requirements.
Looking ahead
The credential boom could fundamentally shift how workers develop and signal skills – but only with the right infrastructure and information.
Rigorous, occupationally aligned credentials can provide genuine alternatives to four-year degrees, with the largest benefits flowing to workers who need them most. That’s the promise. The peril lies in proliferation without quality control: workers investing time and money in credentials that look legitimate but deliver no labour market returns.
The patterns we document – sharp differences between relevant and irrelevant credentials, between certifications and badges, and between non-college and college workers – all point to the same conclusion: one-size-fits-all approaches won’t work. Effective credential markets need differentiation, transparency, and clear information to guide worker investments.
Perhaps most importantly, our finding that non-degree credentials deliver the largest returns to workers without traditional degrees suggests real potential for modestly reducing educational inequality – if we can solve the access paradox. Right now, the workers who benefit most are least likely to earn credentials. Changing that requires not just more credentials, but better information, lower barriers, and quality assurance that protects workers from low-value programmes.
The pieces for a functioning credential marketplace exist. What’s needed now is the infrastructure to connect them: interoperable data systems, transparent quality signals, and occupation-specific guidance that helps workers invest wisely. Without that infrastructure, credential proliferation risks creating more noise than signal.
Source : VOXeu
