Are Apprenticeships Effective at Producing Employment and Earnings Gains?

Darolia and Turner (2026) ask whether the rapid expansion of apprenticeships in the United States is being matched by credible evidence on program quality, outcomes, and skill development. They examine administrative data from the U.S. Department of Labor’s Registered Apprenticeship system, state-level apprenticeship records, and postsecondary enrollment and completion data. They find that apprenticeships have grown substantially since 2011, expanded into new industries, and diversified demographically, with female participation more than tripling since 2015. However, evidence on earnings gains, skill transferability, and program quality remains limited. Apprenticeship participation is still small relative to associate degree completions in most states, and growth has been largest in lower-wage sectors, raising concerns about long-term mobility.

Why This Article Was Selected for The Policy Scientist

This article addresses an issue of broad and growing importance: how earn-and-learn pathways fit into the future of postsecondary education and workforce development in the United States. As policymakers invest heavily in apprenticeships amid rising skepticism about traditional college pathways, understanding what these programs can and cannot deliver is timely. Darolia and Turner, who have contributed extensively to research on postsecondary finance, training programs, and labor market outcomes, situate apprenticeships within this larger institutional shift. The article is valuable for clarifying the limits of existing administrative data and highlighting fragmentation across states, which constrains generalizability.

Full Citation and Link to Article

Darolia, R., & Turner, J. (2026). Apprenticeships in the United States: Emerging opportunities and evidence gaps. Journal of Policy Analysis and Management, 45(1). https://doi.org/10.1002/pam.70082

Central Research Question

This article asks how well the current U.S. apprenticeship system is understood at a moment of rapid policy expansion and whether existing evidence is sufficient to guide large-scale public investment. The authors focus on a broader diagnostic question: what do we know about the scope, structure, and outcomes of apprenticeships in the United States, and where are the most consequential gaps in evidence? They are especially concerned with whether apprenticeships, as they expand beyond the skilled trades and into white-collar sectors, are delivering high-quality training, transferable skills, and meaningful labor market returns. Implicitly, the article asks whether policy enthusiasm has outpaced empirical knowledge, and what risks arise when expansion proceeds without reliable data on quality, equity, and effectiveness.

Previous Literature

The article builds on several strands of literature. One focuses on the economics of apprenticeship systems, particularly European models, which document relatively strong earnings returns for participants and cost recovery for firms through productivity and retention gains. This literature, including work by Lerman and others, often emphasizes institutional coordination, standardized curricula, and industry-wide skill certification as key features explaining success abroad. A second strand examines U.S. registered apprenticeships, typically using administrative or survey data to compare earnings of apprentices with non-apprentices in similar occupations. Much of this work relies on selection-on-observables approaches and finds positive earnings associations, particularly in construction and manufacturing.

The authors situate their contribution alongside broader research on postsecondary education, workforce training, and human capital formation, including debates over general versus firm-specific skills and concerns about occupational tracking. They also draw on literature assessing program scale-up, which shows that rapid expansion can dilute quality when instructor supply, oversight, or institutional capacity lags behind enrollment growth. While acknowledging these contributions, the authors stress that U.S.-based evidence remains fragmented, uneven across states and industries, and weakly positioned to answer causal questions about returns, skill portability, and long-term mobility—especially in newer, nontraditional apprenticeship sectors.

Data

The article synthesizes multiple administrative and public data sources rather than constructing a single unified dataset. The core source is the U.S. Department of Labor’s Registered Apprenticeship data, drawn from RAPIDS and ApprenticeshipUSA dashboards, which provide information on active apprentices, new entrants, completions, industries, and demographics over time. These data are supplemented with postsecondary enrollment and completion figures from the Integrated Postsecondary Education Data System (IPEDS) to contextualize apprenticeship participation relative to associate degree production. The authors also rely on state-level policy information on tax credits and tuition subsidies, as well as secondary sources describing unregistered apprenticeships and work-based learning programs that fall outside the federal reporting system.

A central theme of the article is that the available data are incomplete. Not all states historically reported individual-level data, unregistered apprenticeships are largely invisible, and naming inconsistencies across work-based learning programs blur conceptual boundaries. As a result, even basic descriptive statistics—such as the total number of apprentices or their distribution across sectors—are subject to measurement error. The authors treat these limitations as a substantive finding rather than a mere caveat, arguing that data fragmentation itself poses a policy risk as the system scales.

Methods

The analysis is descriptive and comparative. The authors track trends over time in apprenticeship participation, entry, and completion, examine industry and demographic composition, and compare apprenticeship completions with associate degree completions at the state level. They also document variation in state policy approaches, such as employer tax credits versus tuition assistance, without attempting to estimate their causal effects. Visualizations and ratios are used to highlight cross-state heterogeneity and growth patterns rather than to test formal hypotheses.

No causal inference methods or randomized designs are employed. The authors are explicit about this choice, positioning the article as a policy overview and evidence audit rather than an impact evaluation. They argue that existing data constraints make rigorous causal analysis difficult in many contexts and that identifying these constraints is a necessary step toward future evaluation. Throughout, the emphasis is on mapping institutional structure, participation patterns, and known unknowns, rather than estimating treatment effects.

Findings/Size Effects

The authors document substantial growth in U.S. apprenticeships since the early 2010s, following a sharp decline during the Great Recession. Growth has been driven primarily by increases in new entrants rather than improved completion rates. While construction and manufacturing still account for a majority of apprentices, the fastest growth since 2015 has occurred in business, finance, healthcare, hospitality, and service sectors. Female participation more than tripled over this period, though women remain underrepresented overall and are concentrated in lower-paying apprenticeship fields.

In size terms, apprenticeships remain small relative to traditional postsecondary pathways. In most states, apprenticeship completions are far fewer than associate degree completions, though a small number of states—such as Alaska, Missouri, and South Carolina—exhibit relatively high ratios. State policy variation is substantial, with at least 28 states offering employer tax credits and 17 providing tuition assistance, yet there is little evidence on how these incentives affect participation or outcomes.

The article highlights several risks associated with current expansion patterns. First, rapid growth without standardized curricula or industry coordination may lead to declining quality. Second, expansion into firm-specific, white-collar roles raises concerns about skill portability and long-term mobility, especially in fast-changing labor markets. Third, the concentration of growth in lower-wage sectors may limit apprenticeships’ role as a broadly mobility-enhancing alternative to college. Importantly, the authors note that while prior studies often report positive earnings associations for apprentices, the magnitude and durability of these gains across industries and populations remain uncertain due to weak causal identification.

Conclusion

The article concludes that U.S. apprenticeship policy is advancing faster than the evidence base needed to guide it. While apprenticeships offer a potentially valuable bridge between education and work, the current system is characterized by fragmented governance, incomplete data, and limited understanding of program quality and long-term outcomes. The authors argue that these gaps are especially consequential as apprenticeships expand beyond their traditional domains and attract substantial public investment.

They emphasize the need for improved data infrastructure, standardized skill frameworks, and rigorous evaluation strategies. In particular, future research would benefit from quasi-experimental designs or randomized trials that can credibly estimate returns to apprenticeship participation, compare apprenticeships to alternative education pathways, and assess skill transferability across firms and regions. Without such evidence, policymakers risk scaling programs whose benefits, distributional effects, and opportunity costs are poorly understood. The article’s primary contribution lies in clarifying these stakes and providing a structured roadmap for the next generation of apprenticeship research.