What Are the Long-Term Effects of Raising the Compulsory Schooling Age on Education and Labor Market Outcomes?

Nelissen and De Witte (2026) examine whether raising the compulsory schooling age from 17 to 18 improves long-term educational and labor market outcomes. They use Dutch administrative microdata tracking individuals from adolescence into adulthood, exploiting a quasi-experimental reform. They find the policy reduced dropout by about 1.3 percentage points and increased high school completion by roughly 0.5 points, with stronger effects for vocational students. By age 31, employment rises about 0.5 points and wages increase roughly 2%. They conclude that extending compulsory education yields modest but meaningful gains, especially for lower-track students.

Why This Article Was Selected for The Policy Scientist

The policy question addressed in Nelissen and De Witte (2026) is central to contemporary debates over human capital formation, labor market resilience, and inequality. As economies increasingly reward formal credentials, whether extending compulsory schooling meaningfully alters long-run outcomes has broad implications for productivity, social mobility, and public expenditure. This is especially timely given renewed international interest in adjusting school-leaving ages amid shifting labor market demands and persistent dropout concerns. The article contributes by re-examining a well-studied policy using recent data and long-run outcomes, extending earlier work by the same authors and others. The administrative dataset is high quality and longitudinal, though external validity may be limited beyond similar institutional contexts. The quasi-experimental design is a credible causal inference approach, strengthening identification relative to standard regression methods.

Full Citation and Link to Article

Nelissen, H., & De Witte, K. (2026). The long run impact of increased compulsory education age: Evidence from administrative data in the Netherlands. Education Finance and Policy. https://doi.org/10.1162/EDFP.a.448

Central Research QuestionThis article examines whether increasing the compulsory schooling age from 17 to 18 produces sustained improvements in individuals’ educational attainment and labor market outcomes. The central question is not limited to immediate behavioral responses, such as reduced dropout, but extends to whether such a policy meaningfully alters long-run trajectories, including high school completion, postsecondary attainment, employment, and wages. The authors explicitly situate the question within contemporary institutional conditions, asking whether findings from earlier reforms—often conducted in different economic contexts—remain applicable today. A key dimension of the inquiry is heterogeneity: whether the policy affects students differently across educational tracks, particularly vocational versus academic pathways. The study also implicitly evaluates whether extending compulsory schooling to the point of likely credential completion—rather than simply adding years—generates measurable returns in adulthood.

Previous LiteratureThe study builds on a well-established literature examining the effects of compulsory schooling laws on educational attainment and economic outcomes. Earlier work, including contributions by Angrist and Krueger and Oreopoulos, generally finds that additional schooling induced by policy reforms can increase earnings and employment, consistent with human capital theory. At the same time, other studies have documented weak or null effects when policy changes do not coincide with meaningful credential attainment, suggesting that the marginal year of schooling alone may not generate returns. This has led to a distinction between the signaling value of credentials and the productivity effects of additional schooling.

More recent research has introduced nuance, emphasizing heterogeneity across student populations and institutional contexts. For example, evidence indicates that vocational students may respond differently to schooling mandates than academically oriented students, due to differing labor market opportunities and baseline dropout risks. The present study extends this literature by examining a relatively recent reform in the Netherlands, thereby addressing concerns about external validity over time. It also builds directly on prior work by Cabus and De Witte, which focused on short-term dropout effects, by expanding the analysis to long-run outcomes. In doing so, the article contributes to an ongoing reassessment of whether compulsory schooling policies remain effective under modern economic conditions.

DataThe analysis relies on comprehensive administrative microdata from Statistics Netherlands, covering the full population over an extended period. These data include detailed longitudinal information on individuals’ educational trajectories, labor market participation, wages, and demographic characteristics. The dataset allows the authors to follow cohorts born in 1988 and 1990 from adolescence into their early thirties, thereby capturing both short-term and long-term outcomes.

A key advantage of the dataset is its administrative origin, which minimizes measurement error relative to survey-based data and virtually eliminates attrition. The data include precise indicators of dropout, high school completion, postsecondary attainment, and employment status, as well as hourly wages for individuals in paid employment. The ability to link individuals across multiple administrative registers enables a comprehensive view of life-course outcomes.

However, the data are limited to the institutional context of the Netherlands, which features a highly structured and tracked education system. While this enhances internal consistency, it may constrain generalizability to countries with different educational institutions or labor market structures. Nonetheless, the richness and scale of the dataset provide a strong empirical foundation for causal analysis.

MethodsThe study employs a quasi-experimental research design that exploits a 2007 policy reform raising the compulsory schooling age from 17 to 18. Treatment assignment is determined by date of birth relative to a cutoff, creating a natural experiment in which individuals born just before and after the cutoff are plausibly comparable. The authors implement a difference-in-discontinuities (RD-DD) design, combining regression discontinuity with a difference-in-differences framework to isolate the causal effect of the reform.

This approach strengthens identification by controlling for potential confounding factors related to age and cohort effects. By comparing outcomes across cohorts and around the policy cutoff, the authors address concerns about seasonality, relative age effects, and underlying trends. The inclusion of an earlier cohort (1988) as a control group allows them to difference out pre-existing discontinuities and improve causal inference.

The empirical specification incorporates a range of control variables and fixed effects, including neighborhood-level clustering, to account for unobserved heterogeneity. The authors also conduct robustness checks, including placebo tests and event-study analyses, to validate key assumptions such as the absence of manipulation around the cutoff and the plausibility of local parallel trends.

Overall, the methodological approach aligns with contemporary standards in applied microeconomics and represents a credible causal inference strategy. While not a randomized controlled trial, the design approximates experimental conditions and offers stronger identification than conventional multivariate regression.

Findings/Size EffectsThe results indicate that raising the compulsory schooling age produces measurable but modest improvements in educational and labor market outcomes, with substantial heterogeneity across student groups. The reform reduces dropout rates at age 18 by approximately 1.3 percentage points in the full sample, with larger effects—around 2 percentage points—among vocational students. High school completion increases by about 0.5 percentage points overall, again driven primarily by vocational tracks.

The effects on postsecondary attainment are more limited. There is no significant increase in college completion for the full sample, though students in general education tracks experience a modest increase of roughly 3 percentage points by age 31. The policy also increases the likelihood of obtaining an adult education credential by about 0.5 percentage points, with stronger effects among pre-university students.

Labor market effects emerge only in the longer run. By age 31, employment increases by approximately 0.5 percentage points in the full sample, accompanied by a roughly 2 percent increase in hourly wages. These effects are concentrated among vocational students, who experience employment gains of about 1 percentage point and wage increases of roughly 3 percent. The delayed nature of these effects suggests that the returns to additional schooling may materialize gradually as labor market experience accumulates and macroeconomic conditions stabilize.

Overall, the findings suggest that the reform primarily benefits students at higher risk of dropout, while having limited impact on individuals already on academic trajectories. The magnitude of the effects is modest but consistent across multiple outcomes.

ConclusionThe study concludes that increasing the compulsory schooling age yields modest but meaningful improvements in both educational attainment and labor market outcomes, particularly for students in vocational tracks. The results indicate that aligning compulsory schooling with the point of credential completion is a critical mechanism through which such policies generate returns.

At the same time, the limited effects on postsecondary attainment and the delayed emergence of labor market gains highlight the constraints of compulsory schooling policies as a standalone intervention. The findings suggest that additional years of mandated schooling do not automatically translate into higher-level educational attainment or immediate economic benefits, especially for students already positioned to complete their education.

The article contributes to the literature by providing updated evidence from a modern policy context and by extending the analysis to long-run outcomes. Its use of high-quality administrative data and a rigorous quasi-experimental design strengthens the credibility of the results. While the findings are most directly applicable to similar institutional settings, they offer broader insights into the conditions under which compulsory schooling reforms are likely to be effective.