Does Exposure to International Students Shape the Long-Term Outcomes of Students?

Avdeev (2025) examines whether exposure to international students affects the long-term outcomes of native students. He uses administrative and survey data from the Netherlands covering about one million students over three decades. The study finds that a 10 percentage point increase in international student share raises natives’ likelihood of cohabiting with non-natives by 5.9%, marrying a non-native by 4.2%, and emigrating by 4.0%. It also increases pro-migration attitudes by about 0.16–0.30 standard deviations. However, he finds precisely estimated zero effects on employment, income, and entrepreneurship, suggesting strong social effects but no labor market impact  .

Why This Article Was Selected for The Policy Scientist

The policy relevance of international student flows extends well beyond higher education, shaping long-term social integration, migration attitudes, and cross-border mobility in advanced economies. As governments reconsider limits on international enrollment, this topic is especially timely because it intersects with broader debates over migration, labor markets, and social cohesion. Avdeev has contributed to an emerging literature on internationalization, and this study advances it by focusing on long-term social outcomes rather than short-term academic effects. The administrative dataset is unusually large and longitudinal, strengthening inference, though external validity beyond similar European systems may be limited. The quasi-experimental cohort design is credible.

Full Citation and Link to Article

Avdeev, S. (2025). University as a melting pot: Long-term effects of internationalization. American Economic Journal: Applied Economics (forthcoming). https://doi.org/10.1257/app.20250329

Central Research Question

This article investigates whether exposure to international students during university affects the long-term outcomes of native students. The central question is not limited to short-term academic performance but instead focuses on whether early exposure to international peers shapes social integration, attitudes toward migration, and labor market outcomes over an extended period. Specifically, the study asks whether variation in the share of international students within university cohorts causally influences natives’ likelihood of forming cross-national social ties, adopting more favorable views toward migration and cultural diversity, and pursuing international mobility. It also evaluates whether these social and attitudinal changes translate into measurable economic consequences, such as employment, earnings, and entrepreneurship, thereby addressing a key policy concern about whether internationalization has unintended labor market costs for domestic students.

Previous Literature

The study builds on two established literatures: research on peer effects in education and the contact hypothesis in social psychology. Prior work on international students has primarily examined short-term academic outcomes, including course performance, graduation rates, and early labor market indicators. These studies generally find small or null effects, but they rarely address longer-term consequences or social outcomes. In parallel, the contact hypothesis literature suggests that exposure to diverse groups can reduce prejudice and foster more inclusive attitudes, though much of this evidence derives from controlled settings such as randomized roommate assignments or structured interventions. This article extends both literatures by linking international exposure in higher education to long-term behavioral and attitudinal outcomes. It also complements earlier studies that employ Hoxby-style cohort variation to identify peer effects, while shifting the focus from academic achievement to social integration and migration-related preferences. In doing so, it reframes internationalization as a social and cultural phenomenon rather than solely an educational or economic one.

DataThe analysis relies on a comprehensive set of administrative and survey data from the Netherlands spanning more than three decades and covering approximately one million students. The administrative data include detailed records on higher education enrollment, labor market outcomes, cohabitation and marriage, and demographic characteristics, all linked through individual identifiers. These data allow for long-term tracking of individuals up to 25 years after university entry. The study also incorporates survey data from the National Student Survey, which captures attitudes toward internationalization during university, and the Longitudinal Internet Studies for the Social Sciences panel, which measures attitudes toward migration in adulthood. The breadth and linkage of these datasets provide a rare opportunity to observe both behavioral and attitudinal outcomes over time. The scale and longitudinal nature of the data enhance statistical power and enable precise estimation of even small effects, while the inclusion of both administrative and survey measures allows for a multidimensional assessment of outcomes.

Methods

The empirical strategy exploits cohort-to-cohort variation in the share of international students within specific university programs, using a quasi-experimental design rooted in the institutional features of the Dutch higher education system. Because admission is largely non-selective and program capacity is not tightly constrained, fluctuations in the proportion of international students across cohorts are plausibly exogenous. The study employs a two-way fixed effects framework, incorporating program fixed effects to control for time-invariant differences across fields and institutions, and cohort fixed effects to account for time-specific shocks. This approach isolates within-program variation over time, mitigating concerns about selection across programs. The identification strategy is supported by extensive balance tests showing no systematic relationship between international student share and observable characteristics of native students or program quality. While the design represents a credible causal inference approach, it does not involve random assignment, and therefore remains subject to the standard limitations of observational quasi-experiments. Nonetheless, the combination of institutional context, fixed effects, and robustness checks provides a strong basis for interpreting the estimated relationships as causal.

Findings/Size Effects

The results indicate that exposure to international students has substantial and persistent effects on social and attitudinal outcomes, but no detectable impact on economic outcomes. A 10 percentage point increase in the share of international students leads to a 5.9 percent increase in the likelihood that native students cohabit with a non-native and a 4.2 percent increase in the probability of marrying a non-native fifteen years after university entry. The same increase raises the probability of emigration by approximately 4 percent, indicating a shift toward greater international mobility. In terms of attitudes, exposure increases support for migration-related policies and cultural openness by roughly 0.16 to 0.30 standard deviations, suggesting meaningful changes in preferences rather than temporary behavioral adjustments.

In contrast, the study finds precisely estimated zero effects on labor market outcomes, including employment, income percentile, entrepreneurship, and workplace diversity, even up to 25 years after exposure. The precision of these estimates allows the author to rule out even modest negative effects, directly addressing concerns that international students might crowd out domestic students or reduce their economic prospects. Additional analyses show that social integration effects are stronger in larger and less competitive programs and are more pronounced when international students come from culturally similar regions, such as other European countries. Overall, the findings point to a pattern in which international exposure reshapes social networks and preferences without affecting economic performance.

Conclusion

The study concludes that internationalization in higher education primarily operates through social and cultural channels rather than economic ones. Exposure to international students fosters lasting changes in attitudes toward migration and increases cross-national social ties and mobility, while leaving labor market outcomes unchanged. These results challenge policy narratives that focus narrowly on economic competition and instead highlight the broader societal implications of international student flows. By demonstrating that universities can influence long-term preferences and behaviors, the study positions higher education as a key site for shaping social integration in increasingly globalized societies. At the same time, the absence of labor market effects suggests that concerns about economic displacement may be overstated in this context. The findings underscore the importance of evaluating internationalization policies using a multidimensional framework that includes social and attitudinal outcomes alongside traditional economic metrics.