What it means to be a "student" varies within and between countries. Apart from the wide variety of school types and school quality that is experienced by young people, there also is, accompanying increased rates of school participation, a growing population of students who work part-time. The theoretical and actual consequences of student work have long been in dispute. This article reformulates the dispute as an empirical question that can be addressed using cross-national testing data and student background information from the Trends in International Math and Science Study (TIMSS). Drawing information from 20 countries with distinctive proportions of students who reported paid and unpaid work, this study first compares their academic achievement in each country. Next, regression analysis is used to control for students' home resources, and estimates are made of the effects of work and the differences in these effects cross-nationally. Finally, hierarchical linear models are estimated in each country so as to control for school effects, and to take into account the fact that working students may be clustered in lower-achieving schools. The results show that work after school, whether paid or unpaid, never positively affects academic achievement. However, after controlling for home resources and school effects, work negatively affects achievement only in certain countries. The article concludes with a discussion of the ways to interpret international differences in the effect of students' work. (Contains 4 figures and 8 tables.)