We extract estimation results on the Mincer earnings function from four earlier studies and add new results from a recent dataset. We analyse differences related to differences in earnings concepts, in sampling frame and differences among studies that cannot be explained. Jointly, the studies...

This paper studies the impact of exam luck on individuals’ education and labor market success. We leverage unique features of the Norwegian education system that produce random variation in the content of the exams taken by students at the end of high school. Lucky students take exams in...

Using an originally constructed dataset that follows 30,000 Italian individuals from high school to the labor market, we analyze whether the gender composition of peers in high school affected their choice of college major, their academic performance and their labor market income. We exploit the...

This paper presents the first longitudinal estimates of the effect of work-related training on labor market outcomes in Switzerland. Using a novel dataset that links official census data on adult education to longitudinal register data on labor market outcomes, we apply a regression-adjusted...

How important is mastering information and communication technologies (ICT) in modern labor markets? We present the first evidence on this question, drawing on unique data that provide internationally comparable information on ICT skills in 19 countries. Our identification strategy relies on the...

Students' choices in education can only be based on expected outcomes. Econometric models that infer expectations based on ex post outcomes impose a rational structure of expectations on school performance and post-graduation earnings. Direct surveys suggest much ignorance and fuzziness. We use...

This paper investigates the economic returns to language skills and bilingualism. The analysis is staged in Kazakhstan, a multi-ethnic country with complex ethnic settlement patterns that has switched its official state language from Russian to Kazakh. Using two newly assembled data sets, we...