Languages use different systems for classifying nouns. Gender languages assign nouns to distinct sex-based categories, masculine and feminine. We construct a new data set, documenting the presence or absence of grammatical gender in more than 4,000 languages which together account for more than...

Languages use different systems for classifying nouns. Gender languages assign nouns to distinct sex-based categories, masculine and feminine. We construct a new data set, documenting the presence or absence of grammatical gender in more than 4,000 languages which together account for more than...

I take advantage of a sharp discontinuity in the probability of admission to an elite university at the admission score threshold, to estimate causal returns to college education quality. I use a newly constructed dataset, which combines individual administrative records about high school,...

I take advantage of a sharp discontinuity in the probability of admission to an elite university at the admission score threshold, to estimate causal returns to college education quality. I use a newly constructed dataset, which combines individual administrative records about high school,...

We apply the unordered monotonicity setting of Heckman and Pinto (2018) to estimate the distribution of response types and the counterfactual outcomes associated with the choice of a STEM or non-STEM college. Instrumental variation is induced by the proximity to universities offering STEM and/or...

This paper extends standard models of work-related training by explicitly incorporating workers' locus of control into the investment decision. Our model both differentiates between general and specific training and accounts for the role of workers and firms in training decisions. Workers with...

This paper extends standard models of work-related training by explicitly incorporating workers' locus of control into the investment decision. Our model both differentiates between general and specific training and accounts for the role of workers and firms in training decisions. Workers with...

We examine whether parental and school investments reinforce or compensate for student performance. Our analysis exploits school-starting-age rules in 34 countries, capturing achievement variation that arises because younger children typically underperform their older peers. Parents respond to...

This paper develops a dynamic model of schooling and occupational choices that incorporates personality traits, as measured by the "big five" traits. The model is estimated using the HILDA dataset from Australia. Personality traits are found to play an important role in explaining education and...

We study the impact of a student's ordinal rank in a high school cohort on educational attainment several years later. To identify a causal effect, we compare multiple cohorts within the same school, exploiting idiosyncratic variation in cohort composition. We find that a student's ordinal rank...