We develop a simple human capital model for optimum schooling length when earnings are stochastic, and highlight the pivotal role of risk attitudes and the schooling gradient of earnings risk. We use Spanish data to document the gradient and to estimate individual response to earnings risk in...

We use admission lotteries for higher education studies in the Netherlands to investigate whether someone's field of study influences the study choices of their younger peers. We find that younger siblings and cousins are strongly affected. Also younger neighbors are affected but to a smaller...

We study the effect of an Iranian educational policy implemented in 2012 that restricted access to higher education for women in 30% of Iran’s public universities, mostly in sciences and engineering. To analyze the effect of the policy, we use a triple difference strategy across gender,...

Studies on the determinants of the demand for higher education typically emphasizethe relevance of socio-economic factors, but leave the spatial dimensions of the prospectivestudents’ university choices largely unexplored. In this study, we investigate the determinantsof university entrance...

We use the method of Dominitz and Manski (1996) to solicit anticipated wagedistributions for continuing to a Master degree or going to work after completing theBachelor degree. The means of the distributions have an effect on intention to continue aspredicted by theory. The dispersions in these...

We investigate the major choice of college graduates where we make choice dependent on expected initial wages and expected wage growth per major. We build a model that allows us to estimate these factors semiparametrically and that corrects for selection bias. We estimate the model on the...

This paper identifies pivotal factors behind individual decision-making in the transition from high school to post-secondary education in the Netherlands. We apply a multinomial logit framework to individual data on post-secondary education choices. Specifically, our modeling approach...

We combine data from the Amsterdam secondary-school match with register data and survey data to estimate the effects of not being assigned to one's first-ranked school on academic outcomes and on a wide range of other outcomes. For identification we use that secondaryschool assignment in...