Targeting the Gender Placement Gap: Marks versus Money

Using data from the Turkish University Entrance Exam, we investigate the extent of the gender gap in college placement, its underlying causes, and potential policy interventions. We estimate preferences using a novel approach which improves our ability to capture substitution patterns and find clear evidence that placement differences are primarily driven by preference differences across gender. We compare stipend subsidies to score bonuses that achieve gender parity. Score subsidies improve the welfare of women almost entirely at the cost of men with similar scores and favor high-income women. Stipend subsidies improve the welfare of women, but at little cost to men and favor low-income women. Our work is the first to show that how gender neutrality is achieved matters to society.