Multidimensional Ability Sorting, Education Choices, and Labor Market Outcomes

The process of skill specialization starts early in the life-cycle, with large differences in the labor market “premium” for different types of skills. Understanding the process of skill formation early in the life-cycle is therefore crucial for understanding later-life inequality. We study how early choices in secondary school affect later college enrollment and major choices, as well as later life outcomes. We focus on high school tracking choices in 10th grade which greatly affect the depth and breath of the curriculum students are exposed to. We study how individuals sort into high school tracks and how these shape college major choices, college graduation, and labor market outcomes. We estimate a sequential choice model of education specialization at each stage using Swedish administrative data. Individuals sort into each stage based on prior specialization choices and three dimensions of ability: cognitive, interpersonal, and grit. We use the model to compare counterfactual policies targeting students at different stages and margins of choice. We show that policies targeting marginal individuals to enroll in STEM college degrees have limited effects as most fail to graduate. In contrast, inducing students into the STEM track in 10th grade both increases STEM college enrollment and graduation. Earlier investments in STEM skills thus lead to much larger benefits than later investments.