Job market paper
I develop a dynamic model of education and labor decisions that seeks to explain student demand elasiticities to the grading standards in their courses, and the high rate of dropout from STEM majors. The positive correlation in the data between wages and terminal GPA suggests that firms offer a wage premium to students with higher grades. College students take this 'price' on grades into consideration when making education decisions: dropout or major switching can be induced by grade shocks. I estimate a structural model using the NLSY97, and find that a difference in grades by 1.0 GPA points can impact the net present value of a college degree by as much as 20% at the time of graduation. If STEM and non-STEM course grading standards were adjusted to be in line with each other, there would be a 11.0% increase in the total number of STEM major graduates, comparable effect to a $2,500 STEM tuition subsidy. Finally, a key observation of the model is that due to imperfect sorting across majors and occupations, large changes in the number of STEM degrees translate into smaller changes in the number of STEM workers.