The lifetime earnings premia of different majors: Correcting for selection based on cognitive, noncognitive, and unobserved factors

This paper constructs a simulation approach to estimate the lifetime returns to various college majors. I use data from the 1979 cohort of the National Longitudinal Survey of Youth and American Community Survey to estimate the parameters which form the backbone of the simulation. I address selection into both higher education and specific major categories using measures of cognitive and noncognitive ability. Additionally, I present the lifetime premia under various assumptions regarding the magnitude of unobservable sorting. I find substantial heterogeneity in the returns to each educational outcome, ranging from $700,000 for Arts/Humanities majors to $1.5million for Science Technology Engineering or Math (STEM) graduates (each premium is relative to high school graduates with no college experience). The differentials are larger when search behavior (allowing for differential unemployment probabilities across majors) is taken into account. Finally, I estimate the major premia separately across three birth cohorts to account for the changing nature of selection into both college and majors over time. •This paper estimates lifetime earnings premia for various college majors.•Results are presented with and without a variety of selection corrections.•There is substantial heterogeneity in lifetime premia across the major categories.•Business and STEM fields have the highest returns, Arts/Humanities majors the lowest.•I find moderate convergence in the various major premia across birth cohorts.