How important is selection into full-time and part-time employment? A new panel data sample selection model for estimating wage profiles

The literature has shown that correcting for self-selection into work is important for the estimation of wage profiles. In this article, we analyze to what extent intensive labor supply choices add valuable otherwise unobserved information to improve wage profile estimates. We develop a panel data sample selection model that allows for discrete choices in labor supply decisions and apply this to high-quality administrative data. Compared to labor supply decisions at the extensive margin, our new approach is able to control for additional unobserved heterogeneity from intensive labor supply choices with important consequences for the existence and direction of selection into (part-time) work. Applied to the data, we find that such information is especially important for estimating part-time wage profiles for women.