Working Paper: NBER ID: w25485
Authors: Yifan Gong; Ralph Stinebrickner; Todd R. Stinebrickner
Abstract: We use unique data from the Berea Panel Study to characterize how much earnings uncertainty is present for students at college entrance and how quickly this uncertainty is resolved. We characterize uncertainty using survey questions that elicit the entire distribution describing one’s beliefs about future earnings. Taking advantage of the longitudinal nature of the expectations data, we find that roughly two-thirds of the income uncertainty present at the time of entrance remains at the end of college. Taking advantage of a variety of additional survey questions, we provide evidence about how the resolution of income uncertainty is influenced by factors such as college GPA and college major, and also examine why much income uncertainty remains unresolved at the end of college. This paper also contributes to a literature interested in understanding the relative importance of uncertainty and heterogeneity in determining observed earnings distributions.
Keywords: income uncertainty; college decisions; earnings expectations
JEL Codes: I21; J01; J3
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| college GPA (D29) | initial income uncertainty (D89) |
| college major (M39) | initial income uncertainty (D89) |
| initial income uncertainty (D89) | unresolved income uncertainty at graduation (D89) |
| what students learn about academic ability (I23) | resolution of uncertainty during college (I23) |
| major choice (M49) | resolution of uncertainty during college (I23) |
| labor market frictions (J29) | unresolved income uncertainty at graduation (D89) |
| personal job preferences (J29) | unresolved income uncertainty at graduation (D89) |
| search frictions (F12) | unresolved income uncertainty at graduation (D89) |