Working Paper: NBER ID: w28353
Authors: Thomas Crossley; Yifan Gong; Todd R. Stinebrickner; Ralph Stinebrickner
Abstract: Unique longitudinal probabilistic expectations data from the Berea Panel Study, which cover both the college and early post-college periods, are used to examine young adults’ beliefs about their future incomes. We introduce a new measure of the ex post accuracy of beliefs, and two new approaches to testing whether, ex ante, agents exhibit Rational Expectations. We show that taking into account the additional information about higher moments of individual belief distributions contained in probabilistic expectations data is important for detecting types of violations of Rational Expectations that are not detectable by existing mean-based tests. Beliefs about future income are found to become more accurate as students progress through school and then enter the post-college period. Tests of Rational Expectations almost always reject for the in-school period, but the evidence against Rational Expectations is much weaker in the post-college period.
Keywords: Income Expectations; Rational Expectations; Probabilistic Expectations; Young Adults
JEL Codes: D84; J01
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Beliefs about future income (D15) | Ex post accuracy (C52) |
| While in college (I23) | Beliefs about future income (D15) |
| Rational expectations (D84) | Income beliefs (D83) |
| Mean-variance test and aggregated distribution test (C46) | Rational expectations (D84) |
| New tests (C12) | Rational expectations (D84) |
| Additional information captured by higher moments of belief distributions (C46) | Violations of rational expectations (D84) |