Working Paper: NBER ID: w29579
Authors: Sara B. Heller; Judd B. Kessler
Abstract: This paper demonstrates that information frictions limit the labor market trajectories of young people in the U.S. We provide credible skill signals—recommendation letters based on supervisor feedback—to a random subset of 43,409 participants in New York City’s summer jobs program. Letters increase employment the following year by 3 percentage points (4.5 percent). Earnings effects grow over 4 years to a cumulative $1,349 (4.9 percent). We find no evidence of increased job search or confidence; instead, the signals help employers better identify successful matches with high-productivity workers. But the additional work hampers on-time high school graduation, especially among low-achieving students.
Keywords: Information Frictions; Skill Signaling; Youth Labor Market; Recommendation Letters
JEL Codes: C93; I21; J2; J48
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Recommendation letters (Y30) | Employment (J68) |
| Recommendation letters (Y30) | Earnings (J31) |
| Recommendation letters (Y30) | Employer perceptions (J63) |
| Recommendation letters (Y30) | Job search confidence (J68) |
| Recommendation letters (Y30) | On-time high school graduation rates (I21) |