Working Paper: NBER ID: w29752
Authors: Daniel J. Agness; Travis Baseler; Sylvain Chassang; Pascaline Dupas; Erik Snowberg
Abstract: People’s value for their own time is a key input in evaluating public policies: evaluations should account for time taken away from work or leisure as a result of policy. Using rich choice data collected from farming households in western Kenya, we show that households exhibit non-transitive preferences consistent with behavioral features such as loss aversion and self-serving bias. As a result, neither market wages nor standard valuation techniques (such as the Becker-DeGroot-Marschak—BDM—mechanism of Becker et al., 1964) correctly measure participants’ value of time. Using a structural model, we identify the mix of behavioral features driving our choice data. We find that these features distort choices when exchanging cash either for time or for goods. Our model estimates suggest that valuing the time of the self-employed at 60% of the market wage is a reasonable rule of thumb.
Keywords: Behavioral Economics; Self-Employment; Time Valuation; Labor Markets
JEL Codes: C93; D03; D61; J22; O12; Q12
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| average value of time (C41) | 60% of market wage (J31) |
| behavioral features (D91) | distort choices when exchanging cash for time or goods (E41) |
| self-serving bias (D91) | undervalue compensation received through negotiation (D46) |
| loss aversion (G41) | overvalue what they give up in transactions (D46) |
| direct measures of time value (C41) | align closely with market wages (J31) |
| indirect measures (C36) | yield only 40% of the market wage (J31) |
| traditional methods of valuing time (C41) | overstate benefits of technologies that increase time use (J29) |
| traditional methods of valuing time (C41) | understate benefits of technologies that save time (O49) |