Working Paper: NBER ID: w23188
Authors: Mariana Carrera; Heather Royer; Mark F. Stehr; Justin R. Sydnor
Abstract: The use of incentives to encourage healthy behaviors is increasingly widespread, but we have little evidence about how best to structure these programs. We explore how different incentive designs affect behavior on the extensive and intensive margins through an experiment offering incentives to employees of a Fortune 500 company to use their workplace gym. Overall the likelihood of joining the gym was not strongly affected by the incentive design. Notably, front-loading incentives to encourage initial participation was not more effective than an incentive kept constant over time. For those who were already at least occasional users of the gym, however, we find more evidence that the design of incentives matters. For this group, front-loading incentives appears to be detrimental relative to a constant incentive, but a novel design that spreads out the incentive budget by turning incentives on and off over a longer period of time is effective.
Keywords: health incentives; field experiment; exercise behavior; workplace wellness programs
JEL Codes: D03; I10; J30
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| incentive design (M52) | gym participation (extensive margin) (Z29) |
| frontloaded kickstart program (M13) | gym participation (extensive margin) (Z29) |
| kickstart incentives (O31) | gym attendance (Z29) |
| constant incentives (J33) | gym attendance (Z29) |
| kickstart group (L26) | gym attendance (post-incentive) (Z29) |
| extended-sporadic design (C90) | gym attendance (Z29) |
| timing of incentives (M52) | behavior (C92) |
| length and structure of incentive program (M52) | gym use (Z29) |