Document Type
Honors Project - Open Access
Abstract
Worker effort supply is influenced not only by rational expectations but also biases such as framing and numerical presentation of incentives. Identifying these effects is essential for designing effective incentives. To study left-digit manipulation, loss aversion, and loss aversion prediction, I conducted an experiment in an online labor market. Participants completed a real-effort typing task, a willingness- to-pay (WTP) task, and a self-reported loss aversion survey. Loss framing and left-digit manipulation had no statistically significant effects on effort, with no interaction effect, suggesting their influence on effort-supply may be overstated or highly heterogeneous. Despite null treatment effects, behaviorally elicited loss aversion aligned with theory on average (λ = 2.41), but showed substantial between-subject variation. Importantly, self-reported loss aversion had a near-zero correlation with observed loss aversion, indicating a disconnect between perceived and realized attitudes towards losses. As a result, there was no evidence of loss aversion prediction as an indicator of higher task valuation. In fact, participants who overestimated their loss aversion made suboptimal WTP decisions. Finally, exploratory analyses revealed gender-based differences in the self-perception of loss aversion. In whole, this research highlights the complex array of factors that influence effort-supply heterogeneously by individual. It also demonstrates difficulties with allowing individuals to self-select incentive structures, due to the instability of stated preferences compared to true behavior.
Recommended Citation
Lemov, Maia, "Losses and Labor: Experimental Insights into Worker Effort Supply and Behavioral Biases" (2026). Economics Honors Projects. 134.
https://digitalcommons.macalester.edu/economics_honors_projects/134
Included in
© Copyright is owned by author of this document
