Reference-dependent preferences under shock conditions: evidence from the lab

How to cite

Presse, Adrian; Nöth, Markus. 2022. "Reference-dependent preferences under shock conditions: evidence from the lab." AEA RCT Registry. AEARCTR-0009525. https://doi.org/10.1257/rct.9525-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Adrian Presse* , Markus Nöth

* Primary investigator

Registry ID
AEARCTR-0009525
DOI
https://doi.org/10.1257/rct.9525-1.0
First registered
Last updated
Status
In development
Trial dates
2022-06-03 – 2022-07-14
Countries
Private
JEL codes
D12, D81, D84, J22
Registry record
http://www.socialscienceregistry.org/trials/9525

Abstract

In this study, we develop and extent the real-effort experimental design proposed by Abeler et al. (2011) to test reference-dependent preferences under dynamically changing expectations. To that end, we manipulate expectations via positive earnings shocks and analyze how this affects effort provision. We further shed light heterogeneity among participants with a focus on gender-specific differences and gamification factors influencing the perception of work effort.
Our study helps to reconcile contradicting empirical evidence from the field, specifically in the cab driver market (Camerer et al. (1997), Crawford and Meng (2011) and Farber (2005, 2008, 2015)).

Intervention

Our intervention shifts the expectations of lab participants regarding their earnings. Individuals have to solve randomly generated counting tasks (work effort) and are free to quit anytime. The final payout is determined by a 50-50 lottery: piece rate or fixed payment. Fixed payment amounts vary across treatments.

Experimental design

The experiment consists of five phases. First, subjects answer socio-demographic questions (gender, age) and make four simple lottery choices to elicit risk preferences. Second, the experimental task is introduced and control questions test subjects’ understanding of the task. Third, subjects perform a real effort task akin to Abeler et al. (2011). Subjects receive for each correct answer a piece rate and are free to quit anytime. The final payout is determined by a computerized 50-50 lottery: subjects either receive their accumulated piece rate or a fixed amount communicated at the beginning of the experiment. Fixed amounts vary (dynamically) across treatments. Fourth, the payout lottery determines the final payout after subjects quit the main task (or exceed task or time limits). Fifth, subjects answer a questionnaire on perceived task difficulty (1-10), work vs. game perception (1-5) as well as two qualitative questions on counting strategies and what factors led to the quitting decision.

Design details

Randomization method
done by computer
Randomization unit
individual
Clusters
120-240 participants
Observations
120-240 participants
Arms
30-60 each
Primary outcomes
work effort

Cited by

No citation data available for this registration yet.

Keywords