Delayed Honesty

How to cite

Wenner, Lukas; Kölle, Felix; Bortolotti, Stefania. 2018. "Delayed Honesty." AEA RCT Registry. AEARCTR-0003594. https://doi.org/10.1257/rct.3594-3.3

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

Lukas Wenner* , Felix Kölle , Stefania Bortolotti

* Primary investigator

Registry ID
AEARCTR-0003594
DOI
https://doi.org/10.1257/rct.3594-3.3
First registered
Last updated
Status
Completed
Trial dates
2018-11-27 – 2020-06-30
Countries
United States of America ()
Registry record
http://www.socialscienceregistry.org/trials/3594

Abstract

Many economic interactions involve the release of private information to other agents. Often, there is a trade-off between truthful reporting, which may be morally desirable and/or socially optimal and lying, which may be individually optimal, i.e., payoff-maximizing. Rejecting the idea that economic agents are purely interested in their own monetary gains and without any ``moral compass'', a recent literature in economics has demonstrated a number of factors which influence an individual's degree of honesty, such as the possible gains from lying, the degree of social observability, and the ex-ante likelihood of different states of the world.

In this project, we investigate how the timing of when the costs and benefits of lying occur, affect people’s degree of dishonesty. In particular, we are interested in the following three questions: (i) How is lying affected when the benefits from misreporting accrue delayed rather than immediately? (ii) How does increased mental engagement with the lie affect misreporting when benefits are delayed? (iii) How does a delay between the time when private information is obtained and the time when people report it affect dishonesty? We study these questions in an online experiment conducted on Amazon Mechanical Turk (MTurk).

Experimental design

We study lying behavior in an experiment conducted on MTurk. Subjects report the outcome of two coin flips, flipped in private unobserved by the experimenters. The payment is based on their report. In six different treatments we vary the relative timing between coin flips, report, and payment.

Design details

Randomization method
done by the experimental software (Qualtrics)
Randomization unit
Participant
Clusters
1240 participants
Observations
1240 participants
Arms
200 participants in treatments 1 to 4, 220 participants in treatments 5 and 6.
Primary outcomes
Fraction of subjects who report "2 Heads"

Cited by

No citation data available for this registration yet.

Keywords