With Risk Comes Reward

How to cite

Siegenthaler, Simon; Nikiforakis, Nikos; Dariel, Aurelie. 2023. "With Risk Comes Reward." AEA RCT Registry. AEARCTR-0011326. https://doi.org/10.1257/rct.11326-1.2

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

Simon Siegenthaler* , Nikos Nikiforakis , Aurelie Dariel

* Primary investigator

Registry ID
AEARCTR-0011326
DOI
https://doi.org/10.1257/rct.11326-1.2
First registered
Last updated
Status
Completed
Trial dates
2023-05-04 – 2023-08-09
Countries
Denmark (), Finland (), France (), Iceland (), Netherlands (), Norway (), Sweden (), United States of America ()
Registry record
http://www.socialscienceregistry.org/trials/11326

Abstract

Pay inequality within companies is rising in modern economies. We design a lab and online experiment to examine team performance when leaders outearn their team members. We show that pay inequality causally undermines the ability of leaders to coordinate their teams (via leading by example), causing bad team performance despite aligned financial incentives. We then study whether exposing leaders to higher risk—measured by the cost they suffer when the team performs poorly—can justify inequality. We find that increased leader risk raises the perceived legitimacy of pay inequality. Remarkably, leaders anticipate the teams’ greater willingness to follow, which is necessary for teams to do well.

Experimental design

We consider 1 treatment without a leader and 5 treatments with a first-mover. In all treatments, the safe payoff is $4 for everyone. In all treatments, the two other group members have a LowPayoff of $2.75 and a HighPayoff of $4.75. The treatments vary in the incentives of the leader.

Change: After collecting these treatments, we decided to add one more treatment to obtain a more complete picture of the effect of leader risk. Specifically, we look also at the case when the leader earns no bonus but also has no risk.

Design details

Randomization method
Randomization done by the computer software: as the participants complete the experiment online, we will always assign the treatment with the lowest number of observations.
Randomization unit
Individual
Clusters
1400 individuals (independent observations)
Observations
1400 individuals
Arms
7 Treatments with about 100 subjects in the US. 7 Treatments with about 100 subjects in the Nordic Countries, or as much as possible given limited availability. Update: We have collected about half of the observations for the Nordic countries. Due to limited liquidity, we will now also include people from the Netherlands and France. The rationale is that these countries are geographically near the Nordic countries (EU), have Gini coefficients similar to the Nordic countries, and are sufficiently large so we are able to collect the remaining observations.
Primary outcomes
Percentage of orange choices (overall) Percentage of orange choices (conditional on the leader choosing orange) Relationship between risk aversion, inequality aversion (+expectations), conformity, and the percentage of orange choices Differences in these variables between US and Nordic Countries subjects

Cited by

No citation data available for this registration yet.

Keywords