x

We are happy to announce that all trial registrations will now be issued DOIs (digital object identifiers). For more information, see here.
Risk Attitudes and Other-Regarding Preferences: An Experimental Study
Last registered on June 29, 2019

Pre-Trial

Trial Information
General Information
Title
Risk Attitudes and Other-Regarding Preferences: An Experimental Study
RCT ID
AEARCTR-0004309
Initial registration date
June 12, 2019
Last updated
June 29, 2019 3:17 PM EDT
Location(s)

This section is unavailable to the public. Use the button below to request access to this information.

Request Information
Primary Investigator
Affiliation
UCSD
Other Primary Investigator(s)
PI Affiliation
UCSC
Additional Trial Information
Status
On going
Start date
2019-05-28
End date
2021-08-31
Secondary IDs
Abstract
We study theoretically and empirically the interaction between risk attitudes and other-regarding preferences. An integrated model of risk attitudes and other-regarding preferences that extends the standard notion of inequity discount to lotteries was proposed in Lopez-Vargas (2014). In this model, a decision maker perceives inequity partly by comparing the marginal risks she and others face. The model predicts that fairness considerations will alter risk attitudes, in particular, a higher tolerance to positively correlated (fair) risks compared to negatively correlated (unfair) risks. Further, the model can explain the behavior by which people help others probabilistically (known as ex-ante fairness). Moreover, in contrast with the existing view of ex-ante fairness based on expected outcomes, the model does not imply that stronger ex-ante fairness behavior is associated with less risk sensitivity. We will study these predictions with evidence from an experiment. In a pilot, the primary findings were that subjects take more risks when outcomes are ex-post fair compared to when they are ex-post unfair. A second set of findings was that stronger ex-ante fairness behavior correlates with less sensitivity to risk. The findings would reject the implications of existing models. To generalizes, extend these prior findings, and test these models based on their predictions; we use a new individual-choice lab experiment.
External Link(s)
Registration Citation
Citation
Feldman, Paul and Kristian Lopez-Vargas. 2019. "Risk Attitudes and Other-Regarding Preferences: An Experimental Study." AEA RCT Registry. June 29. https://doi.org/10.1257/rct.4309-1.0.
Former Citation
Feldman, Paul, Paul Feldman and Kristian Lopez-Vargas. 2019. "Risk Attitudes and Other-Regarding Preferences: An Experimental Study." AEA RCT Registry. June 29. https://www.socialscienceregistry.org/trials/4309/history/48919.
Experimental Details
Interventions
Intervention(s)
This is a lab experiment on individual choice. The design is primarily within-subject, but between-subject comparisons were also considered. The experiment was run at UCSD's Economics laboratory. The subject pool consisted of undergrads, mostly, and all participants were over 18 years old.
Intervention Start Date
2019-05-28
Intervention End Date
2019-06-04
Primary Outcomes
Primary Outcomes (end points)
Risk and Other-regarding individual-level preferences.
Primary Outcomes (explanation)
Secondary Outcomes
Secondary Outcomes (end points)
Changes in risk preferences between different portfolio choice tasks and the correlation between risk attitudes and a preference for ex-ante fairness.
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
We elicited risk and social preferences using 4 types of convex-budget tasks to elicit risk and social preferences.
Experimental Design Details
Not available
Randomization Method
Resolution of uncertainty and whether subjects were `deciders' or `partners', only choices from the former got implemented and affected the latter, were done by the computer program.
Randomization Unit
Randomization is at the individual level. We also varied the task-type order, at the session level.
Was the treatment clustered?
No
Experiment Characteristics
Sample size: planned number of clusters
166 subjects
Sample size: planned number of observations
168 subjects, +/- attrition.
Sample size (or number of clusters) by treatment arms
4 sessions with one order and 3 types with another order. Each session has up to 24 subjects.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
For differences in behavior across portfolio allocations, it is 1 token (33 cents). The higher standard errors are around 3 tokens and the pilot study had 55 subjects. For the correlation test, we use a correlation of .2 between the parameters that measure risk and social preferences. All tests assume a 5% significance level and 80% for the power of the test.
IRB
INSTITUTIONAL REVIEW BOARDS (IRBs)
IRB Name
UCSC OFFICE OF RESEARCH COMPLIANCE ADMINISTRATION
IRB Approval Date
2016-04-05
IRB Approval Number
UCSC IRB Protocol # 2613
Analysis Plan

There are documents in this trial unavailable to the public. Use the button below to request access to this information.

Request Information