Research
Published Papers:
"Moral Perceptions of Advised Actions" (with Lucas Coffman)
Management Science, vol. 65 no. 8 (2019), 3904-3927
Can an organization avoid blame for an unpopular action when an adviser advises them to do it? We present experimental evidence suggesting this is the case -- advice to be selfish substantially decreases punishment of being selfish. Further, this result is true despite advisers' misaligned incentives, known to all: Through a relational contract incentive, advisers are motivated to tell the decision-makers what they want to hear. Through incentivized elicitations, we find suggestive evidence that advice moves punishment by affecting beliefs of how necessary the selfish action was. In follow-up treatments, however, we show advice does not decrease punishment solely through a beliefs channel. Advice not only changes beliefs of what happened, but also the perceived morality of it. Finally, in treatments where advisers are available, selfish decision-makers act more selfishly.
"Desirability and Information Processing: An Experimental Study"
Economics Letters, 152 (2017), 96-99
I study, in an experimental setting, how individuals process news regarding the likelihood of an event that is desirable, but not ego-relevant. I hypothesize that individuals biasedly favor good news over bad news, but find no support for this hypothesis.
Working Papers:
"The Effect of One-Sided Visibility on Adherence to Prosocial Norms Decreases with The Size of The Observed Group"
We hypothesize that the power of (one-sided) visibility to induce adherence to a norm depends on the size of the group that is being observed. Using a lab experiment, we find that participants are more likely to reciprocate trust when their choices and identities are observed along with the choices and identities of only a few, instead of many others. The effect of group size disappears when we repeat the experiment but remove trustees' pictures and only show their choices, which points towards an image-related mechanism. Additionally, when we provide the observed decision makers with feedback about others' behavior in their observed group and ask them to make a choice again, visibility has a larger effect for small observed groups than for large groups, controlling for the informational content of the feedback. Motivated by our main result, we show that a model where experienced embarrassment is dependent on the behavior of others in the observed group produces our main hypothesis if there is a sufficiently high proportion of unconditional transgressors in the population and an additional norm-transgressor has a larger impact on the psychological cost of being seen transgressing a norm than an additional norm-follower.
"Norms in Psychological Games"
Some individuals tend to avoid situations in which they are expected to give, use excuses to hide selfish behavior whenever they are available, and behave prosocially conditional on others behaving prosocially. These findings are inconsistent with the notion of a decision maker that has social preferences. We present a framework, based on psychological game theory, that reconciles the previous literature on social preferences with these findings. In our framework, decision makers do have a notion of what constitutes acceptable behavior in a given situation (injunctive norm) according to some values (e.g fairness, reciprocity), but they are not internally motivated to follow those norms. Instead, individuals are averse to be seen transgressing a norm.
"Obscuring Transgression and Obscuring Conformity" (with Christian Alcocer)
Compared to a situation with full transparency, decision makers are more likely to transgress pro-social norms when an audience cannot be certain whether a transgression-consistent outcome is due to the decision maker's volition or due to other factors. That is, obscuring transgressions decreases conformity. We replicate this result using a binary-choice Dictator Game where unequal outcomes can occur due to chance. We also study behavior in novel treatments where we obscure conformity: The audience is not able to determine whether an outcome consistent with the 50-50 norm is due to chance or the dictator's choice. Based on the idea that avoiding shame is the driver of audience effects in situations where decision makers do not expect to have further interactions with members of the audience, we hypothesize that obscuring conformity has no effect on the likelihood of conforming. We find evidence that supports our hypothesis.
In Progress:
"Rank Incentives for Teams and Gender Congruence: An Experimental Study"
"Self-Serving Selection of Causes"