« Back to Results

Stochastic Choice and Experiments on Decision Making

Paper Session

Sunday, Jan. 5, 2020 10:15 AM - 12:15 PM (PDT)

Marriott Marquis, Del Mar
Hosted By: Econometric Society
  • Chair: Yoram Halevy, University of Toronto

News We Like to Share: How News Sharing on Social Networks Influences Voting Outcomes

Kirill Pogorelskiy
,
University of Warwick
Matthew Shum
,
California Institute of Technology

Abstract

More voters than ever get political news from their friends on social media platforms. Is this bad for democracy? Using context-neutral laboratory experiments, we find that biased (mis)information shared on social networks affects the quality of collective decisions relatively more than does segregation by political preferences on social media. Two features of subject behavior underlie this finding: 1) they share news signals selectively, revealing signals favorable to their candidates more often than unfavorable signals; 2) they naively take signals at face value and account for neither the selection in the shared signals nor the differential informativeness of news signals across different sources.

Audits as Evidence: Experiments, Ensembles, and Enforcement

Patrick Kline
,
University of California-Berkeley
Christopher Walters
,
University of California-Berkeley

Abstract

We study the ability of correspondence studies utilizing fictitious applicants to detect illegal discrimination by individual employers. Employers violate US employment law if their propensity to call applicants back depends on protected applicant characteristics such as age, race, or sex. We establish identification of higher moments of the effects of protected characteristics on callback rates as a function of the number of fictitious applications sent to each job ad. These moments are used to bound the fraction of jobs that are illegally discriminating. Applying our results to three experimental datasets, we find evidence of significant employer heterogeneity in discriminatory behavior, with the standard deviation of gaps in job specific callback probabilities across protected groups averaging roughly twice the mean gap. In two experiments manipulating racially distinctive names, we estimate that at least 70% of the jobs that call back both of two white applications and neither of two black applications engaged in illegal racial discrimination. To assess more carefully the tradeoff between type I and II errors presented by these behavioral patterns, we consider the performance of a series of “auditing rules” for investigating suspicious callback behavior under a simple two-type model that rationalizes the experimental data. Though, in our preferred specification, only 17% of employers are estimated to discriminate on the basis of race, we find that an experiment sending 10 applications to each job would enable accurate detection of 7.5% of discriminators while falsely accusing fewer than 0.2% of non-discriminators. An experiment employing an optimally-chosen application portfolio of races and other resume characteristics would boost the detection rate to roughly 10% without increasing false accusations. Our results suggest illegal labor market discrimination can be reliably monitored at relatively low cost.

Decision Making under Uncertainty: An Experimental Study in Market Settings

Federico Echenique
,
California Institute of Technology
Taisuke Imai
,
Ludwig Maximilian University of Munich
Kota Saito
,
California Institute of Technology

Abstract

We design and implement a novel experimental test of subjective expected utility theory and its generalizations. Our experiments are implemented in the laboratory with a student population and pushed out through a large-scale panel to a general sample of the US population. We find that a majority of subjects' choices are consistent with the maximization of some utility function, but not with subjective utility theory. The theory is tested by gauging how subjects respond to price changes. A majority of subjects respond to price changes in the direction predicted by the theory, but not to a degree that makes them fully consistent with subjective expected utility. Surprisingly, maxmin expected utility adds no explanatory power to subjective expected utility. Our findings remain the same regardless of whether we look at laboratory data or the panel survey, even though the two subject populations are very different. The degree of violations of subjective expected utility theory is not affected by age nor cognitive ability, but it is correlated with financial literacy.

Repeated Choice: A Theory of Stochastic Intertemporal Preferences

Jay Lu
,
University of California-Los Angeles
Kota Saito
,
California Institute of Technology

Abstract

We provide a repeated-choice foundation for stochastic choice. An agent chooses from a menu repeatedly over time, generating a time series of choices. We identify the limit frequency of these choices as stochastic choice. We characterize a tractable model of stochastic intertemporal preferences where the agent repeatedly chooses today’s consumption and tomorrow’s continuation menu, aware that future preferences will evolve according to a subjective ergodic utility process. Using our model, we demonstrate how not taking into account the agent’s preference for early (late) resolution of uncertainty would lead an analyst to underestimate (resp., overestimate) the agent’s risk aversion. Estimation of preferences can be performed by the analyst without explicitly modeling continuation problems (i.e. stochastic choice is independent of continuation menus) if and only if the utility process takes on the standard additive and separable form. Applications include estimation under dynamic discrete choice.

Law of Demand and Stochastic Choice

Simone Cerreia-Vioglio
,
Bocconi University
Fabio Angelo Maccheroni
,
Bocconi University
Massimo Marinacci
,
Bocconi University
Aldo Rustichini
,
University of Minnesota

Abstract

We characterize consistent random choice rules in terms of the optimality of their support. We then study stochastic choice in a consumer theory setting. We prove a law of individual demand for stochastic choice that generalizes the standard deterministic law.

Hard-to-Interpret Signals

Larry Epstein
,
Boston University
Yoram Halevy
,
University of Toronto

Abstract

Decisions under uncertainty are often made with information that is difficult to interpret because multiple interpretations are possible. Individuals may perceive and handle uncertainty about interpretation differently and in ways that are not directly observable to a modeler. This paper identifies and experimentally examines behavior that can be interpreted as reflecting an individual's attitude towards such uncertainty.
JEL Classifications
  • D8 - Information, Knowledge, and Uncertainty
  • D0 - General