Measuring risk aversion to guide transportation policy: Contexts, incentives, and respondents

• Published in: Transportation research. Part A, Policy and practice Vol. 80; pp. 15 - 34
• Main Authors: Dixit, Vinayak V., Harb, Rami C., Martínez-Correa, Jimmy, Rutström, Elisabet E.
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01-10-2015
• Subjects: Congestion pricing; Contextual tasks; Driving simulations; Economics; Experiments; Metropolitan areas; Policies; Pricing; Risk; Risk attitudes; Route selection; Tasks; Transportation; Risk attitudes; Congestion pricing; Contextual tasks; Experiments; Driving simulations
• ISSN: 0965-8564; 1879-2375; 1879-2375
• DOI: 10.1016/j.tra.2015.07.002

•We use Experimental Economics methods to elicit risk attitudes.•We compare impact of context on risk attitudes: lottery vs. driving simulators.•We compare impact of respondents on risk attitudes: students vs. field participants.•Across contexts and respondents the risk taking behavior converge with experience.•Experimental Economics, Stated and Revealed Preferences methods are complementary.
Road pricing may provide a solution to increasing traffic congestion in metropolitan areas. Route, departure time and travel mode choices depend on risk attitudes as commuters perceive the options as having uncertain effects on travel times. We propose that Experimental Economics methods can deliver data that uses real consequences and where the context can be varied by the researcher. The approach relies on the controlled manipulation of contexts, similar to what is done in the Stated Choice approach, but builds in actual consequences, similar to the Revealed Preference approach. This paper investigates some of the trade-offs between the cost of conducting Experimental Economics studies and the behavioral responses elicited. In particular, we compare responses to traditional lottery choice tasks to the route choice tasks in simulated driving environments. We also compare students (a low cost convenient participant pool) to field participants recruited from the driving population. While we see initial differences across our treatment groups, we find that their risk taking behavior converge with minimal repetition.