Article
Customer Protection

Consumers’ Choice Formulation under Risk: A Competence-Based Perspective

Date: 2013
Author: Dong-Jun Min, Marcus Cunha Jr.
Contributor: eb™ Research Team

One of the most general approaches to studying consumer choice under risk is based on the assumption that consumer are rational value maximizers (i.e., people prefer larger over smaller expected returns) and prefer smaller risks over larger risks (Pratt 1964). A standard assumption in this literature is that risk is negatively associated with the attractiveness of an option (March and Shapira 1987). One way via which individuals can reduce uncertainty associated with the expected outcome is by acquiring information that can help them to make more precise estimates of the product value (Aqueveque 2006; Conchar et al. 2004; Erdem and Swait 2004; Peterson and Merino 2003; Taylor 1974). Based on this line of reasoning, consumers’ choice formulation under risk should be based on information that provides a clear ranking-based standard for product evaluation (e.g., a camera’s resolution; a vertical product attribute) rather than on information that does not affect the actual performance of the product (i.e., a camera’s design, a horizontal product attribute). We offer an alternative account of choice under risk which predicts that consumers facing trade-offs between horizontal and vertical attributes shift their focus from vertical to horizontal product attribute information when the purchase involves higher levels of risk. Drawing upon research on the ambiguity effect (Ellsberg 1961; Heath and Tversky 1991; Klein et al. 2010), we show that consumers favorable perceptions of their own competence may lead consumers facing higher levels of risk to more positively value an attribute that reflects their own preference (one that they are knowledgeable and feel competent about) but that does not necessarily improve product performance than an attribute that unambiguously indicates product performance. We test these competing accounts of choice under risk in 5 experiments.