Do Auctions Underestimate Consumer WTP? An Artefactual Field Experiment

Abstract

Auction experiments are commonly used to elicit consumer values for a wide range of items and services. These auctions are theoretically incentive compatible so are assumed to give an unbiased estimate of consumers’ willingness to pay (WTP). However, the vast majority of consumer decisions are made not in auctions but in posted-price settings, such as grocery stores. This study tests whether the two mechanisms yield similar WTP estimates by comparing WTP for honey from a second-price Vickrey auction and the WTP from a posted-price dichotomous-choice mechanism in a within-subject, homegrown-value setting. Results from 115 adult consumers indicate that estimates of WTP generated by an auction are approximately 50% smaller than WTP estimates generated by a posted-price mechanism. We test several potential explanations for this difference in behavior and find no evidence of anchoring or yea-saying effects. The evidence does suggest that the framing of choice in an auction format and a lack of familiarity with auctions are the most plausible explanation for this downward bias.