Exploring the discrepancy between willingness-to-pay and willingness-to-accept with the Bayesian Truth Serum

The willingness-to-pay and willingness-to-accept divergence has been explored for the last 40 years. Researchers have investigated both the explanations for the discrepancy as well as possible ways of reducing it. One research direction of decreasing the WTA/WTP ratio is by implementing incentive-compatible methods within the experimental design. This study tests a relatively new incentive-compatible method, the Bayesian truth serum, over the traditionally employed non-incentivized introspection in an attempt to eliminate the disparity between WTA and WTP for a new product provided by Uber. BTS rewards truth telling by assigning each answer a truth score, given respondent’s endorsement and a prediction of the distribution of others’ answers. For the chosen research context, an altered BTS version is proposed and tested. The results indicate that the method does eliminate the discrepancy, however, removing outliers chance the result’s direction significantly suggesting that BTS is inferior. Discussed are future research directions for improvements of the BTS method and implications.