Downward Hypothetical Bias in the Willingness to Accept Measure for Private Goods: Evidence from a Field Experiment

This paper documents a downward hypothetical bias in the willingness to accept measure for a private good using the Becker‐DeGroot‐Marschak (BDM) method. We first construct a theoretical framework that predicts an upward hypothetical bias for the willingness to pay and a downward hypothetical bias for the willingness to accept measure. Based on a field experiment on livestock trading in China, we compare four experiment treatments, including a hypothetical treatment wherein responses are not consequential, two consequential treatments with probabilistic implementation wherein responses have a 25% or 75% probability of being consequential, and a consequential treatment with full implementation wherein responses are 100% consequential. Consistent with our theoretical predictions, the empirical results show that respondents report significant lower (about 12% lower) values in the hypothetical treatment compared with the consequential treatment with full implementation using the willingness to accept measure. The reported values also increase as the implementation probability increases. Our study contributes to the research on hypothetical bias in the willingness to accept by allowing exogenous changes in the consequential probability.