Korean consumers' willingness to accept (WTA) for GM food are studied in this paper. This study compares hypothetical and nonhypothetical responses to choice experiment questions. We test for hypothetical bias in a choice experiment involving GM rice with differing characteristic attributes and multinomial logit model is applied to predict the estimated results. In general, hypothetical responses predicted higher probabilities of purchasing GM rice than nonhypothetical responses. Thus, hypothetical choices overestimate willingness to accept for GM rice. The results of this paper could contributes to government's GM food policies and subsequent studies, also improving economic welfare of farmers and consumers.
