LAUSR

Abstract China has experienced a series of high-profile food safety scandals in the past few years that seriously challenged public confidence in the domestic food industry. Much attention has been paid to Chinese government's food regulatory and inspection systems. Scant research, however, has been devoted to analyzing Chinese consumers' food safety concerns. This study interviewed 2092 Chinese consumers in Beijing, Shanghai, Guangzhou, Xi'an, Jinan, and Harbin and used the Conditional Logit model, Mixed Logit model, and the Latent Class model to analyze consumer preferences and marginal willingness to pay (WTP) for selected food safety attributes of Fuji apple products. We identified three consumer segments: certification-oriented (65.9%), price and origin-oriented (19.1%), and not interested (15.0%). Results reveal that Chinese consumers, in general, are willing to pay a premium for selected food safety attributes. Consumers' perceptual and attitudinal factors and socio-demographic characteristics are used to determine the sources of preference heterogeneity. Marginal analysis is also conducted to estimate the response of the model and selectione'd probability to potential policy levels, such as increasing the perception of food safety and the trust on labeling and traceability information to improve the evaluation of food safety.