Alternative approaches to treat respondent uncertainty in contingent willingness to pay estimation A theoretical and empirical analysis

Purpose - The purpose of this paper is to theoretically using two new models to analyze the effect of respondents' uncertainty about their stated willingness to pay (WTP) on welfare estimates in the contingent valuation method (CVM) theoretically using two new models, and empirically to reveal consumers' WTP to improve drinking water supply safety (WSS) in China. Design/methodology/approach - In this paper, two alternative preference uncertainty treatment approaches are proposed to estimate consumers' WTP theoretically and they are applied to China's WSS improvement program from a payment card method, which depends on how consumers' certainty level about their valuation is. Furthermore, four regression models are presented to investigate the determinants of consumers' WTP. Findings - Theoretically, the alternative approaches that proposed in this research can remove overestimation bias from traditional CVM method but with lower estimation efficiency. In addition, the empirical results of the uncertainty adjusted models show that the expected WTP to improve drinking WSS is from 0.55 to 0.56 Renminbi yuan/ton, which are lower than the estimates from the conventional standard CVM models. Consumers' preferences for their concerns about WSS, attitudes toward WSS improvement programs, trusts in implement authorities and their knowledge of WSS have significant effects on the WTP for improving drinking WSS and on respondents' uncertainty too. Originality/value - Theoretically to the authors' knowledge, it is the first attempt to compare alternative approaches to treat respondent uncertainty using numerical certainty scale combined with payment card format valuation questions in CVM. Empirically it is the first study at this large scale that investigates consumers' WTP for improving drinking WSS incorporating with respondent uncertainty in China. In addition, to assess consumer preferences for improved drinking water safety and the sources of uncertainty, information on consumers' attitudes toward WSS are considered at the first time.