Alternatives to the Neo‐Classical Theory of Choice

Abstract

Abstract The concepts of accuracy and bias in surveys, like that of elicitation of preferences, presuppose that there are true answers to the questions that contingent valuation (CV) studies address. This chapter reviews a body of research that raises doubts about that presupposition; it focuses on two issues in CV methodology: divergences between willingness‐to‐pay and willingness‐to‐accept valuations, and the sensitivity of reported valuations to the elicitation method. Section 6.1 discusses the status of Hicksian consumer theory, and Sects, 6.2 and 6.3 investigate how far the valuations elicited by CV surveys might be expected to differ between willingness‐to‐pay and willingness‐to‐accept formats, and between different elicitation methods if respondents had Hicksian preferences and if survey instruments were unbiased. Section 6.4 considers the possibility that, although respondents do have Hicksian preferences, certain CV survey designs are systematically biased, and Sects. 6.5 and 6.6 look at some non‐Hicksian theories of decision‐making, and draw out their implications for the valuations generated by CV surveys. Sects. 6.7 and 6.8 review the evidence generated by CV and experimental research, and ask whether this can best be explained by Hicksian theory alone, by Hicksian theory combined with hypotheses about bias, or by non‐Hicksian theories; Sect. 6.9 draws some tentative conclusions.