The impact of information on cognitive valuation decisions with implications for the contingent valuation method

Abstract

Consumer valuation decisions are conceptualized as occurring in a six-stage cognitive valuation process. The six-stage model is advocated as an appropriate framework for economic analyses of consumer valuation decisions. The model applies to decisions made in all types of markets, including actual and experimental markets. A major contribution of the model is that it lays the groundwork for reducing the "black box" nature of the cognitive valuation process. Thus, the model is fruitful for developing refutable hypotheses concerning the impact of changes in market structure on consumer valuation behavior and, in turn, final revealed values for goods and commodities. The six-stage model was applied to analyze the effect of information on nonmarket good valuation decisions in contingent markets. This application is viewed as a special case of the more general problem of determining the impact of market information on values revealed by consumers for nondivisible, multidimensional goods. The contingent valuation method is used to elicit values for nondivisible, multidimensional nonmarket goods such as environmental quality and natural environments. In these valuation situations, the accuracy and reliability of contingent valuation results may be quite sensitive to information effects. An information effect is defined as a change in observable bidding behavior resulting from a change in market information structure. This definition encompasses both desirable and undesirable impacts of information on observable bidding behavior. The six-stage cognitive valuation process model was used to formally model contingent market information effects. A specific result of this modelling was the development of potential research hypotheses. Experimental economics methodology was extended to the design and conduct of contingent valuation experiments. Selected research hypotheses derived from the conceptual valuation process model were tested in laboratory-type and field contingent valuation experiments. The experimental results suggest that the accuracy and reliability of contingent valuation results are sensitive to information type, quantity, complexity, and display. Thus, careful consideration should be given to these variables when designing contingent markets. A particular need is to develop a set of contingent market information structure guidelines which are based on rigorous conceptual modelling and hypothesis testing in controlled experiments. The conceptual and empirical results of this thesis are used to propose an initial set of such guidelines.