Scope Neglect

Scope Neglect in Willingness to Pay

The contingent valuation method (CVM) was developed by resource economists (see Mitchell & Carson, 1989) as a tool for assessing the value of public goods for purposes of litigation or cost–benefit analysis. Participants in contingent valuation (CV) surveys are asked to indicate their willingness to pay (WTP) for specified public goods, and their responses are used to estimate the total amount that the community would pay to obtain these goods. The economists who design contingent valuation surveys interpret WTP as a valid measure of economic value and assume that statements of WTP conform to the extensional logic of consumer theory. The relevant logic has been described by a critic of CVM (Diamond, 1996), who illustrates the conditional adding rule by the following example: in the absence of income effects, WTP for saving X birds should equal WTP for saving (X-k) birds, plus WTP to save k birds, where the last value is contingent on the costless prior provision of safety for (X-k) birds (Diamond, 1996).

Strict adherence to Bayes’ rule may be an excessively demanding standard for intuitive predictions; similarly, it would be too much to ask for WTP responses that strictly conform to the “add-up rule.” In both cases, however, it seems reasonable to expect some sensitivity to extension – to the base rate of outcomes in categorical prediction and to the scope of the good in WTP. In fact, several studies have documented nearly complete neglect of scope in CV surveys. The best-known demonstration of scope neglect is an experiment by Desvouges et al. (1993), who used the scenario of migratory birds that drown in oil ponds. The number of birds said to die each year was varied across groups. The WTP responses were completely insensitive to this variable, as the mean WTPs for saving 2,000, 20,000, or 200,000 birds were $80, $78, and $88, respectively.

A straightforward interpretation of this result involves the two acts of substitution that characterize prototype heuristics. The deaths of numerous birds are first represented by a prototypical instance, perhaps an image of a bird soaked in oil and drowning. The prototype automatically evokes an affective response, and the intensity of that emotion is then mapped onto the dollar scale – substituting the readily accessible heuristic attribute of affective intensity for the more complex target attribute of economic value. Other examples of radical insensitivity to scope lend themselves to a similar interpretation. Among others, Kahneman and Knetsch (see Kahneman, 1986) found that Toronto residents were willing to pay almost as much to clean up polluted lakes in a small region of Ontario as to clean up all the polluted lakes in Ontario, and McFadden and Leonard (1993) reported that residents in four western states were willing to pay only 28% more to protect 57 wilderness areas than to protect a single area (for more discussion of scope insensitivity, see Frederick & Fischhoff, 1998).

The similarity between WTP statements and categorical predictions is not limited to such demonstrations of almost complete extension neglect. The two responses also yield similar results when extension and prototype information are varied factorially within subjects. The account cites [Fig. 2.3A] for a study of WTP for programs that prevented different levels of damage to species of varying popularity (Ritov and Kahneman, unpublished observations; cited in Kahneman, Ritov & Schkade, 1999). As in the case of base-rate ([Fig. 2.3C]), extensional information (levels of damage) combines additively with nonextensional information. The authors argue that the plotted lines should fan out under a plausible theory of value. Neither of the referenced figures is included here.

Finally, the role of the emotion evoked by a prototypical instance was also examined directly in the same experiment, using a heuristic elicitation task. Some respondents were asked to imagine that they saw a television program documenting the effect of adverse ecological circumstances on individual members of different species. The respondents indicated, for each species, how much concern they expected to feel while watching such a documentary. The correlation between this measure of affect and willingness to pay, computed across species, was .97.