Before embarking on my argument I want to describe three practical objections to any attempt to use preference or the WTP associated with it as a basis of environmental valuation. I shall propose that even if preference had normative significance - even if there were some basis for believing it has a kind of normative content - economists cannot as a rule (1) identify preference or (2) measure WTP for it. Even if they could, (3) the benefits involved in measuring WTP for environmental public goods will not equal the costs.
First, preferences are not observable. They are enigmatic. People prefer and are willing to pay for a good because of some quality or qualities it possesses. Every object, however, possesses many different qualities. It is often hard to say what it is exactly - which quality - the individual wants or desires. In an earlier book, I used the example of Girl Scout cookies to illustrate this point. When a nine-year-old daughter of my neighbor asked me to purchase cookies I obliged but I am not sure why. I might have wanted or preferred (1) to support the Scouts; (2) to avoid friction with my neighbor; (3) to appear generous; (4) to spare my own daughter from embarrassment among her friends; (5) to do the right thing; (6) to feel a warm glow that I did the right thing; (6) to avoid guilt; or (7) any of a hundred other reasons. In any case, it was not the cookies; I took them to the office. Preference is not observed; it is constructed from a description of the choice a person makes.3 Different descriptions of the same behavior - from which different preferences may be construed or inferred - are often equally plausible. To interpret observed behavior as performing a choice, one must attribute a preference to the agent. One may then infer that preference from that interpretation. The circularity is obvious. I developed this argument in detail in an earlier book; therefore, I shall not belabor it here.4
Second, one cannot measure WTP for a representative or typical environmental public good. More than a century ago, Alfred Marshall wrote "that we cannot guess at all accurately how much of anything people would buy at prices very different from those they are accustomed to pay for it: or, in other words, what the demand prices would be for amounts very different from those which are commonly sold." Marshall understood the difficulty involved in attempts to project the "demand curve" from points hovering around customary prices. He called such attempts "liable to large error" and "highly conjectural except in the neighborhood of the customary price."5
Today economists have huge data sets on many commodities, such as petroleum, which may provide enough historical information about the relation of price to demand that different economists, using the same data, may project similar demand curves or derive the same range of WTP for those commodities. With respect to environmental public goods, however, there are no such data sets, so the situation may not differ from that which Marshall described. One way to parry this objection, however, would be to see whether economists working independently arrive at anything like the same measures of WTP for a given public good. Replication is vindication. Would these independent teams of economists, each using its own method and gathering its own data, project anything like the same demand curves? In the only instance of this sort of experimental replication I know of, the teams differed by an order of magnitude in their estimates of WTP for an environmental public good.6
Third, the costs the government sinks into any effort to measure WTP for a public good or to "price" any environmental "externality" are likely to outweigh the benefits - a lesson that Ronald Coase elegantly presented, that Duncan Kennedy elaborated, and that has been often repeated.7 Coase wrote that "the costs involved in governmental action make it desirable that the 'externality' should continue to exist and that no government intervention should be undertaken to eliminate it."8
One problem is this. If an agency tries to determine WTP for some good in which anyone has a significant interest - for example, because of damage awards in tort - the estimates themselves become objects of controversy, review, and litigation. At a conference on "contingent valuation" I attended, an environmental attorney who worked for the government, who was also a conferee, met with a number of prominent economists in this specialty. The economists chose among themselves which of them would appear as expert witnesses for the government (to defend a high estimate of WTP for the damaged environmental good) and which would appear for the defense, a petrochemical firm (to testify to a much lower figure). Estimates of WTP themselves become goods for which buyers compete. There is the possibility of rent-seeking, that is, paying to influence an agency decision. "You get the WTP you are willing to pay for," one knowledgeable lawyer said.
IS WTP RELATED TO BENEFIT?
Consider the sentence: "Something's economic benefit is determined by how much people are willing to pay for it."9 I believe this statement stipulates or postulates a definition of "economic benefit." It is not as if anyone has measured economic benefit and WTP separately and shown empirically a correlation between them. If person S chooses or prefers outcome a over b, for whatever reason, does it follow that S obtains (or expects to obtain) more utility from a than b? How would one test an answer to this question?
If utility were observable, then one could measure it empirically to see whether it correlates with WTP or with the satisfaction of preference. Since it is not observable, whether it equals or correlates with anything is anyone's guess. One's person's opinion would seem to be as good as another's.
Why do economists believe that preference or the associated WTP has a meaningful relation to economic value or benefit? The literature sometimes suggests that economic benefit and WTP are related because people choose or are willing to pay for things only insofar as they believe those things benefit them. One finds in the literature the proposal that a meaningful relation holds between WTP and benefit or welfare because "each individual is the best judge of how well off he or she is in a given situation."10 In A Primer for Policy Analysis, Edith Stokey and Richard Zeckhauser, for example, state that "individual welfare is all that matters in policy choices." The preferences of each individual, these authors say, constitute the criterion by which we are to judge his or her well-being. "In the United States, we usually take the position that it is the individual's own preferences that count, that he is the best judge of his own welfare."11
Even if the individual "is the best judge of his own welfare," however, this does not mean that that individual is a better or worse judge of the issues that may concern him or her, namely, whether an outcome is good in general, good from the perspective of the community, or consistent with its standards and goals. When a person judges a social policy or outcome, he or she may take and should take the perspective of the community - to ask what is best not for me but for us. Judgment of this sort, basic to political life, rests on argument and evidence; it may not turn on how the individual thinks an outcome will affect her or him. To be sure, people who believe in some political cause or candidate may contribute to a campaign; they may do so, however, to support that cause or candidate not necessarily to enhance their own well-being. According to many studies, people base their preferences with respect to social outcomes on what they believe is good for society or right in principle, not just (or even primarily) on a consideration of what benefits them personally.12 The individual may be the best judge of what is good for her or him - but that kind of judgment may not always affect choices the person makes. As Kenneth Arrow has written, "the individual orders all social states by whatever standards he deems relevant."13
In a series of influential papers, Amartya Sen discusses "the reasons that may lead a person to have different goals from what she sees as her own welfare, or to choose behavior and conduct that go beyond pursuing her own goals." In such cases, "a person's choice behavior may be constrained or influenced by... rules of conduct (for reasons that Immanuel Kant and Adam Smith described so well)."14 Sen distinguishes between "sympathy" and "commitment." Sympathy occurs when one person's welfare is affected by the position of others. "One way of defining commitment is in terms of a person choosing an act that he believes will yield a lower level of personal welfare to him than an alternative that is also available to him."15
According to Sen, "commitment then involves choosing an action that yields a lower expected welfare than an alternative available action."16 Commitment, Sen concludes, destroys the assumption "that a chosen alternative must be better than (or at least as good as) the others for the person choosing it."17 People who act on moral commitment - who act morally - act on beliefs about what is good in general or appropriate in the circumstances rather than on a view of what is advantageous for them. This "drives a wedge between personal choice and personal welfare, and much of traditional economic theory relies on the identity of the two."18
When Stokey and Zeckhauser state that "individual welfare is all that matters in policy choices" they rule out all the other standards the individual may deem relevant. They allow the person to choose only one thing - her own well-being - or her choices would not count. If people are willing to pay for policy outcomes for reasons other than their own well-being, how does that WTP connect to welfare or utility? How can economists explain the relation between WTP and value or between WTP and benefit?
The pervasiveness of "commitment" or "disinterested" values places welfare economists in a dilemma. They may include them in or exclude them from the cost-benefit valuations they prepare. To include them is to confront at least two difficulties. First, as Matthew Adler and Eric Posner have correctly observed, it would be odd to construe a person's disinterested judgments and preferences as part of his or her welfare function. For example, if a person approves the preservation of the endangered Sri Lankan squirrel on moral or religious grounds, it hardly follows that its protection makes that person better off in any meaningful sense.19 Second, if political beliefs, aesthetic judgments, and ethical commitments are only as good as the WTP for them, this rule would have to apply equally to economists and noneconomists. Economists typically prefer that society maximize utility. Why would this view be privileged, that is, be accepted on its merits? It should be "valued" in terms of the WTP of those who favor or propose it - the same as any other policy view or preference.
On the other hand, to exclude disinterested beliefs or judgments from cost-benefit valuation is also problematical. First, one has to distinguish ideal-regarding from self-regarding preferences. Who knows whether a person who buys Girl Scout cookies does so for self-interested reasons (for the cookies) or for disinterested reasons (for the cause)? Second, to provide a normative analysis of environmental policy, which is preponderantly ideological, without including disinterested values is like trying to stage Hamlet without the Prince of Denmark. The principal character is missing. Cost-benefit analysis can neither include disinterested values (to be a measure of welfare) or exclude them (to be relevant).
Was this article helpful?