ABSTRACT
In spite of the problems discussed above, many economists interpret aggregate net benefits as a measure of social welfare change – choosing the simple approach of disregarding welfare weights altogether. Although understandable, considering how little we know about these weights, this procedure implies a certain arbitrariness. Thus, it would be interesting to perform some sort of robustness check, providing at least a rough hint of the possible importance of that arbitrariness. In this chapter, I will present an alternative method for measurement and aggregation of individual benefits, due to Brekke (1997). While this method is equally simple and equally theoretically plausible, it yields quite different results from standard CBA. I do not intend to promote this alternative method as the “correct” one, but the fact that its results are so different illustrates, I believe, that there is no such thing as a “true” or “economically correct” method for measurement and comparison of individual utility changes. We simply do not know how to do this: any proposed measurement method, including monetary valuation and CBA, relies heavily on subjective, non-verifiable assumptions. The focus of the present chapter is on utility comparisons, not on moral philosophical views about the good society. To simplify the discussion, let me thus concentrate on the former by assuming, for the moment, that every decision-maker accepts a social welfare function of the unweighted utilitarian type (i.e., W = U1 + . . . + Un). The purely normative part of the welfare weight (V ′i) is thus equal for all and can be disregarded. That does not imply that we can disregard welfare weights in CBA, since we must still judge how to compare net benefits between persons: Even if each person’s utility should count equally, we must also “translate” money into utility. In standard CBA, the aggregate net benefits associated with a project is measured by first calculating individual net benefits in monetary units, then adding these to yield a measure of social welfare.1 Alternatively, we
could in fact measure individual net benefits in units of the environmental good, and then add these to yield an alternative measure of social welfare change. Individual net benefits in environmental units can be measured by asking the individual how large an improvement in environmental quality he would require to be willing to pay a certain cost. For example one may ask: “How many additional square meters of protected forest would you at least require to be willing to pay x dollars per year?”. Here, x would be the amount each person would have to pay if the project were implemented. If the respondent requires an environmental improvement bigger than the improvement actually associated with the project, the respondent’s net benefit is negative. This procedure is based on exactly the same reasoning as explained in Chapter 5; the difference is that we use another numeraire. As long as we rank projects only for a single individual, this method and standard monetary valuation will yield exactly the same result. If we sum individual net benefits to yield a social ranking, however, Brekke (1997) showed that the numeraire does matter.2 The reason is that with unweighted aggregation, we implicitly assume that the numeraire in use is equally much worth to everybody. That is, if we use the standard CBA procedure, we assume that money is worth equally much to everyone on the margin; if we measure individual net benefits in environmental units we assume, instead, that the environment is worth equally much to everyone on the margin.3
