The compensation principle and the social welfare function

Abstract

In chapter 2, emphasis was given to the Pareto criterion as a tool for ranking social states, for example, corresponding to different prices. A glance at figure 2.1 shows that the Pareto criterion is useless as a criterion for social choices in many, perhaps even most, real-world situations. In general, one would expect a policy change to produce gainers as well as losers, and the Pareto criterion cannot handle such mixed outcomes. Therefore, in this chapter, two different attempts to overcome this deficiency are discussed: the compensation principle and the social welfare function. Consider first the compensation principle. The compensation principle In order to apply the Pareto criterion, we only need to know whether households are better off or worse off following a policy change. Any project which makes everyone better off passes the Pareto test, while any project which makes everyone worse off is rejected by this test. Unfortunately, most real-world projects produce both gainers and losers, and as noted above the Pareto criterion cannot handle such mixed outcomes. In such cases, one may instead try to apply the compensation principle as a decision criterion. The compensation principle was suggested by Hicks (1939) and Kaldor (1939). Let us consider a project which moves the economy from state A to state B, and assume that some individuals gain from the move while others lose. Moreover, we assume that incomes can be costlessly redistributed across individuals. According to the Kaldor criterion , a project is desirable if, with the project, it is hypothetically possible to redistribute income so that everyone becomes better off than without the project.