Moral Good and Right Conduct: A General Theory of Welfare under Fundamental Uncertainty
Silva Marzetti Dall'Aste Brandolini
The methodological foundation of economic science is the postulate of rationality, which is its conceptual core. Simon (1978, p. 369) considers economics as that science which celebrates human rationality in all the ways that it reveals itself.
Traditionally, the positive conception of economics is distinguished from the normative conception in order to stress that 'what is' is different from 'what ought to be'. Positive economics requires economists to reveal economic laws and uniformities only about analysed facts, and to be neutral about competing moral systems. From this point of view, economics is an exact science, and agents behave rationally if they choose means which permit them to reach their established aims subject to the scarcity constraint. Normative economics, in contrast, allows economists a wider sphere of investigation by requiring them to deal not only with the laws that govern the real world but also with what the real world should or should not be, which depends on society's moral goods or values. Thus, from this other point of view, economics is a moral science, a science of thinking that is conditioned by the moral values established by society.
However, it is not always easy to treat the normative aspect of economics separately from the positive aspect, because the choice of ends, or moral values, may also be connected to the problems of their pursuit. In many cases, in fact, the rational choice of ends also becomes important. Consider, for example, a model where governments pursue partisan objectives because they use the public debt in order to influence the choices made by their successors (Alesina and Tabellini, 1990). When there is no agreement between governments that alternate in power on the composition of public expenditure, this situation may generate socially suboptimal public debt. When, in contrast, they share moral values that represent the foundations of social welfare, the composition of public debt may correspond to the socially optimal outcome. Therefore, we cannot deny that rationality has two main aspects: the aspect of the choice of moral values, and the aspect of right conduct.
Two fundamental conceptions of moral value exist, which justify the existence of different models of welfare economics: the ethics of motive and the ethics of end. The ethics of motive makes reference to subjective values, while the ethics of end recognizes intrinsic objective values. If we take instead the point of view of right conduct, two conceptions of rationality are also distinguished: Bayesian reductionism and rational dualism. These are two different ways of considering uncertainty about economic phenomena. Bayesian reductionism assumes that agents are always able to establish numerical subjective probabilities, and admits only the maximization procedure. Rational dualism, in contrast, distinguishes situations simple enough to be fully understandable by the human mind, where decisionmakers are capable of applying the procedure of welfare maximization (substantive rationality), from situations too complex to be fully understandable (as in the case of non-measurable uncertainty).
In this chapter the main characteristics of competing models of welfare economics are analysed, from the point of view of the moral values on which they are based and from the point of view of the right conduct they suggest, in order to find an answer to the following question: from the point of view of rationality under uncertainty, what are the essential characteristics of a general theory of welfare? In sections 12.2 and 12.3 the main economic welfare models based on the ethics of motive and instrumental rationality are analysed. Section 12.4 examines the ethics of end and considers in particular Moore's doctrine of the ideal in order to better understand his criticism of hedonism and of the Humean view, and his influence on John Maynard Keynes's moral and economic ideas. Section 12.5 considers criticisms of expected utility maximization and justifies procedural rationality. Section 6 outlines the fundamental characteristics of a general theory of welfare (GTW) under conditions of fundamental uncertainty. The concluding section argues that a GTW has to allow all the possible values which a society espouses as well as agents' behaviour under conditions of non-measurable uncertainty.