yet “humanity, generosity and public spirit, are the qualities most useful to others” (p. 191). Would it not be a gross exaggeration, then, to regard self-interest as solely responsible for the success of capitalism?4 Many empirical studies, like those for Japan, show that a punctilious observance of values like honesty, fulfilling contracts, avoiding (excessive) corruption has contributed nearly as much, if not more, to Adam Smith’s “Natural Progress of Opulence”. But even common sense suggests that a significant hold of complementary moral values on the human mind is necessary to shape an efficient society that vibrates with justice and fair play, positively values a sharp reduction in the inequities of human condition, and ensures an abundant provisioning of public goods – all issues at the resolution of which capitalism has been singularly inept.
In what follows both the positivistic and normative aspects of economics are presented. It is shown that a purely positivist view (in which self-interest is the oriflamme of economic forces) must be balanced by an ethics-related view (where considerations of social justice, and a commitment to the welfare of the “voiceless millions” guide human endeavour) to produce a creative synergy of the essential plurality of human motivation and conduct. Only thus can the equality of the human condition be ensured, and those adrift in a sea of powerlessness and poverty brought ‘on board’.
To set the stage for a fuller discussion of the moral issues in economic matters, let us first recapitulate the case for giving a starring role to the self-interest principle as the defender of the economic universe.
III. The ‘Universality’ of the Self-Interest Principle
The assumption of universal ‘selfishness’, regarded as a sure sign of rationality, is the regnant idea in much of modern economics. The point of emphasis of neo-classical economics (of which Benthamite Utilitarianism and Pareto-optimality are the central components) as well as of the positive public-choice theory (which admits ‘market failure’ but still accepts Pareto-optimality as a standard of reference for public policy) is a unifocal quest for efficiency. The moral-economic issues, like distributive justice and helping the poor, wither on the vine of scientific neglect. Even some ‘normative’ public-choice theories tell much the same story: they highlight such moral values as neutrality, unanimity and the priority of liberty to the exclusion of all other worthwhile social goals of what is generally accepted as good life. Yet another feature of this framework of thought is that, in it, a (minimalist) government is not expected to do anything positive to maximise social good; instead, it should be confined to safeguarding “negative freedoms”. Its only task is, therefore, to set rules of conduct that let self-interested individuals pursue freely the ends they value most.
The discussion in this section revolves around: (a) Pareto-optimality, (b) Libertarianism, and (c) Benthamite Utilitarianism; and it shows the moral vacuity of each of these theories. Thus, for instance, Pareto-optimality claims to be totally ‘value free’ and not concerned at all with issues of social justice; libertarianism advocates policy inaction in defence of a special kind of individualistic morality; and utilitarianism manages to maximise social welfare without getting any closer to a just society in which the thought of doing good to others would freely insinuate itself into the minds of men (and women).
i) The “Unimprovability” of Pareto-optimality
A situation is regarded as Pareto-optimal if there is no alternative situation in which at least one person is better-off while none other is worse-off. In other words, it depicts a situation in which the welfare (utility) of someone cannot be increased without reducing the welfare of others. More simply, Pareto-optimality reigns supreme if there is no other state in which everyone can be made better off.5 When there are no ultimate consumers in the model, Pareto-optimality, in a pure production framework, portrays a situation when there is no alternative state of the economy in which there is a greater production of outputs, or a lesser use of inputs [Allen Buchanan (1985)]. A Pareto-optimal state is deemed to be efficient, distributionally neutral, value free, fair and liberal. The central assumption here is that of a well-ordered society – namely, one in which a just distribution of income and wealth has already been accomplished, and where no outstanding socially gnawing interpersonal conflicts of interest strain the rather delicate structure of the free markets. Given this vital assumption, the two fundamental theorems of welfare economics, referred to above, establish a two-way link between Pareto-optimality and competitive equilibrium: assume that all individuals and firms are selfish price-takers. Then every competitive equilibrium is Pareto optimum. Assume, again, that all individuals and firms are selfish price-takers, and that the initial endowments have been suitably redistributed by lump-sum transfers, then Pareto optimum can be achieved by competitive equilibrium [Feldman (1991)]. These theorems, subject to the following restrictive assumptions, prove the above-noted nice properties of Pareto-optimality.6 (i) Market arrangements are the most efficient because all points on the Pareto-efficiency frontier denote an ‘unimprovable’ arrangement of scarce economic resources. But such efficiency guarantees, for whatever they are worth, hold only in the space of utility. (ii) By the same token, these arrangements are the most liberal in the narrow sense that there is no room here for government interference. (iii) Since the very best state of the economy must at least be Pareto-optimal, and in competitive equilibrium, market arrangements must be mutually advantageous and fair. (iv) The Pareto-optimality principle could even be regarded as morally non-controversial, because no one can rationally complain about a move from a Pareto-inferior state (in which someone can be made better off without making anyone worse off) to a Pareto-superior state (in which no one can be made better off without making someone worse off); and because such a move would be in everyone’s self-interest.
On closer examination, however, the story looks too good to be true. The rule will not guarantee efficiency if only a little bit of reality is allowed into the watertight set of assumptions that sustain it. Thus, if public goods (characterised by the jointness of supply, non-excludability, and non-rivalrous consumption) of various kinds must be produced and consumed, the ‘free-rider’ problem (i.e., I do not contribute because others most probably will) and the assurance problem (i.e., when I, though not a freerider, will contribute only if assured that others will also do the same) will not let the market produce these goods. In such situations, appropriate state intervention can almost always improve competitive market (and Pareto optimal) solutions [Stiglitz (1991)]. Similar results will follow if such real-life phenomena as asymmetric information, moral hazard, principal-agency syndrome, multiple equilibria are (as they should be) admitted into the analysis; or if markets are missing or incomplete, or if they are liable to complete breakdown. These problems do not go away even when extended to situations of uncertainty and which involve an inter-temporal resource allocation. This is because such extensions require making some additional heroic assumptions – namely, that (i) all economic agents are identically uncertain; (ii) a complete set of contingent markets exists; (iii) transaction costs are zero. True, with these assumptions the Pareto-optimality rule becomes dynamically valid; but it would amount to little else. In particular, this dynamic version too will still fail to deliver if many key contingent markets (e.g., those for sharing risks) do not exist, the information available to economic agents is not symmetric, etc. To add to the rule’s woes, the second theorem does not settle the issue of distributional equity because Hicksian lump-sum transfers from winners to losers, which must be made to prove the second theorem, are not meant to be actually made. Furthermore, even common sense suggests that “there simply is no way to judge the changes that affect distributions while remaining neutral on distribution questions” [Hausman and McPherson (1993); p. 703].
Indeed, the way Pareto-optimality is defined offers no solution to problems of distributive justice. For instance, this rule is perfectly consistent with a famine-like situation in which some
