Part I
Prolegomena
Chapter 1
Economic Theory and Policy
1.1Introduction
This chapter examines the implications for economic policy deriving from John Stuart Millâs perspective on the nature and function of abstract economics.
Section 1.2 spells out constraints on the applicability of economic theory flowing from its character as a specialized science of âwealthâ (read âincomeâ) based on maximizing behaviour and, accordingly, pertinent to particular institutional frameworks â primarily the competitive capitalist-exchange system but abstracting from what Mill termed the âprogressive movements of societyâ.
Recourse to a body of economic analysis is essential if only to reveal instances where desirable ends cannot be achieved simply because they are self-contradictory. But more specifically, theory is indispensable in the face of the complexities of causality. This prime issue is treated in 1.3. Various cautions regarding the âartâ of applied economics are here introduced.
The fourth section elaborates Millâs position that the scientific character of economics flows from the competitive principle. This generalization if taken literally would severely limit the practical applicability of theory, but it turns out that it excludes only small-number cases rather than monopoly or monopolistic competition. I take the occasion here to summarize Millâs enthusiastic arguments for competition from both efficiency and dynamic perspectives.
A companion discussion in 1.5 broadens the relevance of competitive theory in the light of extensions of its method to institutional structures other than the capitalist-exchange system, provided that maximizing behaviour is at play (precluding custom, gift, or force). That the choice of particular institutional arrangement has consequences amenable to theoretical treatment in terms of the competitive model is illustrated in 1.6 by the impact of productive organization upon efficiency and population growth in alternative institutional contexts.
Millâs famous contrast between âimmutableâ Laws of Production and âmalleableâ Laws of Distribution is introduced in 1.7. This contrast taken strictly â so central, if valid, to a theory of economic policy â is shown to be built on sand. For the inflexibility of the first category turns on the assumption of given knowledge and organization, while the amenability to social choice of the second neglects the consequences of such choice which are as inevitable as are the so-called Laws of Production. The contrast is only meaningful in the weak sense distinguishing between problems of efficiency and of value judgment.
The necessity for âverificationâ of theory before application to the âguidance of practiceâ is the subject of 1.8. And in the following section the inability of economics to generate positive historical predictions is explained. Its function, rather is to recommend measures that encourage âbeneficialâ and counteract âinjuriousâ tendencies.
Two particular issues, pertinent to the theory-policy nexus, will be postponed for later consideration but should be kept in mind here. The first, the subject of 7.2, concerns the technical linkage of economic cycles to the falling profit-rate trend generated by capital accumulation subject to increasing land scarcity â the classical growth pattern â a linkage creating a new justification for socially-desirable government expenditure on amenity and infrastructure, but consistent with the Law of Markets to which Mill held fast in the secular context. Here we will encounter the flexibility of the theory-policy relation in Millâs hands, incorporating allowance for the impact of changing empirical circumstances on the policy implications to be drawn from classical theoretical structures, and providing a case study of his concern to avoid simple-minded applications of economic principles. Moreover, the circumstance that cyclical fluctuations are a direct outcome of the falling profit rate provides a fine instance of Millâs methodological principle to rationalize observed empirical relationships as far as possible in terms of individual behaviour. The falling profit rate itself illustrates the notion of âtendencyâ to reflect not a necessarily observable pattern, but one force among other possibly conflicting forces playing on a particular variable.
The second, considered in 8.2, concerns Millâs chapter âOf the Stationary Stateââsurely one of the most cited in the history of nineteenth-century economics. We there show that Mill in fact gave a positive answer on balance to the question whether growth is desirable. Too many social desiderata turned on growth so that stationariness might be risky from the perspectives of labour displacement, capital loss abroad, resource misallocation, social conflict, and prospects for new technology.
1.2Constraints on the Applicability of Economic Theory
Mill distinguished the âscienceâ from the âartâ of political economy. âScience takes cognizance of a phenomenon, and endeavours to discover its law; art proposes to itself an end, and looks out for means to effect itâ (âOn the Definition of Political Economyâ; 1844d, p. 312). The science of political economy is perceived, in general terms, as treating âthe production and distribution of wealth, so far as they depend upon the laws of human natureâ (p. 318), and, more narrowly, as âconcerned with [man] solely as a being who desires wealth, and who is capable of judging of the comparative efficacy of means for obtaining that endâ (p. 321). The specifications are finally formulated as wealth maximization subject to âtwo perpetually antagonizing principles ⊠aversion to labour, and desire for the present enjoyment of costly indulgencesâ, which âdo not merely, like other desires occasionally conflict with the pursuit of wealth, but accompany it always as a drag, or impediment, and are therefore inseparably mixed up in the consideration of itâ (pp. 321â2). The science, therefore, is based on a specific behaviour pattern; and the social phenomena under investigation are themselves limited in range to the production and distribution of âwealthâ. By âwealthâ is intended our âincomeâ, the âantagonizing principleâ of âaversion to labourâ alluding therefore to a preference on the part of workers for leisure rather than wage income, and that of âpresent enjoymentâ to a preference on the part of capitalists for consumption rather than interest from savings.
An intimate connection is made between the (qualified) income-maximizing hypothesis and the limited range of subject matter treated by political economy. A presumption that different classes of social fact depended âimmediately and in the first resort ⊠on different kinds of causesâ, permitted âdistinct and separate, though not independent, branches of sociological speculationâ (A System of Logic; 1843, pp. 900â1). In political economy, the relevant class of social phenomena was that
in which the immediately determining causes are principally those which act through the desire of wealth; and in which the psychological law mainly concerned is the familiar one, that a greater gain is preferred to a smaller. I mean, of course, that portion of the phenomena of society which emanate from the industrial, or productive, operations of mankind; and from those of their acts through which the distribution of the products of those industrial operations takes place, in so far as not effected by force, or modified by voluntary gift (p. 901; emphasis added).
Income maximization was, accordingly, not one among roughly equal motives but the predominating influence governing a specific range of social phenomena, by which Mill intended primarily but, we shall find, by no means solely, capitalist-exchange phenomena.
Now Mill maintained that the âlaw of [any] effect is compounded of the laws of all the causes which determine itâ on the basis of a methodological principle â that of âcomposition of causesâ â whereby each individual causal law is at play even if its effects happen to be counterweighed by other, conflicting, tendencies (1844d, p. 322). Thus, for example, the Law of Diminishing Returns is operative although the raw cost data may not reveal its presence. This accounts for Millâs insistence on the Law in the course of a discussion of the âcountervailing tendencyâ of technological progress acting in âhabitual antagonismâ to it (Principles; 1848, p. 182). It is in this sense that the âtendency of the returns to a progressive diminution ⊠a result of the necessary and inherent conditions of production from the landâ (p. 187) is to be understood. Ideally, at some future time, there would be a synthesis of the results of all the specialist branches of social science to form a general theory of âwealthâ. In the meantime, economics provided only partial theories; and a deductive model based upon wealth-maximization as the âsole endâ of activity was an approximation suitable for the capitalist-exchange environment but to be qualified in application to particular cases in the light of so-called âdisturbing causesâ, including conflicting behavioural patterns (p. 323). These âdisturbancesâ are in fact the unexplained tendencies, in principle the subject matter of other social sciences, future sciences if not now existing.
Mill makes much in the System of Logic of the restricted scope of political economy, insisting that the practical guidance to be derived from the science was limited to given states of society or conditions of social affairs, for a specialist procedure which required that the range of study encompass a sufficiently homogeneous pattern of behaviour was âliable to fail in all cases in which the progressive movement of society is one of the influencing elementsâ (1843, p. 906). Thus even the isolation of âtendenciesâ required the postulation of a specific empirical environment. And more generally, apart from the need to assure the empirical relevance of the behavioural axiom, universally-applicable propositions were not in order because of the âeminently modifiable nature of the social phenomena, and the multitude and variety of the circumstances by which they are modified; circumstances never the same, or even nearly the same, in two different societies or in two different periods of the same societyâ (pp. 898â9).
We can sharpen our comprehension of Millâs position regarding the restricted scope of political economy by reference to his charge that the Benthamite theory of government based on the self-interest assumption â whereby the actions of the average ruler are determined solely by self-interest, such that the sense of identity of interest with the governed is producible only by his accountability to the governed â was âunscientificâ, considering its supposed predominance across time and space, its universality (1843, p. 892). This sort of objection did not apply to political economy envisaged as a science of relatively narrow scope. Rather, the procedure of distinguishing between income maximization and other behaviour patterns reflected a quest for practicality given the complexity of âcomposition of causesâ, and was valid only in the light of the empirical relevance of the maximization hypothesis in a specific temporal and geographic environment, namely the advanced competitive capitalist-exchange system. After all, the results would be trifling âif the assumed circumstances bear no sort of resemblance to any real onesâ (1844d, p. 325). Even so, the assumption has empirical relevance only as an approximation and might be subject to qualification considering the potential presence of âdisturbing causesâ, including alternative behaviour patterns.
As mentioned, the ideal was a general theory of wealth incorporating a variety of specializations. Apart from political ethology (the science of national character formation) Mill seems hesitant to elaborate what he had in mind, but by the distinction between competition and custom, and the constraint â to be elaborated in 1.4 â that âonly through the principle of competition has political economy any pretension to the character of a scienceâ (1848, p. 239), he presumably intended the character of an independent science, the case of custom being the subject matter of some complementary science of âwealthâ based upon an alternative behavioural axiom. An investigation of wealth turning on altruistic behavior might provide a further example. The objective, therefore, was a âsynthesisâ combining theorems relating to the production and distribution of wealth based upon a variety of different behaviour patterns, and linking the end of wealth creation with other ends of human activity, to arrive at a theory of âthe mutual actions and reactions of contemporaneous social phenomenaâ, or of âsocial staticsâ, which finally would yield a fully-fledged theory of social progress.
Political economy is rather less severely constrained than is suggested by the foregoing. Should motivation other than or in addition to income maximization be apparent in particular applications, âcorrectionâ of the results of pure theoretical analysis will be called for (1844d, p. 323). But âin a few of the most striking casesâ (including the principle of population) such corrections are âinterpolated into the exposition of Political Economy itself; the strictness of purely scientific arrangements being thereby somewhat departed from, for the sake of practical utilityâ. The two âperpetually antagonizing principlesâ will be recalled.
Relaxation of the constraints on the scope of political economy also arises in the treatment of knowledge. (And similarly elementary education and health.) For while the âmaterial fruits [of speculative activity], though the result, are seldom the direct purpose of the pursuits of savantsâ, and to that extent should not fall within the economistâs domain, professional flexibility was desirable: âBut when (as in political economy one should always be prepared to do) we shift our point of view, and consider not individual acts, and the motives by which they are determined, but national and universal results, intellectual speculation must be looked upon as a most influential part of the productive labour of society, and the portion of its resources employed in carrying on and remunerating such labour, as a highly productive part of its expendituresâ (1848, p. 43).
The incorporation of pure science reflected a realization that though income maximization might be irrelevant, so that the issue strictly speaking fell outside the economistâs professional competence, it was too close to home to be neglected on the grounds that it fell to some other science. Nonetheless, it is to a sort of descriptive, as distinct from analytical, economics that the topic is assigned. Mill here forgets that in some contexts he himself appreciates that theoretical science, as well as experimentation and innovation proper, is undertaken within commercial establishments based upon profit calculation. Indeed, it is a major theme (see 1.4) that competition acts as a stimulus both to invention and innovation. I am unable to account for Millâs seeming hesitancy on occasion to treat knowledge within the domain of political economy.
1.3On Deductive Method
The limited applicability allowed economic theory must by no means be understood as gainsaying the indispensability of the Deductive Method. To the contrary, Millâs harshest criticisms were reserved for those who forwarded âexperimentalismâ, failing to appreciate that scientific methods âto accomplish anything worthy of attainment, must be to a great extent if not principally, deductiveâ (System of Logic; 1843, p. 384). The âinfant stateâ of the social sciences could, Mill lamented, be ascribed to this failure. Thus, in discussing âfallacies of observationâ, Mill contrasted misleading surface manifestations of apparent causal linkages and a deeper undercurrent of causation moving in a very different direction â a contrast made much of by Marx (see 10.5.4). Conspicuous examples include the notion that lavish state expenditure is required as a stimulus to industry, an error corrected by the Law of Markets; the common case against free trade, corrected by classical allocation and growth theory (see Appendix A); the ascription of inflationary pressure to upward wage pressure, corrected by the inverse profit-wage theorem (itself turning on the theory of allocation or the cost-of-production theory of value), coupled with Quantity Theory reasoning; and the notion that expansion of the currency is stimulatory which erroneously assumes that inflation is unanticipated.
At the same time Mill certainly did not disparage empirical work. To the contrary, it provided an essential complement to the theoretical exercise at the purely technological level of conception. Thus the interpretation of an observed downward trend of agricultural costs over several decades requires an empirical investigation of the facts relating to agricultural productivity, innovation, changing work attitudes, labour relations, health, skill, business organization and so forth. The function of the âabstractâ science of political economy is to account for the observation at a more profound level involving human decision making, for which reason precisely â in an effort to overcome causal complexity as explained earlier â economic models are restricted to particular behaviour patterns and therefore to particular institutional arrangement where such behaviour is pertinent. Taking the âcanonicalâ Ricardo-Malthus-Mill growth model as a prime instance, we have a relation linking falling agricultural productivity (rising costs at the margin) due to land scarcity to the return on capital â a relationship entailing the inverse profit-wage relation and thus the competitive pricing mechanism in labour and commodity markets â and thence to savings decisions; and a reverse relationship brought into play by certain categories of innovation. Deductive theory alone tells us that the imposition of protective Corn Laws sets in motion the former tendency, a linkage disguised by the statistical data in our example. Accordingly, detailed knowledge of the facts of the case is essential to advance our understanding. Not only does such information reveal the presence of âdisturbing causesâ, thereby contributing to theory verification; it is the only means of arriving at theory refutation and at one remove to model improvement (see 1.8).
The aforementioned theoretical construct presuppose a specific set of behavioural and institutional axioms. But even assuming satisfaction of those axioms, it was unlikely that empirical laws pertaining to the effect of Corn Laws on âwealthâ within the rel...