Postscript to Classical Economics

The classical tradition underwent considerable modification between the publication dates of The Wealth of Nations and John Stuart Mill's Principles of Political Economy. Nevertheless, an important thread of continuity - a common concern with the process of economic growth - linked the work of its main contributors.

Measured against this primary analytical objective the achievement of the classical economists was impressive. The perspective they provided on an economic system undergoing dynamic change was markedly superior to analyses available earlier. Moreover, many of their insights into the causes and consequences of economic growth have proved to be of lasting value. In the mid-twentieth century, students of growth and development have revisited classical literature for inspiration in the handling of a persistent set of problems. The classical economists were, after all, interested in the big questions: the process of economic growth over prolonged time periods and the relation of the resulting distribution of income to its prospects. The relevance of these issues has not diminished since they wrote; in fact, in much of the modern world, the questions to which they addressed themselves are the dominant economic issues. The classical approach, though susceptible to considerable updating and refinement, still has much to offer to mid-twentieth century readers.

It is thus not surprising that the analytical achievement of classical economists has left an impressive legacy. Their imprint on an important strand of modern analysis is conspicuous. Among the more interesting statements of updated classicism is the model of growth in underdeveloped economies devised by W. A. Lewis. His analysis, which is of considerable interest in its own right, provides an excellent illustration of the manner in which variations on classical themes can enrich understanding of a wide range of current issues.1

Lewis views the typical underdeveloped economy as divided into two compartments: a capitalist sector and a traditional subsistence sector. In his dualistic system, the capitalist sector is the source of dynamic stimuli and the rate of growth in the economy as a whole is held to be regulated primarily by re-investment from capitalist profits. Expansion of the capitalist sector, however, involves contact with the subsistence sector which, in Lewis's view, is characterized by backward techniques, low output per head, and a substantial volume of underemployment. In these circumstances, capitalists can draw off labour from the subsistence sector by offering wages sufficient to provide a slight improvement over the low real incomes that workers could otherwise have obtained in traditional agriculture. Moreover, capitalists can continue to tap the labour reserve of the subsistence sector indefinitely on low wage terms.

By substituting sectoral distinctions for the class divisions with which the original classicists worked, Lewis managed to reinstate much of the classical framework for the analysis of distribution and growth. The existence of a subsistence sector as a supplier of wage labour replaces Malthusian population postulates in the original classical model to yield the conclusion that output and employment can expand without raising real wages. Moreover, as in the classical scheme, profits are interpreted as the source of accumulation and expansion. The rate of profit, however, may be eroded for reasons similar to those underlying classical explanations of income redistribution in favour of rents. In Lewis's model, this problem is presented as a shift in the 'terms of trade' of industrial for agricultural products. Thus, for example, should the capitalist sector rely on the subsistence sector to feed the growing wage labour force, food prices would probably rise. The money wage would have to be adjusted upward in order to maintain real wages at established levels. Should this be the outcome, the rate of profit would be reduced and accumulation retarded. These tendencies, however, might be offset by improvements in agricultural productivity or through the creation of a self-contained capitalist enclave which produced its own food requirements and thus by-passed traditional agriculture. Ultimately, in Lewis's interpretation, expansion might reach a stage at which the subsistence sector's labour force would be completely absorbed in capitalistically organized lines of production. At that point'. . . an economy enters upon the second stage of development. Classical economics ceases to apply; we are in the world of neo-classical economics.. . .’2

Though Lewis's analysis is open to criticism on points of detail it has supplied a highly stimulating point of departure for a substantial body of current discussion of problems of underdevelopment. In particular, it has directed attention to some of the unique aspects of economic expansion in underdeveloped economies. Moreover, it is clearly oriented towards the basic questions relevant to the study of underdeveloped economies: the interrelationships between growth and income distribution over a prolonged period of dynamic change. With these considerable points to its credit, it is also of interest to note that an updated classical analysis also inherits some of the deficiencies of its predecessors: problems of short run price determination are slighted and, implicitly, Say's Law is assumed to hold. For Lewis's analytical problem a sophisticated treatment of these issues is no more relevant than it was for the original classicists.

In the modern literature, Lewis's model of growth is perhaps the most explicitly classical in form. But a number of classical themes, on a less comprehensive basis, have crept back into current discussions of long-period growth problems. Several recent analyses of economic growth in advanced economies, for example, have been constructed on the assumption that the distribution of income between the profit and non-profit shares of income is the basic determinant of the rate of growth. In these lines of argument it has been assumed that for the community as a whole, wage and salary earners save little or nothing even though their incomes are well above the classical subsistence level. This conclusion, moreover, is reasonably well supported by recent empirical studies. This is not to suggest, of course, that all wage and salary earners save nothing, but rather that the saving of some members of these groups has been roughly counter balanced by expenditures in excess of income by others (i.e. from withdrawals of past savings to cover retirement or emergency expenditures, through the financing of consumption by credit, etc.). The analysis of saving can thus be reduced to an inspection of the forces governing the behaviour of the non-labour shares of income. Those who have built theories on this basis have usually made the further assumption that income which is not spent on consumption is channelled into investment. All of this sounds very much like Say's Law refurbished in modern dress, as indeed is the case. It is part and parcel of this line of reasoning to regard full employment as a norm.

Other classical themes re-emerge in modern theories exploring the relationship between growth and distribution. In certain recent formulations, the share of profits is held to be the main regulator of the volume of accumulation and of the rate of economic growth. Though the classical belief that the rising rents associated with population growth would tend to erode the rate of profit has largely been discarded, models constructed with classical categories now usually maintain that profits are likely to be squeezed for another reason - the existence of diminishing returns to capital investment. Unless this tendency is offset by 'technological progress' - a possibility that classical writers also allowed for in their worries about the future course of rents -, the rate of profit will be depressed and a situation analogous to that of the stationary state may arise.3

Similar considerations have left their mark on contemporary analyses of growth problems in underdeveloped economies. For example, classical conclusions about the role of profits in stimulating capital formation and economic expansion have been invoked to support the case for the application of capital-intensive technologies in developing countries. Highly capitalistic techniques, it has been maintained, are likely to produce a distribution of income more favourable to profits than would more labour-intensive techniques. This conclusion, like its classical ancestors, implicitly assumes that maximization of the growth in output (irrespective of considerations of effective demand) is the primary economic objective and that profits, once generated, will in fact be re-invested productively.

Perhaps the most ingenious of the recent revisitations to the classical tradition has sprung from the work of Piero Sraffa, the indefatigable editor of the ten-volume edition of Ricardo's works and correspondence. By building on the base provided by Ricardo's attempt to derive a rate of profit from the corn economy without reference to valuation (rather than on the dynamic thrust of classical analysis), Sraffa has constructed a system in which the problems of an economy are viewed in terms of the conditions it must satisfy in order to sustain itself and to grow.4 This approach provides a highly illuminating statement of the technical requirements for economic survival and growth. Again all of the familiar classical puzzles re-emerge but in a form altered by the use of the notations of linear algebra. The substantive problems with which Ricardo wrestled - such as the derivation of natural prices and the determination of the rate of profit - remain and their solution is subject to the same constraints. The analysis of the demand side of market behaviour is truncated and the problem of aggregative instability (though not entirely ignored) is not accorded detailed attention.

While classical motifs have left an imprint on modern analysis, most classical thinkers were more concerned with promoting economic improvement than with advancing the techniques of economic analysis. Though they differed among themselves about specific issues they adopted a common procedure when approaching matters of public policy. For them all, one overriding question - the likely consequences of policy actions on the course of economic expansion - set the context of controversy. Classicists generally proclaimed the virtues of a free market, but they did so on grounds quite different from those invoked by later generations of economists. To members of the classical school, the unregulated market was more important as an engine of growth than as a process for optimizing the allocation of economic resources. Their views contrasted even more sharply with those of the Social Darwinists of the late nineteenth century who held that the struggle of unfettered competition insured that only the fittest and most deserving should flourish and that no sympathy should be wasted on the less fortunate. Theorists in this tradition cannot fairly be indicted, however, on the charge of insensitivity to human suffering. Their concern for the miseries of poverty was genuine. The message of those in the mainstream of classicism was that amelioration of distresses could best be achieved through the enlargement of production. Tampering with the distributional mechanism would simply increase claims on the social product and might even have a negative effect on the desired expansion of output. These conclusions followed from the view that the economic process was governed by laws beyond human control. Mill's re-interpretation of economic laws was required before the classical tradition could begin to erase the stigma of ‘the dismal science'.

Nor, in the circumstances of their times, did classical writers lack, a substantial basis for their suspicion of governmental involvement in economic affairs. The political environment they observed was not one in which governments could be regarded as champions of the general welfare. Before the franchise was widened, no government was obliged - as a condition of its survival - to respond to the concerns of a mass electorate. In these circumstances, it was not unreasonable to argue that the social consequences of diffusing economic power impersonally in unregulated markets were likely to be superior to those produced by a system in which narrowly based governments intervened forcefully in the economy. However plausibly grounded such an attitude may have been in the era classical writers observed, this political case for laissez-faire evaporated with the advent of broadly based social democracy.

To be sure, the classical tradition left some questions - questions that were to be elevated to importance by later schools of thought - unanswered. In particular, the classical frame of reference precluded a full exploration of two issues: the process of market price formation and the problem of economic fluctuations. Part of the neglect of these avenues of inquiry was related to the institutional setting of classicism. In the early stages of Western industrial emergence- when poverty and scarcity were the dominant facts of economic life - it was probably appropriate to concentrate attention on the expansion of output. A sophisticated analysis of demand and its significance in the economic process appeared to be unnecessary as, for most practical purposes, it could be taken for granted that additions to output could readily be absorbed. It was only in `abnormal' circumstances (such as those immediately following the Napoleonic wars) that writers of a classical persuasion diverted their attention from the supply to the demand side of production, and then only temporarily.

Nor did classical analysis attempt to offer a full account of costs and conditions of supply. In the circumstances of the first half of the nineteenth century, the reasons are not difficult to comprehend. These theorists were only dimly - if at all - aware of the complications introduced by the economies of scale brought by high technologies. In their economic world these problems were scarcely visible. Later in the century, however, applications of new technologies to large-scale production bred industrial concentrations which eroded the basis of their natural competitive order. John Stuart Mill caught the scent of this problem, though he did not pursue it far.

Similarly, a careful analysis of the nature of economic fluctuations did not then appear to be warranted. The classical economic universe, though not without disturbances, did not experience the consequences of aggregative instability in acute form or for sustained periods. Malthus sensed that something was lacking in the orthodox approach to this matter, but he failed to build a convincing counter-argument. For the most part, classical writers were content to assume that this problem would take care of itself. Their antipathy toward mercantilist views on hoarding reinforced a faith in the efficacy of Say's Law.

An appreciation of the analytical priorities of classical thought, no less than of the institutional climate of the age, is crucial to an appraisal of the strengths and the limitations of this corpus of theory. Its central focus was on the problems and prospects of economic expansion over an extended time period, with special attention to the interaction between the distribution of income and changes in total output. From this point of view, a detailed examination of short-period changes- whether in individual markets or in the economy as a whole - was not directly relevant. What mattered was the long-term trend and the forces that influenced it. At the same time, the classical tradition did devote part of its attention to certain issues of a short-period character, as was the case, for example, in the extended discussion of the relationships between value and price. These matters were not, however, examined for their own sake. Instead, they drew their pertinence from their relationship to the larger questions of growth and distribution.