Contenu du sommaire : Actualité et spécificité de la pensée classique

After questioning whether it is really important to debate the "specificity" of classical political economy and doubting that there is a single "neo-Ricardian" research programme, the paper points to the valuable elimination of error that has been achieved, for example, in the "capital theory debates". More positively, much has been contributed in the analysis of value and distribution, choice of technique, joint production and fixed capital systems. It is perhaps too soon yet to assess the work done on such questions as the "gravitation" of prices and profits, or the integration of value theory and effective demand theory. The study of non-proportional growth and of empirical and policy matters, it is suggested, represent areas of relative weakness.

In this comment, we discuss some of the reserves expressed by lan Steedman in his assessment of the neo-Ricardian school. Firstly, it is assessed that Sraffa's work should not be identified with the developments of neo-Ricardian school, so that it can be suggested that at least some of the critics addressed to them are not valuable for it. Secondly, it is sustained that the neo-Ricardian school does not represent the whole of the classical contemporary school, so that the limits to its developments do not prejudge of the analytical power of the latter. As far as the notion of an organised economy of production and the account made of a social division of agents draw a sufficiently clear line between classical and neo-classical economics, we must not be afraid, as Steedman suggests, by the jungle of semantic discussions. The problem of mutual compatibility between the needs of reproduction and the rules of distribution of wealth defines an approach that we can consider as an alternative to the usual supply and demand theory. The thus (re)defined classical perspective offers a balance both richer and more diversified than that of the neo-Ricardian program. The work of Nicolas Kaldor is a good illustration of this point. Finally, the relatively rare applications of the Sraffian theory must be related to its very nature, and do not necessary represent a limit of its developments.

This comment on lan Steedman's account of the neo-Ricardian research program contains three observations. The difference in kind between negative contributions of that program (demonstrations) and its positive ones (areas) casts some doubt about the theoretical framework adopted in common. The difference in treatment between the uncomplete positive contributions (which aim at some autonomy of the neo-Ricardian approach) and the already robust ones (presented as technical achievements) suggests some uncertainty about the specificity of that research program. Finally, the restriction of lacunae to applied economics and the ambiguous classification of accumulation do not allow to contrast the field of Sraffian theory with those of Walrasian and Ricardian ones.

It has often been argued that Sraffa's criticism of neoclassical theory did not affect the models of intertemporal equilibrium. But it is now known that intertemporal equilibria converge to long period positions, characterized by a uniform rate of profit in the classical sense, provided certain conditions are met. These appear to be related to the well-known effects of capital theory which undermine the marginal productivity theory. To analyse this problem, the profitability of processes in an intertemporal equilibrium is measured using the prices ruling at the beginning of each period ; it turns out that the rate of profit so defined is equal to the own-rate of interest of the numéraire commodity minus the rate of change of prices of the goods produced as outputs of the process. The rates of profit in this sense are unequal for different processes as long as production has not yet adapted to the long-run conditions of demand. On this basis, intertemporal equilibria are interpreted as processes of gravitation towards long-period positions and it is argued that, therefore, the arguments which apply against a neoclassical determination of distribution and employment in the long-period positions are also relevant for the analysis of intertemporal equilibria.

The relationships between the classical and neoclassical theories are much more complex that the neo-Ricardians use to recognize. Sraffa's study is indeed a clever criticism fo some ancient versions of the marginalist theory, such as those developed by Bo'hm-Bawerk or J.B. Clark, but it would be a mistake to identify them with the modem neoclassical theory which is able to consider production as a circular process and is unaffected by the critique of the theory of capital. Moreover the idea that an alternative theory of prices might not rely on any return to scale assumption is wrong. Sraffa's formalisation for prices of production requires constant returns and, as far as activity levels are concerned, the fuzzy notion of long run position must be identified with that of steady state or regular growth path in order to be compatible with the price equations. Under this assumption the prices of production are nothing but the intertemporal equilibrium neoclassical prices (at least for single-product systems, otherwise the "prices of production" do not generally exist). The turnpike theory studies the possible convergence towards such a steady state : if the answer is negative (as it may happen), it is meaningless to study a regular growth path which represents an unstable state of the economy.

The present paper discusses some long-run aspects of what Keynes called "the other point of view". In the interpretation given it is potential output or productive capacity that tends to adjust to actual output or effective demand, rather than the other way round. Early visions of the two alternative points of view clashed in the discussion between Malthus and Ricardo about the possibility of a "general glut" of commodities. The neo-Keynesian approach to growth and distribution theory is rejected on the grounds that the principle of effective demand is inconsistent with the assumption that productive capacity is permanently utilized at the level desired by entrepreneurs. Sraffa's analysis is then used to show how the cost-minimizing pattern of operation of plant and equipment is determined. In the final section it is discussed how investment generates the corresponding volume of savings both in the short and the long run.

This comment starts from a general discussion of the conceptual viability of the project of "ricardian synthesis" to which belongs professor Kurz's model of capital rate of utilization. Emphasis is made on two points : (i) The first one is the outstanding role played by financial factors, namely the "monetary" interest rate and the price of financial assets in the determination of investment, and thus in the construction of under-employment équilibra, conceived as "normal" situations. This allows to question the possibility of any analytical fusion with a classical theory devoted to the analysis of real capital, and which sees investment as constrained by the prior saving of profit earnings. (ii) Furthermore, the explicit denial by Keynes, after 1931-1932, of any idea of "natural" state or "natural" values makes highly questionable the possibility of making Keynesian short term states converge towards classical long term states. (iii) On this basis, the comment then examines some specific questions emerging from H.D. Kurz's model of capital utilization, such as the construction of an investment function, and its utilization for determining the current rate of utilization.

In my answer to Christian Tutin's comment I make the following points : (i) Keynes was concerned with establishing what he called "the other point of view", according to which effective demand has an impact on the performance of the economic system both in the short and in the long run. This presupposed the refutation of the neoclassical, especially Pigovian, doctrine which maintains that with flexible wages there will be a long-run tendency to full employment. Hence Keynes was not, and could not have been, content with a purely short-run analysis, as some commentators argues. (ii) The refutation of the neoclassical theory necessitated a critique of its notion of the rate of interest as the factor that equilibrates decisions to invest and decisions to save. Keynes's alternative theory of the rate of interest was, however, not fully convincing and cannot bear the burden placed upon it. (iii) Sraffa deserves the credit for having demonstrated that the traditional neoclassical theory of the rate of interest cannot be sustained. Therefore, Keynesians are well advised to take Sraffa's contribution seriously, also because the latter provides the theory of effective demand solid production theoretic foundations. In addition some questions relating to the little macromodel presented in my paper are discussed.

This paper is made up of a few chosen excerpts from a book (to be published shortly by Cambridge University Press) devoted to an analysis of the structural economic dynamics of industrial economies. It concentrates on technical progress more than on capital accumulation, and a central role is given to the evolution of the structure of demand — two phenomena which have long been neglected by dominant economic theory. The study is shown to follow from the Classical and Keynesian approaches to economic reality. As a simplifying device, a pure-labour production model is used and is shown to be of fundamental importance to grasp the basic features of the structural evolution of industrial societies.

Analysis of the role played by the hypothesis of uniformity both in technical and value composition of capital in the Ricardian theory of labor-value. It attempts at showing that the former (which is much more particular that the latter) is not necessary for explaining the profit-wage ratio in a system with given technologies. On the other hand, it is absolutely necessary in a general study of distribution which includes rent and which admits that production methods of all goods may be modified each one independently of the others. A general scheme is suggested whereby the limitations of Sraffa's interpretation concerning the definition of the rate of profit in the Principles are set forth.