Introduction

This book (together with the second volume) collects a representative sample of work that has sprung from Piero Sraffa’s two seminal contributions to economic analysis, namely the reconstruction of the core of the classical approach to value and distribution, and the foundations of a critique of the neoclassical (or ‘marginalist’) system on the grounds of the latter’s treatment of capital. The chapters published here cover a wide spectrum of issues, both of a purely theoretical nature and of a more applied character. Indeed, the possibility of an explanation of distribution other than the one in terms of factor supply-and-demand equilibria, re-opened by Sraffa at the most abstract level of analysis, reverberates in the view one may hold of virtually any aspect of the workings of a market economy — including, therefore, questions more directly suggested by actual experience.

The volumes are a result of the project started several years ago with the international conference on Sraffa and modern economics, held in Rome in 1998, on the occasion of the centenary of Sraffa’s birth. A number of papers presented at the conference are collected here, some of which have been re-elaborated during the long interval that has elapsed, while new ones have been added in the meantime. It is worth noting that much of this material does not deal directly with Sraffa’s own work, and rather provides developments in various thematic areas and directions of the basic contribution of this scholar to economic analysis — which testifies to the vitality of the approach of which Sraffa is a modern interpreter.

The structure of the book reflects the attempt to account for the state of research and debate around the main themes that were directly addressed in Sraffa’s work or have been stimulated by it. Parts I and II are devoted to the two core aspects of the profound revision of economic theory carried out by Sraffa, namely the critique of the basic principles of neoclassical theory and the reconstruction of the theoretical approach of the classical economists. In Volume II, Parts III–V comprise advancements along the path opened by Sraffa and either make analytical points directly connected to his work (especially Production of commodities by means of commodities) or, accepting as a premise the approach he recovered and reinforced, extend the investigation beyond the boundaries of Sraffa’s own analysis. Finally, Part VI contains papers of a historical character concerning Sraffa’s writings and biography.

In more detail, the contents of this volume are as follows.

Parts I and II, entitled, respectively, ‘Capital and the critique of neoclassical theories’ and ‘After Sraffa’s interpretation of Ricardo: historical and analytical issues in the classical approach to value and distribution’, concern the two connected lines of progress in modern economic theory that received a fundamental push from Sraffa’s work, namely the criticism of neoclassical theories, which is the subject of the chapters in Part I, and the reappraisal and development of the classical approach to value and distribution, which are what the chapters of Part II are concerned with. These two basic aspects of the theoretical reconstruction started by Sraffa are mutually connected in various respects that are revealed in the chapters collected in these parts. By questioning the principle of factor substitution, the capital criticism undermines the very foundations of neoclassical theories, and, quite naturally, such a basic objection to the dominant approach has been raised jointly with the awareness that a different, and no less comprehensive, paradigm was available. A further link connecting the capital theory criticism and the constructive work in the field of the classical approach is a strictly analytical one. Indeed, that criticism has already arisen, in Sraffa’s Production of commodities, from a meticulous investigation of the influences that changes in distribution are bound to exert upon relative prices, whatever the theory adopted. Furthermore, the simpler structure of classical theory helps to detect the nature and implications of those influences with a degree of lucidity that would have hardly been attainable within the intricate set of interrelations proper to neoclassical general equilibria.

Owing to the core character of the issues that are dealt with therein, a specificity of both Part I and Part II is that both of them allow for some debate by including, together with several other contributions, two key chapters, for each of which a comment is provided, as well as a rejoinder from the key chapter’s author.

In the first key chapter of Part I, Garegnani proposes an extension of the criticism of the neoclassical theory of capital — to which, as is well known, Garegnani himself contributed in the course of the capital theory debates — to the intertemporal versions of general equilibrium theory, in which sequences of equilibria over time start from given initial endowments of heterogeneous capital goods. Garegnani analyses the properties of these equilibria by means of the saving and investment general-equilibrium schedules he is able to extract from the intertemporal equilibrium system, once the saving and investment flows implicit in the latter are made visible. With the help of this tool, he argues that phenomena such as reswitching and reverse-capital deepening may occur and be responsible for instability and multiplicity of equilibria, even in these versions of neoclassical theory, despite the fact that no single value-quantity of capital appears therein among the stocks of available resources. Schefold’s comment deals with two main points, as does the reply from Garegnani. One is whether the saving and investment schedules used in Garegnani’s procedure are something merely implicit in the intertemporal framework, or are instead an unnecessary addition that, as such, would limit the significance of the critique Garegnani puts forward. The other point concerns the extent to which Gareg nani’s results may derive from the assumptions about consumer behaviour adopted in his model, rather than from capital theoretic problems. Gram also comments on the key chapter by Garegnani. After providing a diagrammatical representation of Garegnani’s argument, Gram dwells upon some of the properties (as well as the weaknesses) of intertemporal equilibrium theory, and in this regard he argues that, contrary to Garegnani’s claim, a capital endowment taken as an arbitrary physical vector is consistent with a uniform effective rate of return. In his short reply, Garegnani counters that the rate of return Gram refers to appears to be reckoned on the demand price of capital goods and not on the supply price, as the relevant equalization condition requires.

In the second key chapter of Part I, Tosato, after reviewing some of the literature, from Walras on, concerning capital formation in general equilibrium analysis, uses a model of general intertemporal equilibrium with capital accumulation in order to study the existence and generic regularity (i.e. local uniqueness) of equilibrium. Tosato also discusses the negative results as to the determinacy of intertemporal equilibria claimed by Mandler, as well as those of Sraffian background claimed by Garegnani and by Schefold, who contend that the uniqueness and stability of general intertemporal equilibria are undermined by capital theory difficulties analogous to those encountered by the old single-capital-magnitude versions of general equilibrium theory. Petri’s comment, and the reply from Tosato, touch on several groups of issues: some analytical as well as methodological problems in Walras’s original system and its subsequent evolution in the temporary and inter temporal versions of the theory; the objections that Tosato raises to the above-mentioned criticisms put forward by Garegnani and by Schefold; and Tosato’s claim of (local) uniqueness for his own model of general inter temporal equilibrium with capital accumulation. A point in Tosato’s reply to Petri is the object of a note by Parrinello, who is brought in by a reference to a work by him contained in the rejoinder to Petri, and of Tosato’s reply. The specific issue of the Parrinello—Tosato exchange is whether a special form of Walras’s Law has in fact to be assumed, once savings and investment flows are allowed to appear in the intertemporal general equilibrium system.

Moving on to the other papers in Part I, Harcourt touches on a number of issues he connects to the capital theory controversy and subscribes to the view that, at the heart of that controversy, there is a conflict between ‘history’ and ‘equilibrium’ — the equilibrium approach being unable, according to Harcourt, to deal with the basic aspects of capital accumulation. Potestio is concerned with the criticism of neoclassical theory as formulated by some ‘neo-Ricardian’ authors. She argues against the possibility of reaching meaningful conclusions about stability, as well as multiplicity of equilibria, when capital supply and demand are conceived as value magnitudes, as that defective treatment of capital deprives the analysis of significance, quite independently of the ‘perverse’ results due to reswitching and capital reversal. Serrano discusses at a general level the questions of the existence and stability of equilibrium in neoclassical theory. He notes that existence proofs are so unrestrictive — up to admitting the possibility of zero prices for some, or even all but one, productive factors — as to be devoid of significance with regard to the validity of the neoclassical explanation of market mechanisms; with regard to stability, Serrano points out how relevant to it capital theoretic problems are, and he distinguishes the problem of the stability of neoclassical equilibria from that of ‘gravitation’ within the classical approach.

In Part II, the first key paper, by Vianello, re-examines some issues relating to Sraffa’s ‘corn model’ interpretation of Ricardo’s early theory of profits.¹ The main point in Vianello’s paper is that Ricardo’s assumption of a wage consisting of corn was far from being new, as it can be traced back to Adam Smith’s influential analysis, in the Wealth of nations, of the effects of a rise in the money price of corn. In the first part of the paper, Vianello precisely exposes Smith’s treatment of that question, and then proceeds to consider Ricardo’s own theory of profits in its connection with Smith’s argument, also taking into account the positions held by Malthus in both regards. The comment by Porta and Vianello’s reply offer an exchange on the textual evidence for the assumption of a corn wage in Smith and other authors. Porta also reiterates, and Vianello reacts to, the interpretation of Ricardo’s Essay put forward by Samuel Hollander in contrast to Sraffa’s.

Hollander, in the second key chapter of Part II, discusses the argument, already advanced by other writers, that Sraffa was ‘led astray’ in his interpretation of Ricardo by a tendency to read Ricardo through Marxian spectacles. Hollander’s sub stantial agreement with that claim is based on the view that Sraffa’s interpretation, though not entirely wrong, provides a ‘trun cated’ version of Ricardo in ascribing to the latter the treatment of the wage as a given — whereas, in Ricardo, Hollander maintains, the wage rate results from the interplay of labour supply and demand, with the latter depending on the pattern of final demand and on the growth rate of capital. The comment by Stirati and Hollander’s rejoinder dwell upon the latter’s claim that, in Ricardo, one can find a systematic, decreasing relation between wage and labour demand based, in particular, on the indirect substitutability deriving from demand functions in commodity markets. In this regard, the consistency of Hollander’s suggested interpretation with Ricardo’s discussion of taxation and of the effects on employment of the introduction of machinery is also questioned.

In the remaining chapters of Part II, Stirati provides a critical overview of the controversy concerning the theory of wages in the classical economists, and she maintains that the two interpretations that have been regarded as the main contenders — the so-called New view and Fix wage interpretations — in fact show important similarities. A third point of view is advocated by Stirati and, differently from both the others, centres on the absence, in Ricardo and other classical economists, of any systematic trade-off between real wages and employment. Levrero argues that, in the classical economists and Marx, the institutional factors limiting wage flexibility are not necessarily in conflict with free competition, and that their existence can be better seen as an effect, rather than a cause, of labour unemployment. Owing to the absence, in their analyses, of a general trade-off between wages and employment, it was natural for those authors to assume that competition among workers could not but operate within boundaries, preventing an otherwise unobservable tendency of wages to fall to zero. Basing her analysis on Sraffa’s unpublished manuscripts, Picchio studies the concepts of subsistence and surplus wage we find there, namely the dual aspect of wages as cost of production and share in net income. Subsistence is a component part of capital, as it represents the ‘bread’ needed to enable the labourer to work, whereas net wages are ‘roses’, to be enjoyed as surplus. Picchio also points out Sraffa’s critique of the neoclassical conception of wages as something promised to induce the labourer to work. Using material from Sraffa’s papers, Gehrke attempts to clarify the significance attributed by Sraffa to the substitution, from the first edition of Ricardo’s Principles to the third one, of the phrase ‘price of wages’ with ‘price of labour’ or simply ‘wages’, which Sraffa refers to in his Introduction to Ricardo. Gehrke shows that, in his notes, Sraffa detects two elements of a possible explanation of that change in Ricardo’s terminology: the transition from wages as subsistence to wages as a share in income, and the introduction by Ricardo of the concept of proportional wages. Aspromourgos considers two manuscripts by Petty regarded as historically significant in relation to the origin of the surplus approach. In the first manuscript, an analysis of the social surplus arises in the context of Petty’s advocacy of an increase of population in Britain; the second manuscript enquires as to the technical means by which to implement the principle that taxation should be addressed to surplus consumption, thus favouring saving. Gilibert concentrates on the analytical role played by Sraffa’s standard commodity, which is that of providing a standard such that the relation between the wage, measured in terms of it, and the profit rate is not affected by prices. Gilibert specifies that that relation is the same as one would get in the case of an economy with a single basic commodity, which should be kept distinct from the case of a single-commodity industry, such as in Ricardo’s corn theory of profits. Bidard provides an ‘absolute’ standard of value with respect to technical changes, namely a composite commodity for which the change in the difficulty of production can be measured on its own, and which therefore can be used to assess how much of a price change is due to a change in the method of production of the standard or in that of the commodity whose price is being measured. Finally, Walsh investigates which concept of rationality is present in classical economics, and how it differs from the notion of rationality adopted in Arrow—Debreu general equilibrium theory. An open question in classical theory is, in Walsh’s view, how to deal with strategic choice (as an aspect of rational choice), as current game-theory methods, besides their purely neoclassical origins, involve problems that make them unsuitable.

As has been said above, the papers collected in Parts III, IV and V in Volume II deal with more specific issues, being concerned with the implications and applications of the critical-constructive developments in pure theory that are the subjects of Parts I and II.

Part III, ‘Technical change, variable returns and normal prices in the classical framework’, contains materials concerning the relations between sectoral outputs and prices in the classical analytical framework. The first paper by Ravagnani discusses a subject that has been controversial since the very aftermath of the publication of Production of commodities, namely whether Sraffa’s solution to the determination of prices requires any particular assumption about returns at the industry level. As the Preface to the 1960 book had in fact anticipated, apparently also reflecting a warning from Keynes, the price theory presented by Sraffa has often been imputed with the need of constant returns, especially from the neoclassical quarters (something which, when the allegedly necessary restriction is referred to as returns to scale, sounds even paradoxical if one looks at the severe assumptions about such returns that neoclassical analysis is compelled to adopt in order to be able to reach definite results). The consistency of Sraffa’s and, more generally, classical normal prices with non-constant returns is maintained by Ravagnani, who also argues for the soundness of a method that separates the study of the multifaceted factors affecting normal, long-run outputs from the analysis of the purely quantitative system of relations between prices and distributive variables. D’Orlando and Nisticò suggest that Sraffa’s assumption of given sectoral outputs is inconsistent with the long-period method. In their view, sectoral outputs can legitimately be left to a separate stage of the analysis only if the long-period method is abandoned and the classical-type theory is reconstructed on the basis of the temporary equilibrium method.

Other papers in Part III deal with a further basic aspect of classical normal positions, which is their role of ‘centres of gravitation’ of actual variables. Against the view that the concept of normal prices cannot be considered as relevant until economic analysis has been able to prove the tendency to such values, Petri claims that the competitive pressure to a normal profit rate is a fact empirically observable, as such not in question, and that what has to be provided by theory is rather a coherent explanation for it — whereas the negative results produced by some gravitation models have to be ascribed, he observes, to the arbitrariness of their assumptions. Petri also argues that the issue of gravitation in the context of classical analysis cannot be assimilated to the problem of the stability of equilibrium in commodity as well as factor markets, which arises in neoclassical theory. Initially considering classical-type models in which the tendency to normal prices does not generally hold, Bellino specifies that plausible changes in the assumptions, as introduced in later formulations of those models, are able to yield opposite results. He points out that an asymptotic tendency to normal prices has also been obtained in intertemporal and temporary equilibrium models (where it has to be meant as convergence to the constancy of relative prices over time). In relation to the process by which the tendency to normal prices is supposed to operate, Ciccone maintains that, differently from what is often believed, to admit the possibility of output variations through changes in the actual rates of capacity utilization is consistent with the classical view of an adjustment of sectoral outputs obtained through capital inflows and outflows.

Part IV, ‘Output and distribution in the long run: a classical-Keynesian perspective’, includes papers concerned with ‘macroeconomic’ themes. Once the demand-and-supply apparatus is rejected in favour of the socially based explanation of distribution proper to the classical approach, aggregate demand quite naturally assumes, at the present stage of economic analysis, the basic role of determinant of the level of social output. Indeed, in the absence of price mechanisms, such as those entailed by neoclassical theories, which would adjust expenditure to any otherwise determined level of output, no valid alternative is left, after Keynes’ lesson, to the opposite course, namely the adjustment of output to demand. The combination of the action of effective demand with a classical determination of distribution, which is a feature common to most of the papers in this part, allows the Keynesian principle to overtake the boundaries of short-run or sticky-price situations into which, despite its author’s original intentions, it was confined by the later ‘neoclassical synthesis’. Having in fact left unquestioned the traditional demand—supply determination of factor remunerations, the analysis of the General theory was unable to oppose a durable resistance against the long-run, competitive tendency to full employment associated to the functioning of the price system accepted at the time, whereas, as is shown in some of the papers here published, adopting a different theory of distribution and prices — a chance that had not yet matured in Keynes’ days — leads one to conceive of the autonomy of demand from output as a general property, which as such holds independently of market forms or the time span the analysis refers to.

The issue is discussed at a general level by Bhaduri, who argues that Sraffa and Keynes, as well as Kalecki, adopt an analogous method of analysis, in which the determination of output is treated separately from the determination of distribution. Bortis represents in analytical terms the consistency of the classical approach with the Keynesian basic ideas by providing a set of relations that combine a socially determined distribution of income with long-run levels of investment and output governed by demand. The long-run influence of demand on output, and therefore on investment, is in turn asserted by both Bonifati and Trezzini in their respective papers. Bonifati also analyses the further factors on which investment decisions may depend, which include the economic and social environment in which firms operate, the complexity of which does not lend to be reduced to any sort of definite investment ‘func tion’. Trezzini points out how the steady-state conditions often assumed in growth models prevent an adequate consideration of the role of demand in the growth process. This argument is here developed with regard to the so-called neo-Kaleckian models, in which a contradiction is shown to arise between the autonomy of demand and the constancy of capacity utilization entailed by the steady-growth paths adopted in those models. Jossa concentrates on the true essence of the Keynesian principle of ‘effective demand’, which in his view lies in the statement that, contrary to Say’s law, demand always creates an equivalent supply, and which, he claims, all interpreters of Keynes except Pasinetti have failed to formulate in those terms. In the papers by Panico and by Nell, money makes its appearance in the analysis. Panico attempts to integrate the ‘monetary’ and the ‘post-Keynesian’ theories of distribution, arguing that the dependence of the profit rate on the rate of interest proper to the former theory, and the relation between the rate of profit and the rate of growth maintained by the latter, can be rendered consistent by the intervention of fiscal policy, and in particular by suitable values of the ratio of public deficit to the level of income. Nell reconsiders the role of the quantity equation in the context of classical theory and argues that, once the process of circulation is properly represented, the causality of the relation proves to run from prices to money, rather than the opposite.

The papers in Part V, ‘Applied and policy themes in the reappraisal of classical economics’, offer some instances of application to specific issues of the approach that inspires, at a more general level, the contributions collected in Part IV. Leon explains the different behaviour of unemployment rates in the US and European economies as due to comparatively lower rates of growth of effective demand in Europe, rather than, as is commonly believed, to higher fiscal pressure and lesser labour market flexibility. Cesaratto emphasizes the crucial role of the state both in the process of accumulation, through the influence on long-run levels of demand, and in the distributive sphere, through welfare policies (including those concerning pensions) and their financing. Montani discusses the present tendency to increasing economic integration on the basis of the Ricardian theory of comparative costs, interpreted in the light of the stages of development theory. Palumbo discusses the so-called theory of the balance-of-payments-constrained growth and criticizes the idea that the conditions of external equilibrium are a sufficient basis for a general theory of long-run output determination. She also develops more general considerations on the multiplicity of factors affecting output growth and the difficulty of providing any quantitative representation of such factors, endowed with sufficiently general validity. Pivetti maintains that the project for the European Monetary Union has been accompanied by the advent of views, such as favour for the independence of the central bank in the determination of the monetary policy, that departed quite radically from those generally held before. According to Pivetti, that change was caused by a complex of both actual and theoretical circumstances, mutually influencing one another. Ginzburg and Simonazzi provide an analysis of the relation between interest rates and inflation during phases of stationary or slowly increasing levels of output in industrialized countries. In particular, they investigate the circumstances in which a tightening in monetary policy in the ‘central’ countries may prompt a process of disinflation through a worsening of the terms of trade of ‘periphery’ countries. Maffeo argues that the expansion of the American economy in the 1990s can be explained by the growth of aggregate demand primarily due to the response of private investment to technological changes, and further fed by public transfers, with no role played by traditionally advocated factors such as increases in the saving propensity or wage reductions.

Part VI, ‘Historical issues in Sraffa’s writings’, includes three papers. Fodor points out how the use of the expression ‘standard commodity’ was common in monetary economics before its adoption by Sraffa, and he formulates the hypothesis that Sraffa himself conceived the standard commodity as a ‘money of account’ in his analysis. Using documents from Sraffa’s papers kept at Trinity College and other materials, Naldi provides a reconstruction of some events related to Sraffa’s writings in the 1920s, including Mussolini’s reaction to the Manchester Guardian article by Sraffa on Italian banks, as well as the exchanges Sraffa may have had with Gramsci on the subjects of Sraffa’s 1925 and 1926 articles. Also referring to materials in Sraffa’s papers, Bellofiore and Potier deal with correspondence emanating from Sraffa’s relations with Keynes and Wittgenstein, as well as with correspondence and notes concerning Production of commodities and reviews of that book.

Thanks are due to Dr Saverio Fratini, Professor Sergio Levrero, Professor Antonella Palumbo and Professor Antonella Stirati for their help in the analysis and organization of the huge amount of material that these two volumes contain.

Last, but not least, we would like to thank the contributors and the publisher for their patience and cooperation.

Note

1 We have the sad duty to inform the readers that Fernando Vianello passed away in August 2009.