Contenu du sommaire : La Théorie Générale de John Maynard Keynes : un cinquantenaire

The reception by French economists in the years following the coming out of General Theory. A delay in understanding. A surprise caused by the first obscurities of keynesian thought Remembering G. Pirou, who also died in 1946. Necessity for a new reading and maybe even a new writing. With his smartness and humour, would Keynes write again what gave him prestige in 1936 ?

The paper deals with the reception of the General Theory among the French economists during the fifteen years which followed its publication. More precisely, the authors try to stress the main theoretical questions French economists raised in reading the General Theory and the main answers they offered. A general examination of these questions and answers leads to emphasize the predominant and recursive theme of General Theory's temporal framework. The whole range of possible interpretations is first reviewed, according to the impact acknowledged to the General Theory from the viewpoint of its contribution to economic dynamics ; in this prospect, the theoretical origin of the authors (neo-classical, « classical », marxist, keynesian, ecclectic, ...) plays obviously an important part. More accurate criteria are then introduced to provide a more thorough description of the attempts which considered that the keynesian approach could be at least developed in a dynamic perspective. According to this viewpoint, several possible types of economic dynamics are distinguished and their contrasted contents are characterized. Finally, the distinctive features of a "French" reading of the General Theory are depicted ; they help to understand how the keynesian approach could be considered as a crucial contribution to economic dynamics.

Keynes' General Theory represents a change in theoretical method. The nature of this change is discussed in a Hicksian perspective. In fact, we take into consideration first what Hicks calls single period theory, and secondly what he calls continuation theory. From the standpoint of single period theory, the most revolutionnary change is the use of the method of expectations which implies that demands and supplies are regarded as determined not only by tastes and resources but also by expectations : equilibrium analysis can be used no only in static conditions, but even in the real world in disequilibrium. Yet, from the standpoint of continuation theory, the rigorous way to interpret Keynes' analysis is to regard the only stationnary solutions which imply stationnary stocks or stocks all growing at a common proportional rate. Therefore the continuation theory is setting forth on the assumption that short period expectations are always fulfilled and so must be adjusted to changes in long period expectations.

It is argued that it is not logically consistent to interpret Keynes' General Theory in the light of Hicks' conceptual approach of dynamics. Hicks' sequential analysis is ultimately based on the static equilibrium method. The basic time unit must be a period characterized by a stationary equilibrium. However such basic time unit is difficult to define in a stock-flow model where the stock equilibrium conditions might be inconsistent when they involve expectations with different time horizons. On the contrary, Keynes' implicit approach to dynamics in the General Theory is to get away from Hicks' stationary equilibrium of the basic time unit. The "day-to-day" key nesian dynamics is based on a basic time unit called the "accounting period" ; it is the time interval within which no change in decisions can be made. At the beginning of each "day", short-run and long- run expectations are formed, decisions are made and results of the previous period are known. At the end of the accounting period, balance-sheets constraints are fulfilled by issuing new assets and liabilities. Financial as well as physical constraints are reevalued. Although still not fully developed and formalized, this approach to dynamics based on a continuous "day-do-day" adjustment of expectations seems to be an alternative to the hicksian pseudo-dynamic equilibrium method.

In both the General Theory and his 1937 writings clarifying the main themes of the book, Keynes refers in various respects to his debts to Irving Fisher. To the modern reader aware of the cleavage between neo-classical value theory based on Fisher's work and the post-Keynesian theories stemming from Keynes' work this indebtedness appears perplexing. Recognising the dominance of the neo-classical intepretations of Keynes in modern times this paper seeks to assess the foundations of Keynes' references to Fisher and to clarify his contribution to the ideas of the General Theory 50 years after its publication.

Neither uncertainty nor the link between liquidity and the rate of interest are lacking in I. Fisher's theory. The distinction between Keynes and Fisher arises less from the analysis of capital evaluation than from the introduction of the concept of the demand for money, inherited by Keynes from Hawtrey. The keynesian statement of a monetary determination of the rate of interest rests on a dichotomy between consumption goods and capital goods, which is lacking in Fisher and Hawtrey as well.

The present paper first emphasizes that General Theory rests crucially on the assumption that the money rate of interest is independent of aggregate income. In this view are analyzed the origin and development of the concept of own- rate of interest, and of the market adjustment associated with that concept. It is shown that, against the appearances, the use of that concept by Sraffa and Keynes reveals two alternative theories of the relation between the money rate of interest and investment ; as a consequence, the scope of Keynes's approach becomes more precise and limited.

Even though the General Theory devotes few to those questions, for most of the commentators, it breaks the traditionnal conception of the nature and the extent of the public economic intervention. The purpose of this paper is to appraise that position from two points of view. Those of the way in which economists have defined the status of the State in relation to the economic system. Those of the compatibility between macro-economic strategies and democracy.

Keynes, like all the economists, refers to the State considered as a political subject, without providing the theory of it. Nevertheless, he introduces a deep break, in particular with micro-economics and "applied economics" according to Walras. His conception of economic agents, divided in "active" and "inactive" ones, questions not only the rights of rent-owners, but also the whole figuration of individuals as owners. It legitimates the use of financial State intervention on investment as pivot of the employment policy. The uncertainties and the limits of the keynesian "macro-policy" do not prevent it from being original as compared with "social economy" praised up by Walras.

This comment aims at backing at the analysis of the innovating nature of post-Keynesian theory, given the part played in it by money, an aspect that G. Dostaler does not approach. This part is studied from the angle of money wages and financial constraints. This makes it possible then to judge the contribution of the Post-Keynesians compared with M. Kalecki as well as the present inadequacy of the theory, as far as developments of Keynes's heterodoxy are concerned.

The aim of the present paper is to show the existence of a theory of monetary circulation in Keynes' thought. It is going to be shown that it is just the fact of having neglected the analysis of money supply and to have defined the money stock as an exogeneous variable, that allows neo-classical theory to consider savings as an independent variable determining the level of investment. The analysis of Keynes's theory of financement shows that he supplied a complete description of the monetary circuit, that he considered the money stock as an endogeneous variable, and that the analysis of the supply of money led him to drop, in his subsequent writings, the marginal theory of distribution, a theory he had followed in the General Theory, in favour of the new mark-up theory of prices and distribution, recently suggested by M. Kalecki.

Three questions are discussed in the comment of Graziani's paper : — it is suggested that the theory of endogenous money, expounded by A. Graziani, should be completed by the explicitation of the system of rules which is at its root ; — the role of financial markets is not only to finance the planned expenditures but also to avoid the bankruptcies due to the discrepancies between expectations and realisations ; — the "circuit approach" does not go very far into the study of the rules of circulation.

Keynes' General Theory analysis is limited by three kinds of assumptions which are reserved for the short period : non-conformity between maximum employment and maximum profit : full employment considered as the limit of the equilibrium situations and as a bottleneck of production. Therefore General Theory is not a model of a growing economy. Investment creates employment but is without effect on the volume and capacity of production, since the scale of production is given and the returns are decreasing. However, in his Collected Papers, Keynes shows that an entrepreneurial economy is caracterized by the growth of current wealth. In such an economy investment overtakes personnal accumulation to overpass the stationary situation of full employment. This takes us to a new examination of the relationships between saving and consumption and of the fundamental choice : to increase one's own wealth or to invest to increase the common wealth. Speculation and anticipations influence the decisions to employ and to produce by two factors : money and fixed capital. The General Theory only considers the consequences of investment, mentioning the user costs, the costs of disinvestment and the replacement costs without considering their effects in the analysis : our purpose is to integrate them in the capital cycle, and thus to abandon the short period hypothesis as well as the analysis in terms of gross — and not net — variables. Therefore from the income circuit we can pass to the gross product circuit, which enables us to find the golden rule of growth : "when the product increases, the income and the consumption increase, but the income increases less than the product and the consumption less than the income".

Our previous hypothesis about a duality of research projects in the General Theory is shown to put to the fore two theoretical notions, usually neglected when reading the G.T. : "Orthodoxy" (conventional Economics) and "Convention" . Keynes built the notion of "Conventional Economics" when he became aware of its inability to deal consistently with involuntary unemployment ; his "pragmatic" project then consisted in making unemployment logically possible inside "Conventional economics" by introducing strong uncertainty in the financial markets : although too high, the level of the interest rate results quite rationally from a "convention". Keynes' "radical" project, which he renounced in 1933, would have been to extend the behavioural consequences of such an uncertainty to goods and labour markets. Rules, institutions and conventions, being rational adaptations to uncertainty we suggest that the future of Keynesian economics depends on the development of the "Economics of Conventions" whereas the present "Conventional economics" may be the whole corpus of economic theories unable to deal with coordinating mechanisms other than market price (or rationing).

O. Favereau assumes with spirit and brio that Keynes would have deepened and disclosed various elements of his economic theory with such an order and stressing as to accommodate it to the most efficient fighting against orthodoxy at the time, according to his own judgment. This idea attributes the heterogeneity of Keynes's economic theory to an hesitation in his practical thinking, and not to obscurities or confusions of an analytical or epistemological type. On some points considered as decisive by Favereau himself, it seems that this question "strategic hesitation or theoretic confusion" may be rather settled in the second way. The following points are here stressed in turn : the qualification of the unemployment state by the notions of market or non-market equilibrium ; the origin of non-calculable uncertainty : time or money ? ; the existence in Keynes of a theory of wealth allocation, alternative to the orthodox market theory.

At the heart of Keynes's theory of financial speculation in the General Theory is the idea that the speculator's expectation is essentially concerned with what the other operators will believe about prices of the Stock Exchange at the next period. This situation is self-referential to the extend that the expectations have to do with the market's own product, i.e. the average opinion of the agents and not with an exogeneous value — such as "the fundamental value". The consequence is that speculation shows special characteristics such as self-fulfilling prophecies, and an indeterminacy of equilibria. Keynes's understanding of these phenomena led him to abandon the traditional view whereby speculation plays a stabilizing role. Consequently he conducts a severe criticism of the agent's strategy always of seeking liquidity. In conclusion we demonstrate that there is a close link between Keynes's findings and those reached in modern studies of speculative bubbles (theory of rational expectations).