post-autistic economics review
Issue no. 27,  9 September 2004
article 1



      issue 27 contents                            PAE Review index                               home page




The economics of Keynes and its theoretical and political importance: Or, what would Marx and Keynes have made of the happenings of the past 30 years and more?*

G. C. Harcourt
   (Jesus College, Cambridge University, UK)

© Copyright 2004 G. C. Harcourt


I start with two propositions: first, that Maynard Keynes and Karl Marx, were they still with us, would have made far more sense of the happenings of modern capitalism of the past 30 to 40 years than do the more modern approaches to macroeconomics of the same period; and, secondly, that Keynes would have sat down and tried again to save capitalism from itself.  (Marx may have rubbed his hands and hoped that its demise, so often predicted by him and his followers, was at last on hand – but I could not bet on either of these.)  It may surprise you that I couple Keynes and Marx together, but I would argue – the evidence is supplied in a fine book by Claudio Sardoni (1987) – that, adjectives apart, when Marx and Keynes examined the same issues in the capitalist process, they came up with much the same answers.  Perhaps, on further reflection, this should not be surprising, for along with Michal Kalecki and Joseph Schumpeter (said by Joan Robinson to have been Marx with the adjectives changed), they have made the deepest, most insightful analyses of the laws of motion of capitalist society in our profession.  (Marx’s views on socialism are another matter, see Harcourt and Kerr (2001a).)


I shall say more about their analyses below.  First, let me clear out of the way why I think the modern approaches are less than satisfactory.  They employ either representative agent models, or Frank Ramsey’s benevolent dictator model, or an emphasis on certain imperfections in the workings of capitalist institutions, such as are to be found in New Keynesian models: sticky wages and prices, imperfectly competitive market structures, asymmetrical information and the like.

Modelling the economy as a representative agent rules out by assumption one of the fundamental insights of Keynes (and Marx), to wit, the fallacy of composition, that what may be true of the individual taken in isolation is not necessarily true of all individuals taken together.  This implies that when looking at the macroeconomic processes at work in capitalism, we cannot presume that the whole is but the sum of the parts.  Indeed it is not.  We have, therefore, to consider the macroeconomic foundations of microeconomics as James Crotty, citing Marx, told us long ago now, see Crotty (1980), and on which Frank Hahn, innocent of all this, is now working, as has been Wynne Godley too for many years.  In fact that great and wise Keynesian, Lorie Tarshis, regarded the use of the representative agent as the greatest heresy of modern macroeconomics and explained why in Tarshis (1980), see also Harcourt (1995; 2001a).


As for the use of Ramsey’s benevolent dictator model, a re-read (or a read for the first time) of his classic 1928 article, “A mathematical theory of saving” together with his own scathing assessment of it1, ought to show how fanciful it is to argue that, in a completely different setting, it could illuminate what has been happening in actual interrelated modern economies in recent decades, in fact, any decades.  As for the New Keynesians, while it is possible to applaud many of their policy conclusions and make common cause with them on them (a plea I first made in 1980 though my paper was not published until 1996-97, see Harcourt (1996-97; 2001a)), I submit that their policies do not always follow logically from their theories.  By basing their results on imperfections, they imply that if the latter were not there in the first place, or were to be removed, all would be well.  But as Marx and then Keynes argued, freely competitive  capitalism with power diffused equally between all individual decision makers and the recipients of such decisions, especially wage-earners, so that, in effect, no one individual has any power, still would not work in an optimal manner.  In particular, it would not necessarily provide full employment of labour and capital either in the short or the long period, so that booms and depressions, inflations and deflations and in certain circumstances deep crises, could still be the order of the day.  An especially astute argument for an aspect of this set of arguments is to be found in Nina Shapiro’s 1997 paper, “Imperfect competition and Keynes”.  She argues, plausibly (but fairness demands that I refer the reader to Robin Marris’s paper, “Yes, Mrs. Robinson! The General Theory and imperfect competition”, Marris (1997), that immediately precedes Nina’s paper in Harcourt and Riach, vol. 1 (1997)), that an economy characterised by freely competitive market structures would have cycles of greater amplitudes and higher average levels of unemployment over time than one characterised by imperfectly competitive market structures.  This insight is shared both by her contemporaries, for example, Paul Davidson and Jan Kregel and by distinguished predecessors, for example, Austin Robinson who always lamented the relative lack of interest by Keynesians in the early post-war years in the systemic effects of market structures, regional experiences and requirements and the like, Michal Kalecki, whose review of The General Theory which alas, though published in Polish in 1936, was not available in full in English until 1982, see Targetti and Kinder-Hass (1982), forcefully makes this point in his usual lucid and succinct way, and John Kenneth Galbraith in his greatest classic, The New Industrial State, Galbraith (1967).   All these economists, together with Marx and Keynes, were analysing how key decisions made in an environment of inescapable uncertainty impact on systemic behaviour.

The thrust of Robinson’s, Galbraith’s and Shapiro’s argument is that anything that reduces the impact of uncertainty on the decisions on the production, employment and, most importantly, accumulation of firms (the most fundamental unit of analysis in Keynes’s macroeconomics, a point emphasised repeatedly by Tarshis, one of Keynes’s most devoted pupils and disciples, see Harcourt (1995; 2001a)), is likely to result in more satisfactory and stable systemic behaviour.  Especially is it likely to beget a higher rate of accumulation on average and so a greater chance of absorbing (offsetting) the level of saving associated, if not with full employment levels of income, at least with high levels, certainly higher levels than would occur in a system characterised by the Marshallian freely competitive structures that Keynes used for most of the time in his models in The General Theory itself.

Though the New Keynesians have mounted vigorous and, to my mind anyway, telling counter-attacks on the new classical macroeconomics within the latter’s own framework, see, for example, Hahn and Solow (1995), they have not themselves completely escaped from the clutches of what Joan Robinson once aptly dubbed “Pre-Keynesian theory after Keynes”, Joan Robinson (1964).  It is true that they have routed the extreme idea associated with the beginning of the use of the hypothesis of rational expectations by the new classical macroeconomists that the world may be analysed as if perfect competition and perfect presight reigned so that the Arrow-Debreu model could be used as the base on which to erect theory and policy.  And it is also true that the rational expectations hypothesis when it is uncoupled from the Lucas vertical aggregate supply curve, is just a hypothesis deserving to be tested.  Indeed, if it were found not to be inconsistent with the facts, and if the world is correctly illuminated by Keynes’s model, coupling them together would serve to reinforce policies of intervention for then thinking alone could make it so, as it were.  Yet, having cheered all this, there are still so many remnants of what Keynes dubbed classical economics present in the New Keynesian approach as to make it logically unacceptable as the appropriate model or even “vision” for starting an analysis of the modern world: that is to say, a world in which foreign exchanges have been floated, sometimes dirtily, often freely, financial markets have been deregulated, credit has been made “available to all”, capital controls have been removed in many economies, labour markets have been made flexible (a euphemism for making the sack effective again by recreating the reserve army of labour after the full employment years of the long boom as the Marxists have it or Golden Age of capitalism as the Left Keynesians dubbed it), international trade has been liberalised at least in some directions, often at the expense of the South and to the benefit of the North, and technical advances have reduced the length of the short run in financial and other markets to hours rather than weeks or months.  For it is not obvious that the equilibrating mechanisms of supply and demand (even if associated with path dependence) with their underlying theme of harmony, balance and voluntary choice are universally the appropriate tools to use.  So let us reiterate the essential lessons that Keynes taught us.


I briefly sketch what I have come to believe is the essence of Keynes's new position, as he saw it himself in 1936 and 1937, as he moved from the Tract through A Treatise on Money (1930) to The General Theory (perhaps we should start from A Treatise on Probability (1921) and The Economic Consequences of the Peace (1919).)  I do not give chapter and verse for what I have to say; it is based on my reading over many years of The General Theory and Keynes's other books, the Collected Writings, especially volumes XIII, XIV and XXIX, and much secondary literature, especially in recent years Robert Skidelsky’s superb three volume biography of Keynes, Skidelsky (1983, 1992, 2000).

The essential characteristics of the Marshallian system as Keynes viewed it was, first, the domination of the long period and secondly, a strict distinction between the real and the money.  In the real system, supposing there to be free competition, the object of the analysis was to determine long-period normal equilibrium prices and quantities, using partial equilibrium supply and demand analysis (but showing in an appendix that the same principles apply in a general equilibrium system, to wit, that equilibrium prices were, as we say now, market-clearing).  The analysis was as applicable to the market for commodities as it was for those for the services of the factors of production.  As for the process of accumulation, there was a supply of real saving, consumption foregone, associated with maximising expected utility choices between present and future consumption, with the rate of exchange reflecting time preference at the margin; and a demand for saving, investment, in which the technical possibilities of transforming present consumption into future consumption at the margin were the key concepts.  The price which cleared this market and set the composition of the national income between consumption and saving/investment was the natural rate of interest, a real concept.

The general equilibrium version would have as a corollary the Say's Law level of long-period overall output, itself a 'simple' summation of the individual quantities of commodities (and employment) associated with the long-period market-clearing prices of each individual market.  So what determined overall employment (and zero, non-voluntary, unemployment) was not an interesting theoretical question, if it were ever even to be asked: only simple summation was required.

When we come to the discussion of the determinants of the general price level – so far only relative prices and quantities have been discussed, neither money nor money prices played any significant analytical role – the quantity theory of money tautology could easily be turned into a theory.  For if M was determined by the monetary authorities, V was given by institutions and historical customs and T was interpreted as the total of transactions associated with the Say's Law long-period equilibrium position, P remained the only unknown.  Moreover, if V and T were given, changing M would, at least as a long-period tendency, change P in the same proportion.  (Keynes would have expressed all this in terms of the Marshallian/Cambridge version of the quantity theory but the story is essentially the same.)  Money, therefore, was only a veil in the long period.

The object of volume II of  a Principles of Economics was to set out this basic theory, analyse the causes of fluctuations around the long-period position (the trade or business cycle) and design institutions which either allowed the economy to return as quickly as possible to the equilibrium position after a shock; or to move as painlessly as possible to a new equilibrium position if the basic real determinants of it – tastes, endowments, techniques of production – themselves changed.  The essential task of the monetary authorities was to ensure that the money rate of interest was consistent with the underlying natural rate of interest which like saving ruled the roost in the process of accumulation.  This, in the crudest, simplest form, was the system on which Keynes was brought up, as he came to see it.

Because of the real/monetary dichotomy, and because he was writing on money, Keynes felt inhibited about spending time on the intricate happenings to output and employment in the short period and over the cycle, "the intricate theory of the economics of the short period".  Nevertheless, in the Tract he recognised them and  especially the evils of unemployment as well as falling prices  – hence he cheeked Marshall about our mortality in the long run – but, analytically, he was looking for institutions and their behaviour which would give price stability and allow the economy to settle at its long-period Say's Law position.  In A Treatise on Money he presented the famous banana plantation parable but he was unable analytically to stop the downward spiral of activity and prices until either the inhabitants had starved to death or there was an ad hoc change in their accumulation behaviour (Keynes 1930; 1971, C.W., vol. V, 158-60).  The endogenous process and its end had to wait for the publication of Kahn's multiplier article in 1931 which also contained “Mr. Meade's relation” – the derivation of the value of the multiplier by concentrating on the leakage into saving.

Keynes replaced the old system by a radically new, indeed revolutionary, system.  As a Marshallian his basic tools were demand and supply functions, now aggregate ones.  His emphasis was on the short period in its own right, suitably adapted for analysis of the economy overall.  (This had been the emphasis, too, in Kahn's dissertation, The Economics of the Short Period (1929; 1989) though Kahn's analysis was microeconomic.)  The dichotomy between the real and the money disappeared in both the short period and the long period (which Keynes ultimately ceased to believe to be a coherent concept in macroeconomics).  Money and other financial assts and monetary institutions entered the analysis from the start  (institutions were only sketched relatively to the rich analysis in A Treatise on Money, a deliberate choice by Keynes).  Aggregate planned expenditures basically drove the system which operated in an environment of inescapable uncertainty.  The latter had inescapable consequences for vital decisions, especially regarding investment expenditures and the holding of money and other financial assets and the form that they took.  Investment dominated and saving responded through the consumption function, the relationship between aggregate disposable income and the distribution of income between the classes on the one hand, and planned consumption expenditure, on the other, intimately related to the (income) multiplier through the marginal propensity to consume.  The amount saved (but not the form in which it was held) was treated as a residual.   Investment was determined by expected profitability, on the one hand, and the money rate of interest, representing the alternative ways of holding funds (and their availability and cost), on the other.  Subsequently, in 1937, finance, especially through the banking system and the stock exchange, was also to play a vital role as, cet. par., the ultimate constraint on investment expenditure.  The money rate of interest therefore ruled the roost and the expected rate of profit (the mei, the counterpart to the natural rate of interest in the old system) had to measure up to it.  The money rate of interest was depicted as the price which cleared the money market by equating the demand for money with its supply, not as the (real) price which equalised desired saving and investment.

The rest state in both the short period and the long period (the latter was ultimately to become for Keynes and his closest followers but economics for economists) could be associated with involuntary unemployment – people willing to work in existing conditions but with the level of aggregate demand such as there not to be sufficient demand for their services.  Nor was there any effective way for them to signal that it would be profitable to employ them; indeed, there would not be unless there were to be a rise (or an expected rise) in real expenditures.  Up to full employment, the outcome in the labour market depended on what happened in the commodity market.  The quantity theory was replaced as an explanation of the general price level by old-fashioned Marshallian short-period competitive pricing, suitably (or perhaps not) adapted to the economy as a whole.  There were therefore at least three 180º turns between the old and the new: investment dominated saving, the commodity market dominated the labour market and the money rate of interest dominated the expected rate of profit.  The forces which would make planned accumulation even on average absorb full employment saving were unreliable and weak, not to be relied on even as tendencies.  Moreover, the general price level was determined by factors other than the quantity of money.


The new system was the base on which Keynes would build his theory of inflation in How to pay for the War (1940; 1980) and his policy proposals for the international world order in the postwar period.  In his superb review article, Vines (2003), of Robert Skidelsky’s third volume of his biography of Keynes, Skidelsky (2000), David Vines makes a convincing case for the proposition that Keynes provided the conceptual basis for modern international macroeconomic theory.  Of course this is not to be found explicitly in The General Theory itself.  That book was mainly concerned with a closed economy model in order to highlight the central theoretical propositions and insights of the new theory.  Nor did Keynes analyse the trade cycle or long-term growth issues systematically in The General Theory and some of his obita dicta asides look rather strange now.

For most of The General Theory Keynes was content to discuss existence and stability propositions in the short period, focussing especially on the factors that were responsible for the point of effective demand at which aggregate demand and aggregate supply, and planned investment and planned saving (more generally, injections into and leakages from the expenditure-production-income circuit) were equalised.  (He said later that if he were to write the book again he would have been more careful to separate out the fundamental factors responsible for the existence of the point of effective demand from the other set responsible for stability and reaching the point through a groping process by business people.  He thought that Ralph Hawtrey had confused the two, see Keynes, C.W., XIV, 27, 181-82.)

In his most stark model, one designed not so much to describe the world as is, as to bring out most simply what was at stake, he assumed, as Jan Kregel (1976) has told us, that short-term expectations concerning immediate prices, sales, costs et al., were always realised and were independent of long-term expectations concerning their future courses, the ingredients most relevant for investment decisions, so that planned investment could provisionally be taken as a given and the point of effective demand established immediately.  In his most sophisticated model of (the same) reality, the independence of the two sets of expectations was scrapped, the point of effective demand was not realised immediately and indeed it changed over “time” as the model of shifting equilibrium came into play.  This last apparatus is in rudimentary form the starting point for the development of growth theory by Richard Kahn and Joan Robinson, Nicky Kaldor and Luigi Pasinetti and the models of cyclical growth by Kalecki (independently) and Richard Goodwin.


Both Marx and Keynes recognised that when financial capital was not moving in tandem with industrial and commercial capital (Marx would and Keynes would not have put it this way), malfunctioning and sometimes crises were likely to occur.  Keynes set out his ideas on this in, for example, the key chapter 12 of The General Theory on the operation and non-operation of the stock exchange and its relationship to real accumulation and activity generally.  Another key step was in his 1937 paper on the finance motive, see Keynes, C.W., vol. XIV, 201-23, on how the banking system in particular holds the key to the realisation of investment plans, taking as given the state of long-term expectations.  The stock exchange also has a key role because the repayment of the bank loans used to finance the setting up of investment projects, the start of the process of accumulation, depends upon the firms concerned being able subsequently to place new issues of shares and debentures at satisfactory prices.  (The demand for the new issues comes, in part at least, from the placement of the new saving created by the new investment.)  The point is that finance and saving are sharply separated by their roles and place – timing – in the process of accumulation.

These ideas were subsequently developed by Hyman Minsky in particular, writing under the rubric of his financial instability hypothesis.  Minsky spelt out ideas, perhaps more implicit in Keynes’s and Dennis Robertson’s writings, that the natural, probably inescapable, cyclical movements on the real side of the economy can be enhanced both upwards and downwards by events in the financial aspects of the economic process, resulting in the greater amplitudes of the actual cycles experienced by economies.  Minsky stressed the feedbacks associated with the disparities between expected cash flows and actual or realised cash flows in the accumulation/production process, on how non-realisation acts on confidence and expectations, enlargening the boom, at least in its early stages, accelerating the downturn and deepening and prolonging the subsequent recession or depression.2


As well as pointing out the implications of disparities in the progress of finance capital in relation to commercial and industrial capital, Marx’s analysis of the inherent contradictions in capitalism are of immediate relevance for our purposes in this paper.  Unlike Keynes and, to a lesser extent, Kalecki, Marx made a clear distinction between happenings in the sphere of production, on the one hand, and happenings in the sphere of distribution and exchange, on the other.  As far as the possibility of and limits to accumulation are concerned, it is conditions in the sphere of production – the length and intensity of the working day, the state of the class war between capital and labour, employer and employee – that ultimately determine the size of the potential surplus created for the realisation of profits and for future accumulation.  Whether this potential is realised or not, though, depends upon happenings in the other sphere of distribution and exchange.  It is here that Keynes, Kalecki and developments based on their contributions come into play: the combination of the theories of investment and of the distribution of income determined by the expanded version of the theory of effective demand decides how much of the potential surplus is realised in actual profits and accumulation, see, for example, Harris (1975; 1978).

These ideas help to explain one of the paradoxes of recent decades.  Monetarism has rightly been called by the late Thomas Balogh (1982) “the incomes policy of Karl Marx”.  Ostensibly, the theory was meant to justify policies designed to rid the system of inflationary tendencies.  In fact, it was associated with the attempt to swing the balance of economic, social and political power back from labour to capital.  (The reverse swing had occurred cumulatively in many advanced capitalist economies during the years of the long boom.)  The means to this end was the recreation of the reserve army of labour, so making the sack an effective weapon again and creating cowed and quiescent workforces and greater potential surpluses for national and, increasingly, international capital accumulation.

What was not realised was that the emergence of heavy and sustained unemployment, initially ostensibly to push short-run rates of unemployment above so-called natural rates and then let them converge on natural rates where inflation could be sustained at steady rates and accelerating rates of inflation would be things of the past, would simultaneously have such an adverse effect on what Keynes called the “animal spirits” of business people, the ultimate determinants of rates of accumulation.  Hence we have had decades in many economies in which inflation has been drastically reduced yet accumulation has been sluggish, certainly well below the levels needed to offset full employment saving and the levels achieved during the years of the long boom itself.  In those countries where this had not occurred, despised Keynesian policies have continued to be used, sometimes unintelligent ones such as those implemented, for example, during the last six years of Ronald Reagan’s Presidency in the USA and now by President Bush the Second.

Since attaining full employment by the use of fiscal policies was no longer on the agenda in the former countries and monetary policies were mainly directed at general price levels and exchange rates, contractionary forces were widely prevalent in these countries, as the politicians and their advisors waited (or said they were) in vain while the impersonal forces of competitive markets allied with monetarist rules allowed the economies to seek and find their natural rates.


I think it is fair to say that Keynes never completely threw off the vision of the working of economies in terms of an equilibrium framework.  He did, of course, argue that government intervention was needed to help attain a satisfactory full employment equilibrium (internal balance) in each economy – left alone, less satisfactory equilibria or rest states would emerge.  This was an essential step towards equilibrium associated with external balance in the international system and the possibility then to take advantage of the classical principles of free trade on which he had been brought up.  (Skidelsky (1992, xv) called him “the last of the great English liberals”.)  The proposals he put forward at Bretton Woods were designed to provide the institutions and the orders of magnitude of, for example, the provision of liquidity that would make all this possible.  That the Americans, principally thorough Harry Dexter White, won out on both the institutions and the orders of magnitude adopted for the post-war period was a tragedy; for this ensured that the Bretton Woods system contained within it the seeds of its own eventual destruction from its very inception.  (How Marx would have laughed!)

One of the major changes in vision since Keynes’s death about how markets, economies, even whole systems work, associated with Keynes’s followers, especially Kaldor and Joan Robinson, is the concept of cumulative causation.  The concept has its origins in Adam Smith (what has not?) and was brought into prominence in the modern era by Allyn Young, Kaldor’s teacher at the LSE, and subsequently championed by Kaldor and independently by Gunar Myrdal, especially in their post-war writings.  The way I illustrate the essential idea of the concept for my students is through the analogy of a wolf pack (I am not a zoologist so I may be completely wrong about how wolves behave; but as I am an economist, at least I think so, let us assume I am right).  There are two major views on the workings of markets, economies, whole systems.  The dominant one is that akin to a wolf pack running along.  If one or more wolves get ahead or fall behind, powerful forces come into play which return them to the pack.  (The parallels with the existence of an equilibrium that is stable, and that the forces responsible for existence are independent of those responsible for stability are, I hope, obvious.)  The other view has the forces acting on the wolves who get ahead or fall behind make them get further and further ahead or fall further and further behind, at least for long periods of time.  This view captures the notion of virtuous or vile processes of cumulative causation.  My contention is that, according to which view is “correct”, makes a drastic difference to our understanding of the world and how specific policies may be perceived, recommended and evaluated.

I illustrate with an example, the case for freely floating exchange rates.  A classic paper arguing for them is by Milton Friedman (1953).  Underlying his argument is the first wolf pack analogy, that in a competitive setting there exists a set of long-period stable equilibrium exchange rates that quickly would be found and then kept by a free float.  Moreover, in this setting the systemic effects of speculation would be beneficial, for speculators with their superior knowledge, intelligence and information would help the system to reach the equilibrium pattern more quickly than in their absence and then sustain it there.

But suppose that the second wolf pack analogy is the correct or at least more correct description of how foreign exchange markets work.  Then there is no set of stable equilibrium exchange rates “out there” waiting to be found and now a float combined with speculative activity will be systemically harmful, accelerating the movements away in both directions of exchange rates from one another and also of systems, at least for long periods of time.  I submit that the second scenario is more akin to what has happened over much of recent decades, and provides a rationale for various schemes suggested to curb the action of speculators.  (My own suggestions may be found in Harcourt (1994; 1995; 2001b).  I had generalised the Tobin tax proposal without, I must confess, being aware at the time of its existence!)

It is not only in markets characterised by cumulative causation processes that speculation may be systemically harmful.  Any market in which stocks dominate flows and expectations about the behaviour of other participants in the market dominate the more usual economic factors – preferences, cost of production – in the setting of prices may experience periods when speculation is harmful.  (The seminal and classic paper on this is Kaldor (1939).)  An obvious example is the stock exchange.  On this we may recall Keynes’s famous description in Chapter 12 of The General Theory of what may happen when “enterprise becomes a bubble on a whirlpool of speculation”, Keynes (1936; C.W., vol. VII, 1973, 159).


Let me close with another example of how Keynes and Keynesian/Kaleckian/Marxian ideas are still relevant for both our understanding and  policy making.  The ideas I present now are based on Kalecki’s famous 1943 paper, “Political aspects of full employment” and the writings of my two greatest Australian mentors, the late Eric Russell and the late Wilfred Salter, both devoted Keynesians, see Harcourt (1997; 2001b) for the arguments and references.

Kalecki set out graphically the vital difference between the political economy of getting to full employment after a deep slump, when all classes are in favour of this, the wage-earners in order to get jobs, business people in order to receive higher profits, the government in order to reduce the risk of serious social unrest, on the one hand, and the political economy of sustaining full employment, on the other hand.  In the second situation, as I argued above, cumulatively economic, social and political power shifts from capital to labour.  The capitalist class, indeed conservative elements generally, get more and more uneasy about the emerging situation.  An environment is created in which, for example, monetarist ideas will be well received, and more than one economist will be prepared to be a hired prize fighter in support of them as government (and central bank) actions.

Is there a possible answer to this, on the face of it, inescapable dilemma in our sorts of economies?  Keynes and his followers recognized that attaining and then maintaining full employment would carry with it cumulatively rising risks of inflationary pressures associated with rising money-wage demands.  It is no accident that Joan Robinson always said that from 1936 on, “Incomes Policy” was her middle name, a perceptive insight no doubt reinforced by having an actual middle name of Violet.  Russell and Salter recognized this dilemma and argued in Australia for a full employment policy that included an incomes policy implemented through our centralised wage fixing body (then the Australian Arbitration Commission).  In broad outline, at a starting point, money incomes were to be adjusted periodically for changes in prices and in overall productivity.  Not only is this adjustment equitable, it is also efficient. 

It is equitable because at the level of the economy as a whole, capital and labour are complements and the impact of their combined activity on overall productivity ought to be reflected in changes in the real incomes of all citizens.  It is efficient because with full employment, such an overall policy discourages low productivity, often declining industries whose time has passed and encourages high productivity, often expanding industries whose time has come.   The result is a regime with higher increases in overall productivity than would occur otherwise, certainly than would occur in a regime characterised by so-called flexible markets, such as are the UK’s and the USA’s pride and joy.  There would be therefore an agreeable quid pro quo for money income restraint in the form of rising real incomes, so providing a possible solution to Kalecki’s dilemma.  There are, of course, all sorts of qualifications and modifications and exceptions to the starting rule – I discuss these in the article referred to above.  Here I wanted to set out the core argument as starkly as possible.


In conclusion, may I say that Keynes and his ideas are still alive and well; that subsequent developments by others complement agreeably his own revolutionary contributions; and that people of good will who wish to see established just and equitable societies world-wide have in these ideas an essential starting point?



End Notes

* A lecture given at the conference on “Keynes and after”, held at the Faculty of Economics and Business Administration, University of Iceland, Reykjavik, on 10 October 2003.


1. In a letter to Keynes (28.6.1928) when he submitted the article to the Economic Journal, he wrote: “Of course the whole thing is a waste of time”.  It had distracted him from “a book on logic … [because] it [was] much easier to concentrate on than philosophy and the difficulties that arise rather [obsessed him]”


2. For some policy implications of Minsky’s insights, see Harcourt (2001a), ch. 15.


Balogh, T. (1982), The Irrelevance of Conventional Economics, London: Weidenfeld and Nicholson.

Crotty, James R. (1980), “Post-Keynesian theory: An overview and evaluation”, American Economic Review, 70, 20-5.

Friedman, Milton (1953), “The case for flexible exchange rates” in Essays in Positive Economics, Chicago: The University of Chicago Press, 157-203.

Galbraith, J.K. (1967), The New Industrial State, Boston: Houghton Mifflin Company.

Hahn, Frank and Robert Solow (1995), A Critical Essay on Modern Macroeconomic Theory, Oxford: Blackwell.

Harcourt, G.C. (1994), “A ‘modest proposal’ for taming speculators and putting the world on course to prosperity”, Economic and Political Weekly, XXIX, 2490-92, reprinted in Harcourt (2001b), 255-62.

Harcourt G.C. (1995), “Lorie Tarshis, 1911-1993: An appreciation”, Economic Journal, 105, 1244-55, reprinted in Harcourt (2001a), 114-30.

Harcourt G.C. (1996-97), “Critiques and alternatives: Reflections on some recent (and not so recent) controversies”, Journal of Post Keynesian Economics, 19, 171-70, reprinted with an Addendum in Harcourt (2001a), 289-99.

Harcourt, G.C. (2001a), 50 Years a Keynesian and Other Essays, Houndmills, Hants: Palgrave.

Harcourt, G.C. (2001b), Selected Essays on Economic Policy, Houndmills, Hants.: Palgrave.

Harcourt, G.C. and Prue Kerr (2001), “Karl Marx, 1818-83” in Harcourt (2001a), 157-68.

Harcourt, G.C. and P.A. Riach (eds.) (1997), A ‘Second Edition’ of The General Theory, 2 vols, London: Routledge.

Harris, Donald J. (1975), “the theory of economic growth: A critique and a reformulation”, American Economic Review, Papers and Proceedings, 65, 329-37.

Harris, Donald J. (1978), Capital Accumulation and Income Distribution, Stanford, California: Stanford University Press.

Kahn, R.F. (1929; 1989), The Economics of the Short Period, London: Macmillan.

Kaldor, N. (1939), “Speculation and economic stability”, Review of Economic Studies, 7, 1-27, reprinted in Kaldor (1960), 17-88.

Kaldor, N. (1960), Essays on Economic Stability and Growth, London: Duckworth.

Kalecki, M. (1943), “Political aspects of full employment”, Political Quarterly, 14, 322-31.

Keynes, J.M. (1919), The Economic Consequences of the Peace, London: Macmillan, C.W. , vol. II, 1971.

Keynes, J.M. (1921), A Treatise on Probability, London: Macmillan, C.W., vol. VIII, 1973.

Keynes, J.M. (1923), A Tract on Monetary Reform, London: Macmillan, C.W., vol. IV, 1971.

Keynes, J.M. (1930), A Treatise on Money, 2 vols., London: Macmillan, C.W., vols V, VI, 1971.

Keynes, J.M. (1936), The General Theory of Employment, Interest and Money, London: Macmillan, C.W., vol. VII, 1973.

Keynes, J.M. (1940; 1980), How to Pay for the War; Activities 1940-44, C.W. vol. XXV, London: Macmillan.

Keynes, J.M. (1973), The General Theory and After, Part I: Preparation, C.W., vol. XIII, London: Macmillan.

Keynes, J.M. (1973), The General Theory and After, Part II: Defence and Development, C.W., vol. XIV, London: Macmillan.

Keynes, J.M. (1979), The General Theory and After: A Supplement, C.W., vol. XXIX, London: Macmillan.

Kregel, J.A. (1976), “Economic methodology in the face of uncertainty”: The modelling methods of Keynes and the Post-Keynesians”, Economic Journal, 86, 209-25.

Marris, R.L. (1997), “Yes, Mrs. Robinson! The General Theory and imperfect competition” in Harcourt and Riach, vol. 1 (1997), 52-82.

Ramsey, F. (1928), “A mathematical theory of saving”, Economic Journal, 38, 543-59.

Robinson, Joan (1964), “Pre-Keynesian theory after Keynes”, Australian Economic Papers, 3, 25-35.

Sardoni, Claudio (1987), Marx and Keynes on Economic Recession, Brighton, Sussex: Wheatsheaf.

Skidelsky, Robert (1983), John Maynard Keynes, Volume One. Hopes Betrayed, 1883-1920, London: Macmillan.

Skidelsky, Robert (1992), John Maynard Keynes, Volume Two.  The Economist as Saviour, 1920-1937, London: Macmillan.

Skidelsky, Robert (2000), John Maynard Keynes, Volume Three.  Fighting for Britain, London: Macmillan.

Shapiro, Nina (1997), “Imperfect competition and Keynes”, in Harcourt and Riach, vol. 1 (1997), 83-92.

Targetti, Ferdinando and Bogulslawa Kinda-Hass (1982), “Kalecki’s review of Keynes’ General Theory”, Australian Economic Papers, 21, 244-60.  Kalecki’s review, “Some remarks on Keynes’ theory”, is 245-53.

Tarshis, Lorie (1980), “Post-Keynesian economics: A promise that bounced?”, American Economic Review, 70, May, 10-14.

Vines, David (2003), “John Maynard Keynes 1937-46: The creation of International Macroeconomics”, Economic Journal, 118, F338-F361.

G. C. Harcourt,
“The economics of Keynes and its theoretical and political importance: Or, what would Marx and Keynes have made of the happenings of the past 30 years and more?”,  post-autistic economics review, issue no. 27, 9 September 2004, article 1,