Navigation – Plan du site
Business Cycles, Money and Economic Policy

Keynes’s Misguided Revolution

La révolution erronée de Keynes
Amos Witztum
p. 287-318


La Théorie générale de Keynes n’est en rien une théorie générale. Au lieu de cela, c’est une théorie sur ceux qui sont en dehors du système. Par ce moyen, Keynes a détourné l’attention de la science économique d’une préoccupation de l’école classique : le destin de ceux qui sont à l’intérieur du système. Même s’il est vrai que dans la perspective néo-classique, les modèles concurrentiels n’ont pas besoin d’être soutenus, les économistes classiques pensaient autrement. Pour des penseurs comme Smith ou Mill, le paradigme concurrentiel présentait tant de défauts qu’il ne pouvait fonctionner sans gouvernement. En déplaçant l’attention vers ceux qui sont en dehors du système, Keynes a implicitement sanctionné le paradigme néo-classique et il a décentré le débat de la responsabilité sociale envers ceux qui sont au travail vers celle envers ceux qui en sont exclus. Prendre soin de ces derniers est certes important, mais quand c’est au détriment des premiers, c’est une erreur.

Haut de page

Texte intégral

1. Revolution in Context

  • 1 In a letter to G B Shaw, Keynes writes: “I find myself to be writing a book on economic theory, whi (...)

1Keynes considered his writings in the General Theory (GT) as revolutionary.1 It was, as it were, a rebellion against what he calls the ‘classical’ school. However, the scope of the debate which emanated from this work suggests that there might have been some difficulties in interpreting the nature of the revolution.

  • 2 Backhouse (2006) provides a good summary of the history of this debate.

2Indeed, the debate about Keynes seems to have been stretching across two extremes.2 At the one end Keynes seemed to have argued the complete abolition of equilibrium analysis. Townshend (1937), Shackle (1967) and Robinson (1974) picked up on Keynes’s own response to the debate in the QJE to claim that the contribution of the GT was to introduce time and expectations into the market mechanism of all goods. Coupled with Keynes’s insistence on the irreducibility of uncertainty by probability—and thus, a form of bounded rationality—this meant the introduction of a non-equilibrium approach to economic analysis.

3Leijohnhufvud (1968) agreed that the contribution of the GT was about market failure but not just because of time and expectations. It was a failure because of the absence of an ‘auctioneer’ and thus, the GT is not merely about non-equilibrium but rather about disequilibrium economics. Together with time and expectations this would be extended into an analysis of inter-temporal disequilibrium.

  • 3 A more localised and focused attempt at interpreting Keynes were concentrated on chapter 2 (and 19) (...)

4In turn, a whole literature of disequilibrium macroeconomics developed (Solow and Stiglitz (1968), Barro and Grossman (1971) and Benassy (1986)) but from the point of view of the methodological debate, as captured by Leijohnhufvud, this was not really a proper modelling of disequilibrium. These models were simply Walrasian models with rationing and as such they did not, in his view, capture the spirit of the GT.3

5At the other extreme, there were the less ‘revolutionary’ interpretations like Clower (1965), Patinkin (1965) and more recently Laidler (1999) where methodology was not at the heart of the discussion. For them, the main contribution of the GT was the role of money in exchange. Due to the fact that people exchange goods with money and labour with money, the presence of unemployment may trigger a failure in the co-ordination of decisions.

6The seeds of this spread of interpretations, in my view, lie in the ambiguities in Keynes’s position on the ‘classical’ school. Skidelsky (1992, 512) writes that “the validity of Keynes’s ‘general theory’ rests on his assertion that the classical theory….is, as he put it in his lectures, ‘nonsense’. If [the classical theory] were true, the classical ‘special case’ would in fact, be the ‘general theory’, and Keynes’s aggregative analysis….empty and redundant”. He then goes on to say that in the post war period, Keynes’s theory was indeed perceived as a special case of classical economics.

7But what exactly is meant by ‘classical economics’? Davidson (2007, 22) asserts that “Classical economic theory…provided the rationale for the laissez-faire or ‘no government intervention in the market place’ philosophy that dominated discussions of how to cure the unemployment problem and promote prosperity.” In his discussion he makes no distinction between classical and neo-classical economics and falsely reads Smith’s idea of the ‘invisible hand’ like most neo-classical economists.

8In fact, there are two aspects of classical economics to which Keynes is referring. Firstly it is the market clearing mechanism and secondly, the perfunctory role which the theory ascribes to money. However, his analysis does not go much beyond the role of money in plugging the gap in the labour market.

  • 4 This point is clearly stated in Chapter 24 of the GT and will be discussed further below.

9In the end, in spite of Keynes’s expressed motivation to contest the universality of what he calls the ‘classical’ conception of the labour market’s mechanism, according to the GT he has not really parted way with that school. For one, he famously agreed that once we are in full employment, the classical school “comes to its own” (GT, 378). But more importantly, in spite of the failing of the labour market (and the unnecessary high demand for liquid assets), he clearly believed that the classical model allocates all the employed resources well.4

  • 5 In spite of clear signs that Keynes wishes to overthrow the analysis which treated money as a resid (...)
  • 6 I consider the case of less than full-employment as equilibrium as well even though the labour mark (...)

10Therefore, implied by this is that Keynes did not question the principles of economic organisation embedded in the ‘classical’ tradition but has simply set out to solve a particular problem which he believed cannot be resolved by itself.5 On the one hand, this raises a question-mark over how general is his theory. On the other hand, it reduces his contribution to a mere problem of adjustment. For a text on economic theory, the GT is disappointingly short of any serious discussion of all those things which seem to have mattered to economists before and after his time: namely, the meaning of wealth, explaining prices, forms of human interactions, explaining allocations and distributions and so on. As a result, his contribution seems more like the following summary: the real problem society faces is that of multiple equilibria6 which are all, in a somewhat peculiar way, efficient. To be precise, Keynes’s views seem to imply that the competitive paradigm always produces efficient allocations with respect to those who participate in the system. As only one of these allocations would also entail full employment, the role of government—in a similar manner to that in modern welfare economics—is to find ways of shifting the economy from one decentralised (and ostensibly, efficient) equilibrium to another. Evidently, the decentralised nature of efficient equilibria suggests that whatever the government does, it must remain exogenous to the system and thus, in spite of Keynes’s use of a notion like socialising investment, he does not propose for governments to become part of the system but merely to induce the system to move in a specific direction. In this sense, he remains completely respectful of the major tenets of the ‘classical’ paradigm.

  • 7 This is not to say that Keynes’s agenda did not go well beyond the GT and when viewed in its entire (...)

11Whether or not this amounts to a real revolution is not entirely clear.7 If indeed one interprets neo-classical economists like Marshall or Walras and the real classical school a la Smith and Mill as advocating no government intervention except for the purpose of supporting competitive structures (i.e. competitive equilibria) Keynes’s position constitutes a revolution. But if one examines the classical position more carefully, this may seem rather as a distraction than a real revolution.

12Both Smith and Mill were acutely aware of the possibility that a decentralised competitive environment may produce outcomes which are socially undesirable even in the absence of involuntary unemployment. Notwithstanding the obvious role of competition in combating misguided conceptions of wealth creation emanating from the Mercantile school (in the case of Smith) and in freeing society from the stifling effects of customs (in the case of Mill), neither of these scholars had any illusions about the social pitfalls of decentralised competitive institutions. This, I believe, stands in sharp contrast to Keynes’s position. While Keynes is only concerned with those who are outside the system (the unemployed), for Smith and Mill there were grounds for concern about those who are in the system. In other words, Keynes’s position had strengthened the view according to which government should not interfere with the working of competition except for providing a guiding hand to lead it—from afar—to the state of full employment. In so doing, Keynes distracted attention from what was the real problem for classical economics—i.e. the working of the competitive paradigm—to a relative minor problem of how inclusive it was.

13The problem of involuntary unemployment was not foreign to all classical economists. For one, the wage-fund doctrine is a dynamic concept and when properly viewed over time, it is easy to see that demand for labour in one period is a result of demand for goods in the previous periods. Moreover, we find that in Mill the concept of direct aggregate demand driven unemployment exists but the cause of unemployment was not the level of the demand but its internal composition. Here too, unemployment would not have been resolved by the lowering of wages as the adjustment mechanism would only restore the average wages which could leave a labour market with some occupations fully employed and highly paid and others, underemployed and underpaid.

14However, we show that the real difficulty of the labour market in the eyes of classical economists like Mill and Smith is not necessarily the unemployed as it is the employed. Given the fundamental difference between the classical and neo-classical economic problem, there is a methodological equivalence between wages below subsistence in the classical school and unemployment in the neo-classical one. Hence, it is evident that classical economists recognised the fact that competitive markets may fail to solve the economic problem; something which has, at the time, alluded neo-classical economics.

  • 8 In modern economics, a competitive equilibrium may be inefficient if we face a problem of incomplet (...)

15As a result, we identify the presence of a ‘bad’ and ‘good’ equilibrium (as opposed to efficient or inefficient).8 In modern economics there may be many equilibria which are efficient and the role of government, from outside the system, is to move the economy from one place to another without interfering with the working of the markets. For classical economists, the move from a ‘bad’ to a ‘good’ equilibrium requires a much more involved government. We distinguish between the two by calling the former ‘exogenous’ government while the latter, ‘endogenous’.

16Moreover, as the nature of ‘bad’ equilibrium highlighted the plight of those who are employed it meant the problem of involuntary employment seems more pressing from that of the involuntary unemployment. Clearly, to resolve these issues there is a greater need for an ‘endogenous’ government. To this effect, Keynes’s contribution has deflected attention from those important insights emanating from the real classical economics. By focusing both on the unemployed and the exogenous government, he has allowed neo-classical economics to completely overshadow the agenda of classical economics which, in many ways, is the more relevant for present day societies.

2. What is General in the General Theory?

17When Keynes introduces the notion of involuntary unemployment he makes the following assertion:

The classical theorists resemble Euclidean geometers in a non-Euclidean world who, discovering that in experience straight lines apparently parallel often meet, rebuke the lines for not keeping straight--as the only remedy for the unfortunate collisions which are occurring. Yet, in truth, there is no remedy except to throw over the axiom of parallels and to work out a non-Euclidean geometry. Something similar is required in economics. (GT, 16)

  • 9 See, for instance, discussions in Togari (2001) and Galbraith (1994).

18This reference to developments in Physics (which have not yet been fully investigated9) may explain the title of the GT but it also explains some of the confusion in Keynes’s position on classical economics. From the way it is phrased here, classical economics is more like Galilean (or classical) relativity; it is not even the special relativity for which the general relativity is an extension. Hence, as Galilean relativity was based on the false assumption of an absolute time, and was, therefore, wrong, so is classical economics: wrong.

19If indeed this was Keynes’s position one would have to conclude that the move towards an analysis which is based on aggregates was not merely an attempt to go beyond microeconomics but a desire to replace it altogether. The existence of the labour market and the money market do not constitute a departure as these are anyway markets expressed in terms of aggregates.

20But one would have thought that such a tremendous shift in the focus of economics’ subject matter merited some defence or explanation, yet none has been forthcoming. Moreover, this dramatic aspiration to equate his contribution to that of Einstein is undermined by his own explicit claim when he addresses the question of the purpose of his theory. Here, unfortunately, we come across a strange ambiguity of a very different kind which is easily detected by comparing chapters 1 and 24 of the GT (to which I will return later).

21In chapter 1 Keynes makes a clear claim that the purpose of his work is to “contrast the character of my arguments and conclusions with those of the classical theory” (GT, 3) and to “argue that the postulates of the classical theory are applicable to a special case only and not to the general case” (ibid.).

22Again we have the ‘special’ case of ‘classical’ economics and the ‘general’ case of the GT. However, here the words special and general are used in a more straightforward manner. Accordingly, ‘classical’ economics is not like Galilean relativity but rather like the Special relativity. It is true in some cases but not always. Evidently, Keynes is referring here to the idea of market clearing mechanisms in the labour market which he attributes to what he calls ‘classical’ economics. It holds in full employment but not otherwise and it seems to constitute the essence of ‘classical’ economics. Consequently, ‘classical’ economics does not recognise the concept of involuntary unemployment.

23There are two issues here: firstly, there is the question of whether or not it is true that ‘classical’ economics does not recognise involuntary unemployment. Secondly, there is the question of whether the GT is really a generalisation of ‘classical’ economics (à la general and special relativity) or is it really a different theory or even not a theory at all but rather a question of policy. I will endeavour to demonstrate that it is not true that the real classical economics did not recognise the possibility of involuntary unemployment. At least in the case of Mill there is a clear discussion of it although the cause is not the level of demand but rather its composition. But unlike Marshall—who recognised involuntary unemployment in the short run—Mill does recognise that the plugging of the gap will not come from the fall in wages. All of this is remarkably close to Keynes’s argument although Keynes clearly offers a different explanation. In this respect it would therefore be difficult to see Keynes’s contribution as a generalisation. As an alternative explanation, however, it would be insufficient.

24If we examine the subject matter of economics as perceived by classical and neo-classical economists we will find that Keynes has moved very far away. It is enough to examine the table of contents of the classical books (and Marshall) to see that all of it is missing in Keynes. No more discussions about the meaning of wealth, the purpose of life, the nature of human interaction and of social organisation, the role of exchange, the organisation of markets, the role of institutions and the distributions of returns. Instead, it is all reduced into a new theory of involuntary unemployment and money.

25It is interesting to note that Keynes himself complained that classical economics does not recognise the real effects of money and it appears in residual chapters after the theory of organisation had been fully explored. His own theory could have been a generalisation if he had reversed the order and after his discussion of the real effects of money he would have returned to study organisation. However, he failed to do so.

26In short, this new theory could be quite a general theory if we accept the novelty of the effects of money on exchange and Keynes’s notion of bounded rationality. However, there is nothing in the GT (or other writings) to shed any light on how this would matter to all these questions which concerned every other economist.

3. Unemployment in Classical Economics: A Lesson from Mill

27The key element in Keynes’s complaint against classical economics is that it fails to realise that unless aggregate demand increases, a fall in nominal wages cannot rectify an excess supply of labour and therefore, produce the idea of involuntary unemployment.

28While it is true that the ‘mainstream’ of British classical economics (i.e. Smith, Ricardo and J. S. Mill) seems to be in agreement with the proposition embedded in Say’s Law, it would be slightly misleading to argue that they would therefore necessarily be unaware of the possibility of involuntary unemployment. There are three main reasons for this claim. Firstly, the real classical economists were not concerned with aggregates and secondly, in as much as aggregates appear in their analysis, they had a different conception of aggregate demand which may shed a different light on Keynes’s problem. Thirdly, they were more concerned about the poverty of those who were part of a system which promised good results than with those who were outside the system. By this, I do not mean to say that they would not care about unemployed people but if the system betrays the employed poor, what would possibly be the merit of ensuring that everyone is employed?

  • 10 I should say at the outset that the wage-fund doctrine which I attribute to J. S. Mill is based on (...)
  • 11 In his definition of capital Mill writes that it is “the accumulated stock of the produce of labour (...)

29In J. S. Mill, for instance, we can find a clear statement which very much resembles the problems which bothered Keynes. To begin with, in as much as the aggregate demand—as a determinant of employment—is concerned, it is interesting to note that Mill’s (and Smith’s) wage-fund doctrine contains Keynesian elements in them.10 It is true that many confuse (including, occasionally, Mill himself11) the wages-fund with capital and therefore, combine employment with savings. However, this is not the real case as the wages-fund includes the demand for all types of labour productive or otherwise. The demand for unproductive labour does not come out of savings and therefore, the overall demand for labour cannot be affected just by savings:

To this, however, must be added all funds which, without forming a part of capital, are paid in exchange for labour, such as the wages of soldiers, domestic servants, and all other unproductive labourers. There is unfortunately no mode of expressing by one familiar term, the aggregate of what has been called the wages-fund of a country. (Mill, [1848] 1909, 343-4, emphasis added)

30So Mill was clearly searching for the equivalent of Keynes’s aggregate demand and like Keynes he saw consumption as an important element in forming the demand for labour. It is true, however, that Mill then suggests that the demand for productive labour is the greater part of the whole and thus, he felt that equating the wages-fund with capital was fine as long as one bore in mind that it is not just capital. It seems that he himself forgot about this when he talks about savings and employment.

  • 12 And even from this we must deduct what is called fixed capital.

31Nevertheless, the question that arises is whether or not there is a similarity between the wage-fund and aggregate demand even if we accept the main business of society to be based on what Mill calls productive labour (i.e. labour that produces commodities). Mill insists that it is not the demand for those commodities which labour produces that is the source of demand for labour (Mill, [1848] 1909, 66-8). Instead, it is the capital—that part of the wage-fund directed at production of goods12—available which is the only source of demand.

32However, we must bear in mind that classical economics was a dynamic rather than a static system. Therefore, the question is whether consumption has anything to do with the amount of capital available. What is it, one may wonder, that determines the size of the wage-fund and consequently, the demand for labour?

33Suppose that we have a wheat growing economy where the landlord and the capitalist are the same person. The technology is such that one needs αW of wheat to produce W tonnes of wheat (for simplicity sake this includes the subsistence of the landlord/capitalist). Suppose too that the amount of labour required for the production of one unit of wheat is λ and the amount of subsistence (in wheat per worker during the period of production) is ω. Suppose that at the end of period t, the wage-fund available would be Wt (assuming that the cost of production for this year’s output came from last year’s wage fund).

  • 13 A complete analysis of Smith’s model can be found in Witztum (2010). In Mill’s recantation he says (...)

34Now, as Smith has clearly recognised in his own writings and Mill realised in his recantation, the capitalists (or anyone else for that matter) is torn between present and future pleasures.13 The question now is what will be the demand for labour in the next period? According to Mill it would be the whole fund which would become the demand of labour in either a form of demand for productive labour (i.e. circulating capital) which is the demand for those who are engaged in the production of wheat or, in the form of unproductive consumption through the hiring of unproductive labour (butlers, servants, etc.). Notice however, that there is another form of unproductive consumption where people consume past wheat directly. As such, this would reduce the size of the fund available for demand of labour. Indeed, Mill acknowledges this point when he deducts the fixed capital from the fund available to the hiring of productive labour but fails to acknowledge a similar situation on the unproductive side of consumption. This point, however, he does acknowledge in his recantation.

35Let γ be the proportion of the wage-fund which people wish to direct at luxurious consumption and β of it, they wish to consume directly (i.e. not through the hire of unproductive labour but by throwing wheat-eating parties). Given that these represent present indulgence they may take precedence over decisions about the future. Hence, the amount of wheat devoted to the hiring of labour in the next period would be:

36Kt+1= Wt (1 - βγ)

37Where the demand for productive labour is Kt+1P= Wt (1 - γ) and the demand for unproductive labour is Kt+1UP= Wt γ(1 - β)

38For the economy to reproduce, Kt+1P= Wt (1 - γ) ≥ (α + λω) Wt. This, however, depends on (1 - γ) ≥ (α + λω). The left hand side is comprised of behavioural parameters and the right hand side of technological parameters. There is, therefore, nothing to ensure that the economy would grow. If present enjoyments among the owners of the surplus are stronger, the left hand side would be very small and this would mean that the economy would hire too few productive labours to even reproduce. This, in turn, would mean that the wage fund of the next period will be smaller and as a result, the total demand for labour will also be smaller. In such a case, the fall in demand for labour in (t+1) is due to the composition of demand in period t. Moreover, even if the economy is capable of reproducing itself, an increase in demand for direct luxurious consumption (β) will reduce the overall demand for labour. So the composition of aggregate demand may have both an immediate and a lagged effect on the demand for labour. This means that the composition of aggregate demand may reduce the demand for labour but we are still left with the question of whether this would cause unemployment or merely a decline in wages.

39Indeed, Mill was very much aware of the difficulties of adjustments. When he argues that wages are determined by the ratio between the wages-fund (aggregate demand) and the number of workers seeking employment, he qualifies this statement:

The demand for labour in any particular employment is more pressing and a higher wages are paid, when there is a brisk demand for the commodity produced; and the contrary when there is what is called stagnation; the workpeople are dismissed and those who are retained must submit to a reduction in wages; those in these cases there is neither more nor less capital than before. (Mill, [1848] 1909, 344, emphasis added)

40Mill is saying here two things which are relevant to Keynes. Firstly, he recognises that even for a given stock of capital (aggregate demand), there may be some conditions where the individual components of the aggregate demand would be different for the same whole. The demand for one commodity may slack. “A manufacturer” writes Mill, “finding a slack demand for his commodity, forbears to employ labourers in increasing a stock which he finds it difficult to dispose of”(ibid.). This means that for any given wages-fund, it is still possible that the demand for some commodities would fall. This, he believes would lead to a fall in demand for labour and the dismissal of workers while those who remain will experience a decline in wages. Notice here that labour is dismissed and wages are falling yet labour is not re-instated. Its re-allocation is not the mechanism of adjustment.

41The reason for this is that Mill recognises the specificity of the labour supply and what he talks about is the labour in a specific occupation which would become unemployed even though there may not be any changes in the average wages which would reflect the ratio between the total wages-fund and the total supply of labour.

42Indeed, Mill recognises that this would not be a long run situation as the capital made idle in the declining industry would be employed in a rising industry and will thus increase the wages of this particular occupation. However, note that this does not mean that employment would be restored, only that the average wages would return to its original level. Hence, Mill was well aware of the possibility of involuntary unemployment except that in his case, this is not a failure of the level of aggregate demand but rather of its composition.

43It is interesting, and curious, to note that Marshall too discussed the specificity of labour in a very similar way and notes that “[t]he vertical movement from one grade to another is seldom very rapid or on a very large scale” (Marshall, [1920] 1952, 217). However, Mill is far nearer to Keynes by saying that the adjustment mechanism would be the movement of capital rather than labour while in Marshall, he does believe that in the long run, the shift across grades of occupation would be sufficient to plug gaps in the disaggregated labour market.

44Hence, while it may be easy to see how Keynes’s contribution is different from the neo-classical analysis of the labour market, it is less so when viewed from the real classical perspective. Neither the idea that demand (at least in composition if not in levels) matters in the labour market nor the failure of the clearing mechanism were ideas which were remote or foreign to classical economists. Clearly, the novelty of Keynes’s analysis is in his emphasis on this aspect of the analysis. So to see whether or not his analysis is indeed the claimed generalisation we should now examine a bit closer the economic problem which Keynes wished to solve.

4. The Subject Matter of Economics in the General Theory

45We already noted a certain ambiguity in Keynes’s conception of the economic problem which underlies his theory. On the one hand he seems eager to overthrow ‘Euclidean’ classical economics with an entirely new economic theory, on the other, however, he only wishes to correct some of the axioms of classical economics which only hold in special cases. But, it is not until chapter 24 of the GT that Keynes reveals the real objectives of his theory:

The outstanding faults of the economic society in which we live are its failure to provide for full employment and its arbitrary and inequitable distribution of wealth and incomes. (GT, 372)

46Here Keynes clearly states a purpose for his economic analysis even though the title of the chapter suggests a reverse causality (from the analysis to the objective). But while unemployment might have been a trigger, could it have generated a real new general theory of which unemployment is only one of the issues?

47If one were to summarise the theory of the GT, —inevitably in a slightly crude manner though sufficiently accurate—one can say that Keynes’s GT suggests that the presence of money in the economic system means (a) that the clearing mechanism of the labour market may falter and therefore, the only way to plug the gap is through an increase in demand for labour; (b), that the increase in demand for labour can be directly derived from an increase in aggregate demand; and (c), that the exaggerated demand for money keeps interest high and therefore allows the gap in the labour market to persist.

  • 14 Which is also consistent with the idea that in full employment the classical school is coming to it (...)

48So in spite of Keynes’s discussion of human motivation and the idea that money may have real effects it is difficult to say that Keynes’s theory is about anything else other than unemployment. While in his other writings the role of money in exchange is important, in the GT it seems that had the mechanism of the labour market not failed, money would have only influenced prices rather than output but this had already been known since the days of Hume.14 In fact, Keynes’s analysis is focused on two markets alone, the labour and money markets, and the relationship between them. He also believes that as the labour market is where the wages are determined and the money market is where interest rate is determined, the analysis of these two markets will provide insight into the problems of distribution. However, whether or not there is an equitable distribution of earnings according to individuals’ contribution is beyond the scope of his work.

49While Keynes does offer an insight into what motivates people, his analysis lacks one of the most important elements in economic analysis: the question of interaction and organisation. If people behave the way he describes (both in their aggregate demand for goods and their demand for money), how would this affect the major claim attributed to classical and neo-classical economics that competition is the best means to co-ordinate the behaviour of individuals?

50It is not impossible to treat Keynes’s theory as a novel general theory but for this to be true we must perceive the aggregate model he proposes as an alternative to microeconomic analysis and not its complement. My feeling is that this is not how Keynes saw his own contribution. To claim, as did a few of his followers, that Keynes’s contribution is embedded in his the idea of bounded rationality where the presence of money leads to a break down in equilibrium analysis, should have led to a novel and general theory of markets and co-ordination. But this is not what Keynes has done.

51In fact, as far as the allocation of resources, market interactions, and distribution are concerned Keynes stays firmly with the ‘classical’ school. He has not really abandoned microeconomics:

To put the point concretely, I see no reason to suppose that the existing system seriously misemploys the factors of production which are in use…..When 9,000,000 men are employed out of 10,000,000 willing and able to work there is no evidence that the labour of these 9,000,000 men is misdirected. The complaint against the present system is not that these 9,000,000 men out to be employed on different tasks, but that tasks should be available for the remaining 1,000,000 men. (GT, 379)

52Coupled with the idea that in full employment, the ‘classical’ theory ‘comes to its own’, what we have here are two points of equilibrium, or, perhaps, two allocations generated by competitive institutions. In one of them the labour market clears and in the other, it does not. However, in terms of allocating employed resources, they both seem to be working equally well and in exactly the same manner. Yet in the one case, some people remain outside of the system while in the latter they are all included.

53So the focus of Keynes’s attention is not the organisation of economic activities because these are served well by the recommendations of (neo-)classical economics. He is also not particularly concerned with the effects which the two types of situations in the labour market would have on the efficiency of competitive organisation. Instead, it is simply to shift the economy from a position where the domain of the classical model includes only a subset of the labour force to a domain where it includes the whole labour force.

54This means that Keynes did not really intend to break away from ‘classical’ theory or, for that matter, to offer any alternative to its thinking. Instead, through the complex conception of aggregates and money, he proposes that it is up to the government to help shift the economy from one equilibrium to another.

55As Keynes seemed to have kept the bulk of ‘classical’ economics and chose to focus on unemployment via the influence of money on the system, his theory was certainly new but was in no way general. It did not generalise our understanding of how resources are allocated nor did it add anything to our understanding of how returns are distributed, nor, for that matter, do we learn anything about the nature of exchange or institutions.

56Marshall, for instance, thought that “Economics is a study of mankind in the ordinary business of life; it examines that part of individual and social action which is more closely connected with the attainment and with the use of material requisites of well-being” (Marshall, [1920] 1952, 1). More specifically he says: “But, for all that, the steadiest motive to ordinary business work is the desire for the pay which is the material reward of work.”(Marshall, [1920] 1952,14)

57In this respect, Marshall coins the basic principle underlying neo-classical economics: the desire to maximise material well-being. It is, however, perceived from an individualistic perception which means that the economic problem which modern society faces is how to organise its activities in such a way so that all these rational wants may coincide.

58To some extent, one may say that Keynes is very much part of this general interest in economics as material well being will never be maximised in the presence of involuntary unemployment. But this is, of course, quite wrong. Full employment may not deliver the highest level of output if the mechanism of resources allocation and distributions of returns is flawed so that society is not making the best use of its available resources. Naturally, if Keynes accepts the working of the neo-classical model as part of his system then it is true that output will be greater in full employment but this would mean that his theory is merely tweaking at the edge of the neo-classical paradigm. It may be novel but by no way general. Instead, it is as if he accepts that ‘classical’ economics is general and he only offers a means of ensuring that it achieves the maximum output.

59If, however, we relate this agenda to the economic problem with which real classical economists were dealing then the story would be very different indeed. For classical economics, where the problem was that of growth maximisation, a decline in the labour force is the equivalent methodological infliction to that of involuntary unemployment.

4.1 A View from J. S. Mill

  • 15 While it is true that Mill believes that workers are partly to blame for the competitive low wages (...)

60For J. S. Mill, this was an endemic problem of the competitive system. “In this country” he writes about England, “there are few kinds of labour of which the remuneration would not be lower than it is, if the employer took the full advantage of competition” (Mill, [1848] 1909, 343). In other words, if competition ruled supreme, the wages would have been so low that even employers feel that they cannot be so mean and deprive workers from what may be considered as a “reasonable wage”.15 This means that competitive wages would be below subsistence and while in the long run this may raise the average wages due to a fall in workers’ number, it is an immoral position to take with regard to the current labour force.

61For Mill, competitive equilibrium, if left to itself, is a social evil for parallel reasons which prompt Keynes to write his treaties. It arises from concerns for the unemployed (in Mill’s case, to those who earn wages below subsistence) and the fairness of the distribution of income. He writes:

If the choice were to be made between Communism with all its chances, and the present [1852] state of society with all its suffering and injustices; if the institution of private property necessarily carried with it as a consequence, that the produce of labour should be apportioned as we now see it, almost in an inverse ratio to the labour —the largest portion to those who have never worked at all, the next largest to whose work is nominal, and so in a descending scale...; if this or Communism were the alternative, all the difficulties, great or small, of Communism would be as dust in the balance. (Mill, [1848] 1909, 208)

  • 16 See a full discussion in Witztum (2005).

62Obviously, that which concerns Mill in the working of competition in a world of private property is the fact that people are not remunerated in relation to their contribution. The reason why this is sufficient to make him discard of this form of organisation is that it violates the freedom of individuals. Not to be able to enjoy the fruits of one’s labour, or faculties, is an impediment to the development of individuality.16 It is almost the same as current justifications of inequality, which claim that not to allow entrepreneurs to enjoy the benefits of their innovations will stifle innovation.

63The question, which immediately arises, is whether this is due to the unfortunate distribution of property or that it is inherent to such a system. Allow me to quote at some length:

That all should indeed start on perfectly equal terms is inconsistent with any law of private property: but if as much pains as has been taken to aggravate the inequality of chances arising from the natural working of the principle, had been taken to temper that inequality by every means not subversive of the principle itself; if the tendency of legislation had been to favour the diffusion, instead of the concentration of wealth…; the principle of individual property would have been found to have no necessary connexion with physical and social evils. (Mill, [1848] 1909, 209)

64In simple terms this means that the only way for private property and competition to be consistent with the principle of justice is through active interference in the natural distribution of wealth and income to which it gives rise. An interference, which will not only be concerned with the re-distribution of wealth but first, and foremost, be concerned with the distribution of income: “To judge of the final destination of the institution of property, we must suppose everything rectified which causes the institution to work in a manner opposed to that equitable principle, of proportion between remuneration and exertion…” (ibid.).

65So in the case of Mill, the ‘bad’ equilibrium is equilibrium of ‘laissez-faire’. It is equilibrium without government. In such equilibrium, not only is reward inversely correlated with effort but wages would normally go down to below subsistence levels. The reasons why this equilibrium is ‘bad’ are both equity and efficiency. The possible dwindling of the labour force due to wages below subsistence means lower growth. But low wages (as the ultimate notion of inverse relationship between effort and reward) also mean the lack of personal development. This, according to Mill prevents people from developing their individuality and thus, prevents them from becoming really free.

  • 17 Mill argues that there can be no ‘consumer sovereignty’ when it comes to education and by implicati (...)

66The ‘good’ equilibrium would be the one where government is active in the system. But this activity does not come from manipulating demand (either directly or through monetary policy) but rather through active involvement in the distribution of ownership and income, and the provision of education which is not provided by any market mechanism.17

67In spite of Keynes’s initial claim that he is concerned with the ‘arbitrary and inequitable distribution of wealth and income’ the only thing he can say about taxation is that his ‘new’ theory would allow the government to use taxes without worrying too much about the effect it may have on savings (GT, 372). In other words, unlike Mill, Keynes’s government is very much outside the system while Mill’s government has to be more active to ensure that the ‘good’ equilibrium emerges.

4.2 A View from Adam Smith

68But while Mill was well known for his ‘socialist’ inclinations, no one can blame Adam Smith for such leanings. Yet, even the ‘father’ of Anglo-Saxon ‘laissez-faire’ seemed to have come up with the two types of equilibrium and he too, considered government involvements as crucial.

  • 18 The progressive state is in reality the cheerful and hearty state to all different orders of socie (...)

69In Smith’s case we have an economy where people exchange surpluses. The collective surplus of the economy is what Mill calls the wages-fund which, as we noted earlier, really corresponds to Keynes’s aggregate demand. However, Smith does not take an aggregate perspective. For him, there are three possible outcomes for competitive interactions: growth, stagnation or decline.18 In terms of the methodological equivalent to involuntary unemployment, clearly decline is the case where the equilibrium fails. That which identifies the symptoms of either state is the reward for labour. “The liberal reward of labour” writes Smith, “as it is the necessary effect, so it is the natural symptom of increasing national wealth” (Smith, [1776] 1976, 91). Also note, “[n]o society can surely be flourishing and happy, of which the far greater part of the members are poor and miserable. It is but equity, besides, that they who feed, cloath and lodge the whole body of people, should have a share of the produce of their own labour as to be themselves tolerably well fed, cloathed and lodged.” (Smith, [1776] 1976, 96)

  • 19 See a lengthy discussion in (Smith, [1776] 1976, 378).
  • 20 For a full exposition of the argument, see Witztum (2009a).

70But what is it that determines these outcomes? According to Smith there are two main usages of the surplus, the demand for productive and unproductive labour. As all industries are dependent upon each other19 there is a need to co-ordinate the demand for productive labour. It is easy to show that all three possible states can emerge as an equilibrium outcome.20 Both the stationary and the declining cases can be seen as the methodological equivalents to the Keynesian problem. Here, in all three cases the competitive allocation works. But in the stationary and declining cases it means that wages are at subsistence level or below which means that the labour force will not grow and the economic problem would not be resolved.

71This, of course, leads Smith to expect governments to interfere. But the kind of interference which he has in mind is far more focused on ‘endogenous’ than ‘exogenous’ activities. By this distinction I refer to those kind of intervention which interfere in the working of the system (like government production, corporate rules, taxation etc), which I call ‘endogenous’ and those which are external to the system (like lump-sum taxes or monetary policy) and propose to shift the system as a whole without affecting its internal allocations. These latter interventions, I call, ‘exogenous’.

  • 21 See a discussion in Witztum (2009b).

72The mere division of book 5 of the Wealth of Nations between those duties of the sovereign which are more or less ‘exogenous’ (defence and justice) and those which are endogenous suggest the importance of the latter. Only 34 pages are devoted to the ‘exogenous’ rules while 90 are devoted to the ‘endogenous’ ones. These ‘endogenous’ duties begin with general requirements of directly building and maintaining infrastructures, continues with the need to provide support for special industries (like, for instance, to help facilitate trade when there is a greater risk involved but there are social benefits from having the trade21) and ending with education. In a famous passage Smith decries the terrible effects of excessive division of labour which is the engine of growth in commercial societies:

The man whose whole life is spent in performing a few simple operations, of which the effects too are, perhaps, always the same…has no occasion to exert his understanding, or to exercise his invention … [he] generally becomes as stupid and ignorant as it is possible for a human creature to become … His dexterity at his own particular trade seems, in this manner, to be acquired at the expense of his intellectual, social, and martial virtues. But in every improved and civilised society this is the state into which the labouring poor, that is, the greater body of the people, must necessarily fall, unless government takes some pains to prevent it. (Smith, [1776] 1976, 782)

  • 22 Another excellent example to Smith’s clear preferences to state action as opposed to spontaneous pr (...)

73Very much like Mill we see an emphasis on the provision of education which is non-market driven as well as regulation to ensure that people do not end up in this de-humanising state to which commercial society drives its agents.22 These trends which Smith identifies have much greater significance to growth than Smith is willing to explicitly admit. As knowledge is an important element in innovation, Smith argues that in early stages, each individual knew a lot but collectively there was not as much knowledge as there is in commercial societies. However, in the latter, each individual—due to extreme specialisation—knows relatively little and innovation, then depends on those who specialise in taking an overview of knowledge. This means that there must be some form of mechanism to allow those who specialise in knowledge to transmit it to specific applications. It is not entirely clear that there is an obvious market mechanism for it. As such, again like in Keynes, society needs to move from one equilibrium to another but this can only be done by government’s direct actions.

74Therefore, in spite of the bravura of revolutionary zeal, in comparison with the real classical economists Keynes seems to generally accept that competitive markets allocate resources well and solve the economic problem. There is simply a small difficulty with the mechanism of one market which can easily be rectified. In this respect, Keynes has actually betrayed the agenda of the real classical economists and played into the hands of neo-classical economics.

5. The Role of Government: Exogenous vs Endogenous Involvement

75In the previous section we have seen how narrow and specific Keynes’s GT is in comparison with the broad agenda of the real classical economists. But to some extent, this may be an unfair accusation. While Marshall pays tributes to J. S. Mill, between Mill’s time and Marshall’s, there has been a profound change in the focus of economic analysis. The slow formation of the neo-classical school from the marginal revolution of 1870s to Marshall has seen a shift from the unease which classical economists felt towards the working of natural liberty (i.e. competitive equilibria) to an almost blind confidence in their efficiency. This has certainly surfaced in the Lausanne school and also, in a distinctly different manner, in the rise of the Austrian school, but is equally present in the more partial equilibrium analysis which dominates Marshall’s work.

  • 23 See a discussion of Pigou’s roots in Sidgwick rather than Marshall in Medema (2009, 59-60).

76So, one may argue, by the time Keynes got to conceive his theory, all the doubts which emerged from the writings of the likes of Smith and Mill have disappeared. But this is not entirely true. Pigou, who inherited Marshall’s chair in Cambridge in 1908, was clearly of the view that there is such a thing as market failure. Namely, the competitive system may produce outcomes which are socially undesirable.23 While not exactly the same, this is as near as neo-classical economics can get to the classical distinction between a ‘good’ and ‘bad’ equilibrium.

77As we said earlier the difference between the classical distinction and the neo-classical one is that in the case of market failure, the inefficiency of the system is due to a disturbance to the working of competition while in the case of a ‘bad’ equilibrium there is no clear disturbance. For instance, in the absence of complete markets (which is what Pigou was beginning to formulate) the competitive outcome will be inefficient. However, if we find ways of removing the problem, efficiency will be restored. Thus, in a world of certainty, merely allocating property rights could, in principle resolve the problem. But in the case of classical economics, the regressive state is not triggered by any market failure. It is simply the result of the usage to which the owners of the surplus (the capitalists) put it. So in modern term this would mean that it is a problem of taste parameters.

5.1 Exogenous and Endogenous government intervention

78The significance of this distinction is that it proposes very different roles for government. A market failure still means that the competitive world would be efficient. Therefore, whatever the government proposes to do it must endeavour to emulate the working of the competitive system. In the case of a ‘bad’ equilibrium, government interference cannot be based on the working of the markets as these are clearly failing. This is yet another distinction to the role of government which I would like to add to the one already introduced, that of exogenous and endogenous interventions.

79As an example let us examine the case of externality (or missing market). In such a case, the inefficiency of the outcome is due to a missing market. As this is just a problem of inefficiency, it means that removing the obstacle for perfect competition to work would resolve the problem. This is clearly an approach based on the presumption that the world of perfect competition is indeed efficient. Therefore, whatever the government does it must not interfere in the internal relationship of the system. Here, there are two options for government intervention. The first, the exogenous approach, could be based on Coase’s Theorem. In such a case, the government may simply allocate the relevant property rights and allow the system to sort itself out by competitive means. It is an exogenous intervention as the government has not interfered directly with any of the mechanisms of resources allocation (i.e., the markets).

80An alternative approach would be to interfere in the working of the allocation mechanism (i.e. the market). Here, however, our second distinction comes into effect. As the benchmark for government activities is the model of perfect competition, the government may interfere in the system by emulating the working of the markets. Hence, if the competitive efficient outcome (where social and private costs are the same) requires the price of the under-priced good to be higher, the government can levy a Pigovian tax which, ostensibly, adds costs to the industry such that the new equilibrium price will be the one that would have emerged under the appropriate allocation of property rights.

81This second approach is clearly an endogenous intervention but it is based on the presumption of the efficiency of competitive equilibrium. If, however, one considered the competitive equilibrium—without any impediments—as socially inferior, government’s intervention would have to be endogenous but not at all based on emulating the market.

82In Keynes we have a slightly more complicated problem. On the one hand, there is a market failure in the labour market where the mechanism of adjustment is not working. On the other hand, we have an equilibrium in the liquid assets market which reflects an unreasonably high demand for liquid assets.

83While the former is clearly a market failure it is not quite clear that the latter to can be so defined. It is not really the case that unless demand for liquid assets is tempered the economic problem will not be resolved. In the end, it is not only through interest rate that the gap in the labour market can be plugged. Hence, we shall not deem the exaggerated demand for liquid assets as a market failure but merely focus on the labour market as the problem.

84If we opted for an endogenous approach then government should have interfered in the mechanisms of allocation, namely in the market directly. In this case, it could have, in principle, subsidised wages so that demand for labour would increase to the equilibrium point. Again, as the objective is to lead the economy to the competitive equilibrium, acting in this manner represents not only endogeneity of government intervention but also the market emulation approach.

85In Keynes, this becomes a bit complex. While it is true that through the money market the government (through the central bank) has become an actor in the economic system, the money market itself is exogenous to the real economy. There can be little doubt that Keynes thought that money should have broad implications for exchange but he has not clarified how this will work when classical economics comes to its own (i.e. in full employment). Hence, from the perspective of economic organisation, the money market remains outside the real economy as its only significant impact is on the labour market.

86Therefore, even though the actions of the government may be conceived as an endogenous act given that the government is part of the market, it is not really so from the perspective of the real economy. Hence, I would argue that in Keynes, the government intervened in an exogenous manner.

87To be precise, the way that the government interferes in the Keynesian story is more akin to what happens in modern welfare economics. Here, with the assistance of the two welfare theorems we can provide a good example to what we mean by exogenous government. The two welfare theorems tell us that all competitive equilibria are efficient and that we can reach of all them without interfering with the working of the markets provided that lump-sum taxes were feasible. Together with a perceived agreement that that which society may choose as the desired allocation must also be efficient, the social function of government is to steer the economy from an actual equilibrium towards the one which is socially desirable without disturbing the working of competitive markets. By the use of lump sum transfers which do not influence any relative price, the government can shift the system from the outside. The moment the government has to use other forms of taxes (our second best) the government becomes endogenous but still with an eye to emulate the market (through the principle behind Ramsey pricing).

88Keynes too, proposes something which is very similar to this exogenous idea of government. The working of the government should not interfere with the working of the allocation system but merely shift the same mechanisms of allocation to a different level of employment.

89The reason I suggest that this is a distraction is based on the distinction we have drawn between the ‘bad’ and ‘good’ equilibrium in the real classical economics. Here, as the presumption that the competitive paradigm, uninhibited, works, no longer holds, it is quite possible that there will be no impediment to competition and the outcome would not solve the economic problem. We saw this clearly in the case where decision by the owners of the surplus (and thus the wage fund) would cause the wages of labour fall below subsistence and as a result lead to a decline in the labour force. Here, there are no market failures but rather a failure of the markets. Therefore, if the government wishes to bring about a change, it will have to become part of the system both in terms of taxation and in terms of direct provisions. This is particularly true in the area of infrastructure and education. More importantly, however, the market will not only have to endogenise its activities but as the competitive model does not work, its policy should not reflect attempts to emulate the markets.

5.2 Involuntary Employment and Endogenous government

90While classical economists are generally critical of the working of natural liberty, there are different issues which are raised in their respective writings. Both Adam Smith and J. S. Mill shared the view that the distribution of income resulting from the free reign of self-interested behaviour is inconsistent with good morals. They have also noted the degrading effects which excessive competition may have on those who participate in it.

  • 24 See a full discussion in Witztum (2010).

91In Smith, the story is motivated by a social explanation to why it is that people end up specialising and trading. In his narrative, this is due to the fact that they are social being who seek social approbation through, initially, an exchange of gifts which finally leads to general specialisation and trade and a situation where material wealth replaces the more obvious means of acquiring social standing.24

92This is a very different narrative to the neo-classical one where specialisation and trade is a solution to the individual’s problem of maximising his, or her, material well being. As people specialise and trade they take a risk as the return is uncertain.

93All of this leads Smith to conclude that from a social point of view, the natural distribution of things which results from specialisation is a problem. One way to overcome the difficulties is by pursuing growth maximisation. Namely, if anything, growth is closely associated with the improvement of the lot of the lower classes rather than with the objective of creating a lot of material wealth:

Is this improvement in the circumstances of the lower ranks of the people to be regarded as an advantage or as an inconveniency to society? … No society can surely be flourishing and happy, of which the far greater part of the members are poor and miserable. It is but equity, besides, that they who feed, cloath and lodge the whole body of people, should have a share of the produce of their own labour as to be themselves tolerably well fed, cloathed and lodged. (Smith, [1776] 1976, 96)

  • 25 And I emphasis again that this holds in spite of claims which some may make about the more complex (...)

94Hence, unlike modern economics where the social problem seems to be an extension of the individual’s problem,25 in Smith, the economic problem seems to have been devised to offset some of the consequences of the natural development of specialisation and trade. Hence, the obvious role of government is to uphold the social interests of society by working against the rules of nature (the competitive model).

95So what kind of government activities would be required here? The most immediate one is to facilitate growth. This may mean direct government interference in the process of knowledge creation and dissemination. As I pointed out earlier, when society progresses, according to Smith, more and more people know a lot about very few things and the only way for society to benefit from this would be by facilitating the development of those who can integrate this knowledge.

  • 26 Again this point is fully explored in Witztum (2010).

96But this is not enough. There is an additional problem which emanates from the relationship between market prices and specialisation. Smith distinguished between natural prices and market prices. Both are equilibrium prices but the difference between them is that only at the natural rate would people become secure in their specialisation. At other prices, people may have to change their vocation.26

97If we assume that the ability to specialise in that in which we feel we are good at is part of our well-being then only at natural prices would the equilibrium support such a happy state. In other prices, competitive equilibrium would force people to change their specialisation. I would like to call such a situation: involuntary employment. This is, in my view, a much worse infliction than involuntary unemployment. Smith did not pursue this point to the full but if the government were to work to support natural price equilibrium, it may need to interfere with the relative prices of the economy. This could either be done through corporate rules which offer protection to those who have acquired a specialisation or through direct intervention by absorbing those who are deemed, by Keynes, as frictionally unemployed.

98Mill too felt the same about the fate of the employed:

If the bulk of the human race are always to remain as at present, slaves to toil in which they have no interest, and therefore, feel no interest—drudging from early morning till late at night for bare necessaries, and with all the intellectual and moral deficiencies which that implies—without resources either in mind or feelings—untaught, for they cannot be better taught than fed; selfish, for all their thoughts are acquired from themselves; without interests or sentiments as citizens and members of society, and with a sense of injustice rankling in their minds, equally for what they have not, and for what others have… . (Mill, [1848] 1909, 373)

99So the real problem of commercial societies is not so much the unemployed as it is the employed. Both Smith and Mill felt that individuals should earn more than their subsistence as growth requires this as well as the development of their human capacity. Mill thought that in the process they will all become more co-operative and realise the folly of material wealth which would remove the need to continuously grow to offset the harms created by competition.

  • 27 See a discussion in Witztum (2005).

100Judging by Smith it is interesting to note that in the early stages of the debate about the benefits of self-interested behaviour and natural liberty, the idea of individual development is less significant. The difficulties of the system of natural liberty is that it produces distributions which are morally unacceptable except when prices are at their natural rate. So here the human character is unchanged but at any equilibrium other than the natural rate, most individuals will not be employed in that which they would have liked to do. As we move forward to Mill we discover that the issue of character development (Ethology) becomes crucial. The complaint which Mill has with respect to income distribution other than it offends the basic principles of ethics (in the sense of the absence of proportionality)27 is that it may not allow individuals to develop their character which is, according to him, part of their human rights. When this character develops, Mill is convinced that a more co-operative streak will emerge. This will lead to a significant change in the nature of human interaction which will also influence economic and social institutions. The role of government here is to facilitate this change by first ensuring that individuals have the conditions to develop their individuality as well as allow the emergence of co-operative institutions. Mill does not pursue this point much but it is evident that for such a transition to happen, government will have to step in to allow the transition from competitively organised institutions to the more co-operative ones.

101The point is picked up by Marshall who, in spite of his neo-classical credentials, seems to carry further the idea of individual developments. For instance, in book 4 chapter 12 Marshall discusses at considerable length the connection between individual character and economic organisation. In his discussion of Joint-Stock companies he notices that in spite of their ability to “expand themselves without limit when the work to which they have set themselves offers a wide scope; and they are gaining grounds in nearly all directions.” (Marshall, [1920] 1952, 303), they also have a “great source of weakness in the absence of adequate knowledge of the business on the part of the shareholders” (ibid.).

102Smith too noticed this difficulty and felt that Joint-Companies are not good for the economy. Mill did see a benefit to such corporations but qualified it by their ability to ensure that the moral fibre of those who work in such corporations is sufficiently strong for them to succeed. Marshall, on his part, believes that this is a question of trust. “There is every reason to hope” he writes, “that the progress of trade morality will continue, aided in the future as it has been in the past, by diminution of trade secrecy and by increased publicity in every form; and thus collective and democratic forms of business management may be able to extend themselves safely in many direction.” (ibid.)

103However, if the ability of people to care for other people’s money (to use Smith’s language) is improving, there is no particular reason why such a development should be confined to private enterprise or joint-stock companies. Indeed, Marshall is quite clear that such a development would be useful for the development of co-operatives or profit-sharing enterprises as well as for governmental enterprises. “The same” he carries on writing, “may be said of the undertaking of Governments imperial and local; they also may have a great future before them” (Marshall, [1920] 1952, 303).

104Groenewegen (1995) documents considerable amount of evidence in support of Marshall’s ‘statism’. Not only that but, “Marshall’s tendency to socialism” he writes, “was unashamedly Millian” (Groenewegen, 1995, 148). Which means that the vision of co-operate activities which supersede competition and in which there is no obvious reason to exclude government was something with which Marshall could easily agree. Groenewegen provides evidence to Marshall’s concerns about poverty, distribution of income and the need for the government to interfere. He concludes that Marshall’s disaffection with laissez-faire as the organisation principle of the economy can be summarised, as is done in Pigou, by “let the State be up and doing” (Pigou, 1925, 336).

105It is, perhaps, interesting to note that in Appendix A of the Principles Marshall offers, all of a sudden, a support to considerable government participation in the markets. “In Germany” he writes, “an exceptionally large part of the best intellect in the nation seeks employment under Government, and there is probably no other Government which contains within itself so much trained ability of the highest order” (Marshall, [1920] 1952, 753). Therefore, “[t]he control of industry by Government is seen in its best and most attractive forms in Germany” (ibid.).

106However, it is true that Marshall qualifies this statement and suggests that the large role of government in the market may be more suitable to the German character than to the English one. Yet he claims that subject to this proviso there is still a lot to be learnt from the German experience.

107Altogether, though, it seems that for Marshall, the idea of development does not necessarily lead to less government. The same obstacles which exist in the management of large corporations exist in government enterprises. Therefore, once the development of human character resolves the problems in the former, it also resolves them in the latter. In terms of our terminology such development which would also lead to government involvement in areas of industry would constitute the ultimate form of endogenous government involvement.

  • 28 See an account in O’Donnell (1992).

108Keynes must have been aware of this line of thinking in Marshall. In the sketches for an Essays on the Economic Future of the World Keynes counts issues like ‘prevention of great fortunes’, ‘theoretical socialistic framework’ and ‘psychology of reward and incentives’.28 But he nevertheless seems to remain committed to the idea of decentralised competitive systems. He claims that the strength of decentralised competition lies in the ‘technical superiority of small operations’ and as a result he believes that large operations are harmful (Keynes, [1931] 1951, 314-15). Subsequently, the state too, being a large operator is an inferior form of organisation.

109But while Marshall, in the spirit of Mill, saw how these disadvantages of large operations may be overcome through the development of human character Keynes seems to ignore it or, at least, have a dis-interested perspective. While he is willing to concede that there will be improvement in human nature he still thinks that there is a need for a decentralised system where people can pursue money reward and self interest: “Though in an ideal commonwealth men may have been taught or inspired or bred to take no interest in the stakes, it may still be wise and prudent statesmanship to allow the game to be played…so long as the average man, or even a significant section of the community, is in fact strongly addicted to the money-making passion.” (GT, 374)

110Thus, even progress, in Keynes’s view, will not really alter the need to allow laissez-faire to be played out: “The task of transmuting human nature” he writes, “must not be confused with the task of managing it.” (GT, 374) His position is simply that the pursuit of self-interest and money-reward is beneficial as it distract the proclivities of humans who are so inclined. “It is better” he writes, “that a man should tyrannise over his bank balance than over his fellow-citizen.” (ibid.)

111So, unlike Mill and Marshall who expected humanity to change and subsequently, the rule of the game to change (Mill more so than Marshall), for Keynes the idea of competitive decentralised decision making is almost unmoveable. Progress, for Keynes, would mean the reduction in interest rate which will eliminate the profits above the normal and help the neo-classical model to shine. To write in this spirit at the time when Pigou has begun to alert the world to discrepancies between private and social interest, seems most peculiar.

6. Conclusions

112Keynes wrote a General Theory and was hoping to revolutionise economic analysis. However, he remained committed to the idea of competitive decentralised decision making as the principle of economic organisation. Consequently, I find it difficult to accept that his theory was in any way sufficiently ‘general’. Nor, for that matter, was it entirely new. The idea of aggregate demand as the source of demand for labour and subsequently a cause of involuntary unemployment has been considered before. I have tried to show that it was embedded in the wage-fund doctrine when one takes into account the fact that the classical model was in essence a dynamic one.

113Ostensibly, the novelty of Keynes’s analysis lies in the idea that in the labour market, the ‘classical’ principle of market clearing would not work. But as we saw, in Mill too there was involuntary unemployment which had been caused by the composition of demand rather than by its level but still, would not be rectified through the normal mechanisms of market clearing. Moreover, from the methodological perspective, the idea of involuntary unemployment as a failure of the economic system corresponds to the classical idea of wages below subsistence. So if the issue of involuntary unemployment is to show how the decentralised system fails society, the point was clearly made by both Smith and Mill.

114More importantly, however, is the fact that for classical economists like Smith and Mill, the system may fail even when there is no market failure. Instead, they seem to be more concerned with the failure of the market. So Keynes’s focus of attention was those who were betrayed by the markets by not allowing them to take part in the system. The real classical economists were more concerned with the fact that markets betrayed everyone and in particular those who were in them even when they appeared to be working smoothly.

115This, of course, leads to the question of how can one deal with such difficulties. We claimed that there are, in principle, two alternatives to the way government may relate to the economic system. The first one was the exogenous way in which the government is attempting to influence the outcomes by shifting the system as a whole. Namely, the government attempts to influence outcomes without affecting the working of the competitive allocation mechanism. The second is when the government becomes part of the system. While it is true that Keynes wanted governments to interfere in the money market and through this affect the entire system, as he held a dual conception of the economy, this amounts to a similar position as the one consistent with the idea of exogenous government.

116By contrast, the real classical view was focused on the other form of government intervention. As the problem lies within the system and those who are betrayed are the players, the only way the government may help is by becoming part of the system.

117For Keynes, if we could somehow release the market clearing mechanism of the labour market, there will be no role for the government. For classical economists this would be the beginning of the problem. It would mean that individuals will need to move around from one specialisation to another with a sufficient degree of flexibility to allow the competitive system to produce its ostensibly beneficial outcome. But for those agents who offer the flexibility of employment, this is, in Smith’s terms, a betrayal of the entire purpose which led them to specialise and trade and thus become dependent of the others. The fact that to-date, we need clear market failures to begin considering the possibility of government involvement suggests that Keynes, unintentionally, no doubt, was very effective in distracting the attention from the important role of endogenous government to marginal discussions of exogenous government.

118It is, perhaps, worthwhile to reflect on the role of government from a historical perspective in a more detailed analysis. I will only say this: we must bear in mind that by the time Smith wrote about a decentralised competitive system, he was writing about a process of emancipation. Capitalism, in those stages of social development was an attempt of power transfer from rulers to the public. There is no surprise that there was, at that stage, much greater reluctance to concede roles to the government which was not a reflection of the people’s power. But over the years, with the political transformations which followed, the idea of allowing the government to play a greater role should have a natural development.

119Indeed, as I have been arguing, neither Smith nor Mill thought that the economy can do well without the government. Both these scholars have recognised that the identity between the private interest and the social interest does not always exit. In their analysis, you do not need a market failure for this to be recognised. Keynes seemed to have accepted none of this. He was, therefore, a conservative who refused to accept that the problem of the economy is organisation and that society needs to worry more about the fate of those who are employed than the fate of those outside the labour market. It is perfectly legitimate to be concerned about the lot of the unemployed as a matter of policy but this cannot really be considered as a general theory of anything.

I am in debt to Arie Arnon, Mauro Boianovski, Richard Jackman and Cristina Marcuzzo for helpful discussion on an earlier version of this paper. The author thanks two anonymous referees. The usual caveat applies.

Haut de page


Backhouse, Roger E. 2006. The Keynesian Revolution. In R. E. Backhouse and B. W. Bateman (eds), The Cambridge Companion to Keynes. Cambridge: Cambridge University Press, 19-38.

Barro, Robert J. and Herschel I. Grossman. 1971. A General Equilibrium Model of Income and Employment. American Economic Review, 61(1): 82-93.

Benassy, Jean-Pascal. 1986. Macroeconomics: An Introduction to the Non-Walrasian Approach. New York, NY: Academic Press.

Clower, Robert W. 1965. The Keynesian Counterrevolution: A Theoretical Appraisal. In Frank H. Hahn and F.P.R. Brechling (eds.), The Theory of Interest Rates. London: Macmillan. Reprinted in Clower (1987), 34-58.

Davidson, Paul. 2007. John Maynard Keynes. New York, NY: Palgrave-Macmillan.

Forget, Evelyn L. 1992. J. S. Mill and the Tory School: The Rhetorical Value of the Recantation. History of Political Economy, 24(1): 31-59.

Galbraith, John K. 1996. Keynes, Einstein, and Scientific Revolution. In P. Arestis (ed.), Keynes, Money, and the Open Economy: Essays in Honor of Paul Davidson. Aldershot: Edward Elgar, 14-21.

Groenewegen, Peter. 1995. Keynes and Marshall: Methodology, Society, and Politics. In Allin F. Cottrell and Michael S. Lawlor (eds.), New Perspectives on Keynes. Durham, NC: Duke University Press, 129-155.

Hoover, Kevin D. 1995. Relative Wages, Rationality, and Involuntary Unemployment in Keynes’s Labour Market. History of Political Economy, 27(4): 653-683.

Keynes, John Maynard. [1936] 2007. The General Theory of Employment, Interest and Money. London: Palgrave.

Keynes, John Maynard. [1921] 1990. Treaties on Probability. The Collected Writtings of John Maynard Keynes, Volume 8. Edited by A. Robinson and D. Moggridge. London: Macmillan.

Keynes, John Maynard. [1926] 1963. The End of Laissez-Faire. In Essays in Persuasion. London: W. W. Norton & Company.

Keynes, John Maynard. [1933] 1972. Essays in Biography. The Collected Writings of John Maynard Keynes, Volume 10. Edited by A. Robinson and D. Moggridge. London: Macmillan.

Kurer, Oskar. 1998. Mill’s Recantation of the Wage-Fund Doctrine: Was Mill Right After All? History of Political Economy, 30(3): 515-536.

Laidler, David E. W. 1999. Fabricating the Keynesian Revolution. Cambridge: Cambridge University Press.

Laidler, David E. W. 2006. Keynes and the Birth of Modern Macroeconomics. In R. E. Backhouse and B. W. Bateman (eds.), The Cambridge Companion to Keynes. Cambridge: Cambridge University Press, 39-57.

Leijonhufvud, Axel. 1968. On Keynesian Economics and the Economics of Keynes. Oxford: Oxford University Press.

Medema, Steven G. 2009. The Hesitant Hand: Taming Self –Interest in the History of Economic Ideas. Princeton, NJ: Princeton University Press.

Marshall, Alfred. [1920] 1952. Principles of Economics. 8th Edition. New York, NY: Macmillan.

Mill, John Stuart. [1848] 1909. Principles of Political Economy. London: Augustus M. Kelley Publishers.

Mill, John Stuart. [1869] 1967. Thornton on Labour and its Claims. In R. M. Robson (ed.), Essays on Economics and Society. Toronto: University of Toronto Press, 631-668.

Moore, George Edward. 1903. Principia Ethica. Cambridge: Cambridge University Press.

O’Donnell, Roderick Macduff. 1992. The Unwritten Books and Papers by Keynes. History of Political Economy, 24(4): 767-817.

Patinkin, Don. 1965. Money Interest and Prices. 2nd edition. New York, NY: Harper and Row.

Raffaelli, Tiziano. 2006. Keynes and the Philosophers. In R. E. Backhouse and B. W. Bateman (eds.), The Cambridge Companion to Keynes. Cambridge: Cambridge University Press, 160-179.

Robinson, Joan. 1974. History versus Equilibrium. Indian Economic Journal, 21(3): 202-213.

Shackle, George L. S. 1973. Epistemics and Economics. Cambridge: Cambridge University Press.

Skidelsky, Robert. 1992. John Maynard Keynes: Volume 2: The Economist as Saviour, 1920-1937. London: Macmillan.

Smith, Adam. [1776] 1976. An Inquiry into the Nature and Causes of the Wealth of Nations. R. H. Campbell and A. S. Skinner (eds). Oxford: Clarendon Press, 2 vols.

Togati, Teodoro D. 2001. Keynes as the Einstein of Economic Theory. History of Political Economy, 33(1): 117-138.

Townshend, Hugh. 1937. Liquidity-Premium and the Theory of Value. Economic Journal, 47(185): 157-169.

Witztum, Amos. 2005. Economic Sociology: The Recursive Economic System of J. S. Mill. Journal of the History of Economic Thought, 27(3): 251-282.

Witztum, Amos. 2009a. Wants Versus Needs: A Smithian model of General Equilibrium. In J Young (ed.), Elgar Companion for Adam Smith. London: Edward Elgar, 141-172.

Witztum, Amos. 2009b. Property Rights and the Social Boundaries of the Firm: Some Lessons from Classical Economics. Mimeo.

Witztum, Amos. 2010. Interdependence and Equilibrium in Adam Smith. History of Political Economy, 42(1): 155–192.

Haut de page


1 In a letter to G B Shaw, Keynes writes: “I find myself to be writing a book on economic theory, which will largely revolutionise…the way the world thinks about economic problems” (Skidelsky, 1992, 520); also quoted in Davidson (2007, 18).

2 Backhouse (2006) provides a good summary of the history of this debate.

3 A more localised and focused attempt at interpreting Keynes were concentrated on chapter 2 (and 19) of the GT or the possibility of non-clearing labour market. However, here too, the focus of attention has been primarily methodological. Hoover (1995) provides a summary of this line of investigation.

4 This point is clearly stated in Chapter 24 of the GT and will be discussed further below.

5 In spite of clear signs that Keynes wishes to overthrow the analysis which treated money as a residual analysis to the real economy (see GT 26) he does not offer an alternative theory for market analysis. Nor does he make the case for the aggregate analysis to supersede microeconomic analysis. There is, perhaps, no better demonstration of the gap between Keynes’s mighty declarations and the reality of his arguments than the essay on The End of Laissez-Faire (1926). If one reads the first part of it one cannot escape the conclusion according to which Keynes is out there to demolish the idea of the working of decentralised decision making. However, each of the following sections further diminishes the attack. Firstly we are told that economics never really argued that private and public interest always coincide. Then, we are told that it is really about the problem that some production units are necessarily big and that there are different domains of individual and collective technologies. Finally, it is all coming down crushing: “These reflections have been directed towards possible improvements in the technique of modern capitalism by the agency of collective action. There is nothing in them which is seriously incompatible with what seems to me to be the essential characteristic of capitalism, namely the dependence upon an intense appeal to the money-making and money-loving instincts of individuals as the main motive force of the economic machine. Nor must I, so near to my end, stray towards other fields. Nevertheless, I may do well to remind you, in conclusion, that the fiercest contests and the most deeply felt divisions of opinion are likely to be waged in the coming years not round technical questions, where the arguments on either side are mainly economic, but round those which, for want of better words, may be called psychological or, perhaps, moral.“ (Keynes, 1926, 322; emphasis added). In my view, if one needs a proof that Keynes was not out there to overthrow classical economics, it is the essay on The End of Laissez-Faire which so poignantly provides it.

6 I consider the case of less than full-employment as equilibrium as well even though the labour markets do not clear in the normal way. The reason for this is that there are institutional reasons which make the labour market stay in this position. It is equilibrium as agents who are outside the system will not be able to influence the outcome.

7 This is not to say that Keynes’s agenda did not go well beyond the GT and when viewed in its entirety, may be seen as revolutionary. However, in as much as the GT was a culmination of his efforts, it did not seem to have taken on board a great deal of what Keynes himself said before.

8 In modern economics, a competitive equilibrium may be inefficient if we face a problem of incomplete markets. Hence, the outcome of competition may not resolve the economic problem but the reason for this is that competition was impeded by the incompleteness of the markets. This is different from the idea of ‘good’ and ‘bad’ equilibrium because here we have two outcomes of competitive markets in none of which has competition been impeded. Yet the consequence of the one is ‘good’ and that of the other is ‘bad’.

9 See, for instance, discussions in Togari (2001) and Galbraith (1994).

10 I should say at the outset that the wage-fund doctrine which I attribute to J. S. Mill is based on elements of his recantation (Mill, 1869 [1967]). In my view, Mill’s position in the Principles was sufficiently vague to allow elements of the recantation to stand. For further analysis see Forget (1992) and Kurer (1998).

11 In his definition of capital Mill writes that it is “the accumulated stock of the produce of labour” (Mill, 1848 [1909], 54). However he makes various qualifications later on to which I refer in the text to suggest that he was using the word ‘capital’ as a shorthand for the wage-fund but was fully aware that the surplus (which he called capital) may be used for other things than the hiring of productive labour.

12 And even from this we must deduct what is called fixed capital.

13 A complete analysis of Smith’s model can be found in Witztum (2010). In Mill’s recantation he says that “he [the employer] starts at the commencement with the whole of his accumulated means, all of which is potentially capital: and out of this he advances his personal and family expenses, exactly as he advances the wages of his labourers” (Mill, [1869] 1967, 645).

14 Which is also consistent with the idea that in full employment the classical school is coming to its own.

15 While it is true that Mill believes that workers are partly to blame for the competitive low wages (due to their number) he distinguishes between the question of population which has to do with future workforce and people who are currently in the labour market. Here, he puts the blame on the capitalists who do not devote sufficient sections of their surplus to hire productive labour. This may sound as a call for savings (increasing the capital) but we must view this in terms of the problem. While any increase in the use of surplus would increase demand for labour, the need to increase the productive employment is associated with increasing future surpluses and is therefore associated with growth rather than output maximisation.

16 See a full discussion in Witztum (2005).

17 Mill argues that there can be no ‘consumer sovereignty’ when it comes to education and by implication, he would have opposed to any market driven provision of it. For more details, see Witztum (2005).

18 The progressive state is in reality the cheerful and hearty state to all different orders of society. The stationary is dull; the declining, melancholy.” (Smith, [1776] 1976, 99)

19 See a lengthy discussion in (Smith, [1776] 1976, 378).

20 For a full exposition of the argument, see Witztum (2009a).

21 See a discussion in Witztum (2009b).

22 Another excellent example to Smith’s clear preferences to state action as opposed to spontaneous provision is the case of the debate about Standing armies versus Militia. See a discussion in Montes (2009).

23 See a discussion of Pigou’s roots in Sidgwick rather than Marshall in Medema (2009, 59-60).

24 See a full discussion in Witztum (2010).

25 And I emphasis again that this holds in spite of claims which some may make about the more complex nature of welfare economics. The reason for this is that Welfare Economics accepts the principle of efficiency as the prime criteria and subsequently, attempts to find solutions to social choice within the context of the solution to the problem of reconciling un-satiated wants with scarcity.

26 Again this point is fully explored in Witztum (2010).

27 See a discussion in Witztum (2005).

28 See an account in O’Donnell (1992).

Haut de page

Pour citer cet article

Référence papier

Amos Witztum, « Keynes’s Misguided Revolution », Œconomia, 3-2 | 2013, 287-318.

Référence électronique

Amos Witztum, « Keynes’s Misguided Revolution », Œconomia [En ligne], 3-2 | 2013, mis en ligne le 01 janvier 2014, consulté le 25 avril 2017. URL : ; DOI : 10.4000/oeconomia.247

Haut de page


Amos Witztum

Economics Subject Area, LMBS, London Metropolitan University, 84 Moorgate London EC2M 6SQ

Haut de page

Droits d’auteur

Licence Creative Commons
Les contenus d’Œconomia sont mis à disposition selon les termes de la Licence Creative Commons Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International.

Haut de page
  • Logo Association Œconomia
  • Logo CNRS
  • Les cahiers de