Keynes’s General Theory After 75 Years: Chapter 24 and the Character of ‘Keynesian’ Policy

Abstract

The Great Financial Crisis led to a revival of interest in ‘Keynesian’ economic policy, understood as temporary, debt-financed fiscal activism in response to a short-run, albeit particularly deep, macroeconomic contraction. But Keynes’s theory led him to diagnose involuntary unemployment as a long-run problem, and one for which competition (via wage and price flexibility) was not a solution. Consequently, his policy proposals were more far-reaching than short-run fiscal activism. The final chapter of The General Theory provides the appropriate point of departure for reflecting on the difference between Keynes’s policy position and what later came to be regarded as Keynesian policy.

I Introduction

If Keynes’sGeneral Theory (1936; hereafter, ‘GT’) had been published four years earlier than it was, so that its seventy-fifth anniversary had come in 2007, one may conjecture that there might have been considerably less interest in the event. But as it now happens, the anniversary follows on the heels of the most serious and challenging economic crisis of the developed capitalist economies in somewhat more than seventy-five years. In a general sense that crisis has stimulated some renewal of interest in the possible relevance of Keynes’s theory, to the extent that the Great Financial Crisis (hereafter, ‘GFC’) has placed in question the capacity of a decentralised competitive economy to more or less continuously – or at least, more or less rapidly – self-regulate in a manner so as to ensure full utilization of resources (most especially, labour), without the need of government policy intervention. Yet no sooner had interest in Keynes revived than there followed a self-consciously anti-Keynesian reaction, the practical catalyst for which was an alleged crisis of fiscal deficits and public debt.

The questions of whether or to what extent the trajectories of deficits and debt following on from the crisis themselves constitute a genuine economic crisis, and to what extent they are directly or indirectly attributable to discretionary government policy, are not addressed here. Suffice it to assert that the largest part of the widened fiscal deficits has been an endogenous consequence of the GFC itself, and that the complicity or otherwise of government policy in the GFC is a controversy fraught with ideological axe-grinding.

Instead, to mark this anniversary, this essay revisits what precisely Keynes’s mature policy views were, following on from GT, on the central issue – for him the central issue – of ensuring a full utilization of resources, or more or less zero involuntary unemployment. It is rather taken for granted in contemporary debates around the relevance or usefulness of Keynesianism that the core of the Keynesian policy position is discretionary, countercyclical, debt-financed fiscal activism. But in fact, this is not Keynes’s core policy position.¹

II Keynes’s Core Policy Position

GT is not much devoted to policy analysis, though the limitations of monetary policy are discussed at a number of points (in particular, GT, pp. 163–64, 202–08, 315–21) and the notion of flexible wages as an instrument for ensuring full employment is systematically refuted in Chapter 19. But as well as those critical policy commentaries, the final Chapter 24 of the book, on ‘… the Social Philosophy towards which the General Theory might lead’ (GT, p. 372), sketches a constructive argument concerning the kind of policy regime Keynes regards as necessary or desirable for the establishment and maintenance of full employment in a liberal society. Fortunately, further detail can be added to that sketch by reference to Keynes policy documents of the 1940s, documents that arose out of planning for the post-War world (Moggridge, 1980, pp. 201–419).

To grasp the genuine character of Keynes’s policy position, as distinct from the contemporary conventional characterization of ‘Keynesian’ fiscal activism, it is first and foremost necessary to recognise that his theory posits an absence of reliable mechanisms in a decentralised competitive economy for ensuring full labour employment, both in the short run and the long run. Furthermore, for somewhat more contingent reasons, Keynes believed that mature capitalist economies would exhibit a kind of ‘natural’, secular tendency to underemployment, with the growth of potential supply capacity persistently tending to outstrip the growth of aggregate demand (GT, pp. 249–54), a view suggested already well prior to GT (Keynes, 1930, pp. 321, 325). In a nutshell, liberal capitalism has a genius for expanding potential production capacity over time, but not matched by an ability to similarly expand aggregate demand. Involuntary unemployment, even under competitive conditions, is a long-run as well as a short-run or cyclical phenomenon. Obviously, merely countercyclical demand management cannot be an adequate policy response if there is a problem of long-period involuntary unemployment, attributable to deficient aggregate demand. Permanent and continuous demand management is required to address such a problem.

One element of the conventional perception of Keynesian policy is, however, more or less sound, in relation to the views of Keynes himself. As already alluded to, he had very substantial doubts as to the adequacy of monetary policy, or monetary policy alone, to provide the necessary or desirable demand management. (The capacities and limits of monetary policy of course have been a subject of debate in the context of the GFC, with respect to the efficacy of monetary versus fiscal policy instruments, and the multiplication of monetary instruments over and above the short-term interest rate instrument, which had become the conventional mode of monetary policy conduct over prior decades.) Keynes was firmly opposed to short-run interest rate activism, in favour of a long-run policy of gradually moving the riskless real rate towards zero, as a permanent policy – the famous or infamous ‘euthanasia of the rentier’ doctrine (GT, pp. 374–77, with pp. 217–21; Moggridge, 1980, p. 377; Aspromourgos, 2004). Although his primary motivation for this policy is ethical, Keynes also saw it as supporting aggregate demand by (permanently) redistributing income towards wage-earners and thereby increasing the propensity to consume and the multiplier. Such a regime of permanent ‘cheap money’ also has obvious advantages with regard to the magnitude of sustainable fiscal deficits (Aspromourgos et al., 2010).

However, the main means to demand management for Keynes are the scale of the public sector and associated recourse to fiscal instruments. The notion of underemployment as a permanent problem of mature capitalism, combined with a lack of faith in the capacity of monetary policy to adapt private aggregate demand to aggregate supply capacity, points to a permanent expansion of public expenditure as the necessary means to full employment. At least this is so when combined with Keynes’s further policy conviction that expansion of private consumption by fiscal means is, in some respects, of dubious feasibility, and in other respects, undesirable, up to a point.

With respect to feasibility, in a 1943 letter to James Meade, Keynes doubts the practicability of countercyclical manipulation of private consumption:

People have established standards of life. Nothing will upset them more than to be subject to pressure constantly to vary them up and down. A remission of taxation on which people could only rely for an indefinitely short period might have very limited effects in stimulating their consumption. And, if it was successful, it would be extraordinarily difficult from the political angle to reimpose the taxation again when employment improved.² (Moggridge, 1980, p. 319; cf. 206–07, 323, 326, 361)

It is ironic that there is a significant kinship between this position and the later permanent income and life-cycle theories of consumption, to the extent that the permanent income hypothesis was utilised as an element in the critique of supposed Keynesian short-run fiscal activism. (That hypothesis does not undermine Keynes’s genuine core fiscal proposals.)

With respect to the desirability of expanding consumption, in both GT and the 1940s policy documents Keynes tends somewhat to favour public investment as the appropriate policy route to full employment (combined with progressive taxation in GT, pp. 94–95, 372–74, 377; cf.Moggridge, 1980, p. 323), as against increased consumption expenditure, private or public (GT, pp. 163–64, 220, 377; Moggridge, 1980, pp. 267–70, 284–88, 320–25, 350). To that extent, public spending oriented towards capital works is to fill the expected gap between the full-employment level of private saving plus tax revenues, and private investment. In the above-mentioned letter to Meade, Keynes also offers a supporting pragmatic argument for favouring investment over consumption, in a countercyclical context –‘it is not nearly so easy politically and to the common man to put across the encouragement of consumption in bad times as it is to induce the encouragement of capital expenditure’– and alludes to the capital saturation notion which was associated in GT with euthanasia of the rentier: ‘it is better … that periods of deficiency expenditure should be made the occasion of capital development until our economy is much more saturated with capital goods than it is at present’ (Moggridge, 1980, pp. 319–20; also pp. 321, 323–24). But in the 1940s documents Keynes also is quite strongly averse to public-debt-financing of this enlarged role of the public sector.

At one point, in a published article, he contrasts ‘the day-dreams of planning’ with ‘the nightmare of finance’ (Moggridge, 1980, p. 264). Keynes’s cautious views on debt financing in these documents is often in the context of discussing proposals to formally institute distinct ‘ordinary’ and capital budgets for the public sector (Moggridge, 1980, pp. 225, 264–67, 275, 277–80, 352–57, 366–69, 405–13). But this aspect is a second-order issue for Keynes, not a matter of substantive principle or policy, ‘in a sense … no more than a matter of presentation’ (Moggridge, 1980, p. 352):

This question [of instituting a formal capital budget] is essentially a question of presentation. It does not enable anything to be done which could not be done without it by means of the existing technique and in conformity with the existing form of the Exchequer Accounts. Nevertheless presentation may be of great importance by bringing out clearly the relevant criteria for policy and by high-lighting what it is desirable that Parliament and the public, and also officials, should understand. (Moggridge, 1980, p. 405)

Keynes also rather vigorously points out the endogeneity of the public sector budget balance to aggregate activity levels, and hence the capacity of a full employment policy to limit adverse balances (Moggridge, 1980, pp. 366, 376–77).

What are the requirements for this combination of policy objectives to be achieved? On the one hand, a growth of public expenditures is required so as to ensure a growth rate of demand-led activity levels consistent with the growth of labour supply. On the other hand, stabilization of the public-debt-to-GDP ratio at some desired level entails a definite magnitude of the public sector primary deficit as a percentage of GDP – a magnitude determined by the difference between the GDP growth rate and the average rate of interest on the outstanding stock of public debt, together with the target debt/GDP ratio itself. (Recall in this context the point made earlier concerning cheap money and debt management.) In this thought experiment the target debt/GDP ratio could be constrained to zero, or a very low number (like Australia’s), or a trajectory of decline, in order to capture the spirit of Keynes’s apparent debt conservatism. Given the share of government expenditure in GDP entailed by the full-employment policy, the debt sustainability condition determines a necessary tax share of GDP, or lower bound of the tax share. The existence of an economically feasible solution for this policy problem with two objectives requires that tax share to be between zero and unity – and of course, a politically feasible solution requires it to be not too close to unity. Political feasibility is evidently a function of social values: a forty-five per cent tax share might be feasible in some parts of continental Europe; it would appear not to be so in the US or Britain.

III Six Qualifications

This summary of Keynes’s core policy view may be filled out a little further by noting six qualifications.

First, however left-of-centre Keynes’s policy position may now look in the context of the political atmosphere that has developed in recent decades, he is very much imbued with the tradition of British liberalism. He therefore seeks devices which will render his policy position consistent with limiting the size of the state, and of central government in particular. To this end he looks to decentralised government entities (including levels of government below the national level), semi-government agencies and quasi-public institutions to partially serve as vehicles for the required spending programme: earlier Keynes had suggested that ‘progress lies in the growth … of semi-autonomous bodies within the State … whose criterion of action … is solely the public good’, a kind of ‘semi-socialism’ as against ‘State Socialism’ (Keynes, 1926, pp. 288–94); in GT (pp. 377–81) he speaks of ‘all manner of compromises and of devices by which public authority will co-operate with private initiative’; and later, investment ‘carried out or … influenced by public or semi-public bodies’ (Moggridge, 1980, p. 322; also p. 352). The recent proposal by Skidelsky and Martin (2011) for an American national investment bank is very much in this Keynes spirit, though we would not endorse their broader interpretation of Keynes’s underlying theory and policy position. (Their investment priorities, naturally enough in the contemporary context, are ‘green’ infrastructure and energy.) But notwithstanding these liberal efforts on Keynes’s part, the logic of his theory means that at least centralised policy coordination remains necessary, since the unemployment problem is attributed to the incapacity of a decentralised system to coordinate full-employment saving with investment (cf.Moggridge, 1980, pp. 268, 275–76, 279–80, 408–10).

Keynes’s kind of ‘third-way’ politics and policy (our term), in the context of the inter-War political spectrum –‘a middle way of economic life’, as he puts it (Moggridge, 1980, p. 369) – is nicely captured in the opening sentences of an open letter addressed to Franklin Roosevelt and published in the New York Times, 31 December 1933:

You have made yourself the trustee for those in every country who seek to mend the evils of our condition by reasoned experiment within the framework of the existing social system.

If you fail, rational change will be gravely prejudiced throughout the world, leaving orthodoxy and revolution to fight it out. (Moggridge, 1982a, p. 289)

And also, somewhat earlier, Keynes (1930, p. 322; cf.Moggridge, 1980, pp. 354–55, 374) writes of ‘two opposed errors of pessimism’ concerning the ‘prevailing world depression’:

the pessimism of the revolutionaries who think that things are so bad that nothing can save us but violent change, and the pessimism of the reactionaries who consider the balance of our economic and social life so precarious that we must risk no experiments.

Notice the use in both statements of the term ‘experiment’, expressive of Keynes’s reform sensibility. (See also his important letter to Friedrich Hayek, concerning Keynes’s ‘middle course’ politics: Moggridge, 1980, pp. 385–88.)

Second, Keynes actually favours recourse to user-pays financing of a large part of his recommended public investment programme, which has the consequence of reducing the required share of tax revenue in GDP, relative to the public expenditure share of GDP:

The more socialised we become, the more important it is to associate as closely as possible the cost of particular services with the sources out of which they are provided, even when a grant-in-aid is also required from general taxes. This is the only way by which to preserve sound accounting, to measure efficiency, to maintain economy and to keep the public properly aware of what things cost. (Moggridge, 1980, pp. 224–25)

Housing is an important part of Keynes’s public investment plans (Moggridge, 1980, p. 350; cf. GT, p. 106), and presumably this was to be at least partly, perhaps wholly, financed by user charges. Such financing by user charges of course might also enhance the political feasibility of enlarging the role of government. The desirability of self-financing public expenditures³ is also raised in the above-mentioned 1943 letter to Meade, providing a further reason for favouring investment expenditures, and coloured with a hint of political pragmatism. Reducing taxation with a view to increasing private consumption,

is a much more violent version of deficit budgeting. Capital expenditure would, at least partially, if not wholly, pay for itself. … [T]hat capital expenditure is capable of paying for itself makes it much better budgetwise and does not involve the progressive increase of budgetary difficulties, which deficit budgeting for the sake of consumption may bring about or, at any rate, would be accused of bringing about. (Moggridge, 1980, pp. 319–20; emphasis added; cf. 350)

The Skidelsky and Martin investment bank proposal also has this user-pays character.

Third, Keynes’s apparent debt conservatism is surprising, for the simple reason that it seems too conservative. But perhaps it is not quite what it seems. If his position is taken literally – as a constraint of no positive aggregate net debt issue, unless funded by user charges – then in a growing economy it implies the ratio of ‘general government’⁴ debt to GDP declining continuously from its initial level. Why would that be desirable as a general principle? It may be that Keynes’s apparent debt conservatism is specifically directed at the particular circumstances of 1940s Britain, a country with debt/GDP ratios at well over 100 per cent throughout the 1930s, and well in excess of 200 per cent by the end of the war. This doubt as to the extent of Keynes’s debt conservatism is somewhat strengthened by a passing comment he makes on the then recently published article on functional finance by Abba Lerner (1943), which takes an extremely relaxed view of debt financing: ‘His argument is impeccable’, Keynes wrote, noting also that Lerner is concerned with ‘a chronic deficiency of purchasing power rather than an intermittent one’ (Moggridge, 1980, p. 320; also pp. 404–05).⁵ Furthermore, in a 1942 memorandum Keynes expresses the expectation that ‘for a long time to come … government debt or government-guaranteed debt would be continually increasing in grand total’; in a 1944 note he appears unconcerned about a growth of ‘dead-weight debt’ (debt issued to finance recurrent expenditures) that is ‘neither large in itself nor out of proportion to the growth of the national income’ (Moggridge, 1980, pp. 278, 366).

Fourth, the preference for public capital expenditure over recurrent consumption expenditure is certainly understood by Keynes as a contingent rather than a general or universal principle. In the above-mentioned May 1943 memorandum Keynes makes very clear that the net capital accumulation he regards as having priority in public expenditure and employment policy, vis-à-vis consumption expenditure (except to allow easing of wartime rationing –Moggridge, 1980, p. 322), has a finite limit. Beyond the time-span within which those investment plans are realised – he thinks, for Britain, a period of approximately fifteen years following the War – consumption expenditure may replace them in the role of generating aggregate demand (Moggridge, 1980, pp. 322–23; also p. 350, where the time-span given is twenty years, and pp. 359–60, 410–11).

Fifth, Keynes is also explicit that at a future date, not expected by him to be far beyond those fifteen years, both all desirable net capital accumulation and all desirable increased consumption will be realised, a ‘golden age’ stationary state (‘depreciation funds should be almost sufficient to provide all the gross investment that is required’). If the available labour supply in that situation is incapable of full utilisation then the time has arrived for adjustment to occur on the labour supply side, by reducing working hours (Moggridge, 1980, pp. 323–24; cf. 260–61, 276). A much earlier Keynes essay also projects this vision of a future human condition of consumption satiation and increased leisure, indicating that it is a considered view on his part, not merely an offhand comment, though the timeframe appears rather longer there (Keynes, 1930, pp. 326, 331). He also makes explicit there that population control is part of the scenario (Keynes, 1930, pp. 325–26, 331; also Keynes, 1926, pp. 277, 292). One could characterise this view as a sort of ‘green’ sensibility; but only ‘sort of’, because, while a definite finite limit to consumption per capita is endorsed, this is rationalised without reference to natural resource depletion. Keynes nicely sums up the whole issue in a 1945 letter to T.S. Eliot:

the full employment policy by means of investment is only one particular application of an intellectual theorem. You can produce the result just as well by consuming more or working less. … How you mix up the three ingredients of a cure is a matter of taste and experience, i.e. of morals and knowledge. (Moggridge, 1980, p. 384)

Kregel (1985) and Brown-Collier and Collier (1995) both convey a rather exaggerated impression of the priority Keynes gives to investment, as a general principle.

Whether or not one is prepared to go along with Keynes’s vision of a stationary consumption state, with further technical progress then realised in the form of greater leisure, the principle that remains valid is that there cannot be an unqualified preference for capital over recurrent expenditure. To the extent that investment is capacity-creating, there must ultimately be some definite proportion between capital accumulation and a sufficient growth of aggregate consumption demand to validate the associated capacity creation. Keynes makes this kind of point more than once in GT; for example:

Aggregate demand can be derived only from present consumption or from present provision for future consumption. The consumption for which we can profitably provide in advance cannot be pushed indefinitely into the future. … The obstacle to a clear understanding is … an inadequate appreciation of the fact that capital is not a self-subsistent entity existing apart from consumption. On the contrary, every weakening in the propensity to consume regarded as a permanent habit must weaken the demand for capital as well as the demand for consumption. (GT, pp. 104, 106; emphasis added)

Sixth, and finally, it has been indicated earlier that demand deficiency for Keynes is not a merely cyclical problem and so requires a permanently enlarged role for government expenditures, rather than a merely countercyclical policy.⁶ His opposition to the use of interest rates and tax rates as short-run policy instruments was also pointed out. Altogether, do these views mean that he would have opposed a short-run fiscal response to the impact of the GFC on activity levels? Well, who really knows how Keynes would have responded to the GFC? One is reminded of the comment of Roy Harrod (1974, pp. 8–9), Keynes’s first biographer and of course a distinguished contributor to the theory of cycles and growth himself, in response to an earlier post-Keynes crisis: ‘It would be most inappropriate for me to stand up here and tell you what Keynes would have thought. Goodness knows he would have thought of something much cleverer than I can think of.’ Nevertheless, one can probably safely say that Keynes indeed would have endorsed a fiscal response, certainly for countries like the US, China, Germany, the UK and Australia.

Keynes’s first-best world would have a policy regime such that significant macroeconomic fluctuations were avoided; one may say, an anti-cyclical policy regime but not a countercyclical one. Nevertheless, in the above-mentioned May 1943 memorandum, while emphasizing the stabilizing role of a permanently enlarged public sector, he allows countercyclical policy in response to fluctuations, at the margins:

If two-thirds or three-quarters of total investment is carried out or can be influenced by public or semi-public bodies, a long-term programme of a stable character should be capable of reducing the potential range of fluctuation to much narrower limits than formerly, when a smaller volume of investment was under public control and when even this part tended to follow, rather than correct, fluctuations of investment in the strictly private sector of the economy. … The main task should be to prevent large fluctuations by a stable long-term programme. If this is successful it should not be too difficult to offset small fluctuations by expediting or retarding some items in this long-term programme. (Moggridge, 1980, p. 322; cf. 288, 326, 353–57, 360–61, 365–66, 410–11)

And further:

Emphasis should be placed primarily on measures to maintain a steady level of employment and thus to prevent fluctuations. If a large fluctuation is allowed to occur, it will be difficult to find adequate offsetting measures of sufficiently quick action. … If the authorities know quite clearly what they are trying to do and are given sufficient powers, reasonable success in the performance of the task should not be too difficult. … I doubt if much is to be hoped from proposals to offset unforeseen short-period fluctuations in investment by stimulating short-period changes in consumption. (Moggridge, 1980, p. 323; cf. 316)

There is one particular species of countercyclical policy that Keynes endorses in the 1940s documents: Meade’s proposal for a formula by which social insurance contributions would be varied in response to deviations of the unemployment rate with respect to some norm (Moggridge, 1980, pp. 208–11, 217–18, 222, 278, 308–11, 318–19, 323–24, 365; see also pp. 365–66 on varying ‘Hire Purchase’ conditions). Note that this is a rule-based countercyclical mechanism; in effect, augmenting the automatic fiscal stabilisers.

IV Conclusion

Perhaps it is the fate of all thinkers who gain a large and disciple-like following to be vulgarised, misunderstood and misrepresented. According to Friedrich Engels, Karl Marx is supposed to have once declared, ‘je ne suis pas Marxiste’ (Institut für Marxismus-Leninismus, 1967, p. 388). In a similar vein, after a 1944 dinner with a group of American Keynesians in Washington, Keynes commented, according to Austin Robinson (1977, p. 58): ‘I was the only non-Keynesian there’. But it is one thing to demonstrate that Keynes’s policy position was rather substantially different from the latter-day image of the character of post-War ‘Keynesian’ policy. Addressing the question of whether Keynes’s position is appropriate today is quite another.

It can only be asserted here that it is, with regard to the role of the public sector, but without thereby necessarily endorsing a general applicability of his debt conservatism in relation to wartime Britain. And in a way, his position on the public sector has been vindicated. The growth in the size of the public sector in developed economies since 1936 is a kind of realization of Keynes’s policy, whether or not it has occurred for exactly his reasons.⁷ The rising share of tertiary services in expenditures as per capita incomes rise, combined with socio-political imperatives to public provision, or at least subsidization, of crucial elements of these services (notably, health and education), appears to have been a key factor in these developments. Furthermore, it was not merely the government budget automatic stabilisers that provided a buffer against the shock to private sector activity levels delivered by the GFC; the sheer size of public sectors limited the damage to overall activity levels, independent of the endogenous impacts of private sector contraction on government revenues and outlays. With regard to public debt trajectories, one may say, with some sorrow, that perhaps the strongest argument for a measure of public debt conservatism, in the kind of world we’re now living in, is so as to ensure that one has plenty of fiscal firepower in reserve for the next economic crisis, whatever its source might be (Aspromourgos, 2010, ‘Some Conclusions’).

Keynes’s cheap money view is also attractive, though deregulated financial markets place perhaps insurmountable obstacles in the way of its implementation today (Aspromourgos, 2011). It may be added in this context that an answer to the question of how Keynes might have responded to the actual occurrence of the GFC, and the issue of regulatory responses to it, is much more elusive. But Chapter 12 of GT– on equity markets, long-term expectation and speculation – as well as other Keynes commentary (e.g., Keynes, 1937, pp. 212–19), expresses a firm view of limits to the capacity of rationality to determinately govern the dynamics of capital markets and asset prices. Frydman and Goldberg (2011, pp. 120–28, 149–60, 249–53) provide an interesting recent interpretation.

In relation to the plausibility of cheap money, and more generally, there is also the issue of inflation. The apparent absence of any specific and additional policy instrument assigned to inflation could be regarded as a striking sin of omission in Keynes’s policy position. This potential difficulty can be dismissed if adapting aggregate demand to the supply capacity of the economy, via the full-employment demand-management policy, ensures also an absence of inflationary and deflationary pressures. If, on the other hand, such pressures can arise independently of aggregate-demand/supply imbalances then there is a need for at least one additional policy instrument. Incompatible distributional claims (e.g., triggered by energy price shocks), traditionally characterised in terms of ‘cost-push’, is the obvious possibility here. Of course, such pressures from distributional conflict may actually be encouraged by operating an economy very close to its supply constraint (cf.Moggridge, 1980, p. 385).

Footnotes

• 1

  Kregel (1985) is an important earlier contribution to clarifying Keynes’s genuine policy views, contrasting them with post-War economic policies. See also Brown-Collier and Collier (1995). Keynes’s more well-known policy proposals for post-War international monetary and exchange-rate arrangements are left aside here. Fundamentally, they may be understood as aimed at establishing an international adjustment mechanism with no contractionary bias in relation to global effective demand.

• 2

  Strictly speaking, this does not logically exclude a possible recourse to permanent tax reductions, in some circumstances. In an important May 1943 memorandum Keynes makes the habitual propensity to consume a function of ‘the form and weight of taxation and other deliberate policies’ (Moggridge, 1980, p. 321). Attempted fiscal manipulation of private investment, to the extent that it operates via altering the cost of capital, in Keynes’s view presumably would be subject to the same limitations as monetary policy.

• 3

  That is to say, public expenditures, including debt-financed expenditures, which are financed by user charges, including debt-financed capital expenditures for which the debt servicing is financed by user charges.

• 4

  This qualifier is to exclude ‘commercial’ government debt, the latter signifying debt funded by user charges in the sense of note 3.

• 5

  Lerner was lecturing at the LSE, 1935–37. Then and earlier, he was extensively involved in discussions of Keynes’s new theory with Joan Robinson, Keynes, Richard Kahn and others, and spent six months in Cambridge (Moggridge, 1973, p. 148; Colander & Landreth, 1996, pp. 87–91). He also reviewed GT and shortly thereafter published two further GT-related articles (Lerner, 1936, 1938a,b). For Keynes’s relations with Lerner, see Moggridge (1979a, pp. 181–82; 1979b, pp. 214–16, 271, 281–82; 1982b, pp. 56), a 1972 interview with Lerner in Colander and Landreth (1996, pp. 88–93, 105–09, 114–17), and for one particularly notorious incident of Keynes publicly repudiating Lerner on debt, also Colander and Landreth (1996, pp. 177, 184–87, 202).

• 6

  Of course, if demand deficiency were a merely cyclical problem, then debt-financed deficits in slumps could easily be rendered compatible with a stationary debt/GDP ratio, via offsetting surpluses in booms. In the 1940s documents Keynes is also far from oblivious to supply-side or ‘structural’ unemployment, and policy to address it: ‘something to be handled forcibly and not something to be defeatist about’ (Moggridge, 1980, p. 357; also pp. 245, 250, 281, 287, 299, 303, 305, 331, 335, 354–55, 358, 381, 383–84, 404n).

• 7

  Apart from the issue of the size of post-War public sectors relative to Keynes’s proposals, Kregel (1985, pp. 39–48) argues that the composition and purpose of the public expenditures did not accord with Keynes’s intentions. Brown-Collier and Collier (1995, pp. 350–55) undertake a similar comparison of Keynes’s position and post-War US policy. See also Clarke (2009, pp. 159–80), which is as well the best available short account of Keynes’s life and thought as a whole.