If we want to know why a car is sometimes driving safely along the road, but sometimes crashes or breaks down, we have alternative ways of explaining it. One explanation is mechanical, showing the necessary effects of engine capacity and efficiency in determining the engine’s exact performance limits. The other explanation concentrates on the human factor, exploring the judgement of the driver in making decisions that may well be subject to external distraction or temperamental readiness to take risks. The theory of effective demand, as conceived in 1932, was more like the first kind of analysis. It was essentially hydraulic, specifying necessary flows and displacements, just as the modus operandi of bank rate, though working in a different way, had also been hydraulic.

What about the General Theory itself, as published in 1936? In protracted revision of drafts of the book, it took on a further dimension, infused with more subjective notions, not only about psychology but uncertainty too. In the process, Keynes was returning to ideas that he had first broached a quarter of a century earlier, in his work on probability, as we saw in Chapter 1. He was also, of course, picking up themes from A Treatise on Money, reformulating insights about the role of expectations. Many early Keynesians – and there was none earlier than James Meade – saw no contradiction in taking expectations as given, then specifying determinate results for the system.

Few economists would disagree that Keynes played a key role in establishing the importance of macroeconomics. Although the term itself was not used until the 1940s, and never by Keynes himself, the study of the system as a whole surely informs his approach. Citation counts in economic journals of the period show that the Treatise had already made Keynes the world’s most cited macroeconomist, displacing the American Irving Fisher. After the General Theory, Keynes’s primacy was unrivalled. His name was cited nearly three times more often than the next-ranking economist (Dennis Robertson), four times more often than Pigou, five times more often than Harrod, Hawtrey or Hayek, six times more often than Joan Robinson, eight times more often than Ohlin and nine times more often than Schumpeter.³⁴

The structure of the General Theory makes clear why this is so. It is divided into six parts. The first concludes by rejecting the direction taken by classical economics since Malthus – a huge and provocative claim, of course. As Keynes privately told Harrod: ‘I want, so to speak, to raise a dust; because it is only out of the controversy that will arise that what I am saying will get understood.’³⁵ This helps explain the sharpness of the distinctions made in expounding the theory of effective demand in the early parts of the book.

If aggregate output is to find a market, and thus make production profitable, there must be an equivalent amount of aggregate demand. Effective demand comprises both consumption and investment. Both of them are the product of active choices or decisions – unlike saving, which is a passive residual. (So Marshall’s argument that saving is simply deferred spending has to be dismissed.) Therefore the part of current income that is not spent on consumption goods must be matched by investment, which is itself spending on capital goods. Otherwise there will be disabling repercussions.

Individual choice in a free market is real. Or rather, the initial decision is real but the overall outcome might be unintended and perverse. For saving and investment are only reconciled in aggregate by changes in income. Saving, like spending, is a two-sided affair. One person’s consumption is another person’s income; one person’s thrift can be another person’s loss of income. Saving, as a product of the collective behaviour of consumers, will necessarily be brought into equality with investment, as a product of the collective behaviour of entrepreneurs. This has to be so since each of them, saving and investment alike, equals income minus consumption.

There will thus be an equilibrium, of sorts. It will not, however, necessarily be at full employment or optimal output, since it is aggregate output itself that will have to accommodate – producing any level of income or output or employment that will equilibrate saving and investment. The final words of the book’s second part point to ‘the vital difference between the theory of the economic behaviour of the aggregate and the theory of the behaviour of the individual unit, in which we assume that changes in the individual’s own demand do not affect his income’.³⁶

The third part of the General Theory, ‘The propensity to consume’, draws together vital parts of the argument. ‘The ultimate object of our analysis is to discover what determines the volume of employment,’ Keynes writes; and his key suggestion is that it is the aggregate demand function that has been overlooked.³⁷ Since demand comprises both investment and consumption, both are necessary. It was nothing new for Keynes to stress the key role of investment. But the analysis of the General Theory, in concentrating on the ‘propensity to consume’ rather than the disposition to save, gives a new twist to the argument. In principle, deficient consumption might, as much as deficient investment, be called the root of the problem, since both stem from attempts to save too much. This third part of the book concludes by calling unemployment ‘the inevitable result of applying to the conduct of the State the maxims which are best calculated to “enrich” an individual …’³⁸

Here is the famous paradox of thrift. It is a paradox because it seems natural to suppose that if individual saving enriches the individual concerned, it must also enrich the community. ‘The error lies in proceeding to the plausible inference that, when an individual saves, he will increase aggregate investment by an equal amount,’ the General Theory tells us.³⁹ It was a story that Keynes had told before, as in the banana parable, with excessive thrift promoting an inadvertent spiral of declining prices, declining profits, declining employment, declining incomes, declining prosperity all round. His account – now more cogent and more coherent – turns on the distinction between what is true for individuals and what is true in aggregate.

Keynes was not the first person in history to have seized on this crucial insight. In the General Theory he is very generous in dispensing acknowledgements to pioneers who had preceded him in identifying the paradox of thrift. In Chapter 23, he has enormous fun in rehabilitating a whole gallery of economic heretics, especially Bernard de Mandeville for his early-eighteenth-century Fable of the Bees, and, of course, Malthus, on whom he had already published a biographical essay. There is also a seven-page tribute to J. A. Hobson, best remembered now for his pioneering analysis of imperialism and best known at that time for his propositions about ‘under-consumption’. Hobson in his old age was obviously deeply gratified to hear that his work represented ‘an epoch in economic thought’.⁴⁰

Keynes actually learnt little from Hobson – but should have learnt more. The General Theory was already in proof before the section on Hobson was drafted with Kahn’s help. Hobson was a freelance intellectual who had never held any professorial post in Britain, though he was an honoured visitor to American universities, notably the University of Wisconsin, where his kind of ‘progressive’ political commitments struck a common chord. He had been closer to Keynes ideologically rather than academically. True, Hobson and Keynes had both come to focus on an aggregate process that could be called over-saving; but for Hobson, taking saving and investment as the same activity, this meant over-investment, with the only remedy an increase in consumption; whereas for Keynes, of course, over-saving meant under-investment, and stimulating investment was the key. The General Theory tactfully brings out this difference, while hailing Hobson as an ‘economic heretic’ who deserved better than the slights of the economic profession.

What Keynes ignored, however, was a Hobsonian insight that he could well have picked up sooner. For Hobson identified his own (valid) contentions about the fatuity of collective over-saving as ‘at root a very simple fallacy, viz. the contention that what anyone can do, all can do’.⁴¹ Hobson called this the individualist fallacy and a favourite illustration was to say that although it might be true that any boy could go from a log cabin to the White House, not all could do so simultaneously.

This idea is sometime called the fallacy of composition. Its centrality can hardly be exaggerated, since it is, in a sense, the general theory behind the General Theory. For example, there cannot be any actual aggregate of excess savings: it is an excessive propensity to save that produces the perverse effects. Keynes sees that each of us individually may try to execute strategies that appear rational to ourselves. But these may prove self-defeating if we all try at once. Yet we can all see, in principle, what would be in our collective self-interest. So there are intractable theoretical reasons why market failure can leave us impotent to achieve what we actually want – unless, that is, government can intervene to save us.

There is a second guiding insight at the heart of Keynes’s intellectual revolution. This harks back to his earliest academic work, on probability. He put this best in an article that he wrote at the end of 1936 for the Harvard University Quarterly Journal of Economics, called ‘The general theory of employment’. Here, uncertainty emerges as the ghost in the Keynesian machine. He suggests that ‘the fact that our knowledge of the future is fluctuating, vague and uncertain, renders wealth a peculiarly unsuitable subject for the methods of the classical political economy’. He sees that, while we can estimate probabilities, however remote, when faced with uncertainty, we move beyond the reach of scientific calculation. ‘We simply do not know,’ Keynes tells us. Yet it is a world pervaded by uncertainty in which we live, and have to make decisions every day, economic decisions not least. So ‘our desire to hold money as a store of wealth is a barometer of the degree of our distrust of our own calculations and conventions concerning the future’.⁴²

Expectations are thus the key. And that necessarily means what we expect from a viewpoint that is before the event. Whether expectations are rational depends on whether they are based on reasonable judgment exercised ex ante, not on how this appears ex post. If we apply such thinking to the economy today, then, we can see obvious objections to any model of ‘rational expectations’ that defines rationality as the market behaviour that is actually fulfilled. At any rate, the application of ‘rational expectations’ to license a view of the market as omniscient seems rather naive.

Our decisions may not be irrational, but to suppose that our expectations can be totally rational is a fallacy. The economy depends on investment, and investment depends on expectations, however limited in their scientific basis. The General Theory itself makes this very clear. ‘It would be foolish, in forming our expectations, to attach great weight to matters which are very uncertain,’ it warns us, invoking the Treatise on Probability.⁴³ The precariousness of our knowledge is one theme, the pragmatic response of falling back upon convention another. Second-guessing the psychology of the market itself becomes the motive force. ‘Speculators may do no harm as bubbles on a steady stream of enterprise,’ allowed Keynes, letting himself off lightly before reproaching Wall Street in particular: ‘When the capital development of a country becomes a by-product of the activities of a casino, the job is likely to be ill-done.’⁴⁴

This is partly an institutional problem, in stacking market incentives in the wrong way. But human nature itself cannot be ignored. Confidence is crucial to expectations; the future depends upon its robustness; yet it is far from rational. ‘Most, probably, of our decisions to do something positive, the full consequences of which will be drawn out over many days to come, can only be taken as a result of animal spirits – of a spontaneous urge to action rather than inaction, and not as the outcome of a weighted average of quantitative benefits multiplied by quantitative probabilities.’⁴⁵ Keynes’s pregnant reference to animal spirits did not, unfortunately, give birth to a more extensive analysis of market expectations, though it certainly pointed to the need for one.

The General Theory is not a handbook of economic policy. Though its insights can be applied to particular conditions, its essential claims are universal. When an increasingly sceptical Robertson told Keynes in 1935 that he thought that much of the theoretical structure was mumbo-jumbo, Keynes responded sharply that ‘this book is a purely theoretical work, not a collection of wisecracks. Everything turns on the mumbo-jumbo …’⁴⁶ Unlike the Treatise, which had special reference to Britain as an open economy, subject to external forces, the General Theory applies in principle to a closed economy, just as the whole world is by definition a closed economy.

‘I have called my theory a general theory,’ Keynes wrote in the preface to the French edition. ‘I mean by this that I am chiefly concerned with the behaviour of the economic system as a whole, – with aggregate incomes, aggregate profits, aggregate output, aggregate employment, aggregate investment, aggregate saving rather than the incomes, profits, output, employment, investment and saving of particular industries, firms or individuals.’⁴⁷ Maybe he believed this at the moment when he put it quite like that; but the real reason why Keynes hit on this name for his book was surely that, with a nod towards Albert Einstein, whom he had met and admired, he wanted to claim its general status as opposed to a special theory, valid only under certain conditions.

Keynes was enough of an intellectual magpie himself not to spoil the sport of other magpies, stealing bright ideas. He was determined not to be possessive about insights that he readily acknowledged he had adapted from others. But he was clearly worried about attempts to assimilate selected aspects of his thinking into orthodox economic theory, as a special case. Thus the multiplier could be accepted – but not the equilibration of saving and investment via output changes. Or liquidity preference could be accepted – but not as the replacement of the traditional theory of interest, only as an interesting modification. So the Keynesian revolution might well be marginalised as a sub-species of ‘depression economics’, dependent on the rigidity of wages and a peculiar ‘liquidity trap’ that temporarily thwarted the proper workings of interest rate. But he envisaged a system that was symmetrical: in principle able to cope with the danger of inflation once full employment was reached.

By the forbidding standards of most treatises on theoretical economics, the General Theory is a good read. ‘The ideas which are here expressed so laboriously are extremely simple and should be obvious,’ says the preface.⁴⁸ Keynes claimed to be writing in a relatively technical way only because his fellow economists needed a professionally thorough de-bamboozling. For himself, he had no doubts that his own escape from his Marshallian education had been psychological as much as intellectual. As he told his unpersuaded friend Dennis Robertson in 1937, ‘I am trying all the time to disentangle myself, whilst you are trying to keep entangled.’ Hence Keynes’s exhilaration, once he believed that he had accomplished his revolution in economic theory – ‘I am shaking myself like a dog on dry land.’⁴⁹ He spoke in Stockholm about the General Theory shortly after its publication. ‘What I have to say intrinsically easy,’ read his lecture notes, characteristically adding: ‘It is only to an audience of economists that it is difficult.’⁵⁰