Rebellion and conformism in Keynes

Peter Lilley

For most of the last forty years the average student of economics in Britain or America has been shielded from any systematic critique of the Keynesian system. He is, of course, given an account—often little more than a caricature—of the obscurantist resistance put up by die-hard 'classical' economists to the original Keynesian Revolution. Indeed, part of the appeal of the Keynesian system is that it presents itself as revolutionary and non-conformist whilst in fact being thoroughly established. Thus it satisfies the twin but conflicting desires of most young men—to rebel and to conform. But students have not, at least until recently, been taught about the genuinely non-conformist schools of thought which have continued to develop detailed criticisms of the Keynesian orthodoxy.

This rather unacademic taboo certainly still operated at Cambridge a decade ago. I only stumbled on its existence because I happened to be too obtuse to grasp the sophisticated Keynesian justification for state control of the economy. My supervisor therefore reluctantly introduced me to some authors who would help me 'to express my outmoded prejudices in rigorous academic form'. But I was first required to swear not to tell anyone that I had been advised to read Friedman and his Chicago colleagues! Thus only under a veil of secrecy could one even discover the existence of an alternative to the Keynesian approach.

The existence of a second distinct, though overlapping, anti-Keynesian school —the Austrian tradition under von Mises and its present doyen Professor Hayek— was an even more closely guarded secret.

In recent years the Anglo-American academic establishment has at last been forced to grant diplomatic recognition to both the Chicago and Austrian schools. Not only have they failed to wither away in academic isolation, they show every sign of flourishing at a time when the established Keynesian doctrine is proving theoretically sterile and politically inoperable.

Before this reversal of fortunes can be understood it is helpful to understand why the Keynesian hegemony came about. The original victory was amazingly rapid compared with most intellectual revolutions. Keynes published his General Theory in 1936. By the outbreak of war it was firmly established and opposition within the academicestablishment had virtually ceased.

The reasons for this sweeping advance are doubtless manifold. However, the General Theory did not catch on because of its lucidity and simplicity. It possesses neither characteristic. Professor Samuelson, a leading Keynesian whose best-selling textbook has done much to spread the Keynesian system, wrote of the General Theory: 'It is a badly written book, poorly organised; any layman who, beguiled by the author's previous reputation, bought the book was cheated of his five shillings. It is not well suited for classroom use. It is arrogant, bad-tempered, polemical and not overly generous in its acknowledgements. It abounds in mares' nests and confusions. . . In short, it is a work of genius'.*

Even acknowledging that genius lay beneath the obscurity and confusions it is amazing that people perceived it so speedily.

One major factor underlying the success of the General Theory was that it provided, or purported to provide, a way out of a very uncomfortable dilemma. Contemporary economics taught the economics profession something they did not want to believe— namely, that the higher the real wage stipulated by workers the fewer would be employed. This implied that unemployment could be cured if trade unions permitted workers in the hardest-hit industries to accept lower wages. Yet academics, being comfortably off and securely employed, felt acutely embarrassed at advising workers to accept lower wages. Keynes explicitly recognised this embarrassment. 'A classical economist may sympathise with labour in refusing a cut in its money-wage . . . but scientific integrity forces him to declare that this refusal is nevertheless at the bottom of the trouble'.1.

In the General Theory Keynes sought to show that rigid wages were not the cause of, and wage cuts not the cure for, unemployment. He claimed to have constructed a General Theory, of which classical economic theory was just a special case valid only at full employment. Small wonder that economists should lap up with enthusiasm a book which claimed to show them how to salvage both their consciences and their existing economic expertise.

To claim to have discovered such a theory does not necessarily mean that it exists or is true. Indeed Keynes's disciples have been arguing about what his theory is for several decades (which may cast doubt on its existence), though they have never doubted its truth ! Incidentally, this lengthy exegesis has finally generated some sort of consensus among Keynesians that, in the words of Leijonhufvud, 'his General Theory is but a special case of the classical theory obtained by imposing certain restrictive assumptions on the latter'. In short, the book which

aimed to reduce classical economics to a special case of a more general theory has suffered the reverse fate. That the economics profession should nonetheless value Keynes's contribution so highly may in part be explained by Harry Johnson's quiP that 'the effort required to open the oyster led those who were successful to overvalue the pearl within::

The very obscurity of the General TheorY was one of its strengths. Disciples could believe it contained a revolutionary alternative to classical economics. Critics could be declared to have misunderstood what Keynes 'really meant'. However, Keynes did subsequently define the elements he considered to be crucial to his theory.**

The twin pillars of his system were his

explanations, of what determines the two components of effective demand—con

sumption and investment.

Consumption is determined by a stable 'psychological law. . . that when aggregate income increases, consumption expenditure will also increase but to a somewhat lesser extent'. Keynes emphasised that this apparently banal 'psychological law was of the utmost importance in the develop

ment of my own thought'.

This 'law' has two contentious implications. First, if real incomes are tending to rise, consumption will tend to lag behind; hence, unless investment expands to absorb the growing savings, total effective demand will not grow rapidly enough to keeP everyone employed. Thus there may be an inbuilt tendency for unemployment to rise as the economy gets richer. Second, Keynes's consumption function assumes consumers to be mechanical and irrational. Their savings are determined purely IV current income not by expected l'uture income or accumulated savings.

Keynes's second pillar, investment, depends on the interest rate which Is determined by 'liquidity preference'. 'Liquidity preference' is a measure of people s desire to hold their assets in liquid iform

(money) at the sacrifice of interest ylelded

*The Development of Economic Thought edited by H. W. Spiegel (Wiley 1952) page 767 +General Theory, page 16 t'The General Theory After 25 Yf!ars' in Ott Econotnics and Society by Harry Johnson **The General Theory of Employment Q.J.E., vol. 51, 1937 by less liquid instruments (bonds). When People were apprehensive they would, according to Keynes, have a greater liquidity preference, i.e. they would prefer to hold money until the interest rate increased to restore a balance between the supply of money and bonds. The higher interest rate would then reduce investment. Conversely increasing confidence would reduce liquidity preference, lower interest rates and encourage investment. Unfortunately according to Keynes the collective liquidity Preference' (which in modern jargon is an aspect of the 'demand function for money') IS highly unstable. There is, Keynes believed, no rational way of knowing what the future holds. So people tend to follow the collective mood which sways irrationally from gloom to elation and their demand for money changes correspondingly.

The Friedmanite critique of Keynesianism has taken Keynes at his word and concentrated its fire on these twin pillars of his system—the consumption function and the demand for money (liquidity preference).

Keynes had stated that his 'consumption function' was a 'fundamental psychological law, upon which we are entitled to depend both a priori from our knowledge of human nature and from the detailed facts of experience'. However, Friedman and the Pre-Keynesian economic theory to which he remained attached had a different view of human nature and he set about showing that his view generated a consumption function more in accord with the 'detailed facts of experience'. The classical view of economic man is that he is, on the whole and O n average, a rational being who attempts to make the most of the opportunities open to him. A rational man will divide his income between spending and saving in the light of his expected future income and his existing assets. This is in marked contrast to Keynesian man whose consumption and saving are determined almost exclusively and quite mechanically by past income,

If the Keynesian view is correct two Predictions follow: first, rich people would save a higher proportion of their income than poor people and, second, if society as a whole gets richer the proportion of national income saved would tend to rise. The first prediction was soon found to be Perfectly true. However, Simon Kuznets, Friedman's colleague at Chicago, showed the second prediction to be false. Savings have been a fairly constant ratio of national income in the long term.

Friedman's achievement was to explain these two apparently contradictory findings and show them to be compatible with his OW n theory of rational consumer behaviour. He realised that snapshot studies of savings Patterns in one period concealed the fact that some people (authors, stockbrokers, farmers) have volatile incomes. In years When their income is high they will appear among the rich and will, if they are rational, save a disproportionate amount for lean Years. Those who are having a bad year will

appear among the poor and will draw on past savings so their current rate of savings will appear negative. Thus mud of the apparent difference between the savings habits of rich and poor, which had seemed to confirm Keynes's mechanical consumptibn function, could be explained as a rational response to volatile incomes.

Friedman elaborated his findings into a consumption function which fitted the data quite well (by no means perfectly but far better than did Keynes's).

Friedman's consumption function differs from Keynes's in that it implies that consumer spending habits are stabilising,a fall in national income will normally lead to an increase in the proportion of income consumed. Moreover, the spectre of underconsumption in the long term as savings rise faster than investment appears decidedly less likely.

Having demolished one pillar of the Keynesian system Friedman set about the other: the instability of the demand for money. In a massive work of scholarship Friedman reconstructed the Monetary History of the United States, 1867-1960. He showed therein that on the whole people's willingness to hold money balances has been remarkably stable. Moreover the main periods of economic instability have been associated with erratic movements in the supply of money which is determined by the monetary authorities. For example the Great Depression of 1929-33 was accompanied (and in Friedman's view made so severe and prolonged) by a contraction of no less than a third in the US money supply. The lesson is clear: whereas Keynesian theory suggests the herd instinct of businessmen causes booms and slumps, Friedman's examination of the facts suggests governments are the major source of instability.

Subsequently Friedman and other members of the Chicago school began to test quantitative equations for the demand for money. They are satisfied that the results show demand for money to be a reasonably stable function of a number of variables. Apart from suggesting that individuals are more rational about the form in which they hold their assets than Keynes believed, these • money demand functions provide an alternative to the Keynesian multiplier for forecasting shortterm economic movements. Often the two techniques result in identical forecasts but when they have differed the monetarist approach has proved more frequently correct.

The Austrian critique of the Keynesian system is in a sense far more fundamental. Hayek rejects the whole Keynesian approach, whereas Friedman merely denies the empirical truth of Keynes's main postulates. Consequently the Austrian attack is to some extent directed at Friedman as well as at Keynes.

Hayek's first objection to the Keynesian approach is its 'conceptual realism' (the tendency to ascribe a real existence to

arbitrary statistical aggregates like Consumption, the Wage Level or Capital). To Hayek the Wage Level is as unreal as a mediaeval 'humour'. It is simply the statistical average of all individual wages. This average cannot enter into actual economic relationships. Only individual wage rates can do that. This may sound pedantic. But constant reference to 'the wage level' has obscured from a generation of economists the fact that unemployment may exist because the structure of relative wages is wrong, even though the average 'wage level' may be 'correct'. We still suffer from this type of thinking manifested in such policies as the £6 wage increase. Blanket implementation of this regardless of the supply and demand situation in different occupations is undoubtedly pricing some workers out of jobs even though shortages of labour persist elsewhere.

Hayek's second criticism derives from his total scepticism about the existence of quantitative functional relationships in economics. The behaviour of myriads of individuals in a constantly changing society is too complex to be encapsulated in an arithmetical formula. Thus Keynes was not merely wrong in his formulation of the consumption function but foolish to believe any fixed relationship between consumption and income should exist. And Friedman, too, will find his consumption and money demand functions need constant modification to accord with changing circumstances. Obviously Hayek's attitude destroys the rationale of most modern econometric work which helps explain why it is unpopular in the profession!

Hayek's third critique of Keynes is directed at his capital theory. Because Keynes treats capital as a homogeneous substance he ignores its complex interlocking structure—remove one component (say a steelworks) and many other components (e.g. the iron and coal mines and steel fabricating plants) become useless. The market can only build up such a properly coordinated capital structure if the market signals (prices and interest rates) give a correct indication of the availability of resources. Keynes's recommendation that interest rates be artificially depressed to stimulate employment ignores this. Low interest rates indicate a cheap and plentiful supply of savings. Entrepreneurs will therefore initiate mutually complementary projects requiring much money and time to complete. This will indeed stimulate employment. But wages and consumption will therefore rise, diverting resources away from capital investment. The entrepreneurs will then find their supply of cheap credit choked off and incomplete capital projects left on their hands. Moreover, even if some plants are completed with the now reduced supply of savings they are of diminished value without the expected complementary plants originally envisaged by the market. A painful period inevitably ensues during which firms are liquidated and their plant and labour are reallocated to form a viable and properly coordinated structure of production. The results of Keynesian cheap money policies are therefore boom and bust.

The Austrian school therefore differs from both Keynes and Friedman in its interpretation of the Great Depression (vide Murray Rothbart's book of that name). Whereas Keynes pins the blame on the irrational herd instinct of businessmen and Friedman blames the Federal Reserve Board for allowing the money supply to contract, the Austrians blame the massive unsustainable credit expansion during the 1920s. And they believe the resultant crisis was perpetuated by attempts to resist the liquidation of discoordinated capital projects.

Arising in part from this analysis of the consequences of unwise monetary policy comes Hayek's most recent departure from both Keynes and Friedman. Whereas Keynes depicts the role of monetary policy as to offset the instability of liquidity preference and Friedman believes it should be to stabilise the growth of money supply, Hayek recommends that government relinquish its control of money altogether! He believes that the legal tender laws which oblige us to use government currency should be abolished, leaving people free to choose the currency they prefer. This would prevent governments from distorting the economy by Keynesian cheap money policies and would be a more effective discipline than Friedman's monetary role.

If proof were needed that Keynes's critics have regained the intellectual upper hand it is that such once outré ideas can now be discussed in Cambridge common rooms without fear of reprisal!