The Hayek-Keynes Debate, 1931-1971

Mises Daily by Sudha R. Shenoy

The roots of current economic ideas and of those guiding wage policy lie in the 1930s, in discussion inspired by the publication of the General Theory. Though Keynes's ideas diverged significantly from the theoretical structure of Pigou and Marshall, with which he was most familiar, "Keynesian" ways of thinking had been fairly widespread in Britain and the United States before the General Theory appeared in 1936.[1] Keynes provided a theoretical foundation for these new ways of thinking.

Since the publication of the General Theory, there has been an extensive elaboration of the theoretical system outlined in or generally associated with it, together with a further development of an alternative system of concepts called the classical system. This was close to a mirror image of the Keynesian system,[2] in the main relationships (e.g., between the quantity of money and total expenditure; between interest, saving, and investment; between the wage level and the level of employment, and so on). But whereas the Keynesian system was couched wholly in terms of aggregates, the so-called "classical" system contained what may be termed a price dimension: the changes in the price "level" associated with changes in the total money stock were held by the classical system to imply equiproportional changes in all prices, and variations in the price level in turn were associated with changes in the level of economic activity. In a sense, the Keynesian approach may be regarded as a logical extension and elaboration of this rather crudely aggregative element in the "classical" system.

Challenge to Keynes

The doctrines generally accepted among English economists contemporaneous with Keynes were challenged, in fundamental respects, by an alternative analysis, developed on the Continent, and propounded in Britain by Professor Hayek. But by the 1940s, the Keynesian approach was almost universally adopted by economists. Initially, many appeared to believe that the "macro" problems of unemployment and depression were solved and that few other major economic problems would emerge. The only problem remaining, it seemed, was the methods required to ensure "full" employment.

"Now that the principle of adequate effective demand is so firmly established," declared Professor Arthur Smithies, "economists should devote particular attention to defining the responsibilities of the state."[3]

The British White Paper on Employment Policy in 1944 and the full employment commitment in the UN Charter reflected this belief, as did the 1946 Employment Act in the United States.[4]

A few dissenting voices warned of trouble ahead. Professor Jacob Viner observed of a report to the Economic and Social Council of the United Nations, National and International Measures for Full Employment, prepared by a group of distinguished economists (J.M. Clark, A. Smithies, N. Kaldor, P. Uri and E.R. Walker):

The sixty-four dollar question with respect to the relations between unemployment and full employment policy is what to do if a policy to guarantee full employment leads to chronic upward pressure on money wages through the operation of collective bargaining. The authors take a good look at the question — and run away.

Effective demand to provide employment was the "key concept" in recommendations that Professor Viner rated as "much more Keynesian than was the final Keynes himself…"[5]

Shortly after the General Theory appeared, Professor W.H. Hutt argued that it was a specific for inflation.[6]

Even Keynes had doubts, a few years after the General Theory. In his essay How to Pay for the War (London: Macmillan, 1940) he warned the trade unions of the "futility" of demanding an increase in money rates of wages to compensate for every increase in the cost of living. To prevent inflation, he insisted,

Some means must be found for withdrawing purchasing power from the market; or prices must rise until the available goods are selling at figures which absorb the increased quantity of expenditure — in other words the method of inflation.

And in a discussion of financing war expenditure:

[A] demand on the part of the trade unions for an increase in money rates of wages to compensate for every increase in his cost of living is futile, and greatly to the disadvantage of the working class. Like the dog in the fable, they lose the substance in grasping at the shadow. It is true that the better organised might benefit at the expense of other consumers. But except as an effort at group selfishness, as a means of hustling someone else out of the queue, it is a mug's game …

The Approach to an Income Policy

Over the following 25-odd years, the early Keynesian theory was further elaborated and refined and a highly sophisticated series of macroeconomic models developed. The 1950s more especially saw the discovery of "cost inflation," in which a rise in wages pushed up the cost level. As prices were determined by costs, and, in crucial sectors of the economy, were "administered" on the cost-plus-markup practice, prices rose to protect profit margins. But since wages were also incomes, the cost and price increases had no deflationary effect, as effective demand rose simultaneously.[7]

In these circumstances a contractionary monetary/fiscal policy would be deflationary: it would lead to socially intolerable levels of unemployment and excess capacity; an alternative measure, directed specifically at rising costs, would have to be devised. If price stability and full employment could both be achieved by keeping wage increases within the limits set by rises in productivity, this implied an "income policy." Further investigation into the implications for the price- and wage-level of linking sectoral wage increases with productivity strengthened the case for a nationally determined "wage policy" covering both relative wage rates and the general wage level. If wages rose in the sectors where productivity was rising, the result would be a rise in demand for the outputs of other sectors, resulting in a rise in their prices.[8]

Economic policy in the United Kingdom and the United States, from 1950 on, reflected the adoption of these views; there was a gradual shift from exhortations, guidelines, and pay pauses to more direct attempts to influence and control wages.[9]

That such direct control of wages and prices would be needed to forestall the "vicious wage-price spiral"[10] resulting from full employment had been forecast by Lord Beveridge as early as 1944.

By the late 1960s and early 1970s more economists came to favor an income policy, some reluctantly (Robbins, Meade, Paish, Brittan, Morgan),[11] others enthusiastically (Balogh, Streeten, Opie).[12]

Lord Robbins's case is particularly interesting. In the early 1950s, he analyzed clearly the inflationary implications of the full-employment policy contemplated by Beveridge: it gave union leaders a virtual guarantee that

whatever [wage] rates they succeeded in getting, unemployment would not be permitted to emerge.[13]

It would give them a continuous incentive to push wages beyond increases in productivity, setting off a "vicious spiral" of "more inflation." This, in turn, might force governments to act directly on wage rates.

The present determination of wages by bargain between employer and employed would be suspended. Wage-fixing by the state would take its place.

He believed, however, that this alternative would be rejected "on the ground that in the end its efficient operation would prove to be incompatible with the continuation of political democracy…."[14]

Seventeen years later he argued for an income policy as temporary "shock tactics," to afford a "breathing space" in which fundamental monetary-fiscal reforms might be "advanced and understood."[15]

Despairing of the good sense of union leaders, he sought to bring pressure on them indirectly, suggesting that businessmen be restrained from granting inflationary wage increases by restrictions on aggregate demand, even to the point of precipitating bankruptcies, thus preventing the payment of higher wages that would simply be recouped by higher prices. A suggested alternative or parallel measure would be to tax inflationary wage increases granted by firms. He hoped that union leaders' expectations of automatic increases in wages would thereby be frustrated. (A similar view is taken by Professors E. Victor Morgan, F.W. Paish, and Sidney Weintraub.)[16]

An alternative type of income policy was proposed by Mr. Samuel Brittan.[17] The government would control the level to which wage rates would be permitted to rise while allowing employers short of labor to offer higher rates, but without pretending to determine relative wage rates on the basis of social justice. Such a policy, he said, must be treated as a supplement to monetary and fiscal policies that provide sufficient demand to prevent unemployment, but prevent the emergence of excess demand. He suggested as a stopgap a temporary price-and-wage freeze until these policies were implemented.

Two possible implications of this suggestion may be considered. First, if such a brake on wage increases is to be more than advice, unions must be willing to accept the guidance of the income authority — implying a permanent watchdog role for the authority (or at least an existence parallel to that of the unions-as-wage-fixers). If the unions refuse to cooperate presumably the authority will have to take over their wage-fixing function.

Secondly, in common with other recommendations for incomes policies, this proposal would perpetuate a given structure of relative wage rates since all the rates to which it applied would be allowed to rise only by a given percentage (save in "labor scarcity"). This relative wage structure today reflects not so much the allocative forces of the market but the relative power or "pushfulness" of the different unions.[18] Can we assume that they would be content to retain indefinitely whatever relative positions they had achieved at the moment the income policy came into existence?

"Micro" Dimensions Acknowledged

The common thread running through these discussions is the alleviation of specific wage-rate maladjustments. They have moved some distance from the aggregative analysis. The "macro" problem of adequate demand management has, it now appears, a "micro" dimension: that of establishing (or obtaining) an "appropriate" scale of prices. In other words, from the viewpoint of practical policy, the "macro" problem of a persistent upward push (or pull) on the price "level" is now seen to have "micro" roots, in the specific "pricing" methods used by specific groups of workers. "Macro" measures acting on aggregate expenditure may have allowed us hitherto to ignore this basic micro discoordination,[19] but events have seemingly brought the issue forward unavoidably. "Macro" measures, it would appear, may offset micro problems but are no substitute for appropriate micro solutions.

The significance of coordination at the micro level appears here, in the light of a third type of analysis, which Professor Hayek developed, on foundations laid by the "Austrians": Menger, Wieser, and Böhm-Bawerk, culminating in the works of Mises. Hayek concentrated on an analysis of the structure of relative prices and their interrelations. He did not adopt the framework of a general-equilibrium system, nor treat price changes as elements in a "dynamic" shift between two general equilibria. He regarded prices rather as empirical reflectors of specific circumstances and price changes as an interrelated series of changes in these "signals" that produced a gradual adaptation in the entire price structure (and hence in the outputs of different commodities and services) to the constant, unpredictable changes in the real world. Pricing, in short, is seen as a continuous information-collecting-and-disseminating process, but it is the institutional framework that determines both the extent to which, and the degree of success with which, prices are enabled to perform this potential signaling or allocative function.

This "Austrian" analysis constitutes a substantial break with classical economic theory from Adam Smith to J.S. Mill. It differs also both from the doctrines of the English economists after Mill and from the theoretical preoccupations of the Lausanne School with the conditions of general equilibrium.[20]

Is There a Price "Level"?

In his first English work, the four lectures published as Prices and Production,[21] Hayek questioned the concept of a price "level" — i.e., a relationship between the total money stock and the total volume of production, variations in this "level" being associated with variations in aggregate output. He argued that such a concept failed to show that there were specific influences of changes in the stream of money expenditure on the structure of relative prices, and hence on the structure of production.[22]

These price and output changes, he maintained, occurred irrespective of changes in the price level. Hayek's analysis implied that if "the" price "level" is held "stable" by offsetting monetary measures, under conditions where the relative price changes would result in a falling price "level," the real dislocations would be the same as if prices were made to rise by monetary measures, if otherwise they might have remained "stable." In either case, the outcome is a painful correction of the preceding real misdirection, i.e., a "depression."

During the 1920s, the widespread theoretical and policy influence of the "stabilizations" meant that considerations of the kind sketched by Professor Hayek were not incorporated into either theoretical or policy analysis; consequently, the price "level" "stability" of the period was read as implying a lack of maladjustment in the underlying price structure. (This is an extremely oversimplified summary of a complex historical situation, the specific conditions of which were not uniform in all countries.)

Theoretically and practically, it may be argued that in conditions of "depression" there is little choice save to augment the level of monetary expenditure to the highest possible degree. Hayekian analysis, while readily conceding that depressionary symptoms may thus be overlaid, would argue that the problems are then transformed into those arising out of a situation where every reappearance of recessionary symptoms has to be met by ever-larger increases of monetary expenditure, eventually issuing in the "stagflationist" dilemma.

This is not necessarily to say that the specific policies pursued in the 1920s and 1930s, or the economic and monetary framework of the time, represented an approximation to the Hayekian ideal. Hayek has said with regard to the period 1927–1932:

[U]p to 1927, I should, indeed, have expected that because, during the preceding boom period, prices did not rise — but rather tended to fall — the subsequent depression would be very mild. But, as is well known, in that year an entirely unprecedented action was taken by the American monetary authorities, which makes it impossible to compare the effects of the boom on the subsequent depression with any previous experience. The authorities succeeded by means of an easy money policy, inaugurated as soon as the symptoms of an impending reaction were noticed, in prolonging the boom for two years beyond what would otherwise have been its natural end. And when the crisis finally occurred, for almost two more years deliberate attempts were made to prevent, by all conceivable means, the normal process of liquidation. It seems to me that these facts have had a far greater influence on the character of the depression than the developments up to 1927, which, from all we know, might instead have led to a comparatively mild depression in and after 1927.[23]

$32 $25

Shortly after the publication of the first edition of Prices and Production, Professor Hayek published (in Economica) the first part of a long, substantive review of Keynes's Treatise on Money.[24] This provoked a reply from Keynes, followed by a rejoinder, before the publication of the second part of the review. Hayek criticized Keynes for his neglect of the real structure of production, arguing that Keynes's predilection for concentrating on the immediate and purely monetary phenomena accompanying changes in money expenditure, together with his penchant for aggregative macro concepts (total profits, total investment), had led him into contradictory or untenable conclusions. Keynes apparently held that if there were no entrepreneurial profits or losses in the aggregate, total output would be held constant. Hayek replied that if profits in the "lower" stages of production (nearer consumption) were exactly counterbalanced by losses in the "higher" stages, there would be a contraction in the capital structure and a fall in output and employment — even though there were no aggregate profits or losses.

In his reply, Keynes failed to take up the numerous substantial criticisms made by Hayek. The main point of interest is his explicit statement that "in my view, saving and investment … can get out of gear … there being no automatic mechanism in the economic system … to keep the two rates equal." Hayek's reply to this was based on his analysis of the relative price structure:

Mr. Keynes's assertion that there is no automatic mechanism in the economic system to keep the rate of saving and the rate of investing equal … might with equal justification be extended to the more general contention that there is no automatic mechanism in the economic system to adapt production to any other shift in demand.

Further Implications of Hayekian Analysis

There are further implications of the Hayekian approach:

1. If the current level of output and employment is made to depend on inflation, a slowing down in the pace of inflation will produce recessionary symptoms. Moreover, as the economy becomes adjusted to a particular rate of inflation, the rate must itself be continuously increased if symptoms of a depression are to be avoided: to inflate is to have "a tiger by the tail."

2. To limit price or wage-rate increases by an income policy is to freeze a particular set of price and wage-rate interrelationships while underlying circumstances of supply and demand are continually changing. This is like the "stability" of a set of defective gauges perpetually pointing to the same set of readings. It reinforces other institutional factors preventing the specific changes in relative prices and wage rates necessary to the maintenance of "full" employment. Or, to put this same point from a different angle, if "full" employment is to be maintained at union-determined wage rates (which are inflexible downwards), all other prices and wage rates must be adjusted to them: other prices and wage rates must be set at, or reach, levels consistent with this objective. Even if union-determined wage rates were held down to a maximum percentage increase, it still does not follow that the same percentage increase, or a lesser increase, in all other prices and wage rates would suffice to achieve "full" employment. This is why it may be necessary for incomes to rise faster than output, even to secure that increase in output.[25]

   

3. The major objection to an income-policy approach is that it merely freezes a given array of prices and wage rates. It does nothing to bring about coordination or to introduce coordinative institutions into the labor sector. So long as the discoordinative potential of such nonmarket institutions as unions is not tackled, the problem will recur again and again. There may be no substitute for a very painful reshaping of institutions or other means of bringing within the ambit of the pricing system wage rates made impervious to market forces.[26] The alternative is a permanent income policy: an all-around fixing of wage rates and prices, i.e., an effectively centrally controlled economy (with all its problems), although we may, as it were, "back into" this situation unintentionally. This is to say nothing about whether it is desirable politically.[27]