A CRITIQUE OF KEYNES PART ONE

Herewith the first part of as lecture I gave at Williams College a quarter of  century ago.  Although much has happened since, the central thesis is, I believe, even more pertinent now than it was then.  This is the first of three parts.

                                                            A Critique of Keynes

It is my purpose this afternoon to urge upon you the virtues of socialism. I shall argue that at the present time, the proper way to manage our economic life is through social planning founded upon collective management of the means of production. I shall not advance this proposition in the customary way, however, by telling you affecting tales of human suffering or by thundering at the injustices of the capitalist system. Human suffering we have aplenty in our society, and there is more than enough injustice to support a multitude of tirades. But I have charted a quieter course for myself this afternoon. I shall review with you something of the history and development of economic theory during the past two centuries; in order to show you why even the most sophisticated elaborations of modern neo-classical econometric model-building no longer serve to keep us from disastrous economic trouble. My argument will be quite simple and rather abstract, as benefits a philosopher, and there will be very little in it to puzzle or put off those among you who have not yet undertaken the study of economics. The heroes of my story are an Anglo-Jewish stock broker, an émigré from Soviet Russia, the son of an Italian jurist, and a left-Hegelian romantic philosopher -- which is to say, David Ricardo, Wassily Leontief, Piero Sraffa, and Karl Marx.

I begin by voicing what is, I hope, our shared agreement that the American economy is in a godawful mess. We will also agree, I hope, that the establishment of professional economists is no longer able to speak with solid, collective scientific confidence about the causes of our present disaster and the alternative possible cures. The present disarray of the economics profession is manifest in the self-doubts expressed in technical journals and presidential addresses to associations of economists, in the absence of consensus among those middle-of-the-road economists who are routinely looked to by both major parties for guidance and counsel, and in the bizarre emergence into respectability and even hegemony of such pseudo-science -- Voodoo economics, our vice-president once called it -- as supply-side theory.

 The profession of economics – or political economy, as it was once called – has gone through a succession of stages of confidence and self-doubt during the past two centuries. The triumphant reception of the elegant and powerful theories of David Ricardo in the first quarter of the nineteenth century gave way to increasing doubt and confusion as the century wore on, both because of the internal theoretical inadequacies of Ricardo’s principles and because of a succession of increasingly violent business cycles of boom and bust which threatened both social peace and the continued growth of capitalist enterprise. The new marginalist theories of the last third of the nineteenth century laid the theoretical foundations for a moral justification of capitalism, as set forth most famously in the work of J.B.Clark, and also for the scientific conclusion that in a properly functioning free market, there could be no sizeable long-term involuntary unemployment. The great depression after the first world war strained belief in this bit of economic science to the breaking point, for as the depression grew deeper and unemployment in the United States, for example, exceeded a quarter of the work force, it became more and more difficult to maintain that this violent deviation from rational, efficient full-employment was simply a short term frictional imperfection or a perverse consequence of sub-optional collective bargaining.

With the macro-economic theories of John Maynard Keynes, the economics profession entered a new era. If Keynes was right, then it ought to be possible to salvage capitalism by managing the swings from boom to bust with counter-cyclical actions by the central government. The state, it seemed, was not to be confined under capitalism to the role of night-watchman. Instead, it had a much more important task to perform, as the regulator of effective demand in the service of full employment growth.

 The great test of Keynesian theory came with the end of the second world war. In the United States and Western Europe, the experiment was made of capitalism with the state as a macro-economic governor. For a quarter of a century, the theory seemed to work with brilliant success. Unemployment was low, growth was steady and rapid, inflation was kept to an acceptable level and the inevitable business contractions were softened so as to be nothing more than pauses on the way to ever greater growth. It was during this period – the golden age of the profession of economics – that Nobel prizes began to be distributed to these physicians of the body politic. Clearly, ideology and politics had given way to science.

 Alas, the reputation of economics, like the housing and automobile markets, has soared to ever greater heights, only to plunge to new depths. As unemployment creeps toward ten percent and the United States, once proud leader in per capita gross national product, dips to tenth place behind even Italy, once the weakest partner in the Atlantic Alliance, it is perhaps time to stand back and reflect a bit on what has gone wrong.

My message today is scarcely original, but it is, I believe, true. I shall argue that it is fundamentally irrational to permit the direction and shape of our economic life to emerge as the unplanned consequence of a myriad of private decisions. For reasons which are simple but theoretically quite fundamental, an industrial economy in which there is both population growth and technological innovation cannot be relied upon to grow in a crisis-free manner. What is more, efforts to shape or direct an essentially private economy by fiscal or monetary policies of the central government are doomed to fail. The only rational solution, I shall suggest at the end of my talk, is a thoroughly socialised economy in which the course of capital accumulation and allocation, as well as the pattern of distribution of the social product, is made the object of collective social choice. 

 Let me begin my story almost two centuries ago, with the classical vision of a capitalist economy as a free market system in which the pattern of prices, of capital accumulation, of distribution, and of growth emerges as the unintended consequence of the interactions of rationally self-interested workers, consumers, and entrepreneurs. This vision, first adumbrated  by Adam Smith and brought to rigorous fulfillment in David Ricardo’s PRINCIPLES OF POLITICAL ECONOMY AND TAXATION, rested upon a number of simplifying assumptions, each of which in a different way made it possible for Ricardo and his followers to draw powerful theoretical conclusions from the premises of perfect competition, rational self-interested, and a free market.

 The assumptions fall into three categories: behavioral assumptions about the criteria or guidelines followed by the several classes of agents in their economic decisions; knowledge assumptions about the kind, degree, and accuracy of the information available to economic agents; and a rather special assumption about the nature of money.

 The behavioral assumptions are these: capitalists were assumed to be motivated entirely by a rationally self-interested pursuit of profit, which they measured in monetary terms. Confronted by a choice among several alternatives, capitalists could by and large be counted on to choose the most profitable. Furthermore, capitalists were conceived as perfect accumulators, so to speak. “Accumulate! Accumulate! That is Moses and the Prophets!” said Marx of the capitalist class. In effect, what this meant was that the total private consumption of the capitalist class was so small, in comparison with the aggregate output of the economy as a whole, that it could, for purposes of economic calculation, be treated as effectively zero. In part, of course, this assumption was justified by the relative smallness of the capitalist class as a group. Even a high standard of living per capitalist does not add up to much in the way of aggregate consumption. But more important was the behavioral assumption that capitalists, for whatever reasons, were driven by a desire to accumulate capital as rapidly as possible. To this end, they reinvested almost their entire profits in an expanded scale of operations. The classicals and Marx viewed this process of reinvestment as the primary engine of economic growth.

 The landlords – who loom larger in the writings of Smith and Ricardo than in those of Marx – are by contrast conceived as grasshoppers, as perfect consumers. They save nothing, and support a large population of unproductive servants. As a consequence, of course, the portion of profits diverted to them as rentals drops out of the economy and plays no role in the process of growth.

Finally, the workers are presumed to live at or near the subsistence level, with the consequences that as a class they do not save. In the short run, to be sure, individual workers may succeed in saving enough to rise into the class of small entrepreneurs, and during the transitional periods of labour shortage wages may rise significantly above subsistence for large numbers of workers; but both Ricardo and Marx thought that over the long run, external pressures of population or unemployment would hold wages at a level at which worker saving could not regularly take place.

 The effect of these behavioral assumptions, you will immediately realize, is to make certain theoretical calculations extremely simple. The quantity of capital directed toward investment will be exactly equal to aggregate profits. Effective final demand for consumer goods will simply equal rents plus wages. And so forth. To an extent that is not always recognized, particularly by those outside the profession of economics, the distinctive a priori character of the reasoning of classical economic theory derives directly from these behavioral assumptions.

 In addition to the behavioral assumption, classical theory makes a number of powerful simplifying knowledge assumptions. Consumers and sellers are presumed to have effectively a perfect knowledge of the market – what is being sold, where, and at what price. Entrepreneurs have perfect knowledge of the available production techniques, of the prices in factor markets, of the risks and constraints of different modes of production. Entrepreneurs know also how other entrepreneurs are doing, so that superprofits anywhere in the economy will fairly quickly draw interested investors to the favored sector. The result of these assumptions is to consign to the margins of the theory any consideration of market imperfections.

 The familiar model of a self-correcting economy in dynamic equilibrium, efficiently allocating available resources in response to consumer demand, rests upon these twin pillars of simplifying behavioral assumptions and the assumption of perfect knowledge. The theorems of the classical model, and also those of the neo-classical model as well, can be proved only because the knowledge and behavior of economic actors is conceived in this manner.

One final theoretical assumption underpins the classical theories, an assumption concerning the nature of money. It is an odd fact – odd, I might say, to the point of absurdity – that there is no place for real money as we know it either in the economic theories of the classicals or in the marginalist theories of modern economics. Ricardo conceived of money as simply one commodity among others – gold, of course, in the world he was looking at – whose value was determined, as was the value of every commodity, by the conditions of its production. An ounce of gold exchanged for so much linen, coal, or corn, he thought, because the conditions under which it was mined and refined bore a certain relation to the conditions under which linen was woven, coal dug, and corn grown. Just as a change in the techniques for growing corn would affect its value relative to other commodities, although in ways more complex than Ricardo had originally thought, so too a change in the techniques for mining and refining gold would alter the rate at which gold exchanged with other goods.

 The assumption of commodity-money, as it was called, an assumption which Marx shared, makes even modern capitalist economic transactions essentially sophisticated acts of barter. The grounding of the economic system in a physical base of commodity production is thereby rendered transparent, with the consequence that such phenomena is inflation, credit squeezes, the so-called money-illusion, and so forth can play no theoretical role in the analysis of the economy.

 The classicals concluded from their simplified model that in the absence of distorting interventions from the central government, a capitalist economy would function efficiently in long-run equilibrium. Marx sought to demonstrate that capitalism, even in this extremely simple conception of it, is internally unstable and prone to operational breakdowns. But Marx shared with the classicals all of the behavioral and knowledge assumptions I have just sketched.

 One of the first major departures from this simple model appears in John Stuart Mill’s PRINCIPLES OF POLITICAL ECONOMY, a book which, in many editions, was the leading economics textbook of the nineteenth century. In a little chapter entitled “Of Competition and Custom,” Mill introduces a new element into political economy which, I shall argue, destroys classical theory and eventually leads to an entirely different understanding of capitalism. It is completely typical of Mill that at one and the same time he understands the theoretical significance of what he is doing, and yet treats it as a mere adjustment to a theory which he is otherwise prepared to continue supporting. This paragraph, from the opening page of Mill's little chapter "Of Competition and Custom," is worth quoting:

 "So far as rents, profits, wages, prices, are determined by competition, laws may be assigned for them.  Assume competition to be their exclusive regulator, and principles of broad generality and scientific precision may be laid down, according to which they will be regulated.  The political economist justly deems this his proper business:  and as an abstract or hypothetical science, political economy cannot be required to do, and indeed cannot do, anything more.  But it would be a great misconception of the actual course of human affairs, to suppose that competition exercises in fact this unlimited sway.  I am not speaking of monopolies, either natural or artificial, or of any interferences of authority with the liberty of production or exchange.  Such disturbing causes have always been allowed for by political economists.  I speak of cases in which there is nothing to restrain competition;  no hindrance to it either in the nature of the case or in artificial obstacles;  yet in which the result is not determined by competition, but by custom or usage;  competition either not taking place at all, or producing its effect in quite a different manner from that which is ordinarily assumed to be natural to it."

 In his development of this idea, Mill emphasizes the customary character of groundrent, and also to a lesser degree of market prices for consumer goods. Now, I wish to suggest that once custom has been acknowledged as a determinant of any of the central economic variables, the entire edifice of classical economic theory must surely crumble. Before turning to the way in which Mill’s little point about custom has been expanded into a full-scale theoretical transformation of the classical theory, let me explore for a bit the significance of the elementary example Mill offers.