Quarterly Journal of Austrian Economics

Home | Mises Library | Why Your Grandfather's Economics Was Better than Yours: On the Catastrophic Disappearance of Say's Law

Why Your Grandfather's Economics Was Better than Yours: On the Catastrophic Disappearance of Say's Law

Tags Austrian Economics OverviewOther Schools of ThoughtProduction TheoryValue and Exchange

02/25/2019Steven Kates

I would like to thank Joe Salerno for his kind invitation to present the Ludwig von Mises lecture for 2010, in this way giving me the opportunity to present these views on the classical theory of the cycle and Say’s Law to a wider audience. I would also like to thank my good friend, Peter Smith, who provided excellent and sympathetic advice on an earlier draft of this paper.

Ludwig von Mises Lecture
Austrian Scholars Conference
Ludwig von Mises Institute
Auburn, Alabama
March 13, 2010

It is a great honor for me to have been asked to present the Ludwig von Mises lecture here at the Austrian Scholars Conference.

Let me begin with a story. When I came to select my list of the ten most influential economists of the twentieth century in an article published in the Canberra Times on December 1999, an article which can still be found on the Societies for the History of Economics website, the economist I chose as the most influential—not the best nor the greatest, mind you, but as the most influential—was John Maynard Keynes. No one, I regret to say, has had more influence than Keynes.

Then, second on my list, was Friedrich Hayek, placed there because of his recognized relevance for the economies of Eastern Europe that were then emerging from beneath the horrors of their communist regimes.

But third was Ludwig von Mises, who might have just as easily been second, about whom I wrote these words:

Ludwig von Mises took the fight up to the socialist dogmas of the early twentieth century and showed on paper that no economy could ever solve the problem of allocating resources without a price mechanism, free markets and private property. Who doesn’t know it now? He knew it eighty years ago.

Ludwig von Mises is an economist for whom I have had the greatest imaginable regard which is why having been given this opportunity to speak to you today means as much to me as it does.

And in beginning this address, I would like to mention something that Mises and I have in common. He had been for twenty-four years the economist for the Austrian Chamber of Commerce. Well, as it happens, I had myself been, also for twenty-four years, the economist for the Australian Chamber of Commerce.

And while to some extent this is mere coincidence, I believe that for both of us, as the economic representative of the business communities in both of our countries, even though more than two generations apart, it was this experience that allowed us to understand the workings of an economy with certain kinds of insights that may generally not be appreciated by others.

But it was one aspect of my work that ended up having an immensely large impact on my life, and that is the discovery of Say’s Law for myself. It is because I reinvented this principle that I believe I understand it so well.

And what happened was this. As part of the way in which the Australian economy is managed, we have what was once known as the National Wage Case. It is a court case in front of a panel of industrial relations judges who at the time determined the level of wages for something like ninety percent of the working population.

And as part of the union claim for higher wages, it was always argued that increased incomes would be good for the economy because it would increase demand. I would counter this by pointing out how useless it would be for businesses to find their revenues increased through first increasing their costs by an equivalent amount. And then, a year after I had formulated this argument, I came across the identical argument in a passage in an essay by John Stuart Mill, published as long ago as 1844. This is what Mill wrote:

The utility of a large government expenditure, for the purpose of encouraging industry, is no longer maintained.... It is no longer supposed that you benefit the producer by taking his money, provided you give it to him again in exchange for his goods. There is nothing which impresses a person of reflection with a stronger sense of the shallowness of the political reasonings of the last two centuries, than the general reception so long given to a doctrine which, if it proves anything, proves … that the man who steals money out of a shop, provided he expends it all again at the same shop, is a benefactor to the tradesman whom he robs, and that the same operation, repeated sufficiently often, would make the tradesman’s fortune. (Mill, 1874 [1974])

Although it would be years before I would work this out, what Mill wrote is based on a proper understanding of Say’s Law. High levels of public spending do not encourage industry. Spending does not of itself create growth and employment. You cannot make an economy prosper through expenditure but only through value adding production. Demand does not drive an economy forward, nor does demand deficiency cause recessions.

It was this most fundamental of all economic propositions that Keynes deliberately and willfully destroyed. Say’s Law has, for all practical purposes, now disappeared from economic discourse and policy. And until it returns, the ability for the economics profession to provide sound and sensible advice during recession will remain sharply constrained. But to understand what Say’s Law means one must first understand the role Say’s Law played in the Keynesian Revolution.


The Keynesian Revolution, and therefore the origins of virtually all macroeconomic theory today, can only be understood in relation to Keynes’ coming across Malthus’ economic writings in 1932. In particular, it was his reading of the Malthus side of the Malthus-Ricardo correspondence, which had been unearthed in 1930 by his close associate Piero Sraffa, that turned Keynes’ mind to the possibility of demand deficiency as a cause of recession. Until that time, economists had been near unanimous in arguing that insufficient demand as a cause of recession was fallacious.

There has been universal recognition amongst historians of thought that something does happen in late 1932 to turn Keynes in a new direction. Yet not one of the works devoted either to understanding the nature of the Keynesian revolution nor to examining the road between the Treatise on Money published in 1930 and the General Theory published in 1936, has suggested that the reason for this change in direction occurs specifically because Keynes was at that time updating his essay on Malthus.1 Indeed, there is no reason given of any kind why at that particular moment Keynes came to the conclusion that demand deficiency was the missing link in the theory of the cycle. Yet it is as close to a certainty as one can have in such reconstructions that Keynes would never have written the General Theory as he did, focusing on demand deficiency, had he not become deeply interested at the end of 1932 in Malthus’ economic writings. It was Malthus, of course, who had been the leading advocate in the nineteenth century of demand deficiency as a cause of recession and of increased levels of unproductive spending as the cure. Reading Malthus’ letters to Ricardo, and then the text of Chapter VII of Malthus’ Principles, ought to be recognized as the single most important reason why Keynes was to write what he wrote in the way he did.

Recognizing that this was the inspiration should make it easier to understand what the intent of the General Theory was and to understand the nature of the change in economic theory that occurs as a result. In the General Theory Keynes is very clear about what he has learned from reading Malthus.

The idea that we can safely neglect the aggregate demand function is fundamental to the Ricardian economics, which underlie what we have been taught for more than a century. Malthus, indeed, had vehemently opposed Ricardo’s doctrine that it was impossible for effective demand to be deficient; but vainly. For, since Malthus was unable to explain clearly (apart from an appeal to the facts of common observation) how and why effective demand could be deficient or excessive, he failed to furnish an alternative construction; and Ricardo conquered England as completely as the Holy Inquisition conquered Spain. Not only was his theory accepted by the city, by statesmen and by the academic world. But controversy ceased; the other point of view completely disappeared; it ceased to be discussed. The great puzzle of Effective Demand with which Malthus had wrestled vanished from the economic literature. (Keynes, 1936, p. 32, emphasis added.)

It was the “great puzzle of Effective Demand” that Malthus had been wrestling with which had disappeared and it was this that Keynes was intent on restoring to economic theory.

Nor was Keynes wrong on the implications of Say’s Law to his contemporaries. It was precisely this issue that is the dividing line between pre-Keynesian economics and the economics that has dominated theory ever since. Mainstream economists before 1936 had actively denied any role for aggregate demand in understanding the business cycle. Although there had been some attempts to overturn the law of markets, demand deficiency as an explanation for recession was until then almost entirely the province of cranks.2 The two most important diagrammatic innovations of the 1930s were the IS-LM curves published by Hicks in 1937 and the Keynesian cross diagram first published by Paul Samuelson in 1939 (see Schneider 2010). Both were developed in response to Keynes’ General Theory, and both feature in economics texts to this day.

The problem of recession as conceived in the General Theory was that an economy, once it has passed a certain level of production, will run out of demands for the goods and services it produces.

This is not excess supply for individual goods and services, the “particular glut” whose existence no one had ever denied, but an actual excess supply of all goods taken together, that is, a “general glut.” Keynes made the possibility of demand failure the culminating point at the end of the introductory chapters of the General Theory.

The celebrated optimism of traditional economic theory, which has led to economists being looked upon as Candides, who, having left this world for the cultivation of their gardens, teach that all is for the best in the best of all possible worlds provided we will let well alone, is also to be traced, I think, to their having neglected to take account of the drag on prosperity which can be exercised by an insufficiency of effective demand. (Keynes, 1936, p. 33)

The possibility of a failure of effective demand is the very point behind the Keynesian-cross diagram, IS-LM curves or the AD-AS relationship. It is taught to undergraduate economists worldwide, and is embedded almost universally in our present policies designed to pull economies out of recession. And while other possible explanations for recession are now usually discussed as well, demand failure remains the single most important concept most economists are taught in relation to the causes of recession and involuntary unemployment. It is the argument that recessions can best be understood as occurring because of a fall in aggregate demand that continues to mark economic theory to this day, along with the implication that stimulating demand through deficit spending is the optimal approach to take in dealing with recessions when and where they occur.

Aggregate demand is intrinsic to the modern understanding of the level of economic activity. The implication is that it is the level of aggregate demand that is responsible for the level of output, the rate of economic growth and the number of persons employed. An insufficient level of aggregate demand is held generally responsible for high levels of unemployment and it is almost universally accepted that deficit financed public spending can permanently raise the level of output and thereby lower the rate of unemployment. There is an aggregate supply curve associated with aggregate demand, but its principal role is the determination of the rate of inflation. Production levels are not determined by supply capabilities but by the willingness of individuals to buy what has been produced with the incomes they have received.

Indeed, the issue went farther than this. Keynes argued that if Say’s Law were valid, continuing and persistent unemployment simply could not occur and this was unrecognized by classical economists whom he was about to correct. As he wrote:

Say’s law, that the aggregate demand price of output as a whole is equal to its aggregate supply price for all volumes of output, is equivalent to the proposition that there is no obstacle to full employment. If, however, this is not the true law relating the aggregate demand and supply functions, there is a vitally important chapter of economic theory which remains to be written and without which all discussions concerning the volume of aggregate employment are futile. (Keynes, 1936, p. 26, emphasis added.)

For the vast majority of the economics profession even now, this is the way in which Say’s Law and its implications are understood. It is the very meaning of the Keynesian Revolution. Mises made the same point in 1950:

Lord Keynes’s main contribution did not lie in the development of new ideas but “in escaping from the old ones,” as he himself declared at the end of the Preface to his “General Theory.” The Keynesians tell us that his immortal achievement consists in the entire refutation of what has come to be known as Say’s Law of Markets. The rejection of this law, they declare, is the gist of all Keynes’s teachings; all other propositions of his doctrine follow with logical necessity from this fundamental insight and must collapse if the futility of his attack on Say’s Law can be demonstrated. (Mises, 1950 [1980])

It is precisely here that we find the division between the economics of the classics and virtually all modern economic theory, especially of the mainstream variety. As was recognized at the time, and as Mises clearly notes, Keynesian economics, that is all of modern macroeconomics with its focus on aggregate demand, must collapse if the attack on Say’s Law turns out to be wrong.

It is only to be regretted that Mises did not recognize how singularly important it was to hammer home this point. He treated Say’s Law as so obviously valid, beyond any possibility of argument, that I suspect he found it impossible to understand how anyone who called themselves an economist could accept Keynesian theory. All he ever directly wrote on Say’s Law he contained in a brief article in a collection of essays. But as for the validity of Say’s Law, he could not have been more clear:

The exuberant epithets which these admirers have bestowed upon his work cannot obscure the fact that Keynes did not refute Say’s Law. He rejected it emotionally, but he did not advance a single tenable argument to invalidate its rationale. (Mises, 1950 [1980])

Mises accepted Say’s Law as unquestionably valid, as part of the “perennial laws” of economics. But because he found rejection of Say’s Law inconceivable he did not do what he might otherwise have done, which was to explain why it must remain an integral part of the bedrock foundation of economic theory if that theory is to provide us with the guidance needed when recessions and high unemployment occur.


What is relevant about Say’s Law cannot be contained within a single statement. Say’s Law, if it is to be understood in full, must be understood as a series of related propositions which when taken together constitute the basic ingredients of the classical theory of the cycle. The most extraordinary of the many ironies that have surrounded this issue since Keynes first pronounced on it in 1936 is that Say’s Law was the foundation stone within classical theory for understanding why a cycle exists at all. Keynes’ argument was that belief in Say’s Law meant that classical economists assumed there was never at any stage an obstacle to full employment. The reality is that Say’s Law was an integral part of the explanation of why in fact unemployment actually occurred.

Keynes, in attacking “Say’s Law” in 1936 was not attacking some one-sentence statement of principle. In attacking Say’s Law, he was attacking the entire classical theory of the cycle. Unless this is understood, it is impossible to understand in full exactly what Keynes was able to do. The propositions associated with Say’s Law need to be seen as the constituent elements of the classical theory of the cycle and to understand why this was so, it is necessary to enter into some of the early history of economic theory itself.

What became the classical theory of the cycle was formed during what is now known as the “General Glut” debate that lasted from the publication of Malthus’ Principles of Political Economy in 1820 through until John Stuart Mill published his own Principles of Political Economy in 1848. Malthus was, in 1820, the single most famous economist in the world. His 1798 publication, On Population, had been an international sensation. As a result, when he published his text on economic theory, it was not just another text but a work that would instantly attract the widest attention.3

What in particular distinguished Malthus’ arguments from virtually all other writings on economic issues at the time was his belief that the recessions experienced by England at the end of the Napoleonic Wars had been caused by oversaving and demand deficiency. And so a debate was commenced across the whole of the economics community of the time, with a raft of books on economic theory published over the next few years in which much of the argument centered on a discussion of what Malthus had written. All agreed it was possible to have an excess supply of individual goods and services. The question was whether there could be an excess supply of all goods and services taken together.

Importantly, it was not a debate over whether recessions and large-scale unemployment were possible. On this there was obviously unanimity. The only question was whether recessions, when they occurred, were the result of too much saving and too little effective demand. That this could never be a realistic explanation was ultimately accepted by the whole of the mainstream of the economics community.

Moreover, during classical times there was no economic principle known as “Say’s Law.” The term would not be coined until the twentieth century or enter economic discourse until the 1920s (see Kates, 1998, pp. 148–149). There was Jean-Baptiste Say’s théorie des débouchés, known in English as the “law of markets,” which stated that demand was constituted by supply. It was the law of markets that was employed as part of the response to Malthus’ views but as only one strand in a far more complex series of counter arguments. It was a crucially important part of the argument, but it was only one of the arguments in a longer chain of reasoning. It was the entire set of counter arguments that when taken together became the related propositions that formed the classical theory of the cycle. Leaving Say’s Law in Keynes’ desiccated form of words—“supply creates its own demand”—not only reverses the point that classical economists had tried to make—that demand in real terms can only be derived through the production of value adding goods and services—but ignores every other related aspect that was central to an understanding of the classical theory of the cycle.

By discrediting the crucially central idea that demand is formed on the supply side of the economy, the related propositions that had emerged from the debate over Malthus lost their coherence. The publication of the General Theory caused the entire classical perspective on the business cycle to disappear. The propositions presented below are therefore intended to reassemble the arguments that were at the core of pre-Keynesian business cycle theory and need to be seen as the full meaning of Say’s Law as it emerged during the General Glut debate.4 They are also put in a form so that the entire argument can be seen as a full and complete response not just to Keynes and the arguments of the General Theory, but also as a reply to modern macroeconomics to the extent that it continues to rely on demand deficiency to explain why recessions occur.


The related propositions that make up Say’s Law are discussed below, along with concrete examples from the pre-Keynesian literature to demonstrate their importance as integral components of classical thought.

Proposition 1: Recessions are never due to demand deficiency. An economy can never produce more than its members would be willing or able to buy. A general glut (i.e. general overproduction) is impossible. Neither high levels of saving nor the redirection of resources into higher levels of capital formation cause recessions to occur.

This is the starting point for any understanding of the pre-Keynesian theory of recession and Say’s Law. Four examples of how this statement was an integral part of economic theory across the entire classical period will help put the law of markets into its proper context.

First Adam Smith. He specifically denies that there is any danger from oversaving and that a community has anything to fear from the saving of its more provident members. It was this argument that Keynes specifically set out to deny.

What is annually saved is as regularly consumed as what is annually spent, and nearly in the same time too; but it is consumed by a different set of people. That portion of his revenue which a rich man annually spends, is in most cases consumed by idle guests, and menial servants, who leave nothing behind them in return for their consumption. That portion which he annually saves, as for the sake of the profit it is immediately employed as a capital, is consumed in the same manner, and nearly in the same time too, but by a different set of people, by laborers, manufacturers, and artificers, who reproduce with a profit the value of their annual consumption. His revenue we shall suppose, is paid him in money. Had he spent the whole, the food, clothing, and lodging, which the whole could have purchased, would have been distributed among the former set of people. By saving a part of it, as that part is for the sake of profit immediately employed as capital either by himself or by some other person, the food, clothing and lodging, which may be purchased with it, are necessarily reserved for the latter. The consumption is the same, but the consumers are different. (Smith, 1776 [1976], p. 359)

A second example is Alfred Marshall writing in a publication co-authored with his wife, Mary Paley Marshall, in 1879. Here it is made abundantly clear that deficient aggregate demand is not the proper explanation for depression.

After every crisis, in every period of commercial depression, it is said that supply is in excess of demand. Of course there may easily be an excessive supply of some particular commodities; so much cloth and furniture and cutlery may have been made that they cannot be sold at a remunerative price. But something more than this is meant. For after a crisis the ware-houses are overstocked with goods in almost every important trade; scarcely any trade can continue undiminished production so as to afford a good rate of profits to capital and a good rate of wages to labour. And it is thought that this state of things is one of general over-production. We shall however find that it really is nothing but a state of commercial disorganisation. (Marshall and Marshall, 1879 [1881], p. 154)

And lest it be thought that this is the early Alfred Marshall which was later subsumed by a different point of view, in a section introduced into the fifth edition of the Principles in 1907 he emphatically made the point again. Note that problems on the demand side are seen only to exacerbate a problem that has been due to other causes.

It is true that in times of depression the disorganization of consumption is a contributory cause to the continuance of the disorganization of credit and of production. But a remedy is not to be got by a study of consumption, as has been alleged by some hasty writers. (Marshall, 1907 [1961], p. 711n)

Finally, Friedrich Hayek. His 1931 article, “The ‘Paradox’ of Saving,” is a full-scale discussion, more than 40 pages in length, on the arguments of Catchings and Foster who during the 1920s and 1930s had argued that over-saving was the cause of recessions. Hayek’s opening paragraph is not only an attack on the belief that excess saving is a cause of recession, but he also specifically refers to the théorie des débouchés as providing the appropriate position. And while Hayek had his own theory of the cycle, the article is not in the least dependent on such views. It is nothing other than a straightforward statement of the classical position. Hayek wrote:

The assertion that saving renders the purchasing power of the consumer insufficient to take up the volume of current production although made more often by members of the lay public than by professional economists, is almost as old as the science of political economy itself. The question of the utility of ‘unproductive’ expenditure was first raised by the Mercantilists, who were thinking chiefly of luxury expenditure. The idea recurs in those writings of Lauderdale and Malthus which gave rise to the celebrated Théorie des Débouchés of James Mill and J.B. Say, and in spite of many attempts to refute it, permeates the main doctrines of socialist economics right up to Tugan-Baranovsky, Thorstein Veblen and J.A. Hobson. But while in this way the idea has found a greater popularity in quasi-scientific and propagandist literature than perhaps any other economic doctrine hitherto, fortunately it has not succeeded as yet in depriving saving of its general respectability. (Hayek, 1931, pp. 74–75, bolding added)

It is highly noteworthy that it was only five years later that the General Theory would in fact do what Hayek had feared, and “deprive saving of its general respectability.”5

Proposition 2: Demand is constituted by supply. Aggregate demand is not independent of aggregate production but is identical with it. A community’s purchasing power is constituted by its value added. Aggregate demand can only increase when the value of the goods and services produced is greater than the value of the inputs used up in the production process.

This proposition is a restatement of Jean-Baptiste Say’s original théorie des debouches, wrongly characterized by Keynes as “supply creates its own demand.” Moreover, the statement that “demand is constituted by supply” may be the most important concept in coming to grips with the classical theory of the cycle, but because it is so foreign to modern macroeconomic thought, it may also be the most difficult. Yet it was fully accepted by pre-Keynesian economists.

Here is James Mill, in the first presentation of what would become the classical theory of the cycle, explaining the significance of this principle. He could not be more emphatic nor does he leave any doubt about just how crucial he believes this principle to be.

No proposition however in political economy seems to be more certain than this which I am going to announce, how paradoxical soever it may at first sight appear; and if it is true, none undoubtedly can be deemed of more importance. The production of commodities creates, and is the one universal cause which creates a market for the commodities produced. (Mill, 1808 [1966], p. 135)

Moving forward a century, the same concept is found in the following passage from one of the most widely used economic texts ever published, in which this principle is stated in very clear terms:

It is only because our exchanges are made through money that we have any difficulty in perceiving that an increase in supply is (not “causes”) an increase in demand…. An increase in the supply of cloth is an increase in the demand for other things; and vice versa, an increase in the supply of anything else may constitute a demand for cloth. What is divided among the members of society is the goods and services produced to satisfy its wants; and the same goods and services are both Supply and Demand. (Clay, 1916, p. 242)

The notion of aggregate demand separate from aggregate supply was foreign to pre-Keynesian economic thought. Aggregate demand grows at the same rate and by the same amount as aggregate supply, and will not grow unless supply has grown. It is not, however, just any production that will lead to an increase in aggregate demand.  What creates demand is the production of forms of output for which enough buyers can be found to cover in aggregate the entire costs of production. Only if the goods and services produced can be sold for more than was paid for the inputs that went into their production can it be said with certainty that value has been added during the production process. Conversely, if the goods and services produced do not create more value than is used up in the production process, there can be no increase in aggregate demand because there has been no increase in aggregate supply in any relevant sense.

Proposition 3: The process involved in purchase and sale is the conversion of one’s own goods or services into money and then the re-conversion of the money one has received back into other goods and services. There is no implication of a barter economy. Money is intrinsic to the processes involved.

At the very core of the classical propositions surrounding Say’s Law is an appreciation that money is infused with value only by being received in exchange for value adding production. The process is one that may be characterized in the formula C–M–C’ where the set of goods or services in one’s own possession (C) is converted into a different set of goods or services (C’) by the sale of what one owns for money (M) and then the reconversion of the money received into what one wishes to buy.6 Keynes had accused classical economists of confusing a barter economy with the operation of a money economy, but from the first statements on Say’s Law by Say himself that had never been the case. Here is J.B. Say, in the fourth edition of his Treatise,7 trying to explain the obvious.

Should a tradesman say, “I do not want other products for my woollens, I want money,” there could be little difficulty in convincing him that his customers could not pay him in money, without having first procured it by the sale of some other commodities of their own. … You say, you only want money; I say, you want other commodities, and not money…. To say that sales are dull, owing to the scarcity of money, is to mistake the means for the cause; an error that proceeds from the circumstance, that almost all produce is in the first instance exchanged for money, before it is ultimately converted into other produce. (Say, 1821, pp. 163–165) 

But more importantly, the process lay in ensuring that those who produced made sure that they created value in the process. Demand was only constituted by the value added that arose from the sale of goods or services to others. If output could not be sold at prices that repaid the costs of production, then no value added had occurred. That this frequently did take place provided the core insight into the classical theory of the cycle. That demand was built on productive activities was also pointed out by Mises, who was explicitly following Say in making this point:

Commodities, says Say, are ultimately paid for not by money, but by other commodities. Money is merely the commonly used medium of exchange; it plays only an intermediary role. What the seller wants ultimately to receive in exchange for the commodities sold is other commodities. (Mises, 1950 [1980])

To understand demand being constituted by supply, it is necessary to recognize that in a properly functioning economy, purchases are effected with the revenue from the previous sale of goods and services or with money borrowed from others who have earned incomes by producing. For those who earned their incomes from the sale of goods and services, the process is direct. The creation of value and the sale of what had been produced provided the income for the purchase of other goods and services. For businesses investing borrowed funds, the purchases are effected through the transfer of funds through a saving-investment process. For governments, purchases are effected through revenues raised through taxation of the incomes of those who had sold goods or services to the market.

Proposition 4: Recessions are common and result in high levels of involuntary unemployment.

It really ought to be unnecessary to point out that this proposition ought to be completely non-controversial. It really ought to have been inconceivable to have suggested, as Keynes did in 1936, that economists until then had had no explicit theory of involuntary unemployment and recession. Yet one of the consequences of the publication of the General Theory was the belief that classical economists had no theories to account for recessions and involuntary unemployment. It is therefore necessary to make the explicit statement that classical economists did indeed have such theories of recession and they most assuredly did understand that involuntary unemployment was a frequent feature of economic life. The theory of the business cycle had been developing for over a century by that stage, so that for Keynes to have stated of his fellow economists that they had no theory of involuntary unemployment was absurd.

A compendium of all of the theories of the cycle is found in a League of Nations publication by Gottfried Haberler, titled Prosperity and Depression whose first edition was published in 1937, the year following the publication of the General Theory. The first words of the preface ought to make it absolutely plain that recession and unemployment were amongst the most important questions under examination by the economics community of the world during the 1930s, and had been for generations:

This book has its origin in a resolution adopted by the Assembly of the League of Nations in September 1930 by which it was decided that an attempt should be made to co-ordinate the analytical work then being done on the problem of the recurrence of periods of economic depression. The literature concerning economic depressions and what is currently and somewhat loosely described as the trade cycle is abundant…. It is apparent from the persistence with which depressions occur, from the gravity of their economic and social effects, and from the growing consciousness of that gravity, that – however abundant the literature on the subject, however elaborate and specious the theories – our knowledge of the causes of depressions has not yet reached a stage at which measures can be designed to avert them. (Haberler, 1937, p. iii)

That what ought to have been seen as absurdly improbable was nevertheless accepted from the moment it was first published is an issue that demands the attention of historians of ideas. Here we merely note that Keynes’ statement, that economists before him had no theories to explain recessions and unemployment, is false as a moment’s reflection ought to have led anyone to recognize at the time, just as it ought to be recognized today.

Proposition 5: Recessions are due to structural problems of one kind or another. In particular, recessions occur where the structure of supply does not match the structure of demand. Recessions occur when the pattern of demand is different from the actual composition of output so that a significant proportion of the goods and services put up for sale remains unsold.

For anyone basing their understanding of these issues on Keynes’ writings, it is something of a surprise to discover that the law of markets was at the very centre of the classical theory of the recession and, in fact, provided the foundation for the theory of the cycle as understood by classical economists. Because demand was constituted by supply, cyclical activity was understood to be the result of individuals and businesses producing what could not be sold at prices which covered costs. Why this might happen was the underlying issue, but that it frequently did happen, of this no one had the slightest doubt. The more than one hundred year classical literature on the nature and causes of the business cycle is a testament to the recognition that pre-Keynesian economists gave to unemployment and recession.

Torrens, writing in 1821 in a direct response to the arguments presented by Malthus, makes the point as explicitly as it is possible to make it. The classical theory of the cycle was built on these very concepts. Demand is constituted by supply but only so long as supply consists of what those with incomes to spend want to buy. Keeping demand and supply properly proportioned was the imperative, but once that had been achieved all went well. It was when the proportions were not maintained that recessions would occur. Torrens firstly notes that there is no possibility that supply will ever outrun demand.

So long as the proportion is preserved, every article which the industrious classes have the will and power to produce, will find a ready and profitable vend. No conceivable increase of production can lead to an overstocking of the market…. Increased production will create a proportionally increased demand [sound familiar?] …. (Torrens, 1821 [1965], pp. 370–372)

What is particularly notable is that Torrens uses almost the very words Keynes would use to summarize Say’s Law. “Increased production will create a proportionately increased demand” is the lineal ancestor of “supply creates its own demand.” Torrens is invoking Say’s law of markets to show that demand deficiency is never a problem. But he does not conclude from this that economies cannot therefore go into recession or that there are no obstacles to full employment. He instead uses this very principle to explain why recessions occur. Following on from the above passage, Torrens immediately sets out the consequences should something happen to disturb the balance between the structure of production and the structure of demand.

This happy and prosperous state of things is immediately interrupted when the proportions in which commodities are produced are such as to disturb the equality between effectual demand and supply…. Then gluts and regorgements are experienced. (Torrens, 1821 [1965], pp. 370–372)

Torrens was not the first to make this point, but he made it very well. A lack of proportion between supply and demand is the cause of a descent into recession. The problems of recession are due to structural problems in an economy, not because of a failure of demand. And it required an understanding of the law of markets to understand that recessions occur when what has been produced does not coincide with what those with incomes want to buy.

In these passages, Torrens captured the theory that became during the following century the common ground amongst the economics community in discussing the business cycle. Recessions and depressions were due to structural problems. Haberler, in his Prosperity and Depression, provided a synopsis of the theory of the cycle as it had been understood until then. In summarizing the views of the economic profession of his time, he wrote:

An expansion or contraction may be interrupted on the one hand by an accident…or it may on the other hand itself give rise to maladjustments in the economic system…. Most cycle theorists have tried to prove that the second type of restraining force is all-important. (Haberler, 1937, p. 245)

This is Torrens once again. It is this maladjustment in the structure of production, where demand and supply are out of proportion with each other, that was the fundamental explanation for recession. Demand deficiency played no part in the process within orthodox theory.8

Where demand was crucial was in relation to the structure of demand relative to supply, that is, in situations where what buyers would have been willing to pay the full costs of production for did not match what suppliers had actually put on the market. Starting from the proposition that demand is constituted by properly proportioned supply, recessions are caused by events that mislead producers into producing goods and services that cannot be sold at cost covering prices.

Proposition 6: Partial overproduction of individual goods and services occurs continuously within economies and can lead to a general downturn in an economy. The transmission mechanism is from a reduction in earnings in some sectors of the economy where sales have been below expectations to a fall in demand in other sectors and therefore to a wholesale downturn in activity.

Walter Bagehot, as editor of The Economist, wrote one of the most influential nineteenth century works on the operation of the money market. As part of this work, he included a chapter on the nature of the business cycle, in which he described the evolution of a general downturn built out of a downturn in one part of the economy. Given Keynes’ accusation that classical economists had ignored monetary factors and their effects on economic activity, it should not go unnoticed that the following is from Bagehot’s Lombard Street which had as its subtitle, A Description of the Money Market. What Bagehot wrote was this:

No single large industry can be depressed without injury to other industries; still less can any great group of industries. Each industry when prosperous buys and consumes the produce probably of most (certainly of very many) other industries, and if industry A fail and is in difficulty, industries B, and C, and D, which used to sell to it, will not be able to sell that which they had produced in reliance on A’s demand, and in future they will stand idle till industry A recovers, because in default of A there will be no one to buy the commodities which they create. (Bagehot, 1873, 121–122)

The essence of this process is the creation of an economic downturn built upon the systematic failure of producers to sell what they have produced in their own markets. This is not a description of a Keynesian multiplier but a trail of purchase and sale between different producers. It accepts that when the recovery comes there may be different firms and industries in different proportions. But the conception that lies behind it is that the pieces in the economy must interlock as firms provide a market for each other with the entire structure ultimately aimed at producing goods and services for final home consumption.

Proposition 7: Monetary factors, most notably structural imbalances in the market for credit, can also be and often are an important cause of recession. Even where monetary instability has not been the originating cause of recession, monetary factors will often deepen a recession brought on for other reasons.

It is because Keynes argued that classical economists thought only in terms of real variables that such an obvious statement even needs to be made. It was, in fact, the specific conclusion reached by Becker and Baumol that ought to have put this issue to rest for all time, and also have raised some questions about the foundations of the Keynesian economic theory that had been built on the rejection of so flimsy a straw man. Becker and Baumol could not have been more explicit in dealing with this caricature of classical theory, which they labeled “Say’s Identity.” In discussing what they term “the clearest statement on the point”—Mill’s second essay in his Essays on Some Unsettled Questions of Political Economy—they wrote:

It is all there and explicitly—Walras’ Law, Say’s Identity which Mill points out holds only for a barter economy, the “utility of money” which consists in permitting purchases to be made when convenient, the possibility of (temporary) oversupply of commodities when money is in excess demand, and Say’s Equality which makes this only a temporary possibility. Indeed, in reading it one is led to wonder why so much of the subsequent literature (this paper included) had to be written at all. (Becker and Baumol, 1952, p. 374.)

Monetary factors can and do cause recession. It is stating nothing but what ought to be obvious, that classical economists were fully aware that monetary factors were often part of the process even when not the initiating factor in causing recessions to occur.

The approach to economic policy becomes very different if one begins from a classical perspective rather than from one built that commences with demand deficiency. These different perspectives are part of the matrix of ideas that were part of the structure of understanding that existed under a theory of the cycle built on classical foundations.

Proposition 8: Because recessions are not due to a failure of demand, practical solutions to recession do not encompass increased levels of public spending. While such expenditure may provide some limited benefit if spending is concentrated on value adding goods and services, such expenditure is merely a palliative rather than a cure.

The policy consequences of Keynesian theory have over the years provided ample evidence that on this matter classical economists were correct. There has been no instance of a peacetime increase in public spending during recession that has led to recovery. Reductions in taxation have a different effect on economic outcomes, and can be consistent with classical principles in generating economic growth. Increases in public spending, however, are not. John Stuart Mill’s statement, found at the start of this paper, is about as clear-cut as one could find.

The utility of a large government expenditure, for the purpose of encouraging industry, is no longer maintained.... It is no longer supposed that you benefit the producer by taking his money, provided you give it to him again in exchange for his goods. (Mill, 1874 [1974])

The stimulus packages that have been associated with attempts to revive economies internationally following the onset of the Global Financial Crisis, especially in the US and UK, have been failures. The absence of signs of success and the growing problems related to the rising levels of public debt are indications that these Keynesian policies did not work as their advocates suggested they would. The outcomes of these stimulus packages ought to be recognized as the major test of Keynesian theory and policy that they have been. Based on this experience, the macroeconomics that is almost universally taught should be recognized as of no theoretical or practical value. Ridding economic theory of the aggregate demand curve should be the single most important theoretical issue of our time.9

If public spending and deficit finance are recognized to have failed, just as they failed in Japan during the 1990s and in the United States during the Great Depression, support for Keynesian theory and policy should erode and a search for an alternative theoretical approach should commence. The proper place to begin such as investigation is amongst the long-forgotten theories of the cycle, which were discarded after the publication of the General Theory. There should be a newfound recognition that perhaps, after all is said and done, that so far as Say’s Law is concerned, Keynes was wrong and the classical economists were right.


It was Keynes himself who made it clear that the economics of the General Theory was to be seen as a refutation of Say’s Law. Recessions, he wrote, were caused by a deficiency of aggregate demand.

This was utterly contrary to mainstream classical thought. Classical economists understood that economies are not driven by demand but by value adding production, which is what they referred to as “supply.” They understood perfectly well that raising demand without an increase in the level of value adding output cannot be an answer to recession and unemployment.

This was summarized by classical economists in various ways: demand is constituted by supply, there is no such thing as a general glut, overproduction is an impossibility. However, the most remarkable short statement, not just on the nature of aggregate demand but also on the related issue of how recessions occur, can be found in Ricardo’s reply to Malthus in a personal letter written on October 9, 1820. Ricardo was writing a few months after Malthus’ Principles had been published: “Men err in their productions, there is no deficiency of demand.” (Ricardo, 1951–73, p. 277)

This is, first of all, a statement on the causes of recession: “men err in their productions,” that is, there is some kind of market disequilibrium which has occurred across the economy. And beyond that, it is a statement of what does not cause recessions: “there is no deficiency of demand.” Whatever might have caused the recession, it is not due to a lack of demand. What is found in Ricardo’s short statement is in summary form the entire classical theory of recession with its explicit rejection of demand factors as their cause. To understand what Say’s Law really means and why it matters, this is what you need to know.

What if Ricardo’s short and to-the-point statement were at the core of modern macroeconomics in the way it was at the core of the classical theory of the cycle? Here there is no ambiguity of meaning, none of the uncertainty that currently exists over what “supply creates its own demand” does or does not mean. Ricardo’s brief statement of classical principle means that when recessions occur, they cannot be understood as a consequence of too little demand but should be understood as some sort of derangement within the

market process. Were this understanding at the core of modern macroeconomics, policy makers would have no excuse for the levels of deficit spending that occurred after the commencement of the Global Financial Crisis but would understand that far different measures are needed to get markets and an economy back on track.

My book, Say’s Law and the Keynesian Revolution, which covers in far more detail all that has been discussed in this paper, has as its subtitle, How Macroeconomic Theory Lost Its Way. Macroeconomics replaced the classical theory of the cycle in the 1930s and has been Keynesian ever since. No metaphorical statement on the death of Keynes or of Keynesian economics can be true so long as aggregate demand maintains its presence at the core of macroeconomic theory and policy. Macroeconomics, with its focus on aggregate demand, has been systematically misleading economists since the 1930s. Because of the near universal acceptance of Keynesian theory within the mainstream, economists have repeatedly formulated policies around the need to stimulate demand during periods of high unemployment. Keynesian economics has, however, not had a single peacetime success but has recorded many, many failures to which one more can now be added. It is the very concept of aggregate demand that must be removed from economic theory. Its pervasive presence has caused a blackout curtain to fall across the whole of macroeconomic theory making it all but impossible to understand the underlying workings of an economy or to provide useful advice when recessions occur.

The use of public spending and deficit finance to deal with the Global Financial Crisis has been massive and worldwide. This ought to be recognized as having been a decisive test of the validity of Keynesian theory and policy. These policies have been tried to their utmost limits in the United States and elsewhere and should be recognized as having been an abject failure. A return to an economic theory based around a proper understanding of Say’s Law and the classical theory of the cycle should be the direction in which economic theory now moves.

  • 1. Keynes was at the time completing the essay for inclusion in his Essays in Biography which would be published the following year.
  • 2. Keynes discussed a number of these in the General Theory, referring to them as his “brave army of heretics” (Keynes, 1936, p. 371), a band of brothers that included Bernard Mandeville, Malthus, Major Douglas, Silvio Gesell and J.A. Hobson.
  • 3. As an interesting parallel, Keynes, too, was the most famous economist of his time after having written his Economic Consequences of the Peace at the end of World War I. It had also been a worldwide sensation in its time.
  • 4. For historical accuracy I will note that these arguments had been first brought together by James Mill in 1808 where he, too, specifically invoked the théorie des debouches to explain why demand deficiency is a fallacious explanation for recession. It is for this reason that in my view James Mill had been the first to properly state “Say’s Law” (see Kates, 1998, pp. 24–29).
  • 5. The existence of this critique of Catchings and Foster may also help explain why Hayek, having invested the time and effort in dealing with their arguments, almost completely ignored Keynes’ attempt to achieve the same result. By the time Keynes wrote, Hayek may well have found trying to explain the fallacies in such reasoning completely stale. To have bothered responding to Keynes in the detail required would have for him involved going over old ground.
  • 6. This is the formula used by Marx to explain the classical mechanisms associated
    with the law of markets but was used by him as a criticism.
  • 7. It is the fourth edition that has been the one translated into English because that was the latest edition available when Malthus published his Principles in 1820. There would be a fifth edition in French that has not been translated.
  • 8. There was, however, an under-consumptionist literature which argued that too little demand from consumers was the systematic cause of economic recession. This was at the time almost entirely the province of economic cranks, as Keynes’ reference to the “brave army of heretics” plainly shows. Hobson was seen as the leading exponent of this view as both Keynes (Keynes, 1936, pp. 364–370) and Haberler (1937, p. 115) make clear.
  • 9. The downturn in activity, because the cycle is cyclical, will end at some stage with an upturn. What is evident already, however, is that the spending programs which have been introduced have not been factors in generating recovery. Indeed, not only have they been of virtually no use in creating a net addition to employment they have also coincided with deteriorations in economic conditions generally that have been unexpected by those who introduced the stimulus programs. The argument has been made that economic conditions would have been even worse than they were had these programs not been introduced even though at the time of their introduction, the expectation was that there would be a generally rapid upturn in activity and the labor market. None of this has occurred, as anyone looking at these programs from a classical perspective would have expected.


Bagehot, Walter. 1873. Lombard Street: A Description of the Money Market. London: John Murray, 1919.

Becker, Gary and William J. Baumol. 1952. “The Classical Economic Theory: The Outcome of the Discussion.” Economica 19: 355–376.

Clay, Henry. 1916. Economics—An Introduction for the General Reader. London: MacMillan and Co.

Haberler, Gottfried. 1937. Prosperity and Depression: A Theoretical Analysis of Cyclical Movements. Geneva: League of Nations.

Hayek, Friedrich A. 1931. “The ‘Paradox’ of Saving.” Economica 11 (Old Series): 125–169. In The Collected Works of F.A Hayek, vol. 9: 74–120.

Kates, Steven. 1998. Say’s Law and the Keynesian Revolution: How Macroeconomic Theory Lost its Way. Edward Elgar: Cheltenham.

——. 1999. “The Ten Most Influential Economists of the Twentieth Century.” Canberra Times, 21 December.

Keynes, John Maynard. 1933. The Collected Writings of John Maynard Keynes. Vol. X: Essays in Biography. Donald Moggridge, ed. London: The Macmillan Press Ltd.

——. 1936. The Collected Writings of John Maynard Keynes. Vol. VII: The General Theory of Employment, Interest and Money. Donald Moggridge, ed. London: The Macmillan Press Ltd.

Marshall, Alfred. 1961. Principles of Economics. 2 vols. 9th (variorum) ed. London: Macmillan and Company Ltd. for the Royal Economic Society.

Marshall, Alfred and Marshall, Mary Paley. 1879. The Economics of Industry. 2nd ed. London: Macmillan and Co., 1881.

Mill, James. 1808. Commerce Defended. 2nd ed. In Winch, Donald, ed. James Mill: Selected Economic Writings. Edinburgh: Oliver and Boyd Ltd, 1966, pp. 85–159.

Mill, John Stuart. 1874. “Of the Influence of Consumption on Production.” In Essays on Some Unsettled Questions of Political Economy. 2nd ed. Clifton, New Jersey: Augustus M. Kelley, 1974, pp. 47–74.

Mises, Ludwig von. 1950. “Lord Keynes and Say’s Law.” In Planning for Freedom: And Sixteen Other Essays and Addresses. 4th ed. South Holland, Ill.: Libertarian Press, 1980. Quotations are taken from the website version of this article: http://mises.org/daily/1803.

Ricardo, David. 1951–1973. The Works and Correspondence of David Ricardo. Vol. VIII: Letters, 1819–June 1821. Piero Sraffa, ed. Cambridge: Cambridge University Press.

Say, Jean-Baptiste. 1821. A Treatise on Political Economy; or the Production, Distribution, and Consumption of Wealth. C.R. Prinsep, trans. London: Longman, Hurst, Rees, Orme, and Brown.

Schneider, Michael. 2010. “Keynesian Income Determination Diagrams” in Famous Figures and Diagrams in Economics. Mark Blaug and Peter Lloyd, eds. Cheltenham: Edward Elgar.

Smith, Adam. 1776. An Inquiry into the Nature and Causes of the Wealth of Nations. Edwin Cannan, ed. Chicago: University of Chicago Press, 1976.

Torrens, Robert. 1821. An Essay on the Production of Wealth. New York:

Cite This Article

Kates, Steven. "Why Your Grandfather's Economics Was Better than Yours: On the Catastrophic Disappearance of Say's Law." The Quarterly Journal of Austrian Economics 13, No. 4 (Winter 2010): 3–28.