Quarterly Journal of Austrian Economics

Home | Mises Library | Why Keynesian Concepts Cannot Be Used to Explain Pre-Keynesian Economic Thought: A Reader’s Guide to Classical Economic Theory

Why Keynesian Concepts Cannot Be Used to Explain Pre-Keynesian Economic Thought: A Reader’s Guide to Classical Economic Theory

The Quarterly Journal of Austrian Economics
11/10/2014Steven Kates


ABSTRACT: This note is built around five issues that center on the damage to economic theory caused by Keynes and his General Theory. They have been brought to mind by Machovec in his paper, “Our Classical Macro Heritage” (2014) but there are other similar papers in which classical economic thought is explained using Keynesian concepts with the actual intent of classical economists distorted if not actually lost. This article is thus a reader’s guide to classical economic theory, and in this I am referring to mainstream economics from the first publication of Adam Smith’s Wealth of Nations in 1776 through until the 1871 seventh edition of John Stuart Mill’s Principles, the last published during his lifetime. And while there are a larger number of issues that could be covered, only five will be dealt with here. These are: the lack of an understanding of Say’s Law and why it is important; the need to distinguish between the real economy and money at every stage of an economic analysis; the classical meaning of saving and investment; the inappropriateness of using Keynesian concepts to interpret pre-Keynesian economic thought; and the importance of understanding the history of economic thought if one is to understand modern economic thought. There is finally a discussion of the need to return to classical pre-Keynesian theories of the cycle, of which Austrian economics is the sole manifestation that had been able to remain intact in the wake of the Keynesian Revolution.

KEYWORDS: classical economics, Keynesian economics, macroeconomics, Austrian economics, Say’s Law, saving and investment, market rate of interest, natural rate of interest, history of economic thought, John Stuart Mill
JEL CLASSIFICATION: B12, B13, B22, B25, B30, B50, B53, E12, E14, E21, E22, E32, E62, H12


The General Theory was published a very long time ago and while there are a number of active heterodox traditions, there is no economist who has not been taught macroeconomic theory based on Keynesian economic thought. And what makes a model Keynesian is that it is built around variations in aggregate demand as the major explanation for variations in output and employment. Indeed, the very term aggregate demand does not enter economic discourse until 1936. It is a term with a vastly different meaning from the ancient “effective demand,” which goes back at least as far as the eighteenth century. Effective demand is concerned with what must occur if someone is to turn their wishes into an ability to buy products. Aggregate demand is instead the summation of all of the individual demands that occur across the economy. The use of aggregate demand in any of its various forms is not only not a classical concept but is fundamentally contrary to everything that classical economists were trying to explain. Any economic explanation that depends on variations in aggregate demand should be recognized as Keynesian and not in any sense classical. With this cautionary note, let me turn to the five issues where Keynesian theory can be seen as seeping into discussions that are supposedly built on different economic traditions.


Let it be stipulated that, just as Keynes wrote, Say’s Law was part of the very foundation of classical economic theory, universally accepted with no important dissent amongst the mainstream. Moreover, this was not a theory that had seen its day in the nineteenth century but remained universally accepted right through to the publication of the General Theory in 1936. Nor were the associated issues trivial but a critical part of the machinery in the engine room of ideas that economists kept in active repair. All agreed it was valid and important. Therefore, if one is to discuss Say’s Law, one must have a definition of its meaning that would sound like a substantive principle and be capable of explaining why it might have been defended with such tenacity over such a long period of time by some of the greatest economists who have ever lived.

To understand Say’s Law, one needs to understand it within the context of its first statement. In 1808, James Mill wrote a refutation of an argument that had been put up by one William Spence. Spence (1807) had argued that an economy is driven by demand.Therefore, in his view, the loss of trade caused by the Napoleonic Wars could be overcome by encouraging additional spending, in this case by the landed aristocracy. Mill ([1808] 1966) had replied that an economy was driven by supply. Demand deficiency, he argued, is never the cause of recession and increasing demand without first increasing supply could never raise employment levels. In making his point, Mill took up an argument found in the first edition of Jean-Baptiste Say’s Treatise, in which Say had argued that demand is constituted by supply. If demand is merely a derivative of supply, then it has no independent existence apart from supply itself. The core point was to deny demand deficiency as a cause of recession. It is not necessary here to agree or disagree with Mill; it is merely necessary to understand what the actual issue is.

This earlier discussion had little impact on the development of theory. It was not until the “general glut” debate, touched off in 1820 by the publication of Thomas Malthus’s Principles of Political Economy, that this issue moved onto the center stage of economic controversy. Malthus, like Spence, argued that demand deficiency, caused by oversaving, was the cause of recession requiring an increase in demand to increase employment. The general glut debate brought with it an enormous outpouring of comment with the three leading protagonists against Malthus being James Mill, David Ricardo and J.-B. Say. Say’s most notable work in this regard is his Letters to Mr Malthus ([1821] 1967). Again, it does not matter if one agrees with Say or not. What matters is that one understands what the issue was. The best summary statement of the classical position was offered by Ricardo in a personal letter to Malthus in October 1820 in which he wrote, “men err in their productions, there is no deficiency of demand” (see Kates, 1998, p. 49). That is, when recessions occur, it is because there have been mistakes made in production, not because there is not enough demand.

This is obviously a serious issue over which people may disagree. The fact of the matter is that economic theory from 1808 through until 1936 explicitly accepted this principle, which since the 1920s (and only since the 1920s) has been referred to as Say’s Law. Since 1936, it has been rejected almost universally within economic theory. The great theoretical innovation brought into economics by Keynes was “aggregate demand” which had had absolutely no presence within economic theory up until then (Kates, 1998, 2010). The term that had a distinct presence going back into the eighteen century was effective demand, that is, a wish that could be effected by one’s own income. One’s own income was, of course, determined by what one could produce and sell.

Central to the process was an understanding that all economic activity began on the supply side. This was the conception underlying Say’s Law:

produce something → sell it for money → buy something else

In an exchange economy, one must produce goods or services, sell what one has produced for money, and then use that money to buy other things.1 Buying thus depended on selling. Knowing what to produce was (and is) the major conundrum for an entrepreneur. The absolute necessity is to find goods or services to produce that can be sold at cost-covering prices. Those who contribute to production then share in the proceeds (profits, rent, wages, interest) but it is these receipts that become the incomes used to buy what had been produced by others.

The classical theory of recession was focused on what might stop this directional flow. As Ricardo noted, “men err in their productions,” thus focusing on the second stage of the three. This was a rudimentary version of the theory that would dominate the economic explanation for recessions for the next century. And what the theory as it finally developed did was focus on the first two points, why what had been produced could not be sold at prices that covered costs, or why the financial system had broken down causing major dislocations within the productive system. Not buying (that is, too much saving) as an explanation for recession was ruled out.

Whatever else one might think about this interpretation, it has the benefit of fitting in with the facts of the general glut debate and it provides a more than plausible explanation of what takes place in recession. And to appreciate just how enduring this form of explanation was, Haberler, in his Prosperity and Depression (1937), provides exactly the same theory as had Ricardo in 1820. It was far more elaborate and more detailed, but for all that, it was the same basic framework. One could easily apply this framework to the Global Financial Crisis in a way that trying to explain the very sudden descent into recession being due to an equally sudden fall off in demand could not. A fall in demand was obviously the consequence of something else; it was not the initiating cause.


The Keynesian Revolution has been responsible for a great deal of harm in economic thinking, but possibly most insidious—even more damaging than the introduction of aggregate demand—has been the fudging of the distinction between the real side of the economy and the money side. Keynes was very explicit about this. He complained that classical economists had discussed the nature of recession in real terms and then brought in money only at the very end.

The conviction, which runs, for example, through almost all Professor Pigou’s work, that money makes no real difference except frictionally and that the theory of production and employment can be worked out (like Mill’s) as being based on “real” exchanges with money introduced perfunctorily in a later chapter, is the modern version of the classical tradition. (Keynes [1936] 1973, p. 19)

Keynes therefore brought in money from the very start, a major error and source of confusion to this day. Classical economic theory, just as Keynes stated, did indeed build up from a real foundation before introducing money into the story. It was necessary to understand the nature of what was taking place in terms of the value adding activities of those who contributed to the production process. The factors of production were treated as an important part of the story, with significant discussion about the role of land, labor and capital. And while land and labor are almost automatically thought of in real terms, capital has that dual meaning of finance as well as humanly-produced inputs, helping to cloud the issue.

By sliding into a focus on monetary aggregates, the real side of the economy was firstly pushed into the background and is now almost entirely forgotten. The notion of value adding as an assembly of a specific set of inputs—inputs which already have value—into products which have even greater value, has virtually disappeared. Value adding is discussed, if it is discussed at all, in relation to the national accounts. Recognition that economic growth means nothing else other than that resources have been reconfigured into a set of outputs greater in value than the resources that had been used up—resources that include capital and labor—is nowhere to be seen. The notion of a surplus is no longer discussed. The conception, especially in macroeconomics, is now almost entirely related in monetary terms. The pervasive “Y” of GDP—whether conceived as income, output or production—is calculated in money terms, even where it is a “real” variable. Moreover, it is because of the use of economic modeling that the once hard and fast distinction between the real side and the monetary side of the economy has become more than blurred. And nowhere may this be more important than in trying to make sense of the nature of saving and investment and how they impact on the operation of an economy.


Saving and investment are now almost always thought of in money terms. It is the height of abstract thinking to make the switch to thinking of saving only in real terms (that is, as an inventory of productive inputs) but it is the way every major classical economist thought. I cannot, of course, guarantee that there was never such a confusion within the classical school, but for those whose works are still read, both saving and investment were conceptualized in real terms.

The confusion starts because when individuals think about their own saving, they inevitably think about it in terms of money. Everyone thinks of saving as their putting money into a financial instrument of some kind—a bank account, a bond, the share market—and consequently conceives of saving only in this way. It is this that John Stuart Mill was discussing when he noted that for most people, they follow savings into the strong box and therefore see savers as doing no good while those who spend their way through their inheritance are seen to be providing a positive stimulus to the economy.

Yet it is not possible to think of national saving in such terms. A nation does not save by placing money into financial institutions. A nation saves by using the resources available to it in productive value-adding ways. Saving similarly consists of resources that are not eaten up with no future benefit but are channeled into uses that will add to the future flow of output. To a classical economist, saving and investment are equal because they are the same thing. The residue of this concept is still found in the Keynesian I=S.

Yet the change wrought by Keynes was to shift this equality from the real side to the monetary side and thus leave open the possibility that S and I would not be equal because individuals had chosen to put money aside but no one had chosen to borrow the money for investment purposes. Saving is now seen as a flow variable when in classical times saving constituted that massive stock of potentially productive assets that exist in every economy. The flow of new capital in any year is a minuscule proportion of the capital that already exists. Saving consists of using some proportion of the existing stock of land, labor and capital to produce towards a future return rather than using those resources for current consumption.

The classical literature on saving and investment cannot be read without thinking in real terms. Possibly more than anywhere else in thinking about economic matters, the potential mischief caused by failing to make this distinction is more destructive of clear thinking. The need to make this distinction was recognised throughout classical times and made explicit by Wicksell with his two categories of the money rate of interest and the natural rate of interest. The money rate of interest referred to the supply and demand for money and credit, which would find equilibrium at some level. The natural rate of interest referred to the real resources available for investment purposes. This is a distinction no longer maintained. The result is a muddled concept of investment so that, amongst other things, artificially low money rates of interest are now seen as a positive stimulus to economic activity. No classical economist was likely to make such a mistake.


It should be clearly understood, since Keynes made it the central argument of The General Theory, that the aim of his book was to rid economic theory of Say’s Law. In doing so, he was bringing into the economic theory of the cycle the long-rejected notion of demand deficiency. It is thus crucial if one is to explain the classical theory of the cycle that notions that are found in a standard Keynesian macro model are not used to interpret the classical way of thinking.

The problem of this approach goes back almost to the beginning of Keynesian macro with John Hicks’ “Mr Keynes and the Classics” (1937). It was this article that introduced IS-LM to economic theory and in which Hicks used the same model of aggregate demand to explain both the classical and Keynesian approach. Yet with the model entirely demand-side, it ought to be obvious that if one were actually to understand what pre-Keynesian economists were attempting to argue, and to do so within their own terms, that it would be impossible to use a model built on variations in demand. With IS-LM the supply side of the economy disappeared. There is saving and the demand for investment goods buried within one curve and the money market in the other. But for both, the effect of shifts in either curve are translated into changes in rates of interest on one axis and changes in aggregate output on the other. At no point is there any representation of the supply-side of the economy where decisions to produce are considered. It is demand that will automatically elicit a supply and determines the level of activity.

Going further, the classical theory of the cycle was structural in nature. The economy was shaped by the various entrepreneurial attempts to determine what can be sold, with changes in the demand for particular goods and services reflected in the supplies required of other goods and services needed as inputs. Recessions were not due to a fall in demand. They were due, to put it at its simplest, to entrepreneurial error. For some reason, the array of goods and services produced—and this was not just final goods and services but included every form of input into the production process—were to a much larger degree than was normally the case, not the goods and services that others wished to buy. The economy was therefore out of phase, with some part of the economic structure unable to earn a positive return. The result was a downturn in activity and a higher rate of unemployment that would persist until these structural distortions had disappeared.

The aggregated Keynesian apparatus is virtually useless in trying to make sense of what is actually taking place in terms of the underlying structure. With the economy having turned down, entrepreneurs experience a falloff in demand but that is the effect and not the cause. Something else had led to the downturn. Describing the lack of a proper structural alignment of purchase and sale as a fall in aggregate demand hides everything of any importance if one is to understand what is taking place. So to use statements about a mismatch between aggregate intended investment ex ante and aggregate actual investment ex post as a proxy for the classical concepts employed by John Stuart Mill is totally inappropriate. John Stuart Mill, as well as the entire classical school, never thought in this way and would have rejected such classifications (see Kates, 2014, ch. 14 for a full discussion). What, therefore, is one to make of a passage such as this:

The notion of an ex ante macro demand deficiency, therefore, was characterized in Mill’s Principles not only as a “chimerical supposition,” but worse, as a “fatal misconception” (in Kates, 1998, pp. 67, 68). Mill had undoubtedly [!] based his conclusion on the following logic: Since people do not earn money simply to burn it, they were resolved, during the design and initial execution stages of production to spend/invest all their income from goods sales and/or labor. In addition, banks had originally planned to lend all their deposits (beyond their internally-established levels of cash reserves). Consequently, aggregate demand and aggregate supply, ex ante, were necessarily equal. But the best laid spending plans of mice and men will be revised downwards in response to new, worrisome information, causing end-state aggregate demand to shrink due to unplanned hoarding (either by individuals or banks)….

In such cases the anti-contraction equality between AD and AS that existed ex ante will falter ex post, spurring production cutbacks (reductions in aggregate supply) until AS is again equal to AD, but to an AD lessened by the appearance of hoarding, yielding, thereby, a new, inferior macroequilibrium that is characterized by lower national income and higher unemployment. This is consistent with The General Theory… (Machovec, 2014).

That is some sleight-of-hand to equate Keynes’s argument with Mill’s! And given that I am quoted in the middle of this passage, I am at a loss to know what else I could possibly have written to show just how alien Mill and Keynes truly are.

Most notably, since a classical model never conceives of recession as a demand-side phenomenon, the notion of hoarding cash was an empty set. There were times when business confidence was lower than usual, and there were times of crisis when the demand for money reached a frenzied peak, but these were effects of some other cause. To think of hoarding as a classical issue is to misunderstand in the most profound way the points that classical economists, and Mill in particular, were trying to make. Keynesian demand-side concerns are the absolute antithesis of classical economic thought.


Yet for all that, it is welcome to find someone trying to go back to the classics to try to tease out some conceptual fundamentals. I just fear that the proliferation of Keynesian concepts has made Mill and the classics virtually incomprehensible. Understood properly, the pre-Keynesian theory of the cycle introduces economists to a very different way of thinking about economic issues. Making the effort to understand the way in which earlier economists approached various questions, if it does nothing else, would help deepen our own understanding of the problems that confront us today, since the contrast would bring out more sharply the contours of our own unexamined assumptions embedded within the frameworks we use. That it may go further, and introduce us to other possibly more fruitful approaches to economic issues, is an even more important reason to read literature from a different era.

But historians of economic thought also have their value in trying to ensure that when the writings of older economists are brought into the conversation, there is an expertise available to provide critical oversight. Someone who has read the classical literature as well as other commentaries on that same literature, brings important background information to economic debates. But to become engaged in such issues requires an enormous amount of work. A proper study of the history of economic thought is as difficult as any single area in the whole of economics. It requires an incredibly intense effort through patient sifting of ideas based on an immense amount of reading and debate. A description of the kinds of attributes necessary for an historian of economic thought is provided by O’Brien (2007: 27–35) and they are very exacting and rare.

In the paper that has instigated this response, we are, however, dealing with something altogether different. It ought to be perfectly clear that one could not “fully describe the pre-1900 evolution” of anything in a brief paper, never mind doing so for what is described in the final summary as “the pre-1900 evolution of the foundational concepts that have shaped the corpus of macroeconomic theory, namely, Say’s Law, hoarding, and credit” (p 54). Each of these areas is so vast as to make any short statement less than a book in length merely that, a short statement. I am afraid that Machovec has underestimated how difficult the task he set for himself is. Yet oddly, Machovec’s second purpose is perhaps even more problematic than the first:

The breadth, depth, rapid, and totally unexpected onset of the financial paralysis of 2008–2009 brought the modeling/forecasting component of macroeconomics under fire, and the resultant hostility threatens, undeservedly, to taint everything with a macro label. Hopefully this paper will sharpen our knowledge of classical reasoning, including the errors, and thereby fortify our confidence in the macroeconomic wisdom that has flowed from the century and a half of reflection that occurred between Smith and Keynes... (Machovec, 2014, p. 304).

Speaking for myself, the last thing in the world I would like to see is anything resembling a fortification of macroeconomics as it now is. It should be pulled down like the rotting hulk it is. It was not the onset of the financial crisis that has discredited macroeconomics but the disastrous Keynesian response which has attempted to generate a revival from the demand side. The stimulus is the cause of our current problems; the financial side of our troubles disappeared no later than the end of 2009. It is the failure to deliver recovery that has discredited macroeconomics, exposing its useless aggregations and vapid policies as the empty theoretical structures they fully represent.

It is here that a proper understanding of Say’s Law and the classical theory of credit would be invaluable (see Kates [2014] for my own personal guide through these issues). Modern macroeconomics has almost nothing to do with Smith or classical thought. It is Keynesian aggregate demand that is at macro’s core. Unless aggregate demand is rooted out as the way we approach economic thought, I fear we shall never understand either what has caused the problems we have nor be able to provide a proper policy guidance to get us out of the mess Keynesian macro has created.


The Global Financial Crisis and the abject failure of the stimulus have unsettled macroeconomic theory and policy. Whether founded on fiscal policy or on monetary adjustments—both of which have been used to generate recovery—the outcome has been unarguably bad. Neither the spending of money nor the adjustment of interest to near-zero levels—both of which being forms of demand-side stimulus and both of which having contributed to massive increases in the size of the deficit—have shown not the slightest tendency to generate jobs and growth. The failure now associated with macroeconomics is not due to a failure to forecast the downturn but the inability to formulate policies that would return the economy to robust rates of growth and low rates of unemployment.

In amongst the fallout from the GFC was a shifting of the blame onto a supposed disappearance of regulation and the introduction of free market policies. It is the market economy that is the supposed villain. The hunt, therefore, has been to find a theoretical foundation for an understanding of the economy as a whole. The kind of certainty that everything was under control that existed not all that long ago was expressed by Akerlof and Shiller at the height of the GFC in 2009:

A repeat of the Great Depression is now a possibility because economists, the government and the general public have in recent years grown complacent. They have forgotten the lessons of the 1930s. In those hard times we learned how the economy really works....

In the middle of the Great Depression John Maynard Keynes published The General Theory of Employment, Interest and Money. In this 1936 masterwork, Keynes described how creditworthy governments like those of the United States and Great Britain could borrow and spend, and thus put the unemployed back to work.” (Akerlof and Shiller, 2009, p. 2, bolding added)

The attitude shown in these words seems to be akin to what must have been the prevailing mood as the armies of Europe marched off to war in 1914. The glib certainties of the time have now been replaced by a more somber appreciation of what we are up against. There is now a search taking place for an improved theoretical structure that will better explain the past and give guidance for the future. In the meantime, macroeconomics as we have known it continues to be taught. But there is less confidence that what we have is the final word. Fortifying the macro wisdom of 2009 ought to be off the agenda.

I, of course, have my own view. It is perspective that has grown out of the classical tradition of John Stuart Mill and is one that, while distinct from the Austrian tradition, is in most ways consistent with it. Haberler’s Prosperity and Depression presents that classical tradition from which the only strand that was able to escape past the Keynesian Revolution has been what is now described as Austrian economics. Properly understood, the classical theory of the cycle, which had a continuous evolution from the time of Adam Smith until the publication of The General Theory in 1936, is the thread that must be picked up. This is the theoretical structure that would explain both what went wrong and what is needed to achieve a reversal of our current misfortunes.

  • 1. This is referred to by Marx as C-M-C’, where one set of commodities (C) is exchanged for money (M) before the money is exchanged into another set of commodities (C’).


Akerlof, George and Robert Shiller. 2009. Animal Spirits: How Human Psychology Drives the Economy, and Why It Matters for Global Capitalism. Princeton, N.J.: Princeton University Press.

Haberler, Gottfried. 1937. Prosperity and Depression. Geneva: League of Nations.

Hicks, John R. 1937. “Mr Keynes and the ‘Classics’: A Suggested Interpretation,” Econometrica 5, no. 2: 147–159.

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

——. 2010. “Why Your Grandfather’s Economics Was Better than Yours,” Quarterly Journal of Austrian Economics 13, no. 4: 3–28.

——. 2014. Free Market Economics: An Introduction for the General Reader. 2nd ed. Cheltenham, UK: Edward Elgar.

Keynes, John Maynard. 1936. The Collected Writings of John Maynard Keynes. Vol. VII: The General Theory of Employment, Interest and Money. In Donald Moggridge, ed. London: Macmillan Press, 1973.

Machovec, Frank M. 2014. “Our Classical Macro Heritage,” Quarterly Journal of Austrian Economics 17, no. 3: 273–312.

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

Mill, John Stuart. 1871. Principles of Political Economy with Some of Their Applications to Social Philosophy. 7th ed. In W.J. Ashley, ed. London: Longmans, Green, and Co., 1921.

O’Brien, Denis P. 2007. History of Economic Thought As an Intellectual Discipline. Cheltenham, UK: Edward Elgar.

Ricardo, David. 1951–1973. The Works and Correspondence of David Ricardo. In P. Sraffa, ed. Volume VIII: Letters, 1819-June 1821. Cambridge: Cambridge University Press.

Say, Jean Baptiste. 1821. Letters to Mr Malthus on Several Subjects of Political Economy and on the Cause of the Stagnation of Commerce to Which Is Added a Catechism of Political Economy or Familiar Conversations on the Manner in which Wealth Is Produced Distributed and Consumed in Society. John Richter, trans. New York: Augustus M. Kelly, 1967.

Spence, William. 1807. Britain Independent of Commerce; Or, Proofs, Deduced from an Investigation into the True Causes of the Wealth of Nations, that our Riches, Prosperity, and Power, Are Derived from Resources Inherent in Ourselves, and Would not be Affected, Even Though Our Commerce Were Annihilated. 2nd ed. London: W. Savage, Bedford Bury for T. Cadell and W. Davies.

Cite This Article

Steven Kates. "Why Keynesian Concepts Cannot Be Used to Explain Pre-Keynesian Economic Thought: A Reader’s Guide to Classical Economic Theory." Quarterly Journal of Austrian Economics 17, no. 3 (Fall 2014): 313–326.