Richard M. Ebeling, ed.
Economics Education: What Should We Learn About the Free Market?
Hillsdale: Hillsdale College Press, 1995, pp. 109-136.

The Persistence of Keynesian Myths: 
A Report at Six Decades

Roger W. Garrison

The Keynesian legacy is twofold. In the academic arena, Keynesian theory has caused professional economists who are generally sympathetic to the ideal of a free society to doubt that an unbridled market economy can bring us prosperity and stability. In the political arena, Keynesian ideas have served as a justification for bridling the market economy with fiscal and monetary policies. The Full-Employment Act of 1946 and much subsequent legislation have allowed policymakers to use the powers of taxing, spending, and money creation in their attempts to achieve the prosperity and stability that is supposedly out of the market's unassisted reach. Keynesian myths persist today largely because of the uncritical presumption that the economy's sluggishness and instability are to be attributed to fundamental defects in the market system rather than to the perversities of the very policies intended to compensate for those supposed defects. 
        The preface to Keynes's General Theory begins with a warning: "This book is chiefly addressed to my fellow economists."⁽¹⁾ Keynes goes on to say, however, that "I hope it will be intelligible to others." Few economists would share in Keynes's hope; his book would strain the imagination and patience of even the most intelligent layman. Yet, dealing with Keynesian myths requires that we delve into Keynes's book. Policies inspired by it have affected us all; the layman has had to educate himself in economics to understand just how. This article is chiefly addressed to that interested layman. I can only hope that his tolerance for economic abstraction, interpretation, and dispute will see him through it.
        Robert Skidelsky has recently published a 731-page volume entitled John Maynard Keynes, The Economist as Savior: 1920-1937. This comprehensive look at Keynes's middle years is the second volume of a three-volume biography. The first volume, subtitled Hopes Betrayed: 1883-1920, was published eight years ago.⁽²⁾ There seems to be no worry that reader interest might wane before the third volume, which will deal with the final nine years of Keynes's life, appears sometime around the turn of the century. John Maynard Keynes, who was born in the year of Karl Marx's death (1883), shared a birthday with Adam Smith (June 5), and died in 1946 on Easter Sunday, remains an enigma despite—or perhaps because of—the unending attention to every detail of every aspect of his being. Debate continues today about his economic vision—as Schumpeter used that term—and about the policies and reforms implied by that vision as well as many others proposed and defended in his name.
        The unending controversy about Keynes and his message is both fascinating and maddening. It is fascinating—almost self-reinforcingly so—partly because of its endlessness. What did Keynes actually have in mind when he wrote his General Theory? How, after all these years, can there be such enduring scope for interpretation? And in the absence of any consensus, why is his message, whatever it is, thought to be a profound one? Trying to figure it all out can become an obsession. Modern macroeconomists afflicted with a consuming interest in Keynes (I'll admit to having a mild case of it) are not altogether different from the modern devotees of Star Trek or the Andy Griffith Show. The one significant difference, of course, is that neither the Trekkies nor the self-styled Mayberrians have any impact on public policy. Unending debate about Keynes is maddening because the continuing elusiveness of Keynes's message in the academic arena is accompanied by a continuing reliance in the political arena on macroeconomic policy prescribed and implemented on the basis of Keynes's authority.
        It is not my intent simply to enumerate the propositions that could justifiably be described as Keynesian myths and offer reasons for the persistence of each one. As will be argued later, such a myth-at-a-time approach may be unproductive. Instead, I take my cue from a book just half as old as Keynes's: Keynes' General Theory: Reports of Three Decades, edited by Robert Lekachman.⁽³⁾ Many a student of my generation, including myself, got started down the road toward some understanding of Keynes by reading the essays in this volume. I now offer an essay of my own that might be appropriate, should some devotee be up to the task, for the sequel: Reports at Six Decades—and Counting.

Interpreting Keynes
The practice among Keynes's interpreters of attempting to combine their own wisdom with the authority of Keynes has become commonplace. After offering his own interpretation of a particular point, Lawrence Klein concluded that "...Keynes did not really understand what he had written." ⁽⁴⁾ Joan Robinson once remarked that "there were moments when we had some trouble getting Maynard to see what the point of his revolution really was." ⁽⁵⁾ Anatol Murad presumptuously entitled his own book What Keynes Means. ⁽⁶⁾ These and many other interpreters are guilty of writing about what Keynes meant, what Keynes must have had in the back of his mind, what Keynes should have said, or how Keynes would have changed his mind—if only he had lived a few years longer. According to Alan Coddington, Robert Clower is guilty of "reading not so much between the lines as off the edge of the page." ⁽⁷⁾ Some three decades after Keynes published his most influential book, Clower offered an interpretation that, although revealing about the workings of a market economy, was not very true to the scripture. It stood in contradiction to some of the most clearly written passages. Nonetheless, Clower asserted that either this is what Keynes had "at the back of his mind, or most of the General Theory is theoretical nonsense." ⁽⁸⁾ Axel Leijonhufvud's interpretation more nearly resembles the views of one of Keynes's predecessors (Knut Wicksell) and one of his ablest critics (William H. Hutt), but Leijonhufvud's assessment of the business of interpreting Keynes is on the mark: "The impression of Keynes that one gains from [reading his interpreters] is that of a Delphic oracle, half hidden in billowing fumes, mouthing earth-shattering profundities whilst in a senseless trance—an oracle revered for his powers, to be sure, but not worthy of the same respect as that accorded the High Priests whose function it is to interpret the revelations." ⁽⁹⁾
        Well-credentialed scholars can be found on both sides of every issue: Is the General Theory a continuation of the work begun in Keynes's Treatise on Money, or is it a revolutionary break from his earlier book? Is the assumption of fixed or sticky prices and wages critical to Keynes's arguments, or is it merely a convenient means of making those arguments? Does the so-called liquidity trap, which prevents a falling wage rate from reducing unemployment and renders monetary policy impotent, figure importantly in Keynes's case for fiscal activism, or is it only a curious extreme with no historical or practical importance? Does Keynes actually contend that the economy can be suffering from widespread unemployment and nonetheless be in equilibrium, or is he saying that market adjustments in circumstances of economywide disequilibrium may need to be augmented with or supplanted by policy prescription? Is the standard textbook rendition of Keynesianism, which largely abstracts from problems of uncertainty, faithful to Keynes's theory, or is the notion that capital markets are debilitated by pervasive uncertainty an important part of his central message? Finally, was it Keynes's intention to condemn capitalism or to provide us with the appropriate policy tools to keep it in repair?
        One difficulty in getting a clear reading of Keynes on these and other issues stems from his mixing of levels of abstraction. He intermingles institutional considerations, historical givens, and abstract expressions: The wage rate, which is determined by labor unions (largely? wholly?), is an historical given; the resulting aggregate supply is "Z(N)," an abstract and somewhat cryptic function of the level of employment. At some points, Keynes is condescendingly cryptic. When faced with the question of whether monetary expansion could trigger an artificial boom, Keynes borrows some imagery from Ibsen and writes: "at this point, we are in deep water. 'The wild duck has dived down to the bottom—as deep as she can get—and bitten fast hold of the weed and tangle and all the rubbish that is down there, and it would need an extraordinarily clever dog to dive down and fish her up again'" (p. 183). Leijonhufvud takes this passage to be an obvious slap at F. A. Hayek's treatment of forced savings: "But here Keynes is ... a dead duck in shallow water"—meaning that he was dead wrong about Hayek, owing to the shallowness of his own understanding of Austrian capital theory. ⁽¹⁰⁾
        Keynes also intermingles the issues of theory, policy, and reform. How does the capitalist system work? Why does it sometimes—or generally—not work so well? How can it be made to work better? And what alternative system might be preferable? Attempts by interpreters to answer these questions selectively have given rise to Keynesian Hydraulics, Keynesian Kaleidics, and what I have called Keynesian Splenetics. These three interpretations are interrelated on the basis of their addressing different combinations of the questions: (1) John Hicks and Alvin Hansen are responsible for the Hydraulics of modern textbooks. A set of interlocking graphs, representing the real and monetary sectors of the economy, determine equilibrium levels of aggregate income and the interest rate and serve as the basis for prescribing combinations of fiscal and monetary policies that will achieve some income/interest-rate goal.⁽¹¹⁾ (2) G. L. S. Shackle criticized this overly mechanistic view with his own Keynes-inspired Kaleidics: two of the graphs in the Hicks-Hansen model, namely the demand for investment funds and the demand for liquidity, are virtually unanchored in economic reality and are prone to sudden and dramatic change. The successive patterns of demand are no more predictable, in the judgement of Shackle's Keynes, than the successive patterns of shapes and colors seen through a kaleidoscope.⁽¹²⁾ (3) Fiscal and monetary tools, then, are of questionable use and, at best, give us an interim fix. With prospects dim for meaningful and lasting improvement in the market economy's performance, the ultimate solution is to replace our depression-prone capitalism with something more stable. Needed reform is radical in nature but can be introduced gradually. It will involve "depriving capital of its scarcity value," "the euthanasia of the rentier," and a "comprehensive socialisation of investment" (pp. 376-78). When Keynes writes of the prospects for such reform and hence for eliminating what he considers to be the more offensive aspects of capitalism, he adopts a distinctive spleen-venting tone, and hence: Keynesian Splenetics. ⁽¹³⁾
        Adding still another dimension to the business of interpreting Keynes is the fact that the lens through which the critics and interpreters peer is a zoom lens. Henry Hazlitt zoomed in to give us the ultimate close-up—or so it must have seemed at the time—by dealing with Keynes's masterpiece on a virtual page-by-page basis. The propositions so gleaned from the General Theory are neither jointly supportable nor even mutually reinforcing. According to Hazlitt's reading, for instance, saving and investment are (i) two names for the same thing and (ii) cause trouble when they diverge. ⁽¹⁴⁾ Interpreters more sympathetic than Hazlitt have tended to zoom back and look at the big picture. At a distance Keynes's verbiage flows together into an extraordinarily fertile inkblot; the lens becomes more like a mirror. Many a macroeconomist has dreamed up an idea only to see it right there in the General Theory on a subsequent rereading. Sidney Weintraub, a leading Post-Keynesian, initially claimed authorship of the notion of mark-up pricing but latter attributed this stockboy's view of the pricing process to Keynes. ⁽¹⁵⁾
        A final understanding of Keynes's message has emerged neither from a page-at-a-time reading nor from an inkblot-at-a-time musing. So, in a seeming exercise of one-upmanship, Fred Glahe has given us a word-at-a-time rendition of Keynes's book. Working with a digital scanner and word processor, Glahe produced a concordance of the General Theory. ⁽¹⁶⁾ Each and every word Keynes used is listed in alphabetical order together with the word count plus page and paragraph references. We learn, for instance, that Keynes used the word "capitalism" four times but "socialism" only twice; "optimism" six times but "pessimism" only twice. Keynes wrote "always" 52 times, "sometimes" 30 times, and "never" only 18 times. What future interpreters will make of all this remains to be seen. I suspect that the very existence of a concordance says more about Keynes and his interpreters than even the most inspired use of it can reveal about Keynes's intended message.
        Even the title of the General Theory has been a source of dispute. Old debate centered on the full title, The General Theory of Employment, Interest, and Money. Textbook discussions that pit Keynesianism against Monetarism sometimes distinguish these two schools on the basis of a key question about the relevance of changes in the quantity of money: Does money matter? The conventional account—in which Monetarists say "yes" while Keynesians say "no"—is hard to reconcile with the fact that Keynes included "Money" in his title. A more revealing debate concerns the short title. Is "general" to be contrasted with "special," as indicated in the single paragraph that makes up the book's first chapter? Or is "general" to be contrasted with "partial," as suggested by Keynes's rejection of simple supply-and-demand analysis and his emphasis on the interdependencies among markets for goods and for labor? Does the appearance of the word "theory" or the omission of the word "policy" carry its own message?
        Some seem to believe that Keynes's book is fundamentally unconcerned with policy and reform; others believe that the book is, despite its title, a tract for the times. As careful a scholar as Don Patinkin has recently admonished one of his fellow interpreters for dwelling on the issues of policy or reform: the General Theory, after all, is a book about theory, as advertised in its title. ⁽¹⁷⁾ On a Glahe-based reckoning, Patinkin is 74.4 percent correct. Including the plural and possessive variants of the words, Keynes wrote "theory" 236 times and "policy" only 81 times. But we can award very little partial credit here. After all, Keynes used the word "euthanasia" only three times (all three on p. 376), but he used it where it counts. Suppose a veterinarian examines your aging horse and writes a comprehensive report consisting of two parts. The first part is a long and ponderous explanation of what all is wrong with your horse, why you shouldn't expect him to recuperate on his own, and how an enlightened application of veterinary medicine could yield some marginal improvement in his condition. The second part is a short conclusion, in which he recommends that your horse be put to sleep. Now, which part of this report is the most "important," and what is its "central" message? The veterinarian used the word "sleep" only once. According to Patinkin, Keynes's last chapter (where euthanasia appears as an action item) "could have been omitted without affecting [Keynes's] central message.... Chapter 24 (together with the other chapters that make up his Book VI) is essentially an appendage to the General Theory, and one should not let the appendage to a text wag its body." ⁽¹⁸⁾
        A more recent and more telling reading of Keynes's title is Skidelsky's. Albert Einstein published his "General Theory of Relativity" in 1915. Keynes was self-consciously and self-confidently posturing as Einstein's counterpart in economics. "Keynes's identification with Einstein," according to Skidelsky, "is ... too clear to miss." ⁽¹⁹⁾ And the impression he intended to create is the obvious one: Keynes is to Marshall what Einstein is to Newton. A. C. Pigou, successor to Marshall, noted early on that "Einstein actually did for Physics what Mr Keynes believes himself to have done for Economics."⁽²⁰⁾ James K. Galbraith draws from Keynes's rhetoric to detail the parallels between the two general theories—of relativity and of employment. For example, "[t]he Classical theorists resemble Euclidean geometers in a non-Euclidean world who, discovering that in experience straight lines apparently parallel often meet, rebuke the lines for not keeping straight—as the only remedy for the unfortunate collisions which are occurring. Yet, in truth, there is no remedy except to throw over the axiom of parallels and work out a non-Euclidean geometry. Something similar is required in economics" (p. 16).⁽²¹⁾ On this Skidelsky/Galbraith reckoning, the title tells us more about the Delphic oracle's view of himself than about what the High Priests can actually find in his book.

Judging by the Standard of Generality
Self-consciously created or not, the notion of Keynesian/Einsteinian parallels is not an indictment of Keynesian macroeconomics. Devising a more general theory, in economics as in physics, is a worthy undertaking. But Keynes's General Theory is put in its worst light precisely when judged by the standard of generality. If ever there was a special theory, applicable only in the short run and under highly restrictive conditions, it is his. Worse, the implementation of Keynesian policy works to make the short run shorter and the restrictive conditions less likely to prevail. And still worse, Keynesian fiscal and monetary activism interferes with the self-correcting forces in the market economy, causing the economy to be less efficient, less stable, and more depression-prone than it otherwise would be. Perversely, the resulting unemployment of labor and other resources and the discoordination in financial markets become the basis for advocating the continuance of policy activism. In an ill-conceived effort to stabilize the economy, policy activism has Keynesianized the economy; the unintended consequences of the Keynesian medicine have helped give presistence to the myths about the nature of the malady. These indictments of Keynesianism derive from the policy analysis of several modern strands of market-oriented macroeconomics.
        The most telling restrictive assumptions in the General Theory are those that create a hard link between the employment of labor and total income in the economy. The common practice of theorizing in terms of aggregate income but stating conclusions in terms of employment has become so ingrained that the hard link is simply taken for granted. Entrenched analytical convention has a way of dulling our awareness of the problems it skirts. Any theory claiming to be general, however, must soften this hard link. While a change in total income might reflect a change in the overall employment of labor, it might reflect instead a change in the value of labor or a change in the mix of skilled and unskilled labor or a change in the value or quantity of some other productive input. Following typical textbook presentations, total income (Y), for an economy in equilibrium, is given by the sum of three products: Y = RL + WN + iK, where L, N, and K are the classical factors of production (land, labor and capital), and the coefficients R, W, and i are the respective prices (the rental rate, wage rate, and interest rate). For an economy out of equilibrium, we would also have to include a fourth term to represent entrepreneurial profit. To be consistent with the basic Keynesian vision, an increase in income of, say, ten percent must imply an increase in the employment of labor of ten percent plus corresponding increases in the employment of both land and capital. That is, all macroeconomically relevant changes are unidirectional changes in resource utilization and affect only the scale of operation for the economy as a whole. Keynes explicitly assumes (in his often-overlooked Chapter 4) a fixed structure of industry, which precludes relative movements in factor incomes. ⁽²²⁾ He even assumes a fixed mix of labor skills and thereby precludes the possibility that an increase in skilled relative to unskilled workers might increase the effective labor input while the actual number of workers remains unchanged or even decreases. An increased N, then, always means more workers employed. This same structural straitjacket, incidentally, is also worn by Keynes's aggregate supply function, Z(N), mentioned earlier.
        The hard link between N and Y (and between N and Z) blinds us to the economy's ability to adapt to all sorts of relative changes in market conditions and even imposes limits to changes in the economywide scale of operation. "Bottlenecks," as Keynes called them (pp. 300-301 and passim), may develop in some sectors of the economy, preventing the economy as a whole from actually reaching a position of full employment and igniting inflation while unemployment still persists. These bottlenecks have their counterpart in modern textbooks as the (rounded) elbows that characterize the backwards-L aggregate supply curves. They mark the point (or region) where quantity adjustments give way to price adjustments. In a biting footnote, Hayek discussed Keynesian bottlenecks (and corresponding elbows) in the light of Marshallian supply curves: "I should have thought that the abandonment of the sharp distinction between the 'freely produced goods' and the 'goods of absolute scarcity' and the substitution for this distinction of the concept of varying degrees of scarcity (according to the increasing costs of reproduction) was one of the major advances of modern economics. But Mr. Keynes evidently wishes us to return to the older way of thinking. This at any rate seems to be what his use of the concept of 'bottlenecks' means; a concept which seems to me to belong essentially to a naive early stage of economic thinking and the introduction of which into economic theory can hardly be regarded as an improvement."⁽²³⁾
        The nature of changes in the economy's total income is further restricted by assumptions about the movements in wages and prices. Much of the General Theory makes use of the simplifying assumption that both the wage rate and the price level are fixed. Any increase in income, then, is sure to be a real increase rather than only an increase in its nominal, or dollar, value. Interpreters are correct, however, in arguing that fixed or sticky wages are not central to Keynes's theory. Prices and wages that are flexible—even perfectly flexible—do come into play. But Keynes sees the wage rate and the price level as always moving together such that the real wage rate is unaffected. He hedges his bets by claiming, for instance, that (even in the classical view) changes in the nominal wage rate cause prices to change almost in the same proportion leaving the real wage and the level of employment practically unchanged (paraphrased from p. 12). One might think that when the hedges are dropped, all bets are off. But it isn't so. Bottlenecks and relative movements in prices and wages aside, all changes in demand, when the economy is operating below potential (and Keynes believed that it generally is), are taken to imply proportionate changes in employment. Whether the nominal wage rate and the price level are separately fixed or jointly variable, the real wage rate, defined simply as the nominal rate adjusted for price-level changes, is forever constant. Critics are entitled to wonder how the real wage ever got to be what it is. The explanation would have to lie outside of Keynesian macroeconomics, for Keynes offered no special theory of the real wage rate and certainly no general theory of it that subsumed the allegedly special classical case.
       Sins of commission get compounded by sins of omission. Special attention to the time element was essential, in Keynes's view, in making the leap from the classical vision to his own. But the many direct references to the time dimension as well as numerous indirect references involving the uncertainty of the future, the importance of expectations, and the volatility of long-term asset prices contribute more to the book's tone than to the force of its argument. Any explanation of the actual market process that allocates resources over time—whether that process is believed to work well or to work poorly—must draw heavily from capital theory. But capital theory, which deals with the intertemporal pattern of resource allocation within the investment sector, was no part of Keynes's attempted move toward generality. Further, the rate of interest, which was widely believed to govern the allocation of resources both within the investment sector and between consumption and investment, was given alternative duty by Keynes, namely equilibrating the market for liquidity. On the key issue of the intertemporal allocation of resources, it did not even matter that there was no interest rate available to do the job; there was, in fact, no job to be done. The assumption of structural fixity precluded most all of the allocative processes that capital theory illuminates. It is worth noting here that a more legitimate claim to having achieved generality could have been made by Keynes's arch rival Hayek, who had already begun to build a capital-based macroeconomics and to identify those circumstances under which the market mechanism for allocating intertemporally worked well and those under which it worked poorly.⁽²⁴⁾
       How could Keynes expect to arrive at more general conclusions—or any conclusions at all—with such an assumption-encumbered, hedged, and capital-free theory? The key, it turns out, was expectations, which could turn the argument one way or the other depending on where it needed to go. Expectations served Keynes as something of a wild card, a get-out-of-jail card, a license for theoretical free-wheeling. Falling prices can send demand up or down depending on whether they are expected to keep falling or to rebound. A low rate of interest may or may not persist depending on whether it is taken to be an abnormally low rate or expected to become a new normal rate. Changes in the terms of trade could be exacerbating or ameliorating depending on which would better fit Keynes's immediate purpose. Allan Meltzer justifiably faults Keynes for his cavalier treatment of expectations. ⁽²⁵⁾
        Adopting restrictive assumptions that highlight unidirectional and lockstep changes in resource utilization to the exclusion of all relative changes and positing abrupt discontinuities in supply curves are the makings of a very special—and especially implausible—theory. Imposing similar restrictions on price changes relative to wage changes and omitting all scope for adjustments in the capital structure pushes Keynes's theory even further in this direction. It is difficult to conceive of structural fixity and bottlenecks as important aspects of some truly general theory for which classical economics is a special case.
        The generality achieved by Einstein is reflected in his equations by their inclusion of the Lorentz factor (1 - v²/c²)^-1/2, where v is the velocity of the object under study and c is the speed of light. For speeds of bowling balls, cannon balls, and much else, this factor differs insiginificantly from unity and the conclusions of Einstein's more general theory are the same as those of Newton's special theory. (Note, though, that Newton's so-called special theory is the relevant theory in all cases except those in which the speeds under investigation are close to the speed of light.) Keynes hints at an analogous relationship between his general theory and classical theory: If our policies "succeed in establishing an aggregate volume of output corresponding to full employment ... the classical theory comes into its own again from this point onwards" (p. 378). But despite this rhetoric, there is no Lorentz factor at work here that identifies classical theory (which features relative movements of both prices and quantities in the face of scarcity) as a special case of Keynesian theory (which deals only with the economy's overall scale of operation in conditions of economywide resource idleness). Quite to the contrary, the implied analogy between the speed of light and full employment weighs against Keynes perspective on his own theory: as objects approach the speed of light and as the economy approaches full employment, considerations of relativity and of scarcity give preeminence to Einstein over Newton and to Marshall over Keynes.
        In the light of the Keynesian/Einsteinian parallels, we see that Keynes's intent was to make a classical economist who applies supply and demand analysis to the markets for goods, labor, and loanable funds look as foolish as a Newtonian physicist who chases moonbeams in hopes of getting a fix on the speed of light. Keynes actual accomplishment, however, fell short of his intent. He succeeded only in showing how many assumptions, hedges, and omissions were necessary to force-fit economic theory into his own vision of capitalist society. Obvious as it now seems, the idea that Keynes thought he was Einstein has been recognized by only a few scholars—and mostly by those who would count themselves as admirers of Keynes. Thus, the General Theory has not been critically judged by the standard that this view of its author entails.

Posing a Critical Question
Quite independent of what Keynes actually wrote in his General Theory and what interpreters have taken him to mean, there is a critical question that can be posed using the Keynesian macroeconomic magnitudes (consumption and investment)—a question that would draw a broad spectrum of answers from even the most reasonable and well-intentioned economists. We can frame this question in terms of the familiar production possibilities frontier, which always gets some play in the introductory chapters of principles-level textbooks but has yet to become an integral part of macroeconomic analysis. ⁽²⁶⁾ The frontier emphasizes the fact of economic scarcity by showing the various combinations of consumer goods and investment goods that can be produced given the constraints imposed by resource availabilities and the state of technology. (SEE INSERT.) For any point on the frontier, the economy's labor force is fully employed; unemployment (Keynes would attach the adjective "involuntary") is implied by any point inside the frontier. We begin with a wholly private, fully employed economy producing some particular combination of consumption and investment goods that reflects the saving preferences of market participants. The economy is in equilibrium in both the Marshallian sense (market-clearing prices prevail all around) and in the Keynesian sense (all income earned gets spent either by consumers or by investors).

The production possibilities frontier depicts a fundamental trade-off that governs resource usage. The economy's resources can be employed partly in the production of consumer goods (C) and partly in the production of investment goods (I). The investment goods will add to the stock of capital, increasing the economy's capacity for producing both categories of goods in the future. 
      In a fully employed economy, the particular combination of consumer goods and  investment goods actually produced will be represented by a point somewhere on the frontier, such as F1.  In a depressed economy, the production of consumption and investment goods is accompanied by a widespread idleness of labor and other resources. Depression is represented by a point inside the frontier, such as D. 
      A well-functioning market not only keeps the economy on the frontier but also allocates the resources between consumption and investment in accordance with the saving preferences, or time preferences, of market participants. The more saved, the more invested, and the faster the frontier itself expands outward. 
      Saving preferences can change. Suppose, for instance, that consumers become more thrifty. They choose to forego current consumption so as to be able to enjoy higher levels of consumption in the future. The allocation of labor and other resources must be changed accordingly: the economy has to move from F1 to F2, where the production of consumer goods is reduced, and the resources thus freed up are used to increase the economy's productive capacity. 
      The critical question that separates market-oriented economists from Keynesians is: Will the market get us from F1 to F2 in response to an increase in thrift? Austrians, (as well as Classicists and Monetarists) say "yes"; Keynesians say "no." The Keynesians believe that a saving-induced movement away from F1 will result in unemployment as represented by D and that policy activism is required to return the economy to the frontier. The Austrians believe that a saving-induced movement away from F1 can take us to F2 but that government policy aimed at forcing the economy toward F2 (i.e. artificially cheap credit, which stimulates investment) will be self-defeating and will send the economy inside the frontier before the market eventually returns it to F1. That is, genuine saving begets growth; forced saving begets depression.

The Great Depression as the Michelson-Morley Experiment
In 1887, nearly three decades before Einstein published his famous article and nearly two decades before he began to argue that time and space are interconnected, an experiment conducted by Michelson and Morley showed that the speed of light is independent of movements of the source or movements of the observer. The ultimate victory of Einstein's general theory over Newtonian theory, which deals only with a special case, owes much to the facts about the nature of light revealed in 1887. Similarly, the ultimate victory of Keynes's general theory over Marshallian theory, which allegedly deals only with a special case, owes much to the fact of the Great Depression. The key questions that put Keynesian theory to the test seemed obvious: Is the market economy stable? Does it have self-correcting properties? Can market forces translate changes in saving preferences or in attitudes toward uncertainty into investment decisions that make full use of existing resources?
        The depression had lingered on for more than half a decade at the time of the General Theory's debut. Keynesian ideas began to revolutionize economic thinking, and all too soon the answer to these key questions—in academia, politics, and the media—was a confident "No." Still today, glib assertions to the effect that the market does not work well if left to its own devices are backed by nothing more than reminders of the fact of the Great Depression. Academicians, politicians, and journalists have proceeded as if the experience of the 1930s had the same significance for Keynes's theory that the experiment of 1887 had for Einstein's. The continuing reliance on this as-if comparison gives persistence to Keynesian myths, but a consideration of the alternative explanations of the Great Depression reveals that the comparison itself is a myth. In the case of Michelson-Morley, Einstein's accounting of time, space, and mass, given an observed constancy of the speed of light, was simple and elegant, whereas alternative accounts, which made use of modified Newtonian theory, were complex and contrived. In the case of the Great Depression, Keynes's accounting of employment, interest, and money was complex and anything but elegant, whereas alternative accounts, which made use of classical or Austrian—and subsequently Monetarist—theory, were more consistent with the fundamentals of economics.
        The Austrians, for instance, held that the economy could accommodate changes in saving preferences: increased saving would find its way into investment projects, as represented by a movement along a production possibilities frontier. They also argued that ill-conceived macroeconomic policies, nowadays called "stimulant packages," could result in economywide unemployment. If the central bank held the interest rate below its natural level so as to force the economy along the frontier, the increase in investment activity would not be sustainable. Ultimately, the consequence of forced saving is depression rather than growth. This is only to say that, in the economic reality of both the 1930s and the 1990s, "grow" is an intransitive verb: the economy grows—as long as growth is consistent with consumer preferences and the government does not stand in the way. "Grow" as a transitive verb has no application here: the government cannot grow the economy, Bill Clinton's campaign rhetoric notwithstanding. The Monetarists offer alternative accounts that reinforce this general view of the relationship between government and the economy. The "fact" of the Great Depression, then, is not confirmation that the market economy is inherently unstable. Quite to the contrary, it is dramatic evidence of how a market economy can be destabilized by ill-conceived government policy. ⁽²⁸⁾
        Despite the lack of generality of the General Theory and the absence of any actual experience that weighs obviously and heavily in its favor, there is one sense in which Keynes's theory is Einsteinian. When Einstein himself turned his attention away from physics and toward the economic problems of the 1930s, he sounded like Keynes—or worse: Owing to the rapid progress in methods of production, asserted the world's foremost physicist, 

"[o]nly a fraction of the available human labor in the world is now needed for the production of the total amount of consumption goods necessary to life. Under a complete laissez-faire economic system, this fact is bound to lead to unemployment. ... This leads to a fall in sales and profits. Businesses go smash [sic], which further increases unemployment and diminishes confidence in industrial concerns and therewith public participation in the mediating banks; finally the banks become insolvent through the sudden withdrawal of accounts and the wheels of industry therewith come to a complete standstill. ... If we could manage to prevent the purchasing power of the masses, measured in terms of goods, from sinking below a certain minimum, stoppages in the industrial cycle such as we are experiencing today [1934] would be rendered impossible. The logically simplest but most daring method of achieving this is a completely planned economy, in which consumption goods are produced and distributed by the community."⁽²⁹⁾

The Problem Is...
The overarching myth kept alive by Keynesian theory and the "fact" of the Great Depression is simply that market economies do not have the self-correcting properties the classical economists attributed to them and thus are inherently unstable. But we are entitled to speak of myths because of the plurality of reasons given for the unregulated market's instability. In the classroom, I like to play a sentence-completion game—one that reveals just how many reasons there are, each with some basis in the Keynesian vision. "The market economy may have its virtues, but the problem is...." How would Keynes—or a present-day Keynesian—complete this sentence? The problems are: (1) insufficient aggregate demand, (2) an interest inelastic demand for investment funds, (3) an interest elastic demand for money, (4) the "animal spirits" that rule the investment sector, (5) the "fetish of liquidity," (6) wage and price stickiness, (7) money illusion that keeps labor markets in disequilibrium, (8) irrational expectations that affect asset markets, (9) transactions costs, (10) the distribution of income, (11) the public-goods character of risk-taking, (12) the dual-decision character of the earning-and-spending process, (13) the unknowable future, and (14) all of the above. Being anti-Keynesian in the face of this laundry list of problems can involve a lot of hard work. It simply won't do to label as a myth the substantive statements suggested by each item in the list. But dealing with all the issues associated with any one problem, while trying to keep the other dozen problems at bay, can be analytically demanding and rhetorically ineffective. The Keynesian vision itself, which puts all these problems in perspective, has to be called into question.
        In the final reckoning, Keynes's vision was a double vision. The focus of his General Theory, with the exception of the final chapter, was the existing economic institutions and all the problems they entailed. But Keynes's swan song in Chapter 24, which is overlooked or summarily dismissed by almost all interpreters (with the important exception of Allan Meltzer), envisions an economy with a centrally directed investment sector. This economy, by design, has none of the problems that occupied Keynes for the first 23 chapters. Nor does it have any other problems, or so the reader is led to believe by the absence of any discussion in this direction. The Keynesian vision, then, is a highly biased double vision. Keynesian Splenetics, which is based on my own reading of Keynes as guided by Meltzer's "different" interpretation, portrays the General Theory as a lopsided exercise in comparative institutions, one in which Keynes compares capitalism-as-it-is with socialism-as-it-has-never-been. In the context of this understanding, the burden of proof is thrown back to the Keynesians. They must show—both theoretically and historically—that the problems listed above are problems which have better solutions in a socialist system than in a capitalist system. Only by keeping debate on this level of actual comparative-institutions analysis can those of us whose vision favors market solutions to economic problems expect to see an eventual end to the persistent Keynesian myths.

Notes

1. John Maynard Keynes, The General Theory of Employment, Interest, and Money. New York: Harcourt, brace and Co., 1936. (All page numbers in parentheses are references to this book.)

2. Robert Skidelsky, John Maynard Keynes, Hopes Betrayed: 1883-1920, New York: Viking Penguin, Inc., 1986; and idem, John Maynard Keynes, Economist as Savior: 1920-1937, New York: Penguin Press, 1994.

3. Robert Lekachman, Keynes' General Theory: Reports of Three Decades, New York: Macmillan and Company, 1964.

4. Lawrence R. Klein, The Keynesian Revolution, New York: Macmillan, 1961, p. 83.

5. Joan Robinson, "What Has Become of the Keynesian Revolution?" in Milo Keynes, ed., Essays on John Maynard Keynes, Cambridge: Cambridige University Press, 1975, pp. 123-31 (quotation on p. 125).

6. Anatol Murad, What Keynes Means, New York: Bookman Associates, 1962.

7. Alan Coddington, "Keynesian Economics: In Search of First Principles," Journal of Economic Literature, 14:4, pp. 1258-73 (Quotation from p. 1268).

8. Robert W. Clower, "The Keynesian Counter-Revolution: A Theoretical Appraisal," in Clower, ed., Readings in Monetary Theory, Middlesex: Penguin Books, 1969, pp. 270-97 (Quotation form p. 290).

9. Axel Leijonhufvud, Keynesian Economics and the Economics of Keynes, New York: Oxford University Press, p. 35. Yet, MIT's Paul Krugman writes that, after others had failed at the task, "It was left to Keynes to provide a clear explanation of recessions." Krugman, Peddling Prosperity, New York: W. W. Norton and Co., 1994, p. 26.

10. Axel Leijonhufvud, "The Wicksell Connection," in Information and Coordination, New York: Oxford, 1981, pp. 131-202 (quotation from p. 173).

11. John R. Hicks, "Mr. Keynes and the 'Classics': A Suggested Interpretation," Econometrica 5 (April), 1937, pp. 147-59; and Alvin H. Hansen, A Guide to Keynes, New York: McGraw-Hill Co., Inc., 1953.

12. G. L. S. Shackle, Keynesian Kaleidics, Edinburgh: Edinburgh Press, 1974. In his later years, Hicks himself made a move away from his own Hydraulics and towards Shackle's Kaleidics. In what some regard as a recantation, Hicks suggested that the key first-order distinction in Keynes's theory is not between real and monetary sectors but between relationships that are stable (savings out of income and the transactions demand for money) and relationships that are unstable (investment spending and the speculative demand for money). Hicks, "Some Questions of Time in Economics," in Anthony M. Lang, et al., eds., Evaluation, Welfare and Time in Economics, Lexington, MA: D. C. Heath and Co., 1976, pp. 135-51. Both Shackle and the latter-day Hicks emphasize Keynes's summing-up article, "The General Theory of Employment," Quarterly Journal of Economics 51, 1937, pp. 209-23.

13. Roger W. Garrison, "Keynesian Splenetics: From Social Philosophy to Macroeconomics," Critical Review 6(4), 1993, pp. 471-92.

14. Henry Hazlitt, The Failure of the New Economics, New York: Van Nostrand Company, Inc., 1959, pp. 81-85.

15. This particular instance of origination and attribution is documented by Don Patinkin, "On Different Interpretations of the General Theory," Journal of Monetary Economics 26, 1990, pp. 205-43. See pp. 236-37.

16. Fred R. Glahe, Keynes's The General Theory of Employment, Interest, and Money: A Concordance, Savage, MD: Roman and Littlefield Publishers, Inc., 1991. Parts of my discussion here are drawn from my review of Glahe's book, Roger W. Garrison, "Reflections on a Keynesian Concordance," Austrian Economics Newsletter 12:1 (Spring) 1992, pp. 10-12.

17. Don Patinkin, "Meltzer on Keynes" Journal of Monetary Economics 32, 1993, pp. 347-56.

18. Patinkin, "On Different Interpretations," p. 227.

19. Skidelsky, John Maynard Keynes,conomist as Savior: 1920-1937, p. 487.

20. Quoted from a 1936 review. Skidelsky, John Maynard Keynes, Economist as Savior: 1920-1937, p. 585.

21. Also, Galbraith points out the that the early working title of Keynes's book was simply The General Theory of Employment [which became the title of his 1937 summing-up article]. James K. Galbraith, "Keynes, Einstein, and Scientific Revolution," The American Prospect 16 (Winter), 1994. pp. 62-67. If the idea of Keynes borrowing a vision from Einstein seems strange, consider Stephen Hawking's use of a Keynesian theme in his account of creation: "where did the energy come from to create this matter [of the early universe]? The answer is that it was borrowed from the gravitational energy of the universe. The universe has an enormous debt of negative gravitational energy, which exactly balances the positive energy of the matter. During the inflationary period the universe borrowed heavily from its gravitational energy to finance the creation of more matter. The result was a triumph for Keynesian economics: a vigorous and expanding universe, filled with material objects. The debt of the gravitational field will not have to be paid until the end of the universe." Stephen Hawking, Black Holes and Baby Universes and Other Essays, New York: Bantam Books, 1993. p. 97.

22. At issue here is J. S. Mill's fourth fundamental proposition concerning capital: "Demand for comodities is not demand for labor," or, to put it in Keynesian terms, the demand for N does not move in lockstep with the demand for Z. The centrality of capital in macroeconomic theorizing is the theme of Roger W. Garrison, "Austrian Capital Theory and the Future of Macroeconomics." in Richard M. Ebeling, ed., Austrian Economics: Perspectives on the Past and Prospects for the Future, Hillsdale, MI: Hillsdale College Press, 1991, pp. 303-24.

23. Friedrich A. Hayek, The Pure Theory of Capital, Chicago: University of Chicago Press, 1941, p. 374, n.1.

24. Friedrich A. Hayek, Prices and Production, New York: Augustus M. Kelley, 1967; and Hayek, Monetary Theory and the Trade Cycle, New York: Augustus M. Kelley, 1966.

25. Allan Meltzer, Keynes' Monetary Theory: A Different Interpretation, Cambridge: Cambridge University Press, 1988, p. 311.

26. For an attempt to remedy this lack of integration, see Roger W. Garrison, "Linking the Keynesian Cross and the Production Possibilities Frontier," unpublished manuscript, 1994.

27. Hayek, Prices and Production, pp. 49-54.

28. Hayek, Prices and Production, pp. 54-60.

29. Albert Einstein, "Thoughts on the World Economic Crisis," in Einstein, Ideas and Opinions, New York: Crown Publishers, Inc., 1982, pp. 87-91.