• 07 Feb 2007
  • Research & Ideas

Dividends from Schumpeter’s Noble Failure

Before influential Harvard economist Joseph Schumpeter wrote the seminal Capitalism, Socialism and Democracy, there came the difficult-to-digest Business Cycles. Although the book was a failure, professor Thomas K. McCraw, who has written a forthcoming Schumpeter biography, believes Business Cycles developed Schumpeter's thinking on capitalism and ultimately changed the practice of business history. Excerpted from Business History Review.
by Thomas K. McCraw
Business Cycles was Joseph Schumpeter's least successful book when measured by its professed aims and several other yardsticks. Yet the book contains two vital aspects that have largely been overlooked. First, the prodigious research that went into its writing caused a significant change in Schumpeter's thinking about capitalism. It moved him to a more historical and empirical approach that shaped nearly all his subsequent work. And second, much of the book constitutes a preview of modern, rigorous business history. This article explores both of these elements—not in the spirit of rescuing a neglected classic, because the book is not a classic. Instead, Business Cycles is a noble failure that paid unexpected dividends both to the author and to scholarship. A link to the full version of this article from Business History Review is available at the end of this excerpt.

In 1939, Schumpeter published his two-volume, 1,095-page tome, Business Cycles, after more than seven years of concentrated research. He was fifty-six years old at the time and had been a professor at Harvard since 1932. He was well known throughout the world, having published scores of articles, over seventy book reviews, and three books, including the brilliant Theory of Economic Development (1911; English translation, 1934).1

Schumpeter struggled mightily with the research and writing of Business Cycles. As he told his friend and fellow cycle theorist Wesley Clair Mitchell in 1937, "In order to carry out so detailed an investigation as would be necessary I would have to have a whole research staff working for me." To another friend, he wrote, "I am still a slave to my manuscript and for instance … worried last night till 2 a.m.., on such questions as whether potatoes were important enough in Germany in 1790 to count in the business cycle."2

Even as he wrote the book, he pursued many other activities. As the undisputed star of the Economics Department, he entertained a stream of visiting scholars, led several faculty discussion groups, spent prodigal amounts of time counseling graduates and undergraduates, and taught a heavy load of courses. He also devoted considerable energy to a second big project—a book on money—but decided to defer (and ultimately to abandon) that effort. By the time he neared completion of Business Cycles, he was "in a state of perfect exhaustion," as he wrote in June 1937. Early in 1938, he reported to Harold Burbank, chairman of the Economics Department, "I am half dead and certainly entirely dazed from the long hours I must spend on rereading and touching up my manuscript."3

With his typical intellectual curiosity, Schumpeter strays in Business Cycles onto whatever tangents interest him.

Today, research efforts comparable to what Schumpeter was trying to do often employ teams of half a dozen statisticians, economists, and other social scientists. But in the 1920s and 1930s, this model of academic research was just getting started, and Schumpeter worked almost entirely on his own. As his student James Tobin recalled, "He didn't recruit students to help him; he didn't suggest topics arising in his own research to students for papers or dissertations; he didn't try out the ideas or findings of his draft chapters in seminars. That so enormous an achievement was the product of lonely research tells what a great scholar Schumpeter was."4

The design of Business Cycles—a three-country study of the United States, Britain, and Germany, covering the whole capitalist epoch—was simply too big for Schumpeter or any other scholar to handle alone. But his attempt to do it changed his thinking in a profound way. He adopted a much more empirical and historical approach to economics, which informed both Business Cycles and his subsequent work. The change is quite clear in Capitalism, Socialism and Democracy (1942), one of the seminal nonfiction works of the twentieth century.5

The Main Goal Of The 1939 Book

Schumpeter chose the title Business Cycles not only because the topic was then fashionable (it was the central economic puzzle of the time, and had been even before the Great Depression), but also because he wanted to emphasize the economic ebb and flow that defines capitalism. "Cycles," he writes in his preface, "are not, like tonsils, separable things that might be treated by themselves, but are, like the beat of the heart, of the essence of the organism that displays them."6

This may be true enough, but here Schumpeter extends the theme of cycles into something like a determinate paradigm. He attempts the hopeless task of fitting historical patterns of business booms and busts into predictable wave periods of standard lengths. "Barring very few cases in which difficulties arise," he writes, "it is possible to count off, historically as well as statistically, six Juglars [eight- to ten-year cycles] to a Kondratieff [fifty to sixty years] and three Kitchins [forty months] to a Juglar—not as an average but in every individual case."7

That so enormous an achievement was the product of lonely research tells what a great scholar Schumpeter was. —James Tobin

Clement Juglar, Joseph Kitchin, and Nikolai Kondratieff were prominent business-cycle theorists. References to their work appear hundreds of times in the text of Schumpeter's book, and make it extremely difficult to read…. Schumpeter's student Paul Samuelson later commented that the book's arguments "began to smack of Pythagorean moonshine."8

Schumpeter himself had ambivalent feelings about his framework. As he told Wesley Clair Mitchell, "I must repeat again, lest misunderstanding arise, that I file no theoretical claims for the three-cycle scheme. It is primarily a descriptive device which I have found useful." In the text of Business Cycles, he admits that it "is indeed difficult to see" why boom-and-bust patterns might occur at such determinate intervals. Indeed it is, and Schumpeter's own stance is guarded and empirically informed. As he said to Mitchell, "Any such diagnosis stands and falls with the historical evidence on which it rests." Theory is indispensable to understanding business cycles and capitalism itself. But detailed historical analysis is no less so.9

Schumpeter's Rhetoric—and Keynes's

Schumpeter could be a very persuasive writer, but in Business Cycles he appears at his stylistic worst. Read cover to cover (a chore I do not recommend), the book conveys the impression of an author trying to squeeze a profusion of diverse topics into a single work, all in service to an untenable thesis. Schumpeter had grown up reading ponderous German treatises containing sentences hundreds of words long, and his verbosity in Business Cycles contrasts with the tight prose that marks the work of his fellow economist John Maynard Keynes. The book that made Keynes famous, The Economic Consequences of the Peace (1920), is an easy read, a model of concision and argument. Keynes's General Theory of Employment, Interest, and Money (1936) is much more technical and difficult, but no less a rhetorical tour de force.

In neither book does Keynes stray from his main subject. He writes with total self-assurance. He concedes nothing to the opposition. With consummate skill, he sweeps readers along to the conclusions he wants them to reach, even though his evidence is sometimes weak.10

Schumpeter's own refusal to prescribe remedies vastly reduced the appeal of Business Cycles.

None of this was accidental. During the years when Keynes was writing The General Theory, he tried out his ideas again and again within his elite circle of young economists at Cambridge University. He incorporated their many insights, particularly those of Richard Kahn, whose multiplier effect became crucial to Keynes's thesis. Meanwhile, Keynes discarded some of his own wrong-headed arguments and excess verbiage.11

Schumpeter, working almost alone, seldom exposed his work-in-progress to anyone. He badly needed what Keynes had—a peer group that would tell him, "No, this won't quite do." But he did nothing to assemble such a group. Any number of his colleagues and students—Wassily Leontief, Paul Samuelson, James Tobin (each a future Nobel Prize-winner), Gottfried Haberler, Paul Sweezy, Elizabeth Boody—could have helped. But Schumpeter never asked them. Instead, he plunged ahead, putting far too many words on paper and publishing reams of untested ideas and unedited copy.12

In his review of Keynes's General Theory, Schumpeter reproaches the author for calling his theory "general" when it actually applies to a special situation. Worse, under the guise of a "purely theoretical discussion," Keynes's logic moves from the policy he favors to a theory that will support it, a sequence abhorrent to Schumpeter. Throughout The General Theory, Keynes "pleads for a definite policy, and on every page the ghost of that policy looks over the shoulder of the analyst, frames his assumptions, guides his pen." Overall, says Schumpeter, "the capitalist process is essentially a process of change of the type which is being assumed away in this book." In Keynesian and other macroeconomic models, individual entrepreneurs, companies, and industries vanish from the scene. In complete contrast to Schumpeter's approach in Business Cycles, no mention of a single business firm occurs over the 403-page length of The General Theory.13

Yet what appealed to readers of The General Theory was the very aspect that offended Schumpeter: a plausible diagnosis of the Great Depression, and a prescription for its cure. However legitimate Schumpeter's protests may have been—and a flood of scholarship over the next four decades washed away substantial parts of Keynes's argument—most people at the time cared far more about finding a way out of the depression. Keynes provided a good road map, and that was a contribution of incalculable value. Conversely, Schumpeter's own refusal to prescribe remedies vastly reduced the appeal of Business Cycles. "I recommend no policy and propose no plan," he writes. Instead, his book can "be used to derive practical conclusions of the most conservative as well as the most radical complexion."14

With his typical intellectual curiosity, Schumpeter strays in Business Cycles onto whatever tangents interest him. And, in line with his frequent tributes to "science," he gives ample space to dissenting opinions. If this made his book more difficult and less persuasive, then so be it. Striking an almost defiant tone, Schumpeter says in his preface, "The reader will find the structure of the argument complex. To his justifiable groan I have nothing to oppose but the question whether he expected to find it easy." But however much an author's reach should exceed his grasp, it is not by this much.15

His method of writing resembles that of the American novelist Thomas Wolfe, another untamed genius who was turning out mammoth manuscripts at about the same time. Wolfe's fertile brain, like Schumpeter's, teemed with stream-of-consciousness associations that he scribbled down in the heat of composition. But unlike Schumpeter, Wolfe had a marvelous editor—Maxwell Perkins of Scribner's, who reorganized Wolfe's great novel Look Homeward, Angel and cut its length by almost one-third. Had an editor of Perkins's talents applied his skills to Business Cycles, the book might well have fulfilled—and exceeded—Schumpeter's ambitions.16

The reorganized work would likely have appeared not as one book but as three. The first editorial step would have been to strip the three-cycle scheme from the text and publish it separately as provocative hypotheses about business booms and busts. A second book would have drawn off about 250 of the 309 pages in Business Cycles devoted to the 1920s and 1930s and presented that material as an account of recent economic trends.17

The third book would have consolidated and elaborated the splendid passages on the long-term evolution of business.… In that form, the book would have been a successful marriage of history and theory, and potentially a masterpiece. Among other contributions, it would have signaled the birth of truly rigorous business history. Schumpeter had a very strong attraction to history, and after writing Business Cycles, he urged again and again that economists pay more attention to it.18

Read the full version of this article

About the Author

Thomas K. McCraw's biography Prophet of Innovation: Joseph Schumpeter and Creative Destruction is scheduled to be published in April by Harvard University Press. Footnotes

1. The Theory of Economic Development: An Inquiry into Profits, Capital, Credit, Interest, and the Business Cycle (1911; translation into English by Redvers Opie, Cambridge, Mass., 1934).

2. Business Cycles: A Theoretical, Historical, and Statistical Analysis of the Capitalist Process (New York, 1939). Schumpeter to Mitchell, 6 May 1937; Schumpeter to Oscar Lange, 24 Feb. 1937, both in Joseph A. Schumpeter: Briefe/Letters, eds. Ulrich Hedtke and Richard Swedberg (Tübingen, 2000) (hereafter cited as Briefe), 295, 301, 303.

3. Committee on Research in the Social Sciences, Directors' correspondence, letter, Schumpeter to Committee (16 June 1937), in box 4 (P-Z), UAV 737.18, Harvard University Archives. Department of Economics, correspondence and records, 1930-61, letter, Schumpeter to Burbank (17 Jan. 1938), in box Robertson-Schumpeter, UAV 349.11, Harvard University Archives.

4. James Tobin, foreword to Eduard März, Schumpeter: Scholar, Teacher, Politician (New Haven, 1991), ix. Most reviews of Business Cycles were favorable. See E. Rothbarth of Cambridge, Economic Journal 52 (June-Sept. 1942): 223-29; J. Marschak of the New School, Journal of Political Economy 48 (Dec. 1940): 889-94; and Oscar Lange of the University of Chicago, Review of Economic Statistics 23 (Nov. 1941): 190-93. Hans Neisser of the University of Pennsylvania noted several problems with Schumpeter's argument, but conceded that "it will always be a marvel that such a book could be written by one man." Hans Rosenberg of Brooklyn College wrote, "This work cannot be merely read; it must be studied." Neisser, Annals of the American Academy 208 (March 1940): 205-6; Rosenberg, American Historical Review 46 (Oct. 1940): 96-99.

Simon Kuznets, a business-cycle theorist, pioneering macroeconomist, and future Nobel Laureate then based at the Wharton School, wrote the longest and most important critique. He praised Schumpeter for having written a "monumental treatise" that raised all the right questions, but he argued that cycles are quantitative phenomena, and instead of robust numbers Schumpeter had presented "an intellectual diary." He had told of his "journey through the realm of business cycles and capitalist evolution," disclosing "his encounters there with numerous hypotheses, diverse historical facts, and statistical experiments." These efforts were praiseworthy, but they could not substitute for tight quantitative analysis. See Kuznets, American Economic Review 30 (June 1940): 257, 266-71. A response came from Nicholas Mirkowich of the University of California, who pointed out that the book is not about cycles alone, but is a more general theory of economic development, extending the work Schumpeter had begun in his first two books; and contending that readers should not throw out the baby with the bathwater. See Mirkowich, American Economic Review 30 (Sept. 1940): 580.

5. Including Business Cycles, Schumpeter's published work for the remainder of his career totals about two million words. Of this work, Nathan Rosenberg has argued, "The fact is that most of what Schumpeter wrote qualifies as history, both economic and intellectual. More than this, the very subject matter of economics, in Schumpeter's view, is history." Rosenberg, "Joseph Schumpeter: Radical Economist," in Yuichi Shionoya and Mark Perlman, eds., Schumpeter in the History of Ideas (Ann Arbor, Mich., 1994), 56. In his book Schumpeter and the Endogeneity of Technology: Some American Perspectives (New York, 2000), Rosenberg makes a thorough case that Schumpeter was fundamentally an economic historian.

6. Business Cycles, vol. 1, v. This was by no means the first time Schumpeter had made such an argument. See Schumpeter, "Über das Wesen der Wirtschaftskrisen," Zeitschrift für Volkswirtschaft, Socialpolitik und Verwaltung 19 (1910): 271-325; and Harald Hagemann, "Schumpeter's Early Contributions on Crises Theory and Business-Cycle Theory," History of Economic Ideas 11 (2003): 47-67.

7. Business Cycles, vol. 1, 169, 173-74. Schumpeter would likely have been better advised to focus on the basic Juglar cycle, which for his purposes was the most plausible and important of the three. At times in the book he wrote not just of three cycles but of five, and in fact there could be many more than that, as innumerable stock-market analysts have tried to show. But the overall point was the one he made in his preface: Cycles are of the essence in capitalism; and it follows that depressions are an inescapable and even beneficial phase in its evolution. In a letter to the prominent American scholar Paul Homan, Schumpeter elaborated a bit: "For if one thinks of business cycles as the typical form of capitalistic evolution and if one looks upon those long time movements, which are sometimes called industrial revolutions, as one species of business cycles, it is but natural to link up with the cyclical phenomenon practically the whole of the economics and sociology of capitalist society." Schumpeter to Homan, 2 Apr. 1938, in Hedtke and Swedberg, eds., Briefe, 309.

8. Business Cycles, vol. 1, 299; Samuelson, "Joseph A. Schumpeter," Dictionary of American Biography, suppl. 4, 1946-1950 (New York, 1974), 723. This is not to say that the three-cycle scheme has no value at all. See, for example, Allen Oakley, Schumpeter's Theory of Capitalist Motion: A Critical Exposition and Reassessment (Aldershot, U.K., 1990); Christopher Freeman, "Schumpeter's Business Cycles Revisited," in Arnold Heertje and Mark Perlman, eds., Evolving Technology and Market Structure: Studies in Schumpeterian Economics (Ann Arbor, 1990), 17-38; Maria Brouwer, Schumpeterian Puzzles: Technological Competition and Economic Evolution (New York, 1991), esp. ch. 1; and Mümtaz Keklik, Schumpeter, Innovation and Growth: Long-cycle Dynamics in the Post-WWII American Manufacturing Industries (Aldershot, U.K., 2003).

9. Business Cycles, 174, 224. If, for example, a war or natural calamity interrupted a period's prosperity, Schumpeter still counted that period as prosperous in his cyclical schema. Otherwise, he wrote, it would "be inadmissible for a doctor to say: 'Organically this man is perfectly sound. If he is dying that is due to a brick which has fallen on his head.' I strongly feel that we must get thoroughly rid of the prejudice that our [cyclical] phenomena are simple and can be directly handled by simple methods either theoretical or statistical." Schumpeter to Mitchell, 6 May 1937, in Hedtke and Swedberg, eds., Briefe, 301, 303.

10. Schumpeter sent Keynes a copy of Business Cycles, and received a cordial reply. He then wrote in a letter to Keynes, "I cannot visualize you really wading through those two vols—or else I should apologize, not only for my egotistical concentration on my own tale (which you so generously forgive) but also that terrible size due to my wish to deal fully, historically and statistically, with 16 units of what I call the Juglar cycle, pointing laboriously in every instance to where my schema fits the facts and where it doesn't." Schumpeter to Keynes, 3 Oct. 1939, in Hedtke and Swedberg, eds., Briefe, 319-20.

11. On the composition, publication, and reception of The General Theory, see Robert Skidelsky, John Maynard Keynes: The Economist as Savior, 1920-1937 (New York, 1992), chs. 14-16.

12. In the acknowledgments to Business Cycles, Schumpeter thanks a few research assistants and, briefly, two of his Harvard colleagues, Seymour Harris and W. L. Crum, as well as "Professor Gordon, now of the University of California, and Dr. Clausing of the University of Bonn," but not the many other friends and colleagues who could have given him useful advice—not even his wife Elizabeth Boody Schumpeter, herself a very capable economist (vol. 1, pp. vii-viii).

13. Schumpeter, review of The General Theory of Employment, Interest and Money, in Journal of the American Statistical Association 31 (Dec. 1936): 794-95. In The General Theory, Keynes names dozens of economists and describes their ideas in detail—often very critical detail. He takes no note of Schumpeter, and little of other European business-cycle theorists. Keynes did not read German easily, and he paid insufficient homage to the giants of either the German tradition of economics or the Austrian: to the Germans Gustav Schmoller, Werner Sombart, Max Weber, Arthur Spiethoff, and Adolph Löwe; to the Austrians Carl Menger, Friedrich von Wieser, Eugen von Böhm-Bawerk, Ludwig von Mises, and Friedrich von Hayek. Many distinguished economists besides Schumpeter wrote negative reviews of The General Theory, among them Frank Knight, Jacob Viner, A. C. Pigou, Dennis Robertson, and Alvin Hansen, who later changed his mind and became an ardent supporter of Keynes.

14. Business Cycles, vol. 1, vi. Schumpeter goes on to say that he wishes "to make it clear that my analysis lends no support to any general principle of laissez faire."

15. Ibid., v-vi.

16. Perkins also edited most of the work of F. Scott Fitzgerald and Ernest Hemingway. In addition to condensing Look Homeward, Angel, he cut Wolfe's Of Time and the River by about half. See David Herbert Donald, Look Homeward: A Life of Thomas Wolfe (New York, 1987), 202, 294-303; and the long discussion on pp. 464-84 of the controversial work of another editor, Edward Aswell, on Wolfe's posthumously published novels.

17. In 1964, the economist Rendigs Fels of Vanderbilt, a former student of Schumpeter's, condensed Business Cycles by about 50 percent, though not in the pattern I have described here. McGraw-Hill, the original publisher, brought out this abridged edition. It was not very successful.

18. See, for example, "The Creative Response in Economic History," Journal of Economic History 7 (Nov. 1947): 149-59; "The Historical Approach to the Analysis of Business Cycles," in Universities-National Bureau Conference on Business Cycle Research, Conference on Business Cycles (New York, 1949), 149-62; "Economic Theory and Entrepreneurial History," in Change and the Entrepreneur (Cambridge, Mass., 1949): 63-84; and History of Economic Analysis (New York, 1954).

One of the most significant early signs of Schumpeter's movement toward a greater appreciation of history (and sociology) was his appreciative reconsideration of the leader of the German Historical School. See Schumpeter, "Gustav v. Schmoller und die Probleme von heute," Schmollers Jahrbuch für Gesetzgebung, Verwaltung und Volkswirtschaft im Deustchen Reich 50 (1926), vol. 1, 337-88. For a discussion of the importance of this piece, see Richard Swedberg, Schumpeter: A Biography (Princeton, N.J., 1991), 82-89. On the whole, however, Schumpeter's turn toward history as represented in Business Cycles does not signal a revival of the German Historical School. As is well known, Carl Menger and other leaders of the heavily theoretical Austrian School in which Schumpeter had trained at the University of Vienna engaged in a long and bitter Methodenstreit (struggle over methods) with the German Historical School. Schumpeter lamented the Methodenstreit and respected the Historical School; but he regarded it, on the whole, as handicapped in constructing the kinds of comparative histories that would inform theory.