1I must start by alerting the reader that this is not an ordinary book review of Michel De Vroey’s A History of Macroeconomics from Keynes to Lucas and Beyond (I am writing one for the European Journal for the History of Economic Thought). It is instead a comment on two specific, and rather crucial, elements of De Vroey’s analysis and a broader reflection on the common depiction of the evolution of macroeconomics produced by practitioners and adopted by some historians. In such accounts, it is too hastily claimed that macroeconomics is a controversial field, with alternative schools of economic thought describing “each other as wrong from the ground up” (Solow, 1983, 279), failing pray of revolutions and counter-revolutions that spread a sense of disarray and lack of progress. Macroeconomists do like very much this framing even if they may use it to argue that there is consensus and progress, and you can take the accounts of three eminent macroeconomists as recent installments of this: Olivier Blanchard (2000 and 2009), N. Gregory Mankiw (1990 and 2006), and Michael Woodford (1999 and 2009)—it is worth adding Snowdon, Vane and Wynarczyk (1994) and Snowdon and Vane (2005) as exemplars here (see Duarte 2012 for further references and discussion). Such views abound in macroeconomic textbooks, blogs, coffee-break talks at conferences, and academic articles and books, and the list of competing schools can be quite long. As I argued elsewhere (Duarte, 2012, 190-194), macroeconomists hold very loose concepts of “schools of thought” and of “revolutions”, patchily adding to them some notions of paradigms, research programs, normal science, knowledge accumulation, and progress.
2 Michel De Vroey in his book does follow a conventional story of macroeconomic schools: it all started with Keynes, then came the (American) Keynesians of the neoclassical synthesis, then Milton Friedman and the monetarists, and Friedman’s and Edmund Phelps’s natural rate of unemployment, then the non-Walrasian and disequilibrium macroeconomics, followed by Lucas and his new classical followers (in particular, the real business cycle macroeconomists), the new Keynesians reacting to Lucas, and the emergence of the so-called “new neoclassical synthesis” in the late 1990s. He also structures his book around the idea of scientific revolutions, a concept that he borrows from Thomas Kuhn ( 1970). But in contrast to macroeconomists, De Vroey is much more careful in using these notions and he understands that “revolution” is a useful concept provided it is not overly used:
Macroeconomists have been fond of using the revolution metaphor. In addition to the ‘Keynesian revolution’, we encountered the ‘rational expectations revolution’, the ‘monetarist revolution’ (or ‘counterrevolution’). For my part, I find Kuhn’s scientific revolution concept useful as long as it is not overused. (379)
3 Both ideas—of schools of thought and of revolutions—have a long history in macroeconomics. Macroeconomists did not wait for Kuhn to identify revolutions in their field. Keynes opposed his economics with the faulty ideas of the “classics” in The General Theory (1936) and he thought that his book “will revolutionise—not, I suppose, at once but in the course of the next ten years—the way the world thinks about economic problems”, as he wrote to Bernard Shaw in January of 1935 (quoted in Snowdon and Vane, 2005, 13). His followers and other economists quickly embraced the idea of a “Keynesian revolution” in the early 1940s: see Dudley Dillard (1942, 66), Martin Bronfenbrenner (1942, 558), Oskar Lange (1945, 131), and, in a symposium on Keynes’s legacy after his death, Gottfried Haberler (1946, 187) and Alvin Hansen (1946, 182). But Lawrence Klein’s 1947 book, based on his 1944 PhD dissertation defended at MIT under Paul Samuelson, certainly helped spreading the idea of a Keynesian revolution. Klein (1947) clearly defined what he meant by revolution, in the foreword of his book:
The Keynesian theory is viewed in the following pages as a revolutionary doctrine in the sense that it produces theoretical results entirely different from the body of economic thought existing at the time of its development. The ‘Revolution’ discussed here is a revolution in thought, not in the economic policies of governments. (vii)
4The disputes over the legacy of Keynes (1936) and what his book really meant revolved around whether or not he had a general theory and whether the classical theory was a particular case of his theory as he claimed. As De Vroey (chap. 2) discusses, comparing Keynes’s and the classical systems was at the birth of the IS-LM macroeconomics and there were those arguing against Keynes’s claim. While Hicks (1937, 153) maintained that by bringing in both the speculative and the transactions motives of money demand “Mr. Keynes takes a big step back to Marshallian orthodoxy, and his theory becomes hard to distinguish from the revised and qualified Marshallian theories, which, as we have seen, are not new”, he did not use the language of revolution to deny that there was one. But Haberler (1946, 193) did:
We thus reach the conclusion that, as far as the logical content of Keynes’s theory goes, i.e., apart from his judgment of the typical shape of the various functions and of the concrete situations and apart from policy recommendations, no revolution has taken place; The General Theory marks a milestone, albeit a conspicuous one, but not a break or a new beginning in the development of economic theory.
5Nonetheless the idea of a Keynesian revolution did not wither. After Klein (1947), at a time when Kuhn had already published his influential book, Harry Johnson (1971) compared the Keynesian revolution and the monetarist counter-revolution. Johnson, in this Richard T. Ely lecture, conspicuously chose to ignore Kuhn’s concept of scientific revolution. Instead, in his “amateurish” excursion “into the economics and sociology of intellectual change” Johnson (1971, 1) adapted the notion of revolution from politics to his purposes:
Its essence is unexpected speed of change, and this requires a judgment of speed in the context of a longer perspective of historical change, the choice of which is likely to be debatable in the extreme. (ibid.)
the chief problem in identifying revolutions and counter-revolutions and distinguishing them from slower and more comprehensible and rational processes of change in economic thought is to arrive at a judgment of the relative speed of change and the degree to which the speed is justifiable. (ibid., 2)
7 Writing just before Robert Lucas’s major contributions, Johnson (1971) identified three revolutions in economics before the 20th century (the Smithian revolution, the Ricardian revolution, and the neoclassical revolution) and four revolutions and one counter-revolution in 20th century economics (the imperfect-monopolistic competition revolution, the empirical or econometric revolution, the general equilibrium revolution, the Keynesian revolution, and the monetarist counter-revolution). He then argued that there were only three “real intellectual revolutions” (ibid., 3): the Ricardian revolution, the Keynesian revolution and the monetarist counter-revolution. The others are in fact evolutions from previous ideas. He thus certainly contributed to strengthening the idea that there really was a Keynesian revolution. The idea of a Keynesian revolution was convenient also to Lucas and Sargent’s (1979) attack on Keynesian macroeconomics. Moreover, a general idea of revolutions in economics was also popular among dissenting economists who embraced Kuhn in a period characterized as “the crisis of economics” (see Mata, 2004 and 2009; Backhouse, 2010, chap. 9).
8 Despite Johnson having chosen to ignore Kuhn, there was a strong Kuhnian spell in economic methodology and in the history of economics in the 1970s and 1980s, with methodologists and historians searching for episodes of Kuhnian scientific revolutions, grasping with the difficulties of applying Kuhn to economics and embracing Imre Lakatos’s concept of research programs (see Hands, 2001, chaps 3 and 7). Relying heavily on these methodological references, Axel Leijonhufvud (1976) revisited the ideas of schools of thought and revolutions in economics. And the Keynesian revolution and the monetarist counter-revolution survived well in economics side by side with another revolution in macroeconomics: that of Lucas and followers of the rational expectation hypothesis, which is the one stressed by De Vroey in his book.
9 The existence of a Keynesian revolution survived so well that David Laidler had to dedicate an entire book to echoing Haberler and providing a rich analysis that places Keynes as a continuity, and not a breach, of earlier economic ideas (Laidler, 1999). As for the rational expectations revolution, Kevin Hoover (1991 and 1992) argued against its existence, defending that a more continuous development captures better the history of (macro)economics (Hoover, 1991), and that the rational expectations hypothesis is not the critical aspect of the new classical macroeconomics, but instead the market clearing assumption (Hoover, 1992). Lucas himself, in an interview with Snowdon and Vane, pointed that economists loosely used the concept of scientific revolution and he denied the view that he was a revolutionary:
Sargent once wrote that you can interpret any scientific development as continuous evolution or discontinuous revolution, at your pleasure. For myself, I do not have any romantic association with the term ‘revolution’. To me, it connotes lying, theft and murder, so I would prefer not to be known as a revolutionary. (Lucas, 1998, 127)
10 Lucas later reinforced his view that economics in the last two hundred years have seen no paradigm change, with a continuous evolving from the economics of David Hume, Adam Smith, and David Ricardo done increasingly with “better mathematics, better mathematical formulation, better data, better data-processing methods, better statistical methods, better computational methods” (Lucas, 2004, 21-22).
11 Notwithstanding these (and other) arguments, we have De Vroey bringing back to the fore the age-old discussion on whether or not there are Kuhnian revolutions in the history of macroeconomics. De Vroey carefully argues that there were two revolutions, the Keynesian and the Lucasian ones, and he implicitly accepts that monetarism was a sort of counter-revolution (though not a Kuhnian transformation). This makes macroeconomists highly efficient revolutionaries, who produced upheavals in their field roughly every 20 years in the period from the mid-1930s to the mid-1980s (in a sharp contrast to the examples of revolutions discussed by Kuhn, which may take even more than a century to happen). De Vroey does so with a clear thesis that he defends in the book: that after the Keynesian era, Lucas inaugurated a new era in macroeconomics with dynamic stochastic general equilibrium (DSGE) models that his real business cycle followers really turned quantitative. A new way of doing macroeconomics is inaugurated, radically different from the preceding era. As such, the argument seems innocuous.
12 But appearances are deceiving and I want to argue against the straightjacket of schools of thought (echoing Backhouse and Boianovsky, 2013, 180-186) and revolutions in macroeconomics. Both ideas impose a great deal of ex post rationalization to the historical narratives and, thus, can be misleading by praising winners and relegating other authors, some of whom could have been influential at the time, to the dustbin of irrelevance. By taking the schools of thought view of macro (even though adding the disequilibrium macro and talking about the large-scale macroeconometric models) and claiming a Kuhnian scientific revolution with Lucas, De Vroey tells a story of a new way of thinking that opposes the Lucasian era with the Keynesian one, as if quantitative reasoning with general equilibrium models were exclusive of the new era. In contrast, if we abandon the ideas of schools and revolutions, and look, for instance, on economists increasingly using models, as Marcel Boumans (2005) and Mary Morgan (2012) argue, we would see a rather different picture that would place Lucas in the context of efforts by such economists as Ragnar Frisch, Jan Tinbergen, Irma and Frank Adelman, Jacob Marschak, Lawrence Klein, Albert Ando, Franco Modigliani, among many others. In this context, a radical breach with preceding developments would be harder to establish—and why would we look for such a revolution then?
13 This alternative picture would, as a by-product, call into question the association that De Vroey makes of DSGE with Lucas, as there were several earlier uses of dynamic stochastic general equilibrium models (Frisch, Tinbergen, the large-scale macroeconometric models). DSGE is a label with a particular meaning—new Keynesian models epitomized by Woodford’s 2003 book and later developments—and thus it is odd to see De Vroey here, and also Backhouse and Boianovsky (2013), giving it a misleading meaning because they go for the literally technical meaning while ignoring that Lucas or Prescott were not the first to construct dynamic general equilibrium models with shocks. Additionally, abandoning schools of thought and revolutions would bring us other more interesting questions for understanding the practice of the macroeconomists, with theoretical disputes losing its great appeal that it has in De Vroey’s book: how are models built? What are the technical requirements for solving them? How do macroeconomists build facts and argue over the appropriate description of the workings of the economy as a whole? What is the role of expert knowledge in policy prescriptions?
14 By insisting on the relevance of schools of thought and of (Kuhnian) revolutions, historians are condemned to being the ones tidying the “mess” of the past: creating classifications of the schools to pigeonhole diverse macroeconomists, with clear tenets that identify each grouping, and then spending much effort in arguing that there are revolutions, and not more continuous evolutions. Here historians would be those arguing for incommensurable paradigms, as the Kuhnian scientific revolution represents a gestalt change:
Led by a new paradigm, scientists adopt new instruments and look in new places. Even more important, during revolutions scientists see new and different things when looking with familiar instruments in places they have looked before. It is rather as if the professional community had been suddenly transported to another planet where familiar objects are seen in a different light and are joined by unfamiliar ones as well. …In so far as their only recourse to that world is through what they see and do, we may want to say that after a revolution scientists are responding to a different world. (Kuhn,  1970, 111)
Just because it is a transition between incommensurables, the transition between competing paradigms cannot be made a step at a time, forced by logic and neutral experience. Like the gestalt switch, it must occur all at once (though not necessarily in an instant) or not at all. (ibid., 150)
15 The incommensurability of paradigms, central to Kuhn’s ideas, created problems to my reading of De Vroey’s analysis. This is because he chose to add Kuhn to a metaphor that he borrowed from Leijonhufvud (2006):
I like to think of the history of thought as a decision tree with the current generation of economists occupying the highest twigs, not all of which will grow into sturdy branches bearing ample fruit. Down below in the tree are a number of more or less important forks, where our predecessors have decided to go one way rather than another. My intention is to revisit some of those forks, to try to understand how people saw the issues at the time and why the majority chose the path that they took, and then ask how the choices made look in hindsight. If we want to learn from the history of thought, I believe, this is the way we have to go about it. (27)
16 For me, going back and forth to different tree branches does not fit well the Kuhnian understanding that a revolution is a radical worldview change, as if the scientists were transported to another planet with no way back to the old planet. But De Vroey is comfortable in using the tree metaphor to depict the several decision trees that according to him describe the evolution of macroeconomics, and then mark the radical change between the Keynesian and the Lucasian eras with a Kuhnian revolution, thus having a clear patron for the new way of doing economics.
17 Let me close this commentary by once again calling the attention of the reader that because I have not written an ordinary book review, my discussion here fails to praise Michel De Vroey’s achievements (at the same time that it leaves aside other disagreements I might have). This book is the result of many years working on and teaching the history of macroeconomics, and De Vroey does offer the readers a careful analysis of the authors he chose to study (he explains his choices and motivations in the preface). He clearly aimed his book at the practicing macroeconomists and economics students, and thus chose the typical representation of the field, with schools of thought and revolutions, to structure his narrative. But the effort of writing an encompassing (even if partial) history of macro is no easy task. His work does contribute to the few existing books on the history of macroeconomics, opening new perspectives, as he wished his contribution to do. The historiography of macroeconomics remains an attractive and active field, with many open challenges to understanding how macroeconomists work and think.