1Science studies contain a wealth of stories about discoveries and the emergence of an idea at a certain point in time, which was to bear significant changes in a field. In economics, those stories are often initiated by economists themselves and tend to be repeated through review articles, textbooks or interviews. Macroeconomics appears as a fertile ground for such stories—think of the Keynesian multiplier, the rational expectations hypothesis, the birth of endogenous growth theories. Among those edifying stories, the one about the discovery of the Phillips Curve and its critics, laying the ground for the expectations “revolution” in macroeconomics, is probably the most famous.
2James Forder valiantly speaks against this well-anchored narrative which has been told for years among economists. The story could be summed up like this:
In what quickly became a classic paper, Phillips (1958) discovered a negative relation between inflation and unemployment; then, either under the influence of Samuelson and Solow (1960) or otherwise, policymakers treated it as offering a selection of inflation-unemployment combinations from which they could choose, depending on their … aversion to the two evils; much work was done investigating this tradeoff and, because of it, inflationist policy was pursued until Phelps (1967) and Friedman (1968a) revolutionized thinking by pointing out that continuous inflation would change expectations and thereby shift the Phillips curve so that there was no long-run tradeoff; and although this was initially disputed, in due course it was accepted. (1)
3As the title of the book suggests, Forder’s goal is to demonstrate step by step that “each component of that story is false” (ibid.). To that end, he provides detailed proofs documenting how the reality of the study of the relationship between inflation and unemployment was far more different and complicated than—and sometimes quite the opposite of—what the “myth” tells us.
4Forder directs the first blow against the supposed centrality of the “curve of Phillips”: neither did Phillips (1958) discover a relationship between wage change and unemployment, nor was his article “instrumental in promoting the idea” (11). Forder first shows that several economists investigated the correlation between the two variables before—among them, Tinbergen or Klein and Goldberger—and that Phillips’ work was weaker than the ones of his predecessors. Besides, in the following years, the literature did not rely on the work of Phillips. When he was quoted, most of the time the author was rejecting Phillips’ idea that market forces (i.e. tight labour market), rather than non-economic institutional factors, constituted the primary determinant of wages variations. It is not before the end of the 1960s that the “Phillips curve” expression spread and that more acknowledgement was granted to Phillips.
5Then, Forder discredits the common beliefs about Paul Samuelson and Robert Solow’s (1960) article which Milton Friedman contributed to diffuse: the two MIT economists never used the Phillips curve to advocate inflationary policy in order to reduce unemployment. Instead, they focused on the instability of the curve and the way to improve the trade-off, to reach full employment and price stability at the same time. Forder analyses Solow’s and Samuelson’s retrospective view on their own article, as well as the way economists talked about the article in the 1960s. He concludes that in the 1960s, economists did not interpret their article as a defence of inflationary policies. However, Forder shows that this false interpretation emerged progressively in the 1970s and became common thought even if not substantiated.
6However, the most crucial point of contention of the book concerns the assumed “revolution” in the story of the Phillips curve: Friedman and Phelps discovering the role of expectations in inflation mechanisms. Forder offers here a considerable survey of the literature on inflation at the time of Friedman’s 1968 AEA presidential address. This literature focuses mainly on wage determination, and many works took into account the role of inflation in it. In other words, this literature did not confuse nominal wages with real wages, contrary to what Friedman later argued; neither did it calculate any optimal trade-off and advocate inflationary policies. Forder also convincingly details how the importance of expectations was acknowledged before Phelps and Friedman, and how even the latter did not seem to consider, in the 1960s, that his presidential address brought revolutionary insights. Nevertheless, it has to be noted here that Forder did not deal with the success of Friedman’s (1968) article. Contrary to Phillips (1958) or Samuelson and Solow (1960), Friedman (1968) quickly encountered tremendous success as well as many economists opposed to it. By dismissing the idea that the article was innovative, Forder tends to suggest that it was not really important and that it did not profoundly change macroeconomics. However, Forder has dealt with this issue in a later article (Forder and Sømme, 2019).
7Finally, the last part of Forder’s demonstration concerns how policymakers would have used the Phillips curve in the 1960s to promote inflationary policies. Here again, it is a mere myth. By analysing official documents and how economists related the policy issues of the time, Forder proves that the Phillips curve was not central in the policy decisions of the 1960s. Interestingly, the Phillips curve more regularly appeared in reports after 1975, notably to criticise the policies of the 1960s. It is at the same moment that the myth developed and that economists began to interpret the policies of the last decade through the Phillips curve framework. The best examples are authors who, like Haberler or Brittan, did not describe 1960s policies in terms of the Phillips curve in the late 1960s, but began to do so after 1972. Forder dated the appearance of the whole myth in the mid-1970s, within the writings of several economists (Friedman, of course, but also Robert Lucas, James Buchanan, Karl Brunner and Allan Meltzer). With the next decade came the routinization: the story of the Phillips curve was narrated again and again without any factual proofs.
8The reading of the first six chapters (over eight) leaves a feeling of accomplishment: thanks to a careful analysis of a large sample of articles and books, we observe every piece of the well-known Phillips curve story collapsing. Forder brilliantly proves why contextualization is an essential weapon for any intellectual history. Understanding how a discipline evolved and how some ideas diffused needs putting the “classics” of the discipline in the context of the time. By confronting many contemporaneous works, by understanding how grand ideas (from today’s perspective) were received, by studying the economic context of the time, the historian offers a broader and more vibrant picture of the problems at stake in the discipline at the time. It then enables to dismiss the imagined pasts that circulate, rebuilt a posteriori by the scientists.
9Thanks to this great effort of contextualization, Forder shows that the Phillips curve was not central in the 1960s and early 1970s for thinking about inflation and employment. By dismissing every piece of this tale, he sketches an alternative history of the treatment of the relationship between inflation and unemployment, rich, complex and displaying useful insights for today’s debates.
10Nevertheless, the book does not go without limits. First, it has the weaknesses of its qualities: Forder’s demonstration goes through the analysis of a large sample of articles, books, reports, etc. Often, references add on references to make the case, to the point where the reader would be tempted to jump to the conclusion. It sometimes seems like a never-ending enumeration, which makes it hard to follow the argument at stake. The aridity of the numerous lists and enumerations leaves the reader with the impression that this story lacks an institutional and network background. Most of the time, Forder does not connect economists with one another, does not provide us with biographical information, institutional details, research programs, to go beyond the mere fact of enumerations. There is one telling exception in the book, when Forder details the influence of Popper’s falsificationism at the London School of Economics and how it pushed Richard Lipsey to work on the Phillips curve and to popularise it. He thus gives a lively colour to the alternative story he is sketching.
11In the seventh chapter, Forder tentatively proposes some leads on how the myth could spread or persist. First, he reminds us that the famous founding papers of the myth (Phillips, 1958; Samuelson and Solow, 1960; or Friedman, 1968) were not representative of the whole literature of the time on the topic. Retrospectively, today’s economists are more prone to (even if still wrong in) believing the myth if they only read these papers rather than more representative works of the time. Another argument hinges upon the semantic confusion surrounding the “Phillips curve” term. Through the 1960s and 1970s, the “Phillips curve” was used in turn to designate cost-push or demand-pull mechanisms, a relationship between unemployment and either wage inflation or price inflation, short-run and long-run relationships, complex representations of wage and price inflation determination or a basic correlation between only two variables (inflation and unemployment). This considerable semantic floating contributed to misleading economists after the 1980s and thus to the passive diffusion of the myth. Forder also gives importance to some initial “misdatings” (181)—like Friedman and Phelps being the first to talk about expectations—which persisted and strengthened over time. Forder finally underlines that opponents to Friedman poorly answered him—and so gave some credence to the story he developed—and that the inflation of the 1970s favoured the story which Friedman, Lucas or Buchanan were developing.
12All these elements surely have a grain of truth, but Forder lacks a systematic analysis of the diffusion of the myth, as a general lesson to learn for the history of economics. He enables us to understand how the myth circulated and strengthened, but not how it imposed itself to start with. One could legitimately aspire to broader and more decisive conclusions.
13There is a fantastic topic for the sociology of science here. “Imagined pasts” remain understudied (Wilson, 2017) and Forder’s book offers a wonderful field of investigation. It raises questions on the consequences of scientific practices, like naming concepts or imposing literature survey section in published articles. Schools of thought and concept labels offer a more organised view of the discipline, but they oversimplify its complexities and lead to misreading its history. The diffusion of the Phillips curve myth is good focus for seeing these mechanisms at work.
14In the concluding chapter, Forder defends the quality and relevance of the work done by economists in the 1960s on inflation and unemployment against the common picture of naivety drawn in the Phillips curve myth. He sharply criticises the Phillips curve and the fact that its existence was never established empirically, whereas it is central in today’s macroeconomics. This quick discussion actually raises essential questions. The transformation of macroeconomics in the 1970s through the impulsion of New Classical economists (Lucas, Sargent, or Edward Prescott) but also of disequilibrium economics (see Backhouse and Boianovsky, 2012) represented a return of “high theory” to the foreground. To what extent did the history of the Phillips curve parallel this process? The rise of the Phillips curve, its conceptualisation and formalisation could be seen as a rationalisation of the research in the field. It offered a platform for discussing inflation in a unified framework, but at the risk of losing the diversity of inflation explanations that could not easily fit the Phillips curve framework. Then, Forder’s conclusion would have deserved a more in-depth discussion of this process and its consequences for macroeconomics.
15The last issue that would deserve more investigation concerns the concrete consequences of the diffusion of the myth. In the introduction, Forder argues that discrediting the myth represents an essential task because the story constitutes a crucial part of the post-1980 consensus and of “the story of the origins of early twenty-first century policymaking presumptions” (4). Nevertheless, Forder does not much pursue this line of investigation in the rest of the book (although he worked on related topics; Forder, 2001). How did the emergence of the myth influence the change in economic policy in the late 1970s? Was the myth instrumental for economists or policymakers to impose some changes of practices? The Phillips curve standard narrative also has some importance for the explanation of the 1970s stagflation. Today, the “ideas hypothesis” (Romer, 2005) constitutes the dominant explanation: the inflation of the 1970s was the result of bad economic policies inspired by false economic ideas (namely the belief in a long term trade-off between inflation and unemployment). The story of the Phillips curve myth debunked, what is left of this explanation? How should the debunking of the myth transform our vision of the 1970s?
16However, if all these critical issues appear during the reading of Forder’s book, it is a proof of how stimulating it is to read and how it opens new paths of inquiry for the future. Undoubtedly, it constitutes an important step in the field of macroeconomics history, and it will inspire innovative future works.