This book by Michaël Assous and Vincent Carret contains an introduction and eight chapters that can be read independently. Even so, the chapters together tell a coherent story and are far from disjointed. Each is a piece of a puzzle: to fully capture the birth of macroeconomics in the 1930s and 1940s.
Assous and Carret offer three major contributions, contributions that make their book essential to read. The book's originality, as well as its analytical strength, lies in the great erudition of the authors and their detailed, technical discussions of several important macroeconomic models. The book should also renew the understanding of economists, historians and nonhistorians alike, of macrodynamic issues, cycle analysis, the accelerator-multiplier concept, and theoretical proposals for countercyclical policies. Some might even regard it as a handbook for a (very educated) audience of economists
First, this book sheds light on the shadowy or still-ignored areas of the emergence of macroeconomics. By focusing on the first works in macroeconometrics by Ragnar Frisch and Jan Tinbergen, Assous and Carret provide an account that is an indispensable complement to the works of Kevin Hoover, Pierre Malgrange, and more specifically Michel De Vroey (2016). Assous and Carret detail and analyze the contributions of econometrics to macroeconomics, particularly around the issues of dynamics and instability. The authors show chapter by chapter how and to what extent econometricians have been key actors in the construction of this new theoretical and analytical field.
The second contribution is to describe and detail the construction of a community, a community under construction and present at the genesis of macroeconomics, a complex community that cannot be reduced to Keynes and the Keynesians, or to econometricians, or to a dialogue and antagonisms between these two distinct groups. If the book can appear as a dialogue between Frisch and Tinbergen, Assous and Carret have the clever idea of putting this dialogue in perspective, relating it to the contributions of other individuals such as Michał Kalecki and James Meade. The authors show in the course of the eight key chapters that this dialogue takes place notably during the conferences of the Econometric Society, whether in Europe or the United States, the shift in the geographical center of the debates being a direct consequence of the Second World War. The guiding thread of the eight chapters is the various economists' investigation of the concept of stability and the theoretical debates that arise as soon as the possibility of an unstable economic system is raised. But what is missing in the conclusion of the book is a summary identifying the key issues discussed at each conference of the Econometric Society and during the 1930s and 1940s more generally that were behind the selection of papers published in Econometrica. Why did the editors of Econometrica decide to publish this paper rather than that one? This historical perspective would allow each chapter in Assous and Carret's book to appear as an episode or a stage in the history of the constitution of a disciplinary field around a community. This heterogeneous community finds its unity in the adoption of a common language—that of the scheme and then of the model, a term that slowly imposed itself on the community from 1935 onward (134n21). Their demonstration of the construction of the community is quite convincing, especially in chapter 8, which analyzes the conditions for the stability of the economic system put forward by Meade's model in 1937 and the criticisms then advanced by the econometricians, Tinbergen and Frisch, against Meade's proposals. This debate, which emerged at the Oxford meeting of the Econometric Society, was about the construction of Meade's model. Chapters 7, 8, and 9, which focus on the gas pedal principle and its articulation with the multiplier principle, make it clear that the debates between econometricians and Keynesians take place only because they share a modeling approach or models as a heuristic tool. A model should borrow heavily from the formalisms of physics, a model that is conceived as a reduction of the world into mathematical language, but a model that is empirical above all.
Indeed, the third contribution of Assous and Carret's book is to insist on the empirical and even applied dimension of investigations into the instability of macroeconomic phenomena. If Assous and Carret show that the history of macroeconomics cannot be fully understood without looking at the contributions of the community that organizes itself around the Econometric Society, they show, reciprocally, that the history of econometrics cannot be reduced to a history of technical questions about inference methods, about model estimation procedures, about statistical tests and the articulation between theoretical-abstract and concrete-empirical measurement, about the number of degrees of freedom, and so on. Assous and Carret's argument is that the question of dynamics that plagues these economists leads them to propose a number of innovations, both technical and theoretical. The technical and theoretical innovations are often based on equations from physics that can describe an unstable system, and this point has already been made in the works of historians of economics such as Mary Morgan, Marcel Boumans, and Philippe Le Gall, as well as Olav Bjerkholt, Philip Mirowski, and the present reviewer. One of the major contributions of Assous and Carret's book is to dissect the models with a scalpel, to analyze their mathematical formalism and point out analogies and borrowings. This book also offers a journey to the heart of their empirical translation, thus revealing the theoretical innovations they bring. Thus, throughout the chapters, we understand that the formalisms developed are tools for thinking ontologically about the nature of economic systems and for questioning equilibrium as their natural state. It is essential to read chapter 6, which renews the story about Tinbergen. It completes the analysis proposed by Erwin Dekker in his biography of Tinbergen published in 2021. Assous and Carret point out that Tinbergen's empirical analysis of economic crises certainly led him to associate the origin of crises with coordination failures and the impact of shocks on the economic system. They show that Tinbergen, however, starts from the question of wage variations to clearly and explicitly pose the question of imbalance and instability from the perspective of an endogenous analysis. Thus Tinbergen would have thought as early as 1935 of the possibility of a macroeconomic system generating its own collapse.
From this theoretical context, Assous and Carret rightly link the theoretical conclusions drawn from the analysis of the instability of systems and the proposals of the various protagonists in their history in terms of economic policy. This book confirms the thesis put forward by Alain Desrosières's various contributions: the macrodynamic model is as much a heuristic tool for understanding the world as it is a tool for policymaking, a tool for action to change the phenomena. The great strength of this book is to insist that for Frisch, Tinbergen, Meade, and Samuelson, economic theory and empirical investigation are inseparable from economic policy, that they are two inseparable sides of the same scientific work, that of questioning economic stability.
Despite the erudition, despite a remarkable mastery of the primary and secondary literature, despite highlighting often-unexploited archives such as the proceedings of the Econometric Society meetings, Assous and Carret have left us with some unanswered questions.
The guiding thread of this book is to show how much the question of instability is constitutive—from the 1930s—of the new analytical territory that is macroeconomics. This investigation of instability is rightly linked by the authors to the development of dynamics. Assous and Carret reduce the latter to its methodological dimension. But as Frisch (1929) points out, it is difficult to reduce dynamics to a method of analysis, as opposed to statics. The econometricians of this period clearly distinguish two categories of concepts when one wants to oppose static and dynamic: on the one hand, it is a question of ontology—phenomena are static or dynamic by nature; on the other hand, it is a question of methodology—the way of analyzing them is either static or dynamic, without forgetting a third method, which is kinematics. Thus a phenomenon that is static by nature can be the object of a statistical or dynamic analysis (but not kinematics). This distinction appears in the common thread of these eight chapters, or in Assous and Carret's way of dissecting an equation and the interactions between the variables and at the end by linking these formalisms to their ability or inability to account for the nature of economic interactions. The introduction to the book, which lays out the main arguments and themes of these two historians of macroeconomics, should have gone into the definition of dynamics in greater depth, to show how much it was swept away in the course of the conferences and then by Keynes's heirs, in particular Meade, Samuelson, and Klein, who invested in modeling as the language of macroeconomics.
The analysis of dynamics and the borrowings from physics lead the reader to wonder how these macroeconomists defined time, particularly in a context in which the institutionalists propose an alternative to mechanical time, namely, historical time. The notion of time, and what links it to dynamics, will be better explained in the next book by Assous, which he is writing with Alain Raybaut of the French National Centre for Scientific Research, by the more substantial integration, beyond the reference to Lundberg in the introduction, in this history of the macroeconomics of instability and disequilibrium, of the innovations brought by the proponents of the Swedish school.
In fact, without references to American institutionalists or the Swedish school, we are reduced to understanding time through borrowings from physics of the first econometricians. But this understanding appears only in the unspoken aspects of these borrowings from the physical science of macroeconometrics. These borrowings are certainly traced and analyzed step by step by Assous and Carret, but they overlook the link between the nature of economic phenomena and the nature of time. Thus they ignore entropy, a concept that questions econometricians such as Frisch and makes it possible to nuance the assertion that Frisch is a partisan or even the champion of a hyperstable model. A too-rapid analysis in chapter 6 of Frisch's model in “Circulation Planning” (1934) reduces the coordination defects of the Frischian model to the organizational defects of market societies pointed out by Tinbergen. However, Frisch links these coordination failures to the analysis of anticipations and to the deployment over time of strategic games between economic agents, an analysis developed at the beginning of the 1930s and well before the impulse-propagation model of 1933. The incursion into the Swedish school will allow Assous and Carret to complete their analysis of the links between linear models, time, and instability by showing the importance of expectations in understanding the dynamics of economic systems.
Indeed, Assous and Carret show precisely how much more complex the thinking of the early econometricians became as they moved from linear to nonlinear models, the latter allowing Tinbergen to define several endogenous paths of collapse. If Assous and Carret's demonstration is convincing, it deserves to be placed in the context of the history of the linear model. The works of Michel Armatte and Mary Morgan have shown that linearization is part of the control of the chaos of the world, as described by Frisch in his lecture 8 at the Poincaré Institute in 1933.
Assous and Carret show convincingly that it is by wanting to control the chaos of the world that their protagonists are led to innovate. However, the authors should have questioned (and will be able to do so in their next book) the very nature of the empirical questions treated, such as transport and regional economics, that is, goods whose demand is derived from a complex space-time relationship. Indeed, should we not also see in the complexity of the empirical cases that Tinbergen and Frisch deal with one of the sources of their ability to innovate and to think differently? I assume that the complexity of the investigated case studies, as well as the analogies with physics, should have been influential. Similarly, the book being written by Assous and Raybaut, which should deal with the analysis of instability by macroeconometricians after the Second World War, will not be able to ignore the question of the stability of economies according to their level of development, since the missions conducted by Tinbergen and Frisch for the UN in regions as diverse as India and Latin America seem to have challenged their certainties and their framework of analysis. But that is another story. We look forward to reading the book by Assous and Raybaut.
