1We would like, first of all, to thank the participants in the Forum for their comments on our book Banks and Finance in Modern Macroeconomics: A Historical Perspective (2019). They contain many stimulating observations on topics and issues that certainly merit further work. New studies and contributions to the analysis of the problems tackled in our book as well as ‘quibbles’ on more specific aspects are welcome.

2The numerous contributions by the authors we considered, or by other scholars not mentioned in the book, certainly deserve more attention and further study. However, in a work like ours, which covers a span of time of about one century, it is inevitable that a number of authors and questions could not be treated in a fully thorough way. Our book can be regarded as an attempt to stimulate the attention of others to the difficult topics, and questions, with which we deal.

3In our rejoinder we first deal with some general issues raised by the commentators with the intent to make our historical interpretation and reconstruction clearer than it might have appeared in the book. We then briefly deal with some more detailed issues usefully brought up by the commentators.

1. The general picture: our vision

4Dimand correctly notes that our book’s aim was not writing a history of macroeconomics. Our focus on banks and finance in contemporary macroeconomics could not exhaust all the relevant episodes and debates which took place within the discipline during the 20th century. We chose to deal with some fundamental figures and episodes which, in our view, are crucial in the exploration of how banks and finance were dealt with, or were put aside, in macroeconomics.

5Without aiming at a comprehensive history, we deal with such figures and episodes by organizing them around some theoretical axes: banks in the quantity theory of money; banks and finance in relation to growth and development; the controversial relationship between Walrasian general equilibrium and money and finance and, last but not least, the troubled coexistence of asymmetric information with the perfectly competitive benchmark of contemporary macroeconomics.

6Large part of our book is concerned with the role and nature of banks in market economies as captured (or ignored) in the evolution of macroeconomic thought since the beginning of the 20th century. The ‘alchemy’ of banking and the elasticity of credit in the payment system are central topics in our historical reconstruction. If banks appear to take the center-stage in our work, this is essentially for historical reasons. At the time when economists like Wicksell, Fisher, Robertson, or Schumpeter, carried out their analyses, banks certainly played a central role in the process of financing the working of the economy and, in particular, the process of production. Friedman attributed the source of instability and the financial disaster during the Great Depression to the elasticity of banks’ money in a less than 100% reserve regime.

7If financial markets are given relatively less attention, it is so not because we believe that, in contemporary market economies, banks are more important than other financial institutions. It could be argued that nowadays non-banking financial institutions play a more important role than traditional banks. In fact, when dealing with Keynes (chapters 5 and 6) and, later on, in chapter 9, we pointed out the importance of financial markets and the largely unsatisfactory way they have been considered by the mainstream for a long time.

8The lack of focus on the complex relationship between banks, financial markets, the network of inside debts and the real side of the economy is the result of some biased perspectives that, we argue, have been dominant in the post-war mainstream macroeconomics. We do not hold that in the second half of the last century macroeconomists ignored banks and finance altogether, or that they denied their importance for the working of the economy, although we point out that some currents of thought indeed ended up with such an outcome. We argue, instead, that most mainstream macroeconomists failed to articulate coherent and fully developed theoretical models in which banks and finance play a significant role [3].

9The crucial point that we make is that this lack of attention, more or less nuanced or even explicitly underlined, results from challenging theoretical questions that were not or could not be dealt with effectively within the structure of mathematical modelling and the frame of ideas along which mainstream economics developed since the mid-20th century.

10Thus, we do not think and argue that macroeconomists ignored the many issues related to money and financial markets. There was a lively debate on these issues, and we attentively qualify our analysis for single authors, currents of thought or periods, to avoid to oversimplify our reconstruction. Besides, central banks were always somewhat attentive to the issues related to financial stability, even when, like in the years before the 2007-2008 crisis, their models gave little or no space at all to the financial structure of the economy [4]. Our main point is that, at the theoretical level, the complexities and the central role of the financial structure in market economies were substantially erased from mainstream macroeconomics because of a variety of biased perspectives, which we try to elucidate and evaluate in their historical context.

11Such ‘biases’ include the exclusive focus on the liquidity preference doctrine, the set of assumptions on the irrelevance of the inside debt structure in the macro-economy, the hypotheses of well-functioning or perfect capital markets, the conventional reliance on the blurred dichotomy between short and long term, the ambiguous reference to the Arrow-Debreu model and the parallel recourse to representative-agent constructions or, in the extreme version of ‘monetarism without money’, just the simple dismissal of money as substantially irrelevant for explaining business cycles [5]. The lack of attention to banks and financial markets in macroeconomic theoretical models has been widely recognised after the 2007-2009 crisis by a number of leading scholars [6].

12Ours is, of course, a challenging enterprise and we welcome additional research and deeper reflection on the issues we raised. We are aware that other authors should have been considered or, possibly, given more attention. These include the Austrians and their business cycles theory, Myrdal of Monetary Equilibrium (1939), or Hawtrey and his monetary theory of business cycles [7]. However, there are excellent books on monetary and business cycle theories covering much of this literature; books which we quote and on which we rely [8]. There are a number of solid contributions to the history of macroeconomics, among which De Vroey’s book (2016) and numerous Dimand’s contributions, which we appreciate and quote.

13Turning to consider the macroeconomics of the second half of the 20th century, in the book we follow two main lines of research. First, we argue that the ‘original sin’ of erasing banks from the main theoretical argumentation can be traced back to Keynes himself in The General Theory, where banks are almost absent and the focus is on liquidity preference. In this respect we point out a mixed, controversial heritage of Keynes’s major work. We underline the duplicity of Keynes’s message as regards the relevance and complexity of the financial side of the economy. While, in The General Theory, Keynes pays little, if any, attention to banks, he has the merit to raise the problem of the role of financial markets in chapter 12. His contribution, ignored for quite a long time by most mainstream macroeconomists, is not too far from recent developments offered by behavioral financial economics (see chapter 9 of our book) [9].

14A second central theme of our book is the coming to dominance of the Walrasian approach and of what has been called ‘monetary Walrasianism’ (Mehrling, 2011). In our book (chapters 6 and 7), in reconstructing the macroeconomic debate after The General Theory we point out that the initial criticisms of Keynes for having ignored, or downplayed, a number of significant market interrelations because of his Marshallian background were essentially correct. Such criticisms, however, did not necessarily imply an attempt to set macroeconomics within a rigorous Walrasian framework.

15It was only in the post-war period that the ‘Walrasian’ approach prevailed, starting with important macroeconomists like Patinkin. In chapter 7-leaving apart specific individual influences, be those of Lange, Hicks, Arrow or Patinkin-we look at the difficult coexistence and confluence of two major research programs: the general equilibrium frame, taking a precise shape during the 1950s, and the building of macroeconomic models which aim at incorporating both ‘Keynesian messages’ and optimizing foundations. In chapter 8, we try to explain the contradictions and the difficult balance of linking monetary and real analysis both in Friedman and in Lucas. As we show in the book, Lucas repeatedly expressed the view that it is possible to reconcile the Arrow-Debreu frame and the introduction of money into macroeconomic models inspired by it. During all his career Lucas has pursued the program to reconcile Arrow-Debreu and the quantity theory, despite disappointments and retreats to the safer ground of real analysis.

16Although Hahn’s results on the impossibility to reconcile money and Neo-Walrasian general equilibrium soon became well known to the circle of general equilibrium scholars, we conjecture that they were not openly admitted as true, or even clearly recognized by the macroeconomists of the time. However, well before Hahn’s theorems, since the 1940s Patinkin was aware of the difficulties encountered by the program to introduce money into the general equilibrium frame. Moreover, since the late 1920s and the 1930s, major theorists (Hicks, Hayek, Myrdal, Schumpeter) debated the question of how to link the core exchange equilibrium theory to monetary and business cycle analysis. Schumpeter was not satisfied with Walras’ monetary theory, and there was an open debate on the issue since at least the 1920s. In the 1930s Hicks was perfectly aware of the problem. Our reference to the Wicksellian and Walrasian roots of the IS-LM model did not intend to attribute to Hicks the exclusive, or dominant, responsibility to constrain macroeconomics within the straight-jacket of a Walrasian frame. Quite to the contrary, he was more aware than other scholars of the difficulties to reconcile monetary theory with equilibrium notions, be those Walrasian or Marshallian [10].

17In our book we mentioned all these issues and debates only in passing, as it was impossible to deal with them in a more detailed way. We wanted to point out that the problem is much older than the Arrow-Debreu model and ‘Hahn’s question’. The difficulty to unify equilibrium theory and monetary theory dates back to Jevons and Walras, to Wicksell and Marshall. It has been a structural problem of economic theory since the ‘marginal revolution’, which defined the static exchange equilibrium in the absence of money as the core equilibrium concept in economic theory [11]. Thanks to its axiomatic structure, the Arrow-Debreu model simply made the difficulty surface more clearly.

18We concentrate on the Neo-Walrasian approach to macroeconomic theory because it has progressively gained the status of a sort of ideological and methodological banner along with the rise to dominance of New Classical Macroeconomics and the following derivations. In many academic research communities there is a deeply entrenched ideological attachment to a vision of the market as an invisible hand operating to ensure a welfare optimum; a frame of mind which waits to be explored from the perspective of the sociology of science.

19In more general terms, the central theme of our book is the absence of the auctioneer in market economies, and the structural dominance in decentralized markets of conditions of asymmetric information in a context of change and innovation. This is a central aspect of our historical reconstruction. It is explicitly enunciated in the Introduction and we return to it in most chapters. In the absence of an auctioneer and in conditions of asymmetric information and change, monetary transactions and the overall flows of payments take place in time. There are time consuming adjustment processes in payments and in credit markets, not only in banking but also in financial markets in general. The attention to the payment system, to dealers in the money markets, to the institutional system of banks and financial institutions of various nature becomes crucial to understand coordination processes in decentralized markets, and how they may work or fail, or be exposed to periodical crises and disruptions.

20We believe that the remarks above largely respond to De Vroey’s criticisms of the book. His comments, however, require some further considerations. De Vroey’s contribution is an interesting example of his encyclopaedic knowledge of contemporary macroeconomics and a concise exposition of his approach to the history of macroeconomics. Whatever the merits or limits of his point of view, we believe that he fails to perceive the main issues and questions that we want to consider in our book, that is to say how modern macroeconomics dealt with banks and finance and, more specifically, why these topics were virtually ignored or largely downplayed by mainstream macroeconomists.

21On these issues De Vroey makes just two remarks: i) that to expel money from the study of business cycles was a choice motivated by the empirical success of RBC models; ii) that, contrarily to our alleged view, there has been an explosion of DSGE models incorporating finance after 2008. Thus, it seems that until the rise of RBC, money, banks and finance were well entrenched in macroeconomic theory. A point of view different from ours. Besides, De Vroey’s two assertions are rather puzzling: if there was an explosion of research on finance in DSGE models after 2008, perhaps something was missing in the previous generation models. In any case, De Vroey’s remarks confirm rather than reject our narrative. In particular, differently from what De Vroey asserts, we recognize the growing attention to credit and finance in DSGE models after 2008, and even before in relation to the asymmetric information revolution (Ingrao and Sardoni, 2019, pp. 215–220). Our point is not to deny the new interest and the new efforts, but to underline the internal limits and inadequacies which they suffer, because of the theoretical scaffolding on which they are built [12]. We shall return to this aspect later on.

22The discussion of other general and methodological issues raised by De Vroey would require much more space than a short rejoinder. Two points, however, cannot be ignored, that is to say our alleged ‘reconstruction’ of macroeconomics inspired by Marx, Schumpeter and Hayek and De Vroey’s idea of how proper history of macroeconomics should be done.

23We have never advocated some new macroeconomics based on the bizarre mix of Marx, Schumpeter and Hayek [13]. Our reference to these authors has been misinterpreted. We refer to them as three scholars, who dealt with the dynamic, structural aspects of competition, but we do not think that it is sufficient to look back at them for a satisfactory development of macroeconomics [14]. Moreover, it is quite weird to hold that we find inspiration from Hayek and Schumpeter and, at the same time, to regard us as a sort of partisan anti-capitalist militants, put together with the anti-liberal anti-market crowd.

24From the perspective that we have adopted in writing our book, we modestly point out some issues and topics that, in our opinion, need be further developed to provide a better understanding of the working of the economy and financial markets. We certainly do not pretend to have offered a new and well developed macroeconomic theory. In the long passage from our book (Ingrao and Sardoni, 2019, p. 19–20), cited by De Vroey, we make some methodological general points and put forward some suggestions. De Vroey sees our quotation as an example of our erroneous way to do history of macroeconomics. We, instead, strongly believe that this is a legitimate and useful way to approach the history of macroeconomics.

25We think that historians of economics can and should also speak with a critical voice. Not because we want to be ‘partisan’, but because historical reconstruction should also aim at proposing, by attentive judgement, a critical attitude in terms of method and focus, and what each theory or author achieves or may achieve [15]. We tried to have this same critical attitude as much for Keynes as for Friedman or Lucas or others. We do not believe that the history of macroeconomics should be an empyrean discipline which formulates value-free judgements. De Vroey argues that the history of macroeconomics or economics should be like the history of art, but art historians, in their aesthetic evaluations, are far more partisan and conflictual than De Vroey seems to realize.

2. Some more specific issues

26In considering the commentators’ more specific considerations and observations, here we concentrate on a few of them which are more strictly related to the general issues discussed above.

27In this respect, our attention to Gurley’s and Shaw’s contribution and the influence they exerted on others is due to the fact that they criticized what Tobin, in the early 1960s, called the Keynes-Patinkin model and they look at financial intermediaries as agents actively promoting intermediation in imperfectly competitive markets in a different way from the Walrasian auctioneer, or any simplified demand and supply of money. In their book (1960) this approach applies to banks and flexible credit, but it refers also to the emergence and specialization of innovative financial intermediaries [16].

28Our consideration of Gurley and Shaw relates also to the problem of the interpretation of the IS-LM model, raised by Rubin as well as De Vroey. We agree with Gurley’s and Shaw’s criticism of the model because it collapses the whole banking and financial sector into the two equations for the money demand and supply. We are well aware that post-war macroeconomics produced several IS-LM models of differing origins. As regards the filiation of the model either from Marshall or from Walras, we did not enter into this debate; to answer the criticism that we committed the mortal sin of attributing a Walrasian filiation to the IS-LM model, we have a more nuanced and richer interpretation of the connections of neoclassical economics to the Walrasian heritage than the one proposed by De Vroey, and a more nuanced and richer interpretation of Walras himself [17].

29Hicks, the author of the first complete IS-LM mathematical model, is a complex scholar; he was not a pure Marshallian pupil. He closed his own IS-LM article in 1937 by pointing out that his simplified scheme was not satisfactory dynamics (Hicks, 1937, p. 158–159). In Value and Capital, he deemed the Marshallian approach inadequate to deal with equilibrium in time (Hicks, 1939[1965], p. 122–123). Hicks was deeply influenced by Pareto and the Swedish scholars. We could not deal with all this in the book, but we discuss the controversial nature of what general equilibrium meant in the 1930s and in the 1940s, and even in the 1950s before the publication of Debreu’s Theory of Value (1959). And we certainly want to stress the centrality of Patinkin’s effort in Money, Interest and Prices (1956) in the construction of macroeconomics in the early post-war years. Lucas himself, as we recall, recognizes the imprinting of Patinkin’s research program on his own formative years (Lucas, 2004 [2013]). In his book, Patinkin addresses both the Walrasian heritage and its compatibility with the quantity theory of money and various processes of adjustment that are explicitly dealt with in many passages of the book, notably in the disequilibrium model of chapter 13.

30In our argumentation concerning the macroeconomists’ attitude to banks and finance from the 1960s to the 1980s, we tried to carefully qualify our judgements regarding specific currents and authors and certainly we did not cover all the literature that might be considered. Rubin criticizes us for not having considered how in those years a large number of macroeconomists dealt with banks and financial markets, especially in their contributions to the building of macro-econometric models. In particular, he points to our criticism of Modigliani. We agree with his invitation to pay more attention to the importance of the macro-econometric work in the evolution of macro theories. Rubin is right to argue that our book gave this aspect less attention than it deserves. Further studies and a more detailed picture of the debates in 1960s and in the 1970s are certainly welcome. Nevertheless, it must be acknowledged that Modigliani as well as other macroeconomists of the time never rejected or questioned the Modigliani-Miller theorem, which is a major reason why the analysis of banks and finance could not be developed satisfactorily by the neoclassical synthesis (see, e.g., Bernanke et al., 1999). We confess to our preference for the more elaborate Tobin’s theoretical project.

31Tobin, like other major economists, was a complex personality, not easy to classify under simple labels. We, however, maintain our definition of Tobin as a heterodox Keynesian. One cannot overlook his straightforward criticism of the rational expectations revolution, his elaboration, jointly with Brainard, of the ‘q theory’ with its roots in chapter 12 of The General Theory, his outspoken dismissal of the auctioneer, his Keynesian exploration of the debt-deflation process, his effort at addressing the complexities of capital markets and the time structure of macroeconomic models. These are aspects of Tobin’s rich personality, which we have taken into account in our book, although obviously in a summary way. Dimand correctly points out the various influences to which Tobin was subjected in his formative years. We believe that such influences are still visible in the texture of his more mature contributions, but with a dominant attention to the possibilities of disequilibrium paths or slow processes of convergence to equilibrium, in capital markets as well as in other markets.

32In our historical perspective, the efforts to include credit rationing or the financial sector in macro-econometric models in the 1960s and 1970s are remarkable also for their failure to succeed in changing the basic theoretical macroeconomic construct and for their vanishing later on. It is a story of both high aims and ambiguous or non-achieved results. It is a story that we did not tell thoroughly, but we suggested that it should be narrated by taking account of its complexity, difficulties and ambiguities.

33As for the Marshallian currents in macroeconomics, we suggest that there are various ‘schools’ or trends in ‘Marshallian’ macroeconomics to be more precisely qualified and put in a historical context. We mention the disequilibrium literature and we refer to the excellent book by Backhouse and Boianovsky (2013), but we did not enter into a detailed discussion of this literature as it never addressed disequilibrium processes in the financial side of the economy. Dynamic disequilibrium in the financial side was mainly explored following Tobin’s suggestion to incorporate a debt-deflation sequence into some Keynesian models, an effort that we explicitly mention but that developed into a strand of literature confined to a quite restricted niche.

34As for the Marshallian vs the Walrasian approach to macroeconomics, we argue that the main currents of the neoclassical synthesis did not abandon the unhappy mix of a Marshallian short term and quite a neoclassical long term. The long term was basically built on rational maximizing assumptions, and it was interpreted either as a neoclassical growth model or as a Neo-Walrasian system of equations under conditions of fully flexible prices in competitive markets. This is the reason why, in the end, only long-term perfect price flexibility should clear the markets and restore full equilibrium. There was a loose, or confused, Marshallian consideration of the time structure of macroeconomic models and very little attention to the differences between the time structure of Arrow-Debreu models and the highly controversial Marshallian distinction of long and short term [18]. These ambiguities and difficulties prevented the emerging of a coherent Marshallian macroeconomics, and many of its frail constructions collapsed under Lucas’s attack.

35In this respect, it is necessary to address De Vroey’s criticism that we did not sufficiently and clearly differentiate Lucas from Friedman. As a matter of fact, in our reconstruction we detailed some features of Friedman’s methodology, and we do not think that the readers of our book will be misled in believing that Friedman and Lucas are the same because they are dealt with in the same chapter [19]. We quote a passage where Friedman underlines some similarity between his methodology and Keynes’s methodology. The polemical stance of Friedman versus the Keynesians is not our invention, and it is not a ‘bluff’. We attentively deal with Friedman’s and Schwartz’s interpretation of business cycles as the effect of shocks to the quantity of money. However Marshallian may Friedman’s methodology be, his interpretation of macroeconomic instability is not the same as the standard Keynesian interpretation or Keynes’s original view in The General Theory. In the same theoretical line, we believe that Lucas started from a ‘monetarist’ perspective (the vindication of the quantity theory and monetary surprises) to move later on to the RBC perspective [20].

36Finally, we do not agree with the criticism that we downplayed the credit rationing and asymmetric information literature since the 1980s. In the book we did not carry out a comprehensive survey of the literature on these topics, but we devoted attention to Stiglitz’s, Bernanke’s and others’ contributions. However, we also point out the extreme difficulty to keep together asymmetric information, perfect competition and the retention of the Arrow-Debreu model as a sort of benchmark. We have developed our views in a recent paper (Ingrao and Sardoni, 2020).

37We believe that the asymmetric information ‘revolution’ is indeed a major theoretical turn, even though its macroeconomic implications have not yet been fully appreciated. Far from dismissing the asymmetric information literature of the 1980s and 1990s, we believe that its impact and challenges have not been properly evaluated as regards the necessity that it imposes to overturn the easy escape into a short vs long term macroeconomics or the recourse to a Walrasian benchmark.

38We certainly acknowledge the efforts, notably after the global financial crisis, to develop macroeconomic models that take into due account credit and financial markets. We argue however that these efforts are still plagued by their failure to break from the Neo-Walrasian tradition. The Arrow-Debreu theoretical construct, although important and useful in certain analytical contexts (an aspect that we do not deal with in our book) is not the adequate framework to deal with the problems of monetary economies subject to recurrent disequilibrium phenomena. Something that also prominent New Keynesian economists begin to acknowledge (see, for example, Galí, 2018).

39In conclusion, as our commentators observe, our treatment of the topic of bank and finance in modern macroeconomics, not to speak of the history of macroeconomics in general, could have certainly been more thorough, and more authors and specific issues could have been considered. This however would have required much more time (years) and the delay of the publication of our work. Between the two alternatives of being more thorough and publishing the book as soon as possible, we chose the second as we believe that it is more important to stimulate the current debate rather than to try to write a book à la Schumpeter.