Economic activity is by nature directed towards the future and, for a long time, both economists and economic actors have been building models to predict the future and guide their action. The recurrent failure of these models to accurately predict the course of the economy points to an urgent need to change perspective and attempt to understand how, in practice, “economic actors visualize the future and decide how to act in conditions of radical uncertainty” [vii]. In other words, the challenge is to understand the social construction of expectations. In Uncertain Futures. Imaginaries, Narratives, and Calculation in the Economy, Jens Beckert and Richard Bronk provide a collection of 13 case-studies that are focused on economic action oriented towards the future. According to them, the central question addressed in this volume is: “if the expectations of those operating in innovative markets cannot be based on the rational calculation of probabilities based on past data, how do they form the expectations and beliefs on which their consequential decisions depend?” [9]. Both editors are now recognized for their important theoretical contribution to the role of imagination in economics.^{Footnote 1} This volume provides an opportunity to demonstrate the role of imaginaries, narratives and calculation in capitalist economies through a variety of empirical case studies.

Uncertain Futures is organized around a rich introduction that puts in perspective five sections dealing with the nature of expectations, economic forecasting, central banking, finance, and innovative businesses. The authors—eleven men and four women—come from a range of disciplines including economic sociology, political economy, anthropology and psychology. The richness of the book stems from the diversity of the perspectives mobilized and the multiplicity of its entry points—the scientific perspective of the authors, methodology and rule of evidence, theoretical questions, areas of inquiry.

As Jens Beckert puts it, actors need to rely on social constructions such as traditions, norms, institutions, social networks, organizational structures and narratives when making decisions [Beckert 2016]. In this volume, contributors insist mainly on the role of social networks and calculation devices in the elaboration of narratives. The main results concerning the making of narratives and imaginaries that guide economic actors in the fields of finance, economics and business, which are the most studied in the book, are presented in this review.

A first approach to the understanding of decision-making under radical uncertainty is the micro approach of psychology. In chapter 3, David Tuckett combines psychoanalysis, sociology, psychology and neuroscience to analyze the cognitive and affective behavior of money managers when they anticipate gain and loss. His “Conviction Narrative Theory” aims at microfounding the macroeconomic theory of Keynes’ animal spirits.^{Footnote 2} Several other contributors adopt a more interactionist approach. For them, expectations are produced by the interaction between a multitude of individuals with different points of view. For example, Douglas R. Holmes (chapter 8) shows the transformation of the relationship between central bankers and members of the public and market participants. According to him, this relationship has switched from a one-way talk by central bankers, aimed at imposing their vision of the future, to a conversation with the public that co-constructs expectations. Likewise, Werner Reichmann (chapter 5) analyses the embeddedness of economic forecasting. Economic forecasters in German-speaking countries are in relation with fellow forecasters within their institutes, academic economists, policy-makers and business representatives. In Reichmann’s view, this embeddedness has a triple positive impact on the quality of forecasts: interactions bring to light emerging developments that a single person would have missed; they reinforce the social legitimacy of forecasts; and they increase their quality since “they are based on better information and more diverse perspective” [114]. The recognition of the social nature of expectation formation has partially been internalized by economics. In chapter 7, Andrew G. Haldane depicts the transition from the Dynamic Stochastic General Equilibrium (DSGE), the mainstream macroeconomic model used to model uncertain futures, to simulations using Agent-Based Models (ABMs). ABMs are “interconnected systems of individual ‘agents’ who follow well-defined behavioral rules of thumb” [150]. They rely on heterogeneous and interactive agents rather than on a single representative agent, and these models are able to predict non-stationary or multiple equilibria with complex feedback loops. ABMs are supposed to provide a better description of reality in situations of economic stress. The author shows this by counter-factually testing them on data concerning the UK housing market and financial markets. However, the predictive capacity of such models is still under discussion.

The works previously mentioned insist on the collective character of expectation formation. This recognition is meant to better explain how economic actors deal with uncertain futures. But actors participating in the construction of expectations are not equal and there is room for a “politics of expectations” [Beckert 2016: 65]. In chapter 4, Jenny Andersson, reminds us that the content of dominant narratives is produced by an imbalance of power between actors who have their own interest in the prevalence of a specific scenario and its realization. Studying the struggle for the dominant image of the Arctic future (when de-icing occurs), she shows that “images are often dominant not because of their coherence, but rather because of the geopolitical or economic power of the actors producing them” [86]. She also stresses that actors rely on various devices—forecasts, nation branding, historical memory, research—in order to build convincing “fictional expectations” [Beckert 2016].

The importance of calculation devices in supporting the imaginaries, narratives and calculations of economic actors is to be found in many chapters. According to Beckert and Bronk, “a key function of models and other calculative devices is to help economic actors diagnose newly emerging patterns, as well as persistent regularities” [14] even if “perhaps the most important role [they play] is the social one of justifying and legitimating action despite uncertainty about the future” [18]. The contributors explore a great variety of devices by which actors cope with uncertain futures: risk management techniques, finance models, discounted cash-flow models, central bank forward guidance, economic forecasts, business plans, and new era stories as tools to calculate and/or imagine the economic future. For example, Martin Giraudeau (chapter 12) examines venture project evaluation at the American Research and Development Corporation (ARD) between 1946 and 1973. He shows that George F. Doriot, ARD’s managing director, used “mixed methods” for the appraisal of projects: the gathering of “hard” information and technical data on venture proposals in order to conduct a rational analysis on the one hand, and the development of “feelings” about the entrepreneurs in order to sustain the appraiser’s imagination on the other. Judgment and imagination appear to be necessary complements to instruments and calculation.

Instead of looking at the way calculation devices are built, some contributors focus on the effects of these devices. Natalia Besedovsky (chapter 11) uses the notion of “epistemic culture”^{Footnote 3} to show, through the example of credit rating agencies, that calculative practices encapsulate risk conceptions. While the holistic/diagnostic conception of sovereign rating aims to measure the shock absorbing capacity of countries, the technical conception of structured finance rating aims to measure opportunities for profit. Besedovsky goes further by adding that “calculative practices create the objects they measure and are at the same time the concrete manifestations of the abstract concepts they represent” [242]. In this way, the author is close to Liliana Doganova’s argument that “the uncertain future is consubstantial with the instrument of valuation” [280], in particular valuation formula and software. In chapter 13, this latter analyses two case studies where Discounted Cash Flow (DCF) and Net Present Value (NPV) are the main tools used to assess investments—forestry in the mid-19th century and drug development today. In the case of forestry, the first application in 1849 of the principle of discounting future costs and revenues in order to assess present value introduced uncertainty and a new concern for the future in forest management. As a consequence, it appears that what is defined as “risk” or “uncertainty” changes over time. This point is illustrated in chapter 10 by Elena Esposito. She uses Niklas Luhmann’s distinction between “present future” and “future present”^{Footnote 4} to show how the management of uncertainty and risk in the finance sector has evolved since the 1970s with the creation of financial instruments (derivatives, CDOs, etc.) and calculative techniques (Black-Scholes formula, etc.). Given that most of the time there is no “substantial” uncertainty, calculation devices can be used strategically by actors willing to impose their vision of the future, and they can even be used in order to create uncertainty (see also chapter 4 by J. Andersson).

Finally, a last set of contributions deals with the “failure” of “grand narratives” to adequately depict the future. Narratives, imaginaries and calculation can be “wrong” ex post in two ways: either by not having adequately forecasted what happened, or by not having been performative. Robert Boyer (chapter 2) analyses the succession of “expectation regimes” in the socio-economic order during the last decades of the 20th century and tries to explain their failure. The uncertainty of contemporary capitalist systems due to the growing division of labor, globalization and complexity of financial instruments, leads to the advent of “grand narratives” (Japan number one, the New Economy, the omniscience of financial markets, the Green Economy, etc.) meant to coordinate the behavior of heterogeneous economic actors. According to Boyer, the limited lifespan of these narratives is due to their failure to capture key dynamics leading to crisis. Timur Ergen and Benjamin Braun’s chapters echo these results. The former questions the pros and cons of the alignment of expectations. He approaches this issue from the perspective of innovation by studying the case of US energy technology policies in the 1970s and 1980s. Economic actors faced a trade-off between, on the one hand, openness and cognitive dissonance that provide adaptation benefits but also dangers due to possible stagnation. On the other hand, cohesion would bring coordination benefits but also risks due to possible premature lock-in “that would leave parts of the industry seriously wrong-footed when the technology changed due to further innovation” [31]. For his part, Benjamin Braun (chapter 9) analyses the failure in performativity of central bank policies due to the decline in their “epistemic authority”^{Footnote 5} after the financial crisis. Chapter 6, by Olivier Pilmis, also deals with “failure” of the narratives meant to reduce uncertainty, but from the perspective of people building such narratives. This chapter explains how forecasters deal with “errors”. The interviews with economic forecasters show that, quite like magicians facing failure^{Footnote 6}, they use three kinds of justifications: first, there is an “ontological indeterminacy of economies” [131] that prevents anyone from predicting innovation and shocks; second, the narrative scenario exhibiting a set of causal mechanisms matters more than the actual calculated figures; third, the process—following professional routines—matters more than the results. Consequently, the existence of forecasts is not only justified by their accuracy, their coordinating effects or the existence of an industry owing them its very existence.

By making the reader enter the complex worlds of uncertainty management in the economy, Uncertain Futures demonstrates the richness of the contemporary field of research about “futures” in the social sciences. The volume mainly focuses on institutions (business associations, central banks, credit rating agencies, forecast institutes, etc.) and professionals (economists, asset managers, bankers, etc.) in the finance and economics domains. An extension of this perspective would involve investigating more ordinary practices of uncertainty management and developing a comparison between uncertainty management in the economy and in other fields (climate change, demography, etc.). This book offers great insights for any social scientist interested in future management and social action.