Abstract

Narratives have drawn increasing attention from economists and from historians and philosophers of science. Yet little of that attention has made it into the history of economics itself. This essay reviews some of the salient literature on economic narratives and introduces key themes from a 2021 workshop intended to bring that analysis to bear within the history of economics. Four important, but little noticed, functions of narratives emerging from that workshop are highlighted: exploration, explanation, closure, and reopening; and promising areas for future research on the multiple roles of narrative in the history of economic practice are suggested.

1. Introduction

Economics has recently become self-conscious about its use of narrative—one might almost say it has undergone a “narrative turn”—but much of that interest has been in territories other than the pages of HOPE. For historians, economists of the past may have been more or less wedded to the narrative genre of expression, but we have rarely seen that genre as needing its own history in the way the switch into mathematical expression—its timing, its roots, and its effects—has drawn attention. We convened this special issue to think seriously about the role of narrative in past economics. Here we take narrative as a specific genre of account, namely, one that orders and connects various elements together to make sense of their relations (and not merely a textual mode of expression). As it happens, our collection of articles is determinedly twentieth century in focus, and that in itself generates its own puzzle: How did economists use narrative, and bend it to their various useful purposes, during a century when mathematics and statistics became central to economic science? And what about earlier periods? Surely past economists' commitment to narrative needs attention too, and that history remains to be written. But, let us look first at this recent narrative turn, and at the different versions of “narrative economics” now in the marketplace, to see how far they may speak to these twentieth-century examples of the narrative genre in the history of economics.

2. Narrative Economics So Far

2.1. McCloskey's Rhetorical Tetrad in Economics

An important, groundbreaking path was set by Deirdre McCloskey, whose determined assault on the writing of economics in The Narrative of Economic Expertise (the subtitle of her 1990 book) created a vision for what should constitute good economics.1 McCloskey (1990a), trying her best to school economists in rhetoric, laid out the tetrad of “true facts,” “true logic,” “true stories,” and “true metaphors,” from whence “the combination yields truth for science and wisdom for policy” (4). In her vision, the four elements of this tetrad each need to be allowed to discipline the others; they provide the critical partners that keep the other elements up to the task of doing good science. In the case of economics, she aligned models with metaphors, and stories with “stylized facts” or “histories,” suggesting that modernist economics prefers to work with “high standards of logic, low standards of fact, and no explicit standards of metaphor and story” (23).

While the ways in which this “disciplining,” or correcting/constraining, goes on within the tetrad were illustrated in McCloskey by examples taken from economics, it was not evident that there were any general processes in her account by which all four elements correct each other or work together. Indeed, if we consider the tetrad as containing a box of tools for doing economics, it is not so obvious that two of those tools, logic and facts, are separately constituted from the others: namely, models and histories. While models may have qualities of being metaphors (and may even be formed from analogies), economists surely understand them as also relying on logic in their makeup and usage. And while narratives may well contain facts, or particularities with fact-like qualities (as in histories), narratives may also align with non-fact-like theorizing.

As McCloskey (1990a: 10) notes, “Metaphors and stories, models and histories, subject to the discipline of facts and logic, are the two ways of answering ‘why,’” and more, the “metaphorical and narrative explanations answer to each other” (McCloskey 1990b: 5). So, within this account is the notion that narratives provide explanations (answers to “why” questions), and so do models; and her examples (most often from economic history, her specialist field within economics) sometimes see them as working in contradiction to each other, other times perhaps together. One version of such practical interconnections is found under the label “analytic narratives” (Bates et al. 1998), a recent self-conscious practice of narrative and models used in interconnecting and complementary modes in economic history (and now political science). In this approach, the economic historian, following a question, answers by “applying” or using a model to account for a set of historical events and figuring out whether the explanatory narrative told using the help of the model is complete or if another model needs to be brought in to cover the historical elements not explained by the first round, and so forth. For example, Bates's (1998) explanation of the activities and then collapse of the International Coffee Organization, a cartel of the late nineteenth through twentieth centuries, progressed through a sequence that builds an analytic narrative explanation of the historical events:

  1. Establish a historical start point (a description of the organization and context), then . . . 

  2. Apply a model from game theory (itself labeled with a metaphor plus narrative: the “chain store paradox”), then . . . 

  3. Return to the historical evidence to see what is not yet explained, then . . . 

  4. Apply Chicago regulation theory, then . . . 

  5. Return to remaining leftover historical questions, then . . . 

  6. Apply cartel rivalry model, then . . . 

  7. Etc.

In this example, the constraints of models on history and history on models are sufficiently tight that the inference gap (i.e., what is left in the history that remained “unexplained” by applying the models) is sequentially reduced.

Another part of the tetrad that involves facts and narratives in economic and political history is the world of counterfactual narratives. In Fogel's (1964) counterfactual that imagined no railroads had been built in the mid-nineteenth-century United States, the alternative facts are very tightly controlled and mapped together into a viable narrative: canal building and river widening were geographically imagined to create a narrative of the development of an alternative transport system. The close attention not just to facts and narrative but to their tight fit together provides an exemplar for the recipe for plausibility provided by Hawthorn (1991). In other counterfactuals, the space between the world we have lived through and the alternatives that might have been leads to implausible narratives. The counterfacts (of the imagined world) yawn so much wider from the actual world, and without paying attention to the implications for the narrative account of what happens on the way and its final outcome, that the tetrad controls break down. We see this yawning gap in some of the extensive counterfactuals that rewrite the geographical history of industrialization, for example, in Unmaking the West (Tetlock, Lebow, and Parker 2006).

Another set of interconnections within the tetrad were opened by Morgan (2001, 2012), who took models to be constructed as logical kinds of objects whose usage involved narratives. In using (rather than constructing) models, economists developed “exploratory narratives,” raising questions of the form “What happens if?” and answering them by generating narratives about what would happen in “the world of the model.”2 Thus models and narratives are directly related in usage, the form of the model constrains the narratives that can be told, and because the narrative is generated by the logical elements of the model, it must therefore be consistent with the model. But such narratives are not necessarily constrained by the facts of history or are not necessarily plausible in the hypothetical world of theory or in the real world. This is in contrast to the account from McCloskey, for whom stories can be supportive or not of the other elements in the tetrad and for whom facts are most usually the particular facts of histories.

2.2. Conviction Narratives and the Public Domain

In the work of literary theorists, narrative is often an umbrella term for all kinds of stylistic elements of writing, and its extensiveness parallels McCloskey's rhetorical account. In other branches of narrative theory or “narratology,” it is taken more narrowly to mean a particular genre of writing with a specific narrative structure, as found in novels (or nonfictional stories such as biographies or medical cases). This latter genre notion from narratology is the starting point for the definition that Morgan has used for discussions of narratives in science, namely, using the interrelations between a series of elements or events to produce an integrated account of the relevant phenomena (see Morgan 2017 and Morgan, Hajek, and Berry 2022).

The former, more generous, notion of narrative features in Robert Shiller's (2019) book Narrative Economics. At the beginning, he tries to have it both ways, narrow and broad, for he claims that a narrative (quoting the OED) is “a story or representation used to give an explanatory or justificatory account of a society, period, etc.,” but it can also be “a song, joke, theory, explanation or plan that has emotional resonance” (xviii). And in fact, some of his examples seem to be headlines, or threats, or injunctions rather than accounts with a clearly narrative structure. His book takes a very different focus from McCloskey's, namely, on the public usage of economic accounts, for which he states early on that “we know that the fluctuations and differences in economies are substantially driven by swirls of multiple narrative epidemics” (x). His “key proposition . . . [is] that economic fluctuations are substantially driven by contagion of oversimplified and easily transmitted variants of economic narratives” (26), where those narratives have come from economics but are also subject to random mutation in usage and transfer.

Providing historical evidence for this account is challenging in at least three dimensions. One problem is that it is rather hard to produce the facts about the channels by which these narratives turn from academic into popular usage (it is not even clear if they all start in academic work). Another problem is how to get a good handle on the process by which these stories mutate; perhaps the literature on how rumor and gossip change as they spread through a group might have been interesting comparators to the nature of those random mutations posited by Shiller. The most important lacuna lies in the gap between these narratives and people acting on them: how, why/why not, and when those actions are prompted seems an obviously important and interesting part of his agenda. It cannot all just be sentiment, for to fulfill his thesis, these narratives have to be performative, as Giraud (2021) points out in his considerable review of Shiller's book.

Shiller's thesis about the public viral life of economic narratives initially seems a long way from our agenda of understanding how economists of the recent past have used narratives in their own work—and our questions about the contexts, appearances, and purposes of such usage. But there is an interesting connection in the parallel and overlapping notion of “conviction narratives,” with its literature on how people come to believe something enough to act on it in a radically uncertain world (see Tuckett and Nikolic 2017). The theoretical backgrounds to these ideas come from narrative theory and psychology, and the label provides a space in which commentators from both sides discuss how economic ideas and economic actions are linked via beliefs and by narratives. This is not traditional rational choice theory, but rather a look at how individuals theorize their understanding of situations and associated actions, as well as how actions are motivated in real economic and financial markets. In this framework, narratives are seen/found as devices for decision-making in conditions of radical uncertainty; indeed, narrative may form the integrating device that brings economic beliefs into direct actions. Thus, Tuckett and Nikolic (2017: 502) write, “Conviction narratives enable actors to draw on their beliefs, causal models, and rules of thumb to identify opportunities worth acting on, to simulate the future outcomes of their actions and to feel sufficiently convinced to act. The framework focuses on how narratives and emotion combine to allow actors to deliberate and to select actions they think will produce the outcome they desire.”

Tuckett, whose professional credentials span psychoanalysis and finance, draws upon an earlier interview study of fifty-two asset managers in 2007 and 2011 (i.e., before and after the great crash of 2008–9) to develop his analysis. He describes conviction narratives as accounts that enable economic actors to “make meaningful sense of situations, which means to identify opportunities for action based on implicit causal explanations attached to observations”; to “simulate alternative representations of the future outcomes of actions”; and to “communicate” and “articulate and support” their actions (Tuckett and Nikolic 2017: 506). His more recent ethnographic/field study (Tuckett et al. 2020), gathering “narrative evidence” of the information used by central banks from their own field experts,3 discusses how such banks use that evidence in both framing narrative(s) for their own decision-making and in making suggestions for their communications that might follow this performative action agenda. His account fills in for us how expert knowledge of what is currently known, and what might happen in the future, can be used in expert-based narratives that are (or might be) then relayed to the market as narratives on which agents can be convinced to act. Paul Collier, a development economist, has also recently become interested in conviction narratives, considering contexts in which governments, national or local, are encouraged to act positively by creating narratives that enable them to believe that change is possible. They cannot act to improve their neighborhood, overcome their development problem, and so forth, until they believe in a narrative that sets out that possibility in a course of action that they can undertake (see Collier and Tuckett 2021).

There is an interesting intersection here with Storylistening: Narrative Evidence and Public Reasoning (2021), an account from Sarah Dillon and Claire Craig on how careful attention to the science narratives used in the public domain can help explain how scientific research, and especially science-based policy actions, are reasoned about in that domain. They see narratives as key to that communication space, as perhaps we have seen in public health reasoning throughout the COVID pandemic. Theirs is a very different agenda than Shiller's, for while his attention is on the narratives that circulate in the public, their attention focuses on the scientific narratives that might circulate in the public domains. We might clarify that distinction by contrasting the information that medical experts told the public about COVID, which circulated within the public, against those non-medical-expert-based accounts that also circulated within the public. Of course, for the public, these might not be clearly distinguishable by source. Dillon and Craig's agenda is to bring the strong resources of narrative into play as the means of communication that should be harnessed in that expert communication space.

2.3. Agency and Narratives

Marina Bianchi and Roberta Patalano, in Storytelling and Choice (2017), draw on narrative theorists (particularly of the 1980s) as well as developmental psychology (rather than on cognitive psychology as Tuckett), and while they aim at some of the same targets as Tuckett et al., their focus is on microeconomists' theoretical interests rather than the domains of economic action. For Bianchi and Patalano, the problem for economists is, given that people do not think and behave according to simple rational choice theory, how do they think and behave? In answering this, they focus on the developmental psychology of how people learn to think about decisions and associated actions, along with the importance of narrative in that process. This leads them into discussion of the imaginative elements of narrative and the ability of narratives to provide a frame for how economic agents think about decisions and so move to actions. In their analysis, narrative brings together what is known (“as is”) with what might be possible (“as if”), making narrative a reasoning device between the two and so constitutive of the links between choices and actions. On the economic science side, they suggest this account of agents' reasoning combines logical-scientific or “paradigmatic” dimensions with “ordering experience” in “constructing reality” (Bianchi and Patalano 2017: sec. 4.2). The observation that more is involved in economic choices than simple rationality reflects the emotional commitment that agents make, a commitment generated by agents' narratives and picked up in economists' use of those narrative accounts in their own theoretical explanations.

While Bianchi and Patalano are concerned with the importance of narratives for the agents' behavior envisaged in economists' theories and models, Bruna Ingrao's (2018) seminal essay transfers this interest onto how economists use narratives in their own work. So, these three authors share an interest in cognition and the importance of narrative in cognitive processes and reasoning, but for Ingrao, this is also relevant for the narrative practices of the economist, where narratives appear in conjunction with mathematical modeling as part of a three-step cognitive, and community, process. First, the mathematical skeleton of the model is labeled with names and meaning to create a fictional economic world, and then, second, there happens “narrative arguing” with the model (building on Morgan 2001, 2012), which she portrays as the conceptual framing of the social interactions involved. Finally, “the fictional characters interacting in the model's scenarios may be transfigured into situations related to actors and events in real economies” (Ingrao 2018: 114), that is, narrative is a potential resource in bridging the inference gap between model and world. Thus, Ingrao sees narrative as having a cognitive value or function for the economists themselves in their theorizing/modeling (and perhaps in applications), not in perceiving the cognitive role of narrative for the agents that economists are modeling (as for Bianchi and Patalano), nor to give a cognitive account of narrative for agents acting in the world (as for Tuckett in his work).

But, as Ingrao (2018: 115) recognizes, “if the stories told on models are a crucial component in the cognitive activity of building theories, their status is highly controversial.” And she remains pessimistic that there is any recognizable, and valid, inference bridge between these fictional narratives and the real world of events and policy discussions (123). Instead, she points to another practice, the role of the “learned narrative prose” that is typically not recognized as “‘scientific’ discourse” in economics, yet is not only “crucial for bridging the gap between the imaginary worlds in theoretical economics and the reality of economics events” but also plays a critical role in “conceptualization,” which requires “coherence in specialized language” beyond the “mathematical objects” of economists' theory building (126–28). In this final element, we can perhaps see some alignment with McCloskey's tetrad argument, namely, that the four rhetorical elements of stories, facts, logic, and metaphors (models) constrain each other. But in Ingrao we have insights into the very close relations (and their cognitive function) between models/logic and (fictional) narratives in theorizing; and into (for her) the unassailable gap between those models/logic and real-world narratives; and into a third and hardly noticed, but potentially important, separate relation she remarks between prose narrative and concept-making.

Ingrao's account of the multiple roles of narratives in economics is consistent with a more general argument made by Morgan (2022), who presents narrative-making as having an umbrella role as a sense-making technology for science, a “general purpose technology” that scientists turn to when they want to make sense of a variety of elements/events. As such, it can be found in a variety of empirical, theorizing, conceptualizing sites, including modeling, simulation, and experimental and interpretative analytical tasks.4 This is not a claim that narrative is in all sciences all the time, but rather that the reach of narrative-making in scientific practice is surprisingly wide. And once narrative construction is done, then the narrative forms the representation of the knowledge claims of that site/science, thence to be used in reasoning and inferential domains.

While McCloskey had a tetrad of elements and liked to present these as something like science versus literature, each side of which had to discipline the other in creating knowledge, Bianchi and Patalano pick up several other elements that point us into the internal theorizing uses of narratives in economics. One is the factual versus fictional alternatives, the other is the “as is” versus the “as if” line. It seems easy to present these as dichotomies, but in fact, there are other ways of understanding these narrative possibilities. Our experience of narratives in the history of economics, and drawing on the articles in this issue and the workshop that preceded it, suggests that these fictional/factual alternatives and as-is/as-if lines are better conceived as good companions than alternatives, just as we have already found to be so for elements of the tetrad. The use of models to generate narratives in simulation studies, for example, may generate as-if narratives for the model in a spare construction depicting something about some bit of the world. When put to work, uses of the model may generate narratives that are quite believable when using some parameter values, believable enough to think about some real economy in some real historical period, but the model may also generate other narratives that are quite ridiculous when using other parameter values. This is exactly what happened in the first, hand-cranked, model simulation in economics, when Samuelson tried out his little Keynesian multiplier-accelerator model in the late 1930s. His simulations from adding units of “government spending” generated exploding economies for some parameter values, almost no change for others, and different sized cycles for others (see Morgan 2012).

Bianchi and Patalano also pick up another element that we find in our articles and in our experience, namely, that there is a hazy line between particular and general accounts. While McCloskey rightly gives a good deal of space to the factual side of histories in narrative, that level of particularity is not necessarily a feature of narratives in economists' usages, which may be about particulars, or may be general claims about the way the world works, but may be, and perhaps more often are for economists, at some level in between. Our narratives are often about kinds of situations, kinds of actions, kinds of processes, kinds of markets, kinds of consumption behavior (see also Morgan 2017). One only has to recall Marshall's supply and demand diagrams in his text: once he moved from the simple, base-case, crossed-curve diagram, his footnotes explored other kinds of markets, showing their associated shaped curves, telling little narratives about the different kinds of processes of adaptation that would be found when price changed (for example) in each kind of case. Narratives seem to be the medium that enables economists to slide between, or perhaps glue together, particular and general levels, theories and evidence, present and future, explanations and predictions, and so forth.

And, of course, as this Marshall example makes clear, narratives are not just found in texts in economics of the twentieth century; they are found in and with diagrams, graphs, models, and pieces of mathematics. To return to our opening paragraph, narrative is not one mode of expression any more than mathematics is. Mathematical expression opens up a set of powerful modes of representing and reasoning—there is not one mathematics but many kinds, fulfilling many purposes. Narrative expression does the same, with the same variety.

3. Narrative Economics Continues

We convened this special issue to introduce the HOPE audience to the analysis of narratives in economics and to push that conversation further. In reflecting on existing literature and the contributions to our special issue, we found it helpful to focus on the functions that narratives can serve. Broadly speaking, we have organized these into two dyads: exploration/explanation and closure/reopening. Exploration refers to how models, theories, conceptual relationships, and so forth can generate new narratives that reveal the possibilities and characteristics of their underlying source. Explanation runs the other way around: economists start with a sequence of events and add elements (models, theories, concepts, etc.) to generate a new narrative that renders the original events intelligible in novel ways. Closure refers to the sense of completeness one has when questions posed within a narrative of events (e.g., “Why did B follow A?”) have been satisfactorily answered. Here we notice particularly how shorter narratives (“narrative chunks,” to borrow a phrase from Tuckett and Nikolic [2017: 505]) become valuable tools or components in bringing closure to a larger narrative. Finally, reopening, as the name suggests, considers how alternative narratives can reinvigorate questions that had previously seemed closed.

3.1. Exploration

One central function for narratives is exploration. A researcher has an idea about how this bit of the world works (an informal verbal description, or perhaps a formal model), and she asks what would happen in this bit of the world if X or Y changed. Working through an informal thought experiment or a more formal model experiment creates a narrative that reveals some of the possibilities in that account of the world. These are more than chronologies, for the researcher has already established connections between the elements of this hypothetical world, and those connections underwrite the course of the narrative. By creating these narratives, the researcher is exploring the possibilities of the underlying model or theory, supplying it with new external conditions or simply tracking it along a new path.

Exploration derives from the as-if narratives considered earlier. We can see this process most clearly in simulations, as discussed above in the example of Samuelson's work, though of course computers have massively increased the potential of these explorations compared to his hand-cranked versions. But economists can explore theoretical relationships as well, just as Marshall's exploration of supply and demand curves shows us (a pedagogical technique now standard in microeconomics textbooks). Exploration can be part of the research process (showing a researcher the consequences of her ideas), but it can also be didactic: helping colleagues or students grasp the ramifications of a theory or model. In Lucas Casonato's contribution to our special issue, he examines how Israel Kirzner used a series of simple stories about hunters to elucidate his concept of “alertness” in economic activity. Kirzner walks his readers through a series of scenarios, showing how different distributions of alertness can affect the actions of protagonists and economists' analysis of the situation.

Many narrative explorations are tightly focused, exploring variations developed from a single model or theoretical relationship (as with Marshall's supply and demand curves). But they can also be broader, involving multiple elements in combination, such as a core theory plus key assumptions, expectations, or theoretical presuppositions. In this case, the researcher is exploring the entire assemblage of elements. Macroeconomic forecasts are a perfect exemplar of such broad explorations with heterogeneous inputs. Forecasts may depend on an underlying core model, perhaps with side models for certain elements. But economists regularly supplement those models with rules of thumb, general economic intuitions, expectations about the likelihood of exogenous events and their possible effects, and so forth.5 In these cases, forecasters explore the combined consequences of a more or less rich set of varying components. In Laetitia Lenel's contribution to our special issue, she describes how researchers at the International Monetary Fund (IMF) created sets of hypothetical future scenarios for the national economies of its member states based on a range of assumptions, possible policy actions, posited theoretical relationships, and, eventually, economic models. IMF officials used these narrative explorations to identify optimal (in their view) national policies and to pressure member states to follow those recommendations.

3.2. Explanation

A second common function for narratives is explanation. Narrative explanations (to borrow a term from Morgan [2017]) are the converse of explorations. Whereas explorations start with theories, concepts, or models and use those to generate a series of events (forming the narrative), explanations begin at the other end. Researchers start with a sequence of events that they hope to make more intelligible by introducing theories, concepts, or models, enabling them to create a narrative explanation—that is, a narrative that explains the linkages between the original events. As with explorations, explanations can rely on a single model or theory for their explanatory power, or they might draw on a variegated set of components, possibly including other narratives (narrative chunks). For example, besides analyzing Kirzner's narrative explorations of hunting, Casonato also describes his narrative explanations that use the concept of “information-knowledge” to illuminate everyday stories—a mother realizing she could have made a toy that she just bought or a wedding guest startled to hear his name called over the sound system. Both of these narrative explanations involve the introduction of a single key element (information-knowledge) that sheds new light on the original tales. But narrative explanations can also depend on more complex, heterogeneous elements. Harro Maas's contribution to this issue examines Alfred Marshall's efforts to explain the course of British prices and industry in the nineteenth century, where Marshall's narratives derive their explanatory power from a much larger set of tacit and explicit ideas about human action and economic activity.

At times, a narrative explanation may be the final product of a research project. In Maas's account, for example, Marshall's primary aim was to render British economic history intelligible by constructing a satisfying narrative that provided convincing connections between its key events. In other cases, having constructed an effective narrative explanation, an economist might use that success as evidence for the value or power of a core theory or model deployed within it. For example, Milton Friedman and Anna Schwartz's (1963) famous monetary history of the United States posited causal relationships between the money supply and various other economic factors that were then used to construct a narrative explanation for key episodes in US economic history over nearly a century. In turn, however, the success of that narrative allowed Friedman and Schwartz to argue that their posited causal relationships had been validated by historical events. Their account has been more recently adopted as the “ancestor” of what Christina Romer and David Romer (1989) call the “narrative approach” to economics: using “narrative evidence” (i.e., firsthand textual accounts of relevant actors in a period and problem of interest) to help identify causal factors in econometric work.

Such reciprocity between narrative explanation and theory or model validation is a common feature of economic research, where economists aim to establish more general relationships and concepts that can subsequently be applied in other situations or where particular explanations are found to have more general relevance. For example, an explanatory account of industry exit in the steel castings industry, based on both narrative and econometric evidence, became an exemplar for other cases of exit (and so informed policymaking), but it also moved into a more general account using game theory (see Morgan 2019). In our special issue, Jeff Biddle describes how Zvi Griliches made use of narrative evidence and created narrative explanations for the adoption of hybrid corn and the increased use of fertilizer, respectively, that became exemplars for modeling the adoption of new technologies in agriculture. Or again, our contribution from Alexandra Quack and Catherine Herfeld examines how the political scientist William H. Riker used game-theoretic models to create narrative explanations of coalition formation in order to show his colleagues the potential of game theory for political science. As Quack and Herfeld explain, Riker's strategy followed a common pattern of using narratives to help in the “translation” of models to a new domain. More generally, Morgan (2017: 95) has argued that narrative explanations that have been built upon “abstract, conceptual, or theoretical social science ingredients” can enable those elements to be extracted and reapplied in new domains. The particular narrative explanation (the explanation of these specific events) validates those ingredients and extends their application. We learn how to use them effectively in a new context and thus, as Wittgenstein would have said, extend their meaning.

As Morgan (2017: 93–96) points out, any knowledge-producing activity in the sciences involves some components that can be reapplied in other contexts, but precisely what can travel from one situation to another depends on the field in question or even the particular account. In economics, scholars have often moved from a particular account (this narrative) to a more generic core (model, theory, general law, etc.) thought to explain both these events and many others. This generic core is abstract and skeletal on its own; it must be embedded into a narrative in order for its potential and significance to be apparent. Hence the essential reciprocity of narratives (on the one hand) and theorizing or model construction (on the other). Narratives provide the living flesh to transform the bones of theory into an animate creature. Yet for economists and many other social scientists it is often the bones that draw their interest: these are typically what are extracted and transported to a new context.

Not all fields take the same approach. In anthropology or history, for example, scholars are apt to emphasize the contingency and specificity of their narratives. The value of a new account of the American Revolution typically lies in what it tells us not about revolts in general but about these particular events in this specific time and place and their consequences moving forward. What travels from one historical account to another is less a skeletal theory than new methods: new kinds of sources to interrogate, new questions to be asked, new ways of relating evidence, new perspectives to be taken. If economists are anatomists, looking through the particular to the generic, historians are more akin to artists, borrowing tools and techniques to craft a new sculpture that is both illuminating and informative of its own particular subject.

Of course, the simple binaries here are misleading; we are considering the balance of emphases, not rigid oppositions. Economists do borrow tools and techniques (regression analysis, model-building strategies), and historians rely upon and generate rules of thumb or informal theories to explain individual and social action. And of course the balance may shift from one subfield to another or one practitioner to another. Stapleford (2021) has argued that economists using big data for market design are less interested in creating generic models or theories than in developing new methods and tools for market design and analysis. Their narrative explanations aim to validate their methodology rather than the resulting model (which is highly specific to the project domain). Likewise, in our special issue, Maas argues that Marshall sought primarily to create a convincing and accurate narrative explanation of British economic history, one that relied on not a single core model or theory but a mélange of explanatory elements. What traveled, in Maas's account, was less Marshall's theoretical explanations than his general approach to economic reasoning and explanation, which Maas sees reappearing in John Maynard Keynes's preference for “mazes of logic” over strict statistical models.

In many cases, as we move from the finished product (the published research) to consider the research process itself, the oppositions between particular and generic or exploration and explanation begin to fade in favor of a more iterative process wherein economists move back and forth, bootstrapping their way to richer accounts. Faced with a particular set of events, an economist might explore a generic hypothesis, tinkering with a theoretical relationship to see what narratives result and then testing those as narrative explanations against the original events. She might then fiddle with the hypothesis or adopt a new assumption and explore the resulting terrain once more. Though at any given time we might say an economist is exploring or explaining, the research process itself almost always entails both.

3.3. Closure

We have been talking here of “convincing” narrative explanations. But what do we mean by that? In the philosophy of aesthetics, Noël Carroll (2007: 1) defines “narrative closure” as “the phenomenological feeling of finality that is generated when all the questions saliently posed by the narrative are answered.” Closure is that feeling at the end of a traditional detective story when the sleuth's narrative explanation draws together various events from the story into an overall, logical account in which each sequence makes sense, any countervailing evidence has been explained away, and irrelevant events have been shown to be unrelated. To create a convincing narrative explanation is to seek narrative closure for a set of events.

Much of scientific research can be characterized as a quest for narrative closure in this respect, and the detective story provides a useful analogy for illuminating the complexity of that task. There are, first of all, two main narratives: the detective story itself and the narrative explanation of the events linked to a crime or puzzle. These correspond to what Robert Meunier (2022) has called the “research narrative” in a scientific article (the narrative of what research was carried out) and the “narrative of nature” (the narrative of what happened at the site of that investigation). Both of these narratives must be brought to closure: if we have lingering questions about what the detective (researcher) has done, we will be less convinced by the narrative explanation, perhaps worried some relevant evidence has been overlooked or a critical step missed.

Second, reflection on the characteristics of good detective stories reveals that their explanatory power derives from a heterogeneous set of elements: artifacts, signs (the footprint in the damp earth), experiences (fragments of an overheard conversation), experimental tests (blood analysis), and many other narratives (testimonies, interviews, etc.). To link these together, the detective creates narrative chunks (perhaps the thief left his footprint on the lawn and sliced his hand while forcing the window lock) that must be convincing on their own terms and coherent with the other narrative chunks the detective has tried to stitch together into an overall explanation.

Besides forming chunks within a larger story, narratives can serve supporting roles, such as providing additional direct evidence (the maid who saw a figure creep across the lawn) or hints for further investigation (the alibi that doesn't quite check out). The detective, therefore, is constantly gathering, assembling, creating, and testing narratives, and the success of her efforts depends upon the soundness of her overall explanation and the soundness of its individual narrative chunks (supported in some cases by narrative evidence). This use of narrative chunks runs parallel to discussions about the criteria for narrative in law. The latter seeks closure by combining all evidence, elements, and chunks into a consistent overall story that can viably/plausibly attribute agency (or in our economic cases, relational or causal claims) to specific actors in legal cases.6 Our consideration of narratives in economics, therefore, must examine overall narrative explanations and explorations but also the narrative chunks that compose them, as well as how those narrative chunks can either contribute to closure or in fact raise questions that reopen established accounts.

As in detective stories, narratives can help close explanations in multiple ways, such as supplying missing information, suggesting promising targets for study, guiding the construction of economic models and justifying their features, or accounting for events that lie outside the scope of current theory. In our issue, several authors describe how collecting, building, transforming, and analyzing narratives can be part of bringing closure to a larger narrative explanation, often taking the economist back and forth between the generic and the particular as described above. In Biddle's study of Zvi Griliches's construction of a narrative explanation for the adoption of hybrid corn, for instance, he shows how Griliches began by reading histories of hybrid corn development and interviewing participants about their own actions and decision-making. Griliches then synthesized these primary narratives (i.e., narrative evidence in the sense of Romer and Romer 1989) into a general narrative about the spread of hybrid corn recast in familiar economic terms (“firms rationally pursuing profits”). In turn, Griliches used this general narrative to specify an econometric model for hybrid corn adoption that he could test with available data. Thus particular narrative evidence (histories and interviews with participants) was used to construct a general narrative chunk (a generic story about hybrid corn adoption) that could then specify an economic model to be developed and tested against historical data. The model, with its parameters tuned by historical data, told a particular narrative (the course of hybrid corn adoption in this time and place), but its structure offered a more generic tool. In a similar way, Quack and Herfeld describe how Riker used narratives to match an n-person, zero-sum game to his target system (the formation of political coalitions), to stitch together several static games into an overall dynamic account, and to build an explanatory narrative by first creating a schematic narrative of a particular historical episode framed by his generic game-theoretic model. By weaving back and forth between particular and generic accounts, narratives can thus play a variety of roles throughout the research process, all of which are intended to secure the closure of an overall narrative explanation.

In the heterogeneity of the elements these authors combine in their accounts, we can see examples of the McCloskey tetrad at work, both in how her four elements (facts, logic, models, and stories) constrain each other but more commonly in how they creatively shape each other. Griliches combines narratives from field participants with mathematical models and statistical data to generate viable models of disequilibrium adjustment processes in technological change. Riker uses narrative evidence to structure static game-theory models that are then connected via other narratives to form a dynamic model. Similar combinations appear in our other papers. In Daniel Kuehn's account of Warren Nutter's research on Soviet economic growth, we learn how Nutter used “travelers' tales” to fill in the gaps of missing Soviet data but also to understand how that data system was constituted. As McCloskey had argued, good science relies on effective rhetoric intertwining multiple forms of evidence and arguments.

Closure can also be relevant to narrative explorations, especially those intended as forecasts or possible future scenarios. Here, the task is to connect the predictions or hypothetical scenario back to the actual conditions of the world. The core questions are “What will happen next?” or “How might we get from here to there?,” and the economist achieves success insofar as her narrative exploration can bring convincing closure to these questions. Often, narrative chunks will be key in linking the present to these narrative explorations. Lenel's study of the IMF in this issue provides an excellent example. IMF staff first considered how different growth rates in member countries might affect current account balances, an exploration that created various hypothetical future scenarios. They then drew on their expertise to devise narrative chunks showing how particular policy actions might take each country from its current situation to the growth rates specified in a given scenario, thereby linking policies to scenarios. The IMF's scenario drafting occurred in an agonistic environment, however, as the staff's narrative chunks were challenged by some of its member countries, notably the United States. The staff's efforts to bring closure to their scenarios were thus frequently resisted by critics who sought to reopen the exploration, forcing the IMF to develop new strategies and techniques, including the creation of elaborate econometric models. Yet the IMF staff were never able to escape the central role of narratives in binding the econometric models to the present situation since narrative chunks provided the rationale for certain parameter values and initial values for variables.

Lenel's article comes closest to the concerns of those scholars (discussed in sec. 2.2) who have emphasized the role of narratives in persuasion and motivating action (i.e., Shiller, Tuckett and Nikolic, Dillon and Craig, and Collier). Lenel takes pains to emphasize the struggle of IMF staff to create narratives of future outcomes that will carry conviction in the public domain and so reach into policy discussion and actions; if only those governments, civil servants, central bankers, and so on would listen carefully to their IMF stories and be persuaded by them! But of course persuasion is relevant to economists beyond forecasts or predictions. Casonato suggests that Kirzner turned to narratives in part to persuade mainstream economists of the value of Austrian approaches. In the more personal space of economists developing their own conviction narrative, perhaps we can include Maas's analysis of how Marshall developed his economic history, a narrative explanation of salient events, carefully created on elaborate charts. Perhaps we can see therein an economist's equivalent of the narratives that are everywhere in the natural historical sciences, a kind of social historical science. Successful narrative closure (with its phenomenological feeling of “completeness”) carries persuasive force, whether we are convincing others or ourselves.

3.4. Reopening

The very importance of closure of course invites its opposite: critical efforts to reopen narrative explanations by exposing their inadequacies. Such was the aim of the IMF's critics in Lenel's history, who constructed their own counternarratives to destabilize the authority of IMF recommendations. Indeed, the creation of a counternarrative is one common strategy, perhaps an essential strategy, for reopening an existing narrative explanation. Edward Leamer (1983: 31) began his famous critique of applied econometrics by contrasting the story that econometricians “would like to project” of themselves as agricultural experimenters with “a more accurate one” of a farmer who notices that he gets higher yields under trees where birds are roosting but is unable to demonstrate statistically that the bird droppings are responsible for his increased success. Within our special issue, we see this strategy most directly in Ibanca Anand's discussion of Evsey Domar. Though Domar is most well known for developing the Harrod-Domar model of economic growth, Anand explores his work in comparative economics and economic history. In her account, we see Domar's emphasis on narrative ambiguity and his predilection for showing how multiple narratives can be constructed from the same evidence, underscoring his resistance to premature narrative closure. For Anand, Domar's outlook reflects a parallel resistance to common Cold War binaries about capitalism and communism, which Domar worried could hide a much less definitive dichotomy in reality.

Though counternarratives (explaining the same events in a new way) are an effective means of reopening a narrative explanation, economists may also use new narratives to cast doubts on the explanatory reach of dominant models or theories. For Casonato, the stories told by Israel Kirzner are strategic tales that highlight the limits of neoclassical theory. The richer psychological space afforded by narratives allows Kirzner to introduce concepts (such as “alertness”) that have no place in neoclassical economics but that have explanatory power in his narratives. In Kirzner's work, we see echoes of the psychological analysis of agency and narrative. On the one hand, like Bianchi and Patalano, Kirzner uses narratives to reveal the decision-making processes of agents (the characters in Kirzner's stories). At the same time, like Ingrao, Casonato shows how Kirzner himself uses narratives in his own reasoning, to illuminate Austrian theory using his simple narratives (of hunters, mothers, or wedding guests) and then to connect those narratives to analogous stories about familiar economic actors (such as corporate managers).

The ability of new narratives to “open up” novel questions also appears in Kuehn's examination of Nutter's research on Soviet economic growth. The travelers' tales that Nutter used informed his analysis of Soviet economic statistics. For Nutter, these tales reopened previous narratives about Soviet economic performance by providing new perspectives on the underlying economic data. By filling in missing background information about the production of Soviet statistics (including narrative explanations that were sent to him by Soviet statisticians about the gaps in their data), the tales facilitated a revisionist account, even though Nutter was also aware of the considerable uncertainty that remained.

Just as looking closer at the research practices of economists revealed looping cycles of exploration and explanation, we find a similar pattern with reopening and closure. New evidence or a new perspective leads an economist to reopen a previously closed narrative but only so that she can construct a new narrative closure, perhaps by tweaking or modifying the previous account. In the process of devising this new account, she may cycle through several rounds of tentative closure and reopening as she tries out novel arguments, critiques them, seeks feedback, and so forth. A successful reopening seems to demand an alternative counternarrative, which carries with it at least a partial sense of closure. Still, just as a published article might present an overall trajectory toward exploration or explanation, so too with narrative closure/reopening. Though most economists seek satisfactory closure—a convincing narrative—not all do so: Anand argues that Domar strove to emphasize the ambiguity of his historical and economic narratives. In Domar's eyes, recognizing our inability to achieve a final closure grants us a more accurate view of our knowledge and its limits.

The cycle of closure and reopening occurs on a communal level as well. Some economists may seek to close a narrative only to face resistance from their peers, just as US officials resisted the closure provided by IMF staff in Lenel's essay. Such attempted closure, resistance, and reopening may extend over time: Friedman and Schwartz's (1963) Monetary History faced initial resistance (e.g., Tobin 1965) and later counternarratives (e.g., Romer and Romer 1989) that were themselves resisted (Hoover and Perez 1994). Ultimately, narratives only truly close when a community has ceased to question them, a stability that of course may only be temporary.

3.5. Quantification, Mathematization, and Narrative in Twentieth-Century Economics

As we noted in our introduction, histories of economics in the twentieth century often focus on the mathematization of economics or the growing centrality of statistical data and their analysis. What happens when we turn our attention to narratives instead? Perhaps not surprisingly, several of our contributors focused on economists who were skeptical of formalism and overly simplistic quantified analysis. The Marshall of Maas's account is not the ancestor of Samuelson but that of Keynes and Keynes's critique of Tinbergen. Anand's Domar is likewise attentive to the limits of formal models and statistical data, and Casonato's Kirzner uses narratives to critique neoclassical information theory while emphasizing his own concept, alertness, that by definition cannot be captured by a model.

Yet other contributions show us that narratives have by no means been opposed to modeling, mathematics, or statistical work in the twentieth century. For Kuehn's Nutter, narratives are essential for properly understanding and using statistical data. In Biddle's essay, we see Griliches using narratives to construct econometric models and to connect them to the farmers and seed companies whose actions he hoped to explain. Quack and Herfeld show how narratives were essential to Riker's efforts to win a place for formal game-theory models in political science, and Lenel's examination of the IMF describes how econometric models became key supports for the staff's attempts to create persuasive narrative explorations of the global economy.

None of this should be surprising in light of the preceding analysis, of course. Economists ultimately seek to understand connected sequences of events, that is, narratives. They are thus constantly constructing narratives (explorations) or attempting to make them more intelligible (explanations). They seek convincing narratives without loose ends (closure) and are constantly probing accepted stories for weaknesses (reopening). They stitch together narrative chunks or combine them with other elements such as models or statistical data to build a larger story. In the process, they may use additional narratives as evidence or raw material to be transformed (just as Griliches interviewed farmers to construct a generic story about hybrid seed adoption). Narratives are not opposed to mathematics or statistics, nor indeed to the more recent experimental work and use of big data; rather, those modes of expression and approaches to economic work generate other tools for building and supporting narratives. Narratives lie at the heart of economic practice, as our historians show—even if they do not always make it into final publications.

4. Where Next with Narrative Economics?

Let us go back to our beginning, to wonder about the role of narrative in economics over its longer history. Our account and the articles in this collection directly, or indirectly, raise several questions for us to ask ourselves as historians of economics.

First, while we have outlined some ideas about how and where narratives fit into twentieth-century economics, we have not offered any suggestions or ideas about the history of the previous use of narratives and their modes of expression. What did the narrative elements look like in economics of past centuries? Can we map these modern characteristics of their usage—exploration, explanation, closure, and reopening—onto the potentially different discourse of narrative usage in past economics? Did narratives come in chunks that were joined together, or were they rather free flowing? And if there was radical change from a social science of words to one reliant on technologies of investigation and analysis, was that a step change or was that transition gradual? Chasing down, analyzing, and creating materials for the history of narrative in economics is one considerable, and largely open, agenda.

Second, and as part of that agenda, the recent and current literature on narrative and economics that we surveyed briefly in section 2 treats the public domain of economic narratives and the “internal” domain of economic reasoning rather differently. The two strands of Shiller and Tuckett et al., versus that of Bianchi, Patalano, and Ingrao, are focused not just on different economists' habits of discourse but on their different audiences. Of course we know that economists of the past were heavily involved in public reasoning, which raises questions about the historical relevance of separating public and scientific discourse. Did the economists of the past even see such a distinction between the public use of their work and their internal imaginings about the behavior of people, countries, companies, workers, and so forth? What were the changing shapes of economic narratives in these domains, and how did they shift as economists altered their core beliefs about economic matters?

Third, and closely related to that second point, we go back to McCloskey's agenda of rhetoric and the persuasive power of narratives, either broadly conceived as in her own work or more narrowly conceived as Ingrao's “learned narrative prose.” What makes economic narratives persuasive? Of course, persuasive means are partly matters of broader cultures of writing and argumentation, which vary between places and have changed over time, so there is surely no one answer but many histories to be traced out. Can we indeed extract the narrower learned scientific narratives from the broader rhetorical usages (Shiller's territory), and how did economists of each time do so? Just to take an example: Malthus, in writing his narrative theories of population at the end of the eighteenth century, inveighed against the overblown rhetorical strategy of contemporary utopian accounts of the future of society. He explicitly did so in order to draw the contrast to his own dry, learned, scientific narratives of the population problem, along with his pessimistic forecasts of cycles of famine and relative plenty. By a century later, for cycle theorists such as Mitchell, Kondratiev, Kuznets, and others, persuasive narratives combined learned scientific prose alongside statistical graphs, for by this time, economic science, qua science, eschewed the flourishes of public rhetoric (see Morgan 2021 on these contrasting narrative qualities). What constituted persuasive argument clearly changed from the Scholastics through the early modern period to the classical economists. Probably they all relied on narratives? But where, when, and how did their arguments rest on wider changes in rhetorical fashions that employed narratives as a key element? For example, how did the narratives associated with public policy arguments for free trade in Adam Smith differ from his use of narratives in his accounts of the relation of beliefs and imagination to sympathy, to pick up Bianchi and Patalano's agenda? Perhaps as a final challenge, we might offer Defoe's Robinson Crusoe narrative for analysis. It has been framed as the first novel in English, and it has equally been treated as a semifictionalized account of a set of historical events of the period. But for us as historians of economics, surely its most famous outing was its role in Edgeworth's account of exchange behavior, which prompted the formation of one of the most well-used and developed mathematical artifacts: the Edgeworth-Bowley box. For historians of economics, the next steps in the narrative turn may be to move beyond functional analyses of narratives to consider how they change over time and context, examining shifts in style, genre, audience, forms of evidence, and strategies for closure, to see in short what the histories of such narratives can tell us about past economic practice.

We thank the following for their engagement with the agenda of this introductory article: Deirdre McCloskey, Bruna Ingrao, Marina Bianchi, Jason Leonard, Steve Medema, David Tuckett, and Paul Collier; the many other scholars who answered our call for papers and who participated in our online workshop discussions in 2021, and Guillaume Yon, who ran that online workshop; participants at a Duke/CHOPE discussion of this chapter in 2022; those who refereed papers for this special issue; and the editor and managing editor of HOPE, Kevin Hoover, who invited us to organize and produce this special issue, and Paul Dudenhefer, who dealt so kindly with our manuscript. Finally we thank our article authors, from whom we have, as expected, learned more than meets the eyes.

Notes

1.

One of many sources from a wide range of writings on the rhetoric of economics, McCloskey 1990a provides the materials for analysis here because of its explicit attention to narrative.

2.

In her article on differential equations, McCloskey (1991) explored this sort of usage.

3.

On “narrative evidence,” please see sec. 3.2.

4.

This variety can be seen in the chapters on narrative in different modern sciences in Morgan, Hajek, and Berry 2022.

5.

See, e.g., Lenel’s (2021) account of the forecasting efforts by the Harvard Economic Service during the interwar years.

6.

See the discussion in MacCormick 2005. There is some overlap between these legal criteria and the criteria for explanation when using case study reports, which generally include multiple kinds of evidence (see Campbell 1975).

References

Bates, Robert H.
1998
. “
The International Coffee Organization: An International Institution
.” In Bates, et al
1998
:
194
230
.
Bates, Robert H., Avner Greif, Margaret Levi, Jean-Laurent Rosenthal, and Barry R. Weingast
1998
.
Analytic Narratives
.
Princeton, N.J.
:
Princeton University Press
.
Bianchi, Marina, and Roberta Patalano.
2017
.
Storytelling and Choice
.
Rounded Globe
. https://roundedglobe.com/html/1dd3bc40-0775-4182-87a1-d20e0529076f/en/Storytelling%20and%20Choice/.
Campbell, Donald T.
1975
. “
‘Degrees of Freedom’ and the Case Study
.”
Comparative Political Studies
8
, no.
2
:
178
93
.
Carroll, Noël.
2007
. “
Narrative Closure
.”
Philosophical Studies: An International Journal for Philosophy in the Analytic Tradition
135
, no.
1
:
1
15
.
Collier, Paul, and David Tuckett.
2021
. “
Narratives as a Coordinating Device for Reversing Regional Disequilibrium
.”
Oxford Review of Economic Policy
37
, no.
1
:
1
16
.
Dillon, Sarah, and Claire Craig.
2021
.
Storylistening: Narrative Evidence and Public Reasoning
.
London
:
Routledge
.
Fogel, Robert W.
1964
.
Railroads and American Economic Growth: Essays in Econometric History.
Baltimore
:
Johns Hopkins Press
.
Friedman, Milton, and Anna Jacobson Schwartz.
1963
.
A Monetary History of the United States, 1867–1960
.
Princeton, N.J.
:
Princeton University Press
.
Giraud, Yann.
2021
.
Review of Narrative Economics: How Stories Go Viral and Drive Major Economic Events
, by Robert J. Shiller.
Journal of the History of Economic Thought
43
, no.
4
:
665
70
.
Hawthorn, G.
1991
.
Plausible Worlds
.
Cambridge
:
Cambridge University Press
.
Hoover, Kevin D., and Stephen J. Perez.
1994
. “
Post Hoc Ergo Propter Once More: An Evaluation of ‘Does Monetary Policy Matter?’ in the Spirit of James Tobin
.”
Journal of Monetary Economics
34
, no.
1
:
47
74
.
Ingrao, Bruna.
2018
. “
Models in Economics: Fables, Fictions, and Stories
.”
Annals of the Foundazione Luigi Einaudi
52
:
109
32
.
Leamer, Edward E.
1983
. “
Let's Take the Con Out of Econometrics
.”
American Economic Review
73
:
31
43
.
Lenel, Laetitia.
2021
. “
Searching for a Tide Table for Business: Interwar Conceptions of Statistical Inference in Business Forecasting
.” In
Exploring the History of Statistical Inference in Economics
, edited by Jeff Biddle and Marcel Boumans.
History of Political Economy
53
(
supplement
):
139
74
.
MacCormick, Neil.
2005
.
Rhetoric and the Rule of Law: A Theory of Legal Reasoning
.
Oxford
:
Oxford University Press
.
McCloskey, Deirdre N.
1990a
.
If You're So Smart: The Narrative of Economic Expertise
.
Chicago
:
University of Chicago Press
.
McCloskey, Deirdre N.
1990b
. “
Storytelling in Economics
.” In
Narrative in Culture
, edited by C. Nash and M. Warner,
5
22
.
London
:
Routledge
.
McCloskey, Deirdre N.
1991
. “
History, Differential Equations, and the Problem of Narration
.”
History and Theory
30
:
21
36
.
Meunier, Robert.
2022
. “
Research Narratives and Narratives of Nature in Scientific Articles: How Scientists Familiarize Their Communities with New Approaches and Epistemic Objects
.” In Morgan, , Hajek, , and Berry,
2022
:
247
66
.
Morgan, Mary S.
2001
. “
Models, Stories, and the Economic World
.”
Journal of Economic Methodology
8
, no.
3
:
361
84
.
Morgan, Mary S.
2012
.
The World in the Model
.
Cambridge
:
Cambridge University Press
.
Morgan, Mary S.
2017
. “
Narrative Ordering and Explanation
.”
Studies in History and Philosophy of Science
Part A,
62
:
86
97
.
Morgan, Mary S.
2019
. “
Exemplification and the Use-Values of Cases and Case Studies
.”
Studies in History and Philosophy of Science
78
:
5
13
.
Morgan, Mary S.
2021
. “
Narrative Inference With and Without Statistics: Making Sense of Economic Cycles with Malthus and Kondratiev
.” In
Exploring the History of Statistical Inference in Economics
, edited by Jeff Biddle and Marcel Boumans.
History of Political Economy
53
(
supplement
):
113
38
.
Morgan, Mary S.
2022
. “
Narrative: A General Purpose Technology for Science
.” In Morgan, , Hajek, , and Berry,
2022
:
3
30
.
Morgan, Mary S., Kim M. Hajek, and Dominic J. Berry, eds.
2022
.
Narrative Science: Reasoning, Representing, and Knowing since 1800
.
Cambridge
:
Cambridge University Press
.
Romer, Christina D., and David H. Romer.
1989
. “
Does Monetary Policy Matter? A New Test in the Spirit of Friedman and Schwartz
.”
NBER Macroeconomics Annual
4
:
121
84
.
Shiller, Robert J.
2019
.
Narrative Economics: How Stories Go Viral and Drive Major Economic Events
.
Princeton, N.J.
:
Princeton University Press
.
Stapleford, Thomas A.
2021
. “
Revisiting the Past? Big Data, Interwar Statistical Economics, and the Long History of Statistical Inference in the United States
.” In
Exploring the History of Statistical Inference in Economics
, edited by Jeff Biddle and Marcel Boumans.
History of Political Economy
53
(
supplement
):
175
203
.
Tetlock, Philip E., R. N. Lebow, and G. Parker, eds.
2006
.
Unmaking the West
.
Ann Arbor
:
University of Michigan Press
.
Tobin, James.
1965
. “
The Monetary Interpretation of History
.”
American Economic Review
55
, no.
3
:
464
85
.
Tuckett, David, Douglas Holmes, Alice Pearson, and Graeme Chaplin.
2020
. “
Monetary Policy and the Management of Uncertainty: A Narrative Approach
.”
Bank of England Staff Working Paper
870
.
Tuckett, David, and M. Nikolic.
2017
. “
The Role of Conviction in Decision-Making under Radical Uncertainty
.”
Theory and Psychology
27
, no.
4
:
501
23
.
This content is made freely available by the publisher. It may not be redistributed or altered. All rights reserved.