1 Financial Models

11 Financial Models Ekaterina Svetlova Chapter 11 in C. Borch and R. Wosnitzer (eds), The Routledge Handbook of Critical Finance Studies (2021), London and New York: Routledge, 228-243 Abstract The chapter presents two modes of critical discussions about financial models. One mode refers to criticizing finance as a science involved in developing and spreading unrealistic models that are detached from the complex reality of economic life. The proponents of this strong critical position take the “ideological” or even “political” standpoint and show how finance is always enmeshed with social and political power and contributes to inequality, unjust risk distribution, market crashes, etc. (critical as political). The other approach—as pursued, for example, by social studies of finance—shifts the attention from how financial models fail to adequately represent the economic reality to how they are used and shape, or perform, reality. The performativity approach questions the traditional representational view of finance and aims for a deeper understanding of financial practices and their interplay with models and theories (critical as analytical). The connection between the two modes in form of politics of performativity is explored. Introduction In the aftermath of the 2008 crisis, financial models have been heavily criticized for being dangerous and not sufficiently helping people to make sound investment decisions. Models were accused of causing the turmoil or, at least, of failing to give advance warning. The arguments behind these accusations are familiar: financial models are abstract and unworldly constructs so that their users are predestined to be misguided. Thus, the argument goes, as insufficient models became widespread tools for decision-making in financial markets, the vast majority of market participants were seduced by their mathematical sophistication and blindly followed them towards alleged safety. Financial models “behaved badly” while confusing “illusion with reality” (as the title of the book written by the famous quant turned publicist and educator Emanuel Derman (2011) suggests). The general problem related to this blindness was discussed as the so-called “model-based herding”: markets might start to move in resonance, and such a development could threaten their stability, causing bubbles and crashes as we partly observed in 2008. Furthermore, the more recent severe stock market corrections have again raised the question of how dangerous financial calculative technologies are for markets’ stability and whether humans are losing control. Nasdaq CEO Adena Friedman said on CNBC that “humans are definitely in charge of the decisions in the market” and that “the algorithms are written basically on the back of a human decision.” At the same time, CNBC quoted influential banking analyst Dick Bove who claimed that “the United States equity markets have been captured by out-of-control technological investment systems.” Thus, mathematical models and algorithms turned out to be at the epicenter of critical debates about finance and its societal implications. However, in order to better evaluate this debate, we have to ask What does it mean to be critical? This question has been prominently addressed in the field of critical management studies (Fournier & Grey, 2000), not in the field of critical finance though. From my point of view, there are two understandings of “being critical” that have transpired in finance studies so far (see also Curran, 2018). One refers to criticizing finance as a body of knowledge, as a science involved in developing and spreading unrealistic models that are detached from the complex reality of economic life and thus prone to provide a misleading advice. While focusing primarily on the issue of “truth” and representation as well as the epistemological standards of assessing financial models, this critical finance perspective has perceived the problems associated with financial modeling from the position of philosophy of social sciences. This approach was at the origin of what I call the “strong” critical finance perspective discussed in the introduction to this Handbook as “an explicitly strong angle” (Bay & Schinckus, 2012; Frankfurter et al., 1994; McGoun, 1993, 1997). Importantly, building upon the “critical theory” (e.g., Horkheimer, 1982[1937]), the proponents of this position took the “ideological” or even “political” standpoint in the critical debate on finance. They claimed that the false assumptions implied in financial modeling are never value-neutral, and that finance is always enmeshed with social and political power and thus contributes to inequality, unjust risk distribution, market crashes, etc. The other approach—as pursued, for example, by social studies of finance (SSF)—takes a different, more subtle, stance on “being critical.” It shifted the attention from how financial models fail to adequately represent the economic reality to how they are used and shape (or perform) reality. For some observers, e.g., Mirowski and Nik-Khah (2007) and Curran (2018), this move lacks any critical “bite” and rather downplays the essential role of economists and financial theorists in facilitating (or neglecting) major economic instabilities and inequalities. Furthermore, while understanding performativity as an interventional account of financial science (“models and theories shape reality”), some detractors of the SSF wonder why scientists do not interfere with market practices more actively and do not clearly pursue a critical, i.e., ideological, agenda, during those interventions, as the concept of critical performativity (Alvesson & Spicer, 2012; Spicer et al., 2009) suggests. What is often missed in this debate is that the theoretical and methodological move carried out by the SSF scholars was more profound than just the shift from representation to performativity. It was a move to investigate finance as a practical field, not (just) as science, and to focus on market practitioners such as traders, security analysts, hedge fund managers, and merger arbitragers. The financial market professionals, next to financial mathematicians at universities and business schools, became important subjects of the intensive empirical investigations. Thus, what the SSF brought to attention is that financial models travel both ways—from academia to markets and back—and are used not because they give market professionals true knowledge about markets (or “reality”), but because they fulfill other, non-epistemic functions: they help users to make or legitimize decisions, present themselves as experts, communicate with each other, etc. Though slow (due to their origin in the STS and sociology of knowledge), the SSF re-shifted its focus from knowledge production to acting sensibly and decision-making under market uncertainty as central activities in which models are involved and represent just one, though important aspect. For the SSF, financial market practitioners are not “F9 model monkeys” (Tett & Thal Larsen, 2005) who uncritically adopt models imported from academia and ignore their deficiencies but pragmatic users who combine models with judgments, emotions, narratives, etc., in order to make financial decisions. In this context, “being critical” means to question the traditional functional and representational view of finance and to aim for better, i.e., deeper understanding of financial practices and their interplay with models and theories. For me, this approach subscribes rather to “critical thinking” (akin to “Socratic questioning”) that implies scrutinizing the mainstream ideas, not taking them for granted, comparing theory to observations, and forming judgment while being guided by practice (e.g., Beyer, 1995). The main trait of this bottom-up critical position is that its proponents don’t patronize practitioners but take their views and procedures seriously. By doing so, the SSF scholars “complicate” the link between models and reality. They don’t know in advance what they will say at the end of their analysis, whereas the top-down “ideology” always knows what it claims. “Critical thinking” is not about naming and criticizing phenomena but about understanding how these phenomena come about giving the investigation a subtle critical gesture. There is a difference between critical as political and critical as analytical that has never been seriously thematized with respect to finance, and maybe this chapter and the Handbook more generally will get it off the ground. In this contribution, I will juxtapose the views of the “strong” critical finance and SSF on financial models, compare their conclusions, and, based on this comparison, identify some promising directions for the future critical discussion. Primarily, in my view, the potential of politics of performativity (Boldyrev & Svetlova, 2016: 10) should be explored in-depth in order to understand the potential of the interventional and more generally the critical approach to finance. “Strong” Critical Finance: Still Waiting for the Last Finance Professor Though the discussion about unrealistic and useless models in finance that accelerated after the crisis of 2008 might have appeared refreshing and radical to some financial market observers, it was certainly not new for the “strong” critical finance scholars. Already in the 1990s, when “strong” critical finance made its first steps, the critique of financial theory and models was at the core of their debate. One of the initial meetings of the group—the roundtable at the Financial Management Association (Frankfurter et al., 1994)—devoted most of its time to the discussion of the gap between the established financial models such as the Capital Asset Pricing Model (CAPM), the Efficient Market Hypothesis (EMH), and the Modigliani and Miller model, and the real-life phenomena that can be observed in markets. The roundtable participants criticized financial modeling for its reliance solely on the assumptions of utility maximization, calculable risk, perfect competition, and frictionless markets. The critical finance scholars focused on what models cannot explain or which aspects of reality they unjustifiably exclude. For example, the neoclassical Miller and Modigliani model does not account for the abnormal profitability of some corporate investment opportunities and ignores the possibility of companies’ bankruptcy (Frankfurter et al., 1994: 178f.); such models, so the critical finance scholars argued, cannot explain real people’s behavior. This gap between models and reality was conceptualized by the Frankfurter et al. circle as “hyperfinance,” the finance namely that lost any touch with economic reality and just refers to nothing but itself (McGoun, 1997). Based on Baudrillard’s work, financial theories and models were considered to be simulacra, namely copies of reality that became completely detached from the original and thus actually stopped being even copies (McGoun, 1997; Schinckus, 2008). For the “strong” critical finance proponents, a good example of simulacrum is the theory of “fair” value. According to the established finance theory, the “fair” value is a correct, or true, value of a company based on its “fundamentals” and thus, the “fair” estimate of what the asset is worth today and how much investors should be ready to pay for this asset, assuming that markets are efficient; the price, or what an investor effectively pays when buying a security, might deviate from the intrinsic value but always fluctuates around it (Koller et al., 2010: 337). For McGoun (1997), however, the “fair” value is a useless concept, a simulacrum with no anchor in the corporate reality because investors cannot determine the “fair” value due to the uncertainty of future cash flows; thus, stock prices don’t have anything to do with corporations’ fundamentals and fluctuate widely. Thus, for “strong” critical finance, financial markets are a self-referential postmodern game where people primarily speculate and trade for the sake of trading without real purpose, causing the fluctuations that cannot be captured by the existing financial theory. Keasey and Hudson (2007) labeled the mainstream finance a “house without windows,” and Frankfurter (2006) notoriously wrote about the EMH and the CAPM as “blind” religion and ideology. Thus, the “strong” critical finance debate about financial models has been clearly of philosophical nature and addressed the question about the relation between models and reality as one about the nature of knowledge in finance: How can financial models produce knowledge despite their unrealistic assumptions, insufficient forecasting power, untrue representations, and other epistemological flaws? The criticism in this context meant blaming models for being imperfect and proclaiming them useless due to their non-conformity with “reality.” The future of financial theory and modeling was painted as doom and gloom: I see the finance field going down cointegration paths and GARCH [Generalized AutoRegressive Conditional Heteroscedasticity] and ARCH [AutoRegressive Conditional Heteroscedasticity] and garbage model trails and coming eventually to a very untimely end in a dark alley in St. Louis, because there will be nothing left. There will be this one person with a few bags of belongings and dirty hair, who will say, “I am the last finance professor.” But nobody will pay him anymore, because everyone realizes that what he did was silly. (Frankfurter et al., 1994: 100) Though posing important critical questions about the epistemic value(s) of financial models, the critical finance unfortunately missed the opportunity to bridge its discussion with the extensive work in philosophy of science and STS on modeling as a scientific activity. The philosophical and STS literature has been explicitly concerned with the question of how models help to acquire scientific knowledge about real-world phenomena (target systems); at the same time, financial models have never been at the core of their attention. Still, the philosophical debate about models as representations (Giere, 1988, 2010; van Fraassen, 1980), the idealization, and de-idealization accounts (Cartwright, 1989, 1999; Mäki, 2009; Morgan & Knuuttila, 2012), particularly a prolonged discussion about the unrealistic assumptions of economic models and the poor correspondence of these models with the target system (Hausman, 1992; Reiss, 2012), could have stimulated and enriched the critical debate about financial models initiated by the “strong” school, especially the debate on the link between models and reality as well as on the “truth” issue. It is particularly interesting that some philosophers have claimed that models can be useful even if they do not expose any direct connections to the real world. Accounts of models such as credible worlds (Sugden, 2000), parables (Cartwright, 2008), fictions (Frigg, 2010; Godfrey-Smith, 2009), and make-believe (Toon, 2012) suggest that models are not created by observing a target system and stripping out complicating factors (“idealization”) but by imagining a model world that could be true and thus allowing for meaningful communication and understanding of essential relationships between parameters. These concepts correspond to McGoun’s (2003) suggestion to see the CAPM and the Black-Scholes option pricing model not as positive or normative models but as metaphors, or “useful frameworks”, that rather serve as rhetorical devices and not as scientific explanation tools. Again, the cross-fertilization between philosophy of science and financial economics might have helped to develop an interesting account of financial models as instruments of fictional truth (cross-referring to Beckert’s (2016) concept of fictional expectations). At the same time, the “strong” critical finance scholars could not help but open their discussion to sociology of financial markets. They started to pose the question why models that are completely detached from reality have been strongly influential in the academic world, even received Nobel prices (e.g., the CAPM in 1990), in other words, why we still have not seen the impoverished last finance professor yet. They pointed out that the reasons might be “sociological” or “cultural,” e.g., the scientization and quantification of finance in academia, particularly in business schools, and claimed the necessity to “undertake research on the process of research itself” (McGoun, 1993: 174) without seriously following this empirical program. With regard to the practice of markets, the “strong” school made an interesting observation that investment professionals are rather skeptical about the “unrealistic” models and thus reluctant to apply them in their practice (e.g., McGoun, 1993 on CAPM) as models don’t “survive contact with the real world” (Coleman, 2014). Based on interviews with fund managers, Coleman claimed that financial theories and models don’t play an important role in the professional investors’ decision-making. The reasons for the denial are manifold: (a) false assumptions (e.g., risk-return trade-off, market efficiency, rational expectations, and “fair” asset pricing) that are not supported by empirical tests and (b) the unavailability of the relevant data. Thus, when relying on models, financial decision-makers fear to lose sight of the markets while sitting in “the house without windows.” In sum, the critical finance researchers pointed out to some important critical issues related to financial modeling and formulated the promising research program that they missed to fully implement. The “strong” critical tradition suggested that finance should open itself towards a variety of new topics and disciplines, e.g., philosophy, ethics, and art (Bay & Schinckus, 2012), and pay attention to broader social implications of finance and contextualize finance within society, markets, and organizations. Importantly, in order to understand the “notoriety” of unrealistic financial models in academia and practice of markets, the critical finance scholars envisaged the application of qualitative methods of empirical research (Bettner et al., 1994; Frankfurter et al., 1994). Coleman (2014: 235) suggested that the first step towards the new financial paradigm would be bridging the gap between models and reality “by setting out the actual behavior of markets, investors and managers.” At the same time, the critical finance scholars seldom carried out any significant empirical work despite proclaiming its necessity. They rather directed their efforts towards the replacement of one dogma (classical finance) by the other ideology, e.g., the Marx-inspired criticism of wealth maximization, the casino capitalism, the too loose control of markets and capital, the uneven distribution of financial resources, etc. They envisaged the radical, politically engaged program (akin to the “critical theory”) and proposed the theory of fair markets (Frankfurter, 2006) as a very general alternative to the CAPM/EMH paradigm. Curiously though, the most important points of their program, for example, interdisciplinarity and devotion to the deep empirical research were taken up and put into life by the SSF to which I turn now. SSF: Power and Failure of Financial Models The SSF is an emerging interdisciplinary field that applies the findings and the methodological apparatus of various social sciences (sociology, anthropology, geography, political economy, etc.) to the analysis of financial markets. A significant effort of the SSF scholars has been dedicated to the critical analysis of how modern economic life is shaped by mathematical models. The “critical” in the SSF context is rather understood as careful bottom-up questioning of the status quo while paying particular attention to the practice of markets where financial models are applied. Let me highlight some important SSF insights that are crucial for the critical debate on modeling. Taking Practitioners Seriously The SSF scholars demonstrated that the academics-practitioners gap is not as strongly pronounced in financial modeling as in other disciplines. Finance has been traditionally characterized by the striking proximity and entwinement of theory and practice. Financial models have constantly traveled between, or simultaneously “inhabited,” two worlds: academia and the financial industry. Indeed, some models such as the Black-Scholes model were created at universities and then adapted in the practice of markets. However, today, it is not seldom that models are developed by practitioners. This trend became especially distinctive in the last decade of the twentieth century, when many academics trained in mathematics and physics were hired by investment banks as “quants” or financial engineers (Patterson, 2010), and was re-enforced now with the development of high-frequency trading or HFT (on high-frequency trading, see also Lange’s chapter in this Handbook). Financial modeling today is a constantly revolving process between science and practice the borders of which are nearly non-existent. These insights led the SSF scholars to recognize the prominence of practitioners in financial modeling. As the primary goal of financial market participants is not to acquire knowledge but to make decisions, it is dissatisfactory to consider their practices as purely “epistemic,” science-centred practices. Whereas scientists can live in the “small” world of their idealized models and work with models that are caricatures of reality or just “credible worlds” (Spears, 2014), practitioners need models that guide them through the world and enable them to decide and act. Model use in practice of markets principally differs from scientific modeling. Financial market participants understand their decision-making as a constant process of discovery where “equipment matters” (MacKenzie, 2009: 13). The SSF rejects both the over-calculative and under-calculative views on modeling in markets (Beunza & Garud, 2007). Whereas the so-called over-calculative position suggests that blind applications of formal financial models dominate the markets constantly rendering them at the edge of an imminent disaster, the under-calculative view, on the contrary, considers calculations and modeling to be useless or unimportant as financial agents rely primarily on social resources such as norms, institutions, and networks. But how can we claim that investment professionals—such as Coleman (2014) suggests—completely ignore or neglect models when the latter have become important and ubiquitous tools in all fields of financial markets? In today’s financial world, to model or not to model is not the question. Thus, the SSF hase striven to develop a third—integrative—view of financial markets and modeling. For example, analyzing the use of models by merger arbitrageurs, Beunza and Stark (2012) provide an empirical example of model use as the interdependence between the social and the calculative. The authors claim that no aspects—neither social (mimesis, networks, institutions) nor calculative (models)—should be neglected when explaining decision-making in financial markets. Rather, the calculative, social, and technical aspects should be simultaneously taken into consideration. The analysis of this “collectively constructed calculative technology” (Beunza & Garud, 2007: 19) has become the key programmatic issue of the SSF and represents their critical stance. Performativity, Counter-Performativity, and the Indeterminacy of Model Effects Taking the integrative view as point of departure, the SSF developed a distinct account of the relationship between models and “reality,” namely the account of performativity. According to the performativity concept, the relationship between financial models and economic reality is understood not as passive (representation), but as active. Financial models have effects: they influence or even constitute what they aim to represent. The careful analysis of these effects—and herewith of the link between models and reality—is an important critical gesture of the SSF. The critical stance of the performativity concept becomes particularly obvious if one avoids its simplistic reading. Indeed, performativity was often understood as the one-way travel of theoretical financial models into the realm of markets without any contingencies and “problematicness” behind this travel and its effects (Curran, 2018). However, the performativity scholars rather strive to understand and show empirically how economic phenomena and events come about and how they are shaped by financial modeling and technology. By doing so, they explicitly highlight the non-deterministic character of those processes. This non-determinism is important for the “critical” finance discussion for two reasons. First, it allows performativity theory to criticize the mainstream approach to economics (Esposito, 2013: 108) bringing to the fore the contingency of market participants’ expectations, the reflexivity of their behavior, and the radical uncertainty of markets, the issues namely that are widely neglected by the mainstream economic theory. Second, the performativity concept provides a nuanced answer to one of the central “critical” questions about modern financial markets: How endangered and “crisis-prone” are they? As already indicated, the common discussion of financial crises claims that the widespread thoughtless use of models creates specific risks that are relevant for the markets as a whole, namely model herding. Those risks are described in the literature as “second-order dangers” (Holzer & Millo, 2005), “model risk” (Esposito, 2013), or “resonance” (Beunza & Stark, 2012), and relate to a distinct new form of interdependence among market participants which is mediated by models. The performativity scholars, however, do not focus on blaming models or markets but show in careful empirical analysis how model-based actions can provoke—or avoid—herding. The non-deterministic character of performativity—and thus of models’ effects on markets—is emphasized in two SSF concepts: cultures of model use and counter-performativity. First, in my book Financial Models and Society (Svetlova, 2018), I demonstrate how indeterminant performative effects unfold in the multifaceted interplay between users and models in the practice of markets or flexible cultures of model use. Investors might apply their formal tools but ignore the tools’ recommendations in the very process of decision-making or, relying on own judgments, narratives, emotions, social observations of the others, etc., “overlay” the decisions models prescribe. Furthermore, financial analysts frequently use their dividend cash flow (DCF) models as “opinion proclaimers,” i.e., they apply models to express their pre-formed opinions about the market or a security; in other words, financial analysts play with parameters and numbers in the model until the model fits the subjective “fair” price or the users’ subjective views more generally. Qualitative “overlay” and opinions’ representation do not mean that the valuation process is completely detached from reality of markets as the “strong” critical finance suggests. On the contrary, the SSF empirical findings demonstrate how valuation is anchored in the reality by means of models. The empirical accounts rather point out that, in many cases, financial models do not provide direct prescriptions for decisions, and, therefore, the link between models, decision-making, and market events is not as straightforward as the ongoing critique of financial models indicates. We find large “pockets” where human judgment and stories are as important as the complicated formulas and algorithms, and their interplay produces uncertainties and unforeseen consequences for financial decision-makers. Generally, performativity analysis contributes to better understanding of the uncertain nature of economic events and critically questions probabilistic calculus and rationality of agents which are at the core of mainstream economics. The second concept that points to the indeterminacy and ambiguous character of models’ effects is counter-performativity. Models may create a reality that they describe but, at the same time, can be counter-performative producing as well: a very particular form of misfire, of unsuccessful framing, when the use of a mathematical model does not simply fail to produce a reality (e.g., market results) that is consistent with the model, but actively undermines the postulates of the model. The use of a model, in other words, can itself create phenomena at odds with the model. (Bamford & MacKenzie, 2018: 100) Those misfires can appear, on the one hand, when performative “felicity conditions” are not fulfilled (MacKenzie, 2007: 70). As witnesses and a priest must be present at the wedding ceremony in order to “bring about” husband and wife, the Black-Scholes model in the 1970s—in order to produce effects—had to acquire sufficient authority to warrant users’ beliefs, possess sufficient cognitive simplicity, and be publicly available and supported by an appropriate technology so that a sufficient number of market participants could start to use it. If all those conditions were not fulfilled, the Black-Scholes model might have failed to create the billion-scale option market. Also, when circumstances changed in 1987, the model produced counter-performative effects: its use created “skewed” patterns of implied volatility that contradicted the straight line posited by the model (Bamford & MacKenzie, 2018). On the other hand, more generally, the misfires are intrinsic to every performative process: “[B]reakdown is constitutive of performativity (performativity never fully achieves its effect, and so in this sense ‘fails’ all the time)” (Butler, 2010: 158). Performativity is closely related to performance, or staging, that is, by its very nature, devoid of any planning and control; artistic performances are unique and unrepeatable—and thus to some extent always unpredictable and surprising. There are no pre-defined circumstances in which a performance succeeds. Financial models participate in financial markets’ performances as communicative tools, persuasion devices, and props for “scientific” and “objective” knowledge helping to feign (or perform) investment decisions as rational and legitim. Those theatrical acts of persuasion and convincing staging—like performances in a real theatre—might not be liked or “believed” by audiences and thus fail, or misfire, at any moment. In this sense, models might produce unintended or unexpected results. This proposition can be best illustrated using the example of herding. Herding, Anti-Herding, and Financial Markets Stability The widespread use of similar models might lead to herding. Beunza and Stark (2012) demonstrate, for example, in their account of reflexive modeling that, while constantly observing and backing out the spread plot that represents the market consensus, merger arbitragers might lock themselves into the thinking of the market and connect themselves to other financial actors. Hence, individual errors, interlocked in the process of model application, might be amplified and produce resonance of decisions. Because models become themselves a part of the very phenomena they describe and their use and effects are constantly observed by other market participants, models unintentionally co-produce unwanted (or “critically” questionable) market phenomena, e.g., the misestimation of market risks in the case of the Long-Term Capital Management (Holzer & Millo, 2005), the GE-Honeywell merger failure (Beunza & Stark, 2012), “the correlation crisis” and faulty valuation of structured products prior to financial crisis (MacKenzie, 2011), or the drastic price fall in August 2007 due to “quant quake” (Tett & Gangahar, 2007). Still, in normal times, the individual(ized) applications of financial models might produce dissonance, i.e., the divergence of opinions among market participants (Beunza & Stark, 2012). The analysis of cultures of model use as well as the counter-performativity account supports this view and relativizes the fear of an imminent market collapse due to blind usage of identical models: the anti-herding tendencies can be created in the process of model use. As every decision implies the “undoing” of models, there is always a moment of flexibility in financial decision-making. The empirical patterns of model application demonstrate that, while using models to structure decisions, observe markets, or express opinions, investment professionals are free to follow or not follow the model prescriptions, to suspend or “game” them. There are many individually fashioned styles of using one and the same model. For example, the DCF can be applied by fund managers to anchor and communicate decisions, by investors to “reverse engineer” the market, and by financial analysts to express their judgment. Exactly because the styles of model use differ, there is no way that the different users manipulate models absolutely identically and derive the same results. Various strategies of “model overlay” and “opinion proclamation” can be applied by market participants to disagree with the market or, at least, to question the market’s views. Thus, the various strategies of model use give rise to forces that counteract herding tendencies. Cultures of model use do not automatically promote a particular behavior in financial markets but can re-enforce disagreement. Also, the counter-performativity account highlights that models can produce unexpected results. Recently, Borch (2016) stressed that we have to take into consideration the further technological aspect of markets: the algorithms. He claimed that, due to the increased importance of algorithms in financial markets, the herding tendencies have further amplified. He questioned Beunza and Stark’s (2012) and my view that model use frequently generates dissonance and claimed that the interdependences we observe in the HFT segment are interdependences among algorithms, and that human oversight and human control generally play a lesser role in the modern financial markets. Still, the emotional interference of traders with their algorithms cannot be damped completely (Borch & Lange, 2017). This discussion of who is in control of markets is at the core of the “critical” dispute about modern finance and its technology. The nuanced analyses of model use produced in the SSF allow to lead an evidence-based discussion on the issue and not to jump to pre-formed conclusions or adhere to blaming. As already said, the SSF critical approach is akin to “critical thinking” that does not accuse financial market practitioners of being thoughtless and blind but takes their views seriously as a point of departure for a balanced reality-rooted analysis of financial practices. Furthermore, the SSF does not produce simple recipes. From their point of view, the traditional solutions for improving financial market stability, e.g., to “conquer” models as an evil, to ban them and go back to “intuition” (Derman, 2011) and “common sense” (Triana, 2011), or to generally apply “fewer” models as the “strong” critical finance suggests, seem to be off target. First, given their ubiquitous use, it is unrealistic to eliminate models and algorithms from modern markets. Second, models as such do not represent a danger because in fact they do not dictate decisions; they are always combined with judgment, emotions, and tacit knowledge in the practice of their use. The problem is not about the re-introduction of human judgment into the nearly fully automated and formalized markets. We can hardly find a model or an algorithm used without a human component at one or the other stage: social and organizational elements are constantly “folded” into a market (Muniesa, 2007). These observations relativize the accusations of the “strong” critical finance proponents towards financial models as being disconnected from the real world; rather, models are always connected to markets through fulfilling various non-epistemic functions in the practice of their use (e.g., Millo & MacKenzie, 2009; Svetlova, 2018); those non-epistemic connections and effects should be analyzed in more empirical details. Thus, if the goal of critical finance is to understand how to arrive at the “better,” i.e., more stable and less crisis-prone markets, the SSF suggests focusing on an in-depth analysis of various “qualculative” practices (Cochoy, 2008) and the modi of differently combining human judgment and modeling. The cultures of model use in a merger arbitrage department take different forms than those that they take in an HFT company or in asset management. Critical finance should avoid generalizations and investigate these practices in their own right. The governability of modern markets depends on the proper understanding of cultures of model use and counter-performativity which simultaneously produce order and disorder, resonance and dissonance. A “restorative regulation”—understood as the correction of purely technical malfunctioning of markets (Engelen et al., 2012)—can hardly be applied to financial markets understood as socio-technical agencements. Some Thoughts on Future Research At the core of the critical finance studies seems to be the issue of politics of performativity (Boldyrev & Svetlova, 2016; also Cabantous et al., 2016): How being analytical does not preclude being political and vice versa? This might continue to be the central question for the future research in the field. I think that the more nuanced approach to performativity presented in this chapter suggests some directions. It clearly rejects the “simplified” reading of Callon’s (1998) and MacKenzie’s work as one implying “the conflation of economic models with economic reality” (Curran, 2018: 493) in a quasi-automatic way and thus neglecting the central critical issue with regard to models: they are false and misleading and thus a significant factor of market turmoils. Performativity studies in my understanding do not ignore the gap between models and reality but suggest that financial market participants close this gap in situ of markets, in the process of model construction and model use by means of narratives, interpretations, pragmatically addressing various audiences, pre-formulating the anticipated model results, etc. Financial models are constantly connected to the markets in current decision-making situations; they calculate but are also suspended, confirmed, or questioned with regard to their results in those immediate, real-time connections to the market. They might succeed but might also produce the unexpected and unintended effects that undermine their own predictions. In other words, there are many—often unforeseen—ways of how they change the economic world and financial markets. These insights rather call not for performativity being generally “more critical” but for more analytical rethinking of the concept, its theoretical value, and political consequences. First of all, as already mentioned, performativity theory might enrich the mainstream economics by clarifying the social nature of radical uncertainty and critically analyzing such phenomena as market fluctuations and herding, asset valuation, risk-taking in banks and corporations, as well as credit ratings. Furthermore, the performativity studies might want to clearer acknowledge that, while moving into the realm of markets or even being created directly in banks, asset management, or HFT companies, financial models and algorithms find themselves in an explicitly non-epistemic context. Thus, the performativity debate should be less about “epistemic cultures” or “knowledge cultures.” While analyzing practitioners’ decision-making as searching for productive methods to simultaneously calculate and suspend calculations, we are not merely talking about “the other kind of knowledge,” the specific “tacit knowledge,” or “know-how,” but about a process where knowledge produced by models is just one component of real actions as immediate involvements with the complex, constantly evolving world. Financial models are applied as not purely epistemic devices in the practice of financial markets; thus, their examination from the critical finance perspective should take this issue into consideration and ask: which non-epistemic functions do models fulfill and with which effects? Such an investigation opens up to multiple contingencies and counter-performative effects that are implied in the situational bridging the gap between models and markets and pays attention to materiality and processuality of financial markets. The performativity studies can show how models’ influence in non-epistemic contexts is mediated by the institutional environment, fictional (and often ideological) narratives, political interests, and the power of the involved actors. Thus, bridging the gap between models and reality is often about “the struggles of performation,” a series of collective efforts to create and sustain certain realities based on one’s vision that does not have anything to do with academic knowledge. It is also about “the political engineering of sociomaterial agencements that are constituted within and across organizations, institutions and markets” (Cabantous et al., 2016: 3). Those struggles can be uncovered and analyzed in critical finance studies as politics of performativity. Obviously, this analysis is not inherently apolitical as it reveals special powers that mediate between academia, markets, and society. However, it is less judging and interventional and aims to awake the interest of practitioners and invites the latter to reflect on their practice. In accordance with Curran, this program could allow for “developing a critical approach to social science as both a body of knowledge and a series of social and political processes that constantly reshape social and material life in often unexpected ways” (2018: 494) overcoming the juxtaposition of being critical as political and critical as analytical. References Alvesson, M., & Spicer, A. (2012). “Critical leadership studies: The case for critical performativity,” Human Relations 65(3): 367–390. Bamford, A., & MacKenzie, D. (2018). “Counterperformativity,” New Left Review 113: 97–121. Bay, T., & Schinckus, C. (2012). “Critical finance studies: An interdisciplinary manifesto,” Journal of Interdisciplinary Economics 24(1): 1–6. Beckert, J. (2016). Imagined Futures: Fictional Expectations and Capitalist Dynamics. Cambridge, MA: Harvard University Press. Bettner, M. S., Robinson, C. & McGoun, E. (1994). “The case for qualitative research in finance,” International Review of Financial Analysis 3(1): 1–18. Beunza, D., & Garud, R. (2007). “Calculators, lemmings or frame-makers? The intermediary role of securities analysts,” Sociological Review 55(s2): 13–39. Beunza, D., & Stark, D. (2012). “From dissonance to resonance: Cognitive interdependence in quantitative finance,” Economy and Society 41(3): 383–417. Beyer, B. K. (1995). Critical Thinking. Bloomington, IN: Phi Delta Kappa Educational Foundation. Boldyrev I., & Svetlova, E. (2016). “After the turn: How the performativity of economics matters,” in I. Boldyrev and E. Svetlova (eds.), Enacting Dismal Science: New Perspectives on the Performativity of Economics (pp. 1–27). New York, NY: Palgrave Macmillan. Borch, C. (2016). “High-frequency trading, algorithmic finance and the flash crash: Reflections on eventalization,” Economy and Society 45(3–4): 350–378. Borch, C., & Lange, A.-C. (2017). “Market sociality: Mirowski, Shiller and the tension between mimetic and anti-mimetic market features,” Cambridge Journal of Economics 41(4): 1197–1212. Butler, J. (2010). “Performative agency,” Journal of Cultural Economy 3 (2): 147–161. Cabantous, L., Gond, J.-P., Harding, N., & Learmonth, M. (2016). “Critical essay: Reconsidering critical performativity,” Human Relations 69(2): 197–213. Callon, M. (1998). “Introduction: The embeddedness of economic markets in economics,” in M. Callon (ed.), The Laws of the Market (pp. 1–57). Oxford, MA: Blackwell. Cartwright, N. (1989). Nature’s Capacities and Their Measurement. Oxford: Oxford University Press. Cartwright, N. (1999). The Dappled World: A Study of the Boundaries of Science. Cambridge: Cambridge University Press. Cartwright, N. (2008). “Models: Parables vs fables,” Insights 1(11): 2–10. Cochoy, F. (2008). “Calculation, qualculation, calqulation: Shopping cart arithmetic, equipped cognition and the clustered consumer,” Marketing Theory 8 (1): 15–44. Coleman, L. (2014). “Why finance theory fails to survive contact with the real world: A fund manager perspective,” Critical Perspectives on Accounting 25: 226–236. Curran, D. (2018). “From performativity to representation as intervention: Rethinking the 2008 financial crisis and the recent history of social science,” Journal for the Theory of Social Behaviour 48(4): 492–510. Derman, E. (2011). Models. Behaving. Badly: Why Confusing Illusion with Reality Can Lead to Disaster, on Wall Street and in Life. New York, NY: John Wiley & Sons. Engelen, E., Ertürk, I., Froud, J., Johal, S., Leaver, A., Moran, M., & Williams, K. (2012). “Misrule of experts? The financial crisis as elite debacle,” Economy and Society 41(3): 360–382. Esposito, E. (2013). “The structures of uncertainty: Performativity and unpredictability in economic operations,” Economy and Society 42(1): 102–129. Fournier, V., & Grey, C. (2000). “At the critical moment: Conditions and prospects for critical management studies,” Human Relations 53(1): 7–32. Frankfurter, G. (2006). “The Theory of Fair Markets (TFM): Toward a new finance paradigm,” International Review of Financial Analysis 15(2): 130–144. Frankfurter, G. M., Carleton, W, Gordon, M., Horrigan, J., McGoun, E., Philippatos, G., & Robinson, C. (1994). “The methodology of finance: A round table discussion,” International Review of Financial Analysis 3(3): 173–207. Frigg, R. (2010). “Models and fiction,” Synthese 172 (2): 251–268. Giere, R. N. (1988). Explaining Science. Chicago, IL: University of Chicago Press. Giere, R. N. (2010). “An agent-based conception of models and scientific representation,” Synthese 772(2): 269–281. Godfrey-Smith, P. (2009). “Models and fictions in science,” Philosophical Studies 143(1): 101–116. Hausman, D. (1992). The Inexact and Separate Science of Economics. Cambridge: Cambridge University Press. Holzer, B., & Millo, Y. (2005). “From risks to second-order dangers in financial markets: Unintended consequences of risk management systems,” New Political Economy 10 (2): 223–245. Horkheimer, M. (1982[1937]). “Traditional and critical theory,” in M. Horkheimer (ed.), Critical Theory: Selected Essays (pp. 188–243). New York, NY: Continuum. Keasey, K., & Hudson, R. (2007). “Finance theory: A house without windows,” Critical Perspectives on Accounting 18(8): 932–951. Koller, T., Goedhart, M., & Wessels, D. (2010). Valuation: Measuring and Managing the Value of Companies. Hoboken, NJ: John Wiley & Sons. MacKenzie, D. (2007). “Is economics performative? Option theory and the construction of derivatives markets,” in D. MacKenzie, F. Muniesa, and L. Siu (eds.), Do Economists Make Markets? On the Performativity of Economics (pp. 54–86). Princeton, NJ: Princeton University Press. MacKenzie, D. (2009). Material Markets: How Economic Agents Are Constructed. Oxford: Oxford University Press. MacKenzie, D. (2011). “The credit crisis as a problem in the sociology of knowledge,” American Journal of Sociology 116(6): 1778–1841. Mäki, U. (2009). “Missing the world: Models as isolations and credible surrogate systems,” Erkenntnis 70(1): 29–43. McGoun, E. (1993). “The CAPM: A Nobel failure,” Critical Perspectives on Accounting 4(2): 155–177. McGoun, E. (1997). “Hyperreal finance,” Critical Perspectives on Accounting 8(1–2): 97–122. McGoun, E. (2003). “Financial models as metaphors,” International Review of Financial Analysis 12(4): 421–433. Millo, Y., & MacKenzie, D. (2009). “The usefulness of inaccurate models: Towards an understanding of the emergence of financial risk management,” Accounting, Organizations and Society 34(5): 638–653. Mirowski, P., & Nik-Khah, E. (2007). “Markets made flesh: Performativity and a problem in science studies, augmented with consideration of the FCC auctions,” in D. MacKenzie, F. Muniesa, and L. Siu (eds.), Do Economists Make Markets? On the Performativity of Economics (pp. 190–224). Princeton, NJ: Princeton University Press. Morgan, M. S., & Knuuttila, T. (2012). “Models and modelling in economics,” in U. Mäki (ed.), Philosophy of Economics: Handbook of the Philosophy of Science (pp. 49–87). Amsterdam: Elsevier Science. Muniesa, F. (2007). “Market technologies and the pragmatics of prices,” Economy and Society 36(3): 377–395. Patterson, S. (2010). The Quants: How a New Breed of Math Whizzes Conquered Wall Street and Nearly Destroyed It. New York, NY: Crown Business. Reiss, J. (2012). “The explanation paradox,” Journal of Economic Methodology 19(1): 43–62. Schinckus, C. (2008). “The financial simulacrum: The consequences of the symbolization and the computerization of the financial market,” The Journal of Socio-Economics 37(3): 1076–1089. Spears, T. C. (2014). Engineering Value, Engineering Risk: What Derivatives Quants Know and What Their Models Do. PhD dissertation, University of Edinburgh. Spicer, A., Alvesson, M., & Kärreman, D. (2009). “Critical performativity: The unfinished business of critical management studies,” Human Relations 62(4): 537–560. Sugden, R. (2000). “Credible worlds: the status of theoretical models in economics,” Journal of Economic Methodology 7(1): 1–31. Svetlova, E. (2018). Financial Models and Society: Villains or Scapegoats? Cheltenham: Edward Elgar Publishing. Tett, G., & Thal Larsen, P. (2005). “Market faith goes out the window as the ‘model monkeys’ lose track of reality,” Financial Times, May 20. Tett, G., & Gangahar, A. (2007). “System error: Why computer models proved unequal to market turmoil,” Financial Times, August 15, p. 7. Toon, A. (2012). Models as Make-Believe: Imagination, Fiction, and Scientific Representation. Basingstoke: Palgrave Macmillan. Triana, P. (2011). The Number That Killed Us: A Story of Modern Banking, Flawed Mathematics, and a Big Financial Crisis. New York, NY: John Wiley & Sons. van Fraassen, B. C. (1980). The Scientific Image. Oxford: Clarendon Press. Information Classification: General Information Classification: General31