orgtheory.net

We have a late entry to our online seminar of Mackenzie’s An Engine, Not a Camera. Peter Levin, an orghead hailing from Barnard College, has written a review of the book for Social Forces. He’s offered us a sneak peak of the review. Comments welcome!

Peter

Under the auspices of social studies of finance, sociologists are trekking ever closer to the core tools and concepts that have long been the sole purview of financial economics. Among the best of these scholars is Donald MacKenzie, whose earlier work on ways scientists stabilize and produce the world they set out to discover, proves useful background for his latest work investigating the origins and current configuration of arbitrage and financial economics. This book is an account of the rise of finance, drawing from interviews with many of the central figures in finance since the 1970s.

An Engine not a Camera analyzes the effects of economics (and finance in particular) on the execution of financial markets. The book’s central argument is that theories of finance have had a ‘performative’ effect on the markets they purport to describe. That is, financial models of the world shape that world, making it conform to the models. Performativity is not a perverse effect of economics on markets, neither does financial theory lie just in the realm of ideas. Instead, MacKenzie highlights the systematic ways economic theories have been incorporated into the infrastructure of markets (pp. 19-20). Performativity is also posited to varying degrees – developments can be weakly (’generic performativity’) to more strongly (’Barnesian performativity’) performative, reflecting the extent to which theories are generally or specifically called upon, and their relative effects on economic processes.

The three parts to this argument consist of an historical account of the trajectory of modern financial theories; an analysis of the effects of these models in the world between the 1970s and the 1987 stock market crash and its resultant effects on the models; and a test of performativity using the case of Long Term Capital Management (LTCM) and arbitrage.

Chapters 2-4 provide a detailed history of modern finance theory. Most striking are the developments leading simultaneously to a comprehensive theory of asset and options pricing and an increasing dissociation from assumptions that made sense in the empirical world. The ascendancy of “positive economics,” whereby theories were judged by their predictability rather than plausibility, led to a set of simpler assumptions, for example that markets are perfectly competitive. This in turn led to more predictive models. MacKenzie also demonstrates the ties between financial theory and its application. The early creators of the Capital Asset Pricing Model (which measures the relationship between risk and expected returns for securities) and the Black-Scholes-Merton formula (which used historical volatility to determine the ‘appropriate’ pricing for options) actually traded on their models. CAPM provided the theory underlying index funds, beginning in 1968 at Wells Fargo Bank, developed in conjunction with the leading financial economists of the day (p. 84). The Chicago Mercantile Exchange paid Milton Friedman $5000 to write up a theoretical justification of currency futures for presentation to federal regulators (p. 147). These developments show how financial economists were not content to just theorize their ideas – they deliberately realized them.

The book’s strength, and heart, is in chapters 5-7, MacKenzie’s ‘tests’ for performativity of financial models vis-à-vis financial markets. The B-S-M formula enabled options traders to decide how much options should theoretically be worth, given a set of parameters about the historical volatility of the underlying asset. Buying and selling options – especially selling them – was, and is, a tricky business. When Fischer Black produced, and sold, statistical tables of options prices based on B-S-M, he provided a very real-world guide to how the model would work in profiting off of mis-priced options. But the B-S-M formula did more than simply guide options sellers. It actually caused options prices to conform more closely to the formula. This is the strongest case for Barnesian performativity – options prices measurably shifted after popularized exposure to the B-S-M formula to become more in line with the formula. The model that was supposed to depict reality caused reality to move towards the depiction. Hence finance acts as an engine rather than a camera.

Chapter 8 examines performativity in arbitrage more directly, using the case of the hedge fund Long-Term Capital Management (LTCM). Instead of a blind faith in models, or particular negligence on the part of LTCM, MacKenzie finds support for the hypothesis that the fall of LTCM (who gained and then lost over $2 billion between 1994 and 1998) is attributable to more sociological processes of imitation and a collectively self-perpetuating cascade of adverse price movements. He concludes that a main conceptual benefit of performativity would be to complement more sociological “patterns of relatedness” (Knorr-Cetina and Bruegger 2002) – that is, to examine both performativity and connectivity – to understand the globalization of finance.

Engine is an incisive, well-written, and accessible analysis of one of the more technical aspects of financial economics. The strongest arguments are in the analysis of performativity around the development of Black-Scholes-Merton model, its deployment in the options market, and the effects of the 1987 crash on subsequent modeling of securities. Here, he clearly identifies something more than a self-fulfilling prophesy: for at least a brief moment in history, financial economics brought into being a physical market which used, and confirmed, precisely the modeled market conceived by financial economics.

The problems with the book relate to fundamental problems with the theory of performativity and, more specifically, the porting of this concept from the social studies of science to finance. First, as we move away from arbitrage and trading itself, the performativity concept falls apart. It helps us understand how options and arbitrage prices converged upon the models. But to understand the fall of LTCM, or the political greasing of the skids to get trading approved to begin with, performativity, it seems, proves less helpful than more sociological notions of ‘embeddedness’ and influence, or Neil Fligstein’s conception of markets as politics (Fligstein 2001). MacKenzie claims to be no economic sociologist, and purports not to be doing economic sociology at all (p. 25), but more dialogue with economic sociology proper would strengthen the argument.

Performativity would also benefit from more direct engagement with the study of organizations and work. Treating economists as scientists – and by extension, markets as science – does not hold as closely as we might expect, particularly because markets are so social. Indeed, an alternative explanation for LTCM’s troubles, from Roger Lowenstein’s more journalistic account, suggests that LTCM’s competitors actively conspired in the end to take them down (Lowenstein 2000). And Partnoy’s insider account of Morgan Stanley explains their pricing strategies for over-the-counter derivatives using less elegant but perhaps more descriptive terms: obfuscation, deceit, and outright fraud (Partnoy 1997). Indeed, MacKenzie’s account, drawn from interviews and archives, is skewed towards the elegant formulation and execution of financial theories and away from the more day-to-day workplaces where those theories are worked out.

Overall, the book is an accomplishment, written with delightful clarity without sacrificing complexity. If these are the emerging fruits of science and technology studies’ engagement with finance, we should look forward to the feast to come.

Fligstein, Neil. 2001. The Architecture of Markets. Princeton, NJ: Princeton University Press.

Knorr-Cetina, Karin, and Urs Bruegger. 2002. “Global Microstructures: The Virtual Societies of Financial Markets.” American Journal of Sociology 107(4):905-950.

Lowenstein, Roger. 2000. When Genius Failed: The Rise and Fall of Long-Term Capital Management. New York: Random House.

Partnoy, Frank. 1997. FIASCO: The Inside Story of a Wall Street Trader. New York: Penguin Books.