The Sociological Imagination

In contrast to the ahistorical, price-taking, utility-maximizing, omniscient actors which fill the abstract models of neoclassical economics (what Coase (1988: 19) mockingly called “blackboard economics”), economic sociologists have made vast improvements in the study of actually existing markets. Fourcade (2007) breaks the sociology of markets down into three primary camps: network analysts, field analysts, and performativists. While the sociology of markets is most associated with the embeddedness program of Granovetter (1985), this is probably the least “sociological” of the three camps. As Krippner (2001) has persuasively shown, network analysts commandeered Polanyi’s concept of embeddedness and gave it a new meaning making it synonymous with network ties. For reasons explained in the short answer below (See #1), this essay will focus on the other two camps – field analysts and performativists.

Markets as Fields

Fields analysis is the older of the two perspectives and has largely taken a center role in the sociology of markets after the decline of network analysis. One of the strengths of field analysis is that it has its origins in several social science disciplines (Scott 2007: 1-18). Field analysts in each discipline saw importance in different topics/subjects and often employed different methods to answer questions about them, but in all cases they put market institutions at the center of their analysis. Scott (2007: 48), synthesizing several strands of thought, defines institutions as “comprised of regulative, normative and
cultural-cognitive elements that, together with associated activities and
resources, provide stability and meaning to social life.” This is a useful
working definition for our purposes here. Institutions are necessary because economic actors live in an uncertain world where preferences, production functions, and resources are not objectively given to all. Homo economicus may know exactly how to act but the rest of us need some guidance. For Max Weber, institutions served to help actors rationally calculate their actions (Swedberg 1998). In this way institutions include a wide variety of structures such as prices
(Hayek 1948), networks (Granovetter 1985), firms (Williamson 1985), cultures
and states (Fligstein 2001).

Field analysts have focused on market institutions as they are embodied in culture/norms, states, and powerful actors with the ability to set the agenda. The project of field analysts has been to show how larger social structures influence market actors (whether individuals or organizations) as well as how actors compete to shape institutions in turn. Cultural arguments date back to Weber’s (1905) argument that a “protestant work ethic” based on Calvinist precepts was rationalized and transformed into the “spirit of capitalism” which led to the modern world. More recently, McCloskey (2010; forthcoming) argues that the industrial revolution and modernity did not come about because of material (i.e. coal) or legal (North 1990) changes. Instead, she looks at the role of rhetorical changes among elites in Holland and Britain and finds what was really important was their reevaluation of the bourgeoisie. Trade, invention, and manufacturing took on a new meaning as “dignified” which unleashed the creative destruction necessary for economic development.

Among modern sociologists, Meyer and Rowan (1977) famously described institutions as “myth and ceremony.” Rather than the formal structures themselves being the most important (Williamson 1985), they saw institutionalized culture and norms as more influential. On the individual
level, Fligstein (1990) examines individual “conceptions of control” which
frame the way individuals see “problems” and “solutions.” On the national
level, Dobbin (1994) looks at the role of what he calls industrial culture in
shaping the economic organization of railway markets the United States,
Britain, and France over the 19^th century. These cultures “influence policy-making by contributing to collective understandings of social order and
instrumental rationality in different nations, and modern industrial policies
are organized around those ideas (Dobbin 1994: 2).” Similarly, Meyer et al (1997) move from a national to an international context arguing that there a “world culture” exists shaping the way that nations govern themselves and the economy.

States, with their pervasive presence in modern society, make for another obvious candidate for the embodiment of institutions. North (1990), describing market institutions as the “rules of the game,” puts most of his attention on property rights, regulations, and legal rules enforced by states. Dobbin and Dowd (2000), for example, show how early antitrust laws shaped railway firm business strategies for competition. Until the advent of antitrust laws, Massachusetts railways relied on collusion to avoid ruinous “cutthroat” competition. When antitrust laws outlawed cartels, firms first resorted to intense price competition at a cost to their financiers. Eventually, these powerful banks reigned in this competition by threatening to withhold funds from firms who engaged in price-cutting wars. This in turn led to a merger movement which stabilized the industry and led to the market as we know it today. The “varieties of capitalism” literature probably takes state institutions most seriously (Hall and Soskice 2001). They see states as providing institutional complementarities that give firms in that nation a particular comparative advantage. Liberal Market Economies (LME) are based largely on competitive markets and hierarchies, and conform to the transaction cost view of markets. Coordinated Market Economies (CME), on the other hand, rely more heavily on networks to move information and cooperation among firms. In each case, the state determines acceptable market behavior.

While many economic sociologists have tended to focus on one explanation over the other, it would be a mistake to characterize them as competing explanations. This is made most evident by recent work. Fligstein (2001) has done the best job at integrating the two perspectives within his aptly named political-cultural approach to markets. On the individual and organizational level, people have those particular conceptions of control which shape their outlook. At the same time, political organizations (primarily the state) regulate economic activity from the top down. Krippner (2011), looking at how financial services rose to dominance in the American economy, credits both the triumph of the ideology of shareholder value and a potpourri of legal changes in all dimensions of American law. Continuing the tradition of scrutinizing seemingly “noneconomic” areas of social life, Healy (2006) looks at the market for human blood and organs. He finds that it is regulated and made possible by people’s beliefs about donation, state regulation of the industry, and the rules of the organizations involved.

Field analysts, as they study markets, rely on two major concepts imported from Pierre Bourdieu. The first is the concept of fields of action. In neoclassical theory, atomized actors do not orient themselves to any particular time or place. They act in accordance to opportunities to maximize their objective utility wherever they may arise. Network theorists have narrowed down the scope of actors’ orientation to their industry (White 1981) or their organization (Burt 1992), but this is still not accurate. A field can be described as the social space in which actors orient their actions to each other. Dimaggio and Powell (1983) adopted this concept in their seminal article on “instititutional isomorphism” among organizations. They wanted to understand why organizations interacting in the same organizational field tend to converge on one or a few organizational forms. This proposed that occurs through three mechanisms: “1) coercive isomorphism that stems from political influence and the problem of legitimacy; 2) mimetic isomorphism resulting from standard responses to uncertainty; and 3) normative isomorphism, associated with professionalization (Dimaggio and Power 1983: 150).” As previous examples allude to, fields can take many shapes from interactions among members of a profession (Abbott 1988; Guillen 1994; Babb 2001; Fligstein 2001) to national contexts (Dobbin 1994; Hall and Soskice 2001) or even the whole world (Meyer et al 1997).

The other concept, though not always used explicitly, is habitus. According to Bourdieu (2005: 84), “Insofar as he/she is endowed with a habitus, the social agent is a collective individual or a collective individuated by the fact of embodying objective structures.” In other words, individuals within particular fields internalize the institutional structure in which they act. This does not mean that actors are simply “dopes” who act according to institutional constraints. They actively shape them as well (Dimaggio and Powell 1983). Field analysts have been careful to show that this happens both purposely and in ways actors did not intend (Dobbin and Dowd 2000; Fligstein 2001; Zelizer 2010; Krippner 2011). In keeping with the emphasis on allowing actors to act in an uncertain world, Bourdieu (2005: 85) remarks, “Habitus is a highly economic principle of action, which makes for an enormous saving in calculation (particularly in the calculation of costs of research and measurement) and also in time, which is a particularly rare resource when it comes to action.” For field analysts, it is very much about calculation.

Markets as Performance

Performativity is the newest of the three camps, coming out of the Actor-Network Theory (ANT) program of Bruno Latour and taken into economic sociology by Michael Callon (1998). The origins of each camp are important because it helps explain why they are on their present day trajectories. Field analysis, as we see from Weber and Bourdieu, was originally a reaction against the materialism of Marxist theories of the economy. ANT, on the other hand, can be seen as a reaction against the perceived excesses of social constructionism. It is one thing to reject material determinism, it is some altogether different to completely ignore the material aspects of markets. Against this backdrop, performativists have spent the bulk of their time looking at how socioeconomic technologies and “market devices” have shaped the way people interaction in market settings. Technologies, as conceptualized by performativists, can be either material or nonmaterial. Buenza and Stark (2004), for example, examine the way that the material layout of a Wall Street trading room shapes what goes on there. They find that often ignored aspects, like the use of computers and proximity of desks to each other, have huge effects on the way traders perceive opportunities and act on them. The idea of economic technologies also encompasses the knowledge of economists. MacKenzie (2006) traces the development of the Efficient Market Hypothesis and concomitant Black-Scholes Option Pricing Model and its consequent application on the Chicago Mercantile Exchange. He is able to show how the model did not originally fit what was happening on the exchange. Instead of throwing out the model, economists and officials from the exchange worked to change the behavior of market actors to conform to the model and eventually succeeded. MacKenzie concludes that economic models are “an engine” driving markets rather than just “a camera” taking snapshots of a reality from afar.

In the same vein as field analysts, performativists put much emphasis on market
technologies as “calculative devices.” Although Hayek (1948) would argue that
prices serve as transmitters of local knowledge, he discounts the fact that the
prices themselves must be constructed and interpreted by the actors utilizing
them. This is where the various kinds of calculative devices come into play. Take the evaluation of someone requesting a mortgage as an example. The bank representative must find a cost-effective way of determining whether someone is
worthy of a loan. If they have personal knowledge of the applicant then they might be able to judge their “worthiness” at little cost. But how does this work in the context of a national banking system often detached from the local scene? Poon (2009) details the rise of FICO scores in serving as a substitute for personal knowledge. The use of nominal-level scores allowed banks not only to determine who was creditworthy but even parse them out into degrees of creditworthiness. In other words, they make them commensurable (Espeland and Stevens 1998) and thus create new markets where none existed before the advent of the technology. Performativists, in line with ANT, blur the line between actors and the material environment. MacKenzie (2009: 19-25) describes the actor-network at “agencement.” Performativists see it as impossible to separate actors from their technical environment and even go so far as to attribute agency to non-human material objects. All this action happens in what we might call a particular socio-technical assemblage.

It is worth taking a moment to discuss the aspect of performativity that has received the most attention to date among economic sociologist: the idea of economic theory as a kind of self-fulfilling prophecy. MacKenzie (2006) breaks performativity down into three broad categories. Generic performativity occurs when market actors simply use abstract economic theory in the real world. A good example might be the fact that people drive less when gas prices rise. There is little evidence that the market actors involved explicitly know that economic
theory asserts that demand curves slope downwards, but their actions conform to the theory nonetheless. Effective performativity occurs when economics makes a difference in the outcome. For example, Pigouvian theories of externalities and social costs are actively used by regulators when creating taxes on cigarettes or particular pollutants. These two categories of performativity are rather common in the world. The strong program pushes it one step further. Barnesian performativity occurs when market actors make the theory being used “more true.” What exactly does this mean? MacKenzie compares it to the concept of a self-fulfilling prophecy. Economic theories and models might not actually be true in any realistic sense (with their myriad assumptions, we can easily see why this could be the case) but can become true as market actors conform to the theory in question. The aforementioned work on the Black-Scholes options pricing model fits this criteria.

While this is an interesting and novel concept, I think it is a dead end as performativity as a progressive research program goes. The reason is that performativity relies heavily on institutionalist arguments about the greater social structure within which agents (however you want to define them) act. As Fourcade (2007: 1027) argues, “Indeed, once social conditions are taken into consideration, the radical novelty of the performativity argument is somewhat diminished.” MacKenzie (2009: 181) recognizes this fact argues that performativity is a necessary addition to other embeddedness approaches in economic sociology. They focus on the social but ignore the technical, but “Because the ‘social’ and ‘technical’ are inextricably linked in market construction, the two sets of tools will ultimately need to be integrated full: a challenging but important academic task.” It is worth making a side by side comparison from the above descriptions.

A Critical Comparison

When we examine the two camps together, several similarities emerge. The table below looks at them through the lens of the “five Ws” (plus one H) of markets.

First and foremost, we find that both theories put calculation at the center of their analyses. Why do people create and utilize market institutions and technologies (which are really synonymous)? Because it helps them reduce uncertainty and enables them to act within their field or assemblage (again, the two concepts are largely synonymous). This comparison even shows that both theories have some of the same shortcomings. Schneiberg (2007: 51) repeats the critique often lobbed at field analysts that “institutions create stability and have causal efficacy, until they don’t.” Without a comprehensive theory of social change, field analysts have relied on exogenous shocks (“crisis”) or some heroic
“institutional entrepreneur” to usher in social change. Correspondingly, critics of performativity argue that economic theories perform until they don’t. While there is a name for this, counterperformativity, there is no real theory behind it. Both cases require further research and theoretical development to fill in the holes. In this case, the answer may be somewhere in between performativity’s emphasis on agencement in everything and field analysis’s emphasis on the actions of a single individual or organization – what Aldrich (2011) calls “institutional entrepreneurship NOT institutional entrepreneurs.”

The pragmatic question of “how do we go about doing this” still remains. Scott (2007), in tracing the early history of institutional theory, unknowingly makes a prescient point about the future development in the sociology of markets:

Why was the impact of the early institutionalists blunted? Modern-day commentators offer several explanations. The German Historical School no doubt overemphasized the uniqueness of different economic systems and underemphasized the value of analytical theory. Even sympathetic critics acknowledged that Veblen exhibited ‘an explicit hostility to intellectual ‘symmetry and system-building’’ (Hodgson 1996: 211) and that Commons’ arguments were hampered by his ‘idiosyncratic terminology and unsystematic style of reasoning’ (Vanberg 1989: 343). But a more serious shortcoming was the tendency for the work to generate into naïve empiricism and historicism. Emphasizing the importance of the particular time, place, and historical circumstance, institutional analysis came more and more to underline ‘the value of largely descriptive nature and function of politico-economic institutions’ (Hodgson 1991: 211).

The resemblance to contemporary performativity theory is uncanny. Replace “institutionalists” with “performativists” then “institutions” with “technologies” and all the same critiques would still apply. There is no reason to reinvent the wheel with performativity though. Going forward the best strategy seems to be integrating specific insights from performativity into field analysis. This means dropping the excessively historical accounts and inserting more theory without the unnecessary new jargon. It also means dropping the idea that non-human objects have agency as it adds little value to the analysis. Just as field analysts have cherry-picked concepts from Bourdieu (Dobbin 2008), so they should do the same with performativity. To boil it down to one concept, field analysts should come to grips with the materiality of markets and integrate the study of market technologies into their work. MacKenzie (2006) comes closest to approximating this ideal as do Espeland and Sauder (2007).

As I argue in one of the short essays below, the primary limitation unique to a
political-cultural-technological approach to markets is an inability to comment
on the efficiency of particular institutional arrangements and organizational
forms. By avoiding such a discussion, economic sociologists relegate themselves
to the sidelines of major intellectual and policy debates. One does not need to
go back to neoclassical economics in order to formulate such a theory. New
Institutional economics (North 1990) and network analysis (Uzzi 1997) already
do this so it is only a matter of integrating their insights (or alternatively,
Marxist economic theories) into the framework.