[I took my comprehensive exam in economic sociology last week and will be posting the essays on here over the next couple of weeks for your summer enjoyment.]
When I hear the complaint that economists do not take ideas and institutions seriously, I am reminded of an economist (an acquaintance who shall remain nameless) who revels in informing me that sociology is “nothing but Marxists and feminists.” It is true that a great number of economists do not take ideas and institutions seriously as it is true that a great number of sociologists are Marxists and feminists, but the picture does not hold up as well now as it did in the 1960s when this particular economist was in graduate school. In the same vein, sociologists who still insist that economists do not take institutions seriously must willfully ignore the Nobel Prizes awarded to George Stigler [1982], James Buchanan [1986], Ronald Coase [1991], Gary Becker [1992], Robert Fogel and Doug North [1993], Vernon Smith [2002], and most recently, Elinor Ostrom and Oliver Williamson [2010] – all of whom did extensive work on the role of institutions (though not necessarily ideas) in economics.
That said, it is helpful to break economic institutionalists down into three primary groups: neoclassical institutionalists, transaction cost economists, and what I will call cognitive institutionalists. I will examine each in turn and compare them to various sociological approaches to institutions along the way.
Neoclassical Institutionalism
The University of Chicago was home to the first generation of neoclassical institutionalists. It is no surprise that the department known for its economic imperialism was the first to apply economic analyses to areas outside traditional economics. Neoclassical institutionalism (as well as transaction costs institutionalism as we will see) has its origins in the work of Ronald Coase (1988) whose early work on The Nature of the Firm and The Problem of Social Cost began to introduce the idea of transaction costs into economic analysis. Neoclassical institutionalists, while keeping with the neoclassical framework as the name implies, began to explore the ways in which the state could affect the costs of economic transactions. Neoclassical institutionalists studied both the cause and effect of institutions. Those who studied the effect of institutions fall under the umbrella of the Law & Economics (L&E) movement. Posner (1973), for example, examined the law from an efficiency standpoint and came away with the claim that the common law was economically efficient. Becker (1968) examined criminal law through the lens of cost/benefit analysis. He assumed criminals were rational like everyone else and made decisions based on the potential costs and benefits of committing a crime. The law could change crime rates by tweaking the costs of crime.
Those who studied the cause of institutions fall under the umbrella of Public Choice (PC) theory which begins with Buchanan and Tullock’s (1962) The Calculus of Consent where they try to work out the pareto optimal constitutional rules for democracy. They import all the assumptions of neoclassical economic actors and attribute the same assumptions to political actors. Thus, bureaucrats do not look out for the elusive public interest, but rather their own interest in the form of maximizing their budget (Niskanan 1971). Politicians do not seek the optimal level of taxation, but rather they try to maximize tax overall revenues (Brennan and Buchanan 1980). Special interest groups engage in rent-seeking (as opposed to profit-seeking) behavior to stymie competition (Tullock 1967).
Neoclassical institutionalists, despite two key differences, have much in common with neo-Marxist and post-Marxist theories of the state in sociology. Both differences are methodological. While the former builds mathematical models to be tested by quantitative data, the former eschew mathematical model in favor of logic-based arguments which rely on qualitative historical methods for their application. Secondly, public choice takes methodological individualism as its starting point whereas Marxist theories begin with class analysis (See Barrow 1993 and Mueller 2008 for comprehensive summaries). This is where any differences end though. Both public choice and Marxist theories assume that actors, whether individuals or classes, seek out their material self-interests which they already know beforehand. Take the case of South African apartheid as an example. Bonacich (1981), a neo-Marxist, famously argued that the system was a result of higher price white labor trying to avoid competition from lower priced black labor. Lowenberg and Kaempfer (1998) essentially make the same argument from a public choice perspective (with all the associated mathematical modeling). Another commonality is that both largely deal with the institutions of the state and ignore (or dismiss) the impact of culture, organizations, and ideas on actors. Whether the state is simply a conduit for special interests or an actor itself, everything boils down to material interests.
There is also an ideological component to each theory. Marxist theories see capitalism in terms of class struggle, accumulation, and crisis. Markets are inherently unstable and therefore capitalists must intervene though the state to stabilize the economy, restore legitimacy, and suppress the working class (O’Connor 1973). Public choice theories, on the other hand, see the market as largely equilibrating, and with a few exceptions (provision of public goods and regulation of negative externalities), efficient and therefore see less need for state intervention (Buchanan 1977; Mueller 2008). Thus the taken for granted assumptions about markets effect their analyses of state institutions. This also explains why broad cross-fertilization between the two schools has not taken place.
Transaction Cost Institutionalism
As previously mentioned, the work of Ronald Coase also contributed to the growth of transaction cost institutionalism. Looking back, Coase (1988: 15) lamented the fact that neoclassical institutionalists had misinterpreted his article on social cost. Rather than assume a world of zero transaction costs, “What my argument does suggest is the need to introduce positive transaction costs explicitly into economic analysis so we can study the world as it exists.” This task has been taken up by Williamson (1985) and North (1990).
Williamson (1985) opens up the “black box” of the neoclassical firm by moving from the view of the firm as a production function to viewing it as a governance structure than can take many shapes and forms. These structures are necessary because there are varying costs to using the price mechanism versus internal production. Firms exist to minimize transaction costs. He modifies several assumptions from neoclassical economics. The first is rationality. Neoclassical theories assume simple utility maximization by actors. Williamson instead assumes bounded rationality where actors intend to be rational but are limited in their cognitive abilities to do so. Secondly, he assumes that actors will engage in opportunism or “self-interest seeking with guile” if given the chance (Williamson 1985: 47) whereas opportunism is simply assumed away in neoclassical models. He also includes three dimensions for rationally choosing different arrangements: asset specificity, uncertainty, and frequency. Williamson explores classical, neoclassical, and relational contracting. Depending on the frequency (occasional or recurrent) of the transaction and the level of asset specificity (nonspecific, mixed, idiosyncratic), we will get particular kinds of governance structures or institutions.
North (1990) largely uses the same assumptions as Williamson but focuses on state institutions. Institutions, according to North (1990: 3) are the “humanly devised constraints that shape human interaction.” They can be both formal and informal, purposely constructed or evolved over time. North is careful to point out a difference between institutions and organizations. The former are the rule of the game while the latter are the players. “The major role of institutions in a society,” according to North (1990: 6), “is to reduce uncertainty by establishing a stable (but not necessarily efficient) structure to human interaction.” We see convergence because of the interaction between organizations and institutions. There are lock-in effects (path dependence) and a feedback process. Institutions can be changed by entrepreneurs but it all depends on the information they receive and how they perceive it.
The major difference between Williamson and North are their views of informal constraints. Although we tend to focus on the formal rules that structure our lives, North agues that they make up a very small portion compared to informal norms, conventions, and moral constraints. Culture, as conceptualized by North, “provides a language-based conceptual framework for encoding and interpreting the information that the senses are presenting to the brain… In the short run, culture defines the way individuals process and utilize information and hence may affect the way informal constraints get specified. Conventions are culture specific, as indeed are norms. However, norms pose some still unexplained problems. What is it that makes norms evolve or disappear – for example, dueling as a solution to gentlemanly differences? (North 1990: 37, 42-43).” In other words, economists still know very little about cultural evolution or change.
Network analysts have been quick to dismiss transaction cost institutionalism. Granovetter (1985), for example, took explicit aim at the work of Williamson on markets and hierarchies. Of course, Granovetter does not actually challenge his general framework. Instead he argues that in addition to markets and hierarchies, there are network forms of organization. How is this any different from Williamson’s challenge to neoclassical economics on nontraditional contracting? If one steps back to think about it, there is no difference. Network analysis is really just an expansion of transaction cost institutionalism within another discipline which asks different questions. The real challenge comes from the varieties of capitalism approach. Hall and Soskice (2001: 15) argue, contra Williamson (1985), that firm strategy follows institutional structure rather than vice versa. Dobbin (1994: 8) points out that North (1981) makes the same argument as Williamson (1985) but he ignores the fact that North later changed his position on this and critiqued his own earlier work. I think this critique is largely correct and has been integrated into later transaction cost institutionalist arguments.
What of the critiques that Williamson ignores ideas, norms, and culture? Again, these critiques are correct and also apply to network analysis. North (1990) includes cultural factors in his analysis but his argues remain relatively undeveloped. Fligstein’s (1990) “conception of control” and Dobbin’s (1994) “rationalized meaning systems” can be seen as a highly developed versions of North’s (1990) “mental constructs” because they are able to explain cultural change in ways North admits economists cannot. This might be expected given the two disciplines’ diverging research interests. Transaction cost institutionalists are more interested in evaluating the efficiency of various institutional arrangements. Sociological institutionalists such as Dimaggio and Powell (1983), Dobbin (1994), and Fligstein (2001) are more interested in explaining institutional stability and change. North (1990: 83-91) argues that while stability is necessary for complex human interaction but it is not sufficient for efficiency. As noted in short answer question #3 below, sociological institutionalists have been reluctant to comment on the efficacy of particular institutional arrangements. Dimaggio and Powell (1983), in their discussion of coercive, mimetic, and normative isomorphism for example, argue that such changes are not rational in the strict neoclassical sense of the term, but make little attempt to conceptualize ways in which they might be considered rational (i.e. signaling theory, Knightian uncertainty, etc). This affects their visions for capitalist societies insofar as they believe it is legitimate to take positions on the desirability of one arrangement over another.
Cognitive Institutionalism
The last group consists of what I have termed cognitive institutionalism because it is made up of an eclectic group of scholars including Vernon Smith, Elinor Ostrom, Deirdre McCloskey, and the Austrians, who all focus on the cognitive aspects of institutions in a distinctly Hayekian way. Each of them focuses on what may be called the institutions of “spontaneous order.” Hayek’s student, Ludwig Lachmann (1971: 49-50) argued that an institution “provides means of orientation to a large number of actors. It enables them to coordinate their actions by means of orientation to a common signpost.” They are not necessarily embodied in states or organizations and are not always purposefully constructed yet they act as institutions regulating human behavior. Most importantly, they vary depending on the particular circumstances of time and place.
Smith (2003), following Hayek, makes a distinction between constructivist and ecological rationality. Acording to Smith (2003: 468) “constructivism uses reason to deliberately create rules of action, and create human socioeconomic institutions that yield outcomes deemed preferable, given particular circumstances, to those produced by alternative arrangements.” Ecological rationality, on the other hand, “uses reason – rational reconstruction – to examine the behavior of individuals based on their experience and folk knowledge, who are ‘naïve’ in their ability to apply constructivist tools to the decisions they make; to understand the emergent order in human cultures; to discover the possible intelligence embodied in the rules, norms, and institutions of our cultural and biological heritage that are created from human interactions but not by deliberate human design. People follow rules without being able to articulate them, but they can be discovered (Smith 2003: 470).” Smith uses experimental economics to explore not only how participants act in the laboratory setting but also to examine the rules that emerge to govern their interactions (Smith 1991; 2000). This distinction is important to overcome the critiques of rational choice theory. Cognitive institutionalism use a weak form of rationality which Williamson (1985: 46-47) calls “organic rationality.” Actors do not necessarily have to know how institutions work in order to use them and benefit from them.
Others have taken their research to the field. Ostrom (1990; 2000; 2005) examines what she calls “common pool resources” which mainstream economists assume would result in the “tragedy of the commons.” Instead, her field research finds that individuals are able to construct institutions to govern the use of such resources. She moves beyond markets, hierarchies, and networks to explore “polycentric governance systems” in which centers of decision-making overlap yet are formally independent from each other. Chamlee-Wright (2010) utilizes a framework she calls cultural economy. Cultural economy weaves Austrian economics and new institutional economics together with cultural sociology and network analysis. She finds that New Orleans residents are able to rely on community narratives as a tool for action in the wake of Hurricane Katrina when government rules were uncertain or shifting. For Chamlee-Wright, it is very much a story of spontaneous order.
McCloskey (2010) examines the role of rhetoric in changing the perception of commerce in 17th and 18th century Holland and England. Previous transaction cost institutionalist arguments fell short by conceptualizing institutions as “mere constraints.” Alternatively, she takes culture and discourse more seriously than her predecessors.
How do cognitive institutionalists differ from sociological institutionalists? In many respects, there are no substantive differences at all. Cognitive institutionalists take both institutions and ideas very seriously and would be able to speak freely across disciplinary boundaries (as most have advocated). Minor differences still persist which have been influenced by the research questions scholars are expected to pursue in each discipline. With the exception of McCloskey, cognitive institutionalists tend to focus on micro and meso-phenomena while sociologists focus on macro-phenomena. This stems from their methodological tools. Smith uses laboratory experiments while Ostrom and Chamlee-Wright rely on fieldwork. McCloskey fits nicely with sociologists simply because she is making a historical argument. There is no reason why sociologists should only employ one or the other though. Zucker (1977), in one of the most important articles of early organizational theory, relied on a laboratory experiment to make her point about institutionalization. More recently, Hallett (2010) relies on fieldwork to make an argument about institutional “recoupling.”
Additionally, it would be safe to argue that cognitive institutionalists effective study both efficiency and stability – a problem which still plagues economic sociology (Fligstein and Dauter 2007). They do this in a way which does not necessarily cause one particular vision of capitalist society to emerge. Because many of the institutions important to actors are the result of human action but not of human design, it would be pointless to try to reform or improve upon them by replacing them with price mechanisms or state planning. Hayek (1988: 76) was quite explicit about this, arguing “The curious task of economics is to demonstrate to men how little they really know about what they imagine they can design.” Although we can observe the efficiency of different institutional arrangements, there is often very little we can do to change them without unleashing a whole slew of unintended consequences.
Most importantly, cognitive institutionalists have paid special attention to the role of civil society in solving many collective action problems. This expands the study of economic sociology beyond market-centered and state-centered approaches. Dating back to Polanyi (2001), some economic sociologists have paid lip service to culture and society but in reality have only seen culture important in so far as it shapes state actions (Dobbin 1994; Babb 2001; Blyth 2002). Cognitive institutionalists join the moralized market approach (Healy 2006; Quinn 2008; Zelizer 2010) and political-cultural approach (Fligstein 2001) in providing an active role for civil society and blurring the line between economy and society.
The evolution of economic institutionalism has closely mirrored that of sociological institutionalism. Both have reached the point where fruitful cross-fertilization is not only possible but necessary for the growth of each as both have a long way to go. Sociologists who complain that economists do not take ideas and institutions seriously should promptly be reported to the Society for the Prevention of Cruelty to Straw Men.
