Endogenous Preferences and Embeddedness: A Reappraisal of Karl Polanyi
Rodrigues, Joao, Journal of Economic Issues
Economic Activity is motivated by social obligation and regulated by the moral context that governs social life in general.
--J. R. Stanfield, Karl Polanyi and Contemporary Economic Thought
Karl Polanyi's contributions to social theory are widely recognized among the generality of social sciences. In the fields of economic anthropology, historical and economic sociology, and economic history, Polanyi's work, in particular The Great Transformation, is considered classic. (1)
In this article I want to explore the relevance of Polanyi's contributions to economics in general and to the issue of endogeneity of human preferences in particular. Following Bowles 1998, we understand preferences as reasons for behavior or attributes of individuals that (along with their beliefs and capacities) account for the actions they take in a given situation. This is a broad concept of preference that goes beyond tastes to include different types of motives and values--Amartya Sen's (1977) commitments and Albert Hirschman's (1985) meta-preferences. We take preferences to be endogenous in the sense that they are at least partially formed and molded by institutions.
According to Geoffrey Hodgson (1988), this issue of endogenous preferences constitutes a "Rubicon Line" that divides neoclassical economics and even some of its critics from an institutionalist approach.
The assumption of a rational agent with a maximizing behavior and stable and given preference functions forms the core of neoclassical economics, together with the absence of chronic information problems and a focus on economic equilibrium (Hodgson 1999).
Neoclassical economists give various reasons for taking preferences as an exogenous element in their theories. Although it may be recognized that preferences vary across cultures and that consideration of economic and social structures is not irrelevant to assess them, it is argued that maintaining this unrealistic assumption does not do harm to a theory which has the sole purpose of predicting. This is the well-known Friedmanian argument (Etzioni 1985). Another view argues that the analysis of preferences may be of interest, but in the scientific division of work it is not to be taken by economics. A more radical view about preferences is advanced by Gary Becker and George Stigler: "Preferences are assumed not to change substantially over rime, not to be very different between wealthy and poor persons, or even between persons in different societies and cultures" (quoted in Etzioni 1985, 184). This view assumes that each individual has an immutable preference function which is defined by/for himself independently of any social influence. This is what is usually meant when preferences are taken as "given."
The exogenous view of preferences is absolutely crucial to the maintenance of "a remarkably parsimonious postulate: that of the self-interested, isolated individual who chooses freely and rationally between alternative courses of action after computing their prospective costs and benefits" (Hirschman 1985, 7). This ahistorical and abstract picture of the individual is the starting point of an individualistic reading of social reality. "Taking individuals as they are" is synonymous of taking individuals as neoclassical theory imagines them to be.
Presently there are signs that this asocial concept of individuality and preferences that has impoverished economics and diminished its relevance to the analysis of pressing economic problems is becoming a target of open criticism and a topic of debate even within mainstream circles (Gintis and Romer 1998). While this should be greeted as a positive development it is worth noting that many of the recent contributions are made without paying sufficient attention to an important tradition of economic thought that has been raising this issue for more than a century.
The idea of complex and endogenous preferences has always been of course central to the institutionalist tradition (Keller et al. …