This paper explores some interconnections between the "original," or "old," institutional economics (OIE hereafter) and the French economics of conventions (EC hereafter), with the aim of promoting a mutually beneficial cooperation between them. (1) Since its inception, (2) the EC has acquired a certain prominence in France, where it is one of the main approaches that develop an alternative to neoclassical economics, particularly regarding the study of institutions. The conventionalist approach is still not very well known in economic circles outside French-speaking countries, but its contributions are very interesting and deserve greater attention. (3)
At a general level of discussion, one can identify some important points in common between the OIE and the EC. Both approaches emphasize the role of institutions and conventions as nonmarket factors that promote order and coordination in economic life. In part because of this, they see economics as a social science and are very open to contributions from other disciplines. Indeed, it is hard to think of other approaches in economics that are closer to several other social sciences. Some original institutionalists and conventionalists even go beyond interdisciplinarity and defend the unification of all social sciences. Also shared is the view that institutions and conventions have a mental dimension and a behavioral one. Another common characteristic is the emphasis on values and on normative evaluation.
This paper compares the OIE and the EC at a more specific level, regarding two theoretical issues related to the conceptualization of institutions in the OIE and conventions in the EC: cognition and valuation. It also considers the temporality of the relation between institutions or conventions and individuals. Finally, it briefly relates these discussions of cognition, valuation, and temporality to an assessment of the methodological positions of OIE and EC.
Institutions, Conventions, and Cognition
For the EC, a convention is not always a rule of behavior: it may be a mental construct. Olivier Favereau (1986), for instance, distinguished between [convention.sub.1], a collective or social representation, and [convention.sub.2], an interindividual behavior. Batifoulier et al. 2001 uses the clearer terms conventional model of evaluation to designate a convention of thought, and (as already suggested by Favereau) conventional rule to designate a convention of behavior. (4) The first type of convention goes beyond mutual expectations and is a social representation of a collective entity, an image of the relation among the individuals involved. A rule (be it conventional or not) is incomplete, that is, its sense is not objectively given and its conditions of justifiable application cannot be enumerated exhaustively (Favereau 1999, 166). Hence a rule cannot be applied mechanically; it needs to be interpreted (1998, 218; 1989, 292). The mental representation shapes this interpretation.
Andre Orlean has also referred to a convention as a shared reference or representation (1999), but, more specifically, he has developed the notion of "convention of interpretation," which he applies only to the context of financial markets. This convention, also called the financial convention, is "a specific model of interpretation and analysis of the economy" (124), an interpretative model (177), or a model of evaluation (144). In the Stock Exchange, for example, a convention of interpretation defines "the expected forms of economic development, in terms not only macroeconomic, but also qualitative, for example, by positing that certain forms of governance or certain forms of company growth should be privileged" (2004, 40).
Conventionalists like Favereau (1989) and Orlean (1994) treat conventions as "collective cognitive devices." They emphasize the role of conventions in the process of cognition and particularly interpretation. (5)
All this suggests that the conventionalist approach has an important similarity with the original institutionalist views on institutions and cognition. In particular, the EC seems to acknowledge, in the case of conventions, what I have termed the deeper cognitive role of institutions (Dequech 2002, 2005). (6) Not only do institutions provide information to individuals, including the indication of the likely action of other people (I call this the informational-cognitive role of institutions) but they also influence the very way people select, organize, and interpret information. This latter, deeper role, particularly regarding interpretation, is also attributed to conventions by the French conventionalists.
On the other hand, this similarity is not enough to imply that the EC also resembles the OIE in allowing conventions to influence individuals in fundamental ways, so that individuals cannot be taken as given in relation to conventions. It is still necessary to compare the OIE and the EC in terms of their treatment of the temporality of the relation between individuals and institutions or conventions. This broader issue will be addressed later in this paper.
In any case, it is clear that the OIE and the EC differ with respect to the role of habits in cognition and in general. In contrast to the OIE, the EC does not give much prominence to habits, despite a few isolated references to tacit knowledge. (7) The conventionalists apparently want their agents to be fully aware of the conventions they follow. This seems to be part of an effort to avoid portraying individuals as "cultural dopes," as sociology was often accused of doing. Institutionalists (and even some Austrian economists) may, however, fear the conventionalists go too far.
Institutions, Conventions, and Valuation
The OIE and the EC share a strong concern with valuation and normative issues. Valuation could be discussed...