Expectations, interactions between agents and technological regimes

– We argue here two types of interactions within an organization are crucial for understanding how the making of expectations is structured. Our basic assumption is that the making of expectations is realized with different manners in a small firm which works in an entrepreneurial regime than in a large firm of the ologopolistic or routinized regime (see Winter, 1984). We follow the very revelant analysis carried out by Swann and Gilll (1993). In a young small firm with a (very) small number of individuals, in other words when the complexity is weak, the exchange of information among the members is easier than in the case of a larger firm. On the other hand, the process of decision-making is more rapid. The expectations will be flexible. And conversely for the large firms of a routinized regime, because the making expectations is based on (organizational) routines which are the mean to cope with the stronger complexity due to the large number of individual agents. The process of expectations revision will be longer and expectations more difficult to modify. The existence of two different schemes, with regards to the making of expectations, entails some consequences on the stability of growth for each of both technological regimes. We will show the technological regimes have different properties in terms of stability of growth trajectory. The way by which the expectations are made explains these differences. Classification Codes: L2, O0, A1. 1. Expectations and interactions between economic agents: a general framework Economics shares with other sciences the idea the interactions between agents produce effects and new structures. Mainstream thought, founded on general equilibrium analysis, identifies only the interactions built on the relationships in competitive markets. Interactions outside of the market, externalities, may be dealt with by the establishment of property rights. Other approaches focus mainly on the interactions outside of the market for explaining the dynamic properties of evolution. For instance, in the epidemic * Centre Walras, Université Lyon-2, Faculté de Sciences Économiques, 16 Quai Cl. Bernard, 69365 Lyon 07, France. E-mail: [email protected]