The development of new institutional economics (NIE) has been a huge advancement in economics. Despite attempts to integrate concepts such as bounded rationality, however, NIE has predominantly remained a neoclassical, equilibrium-oriented theory which primarily studies institutions only in regard to their effects on allocative efficiency, that is, in regard to transaction and agency costs. This article analyzes the effects of institutions if instead of a neoclassical concept of competition an evolutionary concept of competition in the tradition of Friedrich Hayek and Joseph Schumpeter is used. Based upon earlier work of Wolfgang Kerber (1996, 1997) and Chrysostomos Mantzavinos (2001), it is shown that the institutional framework for markets influences both the dynamics and the direction of knowledge-generation in competition processes, in which innovations are generated and tested in the market. Or to put it differently: The institutional framework influences the selection environment of the market and therefore channels the development of knowledge in evolutionary competition processes. The paper can also be seen as an attempt at a greater integration of institutional and evolutionary economics.
Competition as an Evolutionary Knowledge-Generating Process
Two major strands of evolutionary concepts of competition can be distinguished. The first one encompasses Austrian market process theory, that is, Hayek's theory of "competition as a discovery procedure" as an experimentation process (1978) and the market process theory of Israel M. Kirzner (1997). Much broader are the Schumpeterian approaches (competition as processes of innovation and imitation). Rooted in the basic ideas of Schumpeter (1934) about innovations, the entrepreneur, and economic development, concepts of dynamic competition and the main body of innovation economics have been developed (Dosi 1988; Freeman and Soete 1997; Metcalfe 1998). Particularly through the research of Richard Nelson and Sidney Winter (1982), this innovation economics has also an explicit evolutionary dimension by the application of the (biological) variation-selection-mechanism to innovation and market processes (Nelson 1995). The basic ideas of an integrated evolutionary theory of market competition (Kerber 1997) using valuable elements from all of these strands can be summarized as follows.
This evolutionary approach can be presented best by beginning with Hayek's concept of competition (1948, 1978), which differs radically from neoclassical competition. He criticized that the theory of perfect competition with its assumption of perfect knowledge "throughout assumes that state of affairs already to exist which ... the process of competition tends to bring about" (1948, 92). From the Hayekian perspective, the knowledge of both firms and consumers in real markets has to be characterized as imperfect, subjective, and fallible knowledge, in other words, the agents can never be sure whether their knowledge is correct, and therefore have to search for better knowledge. As a consequence, market competition primarily should be seen as a process of parallel experimentation, in which firms compete with different hypotheses (conjectures) about good solutions for the problems of the demand side and can learn from each other through imitation. The crucial point of this "discovery procedure" (Hayek 1978) is that only through the market test is the knowledge generated of which of the products or services of the competing firms are the superior problem solutions. In that respect, competition can be seen as a "test of hypotheses" (Kerber 1997), in which knowledge is generated and spread by imitation.
A more detailed analysis shows that the products the firms offer can be described as a complex bundle of traits of performance as, for example, the technical features and quality of the product, its image, service, warranties, price, and so on. Each firm has to …