Why There is No Such Thing Like Transaction Costs, After All

Abstract

In recent decades the concept of transaction costs has been very successful in eco-nomic theorizing, yet there is still no workable approach to measurement. I demonstrate that the reason of this situation is twofold: The concept leads into an infinite regress, and under the conditions of an ideal-typical competitive economy all the transactions costs become transformation costs a la North, after all opportunities for profitable arbi-trage have been exploited (i.e. transaction cost become the transformation costs of the services needed to realize the transactions). Only in the latter case exact measurement will be possible. However, as long as individual capabilities and sets of information dif-fer across the set of economic agents, transaction costs cannot be measured, and in fact reflect a specific rent imputable to individual competitive advantages in the realiza-tion of transactions. The rent is a part of entrepreneurial profit that cannot be separated analytically. This assertion implies that one should not follow the standard approach of the New Institutional Economics which focuses on the environment and the object of transactions in order to explain, for example, the boundary between firm and market. Instead, the subject of transactions and her capabilities should be in focus, namely, the entrepreneur. Here non-economic explanations come into play in an essential way.