This paper presents an empirical analysis of the value of a coordinated market exchange mechanism. I present a model of efficient trading mechanisms under uncertainty, and develop a measure of the value of an interfirm trade agreement in the context of sequential ‘buy-versus-produce’ decision-making by firms. The theory is applied to estimate the value of a formal trading institution in the California electricity market, where an interutility power pool has been proposed to restructure the electric power industry I develop an empirical model of the optimal production and trading decisions for a firm in such a pool, and estimate state-contingent willingness-to-trade functions for each of the four major utilities in this market. With this information, I estimate that distribution of future costs that would obtain if an efficient exchange mechanism arbitraged away observed differences between willingness-to-buy and willingness-to-sell among the sample firms. The principal finding is that with the simple, relatively state-independent bilateral contracts observed in this market, the sample firms achieve within 4% of the theoretical minimum expected costs available with a complete state-contingent exchange mechanism. This difference represents an opportunity cost of approximately $250 million per year. I conclude with regulatory and managerial explanations for the absence of a more efficient state-contingent trading mechanism, and implications for deregulating electric power markets.