Measurements of Market Power in Long Distance Telecommunications

The author is an economist with the Bureau of Economics, Federal Trade Commission. The views expressed in this study are those of the author and do not necessarily reflect the views of the Commission or any individual Commissioner. iv EXECUTIVE SUMMARY This study assesses empirically the competitiveness of the long distance telephone market. To do so, it estimates firm-specific long-run demand elasticities for AT&T and its rivals for long distance service marketed to households and small businesses during 1988-1991. A lower-bound for AT&T's long-run demand elasticity is estimated to be approximately-10.1. If AT&T's prices were completely unregulated, this elasticity estimate implies that the upper-bound deadweight loss due to allowing AT&T to set prices in excess of marginal cost would be about 0.36% of total industry revenues in 1991, or $199 million in 1991. While direct estimates of the costs imposed by the current form of regulation are not available, this welfare loss estimate is well below previous estimates of the benefits that followed partial deregulation of the long distance market. Measurements of a firm's demand elasticity provide information on the extent of its market power. In a perfectly competitive industry, each firm has the same costs, price equals marginal cost, and each firm in the industry faces a horizontal demand curve at that price. In such an industry, a firm that attempted to raise prices above its marginal cost would lose all of its customers to rival suppliers. In other words, firm-specific demand curves in perfectly competitive industries are infinitely elastic. However, in industries where products are differentiated or where firms have different marginal costs a firm setting prices above its marginal costs would lose some, but not all, of its customers. Its firm-specific demand curve slopes downward, and its demand elasticity is finite. In general, a particular firm will find that its firm-specific demand curve becomes more elastic as competitors' products become better substitutes and as competitors' costs fall. v Estimating firm-specific demand curves raises a number of specific analytic and econometric issues, each of which is addressed in this study. The first issue is the analytic structure imposed on those consumers demanding long distance service. This study assumes that residential and small business customers' long distance purchase decisions can be characterized by a two step approach. Given the general level of long distance prices, residential consumers and small businesses first decide how much long distance service to purchase. …