Thursday, August 21, 2008
Posted by D. Daniel Sokol
Debra J. Aron (LECG) and David E. Burnstein (LECG) discuss Regulatory Policy and the Reverse Cellophane Fallacy in the telecom context in their latest working paper.
ABSTRACT: A central concern of antitrust analysis is determining whether or not a firm possesses significant market power. The trend in the US and many other countries toward liberalized regulation or deregulation of telecommunications providers has been accompanied by the application of these antitrust concepts and principles to assessment of market power of regulated companies. Regulators are not inclined to deregulate markets without assurance that the incumbent's purported market power would be checked by competition. While economic principles of antitrust analysis certainly apply to the regulated setting, a proper economic analysis of whether a regulated firm has - or more accurately, would in the absence of regulation, have - market power is a significantly different exercise, however, from a typical market power analysis of an unregulated firm.
In this paper we show that applying the usual tools of market power analysis to firms in regulated industries can lead to predictably erroneous outcomes. Specifically, assessing whether a firm has market power by conducting the "small but significant non-transitory increase in price ("SSNIP") test in a regulated industry would lead to the reverse of what is referred to in the antitrust literature as the "cellophane fallacy." The cellophane fallacy occurs when assessing the market power of a monopolist by applying the SSNIP at the monopoly price. Because a monopolist will rationally increase price to a point where other goods or services become substitutable, the test would lead analysts to find more substitutability (broader product markets) than is warranted, and erroneously infer a lack of market power. In the regulated setting, prices that are set by regulatory fiat at below-cost levels would cause the opposite error: what we term the "reverse cellophane fallacy." The uneconomically low prices cause other services to appear to be weaker substitutes than they are and therefore lead to improperly narrow market definitions and erroneous inferences of market power. This in turn leads to a self-perpetuating nature of regulation, in which regulators insist on finding that the incumbent lacks market power before deregulating prices, while the artificially restricted prices lead to an erroneous inference of market power.
We test this hypothesis empirically by examining the relationship between a price-regulated service (local telephone service), and the regulated price in a sample of "markets" (incumbent exchange service areas) in a single state in the U.S. in 2004. We employ a probit regression to evaluate whether, and to what extent, a regulated incumbent's retail prices affect wireline competitive entry after controlling for the effect of the size of service area and the costs of service. We find an independent, statistically significant positive effect of the incumbent's retail price on competitor penetration, all else equal. Our findings indicate that the relative absence of competitive activity in rural areas should not be interpreted as evidence that the incumbent would be able to exercise market power in the absence of price regulation. Rather, a proper competitive analysis must control for, and eliminate the potential errors of inference caused by, the regulated retail price structure.