Is Structural Separation of Incumbent Local Exchange Carriers Necessary for Competition?

In antitrust’s dWhen Congress passed the Telecommunications Act of 1996, it said that the legislation’s purpose was to “promote competition and reduce regulation in order to secure lower prices and higher quality services for American telecommunications consumers and encourage the rapid deployment of new telecommunications technologies.” Half a decade has now passed. The vision held by some of a radically different market structure for local telecommunications has thus far failed to materialize. Collectively, competitive local exchange carriers (CLECs) are gaining market share and accounted for 8.5 percent of access lines in the United States as of December 31, 2000. The New York Public Service Commission has reported that as of December 31, 2000 CLECs supplied 2.9 million lines in that state, for a market share of 20.9 percent. For the first time, more CLEC lines (52 percent) served residences than businesses (48 percent) in New York. Still, many entrants into the market for telecommunications have gone bankrupt or lost a substantial fraction of their market capitalization, and many telecommunications carriers now face substantial debt burdens. Between the fourth quarter of 1999 and the third quarter of 2001, at least twelve publicly traded CLECs experienced negative growth in earnings, and many filed for Chapter 11 bankruptcy protection, including WinStar, Northpoint, ICG, and Covad. ivision of labor, lawyers care a great deal about who wins and who loses cases and why. Economists are not disinterested in these matters, but they care as much if not more about what happens after an antitrust case has been decided or settled. Economists ask: what are the benefits and costs of the antitrust remedy?

The competing explanations for the difficulties encountered by many CLECs are numerous. In particular, however, some argue that the current regulatory strategy (or the current path of “managed competition” adopted in lieu of true deregulation) is not enough and that competitive local telephony will eventuate only if regulators require the incumbent local exchange carriers (ILECs) to place their wholesale and retail operations in structurally separate subsidiaries. In February 2001, the chairman of AT&T, Michael C. Armstrong, publicly advocated such intervention by state or federal regulators or by Congress. By the summer of 2001, regulators in Pennsylvania, New Jersey, and Florida had considered or begun proceedings on the subject, and Senator Ernest F. Hollings, chairman of the Senate Commerce Committee, had introduced the “Telecommunications Competition Enforcement Act of 2001,” a bill that would mandate structural separation of the ILECs.

In Part II of this article, we suggest that faulty business strategies are to blame for many of CLEC failures. Our hypothesis is consistent with empirical evidence that, despite the massive shakeout in the telecommunications sector, some CLECs have actually prospered. Next, we provide anecdotal and empirical evidence that supports our hypothesis. In particular, we review the business strategies of several CLECs and then correlate those strategies with success or failure. Entrants that deliberately built their own networks, carefully analyzing competition and consumer demand prior to entry, were able to increase revenues and continue to attract capital. An overly generous unbundling regime, which rewards CLECs for deferring investment, might be at the root of the CLECs’ problems.

Part III examines AT&T’S alternative hypothesis for failing CLECs and its proposal for mandatory structural separation. According to AT&T, anticompetitive behavior by the ILECs caused the CLECs to fail. Ostensibly to prevent discrimination against nonaffiliated retailers of local service, AT&T and some other CLECs urge regulators to separate structurally the ILECs into wholesale and retail companies. According to its proponents, structural separation would level the playing field between nonaffiliated local retail providers and the ILECs. We explore the meaning of mandatory structural separation, and lay out the purported merits of mandatory structural separation.

In Part IV, we critique AT&T’s diagnosis of local competition and its accompanying structural solution. Anticompetitive practices cannot explain certain market and regulatory phenomena. Despite allegations of anticompetitive practices by the ILECs, some CLECs are thriving. Indeed, we demonstrate that CLECs’ market share has steadily increased in the past three years. Moreover, during the same time period, state regulators have approved RBOC entry into long-distance services—an event that is not consistent with anticompetitive behavior. Next, we argue that mandatory structural separation is inefficient. Its likely costs would exceed any purported benefits. In particular, we examine several potential efficiency gains associated with vertical integration, including, among others, the coordination of investment and production decisions, accountability for product quality, and the ability to make bundled offerings. Mandatory structural separation would jeopardize each of these efficiencies. We conclude Part IV by explaining why mandatory structural separation is not an efficacious remedy under any diagnosis of the CLEC problems. First, there is no systematic evidence of discrimination. Second, behavioral restraints could prevent discrimination. Third, mandatory structural separation will not lower wholesale discounts to CLECs. Fourth, the experience with structural separation in several other contexts has been unsatisfactory or inconclusive.

Part V explains how mandatory structural separation can serve an ulterior motive: it can advance an anticompetitive strategy of AT&T and others to raise the ILECs’ costs of providing local telecommunications services. This strategy of raising rivals’ costs would ultimately increase costs for consumers and reduce investment.

Part VI concludes with the recommendation that policy makers reject proposals for mandatory structural separation of the ILECs.