The AT&T consent decree: in praise of interconnection only.

• Author: Epstein, Richard A.
• Position: The Enduring Lessons of the Breakup of AT&T: A Twenty-Five Year Retrospective

1. THE BELL DECREE AND THE CORPORATIST MINDSET II. "DEREGULATION" IN A NETWORK INDUSTRY: DO GREENE AND BAXTER MIX? III. STRUCTURAL VERSUS CONDUCT REMEDIES IV. CONDUCT AND STRUCTURAL REMEDIES FOR NETWORK INDUSTRIES V. A SUMMING UP I. THE BELL DECREE AND THE CORPORATIST MINDSET

   In this short Article, I shall return to some of the issues that I addressed in my short book, Antitrust Consent Decrees in Theory and Practice: Why Less is More. (1) The basic theme of that book is that the success of consent decrees, and indeed the resolution of all large antitrust cases, follows a clear pattern. The more ambitious the decree, the worse matters are likely to turn out. The reasonable response therefore is to cut back on ambition in order to execute modest plans well. Perhaps the most vivid illustration of a consent decree process gone wrong is the breakup of AT&T. (2) Therefore, holding a conference that addresses the strengths and weaknesses of that decree twenty-five years later offers a propitious occasion on which to examine this fundamental restructuring of the telecommunications industry.

   At the time of its adoption, the 1982 decree was generally lauded as a rebuke to the old corporatist way of doing business. (3) That system of self-conscious industrial policy dominated New Deal thinking. Its operation rested on three legs. The first was a strong government willing to create and preserve monopoly profits. The second was a system of strong labor unions, bolstered by the protections of the Norris-LaGuardia Act (4) and National Labor Relations Act, (5) that shared in those gains. The third was a corporate and antitrust culture that blessed these accommodations. The corporate law did not see shareholder maximization as an exclusive goal, and thus fostered accommodations with labor and other political constituencies. In addition, the antitrust law often punished competitive behavior or insulated anticompetitive behavior from judicial scrutiny. This system offered a cozy comfort to all the participants, and it promised stability in institutional arrangements that could not have been achieved in a competitive market where new entry and exit would quickly erase the monopoly profits for both the firm and its union. (6) But the defenders of the system thought that they had created the vaunted stability long prized by regulators of all stripes, who seek to insulate their own preferred constituents from the vicissitudes that plague the rest of the world.

   That so-called stability is in fact an illusion by any system-wide measure. (7) Of course, the regulated industry and its constituents are protected against both price fluctuations and new entry. But the model is not sustainable in the long run for three reasons. First, its ostensible certainties cannot be replicated system-wide. Uncertainty is an inescapable feature of all complex social systems. The only question is who will be forced to bear its consequences. The effort to insulate one group from the uncertainty increases the level of uncertainty borne by everyone else in the system, for all the initial variation is now forced on that fraction of the economy that is denied the protection afforded selectively to the regulated industry. Thus, if economic circumstances move sharply, these parties will be forced to bear the price declines in their own industries and to subsidize the price rigidities in telecommunications. If voluntary markets will tend to equalize uncertainty at the margin for all sectors, regulation tends to force greater risks on certain groups to benefit others. The likely consequence is to prevent the outside groups from making their needed adjustments. They will be on the steep portion of their uncertainty cost curves. Yet, the regulated parties will be spared the initial amounts of uncertainty, which they could probably bear at far lower cost. The result is more uncertainty system-wide, all in the name of price and income stabilization.

   The second risk with these accommodations is that while they insulate the protected groups from a multitude of small shocks, they do not protect them from the few large ones that really matter. Constant levels of protection look good until a large event makes the position of the preferred players untenable. New entry from some unanticipated comer can topple the financial base on which these long-term accommodations rest. In telecommunications, that transformation started with the rise of MCI, founded in 1963, which offered a way around the Bell monopoly, chiefly by exploiting the then new technology of transistors, coaxial cable, and microwave technology. (8)

   Third, the members of any industry--steel, automobiles, tires, agriculture--can try to play the same corporatist strategy. Access to that combined strategy became easier once labor unions received statutory monopolies under the National Labor Relations Act, justified, of course, as a way to "stabilize" the wages and purchasing power of their members. (9) The agricultural price support systems similarly stabilized prices for farmers but forced the public at large to take the full risk of the price fluctuations by buying back excess production at inflated prices. It is no accident that labor and agriculture won exemptions from the antitrust law under section 6 of the Clayton Act of 1914. (10) Nor is it an accident that the broad expansion of the reach of the federal commerce power took place first in labor and then in agriculture. (11) State-wide regulation was subject to too much competitive pressure from other states to succeed. With each additional maneuver of this sort, system-wide uncertainty increases, representing yet another distortion of a system of state monopolies. What is good for the coddled industry is systematically bad for the public at large. Indeed, there is a grand prisoner's dilemma at work here. It would be better for all of these cartel-like structures to disband simultaneously, for the gains that any group got from its own protection were more than offset over time when firms in other industries successfully imitated their strategy. In the long run, no one wins as the social pie shrinks. And on the few occasions where that point was grasped, large moves toward deregulation were possible. Airline transportation was deregulated in 1978 (12) and surface transportation was largely deregulated in 1982. (13) It is not, therefore, entirely coincidental that the Bell consent decree dates to 1982, at the outset of the Reagan years.

   The conventional accounts of that decree treat it as yet another nail in the coffin of the corporatist strategy. In one sense, this proposition must be regarded as true, given that the architect of the system was the late Assistant Attorney General William Baxter, whose anti-corporatist sentiments are beyond dispute. But there is good reason in this instance to think that in some sense he did not quite shed the corporatist heritage. In part, this is perfectly understandable. The great "achievement" of the New Deal corporatist state was its uncanny ability to take competitive industries like agriculture and convert them into grotesque monopolies and cartels organized and propped up by the government. The most evident move in that direction was the decision of the U.S. Supreme Court in Parker v. Brown, (14) which exempted a state-run raisin cartel from the reach of the federal antitrust law on the ground that states were entitled to enact their own industrial policies unless the federal government had intervened to block their decisions. To be sure, the New Deal marked a huge expansion in federal power. Yet, at the same time, it must never be forgotten that it also led to an expansion of state power as well, so long as there was no clash between the two systems. And to reduce the likelihood of that clash, the Supreme Court created its "presumption against preemption," which to this day still increases the risk of dual systems of regulation. (15)

2. "DEREGULATION" IN A NETWORK INDUSTRY: DO GREENE AND BAXTER MIX?

   Start with this proposition: the provision of telecommunications services is not like the production and sale of raisins. Even if pure competitive markets are possible in agriculture, they are not possible in telecommunications, notwithstanding the hype in support of this assertion. Judge Greene, for example, lamented that AT&T was in the worst of both worlds because "the Bell System has been neither effectively regulated nor fully subjected to true competition." (16) This simple observation has powerful implications for the counterattack against the corporatist mentality. If there is no good competitive solution for networks, the strong pro-market intuition may be enough to make the consent decree feasible. It need not be strong enough to make it work, however, because the only choice is a different--and hopefully better--form of regulation, not a pure competitive market.

   Optimism, therefore, has to be tempered in light of the restricted set of feasible alternatives. The minimum requirement for a sound network is that each person situated at any point on the network be in a position to communicate with any other person situated on the network. That problem--universal access--is easily solved under the corporatist model that animated the old AT&T. If every system user is situated under one roof, there is no need to decide who should pay what fees to gain access to networks populated by independent firms and their independent customer bases. All the cross-subsidies are kept internal to a single umbrella firm so that ratemaking can be separated from the problem of establishing interconnections. So shielded, these hidden subsidies can be quite large because the parties who are subject to regulatory taxes usually have no place to go and, given the monopoly nature of the industry, no sure knowledge of the magnitude of the subsidies created.

   Removing this unified monopoly makes it hard to keep the cost subsidies...