The Bell Doctrine: applications in telecommunications, electricity, and other network industries.

• Author: Joskow, Paul L.
• Position: Baxter Symposium

In this article, Professors Paul Joskow and Roger Noll examine Professor Baxter's most "visible" contribution to antitrust policy: the Bell Doctrine. The Bell Doctrine, so named because it underpins the 1983 break up of AT&T into the Baby Bells, can theoretically be applied to other network industries where regulated monopolies control both monopoly segments and potentially competitive segments of the industry. The primary purpose of the article is to examine the economic theory of the Bell Doctrine and to evaluate the current applicability of the Bell Doctrine in light of changes that have taken place in regulated infrastructure industries. After describing the development of the Bell Doctrine and reviewing its application in the U.S telecommunications industry, the authors survey reforms in other industries and in other countries. The recent reforms in the electric power industry are closely examined as a case study. Professors Joskow and Noll conclude that the Bell Doctrine continues to dominate the policy analysis of reforms in network industries across the globe, even if easy application of the Bell Doctrine's principles and proposed remedies is no longer possible.

William Baxter's contributions to antitrust analysis and policy are numerous and impressive, but none is more visible than the divestiture of the American Telephone and Telegraph Company (AT&T) in the settlement that terminated United States v. AT&T.(1) The fundamental theory underpinning the Modification of Final Judgement (MFJ), a theory named the "Bell Doctrine" by Professor Baxter,(2) is that regulated monopolies have the incentive and opportunity to monopolize related markets in which their monopolized service is an input, and that the most effective solution to this problem is to "quarantine" the regulated monopoly segment of the industry by separating its ownership and control from the ownership and control of firms that operate in potentially competitive segments of the industry.

At the time of AT&T's divestiture, the vast majority of telephone subscribers had no reasonable alternative to monopoly local telephone companies for obtaining access to the national telecommunications network and for acquiring most telecommunications equipment and services, including ordinary direct dial long distance. According to the Bell Doctrine, local access monopoly permitted the only fully integrated telecommunications companies, AT&T and GTE, to monopolize long distance service, information services, and equipment that is used by telephone companies (e.g., switches) and customers (e.g., telephones). The point of the AT&T divestiture was to eliminate the incentive and opportunity of Bell Operating Companies (BOCs) to use their control over local access to monopolize these other potentially competitive markets. The expansion and enhanced enforcement of regulatory rules requiring AT&T to give competing suppliers of long distance service access to its local networks was viewed as an ineffective public policy alternative to full separation of the ownership and control of potentially competitive segments from what were then thought to be durably monopolized segments. Moreover, the benefits of enhancing competition by requiring divestiture were assumed to be significantly greater than any economies of vertical and horizontal integration that AT&T argued would be lost as a consequence of vertical and horizontal divestiture.

The Bell Doctrine theoretically applies to a variety of other industries where incumbent, vertically integrated, regulated monopolies control both monopoly segments and potentially competitive segments. Among these industries are electricity, natural gas, railroads, and other transportation sectors. Although each is unique in some important ways, all have several significant similarities. Each industry is constructed as a large (sometimes national) network for transporting its fundamental product (information, energy, cargo) between any two feasible points of origin and destination. Moreover, each network is conceptually separable into several basic functions: local distribution networks that provide service to end-users, long distance links among the local distribution networks, assemblage or creation of the fundamental product, and management of the operation of the system so that the transportation function can be efficiently undertaken. In each case, some of these components are provided by independent companies or government enterprises. In every case, some transactions are fully handled by a single company, while other transactions are handled by multiple companies that successfully coordinate in providing an integrated product. For example, originating, switching, and terminating a local telephone call is usually handled by a single company, whereas a long distance telephone call may be handled by three or more separate companies. Likewise, a single railroad may pick up, transport, and deliver a shipment, or a shipment may be arranged by an independent freight forwarding company while being physically transported over several railroads and even, for part of the journey, a trucking company.

Perhaps most significantly for the policy debate, each of these industries has elements that arguably are "natural monopolies" in two senses. First, some specific components of the underlying technology exhibit economies of scale over a range of output or capacity that is comparable to the magnitude of demand. The best examples are local distribution networks in electricity, gas, and telephone service, and the long distance links in gas pipelines and rail transport between some origin and termination points. Second, as networks, these industries require some standardization and coordination throughout the entire interconnected system in order to operate efficiently. Examples are standard gauges in railroads, standard current characteristics in electricity, standards for the physical composition and pressure of natural gas, and standards for the physical characteristics of information streams in a telecommunications network.

The fact that components of the network have natural monopoly characteristics does not necessarily imply that any part of the industry ought to be a monopoly. With respect to scale economies, the physical links that exhibit declining average costs per unit of capacity may not be the most important element of the cost of providing service. Moreover, if customers desire different qualitative attributes in their service (such as speed, reliability, or special service features), they may prefer two or more product-differentiated firms with somewhat higher costs and prices than one firm that otters a homogeneous product that is not the most desirable service for anyone. With regard to standards and coordination, the alternative to monopoly ownership is some sort of industry-coordination organization. Examples area standards regulatory agency (such as the Federal Communications Commission with respect to the technical characteristics of broadcast signals), or an industry joint board for coordinating technical standards and operation (such as the IEEE committees that set standards for computers and related new information technology products and the National Electric Reliability Council (NERC) which sets standards for the operation of interconnected high voltage electric power networks). The relevant policy issues pertaining to natural monopoly in all cases are whether the scale economies and coordination requirements are so overwhelmingly important and complex that they require a monopoly in at least some component of the industry, taking into account the reality that monopoly leads to the necessity for economic regulation that may cause other inefficiencies.

The purpose of this article is to set forth the basic economic theory of the Bell Doctrine and to explore the extent to which it remains applicable to telecommunications and these other infrastructural industries. This exploration is of more than historical interest. Since the AT&T divestiture, regulated infrastructure industries have changed in three major ways. First, in some cases regulation has been eliminated or vastly weakened, so that a fundamental premise of the Bell Doctrine--that regulation of the monopoly segments of the sector give regulated firms profit incentives to foreclose competition in the competitive segments--may not be true. Second, where regulation remains, many regulatory agencies, both federal and state, have changed the method for regulating prices and imposed requirements for network access. These changes raise the question of whether the new forms of regulation sufficiently diminish the incentive or ability of a firm to engage in anticompetitive vertical foreclosure so that the remedies component of the Bell Doctrine is no longer applicable. Third, technological progress has been quite dramatic in some industries. This progress raises questions about whether the monopoly segments of the industry, if they ever were a natural monopoly, still are effectively insulated from competition. In addition, technological changes that have opened opportunities for competition in some segments simultaneously may have made the neat structural remedies that were applied to AT&T more costly and difficult to implement. For these three reasons, some analysts have come to doubt that the monopoly segments can still be used effectively for anticompetitive purposes without inviting either successful competitive entry or detection and prosecution of anticompetitive activities through regulatory and antitrust sanctions.

Our fundamental conclusion is that whenever a monopolist enters a vertically-related, competitive market, controversy over whether the monopolist is behaving anticompetitively is virtually terrain to arise. If the monopolist is regulated, these controversies are likely to cause unending regulatory disputes and ongoing regulation of the...