The national information infrastructure (NII) vision embraces three communications components (exclusive of information libraries and other content repositories): internodal networks, access nodes, and end-user access to the nodes. Internodal networks and access nodes are currently subject to competitive supply. End-user access is the only NII component that remains monopolized virtually everywhere in the United States. Such access is costly to provide. Costs of new or improved end-user access facilities must be justified by usage in a two-way mode, in an environment that will probably evidence substantial price elasticity. Nevertheless, many risk-taking access providers are preparing the technologies with which to offer economic, reliable end-user access, confident that a strong market will develop. Still, the in-place, monopoly-provided, end-user access seems to be useful for a great many consumers. Increased consumer demand for the greater capacity and speed of digital technologies within the NII will be conditioned largely by the behavior of public policy makers vis-à-vis the entrenched monopolies. In this paper we discuss the opportunities for and obstacles to replacement of the monopoly for end-user access.
It has long been recognized by economists that a monopoly has less incentive to innovate than does a firm in a competitive market. Firms in a competitive market will seek to gain a competitive advantage over their rivals through innovation and differentiation, thus enhancing consumer surplus as well as producer surplus. Events in the long-distance and equipment segments of the telecommunications industry seem to have confirmed the result predicted by economic theory. In the inter-LATA long distance market, AT&T's market share has dropped from over 80 percent to less than 60 percent since 1984, and the number of carriers has risen to over 400 1. The facilities-based long distance carriers increased their deployment of fiber-optic transmission facilities fivefold between 1985 and 1993 2. Long-distance rates have fallen more than 60 percent 3. Similar gains have been realized in the demonopolized market for telecommunications equipment. Now handsets can be purchased for less than the price of a pair of movie tickets, and businesses and consumers are linking ever more sophisticated equipment (computer modems, PBXs, cellular phones, and handsets that have computer memory) to the public switched-telephone network.
Theoretically, the same benefits would accrue to consumers from competition in the remaining monopolized local exchange market as have been realized in the equipment and long-distance market segments (efficiency, diversity, innovation, and price reductions). But the grip of government-sanctioned monopoly remains powerful. Only slowly has it been understood that local exchange monopolies may temporarily drive down costs but never as far as would competition: They may innovate, but not as swiftly as competitors would; they may improve service quality, but not as readily or effectively as they would under competition. The full benefits of the NII require efficient exploitation of all telecommunications technologies, including photonics, coaxial cable, fiber optics, wireless, and even the existing twisted pair technologies, linking subscribers to one another via a "network of networks." A competitive market for local exchange services is essential to such efficient exploitation of all technologies.
Moving from theoretical concepts to the practical reality of creating local exchange competition requires certainty and flexibility: certainty to inspire investor confidence, and flexibility to respond to constantly changingand largely unknownmarket realities. Achieving balance between certainty and flexibility means opening up existing monopolized market segments as much as possible without threatening the long-term stability of competition itself. Both industry and public policymakers must take this challenge seriously if the NII promise is to be realized by 2000or ever.
Revenue for the local telecommunications market in 1993 exceeded $88 billion 4. Of that total, 99 percent was captured by mature, traditional local telecommunications carriers. The 1 percent of the market liberated by new entrants was composed almost entirely of private line and special access services; they have not yet begun to reach a mass market 5.
Still, competitive access providers are optimistic about the future of competition, building or operating networks in some 50 or 60 metropolitan statistical areas. A recent report by the Federal Communications Commission indicates that over 10,070,308 miles of fiber has been deployed in the United States by local, long-distance, and urban telecommunications providers, and shows that the rate of deployment by competitive access providers far exceeds that of the incumbent local exchange carriers 6. The largest competitive access provider, Teleport Communications Group (TCG), has installed SONET-based, self-healing, two-way ring networks capable of transmitting information at a rate of 2.4 gigabits per second. ISDN is provided over these networks, and TCG offers frame relay at up to 1.5 megabits per second and ATM (switched data) service at up to 155 megabits per second. With more than 167,314 miles of fiber throughout 22 networks and a strong switch deployment program, TCG is technically positioned to serve larger markets 7.
Cable television operators, whose networks now pass 97 percent of the nation's households and provide television service to more than 65 percent, are the obvious "other" end-user access provider 8. Cable companies are upgrading their distribution plant and must provide for switching to offer local exchange services. Experiments, such as the Motorola-TCI-TCG trial of residential service using radio frequencies over fiber-optic coaxial cable, will identify technological requirements. The promise of wireless subscriber loops has also drawn considerable interest from a wide range of industry participants. Bidders in the FCC's recently concluded Personal Communications Service (PCS) spectrum auctions committed more than $7 billion for licenses to build and operate local telecommunications networks 9. A recent Merrill Lynch report predicts that one of the successful PCS bidders would have a 5 to 8 percent penetration of the local exchange market by 2004 10. There is no shortage of potential entrants seeking to provide local exchange services. Meeting consumer demand will mean the employment of a variety of distribution systems, both broadband and narrowband, wireline and wireless.
However, it is not yet certain what customers want and when they want it. Though studies such as a recent Delphi study of industry executives and academics have projected 2010 as the outside year in which a mass consumer market will exist for network-based interactive multimedia products and services provided over switched broadband, market research is notably thin on the subject of what people will pay for "infotainment" or household management services 11. As with the case of previous telephone, television, computer, and audio products, much investment rides on the premise that supply will create demand.
Despite the limited deployment of broadband networks, consumers have not been especially hindered in their attempts to establish themselves as providers, as well as users, of information. Internet platforms and a plethora of electronic bulletin boards allow consumers to "publish" information available on demand by other consumers. Aside from providing the conduit over which the information travels, the network operator has no role in the content of the traffic speeding over its lines. The development of local network competition will only hasten the development of more information services and more gateways through which consumers can share information.
The prospects for local competition, therefore, depend in part on the perfection of new technologies, but more importantly on reduction of their cost. However, factors other than technology itself pose greater obstacles to deployment of competitive choices and must be dealt with expeditiously if investment is to accelerate. Before the various broadband and other end-user access technologies can be widely deployed, legal, technical, and economic hurdles must be overcome in every state. Furthermore, even though technical interconnection and interoperability of networks have been shown to be feasible without serious difficulties, many aspects of the seamless interoperability of competitive networks remain to be resolved. Among them are a number of portability and central office interconnection arrangements. Finally, economic interconnection of competing local networks has to be achieved so that new entrants can develop their own services and prices that maximize the revenues from their own networks.
Regulatory barriers remain the threshold barrier to entry into local telecommunications services. Only seven states have authorized or permit competitive local exchange carriers to provide service: New York, Illinois, Maryland, Massachusetts, Washington, Michigan, and Connecticut. Making competition legal means removing a series of entry barriers imposed by different government agencies whose interests are not necessarily congruent. The primary issues that must be resolved include exclusive franchises, qualifications for entry, franchise fees, and access to rights-of-way and to buildings. Market entry that is conditioned upon demonstrating that the incumbent carrier is providing inadequate service is an unreasonable burden, as are franchise fees or other payments to local government as a condition of operating, if the incumbent does not pay equivalent fees. But the most difficult ongoing issues will be the access issues, because that is the area where the incumbent monopoly has the ability and the incentive to encourage unequal access. Incumbent LECs enjoy generally unlimited access to public right-of-way and often control the rights-of-way needed by entrants. They also have established access to building entrance facilities, at no cost. New entrants must have access to those rights-of-way at the same rates and on the same terms and conditions as the incumbent.
In 1995, 13 states enacted legislation removing barriers to entry and endorsing local exchange competition. Federal legislation preempting state entry barriers and setting guidelines for local competition is again under consideration. The technical and economic aspects of network interoperability are rising to the forefront of public policy issues that will condition NII development.
At the end of the nineteenth century and continuing into the early years of this century, local exchange "competition" did thrive. But rather than a "network of networks," consumers faced a tangle of discrete networks, all of them talking at once but none of them talking to each other. To be certain that they could reach everyone with a phone, customers had to subscribe to the local exchange service of every provider.
As that experiencewhich gave way to government-sanctioned monopolydemonstrated, the key to an efficient and effective "network of networks" is interconnection. Adjacent (i.e., "noncompeting") carriers interconnect seamlessly with each other today and have done so for more than 80 years. Now competing local exchange networks must be able to interconnect with the incumbent local exchange network on the same terms and conditions as the incumbent interconnects with adjacent carriers and with itself. At a minimum, the technical interconnection arrangements should include the following.
Central office (CO) interconnection arrangements are the physical facilities that connect a competitor's network to the local exchange carriers' (LEC) network. An efficient cost-based CO interconnection arrangement consists of three elements:
Although the long-term NII vision presumes that existing copper pair will be replaced (or upgraded) by broadband or at least enhanced narrowband access, a transition period during which local exchange competitors are maximizing their return on other network investment (in internodal transport and access nodes) will occur. These competitors will be able to offer service to a mass market during the interim period only by reselling existing subscriber loops under monopoly control. Currently, whether providing single-line basic local exchange service or multiline local services such as Centrex, LECs usually bundle together the "links" and the "ports" components of the loop. LECs must unbundle the local loop elements into links and ports, individually pricing each element at cost-based rates.
A LEC's tandem switching offices and end offices are interconnected via the LEC's interoffice transmission network. Adjacent LEC switching offices are also interconnected via the same network for the distribution of intra-LATA traffic. Routing through the interoffice network is governed by a local exchange routing guide (LERG) unique to each LATA. The LERG prescribes routing of all traffic between and among all LEC and interexchange carrier (IXC) switching offices. Cost-based interconnection arrangements at tandem offices as well as end offices are necessary, and the Class 4 and 5 switches of local competitors must be incorporated into the LERG on the same basis as (and with the same status as) the LEC's end and tandem offices. This will enable a competing local exchange carrier to efficiently and effectively route calls originated by customers on its network to the LEC for final termination at the premises of a LEC customer, and vice versa. This includes access to Signaling System 7 (SS7) with a line information database (LIDB). In an SS7/LIDB environment, routing, translation, service, and account information is stored in centralized databases that LEC and IXC switches can access through service control points (SCPs). Local competitors must be able to interconnect and query the LEC SS7 databases at the SCPs in a nondiscriminatory manner and under equitable terms and conditions.
Competitive local service providers must be allowed to have their customers' telephone numbers included in telephone directories, directory assistance, LIDB, advanced intelligent network (AIN), 800-number, and other databases. Their access to such resources must be equal in price, functionality, and quality to those of incumbent local telephone providers.
Numbering policy must be broadly developed and administered in a competitively neutral manner. The administration of area codes provides an excellent example of the problems that can arise under LEC-administered code assignments. In recent years, as competition has developed, the incumbent LECs have proposed a deviation from their historic practice of assigning customers to a new area code according to geography when the existing area code numbers were approaching depletion. That is, customers in one part of the existing area code territory would retain their area code, while customers in an adjacent area would be assigned a new area code. The new approach favored by some local exchange carriers is to assign a different area code to customers of the new providers (wireless and competitive access providers), allowing only the LEC's customers to remain undisturbed by the new area code assignments. The anticompetitive effects of such a plan are not hard to imagine: customers of the new carriers are difficult to locate, and only customers of the new entrants must incur the expense of changing their numbers (e.g., letterhead, business cards, advertising). Such expenses could be a significant deterrent to a customer who might have found it otherwise economical to switch to a new entrant provider.
The ability of customers to change service providers and to retain the same local telephone number at the same location (service provider number portability), without having to dial extra digits and without being burdened by "special" actions, is a critical component of the economic viability of local exchange competition. Interim number portability mechanisms (such as remote call forwarding) are an inferior form of number portability that impairs a new market entrant's service, and such impairment should be reflected in interconnection charges.
Local exchange telephone companies maintain a detailed set of administrative, engineering, and operational practices and procedures. These coordination practices must extend to new local competitors if competition is to emerge. In addition, the traditional local telephone companies maintain a detailed set of practices governing revenue pooling, intercompany settlements, and other financial interactions between and among established local carriers. These also must be extended to local competitors. Finally, any other intercompany notification procedures, standards-setting procedures, and the like, through which LECs may now or in the future interact with one another, must be extended to include local competitors. These requirements may appear reasonable, but the task of integrating such behaviors into huge corporations with many employees and a monopoly mentality will be daunting.
The ultimate determinants of the amount and extent of local exchange competition will be neither legal nor technical, but economic. The ability of competitors to meet consumer demand depends ultimately on economic interconnection arrangements for competing local networks. The new entrant must install capacity to accommodate peak-hour traffic, even though initially it has little inbound traffic. Arrangements for reciprocal compensation for the mutual exchange of local traffic should allow both carriers to recover the incremental cost of capacity. This cost is low: about 0.2 cents per minute, on an average basis 12. Given the trivial cost of supplying incremental capacity, it makes practical sense to implement a "sender-keep-all" arrangement, like that now used by Internet providers, rather than to impose explicit charges on terminating carriers. Sender-keep-all is also administratively simple.
Such arrangements should minimally be in place until data-based number portability is established. The lack of number portability and the substitution for it of measures such as remote call forwarding impair the new entrant's ability to use its network to its full capabilities and to sign up customers. Remote call forwarding and other so-called interim number portability measures entrench the incumbent's position and give it an economic advantage because all calls are routed through the existing LEC's network. Under current regulation, this means that the LEC as intermediary could collect terminating access charges from interexchange carriers for calls destined to a customer on a competing local exchange carrier's network. Although arrangements to overcome this economic inequity might be made, they would be complex and costly. It is far more efficient to plan for and expeditiously implement a long-term solution to the number portability problem. This must be a data-based solution that does not put any telecommunications service provider at a disadvantage and allows all providers to maximize the utility of their networks.
The long-held policy objective of universal service must be addressed before competitors can compete effectively throughout mass markets for local telecommunications services. This will make it possible for efficient competition to eliminate uneconomic or noncompetitive subsidies embedded in telecommunications pricing structures over a reasonable transition period. This is an expected result of competition. Incumbent LECs argue that retail rates (especially for residential consumers) may be priced substantially below costs and will be subject to upward price pressure. But again, as distribution networks achieve economies of scale, incremental costs are expected to fall, mitigating such price pressures.
Nevertheless, social policy pressures to maintain average prices for similar offerings across the country and between urban and rural subscribers will constrain providers' pricing practices for a transition period of at least 10 yearsuntil competitive facilities-based distribution networks reach a substantial portion (say 30 to 40 percent) of rural subscribers. During this period (and for certain communities, perhaps permanently) universal service must be assured. The ability to obtain access to the public switched network voice-grade service with tone dialing on a single-party line, at affordable rates, together with operator assistance, directory listings, and emergency service, is the minimal level of residential service and may require subsidies for some customers. If public policymakers seek to mandate a higher level of service, such as broadband access, the costs will soar; currently there seems to be a political consensus that market forces should define what "basic" service requires subsidy.
Subsidies to preserve universal service must be explicitly identified and administered through a provider-neutral fund so as to minimize their cost and maximize their efficiency. All telecommunications service providers should contribute to the subsidy fund based on their share of the market. Subsidies should be portable among carriers; that is, all local exchange carriers must have the opportunity to receive a subsidy when selected as a service provider by a subsidized customer.
The prospects for local competition are improving. In 1995, 13 states enacted legislation making local exchange competition possible, and most of these laws directed state regulators to provide for the requisite technical and financial interoperability and equal access. In nine other states, the critical issue of reciprocal compensation for the mutual exchange of local traffic is being addressed and interim agreements have been concluded between new entrants and LECs. Up to 15 additional states will soon develop rules for local exchange competition. Federal legislation setting the standard for local exchange competition and giving all major industry groups something of what they want in a competitive environment has passed the Senate; many observers give it at least a 50 percent chance of becoming law in 1995. Three trials designed to assess the technical requirements for telecommunications number portability are under way. And activity in the courts has sharply accelerated as parties seek to end existing legislative or judicial curbs on their ability to address each other's markets.
Among the restraints under attack are the line-of-business restrictions on the Bell operating companies (BOCs) entering the interexchange market or manufacturing, included in the consent decree that divested the BOCs from AT&T. These restrictions were intended to endure so long as the divested regional BOCs (RBOCs) retain control of the local exchange bottleneck, which gives them the ability and incentive to impede competition in adjacent markets. The "VIIIc" test applied by the court overseeing the decree requires the U.S. Department of Justice to report that there is "no substantial possibility" that a BOC seeking relief from the decree would be able to impede competition in the adjacent, restricted market. Some members of Congress seek through legislation to abrogate the decree entirely, eliminating the role of the DOJ, without evidence that the local exchange bottleneck has actually been broken. Others would substitute a more liberal testthe "public interest" testfor VIIIc and make the FCC rather than the DOJ and the court responsible for its application. Still others would have the DOJ involved in an advisory capacity. And some, like the Clinton Administration, prefer to retain the Justice Department's role 13.
Several BOCs are seeking relief from the decree directly from the court. On April 3, 1995, the DOJ filed a motion to permit a trial supervised by the DOJ and the court in which Ameritech could provide interexchange service after the development of actual competition, including facilities-based competition, and substantial opportunities for additional competition in local exchange service 14.
The questions of how much and what type of competition warrants relief will be the key to whether or not durable and robust local exchange competition is possible. Even if all the "paper" requirements for local exchange competition are met, the incumbents can and do manipulate each and every aspect of access to their essential facilities. As BOCs have demonstrated frequently, they can frustrate and thwart competitors in numerous ways: by imposing different and unnecessary physical interconnection requirements for competitors; by engaging in pricing discrimination, tying arrangements, and selective cost de-averaging; by imposing excessive liabilities on customers who wish to change to a competitive carrier; by engaging in sales agency agreements that effectively foreclose markets to new entrants; and more. As a result, a public policy that seeks the full benefits of competition for the greatest number of telecommunications customers must include safeguards that constrain anticompetitive behavior and sanction the parties having market power that engage in it. A rigorous effective competition test prior to BOC relief is such a safeguard.
The other required safeguard is continued regulatory oversight of the firms having market power, with lessening intervention as that market power diminishes. During the development of interexchange market competition, AT&T, the dominant interexchange carrier, endured economic regulation while its competitors were free to price their services to the market. A similar approach is necessary in the local exchange market. Incumbents may be granted pricing flexibility, so long as they do not reduce prices of competitive services below their incremental cost; but cross subsidies from captive customers must be prevented. New entrants must face minimal regulatory burdens because, lacking market power, they cannot harm the public interest. However, new entrants must contribute to the maintenance of universal service.
By any measure, local competition does not exist today, but a more competitive environment is being created. Full-scale competition will require huge investments, which in turn calls for relative certainty as to the pace and outcomes of regulatory and legislative actions. If new entrants have technological innovations to exploit, their ability to complete cost-effective deployment depends on their relative financial strength vis-à-vis the incumbent monopolies. The latter often can manipulate financial faucets by means of their negotiated agreements with state regulators, whereas competitors depend entirely on risk markets for their capital.
Competitors are engaged in a historic process on the legislative, regulatory, and judicial fronts. There is no doubt that the barriers to entry will fall. The safeguards to control residual market power as local telecommunications competition emerges will be lodged in different government agencies, as they currently are, but the guidelines will become clearer as the benefits of competition become more obvious and more widespread.
The final determinant of the pace at which these policy adjustments occur is consumer demand. Demand will surge when "infotainment" is varied and priced to mass market pocketbooks.
1 Industry Analysis Division, Common Carrier Bureau, Federal Communications Commission. 1995. "Long Distance Market Shares: Fourth Quarter 1994," Federal Communications Commission, April 7.
2 Industry Analysis Division, Common Carrier Bureau, Federal Communications Commission. 1994. "Fiber Deployment Update: End of Year 1993," Federal Communications Commission, May.
3 "FCC Monitoring Report, Table 5.5." CC Docket No. 87-339, June 1995.
4 Federal Communications Commission. 1994. Statistics of Communications Common Carriers.
5 Connecticut Research. 1993. "1993 Local Telecommunications Competition . . . the 'ALT Report.'" Connecticut Research, Glastonbury, Conn.
6 Industry Analysis Division, Common Carrier Bureau, Federal Communications Commission. 1994. "Fiber Deployment Update: End of Year 1993," Federal Communications Commission, May.
7 It has been announced that TCG will be contributing to a new venture of Sprint, Tele-Communications, Inc., Comcast Corporation, and Cox Cable that will package local telephone, long-distance, and personal communications with cable services into a single offering for residential and business consumers.
8 National Cable Television Association. 1993. "Building a Competitive Local Telecommunications Marketplace," National Cable Television Association Position Paper, October.
9 Federal Communications Commission Bulletin Board, March 13, 1995. Wireless Co., a consortium formed by cable television operators TCI, Cox, and Comcast, and long distance carrier Sprint, successfully bid over $2 billion for licenses in 29 markets.
10 Merrill Lynch & Co. 1995. "The Economics of the Sprint/Cable Alliance," Merrill Lynch & Co., February 10.
11 Kraemer, Joseph S., and Dwight Allen. n.d. "Perspectives on the Convergence of Communications, Information, Retailing, and Entertainment: Speeding Toward the Interactive Multimedia Age," Deloitte Touche Tohmatsu International, p. 13.
12 Brock, Gerald. 1995. "Incremental Cost of Local Usage," prepared for Cox Enterprises, March, p. 2.
13 See S. 652, "Telecommunications Competition Act of 1995"; H.R. 1555 Discussion Draft, May 2, 1995, "Communications Act of 1995"; and H.R. 1528, "Antitrust Consent Decree Reform Act of 1995."
14 United States of America v. Western Electric Company, Inc. et al., and American Telephone & Telegraph Company, Civil Action No. 82-0192.