Justice Stephen Breyer, AT&T v. Iowa Utilities Board1
The Federal Communications Commission (FCC) is expected to vote this week on new rules governing competition for local telephone service. To the average consumer, the debate on this issue probably seems obscure and impenetrable--filled with technical jargon and acronyms. Yet the consequences of this decision may determine not just the future of competition in this market, but prospects for new technology and for the U.S. economy as a whole.
At its core, the issue presents a stark choice between two very different visions of competitive markets. The first--embodied in the current Clinton-era rules--relies on regulations to require incumbent providers to share parts of their network with potential competitors. The second approach would encourage the development of competing networks with a minimum of regulation. Despite loud protests by many firms that now benefit from the current, regulation-based policies, the second path offers consumers better prospects for real choice and greater availability of advanced telecommunications services.
Specifically, the FCC should rewrite its rules so as to limit mandated unbundling to those network elements that are true bottlenecks. In addition, new investment in infrastructure for high-speed services should not be subject to regulation. Finally, the commission should ensure that state and local regulators do not nullify its decisions by imposing additional regulation on U.S. telecom markets.
The roots of the current debate go back seven years to the Telecommunications Act of 1996. One of the goals of that legislation was to open local telephone service to competition, much as had been done in the long-distance telephone market in the 1970s and 1980s. As a first step, the legislation stripped local telephone companies, known as local exchange carriers (LECs), of their legal monopoly on telephone service, overturning state rules that made it illegal to compete with an incumbent LEC. But it went farther: providing for rules requiring LECs to make parts of their networks available to potential competitors.
The basic idea behind this "unbundling" requirement was deceptively simple: While many firms would be able to compete with the formerly monopoly LECs in many or most aspects of their business, the LECs still controlled a number of key network elements that could not be economically duplicated and without which a competitor could not operate effectively.
The 1996 act left to the FCC the job of defining which of these network elements would be subject to unbundling. It required, however, that the commission consider whether access to an element was "necessary" to a competitor and whether its absence would "impair" the competitor's ability to compete.
Taking an expansive view of its authority, the FCC in late 1996 adopted extremely inclusive rules specifying a large and detailed set of unbundled network elements (UNEs). The UNE list included everything from wires to individual users and switching equipment to directory assistance and telephone operators. It also permitted competitors to lease what is known as the UNE-platform (UNE-P)--basically the entire LEC network--as if it were one element.
The LECs cried foul, appealed the decision in court, and won. In a 1999 decision, the U.S. Supreme Court struck down these rules, rebuking the FCC for ignoring the "necessary" and "impair" wording in the statute.2
Later that year, the FCC issued revised rules on unbundling. Except for a few minor changes, including the elimination of operators and directory service from the list, these new rules looked remarkably like the old rules. The issue went back to the courts, and in May of last year, the rules were once again struck down. In a strongly worded decision, the D.C. Circuit Court of Appeals found that the commission had again failed to consider meaningfully the "necessary" and "impair" language.3 Instead, the FCC in effect merely looked at whether mandating access to a particular part of the network would help competitors--in violation of both statutory language and economic common sense. The court also criticized the FCC for defining elements nationally, with no examination of the economics of each local market.
The FCC, under the chairmanship of Bush Administration appointee Michael Powell, is now set to take another bite at this apple, with a third version of the unbundling rules expected soon. The decision is a critical one, and the issue is complex.
The public debate on the topic, however, has too often been oversimplified. As framed by many, the choice for the FCC is simply competition or monopoly, with any lessening of the current unbundling rules portrayed as a deathblow to competition. This storyline is a tempting one, nicely fitting a classic black hat-white hat scenario of spunky small competitors versus monopolistic LECs.
The reality, however, is much different. Competition in local phone service, far from being threatened, is in fact quite healthy, and for reasons that have little to do with the FCC's unbundling rules. Revisions of the rules, in fact, could actually increase real competition. The most critical impact of the commission's decision, meanwhile, may have less to do with competition in old-fashioned telephone service than with the development of advanced new technologies and services.
In the first few years after passage of the 1996 Telecommunications Act, a large number of challengers to incumbent LECs entered the marketplace. In 1999, some 300 competitive local exchange carriers (CLECs), with a total capitalization of some $86 billion, were operating in the United States.4 The next few years, however, brought a wave of bankruptcies and liquidations. Today, only 80-100 survive, with a total capitalization of some $4 billion.5 The biggest names in the industry, including Covad, NorthPoint, Teligent, and Winstar, have filed for bankruptcy protection.6
Given these numbers, one might think that local phone competition is on the rocks, with the incumbent carriers holding a steady grip on their monopoly position. In fact, the opposite is true: Despite massive consolidation, competitors hold more of the local phone service market than ever before. According to a report just released by the FCC, as of June 2002, CLECs held 21.6 million--11.4 percent--of all telephone access lines.7 That is a 10 percent increase over the beginning of 2002 and the highest CLEC share ever.
These numbers, however, tell only part of the story. Most notably, they do not include competition from cellular phone and other mobile wireless providers. Once purely a supplementary telephone service and too expensive to use regularly, wireless phones, because of their price and functionality, are becoming a viable substitute for wireline phones. The number of subscribers is vast--some 129 million, a figure approaching the 189 million wireline lines in service.8 And while only about 6.5 million Americans rely exclusively on their wireless phones, with no wireline subscription, some 18 percent now consider their wireless phone to be their primary phone line.9 Most important, even for consumers who do not currently rely on wireless, it serves as a vital check on the market power of wireline incumbents.
Ironically, cellular's competitive success story was not contemplated by the 1996 Telecommunications Act, which barely references mobile wireless services at all, and the service has rarely figured in subsequent efforts by regulators to foster local competition. Instead, after the assignment of spectrum to wireless providers, Washington largely left them alone. In other words, despite the years of squabbling over how to create competition, the biggest success has come from the marketplace, not from regulatory policymaking.
Significantly, wireless firms provide service largely using their own infrastructure, generally without leasing network elements from incumbent LECs,10 but the same cannot be said of more traditional competitors to the incumbent LECs. While CLECs serve 21.6 million lines, only 29 percent use their own local loops.11 Close to half of these are cable companies offering telephone service through their coaxial cable lines. Of the remainder, most use loops obtained through UNE regulations,12 while 20 percent of the total are merely "resold" lines, with little infrastructure owned by the competitive provider itself.
This is the Achilles' heel of local telephone competition today. When competitors are merely leasing facilities owned by others, they provide less of a real choice for consumers. The nameplate may be different, but ultimately, to the extent they lease their product from the incumbent provider, they offer the same old product to the consumer. As Justice Breyer wrote in AT&T v. Iowa Utilities Board, "A totally unbundled world...is a world in which competitors would have little, if anything, to compete about."13 The local competition challenge for policymakers, therefore, is not to increase the total number of brand names from which they can choose, but to maximize the number of facilities-based telecom providers offering them service.
With this in mind, it appears that the concerns of critics are misplaced: The current comprehensive unbundling rules are not only unnecessary, but could actually be hindering the development of more real (i.e., facilities-based) competition. Perhaps more important, beyond their effect on local telephone competition, the rules actually could be hindering progress in other markets--particularly broadband telecommunications--with far more serious consequences for consumers and the U.S. economy.
Last year's D.C. Circuit Court of Appeals opinion, written by Judge Stephen Williams, was based on an economically sensible reading of the text of the 1996 act: The FCC was required to consider whether forced access to each element was "necessary" to competition and whether the unavailability of an element would "impair" the ability of a potential competitor to provide service. The terms, the court wrote, cannot mean only that it would be more expensive or more difficult for a firm to compete if it did not get access to an element.14
Such a reading would be grossly overbroad. Firms in almost any industry--even fully competitive ones--might find it cheaper and easier to use their rival's assets than to build their own. Anyone who starts up a burger stand, for instance, would likely find it cheaper to lease Burger King's kitchens. With thousands of restaurants, Burger King's costs are probably lower, and the competitor would benefit from that. That does not mean, however, that forced access is necessary to achieve competition in the burger market.
Congress, the court reasoned, must have meant a tougher test--especially in a supposedly deregulatory statute. The obvious situation is one in which economies of scale are so strong that it would not make sense to duplicate the asset. A classic example is a bridge over a river: The bridge is essential, but two bridges may be economically unsupportable. The owner of the bridge would then have what is known as a bottleneck facility and, under traditional antitrust regulation, could be compelled to share it.15 While the court did not explicitly require the FCC to apply the same rules used under antitrust regulation, it was clear that simply reducing a competitor's cost is not enough to require forced unbundling.
Harm to Local
Beyond being unnecessary, the current rules actually could hinder competition. The availability of UNEs from incumbent carriers at low regulated prices distorts the "make v. buy" trade-off for competitors, making it more likely that they will lease vital equipment rather than make the expensive--and risky--investment necessary to develop their own facilities.16
The rules also can discourage investment by incumbent providers. Quantifying the effects is difficult, but a recent study conducted by the Cambridge Strategic Management Group (CSMG) and commissioned by Corning, Inc., concluded that UNE rules have reduced the availability of high-speed fiber optics. The authors of the study calculate that while 5 percent of homes would have a fiber connection by 2013, 31 percent would have one if UNE rules did not interfere. Overall incumbent investment, CSMG calculates, would be some $39 billion less than it would be without the rules.17
The harm caused by the current rules, however, goes well beyond competition for traditional telephone service connections. The greatest danger may be to the development of tomorrow's high-speed Internet connections.
High-speed, or broadband, telecommunications allows users to receive and send information over the Internet at many times the speed of a standard telephone connection. Broadband service can be delivered in a variety of different ways, including cable modem service through cable TV networks, digital subscriber line (DSL) service through telephone networks,18 or several wireless and satellite technologies.
The potential impact of broadband technologies is huge. By speeding up the rate at which users can transmit information, a host of new applications and services, ranging from high-definition video to Internet-assisted medical care, become practical.
Aside from the qualitative consumer benefits involved, many observers see broadband as a critical catalyst for reviving the Internet economy, and perhaps the U.S. economy as a whole. The numbers could be large. One study estimated that the total potential benefits to the economy from universal diffusion of broadband service could be nearly $400 billion.19
Broadband service is growing rapidly. An estimated 15 million Americans subscribe to some form of broadband. Cable firms are the leader in broadband service with 9.4 million subscribers, and LECs are a distant second with 5.4 million.20 Yet there are reasons for concern. One is that the relatively inexpensive places to deploy broadband are becoming harder to find, making future growth more difficult. Another is that today's version of broadband is only a start; the real benefits of broadband may come when speeds hit much higher levels.21
Finally, there are competitive concerns. Although broadband promises to be a competitive service, with multiple facilities-based rivals already in the market, the networks (particularly the telephone networks) are not yet able to offer broadband everywhere. As a result, many consumers today can get broadband from only one source--typically, their local cable companies.
Building a broadband network is expensive. Deploying current technology nationwide and upgrading to higher-speed systems could demand investment in the hundreds of billions of dollars. Yet, for DSL service by telephone companies, much of the resulting infrastructure will be subject to the same unbundling rules that apply to traditional telephone service. The result is discouraged investment. In fact, because the potential harm from such rules increases with the riskiness of investment and complexity of the technology, the disincentive to broadband may be even greater than to traditional telephone service.22
Consistent with the last year's remand order from the D.C. Circuit Court of Appeals, the FCC must significantly rewrite its rules on the unbundling of network elements. Unlike last time, however, these changes should not just tinker at the margins. Major change is needed not just to satisfy the courts, but also to provide competitive, advanced telecommunications services to American consumers.
Crafting this new set of rules will not be easy. Given the complexity of the technology and the markets involved, the FCC will have to contend with a multiplicity of details.23 In determining which network elements should remain subject to unbundling, it should adopt a rule similar to the "essential facilities" doctrine used by the courts in antitrust cases. Under this test, only those elements that are true economic bottlenecks for which sustainable duplicate facilities are not economically feasible would be subject to forced access. While the D.C. Circuit Court did not specifically require this, Judge Williams hinted strongly that it would be appropriate.
Under such a rule, many current network elements, including switching and inter-office transport, would likely be crossed off the regulatory list. Much of the local loop infrastructure probably would still be classified as a bottleneck, although that could vary by geographic market. The all-inclusive UNE-P would also be eliminated, as least as currently structured.
New investment in infrastructure for high-speed and other advanced services also should not be subject to regulation. Not only is the disincentive effect particularly strong for such investment, but--because of the wide availability of cable modem service--the LECs also have no monopoly power, making regulation unnecessary. While the Telecommunications Act does not specifically cite consideration of such factors, they can be considered because of the FCC's general authority under the act to "forbear" from subjecting providers to regulation it finds not to be in the public interest.24
It is likewise important that the FCC ensure that state or local regulators do not nullify its decisions by imposing their own unbundling requirements on carriers. To a large extent, telecommunications today is inherently interstate in nature.25 Rules imposed in one jurisdiction can interfere with national efforts to create competitive and advanced telecom networks. The FCC's rules, however, need not exclude the states entirely from decision-making. They could play a role, for instance, in evaluating specific market conditions in local areas to determine whether bottleneck conditions exist.
The FCC's upcoming decision is a complex one, but the principles that need to be followed are clear. All parties agree on the benefits of competition and advanced services. At issue is whether policymakers will continue to pursue those goals through highly prescriptive regulations and mandates, or whether they will reduce regulation and foster marketplace incentives. The better choice, both for consumers and for the economy, is the second path.
James L. Gattuso is Research Fellow in Regulatory Policy in the Thomas A. Roe Institute for Economic Policy Studies at The Heritage Foundation.
6. For an analysis of some of the factors behind this financial "meltdown," see Larry F. Darby, Jeff A. Eisenach, and Joseph F. Kraemer, "The CLEC Experiment: Anatomy of a Meltdown," Progress and Freedom Foundation Progress On Point 9.2, September 2002.
15. It should be noted, however, that this antitrust doctrine itself has been subject to criticism. See Abbott K. Lipsky, Jr., and J. Gregory Sidak, "Essential Facilities," 51 Stanford Law Review1187 (1999).
16. For a full discussion of this, see Randolph J. May and Larry F. Darby, "Comments of the Progress and Freedom Foundation Before the Federal Communications Commission," Review of the Section 251 Unbundling Obligations of Incumbent Local Exchange Carriers, CC Dkt. No. 01-338 (filed April 5, 2002).
17. See Cambridge Strategic Management Group, "Assessing the Impact of Regulation on Deployment of Fiber to the Home," in "Comments of Corning, Inc.," Review of the Section 251 Unbundling Obligations of Incumbent Local Exchange Carriers, CC Dkt. No. 01-338 (filed April 5, 2002).
19. Robert W. Crandall and Charles L. Jackson, "The $500 Billion Opportunity: The Potential Economic Benefits of Widespread Diffusion of Broadband Internet Access," Criterion Economics, L.L.C., July 2001.
22. For an explanation of the potential incentive effects, see National Research Council, Broadband: Bringing Home the Bits, 2002. See also, "Broadband: Should We Regulate High-Speed Internet Access?" AEI-Brookings Joint Center for Regulatory Studies, 2002.
24. Notably, the current decision involves only one of several proceedings the FCC has ongoing regarding broadband regulation. See "Written Statement of Michael K. Powell, Chairman, Federal Communications Commission, Before the Senate Committee on Commerce, Science and Transportation," January 14, 2003.