February 25, 2008 Antitrust Oversight of an Antitrust Dispute: An Institutional Perspective on the Net Neutrality Debate Jonathan E. Nuechterlein1 Several years after its first appearance in the telecommunications lexicon, the term “net neutrality” remains elusive, in part because its meaning varies with the speaker and the speaker’s agenda. But at the highest level of generality, the term describes two distinct types of proposed regulation of broadband Internet access providers. Under one type of proposal, regulators would draw and enforce a line between acceptable network management practices and unacceptable “blocking” or “degradation” of disfavored Internet applications and content. Under the other, regulators would ban a broadband Internet access provider from reaching commercial agreements with particular applications and content providers to provide the sophisticated performance-enhancement techniques—over and beyond best-efforts Internet access—needed to support unusually performance-sensitive applications and content, such as real-time video streaming or multiplayer online videogames. (In a variation on this second theme, regulators would permit such agreements but subject them to “nondiscrimination” requirements.) These two types of proposals are distinct but complementary: net neutrality proponents typically advocate both the anti-blocking rule and a ban on (or close regulation of) business-to-business relationships between broadband networks and applications or content providers. Such proposals will likely be, one way or the other, a principal focus of telecommunications policy for the next decade. They have captured the attention of Congress, where several bills on the topic have been introduced;2 of Senators Barack Obama and Hillary Clinton, who both advocate strong forms of net neutrality regulation;3 of legal, economic, and 1 Partner, Wilmer Cutler Pickering Hale & Dorr LLP; B.A., Yale College (1986), J.D., Yale Law School (1990). I am grateful to Lynn Charytan, Bob Hahn, Bill Lake, Paul Larkin, and Phil Weiser for their helpful comments on earlier drafts of this paper, to Mary Beth Caswell for her expert research assistance, and to participants in the February 2008 Silicon Flatirons “Digital Broadband Migration” conference in Boulder, Colorado, where this paper was presented. Although I have represented broadband companies on net neutrality issues, the views expressed here are my own. 2 See, e.g., S. 215, 110th Cong. § 12(a)(4)(C), (5) (2007); H.R. 5273, 109th Cong. § 4(a)(6), (7) (2006); H.R. 5417, 109th Cong. § 3 (2006). 3 See, e.g., Roy Mark, Clinton Defends Net Neutrality Position, eWeek, Nov. 14, 2007 (http://www.freepress.net/news/28178) (“Sen. Hillary Clinton’s campaign said Nov. 15 her long silence on network neutrality should not be interpreted as waning support for the idea of mandating that broadband providers treat all technology scholars across the ideological spectrum;4 and—of principal interest here—two key federal agencies: the Federal Communications Commission and the Federal Trade Commission. Most discussions of net neutrality focus on the merits of the debate: on the substantive costs and benefits of government intervention in the broadband market. This paper focuses instead on the comparatively neglected institutional dimension of the debate: an inquiry into which federal agencies are best positioned to resolve net neutrality disputes when they arise. As I argue below, the net neutrality controversy is best understood as a classic antitrust dispute about “vertical leveraging,” and the institutions most likely to appreciate the economic complexities of that dispute are the nation’s specialized antitrust agencies: the Justice Department and the FTC. Because these agencies regulate the economy at large rather than a single industry, they are less vulnerable than the FCC to capture by industry factions, they are less likely to develop industry-specific bureaucracies with incentives to keep themselves relevant through over-regulation, and, because of their firm grounding in antitrust enforcement, they are more likely to resolve competition-oriented disputes dispassionately and on their economic merits. I would thus revive in this context the competition-policy model that prevailed for much of the final quarter of the last century: a regime in which antitrust authorities, rather than industry-specific regulators, take the lead in addressing vertical-leveraging claims against providers of telecommunications transmission platforms. network use in a nondiscriminatory manner. . . . ‘Hillary Clinton has been and continues to be a strong supporter of net neutrality,’ Jin Chon, a spokesperson for the Clinton campaign, told eWEEK. . . . Clinton’s silence was the subject of a Nov. 15 teleconference involving several top-ranking Obama campaign officials. The conference came the day after Obama told a large crowd at Google’s California headquarters, ‘I will take a backseat to no one in my commitment to network neutrality.’”). Senators Clinton and Obama have both cosponsored net neutrality legislation that contains what I describe, in Part I.B below, as the “strong” form of access-tiering restrictions.. 4 For representative treatments, see Philip J. Weiser, The Next Frontier for Net Neutrality, 60 Admin. L. Rev. ___ (forthcoming 2008) (http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1080672#PaperDownload); Keeping the Internet Neutral?: Tim Wu and Christopher Yoo Debate, 59 Fed. Commun. L.J. 575 (2007); Edward W. Felten, Nuts and Bolts of Network Neutrality (July 6, 2006) (http://itpolicy.princeton.edu/pub/neutrality.pdf); J. Gregory Sidak, A Consumer-Welfare Approach to Network Neutrality Regulation of the Internet, 2 J. Competition L. & Econ. 349 (2006); Timothy Wu, Why Have a Telecommunications Law? Anti-Discrimination Norms in Communications, 5 J. Telecomm. & High Tech. L. 15 (2006); Robert W. Hahn and Scott Wallsten, The Economics of Net Neutrality, Economists’ Voice (Apr. 2006) (http://ssrn.com/ abstract=943757); Robert D. Atkinson and Philip J. Weiser, A “Third Way” on Network Neutrality, Information Tech. & Innovation Foundation (May 3, 2006) (http:// www.itif.org/files/netneutrality.pdf); Benjamin E. Hermalin and Michael L. Katz, The Economics of Product-Line Restrictions With an Application to the Network Neutrality Debate, Competition Policy Center (July 28, 2006) (http://repositories.cdlib.org/iber/cpc/CPC06-059); Christopher S. Yoo, Beyond Network Neutrality, 19 Harv. J. L. & Tech. 1 (2005). 2 This paper is divided into three main parts. Part I gives a brief primer on the contours of the net neutrality dispute and explains why, at bottom, net neutrality proposals could make sense only as claims about the proper application of antitrust-oriented concepts to the broadband marketplace. Part II then addresses the present institutional arrangements for addressing the net neutrality dispute, why those arrangements are redundant, and why such redundancy is problematic. Parts II.A and II.B discuss the parallel inquiries that the FCC and the FTC have initiated on net neutrality and describes the complex jurisdictional questions those inquiries raise. Part II.C then explains why permitting two peer federal agencies to address net neutrality disputes in parallel would systematically skew broadband policy towards inefficient over- regulation. Among other concerns, each agency would have an effective veto only over the other agency’s judgments that intervention is inappropriate and not over the other agency’s judgments that intervention is appropriate. Part II thus concludes that one, not two, federal agencies should be assigned exclusive jurisdiction to resolve net neutrality issues. Finally, Part III proposes a long-term institutional solution for oversight of the broadband industry. Under the arrangement proposed here, competition issues would be addressed by one of the two antitrust agencies (DoJ or the FTC); consumer-protection issues would be addressed by the FTC’s Bureau of Consumer Protection; and the FCC would maintain jurisdiction over residual, non-competition-related issues within its peculiar expertise. I. WHAT PEOPLE ARE ARGUING ABOUT WHEN THEY ARGUE ABOUT NET NEUTRALITY. One of the main challenges for students of the net neutrality debate is the difficulty of pinning down exactly what that debate is about. Before addressing that issue, I first review the technological context in which this debate arises.5 A. A Taxonomy of IP Networks. The first step is to define “the Internet,” the central subject of all net neutrality proposals. What we call “the Internet” is not a unitary, centrally managed network, but an interconnected set of many thousands of constituent networks. What joins these networks together into the 5 For a more detailed background, see Jonathan E. Nuechterlein & Philip J. Weiser, Digital Crossroads: American Telecommunications Policy in the Internet Age 128-46 (2007 ed.). 3 Internet is that each has voluntarily adopted a common protocol and addressing scheme—the Internet Protocol (“IP”)—that enables its end users to communicate with end users connected to other networks for purposes of exchanging higher-layer applications and content.6 Most of these IP networks are privately owned and operated, and—significantly—their IP infrastructure is often used to provide “managed” IP services unrelated to communications with other IP networks over the publicly accessible Internet. For example, a global IP network provider might allocate some capacity on its network for the routing and transmission of Internet traffic but set aside additional capacity on the same network infrastructure for the provision of high-quality videoconferencing over a closed IP network devoted to a multinational corporate customer. Very roughly speaking, the constituent networks of the Internet fall into three basic categories. First, Internet backbone networks—such as AT&T, Level 3, Global Crossing, and SAVVIS—use long-distance fiber-optic cable to connect other, geographically dispersed networks, including the networks of large businesses, Internet access providers, and other backbone providers. Second, although large businesses often contract directly with a backbone network provider, most end users rely on an access network to bridge the “last mile” gap between them and an Internet backbone network (which in turn connects them to the rest of the Internet). Today, most residential consumers, and essentially all businesses with more than a few employees, obtain Internet access through a high-speed broadband connection. As discussed below, there is much controversy about how competitive the broadband marketplace is now and is likely to become. That controversy lies at the heart of the net neutrality debate. Finally, the third category of IP networks that participate in the Internet are so-called edge networks. These fall into two subcategories. The first consists of “end user” networks, which range from home WiFi networks to corporate LANs (“local area networks”). The second—of greater relevance here—consists of the networks operated by providers of Internet applications and content. In the commercial Internet’s early years, the stereotypical “edge” provider was an entrepreneur who ran a start-up website from a server in his garage. Today, the 6 See Resolution of the Federal Networking Council, Oct. 24, 1995 (quoted in http://www.isoc.org/internet/history/brief.shtml) (“‘Internet’ refers to the global information system that—(i) is logically linked together by a globally unique address space based on the Internet Protocol (IP) or its subsequent extensions/follow-ons; (ii) is able to support communications using the Transmission Control Protocol/Internet Protocol (TCP/IP) suite or its subsequent extensions/follow-ons, and/or other IP-compatible protocols; and (iii) provides, uses or makes accessible, either publicly or privately, high level services layered on the communications and related infrastructure described herein.”). 4 most prominent “edge” networks feature enormous “server farms” and caching facilities built by companies as diverse as service-providers Akamai and Level 3, on-line retailers Amazon.com and eBay, and Internet superpower Google. The largest of these edge networks are sometimes known as overlay networks because they resemble Internet backbones in their global reach. They operate by storing (or “caching”) copies of Web content on servers throughout the Internet, close to end users in many different locations, and deploying high-speed fiber-optic links connecting those servers to central databases. By circumventing points of traffic congestion on the Internet, these overlay networks give end users faster and more reliable access to a given company’s Web content. Although Google and a number of other large Internet companies have built proprietary overlay networks for their own use, many applications and content providers hire third-party providers such as Akamai and Limelight to perform this function. Applications and content providers that pay the substantial costs of this function have long enjoyed a commercial advantage over rivals that do not (or cannot) pay those costs—because, all else held equal, their consumers receive faster and more reliable access to applications and content. As discussed in Section I.B.2 below, the Internet has never been “neutral” among providers in this regard. B. A Taxonomy of Net Neutrality Proposals. Until the late 1990s, almost all residential consumers obtained access to the Internet through dial-up connections over the conventional telephone network. Independent Internet service providers, such as AOL and Earthlink, provided the critical gateway function linking the telephone network with the Internet. Customers would call a telephone number associated with their ISP’s facilities (“modem banks”); those calls would be routed through the telephone company’s circuit-switched network en route to those ISP facilities; and, at the receiving end, the ISP would provide the “protocol conversion” functions needed for communications between the subscriber’s computer and the servers that provide Internet applications and content.7 The telephone company was a more or less passive participant in this arrangement. As a common carrier, it routed calls to different ISPs’ modem banks in essentially the same manner as it routed calls to anyone else. As a legal matter, moreover, the telephone companies were subject to longstanding FCC rules known as the Computer Inquiry requirements. Very roughly 7 See Nuechterlein & Weiser, Digital Crossroads, supra note 5, at 134-35. 5 speaking, these rules enforced common carrier norms by requiring telephone companies to provide the same transmission capabilities to unaffiliated ISPs (and other information service providers) as they provided to their own information service affiliates.8 This technological landscape began to change in the late 1990s as residential consumers began bypassing the circuit-switched telephone network by using the local cable company’s facilities—and the ISP affiliated with that cable company—for high-speed access to the Internet. And with that technological change came a lively policy debate: should cable operators, like telephone companies, be required to “open” their broadband transmission networks to unaffiliated Internet service providers? This “open access” debate persisted on several fronts until 2005, when, after several years of litigation, the Supreme Court finally upheld the FCC’s conclusion that such regulatory intervention would be both unnecessary (because competition among rival broadband providers would protect consumer interests) and harmful (because excessive regulation would dampen incentives for investing in new broadband facilities for underserved residential communities).9 Meanwhile, telephone companies had begun to offer residential broadband connections themselves (through “digital subscriber line” technology) in competition with the cable companies. In 2005, the FCC followed through on its victory in the Brand X case by extending its deregulatory regime to telephone companies—specifically, by eliminating the Computer Inquiry requirements to the extent they applied to a telephone company’s provision of broadband Internet access.10 By then, the “open access” debate had begun to seem almost antiquated. That debate had focused on the rights of independent ISPs such as AOL and Earthlink. It had become clear by the early 2000s, however, that broadband technology makes such ISPs, if not irrelevant, at least much less central to a user’s Internet experience.11 In a dial-up world, you paid a monthly subscription fee to the ISP, not to the telephone company that carried your “local” call to that 8 See id. at 151-55; Robert Cannon, Where ISPs and Telephone Companies Compete: A Guide to the Computer Inquiries, 9 Commlaw Conspectus 49 (2001). 9 National Cable & Telecomm’ns Ass’n v. Brand X Internet Servs., 545 U.S. 967 (2005), aff’g Inquiry Concerning High-Speed Access to the Internet Over Cable and Other Facilities, 17 FCC Rcd 4798 (2002) (“Cable Broadband Order”). 10 Report and Order and Notice of Proposed Rulemaking, Appropriate Framework for Broadband Access to the Internet over Wireline Facilities, 20 FCC Rcd 14853, ¶ 44 (2005) (“Wireline Broadband Order”), aff’d, Time Warner Telecom, Inc. v. FCC, 507 F.3d 205 (3d Cir. 2007). 11 See Nuechterlein & Weiser, Digital Crossroads, supra note 5, at 155, 161-62. 6 ISP, and you blamed your ISP if your Internet connection was slow. In most cases, you could hardly blame the telephone company for poor performance, because it was treating a dial-up call like any other call and was thus dedicating fixed capacity (a voice-grade circuit) for the duration of an Internet connection. Today, however, you pay monthly fees directly to your broadband provider, and if your broadband connection is fast or slow, you assign credit or blame to that same provider; you are unlikely to know or care which ISP entity is connecting that provider’s local broadband network to the broader Internet. And as Timothy Wu points out, “[c]ompetition among ISPs”—the goal of open access mandates—“does not necessarily mean that broadband operators will simply retreat to acting as passive carriers in the last mile.”12 But when the air goes out of one telecommunications policy dispute, the vacuum is soon filled by another. Here, the regulatory energy that used to fuel the “open access” debate is now spent on a similar but distinct debate: “net neutrality.” Whereas open access proposals would have granted ISPs like Earthlink rights of “nondiscriminatory” access to the broadband transmission platform, net neutrality proposals would grant such rights to applications and content providers like Joost and BitTorrent. Beyond that generality, the term “net neutrality” means different things to different people, and the parties to this debate can be vague in defining what exactly they are talking about. As former FTC Chairman Timothy Muris recently observed (paraphrasing Phillip Areeda’s famous remark about the “essential facilities” doctrine), “‘net neutrality’ has become an epithet devoid of any analytical content.”13 Our first task, therefore, is to pin down the content of that term by identifying the major species of net neutrality proposals. 1. The Anti-Blocking Principle At the highest level of generality, net neutrality advocates propose two different types of requirements: a ban on “blocking” or “degrading” of disfavored content or applications over an Internet access platform, and a ban on (or at least close regulation of) contractual deals between broadband networks and content or applications providers for the terms of access to that 12 Tim Wu, Network Neutrality, Broadband Discrimination, 2 J. Telecomm. & High Tech. L. 141, 149 (2003). 13 Statement of Timothy J. Muris, Foundation Professor, The George Mason University School of Law, Before the Workshop on Broadband Connectivity Competition Policy, U.S. Federal Trade Commission, Feb. 28, 2007, at 18; cf. Phillip Areeda, Essential Facilities: An Epithet in Need of Limiting Principles, 58 Antitrust L.J. 841 (1989). 7 platform.14 As discussed below, these two types of proposed requirements are analytically distinct, although they are often blurred together. The first type of requirement—which I will call “anti-blocking” rules—would address efforts by a broadband provider to impede its subscribers’ access to particular Internet content or applications for reasons that a regulatory authority deems impermissible. In February 2004, FCC Chairman Michael Powell became the first major federal policymaker to address that issue when he “challenge[d] the broadband network industry” to honor several “Internet Freedoms” for consumers, including “access to their choice of legal content,” subject to “reasonable limits . . . placed in service contracts,” and a right “to run applications of their choice,” except where “they exceed service plan limitations or harm the provider’s network.”15 The next year, after Powell had left the FCC, the Commission followed Powell’s lead by issuing a non-binding Policy Statement that, in substance, embraced his “Internet Freedoms.”16 The Policy Statement provides, among other things, that consumers are “entitled to run applications and use services of their choice,” such as VoIP or video, “subject to reasonable network management” and “the needs of law enforcement.”17 At the time, the only documented violation of these principles had occurred in 2005, when a small rural telephone company named Madison River Communications blocked its subscribers’ access to VoIP services. It was alleged, and the FCC apparently concluded, that Madison River had blocked these services not for any legitimate network- management purpose, but simply to protect the lucrative access charges it earned for handling 14 Significantly, net neutrality proposals address the terms on which broadband providers offer Internet access service to consumers. Few net neutrality advocates seriously propose that the government disqualify the operator of an IP network from by devoting a portion of its bandwidth to particular applications other than connectivity with other IP networks, such as cable television service or secure teleconferencing networks. See Testimony of Timothy Wu before the House Comm. on the Judiciary, Telecom & Antitrust Task Force, at 7 (Apr. 25, 2006) (http://judiciary.house.gov/media/pdfs/wu042506.pdf) (asserting that “[t]he best proposals for network neutrality rules . . . . leave open legitimate network services that the Bells and Cable operators want to provide, such as offering cable television services and voice services along with a neutral internet offering”). Instead, the net neutrality debate concerns whether, and in what ways, broadband companies may treat different types of data differently in connection with the retail service it provides to consumers in the form of “Internet access.” 15 Remarks of Michael K. Powell, Preserving Internet Freedom: Guiding Principles for the Industry, at 5, Feb. 8, 2004 (http://hraunfoss.fcc.gov/edocs_public/attachmatch/DOC-243556A1.pdf). 16 Policy Statement, Appropriate Framework for Broadband Access to the Internet over Wireline Facilities, 20 FCC Rcd 14,986, ¶¶ 4–5 & n.15 (2005) (“FCC Broadband Policy Statement”). 17 Id. 8 long-distance calls over the conventional telephone network. Madison River quickly suppressed the ensuing controversy by paying a small fine and pledging to stop this practice.18 The FCC stressed in its Policy Statement that it was “not adopting rules.” Curiously, though, it did not hesitate to suggest that it would crack down on any violations of these non- rules. And soon after adopting the Policy Statement, it forced two of the nation’s largest broadband providers—SBC (now AT&T Inc.) and Verizon—to accept the Statement’s principles as binding (though temporary) conditions on the Commission’s approval of their pending mergers with, respectively, AT&T Corp. and MCI.19 For the ensuing two years, however, the debate about whether the FCC should convert its anti-blocking “principles” into industry-wide rules remained quiescent. The major broadband providers claimed that rules were unnecessary because they had no intention of violating the principles in the first place. And few broadband providers expressed any theoretical opposition to the Commission’s antiblocking principles in the abstract, at least to the extent they are applied to conventional cable or wireline broadband networks.20 That period of regulatory quiescence ended when, in late 2007, independent tests suggested that Comcast had manipulated Internet packet headers to suppress its customers’ use of BitTorrent, a peer-to-peer file-sharing application.21 The ensuing controversy vaulted the antiblocking principle once more to the forefront of the FCC’s policy agenda, as the Commission accepted invitations to open inquiries into whether it should enforce the antiblocking principles 18 See Order, Madison River Communications LLC, 20 FCC Rcd 4295 (2005). For an analysis of the Madison River case and its implications for the broader net neutrality debate, see Sidak, Consumer-Welfare Approach, supra note 4, at 416-22. 19 E.g., Mem. Op. and Order, SBC Communications Inc. and AT&T Corp. Applications for Approval of Transfer of Control, 20 FCC Rcd 18,290 (2005). More recently, in connection with approving the AT&T-BellSouth merger, the FCC extracted from the combined company a further commitment not to enter into certain arrangements with Internet content, applications, or service providers for two years. See Mem. Op. and Order, AT&T Inc. and BellSouth Corporation Application for Transfer of Control, 22 FCC Rcd 5662, Appx. F, at 5814-15 (2007) (“AT&T- BellSouth Merger Order”). This latter commitment bears a close resemblance to the proposed “access tiering” ban discussed below. 20 The issue is somewhat more complicated with respect to wireless broadband platforms, given the more extreme scarcity of network bandwidth (i.e., licensed spectrum). See generally Robert W. Hahn, Robert E. Litan, and Hal J. Singer, The Economics of Wireless Net Neutrality, AEI-Brookings Joint Center Working Paper No. RP07-10 (Apr. 2007) (http://papers.ssrn.com/sol3/papers.cfm?abstract_id=983111). In this article, I focus on the net neutrality debate as it applies to wired broadband platforms, including cable and wireline platforms. 21 See, e.g., Jacqui Cheng, Evidence mounts that Comcast is targeting BitTorrent traffic, Ars Technica, Oct. 19, 2007 (http://arstechnica.com/news.ars/post/20071019-evidence-mounts-that-comcast-is-targeting-bittorrent- traffic.html). 9 into binding rules—and, if so, how it should distinguish between “reasonable network management” and the unjustified suppression of disfavored applications.22 That distinction is likely to prove elusive, and the Comcast case shows why. No one argues that Comcast or other broadband providers can take no steps to ensure adequate network capacity for most subscribers by constraining its subscribers’ use of bandwidth-intensive applications. Indeed, the FCC’s Policy Statement conditions a consumer’s right “to run applications and use services of their choice” on a broadband provider’s prerogative to engage in “reasonable network management.”23 Defining that “reasonable network management” qualifier, however, is no easy task. All broadband networks contain potential bottlenecks of shared capacity. During peak usage periods, congestion in these bottlenecks can degrade basic Internet access for all subscribers. Such congestion poses an escalating challenge for network engineers, who must cope with the rapidly growing popularity of high-bandwidth Internet applications such as high- definition video-streaming and peer-to-peer video file-sharing while conserving on costly capital investments.24 Complicating that engineering challenge is an economic peculiarity about the retail market for Internet access. Most Internet access plans today include “all you can eat” connectivity; consumers pay a flat fee for a particular level of bandwidth but do not pay any incremental per-bit price for causing extra data traffic to cross shared network facilities. They pay the same for a 3 Mbps connection whether they use that connection once a day, to download a static webpage, or all day, to download and upload high-definition video files. There are thus no price signals to deter a minority of subscribers from overconsuming network capacity at the expense of the majority. The question in the FCC’s current proceedings is whether it is “reasonable” for a broadband provider like Comcast to treat the use of certain applications (such as BitTorrent) as a 22 See Public Notice, Comment Sought on Petition [of Free Press et al.] for Declaratory Ruling Regarding Internet Management Policies, WC Dkt. No. 07-52 (Jan. 14, 2008); Public Notice, Comment Sought on Petition [of Vuze, Inc.] for Rulemaking to Establish Rules Governing Network Management Practices by Broadband Network Providers, WC Dkt. No. 07-52 (Jan. 14, 2008). 23 FCC Broadband Policy Statement, supra note 16, at ¶¶ 4-5 & n.15. 24 See, e.g., William B. Norton, Video Internet: The Next Wave of Massive Disruption to the U.S. Peering Ecosystem, v1.3, at 2 (Equinix 2007); David Vorhaus, Confronting the Albatross of P2P, Yankee Group (May 31, 2007); Deloitte Touche Tohmatsu, Telecommunications Predictions: TMT Trends 2007, at 6 (2007) (http://www.deloitte.com/dtt/cda/doc/content/dtt_TelecomPredictions011107.pdf); Yankee Group, 2006 Internet Video Forecast: Broadband Emerges as an Alternative Channel for Video Distribution 6-7 (Dec. 2006). 10