The Future Legacy of the AT&T/ T-Mobile Merger Case
Jonathan L. Rubin1
The government’s challenge to the proposed acquisition of T-Mobile’s U.S. wireless communications business by AT&T Mobility could leave a lasting imprint on the annals of antitrust. The epic cases between the government and “Ma Bell” in the 1980s and Microsoft in the 1990s not only moved antitrust policy onto the front pages but also stretched the limits of antitrust analysis. Both cases left their mark on antitrust law, if not on the commercial life of the nation. Might the DOJ’s challenge to the AT&T/T-Mobile merger be a similarly significant antitrust case?
II. THE SIMPLE SCENARIO
On its most basic level, the transaction is a 4-to-3 merger in a nationwide market that seems to conform nicely to the kind of transaction that should be blocked under the Department of Justice (“DOJ”) – Federal Trade Commission (“FTC”) Horizontal Merger Guidelines. The Guidelines consider markets with a Hirschman-Herfindahl Index (“HHI”) between 1,500 and 2,500 to be moderately concentrated while markets with an HHI above 2,500 are highly concentrated. The FCC has calculated the national HHI for the wireless segment for the year 2009 as 2,350, and the average HHI for regional wireless markets as 2,811. More recent estimates put the national HHI figure well over the 2,500 benchmark for highly concentrated markets.
A merger that increases the HHI by more than 200 points in a highly concentrated market is significantly likely to reduce competition and a merger resulting in more than a 30 percent market share is presumed illegal under well-established merger case law. The government alleges that the AT&T/T-Mobile merger would cause an increase of more than 700 points in the HHI nationally and give the merged entity a market share of more than 40 percent in the national and many regional markets, and more than 50 percent in other regional markets.
In its simplest incarnation, then, the case presents a concentrated national market of wireless carriers composed of Verizon (with 106 million subscribers), AT&T (99 million), Sprint (52 million), and T-Mobile (34 million) teed up for a trial over the likelihood of anticompetitive effects in a world in which a combined AT&T/T-Mobile with 133 million subscribers faces only two other national competitors. The government should be reasonably confident that it can muster the evidence to establish a prima facie case. Of course, litigation always entails risks and some considerable probability of losing, but the DOJ and the plaintiff states have made a reasonable bet on this framework to lead to success at trial or to a settlement that defeats the transaction.
Ending the merger on this basis could very well represent a watershed event for the wireless industry, but it wouldn’t elevate the case to epic status. While the simple scenario has the
1 Jonathan L. Rubin, a trial attorney, economist, and internationally recognized expert in antitrust law, is the founder of Rubin PLLC. He can be reached at email@example.com.
advantage of fitting neatly within conventional antitrust analysis, a decision on these grounds won’t break new ground or engender a broader impact beyond the telecommunications sector. In order for that to happen, several additional issues must be allowed to enter into the analysis.
III. ISSUE I: EFFICIENCIES
In a recent comment the Hudson Institute’s Irwin Stelzer opined that AT&T’s executives convincing themselves that the deal would pass review without a challenge is a case of hubris preceding nemesis.2 Equally likely, however, is the chance that AT&T believes deeply that the efficiencies in the transaction will more than make up for any likely anticompetitive effects the government might be able to demonstrate.
A deep-seated belief in the efficiency benefits of the transaction would not be out of character for an operator of a large network facing capacity constraints who sees a chance to solve its congestion problems for years to come in one fell swoop by joining two large networks into an even larger one. Even if the additional capacity now owned by T-Mobile could be replicated at the portion of the purchase cost not allocated to customer acquisition, patching the two networks together could be accomplished practically overnight compared to the years it would take to put comparable new capacity on line and capture sufficient market share to support it.
The problem with following a legal strategy based on such thinking, of course, is that it confuses the idea of engineering efficiency with economic efficiency. The latter, which is the province of antitrust, requires credible economic evidence that the loss of competition will be more than accounted for in tangible economic benefits that flow to consumers. As the Merger Guidelines put it:
The greater the potential adverse competitive effect of a merger, the greater must be the cognizable efficiencies, and the more they must be passed through to customers, for the Agencies to conclude that the merger will not have an anticompetitive effect in the relevant market. When the potential adverse competitive effect of a merger is likely to be particularly substantial, extraordinarily great cognizable efficiencies would be necessary to prevent the merger from being anticompetitive. In adhering to this approach, the Agencies are mindful that the antitrust laws give competition, not internal operational efficiency, primacy in protecting customers.3
Thus, in the face of evidence of anticompetitive effects, the parties’ efficiency defense only succeeds if they are able to demonstrate lower incremental costs of production not achievable by realistic means other than the merger. But the “efficiency” at the heart of the parties’ network-linking plan is not necessarily tied to lower incremental costs (or, what is the same thing, economies of scale). The government could lose if the court accepts the engineering efficiencies of integrating the two networks to put large additional capacity on-line quickly as a substitute for operating efficiencies likely to directly benefit customers in the form of lower prices. If the court did so, however, it would do irreparable injury to the concept of “cognizable efficiencies,” the doctrine that limits the use of the efficiency defense to verifiable, merger-specific net-efficiencies
2 Irwin Stelzer, Antitrust Policy in an Age of Rapid Innovation, Economic Policy Studies / Briefing Paper, Hudson Institute, (October 2011).
3 HORIZONTAL MERGER GUIDELINES, U.S. DEPARTMENT OF JUSTICE AND THE FEDERAL TRADE COMMISSION, 30 (rev. Aug. 18, 2010).
that are not the result of anticompetitive effects. The transaction does enable AT&T to circumvent the discipline of managing organic growth and investment in the face of competition for market share, but permitting the transaction to go forward in the face of demonstrated anticompetitive effects based on the engineering advantage of linking the two networks is more akin to a subsidy rather than a cognizable efficiency. Notably, the merging parties implicitly tout the benefits of consolidating control of T-Mobile under U.S. ownership as a justification for permitting the deal to proceed.
If the parties’ efficiency defense succeeds by any means other than by demonstrating post-merger economies of scale in a magnitude sufficient to overcome the anticompetitive effects—an outcome that seems fairly unlikely—the potency of U.S. merger law could be impaired significantly, with the case marking an unprecedented retreat from current merger control standards.
IV. ISSUE II: THE ADJACENT MARKETS
The simple scenario—with or without cognizable efficiencies—skirts the thorny issues raised by the separate legal challenges to the merger brought by the third largest national carrier, Sprint, and the regional carrier C Spire (formerly, Cellular South). For its part, Sprint may lack antitrust standing to bring its case against the merger if the analysis remains within the confines of a national market with four players in which competition is diminished, to the presumptive benefit of Sprint. It is imperative to its case, therefore, that Sprint succeed in extending the issues to be litigated beyond unilateral and coordinated effects in the wireless markets to include the likely exclusionary effects of the merger in adjacent markets.
Thus, Sprint’s complaint includes allegations both about the horizontal effects of the merger and its vertical effects on the adjacent markets for backhaul (special access land lines between towers and switches largely controlled by AT&T and Verizon), devices and equipment (controlled by manufacturers of handsets, smartphones, tablets, and networking equipment), and roaming services (principally, again, AT&T and Verizon). With a not very compelling story to tell in the roaming market, Sprint’s complaint about vertical effects ultimately comes down to the “alleged backhaul markets,” that is, “each market in which AT&T offers backhaul services and T-Mobile provides wireless services,” and its access to devices and equipment.
What the merger will do is remove T-Mobile as a purchaser in the equipment and backhaul markets, a situation that Sprint alleges will lead to unilateral price increases for backhaul, increased coordinated interaction to raise rivals costs, and exclusionary conduct that will foreclose Sprint’s access to the newest handsets and equipment.
The complaint is short on specifics about the “loss of competition” foreseen by Sprint in the backhaul markets, but all else being equal taking T-Mobile out of the picture as a buyer of backhaul services would ease the competition for backhaul supply and not necessarily portend higher backhaul prices. But Sprint is raising a legitimate point that its competitive disadvantage from having to rely on its integrated rivals for an essential input would likely be exacerbated after the merger. The complaint characterizes the effect as an enhanced opportunity for AT&T and Verizon to coordinate in the backhaul markets and an increased incentive to exclude Sprint from access to backhaul as an AT&T/Verizon duopoly emerges. The allegation also is made that the merger will heighten barriers to competitive entry into the backhaul markets. Whether Sprint’s complaint is sufficiently well-pleaded to survive AT&T’s motion to dismiss, which is pending as of this writing, remains to be seen.
The mechanism of the anticompetitive effects in the handset market is more straight.forward. AT&T will accrete additional buyer power by absorbing T-Mobile’s demand for devices and equipment, a result that Sprint alleges puts it at risk of being excluded from access to the latest devices as manufactures design and customize their products for the larger carriers. Customization is a particularly important way that a buyer can exercise buyer power, because a technological tie between a device and network design can completely close off access to devices and features by rival networks. Sprint’s complaint is also short on detail about the device market and the court may require additional facts to support the extent of its alleged foreclosure there, but the point about AT&T’s additional buyer power is well taken. The problem is even more acute for Sprint’s smaller rivals, whose smaller subscriber base leaves them with little or no negotiating power against device manufacturers.
C Spire, a regional carrier with around 800,000 subscribers, not only challenges the merger on grounds that implicate AT&T’s post-merger buyer power in the device market, but also on the effects in the wholesale market for roaming. Regional carriers are large consumers of roaming services in order to provide their subscribers with nationwide coverage. The loss of T-Mobile as a competitive supplier of GSM roaming services could have a significant impact on all the regional, or “Tier II,” GSM carriers.
Depending on how the court disposes of these adjacent market issues, the decision could be significant for what it says about the standing of competitors to mount a private challenge to a merger. The essence of these complaints is the likely conduct of an integrated competitor who is also a monopoly or duopoly supplier. The set-up is so rife with anticompetitive opportunities that only a legal technicality could derail consideration of the issues in the adjacent markets. Sprint and C Spire may have their work cut out for them before their actions are suitably framed for litigation. But if they can overcome that hurdle—and the cases are not first otherwise disposed of—their challenges could lead to a significant ruling regarding the latitude of private parties to sue to enjoin a proposed merger of an integrated rival that controls an essential input or wields excessive buyer power.
V. ISSUE THREE: THE BIGGER PICTURE AND TELECOMMUNICATIONS POLICY
A broader view of the case would do a better job of accounting for the presence, now and in the future, of the Tier II carriers and their roughly 30 million subscribers. A market definition that ignores such a substantial commercial presence is unsatisfactory to the merging parties on the grounds that it understates the extent of competition that a combined entity would be expected to face. Anticipating this line of defense, the DOJ’s strategy is to plead a national market for government and enterprise wireless services but regional markets for consumer services. These market definitions exhibit concentration ratios as high or higher than the national figures, even with the Tier II carriers included.
But a legitimate question exists over the extent to which the court can, or should attempt to, promote a more coherent competition policy than what currently pertains under the patchwork of Federal Communications Commission (“FCC”) orders and the history of the wireless industry that brought us to this point. Having adopted a policy of relying on competition in the markets for wireless services, the FCC, in this area more so than in other segments of the telecommunications industry, seems to have left the heavy lifting to the antitrust courts and competition agencies.
The competitive effects of the merger on adjacent markets arise at least in part as a consequence of the FCC’s approach to price regulation for special access and the regime of “regulation light” it brings to bear on the wholesale roaming market. Under the latter policy wireless carriers are directed to conclude roaming agreements on terms that are “commercially reasonable.” By contrast, no regulation at all applies to exclusive deals between carriers and device manufacturers.
A decision in this case recognizing the need to protect opportunities for new entrants and the prospects for the continued existence and expansion of Tier II carriers would certainly contribute to a coherent wireless telecommunications policy. The merger could be rejected on the grounds that vesting AT&T with control over such a large integrated platform without providing for adequate means of insuring a policy of open access to certain key inputs would entail too great of a competitive threat. Such a decision would favor the longer-run development of a competitive wireless industry and unify the issues in the adjacent markets within a “systems competition” framework.
VI. THE CHANCES FOR AN EPIC CASE
The Bell System break-up and Microsoft cases arose out of concrete anticompetitive conduct; neither the parties nor the courts set out to litigate an epic case. Indeed, given a choice, courts prefer to do less rather than more. Because narrower grounds are available on which to base a decision, there is little likelihood that the case will achieve the status of the Bell break-up or the Microsoft case with grand pronouncements about the competitive conditions in the wireless industry. Somewhat more likely is the prospect that the case will achieve notoriety because AT&T succeeds on its efficiencies defense, or because new ground is opened for competitors such as Sprint and C Spire to bring a private action to enjoin a proposed transaction.
As in most things, the simplest explanation is usually the best. If the case remains within the confines of the simple scenario, the wireless industry and merger law may be well-served, but a significant impact on antitrust law and policy will have to wait for another day.