The AT&T/Time Warner Merger Case: What Happened and What is Next
By: Steven C. Salop, Professor of Economics and Law, Georgetown University Law Center
The trial of the AT&T/Time Warner vertical merger has now been completed, and Judge Leon has issued his opinion. The government lost. I am disappointed that the Department of Justice was unable to satisfy Judge Leon’s burden of proof. Notwithstanding Judge Leon’s confidence that the merger is a procompetitive transaction by AT&T and Time Warner to compete with firms like Netflix, Amazon Prime, and Google, I remain concerned that the merger will have anticompetitive effects, both on its own and as a result of the subsequent vertical mergers in the Pay-TV industry that the decision will encourage and the opinion may protect.
I am particularly concerned that Judge Leon’s decision will lead to a Pay-TV distribution market dominated by a few vertically integrated firms. As I have written in a recent article, the resulting market structure could lead to a reciprocal licensing outcome with severe anticompetitive coordination effects. These integrated firms might well be able to facilitate credible pricing coordination among themselves with reciprocal program content contracts at high-input prices, supported by most-favored nations (MFNs) contractual provisions that discourage discounting. The higher prices would then be passed on to consumers, allowing the firms to achieve an outcome closer to the cartel outcome in the downstream video subscription market.
This post will offer some preliminary observations of several issues. I will not try to review all the facts of the case. Instead, I want to offer a few preliminary comments about Judge Leon’s overarching skepticism about the government’s theory. I will then turn to the failure of the antitrust agencies to revise the long out-of-date vertical merger section of the 1984 Merger Guidelines (which I will refer to as the “Vertical Merger Guidelines,” or “VMGs”), and the revisions that may occur now. I will also offer some observations on the possible impact of this case on vertical mergers in the Pay-TV industry and vertical merger enforcement. Finally, I will mention the potential for an antitrust case attacking MFNs in this industry.
Judge Leon’s Opinion
Although he suggested that he did not take claimed efficiencies into account in rendering his decision, in light of what he perceived as the lack of anticompetitive effects, Judge Leon’s analysis was premised on his view that the merger would be highly efficient. He expressed confidence that AT&T will achieve a high level of its claimed efficiencies. In fact, the government conceded that the merger would lead to $350 million in annual efficiencies achieved from the “elimination of double marginalization” (EDM), resulting in DirecTV being able to acquire the Turner networks at cost rather than at the higher affiliate fee.
Judge Leon unfortunately overstated the consumer welfare benefits of this efficiency benefit in that he erroneously assumed for two reasons that they will be fully passed on to DirecTV customers. First, normally only a fraction of marginal cost reductions are passed on to consumers. Second, the fact that rival distributors would have higher costs resulting from AT&T’s increased bargaining leverage would tend to reduce the pass-through of the EDM by even more.
While he explicitly disclaims a presumption that vertical mergers rarely are anticompetitive because of efficiency benefits, the government’s conceded EDM benefits supported what seemed like a tacitly accepted presumption of this effect. As he explains,
The Court therefore declines defendants’ invitation to adopt either a per se rule or a presumption that would apply to most vertical mergers. To be sure, the standard for which defendants advocate aligns with the views of a number of authorities, including judges from this Circuit. See, e.g., Robert Bork, The Antitrust Paradox 245 (“[I]n the absence of a most unlikely proved predatory power and purpose, antitrust should never object to the verticality of any merger.”) … Tempting though it may be to agree with my appellate brethren, I need not, and will not, go that far to resolve this case.
Despite these EDM efficiencies, the government argued that consumers nonetheless would be harmed by higher prices on balance. The main prong of the government’s theory was an allegation that the merger would provide the merged firm with the ability and incentive to obtain a higher affiliate fee for Time Warner’s Turner Cable Networks in its negotiations with competitors of DirecTV. As explained by the government’s economic expert, Professor Carl Shapiro, higher affiliate fees would raise the costs of DirecTV’s rival video distributors, which would allow DirecTV to charge a higher subscription fee for DirecTV or retain more subscribers, given that DirecTV has been losing subscribers.
The same analysis would apply to internet (“virtual”) program distributors that compete with AT&T’s DirecTV Now “virtual” distributor. In addition, the merger likely would to lead to somewhat higher prices charged for Time Warner content to other “virtual” program distributors, such as PlayStation Vue or Google’s YouTube TV, that distribute over the internet. It is also the case that AT&T’s wireless business similarly would benefit if affiliate fees were raised or content were withheld from the video packages created or distributed by AT&T’s wireless competitors.
According to the government’s bargaining leverage theory, AT&T would have the ability to negotiate a higher price because a “blackout” of Time Warner’s Turner cable networks would cause DirecTV to retain more subscribers at the expense of the other video distributors. This would mean that the merged Time Warner/AT&T would have less to lose from a failure to reach a negotiated outcome than the Turner networks would have suffered absent the merger. This change in bargaining conditions resulting from the merger is what would increase the bargaining leverage of the merged firm.
Judge Leon was highly skeptical of this anticompetitive leverage theory. Judge Leon was receptive to the parties’ argument that the Turner negotiators would not take into account the interests of DirecTV in making their negotiation demands. He was also receptive to the argument that a blackout would reduce their profits, so a blackout threat would not be credible. He was also receptive to the idea that blackout threats are not credible because blackouts are relatively rare and almost all blackouts are temporary, not permanent blackouts.
Although Judge Leon did not reject the theory outright, he made numerous criticisms of the assumption that Time Warner executives would work to maximize the “joint profits” of the vertically integrated firm, as if it were just an “economists’ assumption,” not a good description of the real world. I was very surprised by Judge Leon’s skepticism on this point. First, it would be economically rational for the Time Warner negotiators to take into account the interests of DirecTV since they are both part of the same vertically integrated firm.
This is not simply an “economists’ assumption” made for convenience. Antitrust law and economics both are premised on the view that the integrated firms work to maximize the “joint profits” of all of its divisions. For example, this was the basis for the Supreme Court’s decision in Copperweld. Moreover, as explained by Professor Shapiro in his testimony, this assumption of joint profit-maximization is also a premise of the EDM efficiency theory, which Judge Leon embraced.
Furthermore, the fact that blackouts are rare and almost all of these are temporary, not permanent, does not mean that the leverage theory is wrong. The leverage theory does not require that blackouts are common, and theory is consistent with blackouts being always or almost always temporary. The leverage theory is premised on blackout threats, not actual blackouts. In fact, programmer blackout threats sometimes occur before the contract expires, as a shot across the bow of the distributor to induce the distributor to agree before the existing contract even expires. These threats are enhanced if the contract expires right before a major event (e.g., the Super Bowl, March Madness, or Nielsen Sweeps Week). They can also be enhanced if the programmer inserts advertisements into its programming to warn viewers of the impending blackout and suggests that subscribers may want to consider shifting to a rival distributor in advance. I discussed this issue in a submission to the FCC in 2010 on behalf of Time Warner Cable.
In legal disputes, the direct analogy to a blackout is the parties failing to settle in advance of completing discovery or before trial. Because of those litigation costs, a very high percentage of business disputes are settled before trial or even before a complaint is filed and discovery gets underway. For that matter, large wars are also pretty rare, but the threat of war always sits in the background. Would Judge Leon also be skeptical that America’s nuclear arsenal provides it with bargaining leverage in that nuclear weapons have not been used for almost 75 years?
Judge Leon was also skeptical of testimony by the rival distributors who were supportive of this leverage theory. As downstream competitors of DirecTV, Judge Leon explained on page 92 of the opinion that he believed that there was a “threat that such testimony reflects self-interest rather than genuine concerns about harm to competition.” In fact, Judge Leon even discounted the leverage concerns expressed to the FCC by DirecTV itself regarding the acquisition of NBC Universal by Comcast, and the bargaining leverage economics submission of its economic expert, Professor Kevin Murphy. He apparently did not apply this same degree of skepticism towards the testimony of the executives of the merging firms. Yet, their testimony obviously might similarly reflect their self-interest in completing the merger rather than consumer welfare.
Competitors’ testimony is worthy of more credibility when the competitors are also customers. This distinction has been recognized in other antitrust cases, where the complaints of competitors are subjected to an “antitrust injury” requirement that their complaints are consistent with harm to consumers. One recent example is Judge Huvelle’s 2011 order regarding the antitrust complaints made by Sprint and C Spire against the then-proposed acquisition of T-Mobile by AT&T.
Judge Leon also rejected the empirical inputs that were used by Professor Shapiro to predict the potential anticompetitive effects of the leverage in this matter. I am not going to review all the back-and-forth on those inputs. (This also would be made more difficult from the fact that the expert reports have not been made part of the evidentiary record.) So it remains simply an interesting question for readers of the opinion — and potentially for the D.C. Circuit if the government appeals — whether Judge Leon’s general skepticism might have affected his weighing of the evidence on these inputs.
Failure to Revise the Vertical Merger Guidelines
This opinion shows the fallout from the antitrust agencies previously failing to revise the 1984 VMGs to account for modern economic analysis of vertical mergers. There are numerous issues that would need to be studied and resolved in a revision. So it would have been considerable work for the agencies. This could have been an impediment for revision.
But the benefit is that revised VMGs would have laid out the analytical framework for enforcers, practitioners and judges. Revised VMGs would have explained the harms and benefits, potential presumptions and how they should be balanced under the antitrust agencies’ view of the Section 7 legal standard.
Revised VMGs would also have taken into account more recent law involving exclusionary vertical restraints, such as the Supreme Court’s adoption of the rule of reason (without a procompetitive presumption) in Leegin, as well as economic analysis that has taken place since Bork’s contribution. They would have discussed the range of probative evidence, including both criticisms and justifications of that evidence. The point here is not that the VMGs necessarily would have come out in one direction or another, but rather that one would hope that the agencies would have taken a serious look at the issues, decided on the proper course, put out their tentative conclusions for comment, and then reached an analytical framework that could be relied on by the court.
Judge Leon cites the 1984 VMGs several times. Had the VMGs been revised, Judge Leon would presumably have cited to the newer version. He would also have had a better understanding of the modern approach to vertical mergers. Judge Leon would have been able to consult these new VMGs, rather than having to attempt to master vertical merger analysis for the first and only time from Professor Shapiro’s abbreviated oral testimony in the context of this high-stakes litigation.
The HMGs have been very influential with judges. Revised VMGs likely would have great influence as well. Guidelines are authoritative statements of how agencies analyze mergers, which is why courts are willing to rely on them. Guidelines also help the agency litigators shape potential cases for presentation to a court, which increases their likelihood of winning cases that they bring and helps the agencies avoid bringing cases that are likely unwinnable or fail to serve the public interest.
The treatment of ease of entry in Baker Hughes provides a possible illustration of the importance of having up-to-date analysis in merger guidelines. In Baker Hughes, the district court rejected the DOJ’s claims that entry (or entry threats) would be insufficient to deter price increases. At the time the HMG’s treatment of ease of entry was very terse, so the district courts received little guidance about the specific analysis. The DOJ argued to the D.C. Circuit that the proper standard for easy of entry is that the entry would be “quick and effective,” a standard that the court rejected. Perhaps in response to DOJ’s loss in Baker Hughes, the 1992 HMGs set out a detailed analysis that formulated three requirements for establishing “easy entry” — that the entry must be timely, likely, and sufficient. Under this standard (which remains even after the 2010 revision of the HMGs), I do not think that any defendant since 1992 has succeeded in winning a litigated case on the basis of easy entry.
Revised VMGs would have indicated whether or not to credit as foreclosure the bargaining leverage theory used by Professor Shapiro in this matter, or whether to require a showing of profitable withholding of the programming that can harm consumers. If it did credit the leverage theory, the VMGs hopefully would have explained the credibility issue, that is, why increased bargaining leverage from a merger is a competitive concern even if failure to reach agreement (i.e., “blackouts”) harms both sides.
While the party threatening the blackout might suffer lower profits if it carries out the threat, mainstream economic analysis has shown that prudently calculated threats are credible (in the sense of “believable”) once the bargaining dynamics are traced through. This analysis of sequential offers and counteroffers is now the modern foundation of the seminal Nash bargaining solution. Similarly, revised VMGs would also have explained clearly that there is a strong presumption that executives in a vertically integrated firm would take actions to maximize the joint profits of the corporation, not just the profits of their own division.
Revised VMGs hopefully would also have explained that a degree of skepticism towards (downstream) competitor testimony is much less warranted in a vertical merger case because those competitors also are the customers of the merged firm. This is important when the competitive concern is that a merger will incentivize the merged firm to raise the input prices that it charges to customers/competitors.
The guidelines’ drafters would have analyzed, and then the VMGs hopefully would have explained, how the agencies would treat EDM. As an efficiency claim, it is likely that the burden would be placed on the merging parties to show that the EDM claims are merger-specific, verifiable, and sufficient to overcome the anticompetitive effects on consumers, just as in the case of the HMGs. The parties would need to explain why they did not have contracts that would account for double marginalization before the merger.
A conclusory answer that there were “bargaining frictions” would seem insufficient, particularly in that the parties were able to negotiate a merger agreement. If other similarly situated firms had such contracts, that fact would also weigh in the balance. If the parties argue that a contract would not have been able to eliminate all the double marginalization, then only that incremental fraction would be merger-specific. As I have discussed in my recent article, the VMGs would also identify and discuss the countervailing forces reducing EDM.
Revised VMGs would also have presented and explained the agencies’ interpretation of the proper application of the Section 7 incipiency standard to vertical mergers, in light of their analysis and the treatment of exclusionary vertical conduct in other contexts where there is no incipiency standard. While a judge obviously would not be forced to apply this interpretation, the VMGs would have been given weight.
Judge Leon’s decision likely will have short-term effects on mergers in the Pay-TV ecosystem. Comcast started a bidding war with Disney for the Fox assets and may still attempt to obtain some of the divested regional sports networks, or it may attempt to acquire other programmers. One also can imagine that Charter, Cox Communications, and Verizon are thinking hard about what vertical mergers to pursue themselves.
The impact of this case on vertical merger law more generally is harder to predict. The big question right now is whether the DOJ will appeal. If they do not, then an issue will be whether or not the opinion will be treated as a thoughtful judicial analysis worthy of following by other courts. If the DOJ does appeal, then it is a question of how the D.C. Circuit treats the case and the opinion.
Unless the D.C. Circuit adopts a Borkian legal standard, I expect the DOJ to continue to take a tough line. If the DOJ does not appeal this case, Judge Leon’s opinion was totally fact-based, so it should not discourage future cases where the government believes that they have sufficient facts.
I also expect that the agencies will now revise the VMGs. I expect the VMGs to include the bargaining leverage theory. I also do not expect the VMGs to presume substantial efficiencies or to set a high burden of proof on the government. To embed that presumption in the VMGs would essentially concede that their AT&T/Time Warner case was defective from the get-go. Weak VMGs would also feed into the false view that the DOJ brought the case at the behest of President Trump.
I expect the VMGs to favor injunctions and divestitures over behavioral remedies. But I hope that the VMGs will also consider the idea that well-monitored behavioral remedies can have some benefits, if they are drafted in a way that they can be revised if they fail to achieve their competitive goal, as I have suggested elsewhere. The VMGs might also consider the appropriateness of setting a lower burden of proof on the government when the proposed remedy is less than a full injunction.
In light of this case, the DOJ might also consider a possible Section 1 case attacking MFNs in this industry. MFNs are vertical contracts that Baker and Chevalier have explained can have severe coordinated and exclusionary anticompetitive effects. MFNs are rampant in the Pay-TV industry. Moreover, the strongest argument for EDM being merger-specific in the AT&T/Time Warner merger (and other vertical mergers in this industry) is the existence of MFNs that discourage discounting and other innovative arrangements.
MFNs help to maintain high prices for content that then drive up subscription prices paid by consumers and deter low-cost, innovative new services. There were rumors in 2012 of a DOJ investigation of Pay-TV MFNs, but there is no public record of its findings. The FCC proposed a prohibition on MFNs in 2016, though Commissioner (and now-Chairman) Ajit Pai voted against the proposal. But, in light of the AT&T/Time Warner case, it may be clearer to the DOJ how these MFNs are harming competition.
Steve Salop is a professor of economics and law at Georgetown University Law Center who has written numerous articles on antitrust and vertical mergers.