The decision regarding the first challenge to a vertical merger brought by the U.S. federal government since 1979 came out recently: it is the ATT / Time Warner merger judgment, decided on 12 June, which can be found at http://www.dcd.uscourts.gov/sites/dcd/files/17-2511opinion.pdf.
The DoJ’s case was that the merger would substantially lessen competition in the video programming and distribution market by enabling AT&T to use Time Warner’s ‘must have’ television content to either raise AT&T’s rivals’ video programming costs or to drive those rivals’ customers to AT&T’s video distribution channels. ATT / Time Warner’s (the merging parties) case was that the video programming and distribution market is in the middle of a revolution where digital players such as Netflix, Amazon and Hulu are integrating video programming and distribution, while companies such as Facebook and Google are syphoning advertising from the TV to the digital space. The merger of AT&T and Time Warner would allow them to catch up to the competition – AT&T could offer Time Warner’s innovative video content and advertising offerings to its customers, while Time Warner would benefit from AT&T’s customer relationships and programming data.
As you may have heard, the judge dismissed the DoJ’s case and allowed the merger to proceed unconditionally. Earlier today, I learned that the DoJ decided to appeal. I apologise for the length of the overview below, but this is a fact-intensive 179 page decision.
The judgment is structured as follows:
A first section provides an overview of the background for the merger.
The video programming and distribution market can be said to have three levels: content creation, content aggregation and content distribution. Content creation may come from studies, networks or the distributors themselves. Content aggregation is pursued by programmers that aggregate content into networks or network groups. Distributors then bundle networks and distribute them to their subscribers.
Some subscription-video providers (e.g. Netflix, Hulu, Amazon Prime) are vertically integrated and able to distribute directly to subscribers, while the merging parties occupy different levels of this market.
Time Warner owns Warner Bros. (a movie, television and video studio), Turner and HBO (a premium subscription video service which relies solely on subscription fees). Turner operates, among other things, ten linear cable networks. It is a traditional programmer pursuing content aggregation – of networks such as TNT or CNN – and licensing them to content distributors. It makes money in two ways: from fees paid by content distributors, and from advertising fees. Both mechanisms create incentives for Turner and HBO to get as much of their networks into as many distributors as possible.
AT&T provides wireless, broadband and pay-tv services to consumer. As such, AT&T is a distributor. There are three key categories of players in the video distribution market:
- traditional multichannel video programming distributors (‘MPVD’). These include satellite providers operating nationally such as AT&T’s DirectTV, cable TV providers operating locally such as Comcast, overbuilders and telcos such as U-verse (which provides services over phone lines);
- virtual MVPDs, such as YouTube and HuluTV, which offer their services over the internet; and
- subscription video on demand services (‘SVODs’), such as Netflix, HULU and Amazon Prime. SVODs are vertically integrated and typically invest billions of dollars in creating original content.
Consumers usually subscribe to traditional MVPDs as part of service bundles (e.g. cable, wireless and phone services). Traditional MPVDs are dominant for access to television content, but their market share is declining steadily as consumers move to virtual MPVDs and SVODs. This reflects industry trends such as: (i) the rise of over-the-top vertically integrated video content services such as MPVDs and SVODs. These providers not only benefit from the internet to provide lower-cost, better targeted services to consumers, but also from reduced negotiating friction and better access to consumer data; (ii) declining MVPD subscriptions and revenues, arising from an increasingly competitive landscape created by the rise of virtual MVPD and SVODs. This may lead to reduced revenues to both traditional content creators and aggregators, at a time when the cost of producing content is increasing; (iii) MPVDs revenues are also under threat as a result of increased reliance on targeted, digital advertising in platforms such as Facebook and Google.
The $US 108 billion merger was presented by the merging parties as a reaction to these industry dynamics. Time Warner would be able to access AT&T’s customer data, allowing it to compete with SVODs in targeting its advertising. AT&T would gain access to extensive video content and an extensive advertising inventory which it can distribute through mobile devices and other innovative ways.
The Government admitted that this will be to the benefit of current AT&T subscribers, who will pay less for access to video content.
A second section discusses the legal standards applicable to mergers, and particularly to vertical mergers. This is set out in s. 7 of the Clayton Act, and prohibits mergers that may substantially lessen competition.
While in horizontal mergers there is a structural presumption of anticompetitive effects, in vertical mergers the Government must demonstrate that the merger is likely to have anticompetitive effects – i.e. the Government has to introduce sufficient evidence to show that the challenged transaction creates a sufficiently probably and imminent risk of such an effect. This requires it to pursue a ‘comprehensive inquiry’ into the future competitive conditions in a given market.
Given the difficulties in pursuing such an assessment, the allocation of burdens of proof is particularly important. The burden of proof operates as follows:
- the Government must establish a prima facie case by establishing the relevant product and geographic markets and showing that the proposed merger is likely to significantly reduce competition in the market. If this requirement is met:
- the burden shifts to the merging parties to show that the prima facie case inaccurately predicts the probably effects of the transaction. One way to do this is to show that efficiencies outweigh the anticompetitive effects. If the prima facie case is thus rebutted, the burden of producing additional evidence of anticompetitive effects moves to the government, and merges with the burden of persuasion which remains with the government at all times.
It is widely acknowledged that many vertical mergers create vertical integration efficiencies between purchasers and sellers. Since in this case the Government admits that there are efficiencies flowing from the merger, this requires a balancing of anti- and pro-competitive effects.
A third section analyses each of the three interrelated theories of harm advanced by the government. All these theories are advanced despite the government acknowledging that the merger will result in significant benefits to customers of the merged entity, mainly in the form of the elimination of double marginalisation created by AT&Ts payment of a profit margin for accessing Turner’s contents.
The relevant markets – which were not vigorously challenged in court – are c. 1,100 local markets for multichannel video distribution provided by MVPDs and virtual MPVDs.
- A first theory of harm was that the merger would allow Turner to charge AT&T’s rival distributors higher prices for its content. The argument is that Turner’s content is a ‘must have’. Should Turner fail to strike a deal with a distributor, it would still benefit from AT&Ts distribution – which, would in turn, expect to gain additional subscribers as a result of its distribution of Turner’s content. This would allow the merged entity to increase its leverage and increase the prices charged by Turner to AT&Ts competitors (the court describes this as an ‘increased leverage theory’).
The court held that the DoJ failed to show that the proposed merger is likely to increase its bargaining leverage. In particular, the government did not argue that the merger would lead to blackouts. Blackouts occur when a distributor loses access to a network’s contents. This may lead to licensing and advertising revenue for a programmer, and the loss of subscribers for a distributor. As such, the Government did not argue that the merger would foreclose AT&Ts rival distributors. Instead, it argued that the merger would lead to higher prices for these distributors’ customers.
However, the court found that the evidence adduced – in the form of internal documents and regulatory submissions of the merging parties – did not support a finding that Turner’s contents were ‘must have’. While this content is popular, the evidence in support for the increased leverage theory does not suffice for the Government to meet its burden of proof that the merger can lead to increased leverage. This is compounded by evidence showing that prior vertical integration of programmers and distributors has not affected affiliate fee negotiations, and, hence, that prior experience does not support the Government’s theory in this case.
Lastly, the deployment of an economic model (by Carl Shapiro), which identified a $0.27 harm to consumer per month flowing from the merger, was found to rest on a number of assumptions that were found to be ‘implausible and inconsistent with the evidence’.
- A second theory of harm was that the merger would create an increased risk that the merged firm would act – either unilaterally or in coordination with Comcast-NBCU – to thwart the rise of lower cost virtual MVPDs, which are currently threatening the traditional video distribution model.
As regards the merged entities’ unilateral conduct, the court began by noting that the Government’s expert own model seemed to find such a theory implausible, because it would be more profitable for the merged entity to continue to license Time Warner content to virtual MVPDs. This was supported by additional evidence that AT&T would have an incentive to license Time Warner content to virtual MPVDs – in order to distribute its content as far and wide as possible, benefit from their growing in relevance, and allow AT&T to go mobile with video content (which is the merger’s main rationale).
As regards the coordinated conduct theory, the Government relies on a key assumption: that as the only two vertically integrated traditional MVPDs, Comcast and AT&T would share an incentive to slow the entry and growth of virtual MVPDs. This would require them to mutually forbear from licensing programming content without prior communication. The court held that the Government did not put forward sufficient evidence to prove more than a theoretical possibility of coordination, and ignored that both AT&T and Comcast stood to lose large amounts of licensing fees and advertising revenues by withholding content from virtual MVPDs. Lastly, AT&T has 100 million wireless subscribers, while Comcast is more of a cable infrastructure. This means that, by licensing Time Warner’s content to virtual MVPDs, AT&T gains a competitive advantage over Comcast. This lack of shared incentives further argues against coordinated effects flowing from the merger.
- The last theory of harm was that the merger entity could harm competition by preventing AT&T’s rival distributors from using HBO as a promotional tool to attract and retain customers – either by foreclosing access or, at least, access on competitive terms.
The court held that the Government’s claim fails for two reasons. First, it failed to show that the merged entity would have any incentive to foreclose rivals’ access to HBO-based promotions – particularly given how reliant on wide distribution HBO’s business model is, since its revenues flow exclusively from subscribers. Second, the Government failed to establish that HBO promotions are so valuable that withholding them restricting to other distributors would drive customers to AT&T.
Comment: In short, it strikes me that the merger was allowed on two main grounds: (i) the Government had the burden of proving that harm flowed from the merger, and that burden was not met; (ii) this assessment was coloured by the assumption that the rational behaviour for the vertically integrated merged entity would be to keep network licensing and distribution separate in order to maximise profits – i.e. that the merging party’s behaviour would remain broadly the same towards third-parties, and that hence there would be no anticompetitive effects.
Once this is established, the DoJ had a steep hill to climb in order to prove that the merged entity would act any differently towards its customers because of increased vertical integration.
For those interested in it, Steve Salop provides an interesting – and rather critical – overview of this decision here.