7 Reasons Why The AT&T/TW Trial Matters So Much The Future of Antitrust (+1 for Appeal).

Starting this week, AT&T and Time Warner get their day in court to prove that their proposed merger does not violate the anti-trust laws. I outlined the basic line of reasoning in the government’s case back shortly after it became clear the government intended to oppose. Since then, the parties have engaged in discovery, lined up their experts, and now filed their pre-trial briefs outlining their arguments on the relevant issues and standards. You can read the AT&T pre-trial brief here, and the DoJ pre-trial brief here.


It’s a lot easier to outline what the parties will try to show, and their differing strategies for trying to show it, than it is to guess how Judge Leon will decide at this point. But while the outcome alone makes this pretty important, it has the potential to massively shape antitrust going forward (assuming antitrust law survives the Supreme Court’s upcoming decision in Ohio v. American Express). Below, I unpack what makes this case so potentially important from a law perspective.


Why Is This Super Important?


Whether the deal goes through or not is super important for all the usual reasons relating to media concentration, competition in telecommunications, and all that other stuff I usually care about here at ToTSF. But the AT&T/TW trial raises a lot of super important questions for the future of antitrust enforcement. Specifically, does antitrust law care about vertical integration or not? If antitrust law no longer cares about vertical integration, or adopts a presumption that makes it harder (than it already is) to prove a case challenging a vertical merger, that will have huge impact on what kind of deals we see move forward — not just in the entertainment industry or telecommunications industry but pretty much in all industries.


By contrast, if the DoJ wins — especially on the coordinated effects argument (I’ll get to that later) — it may put the breaks on major merger activity (at least while Delrahim is at DoJ). It will have ripple effects beyond whether Comcast tries to buy News Corp assets. It would potentially have substantial impact on future acquisitions by platforms such as Amazon, Facebook and Google as these platforms seek to expand from their existing markets into new markets.


Also at stake is the question of whether or not the government needs to consider a proposed behavioral remedy when contemplating a merger, or whether the DoJ can simply decide “nope, its divestiture or nothing.” Likewise, since everyone keeps pointing back to Comcast/NBCU, it raises the question of whether a settlement creates some sort of legal precedent (either with regard the nature of the market or the nature of any proposed remedy). Finally, the rather important question of what you need to prove (and how you have to prove it) is subject to a lot of argument in the pre-trial briefs.


Sorry, I know You Covered This In Your Blog Post Back In November, But Remind Me What “Vertical Integration” Is?


Antitrust generally talks about two types of mergers. In a “horizontal” merger, one company buys out another company that makes the same kind of product or does the same kind of service or otherwise competes directly with the acquiring company. For example, when AT&T tried to buy T-Mobile, that was a “horizontal” merger, since both companies sold wireless mobile voice and data services and were major competitors to each other.


By contrast, a “vertical” merger involves a company buying another company that either makes a critical input for the first company or sells a critical input for the first company. The classic example is if I own a steel mill, and then I decide to buy a coal mine, so I will have a permanent supply of coal for my steel mill. That’s a “vertical” merger because the coal is an upstream input to my steel mill. By contrast, if I were to buy another steel mill, that would be a horizontal merger. This brings us back to issue #1.


Issue #1: Does Antitrust Law Care About Vertical Integration? And Does It Have a Different Standard Than Horizontal Integration?


It’s clear that in theory, either horizontal integration or vertical integration can increase my ability to control prices in the market to the detriment of competition and consumers. If I control the existing supply of local coal, or the existing railroad line used to bring the coal to the steel furnaces, I can raise the price of coal to my rivals. Since I am giving the coal to myself for free* (*yes, I know it’s not really “for free” in that there is opportunity cost, but let’s not get too bogged down), I get to lower my prices and attract more customers at the expense of my rival steel mills. They either have to eat the excess cost I’m charging or raise their price, thus losing customers. Soon I run them out of business and control all the steel mills. BWAHAHAHA!!!!


In horizontal monopolization, I just buy up all the steel mills and become a steel monopoly. BWAHAHAHAHA!!!


But modern antitrust treats these things somewhat differently. For one thing, it turns out that there are lots of coal mines, and that the cost of getting into the coal mining industry is reasonably cheap. So if I start to raise the cost to my rivals, they can switch to other coal suppliers.  So while I am able to drop the price of steel by giving myself free coal (add in the production cost of the coal, but deduct the profit the coal mine would normally include in his price), I can’t effectively drive my rivals out of business. To the contrary, my purchase of the coal mine is in theory an efficiency that lets me lower the price of coal, benefiting all consumers. Yeah consolidation! All praise the University of Chicago!


Most antitrust economists treat all vertical mergers like this highly simplistic example a telecom blogger just made up — except sometimes they use more math. This is why you will hear people say that while vertical mergers can be good because they produce efficiencies — they can never be bad because foreclosure is effectively impossible because I can always assume more coal mines. If that sound abysmally stupid, I invite you to read Justice Gorsuch’s comments in the Supreme Court’s recent antitrust case Ohio v. American Express, where Gorsuch asserts that it is essentially impossible to have vertical foreclosure. Gorsuch also goes on to say that even if it were possible, we would be better off allowing a monopoly which would ultimately “fix itself” whereas a “false positive” would potentially deprive consumers of efficiencies. (Yes, the Koch Brothers totally got their money’s worth on Gorsuch.)


Unsurprisingly, AT&T spends much of its time in its brief arguing that verticals are like good fairy god mothers who only bring us wonderful efficiencies and never, ever confer the ability to foreclose and raise the cost to rivals. As a result, AT&T argues, there is a strong presumption against finding that a vertical transaction violates the antitrust laws.


DoJ argues the opposite. DoJ argues that, under the law, there is no difference in how to analyze a vertical deal v. how to analyze a horizontal deal. Both use the same standard. First, the government must make a prima facie case that the merger is “likely to  substantially lesson competition.” Once a prima facie case is made, the burden shifts to the merging parties (here, AT&T and TW) to rebut (the government then gets to try again on reply). DOJ observes that while the conventional wisdom is that the majority of vertical mergers create efficiencies and don’t threaten competition, that is pretty much the same conventional wisdom for horizontal mergers. So even if the mechanism by which vertical integration would potentially “have the effect of substantially lessening competition” (to quote from the statute) is different from the standard way in which acquisition of marketshare directly in the relevant market may substantially lessen competition, the actual legal standard does not change.


What to watch for: This is one of those legal things that is both incredibly important for the legal mind but really difficult to put into words for normal people. In theory, of course, the government already has the burden of proof (at least for a prima facie case) in showing that the transaction may “substantially lessen competition.” It is also the fact that, as a practical matter, the tools used for showing this for a horizontal merger are more generally recognized and easier to use. What AT&T is looking for is a formal recognition as a legal doctrine of the generally accepted idea that because there are usually multiple sources for inputs needed in the market, antitrust experts and courts have shown skepticism to anti-trust claims against vertical combinations. DoJ’s counterpoint is there is a difference between a rule of thumb explaining why cases against verticals proceed differently and rely on specific circumstances of the relevant markets rather than relying on the standard increase in concentration in the market the way horizontal merger cases do.


If the court were to find an actual legal presumption along the lines sought by AT&T, this would be a very big deal for antitrust generally. But even if the the court does not adopt a formal presumption about the standard, everyone will be looking to see how skeptically Judge Leon evaluates the general arguments about how acquisition of an upstream company could substantially lesson competition in the relevant market.


Issue #2: Statements By The Parties v. Expert Testimony, or Fact v. Theory. 


If you simply flip through the briefs of the two parties, one difference will leap out immediately. AT&T has virtually nothing redacted in its brief. By contrast, the DoJ brief has huge blocks of black ink where the government has redacted confidential business information in the form of statements from the merging companies and from their chief competitors. AT&T spends a lot of time arguing how it is just laughable from a theoretical point of view that they could afford to withhold programming from rivals long enough to force a significant increase in price that would have any impact on competitors or cause any harm to consumers (AT&T in fact says they cannot force any increase in price, but they also argue that even accepting the testimony of the government’s witness the amount they could raise the price is too small to matter). DoJ, by contrast, uses a lot of quotes from AT&T and Time Warner (both directly merger related and from FCC proceedings such as the repeal of the Program Access Rules), as well as from competitors deposed for purposes of the trial, to demonstrate that everyone in the industry both understands how AT&T could use control of Time Warner programming to block new competitive entry and how it could raise the price of programming to existing competitors allowing AT&T to keep prices in the industry (including its own prices) artificially high, to the detriment of consumers.


On the one hand, you would think that fact trumps theory. If the parties have document where they spell out precisely how they can use their control of must have programming to crush competitors, and competitors believe this is true, that should settle the question, right? Turns out, not so much.


Under the statute, we don’t technically care whether the merging parties actually intend to “substantially lessen competition.” The statute prohibits mergers or acquisitions where “the effect of such acquisition may be substantially to lessen competition, or tend to create monopoly.” On the one hand, that is helpful to the government because it does not need to prove any kind of intent. The government does not have to show that AT&T affirmatively intends to crush competitors and drive up prices to consumers, it just has to show that this is a likely outcome (the word “may” as in “may be substantially to lessen competition” is one of those ambiguous words that can range from ‘that seems reasonable’ to “practically certain,” so I’ll just go with “likely” for purposes of this blog post).


But at the same time, that means that the statements of the parties or potential competitors about their assessment of the value of the property is not the last word. A competitor might legitimately be worried that a vertically integrated rival could raise the price to a level that makes it impossible to compete on price, but that doesn’t mean that when push comes to shove it turns out to be true. Similarly, a party might be convinced that it has a product that no one can live without, only to discover that folks can live without it just fine. Additionally, as AT&T will undoubtedly argue, context matters. Saying in the context of an enthusiastic sales pitch that HBO is proven programming and that a rival service such as Playstation Vue cannot possibly compete without it can be dismissed as “puffery” (legalese for “advertising bullshit”) that everyone knows not to take literally.


Nevertheless, as the DoJ will argue, there is a reason AT&T is willing to pay so much for this programming, and the fact that everyone in the industry says the same thing about the value of the programming and the ability of AT&T to withhold it long enough to win any price negotiation with a rival (or refuse to sell it to a potential new entrant) lends credence to the government’s economic experts and undermine the claims of AT&T’s economic experts.


What to watch for: The problem is, that unlike an enforcement action based on what companies did in the past, an action to block a merger under the antitrust laws requires the government to make a prediction about the future. And, as Yogi Bera famously said: “It’s tough to make predictions, especially about the future.”


Which is why the statutory language does not require the government to show that this will definitively lessen competition by allowing AT&T to use Time Warner programming to drive up the cost to rivals, or withhold it from potential new entrants. Under the statute, the government merely has to show that it “may” have the effect of substantially lessening competition or “tend” to create a monopoly. On the flip side, precedent has established that “may” means more than just “there is a possibility.” But as with so many things, this is squishy and hard to define in any real detail.


This is where, in my opinion, the documents and statements collected at deposition matter. Both AT&T and DoJ have economic experts with lots of theory as to why AT&T could or could not actually use their control of Time Warner programming in an anticompetitive fashion. The evidence of what AT&T/DIRECTV have said previously, as well as internal documents relating to the current acquisition and the depositions of competitors like DISH potentially undercut AT&T’s argument that it would be impossible for them to foreclose/drive up costs because they would lose too much revenue. If I were a generalist judge trying to evaluate the competing experts and their respective models I would be like “well, if everyone in the industry thinks this is something that can happen, then the government has met its burden of showing that this “may” substantially lessen competition. Sure, everyone might ultimately be wrong. But if everyone who actually knows this industry (including the folks now telling me the opposite) takes this seriously, than I as the judge ought to take this seriously.”


Issue #3: Do We Still Believe The Basic Cable Market Power Story? What Is The Relevant Market These Days?


As I observed in my blog post outlining the government’s case, the general case against AT&T/TW is based on the history of the 1992 Cable Act and a subsequent gloss from Comcast/NBCU. Basically, Multichannel Video Programming distributors, aka “MVPDs,” aka cable and any multichannel realtime linear programming provider, vertically integrated with cable programming use their power over “must have” programming to raise the price to rivals. In the most extreme case, these vertically integrated MVPDs simply refuse to provide the programming to competitors, making it impossible to compete. Alternatively, if the MVPD wants to maximize value of the programming and prevent rivals from competing on price, the MVPD raises the price of the programming so that it doesn’t have to compete on price against its rivals. There are lots of variations in between, depending on what kind of contract conditions the vertically integrated programmer imposes on the purchaser of the programming.


The Comcast/NBCU deal recognized three distinct markets: traditional MVPD providers, like DISH; over the top (OTT) “virtual” cable providers that provide linear programming similar to traditional cable (indeed, using traditional cable programming), such as YouTube Red or AT&T’s own OTT DIRECTV service; and video on demand streaming services such as Netflix and Prime. The Comcast/NBCU consent decree recognized virtual MVPDs as (at the time potential) competitors to traditional MVPDs, but classed streaming video on demand (sVODs) as a different market (competing with traditional cable video on demand).


Nearly 7 years after the Comcast/NBCU deal, the market has . . . sorta changed. Sure, it is still easy to see how virtual MVPDs compete with traditional MVPDs. What remains unclear is whether sVODs, or efforts by some programming networks to establish their own stand-alone streaming services, compete in the same market. Is the market all online video programming? Or is it still limited to MVPDs and virtual MVPDs? (And is competition limited because programmers won’t sell traditional linear cable programming to sVODs — as the DoJ and FCC found in the Charter/TWC Consent Decree when they imposed limits on the type of contractual provisions the combined firm could impose on independent programmers).


AT&T argues for the broadest possible market definition. This is the classic strategy. Back in the 1990s, cable operators argues that DVDs competed against cable programming. But the idea of the “attention marketplace” as indistinguishable from traditional television viewing has gained traction in recent years, making this broader definition slightly more plausible.


More importantly, of course, there is the whole “cord cutting” phenomena. But what does cord cutting actually show and how does it play into the analysis? As this article points out, cord cutting has increasing with the combination of climbing cost of traditional MVPDs and the availability of lower priced virtual MVPDs. First off, it’s very odd to see a product losing customers actually raise prices in the absence of market power. Which leads to the next question. Are sVODs also part of the “cord cutting” equation? Sure, people who drop traditional MVPDs generally have streaming services and spend more time streaming. But lots of people who have traditional MVPD subscriptions also subscribe to sVODs like Netflix. Usually, if products compete with each other, a consumer buys one or the other (something economists refer to as “substitutability”). If Netflix is genuinely a substitute for DIRECTV, why do consumers get both?


By contrast, we don’t usually see people subscribe to both a standard MVPD and a virtual MVPD, which would support the government’s contention that the relevant market is traditional and virtual MVPDs. But this is also complicated because some people do just drop traditional MVPD and only subscribe to a sVOD (or use a free streaming service such as YouTube). Can the government confidently say that the availability of Netflix and Hulu, with all its original programming, is not having some competitive effect?


But assume the government wins the market definition fight. It still has to prove that AT&T/DIRECTV, once vertically integrated with Time Warner, will have a greater capacity to raise the cost of programming than Time Warner is already doing. The traditional theory, which the government advances here, is that the virtual MVPDs are still a minuscule part of the market and cannot become real competitors without Turner programming. Denying Turner programming, or forcing virtual MVPDs to pay a premium for it so they cannot compete on price, becomes possible because Time Warner no longer has the same incentive to make its programming available to every distributor. The income lost from selling programming to competing platforms is more than made up by keeping customers who would otherwise leave if they could get Turner programming on a virtual competitor, or from capturing customers who leave when the programming is dropped during a “retransmission dispute.”


There is some circumstantial evidence to support the government’s claim. Notably, YouTube Red just paid a sufficiently high premium for Turner programming that it raised its fee from $45 to $50. But the fact that Turner Programming has value does not mean it is “must have” programming. Again, there will be lots of expert witnesses and evidence from parties and competitors for the Judge to weigh and evaluate.


What to watch for: Since 1992, media regulation has depended on a bunch of underlying assumptions, several of which are at play here.  But if the Judge decides that the market is different now, or that there is no longer such a thing as “must have” programming, that would have serious long-term implications for the future of media ownership regulation as well as antitrust. If mobile platforms and streaming services all compete in one big happy market, it potentially becomes much more difficult for a future FCC address things like cross-ownership or even straight up horizontal ownership limits.


Issue #4: Is There A Threshold Amount On Monopolization/Price Increase? Do Efficiencies Counter SSNIP?


If you read the antitrust statutes, you would wonder about the total obsession with whether antitrust behavior raises retail prices to consumers. However, as a number of folks have chronicled recently (see this by Lina Kahn for the scholarly version) antitrust law has narrowed considerably since Robert Bork’s book “The Antitrust Paradox.” I will not go into the whole thing here. Suffice it to say that it has now effectively become a mandatory element to show that the merger will allow the merged company to raise prices by what is called a “Small but Significant Non-Transient Increase in Price” (SSNIP) that gets passed on to consumers. Just crushing your competitors doesn’t count.


As detailed in the parties briefs and here and here, the government’s chief economic expert fond that a combined AT&T/TW could raise the price of Turner programming to bring in $463 Million/year in monopsony profits. The government also notes that AT&T/TW can do other things to make it harder for competitors (whoever they are) to compete, such as pricing or contract clauses that prevent competitors from using HBO as a loss-leader to entice customers to switch.


AT&T does three things in response. First, it  attacks the government’s economic study and claims they can’t possibly raise prices. Obviously, if they win on that, they win. In the alternative, AT&T argues that the government has shown a mere 45-cent/subscriber increase, which no one should care about. AT&T also claims to have lots of “efficiencies” which counterbalance any harm. Mind you, according to AT&T, the efficiencies are that it can do a much better job than Time Warner violating user privacy and delivering addressable ads, making both the content and AT&T’s own advertising business more valuable. How this is supposedly good for consumers, however, AT&T does not elaborate in its brief.


What to watch for: As the government points out, there is no such thing recognized in law as an “efficiencies defense” (whether the benefits outweigh the harms is more an FCC “public interest” balancing test kind of thing). Additionally, there is no such thing as a “de minimis” violation of the competition laws. If the government proves that the merged entity and create a SSNIP, that should do it. But the parties do get to try to rebut the government’s prima facie case, and there is precedent for the DoJ considering efficiencies when it opts to seek a consent decree. Still, the efficiencies here are extremely weak.


Bottom line, it would be a major precedent in antitrust law if the Judge recognized that sure, the merged parties could raise prices, but not enough to worry about (or that the benefits outweighed the harm). I don’t think it’s likely, but AT&T spends a lot of time trying to put it in play. For myself, I would love the opportunity to walk up to every AT&T and Time Warner executives and take 45 cents a month. Oddly, I am not allowed to decide that AT&T and TW executives ought to pay me 45 cents a month because ‘hey, they won’t even notice; it’s like some fraction of a latte.’ (For reasons unclear to me, “the latte” has become the standard unit of measurement for amounts of money supposedly too small to notice.)


Issue #5: Comcast/NBCU: Is It Relevant? Precedent?


Back in 2011, the DoJ permitted Comcast to buy NBCU, subject to a fairly extensive consent decree. AT&T has basically argued that it ought to get the same deal or better, since — as it argues — the Comcast/NBCU deal presented much more danger to competition. AT&T also argues that DoJ’s refusal to offer it the same terms as it offered Comcast/NBCU is super suspicious. DoJ, for its part, argues that the fact that it went after Comcast/NBCU shows it regarded that as a major threat to competition, but that the DoJ does not have to stick to behavioral remedies (i.e., consent decrees that regulate behavior) rather than demand divestiture. (DoJ offered to let AT&T divest itself of either Turner programming or DIRECTV to allow the deal to go through. AT&T was not interested.)


DoJ has tried to distinguish the Comcast/NBCU decision in other ways. Most notably, DoJ has pointed out that when it did Comcast/NBCU, it had the FCC as a partner in the remedy. I will get to that a bit more when I talk about AT&T’s proposed remedy.


What to watch for: Judge Leon actually presided over the Comcast/NBCU consent decree, and took the extraordinary step in a Tunney Act hearing of pushing back on DoJ to show that a consent decree was in the public interest. DoJ did, in fact, lean heavily on the FCC’s involvement in the conditions and enforcement of the conditions, and Judge Leon ultimately signed on. Whether to read that as weighing in favor of viewing the Comcast/NBCU as some kind of precedent in which Judge Leon feels invested, or whether Judge Leon views this as an opportunity to say “I told you so” by finding that the deal needs to be enjoined, remains to be seen.


To the extent we have any information, it looks like Judge Leon actually doesn’t want to talk about Comcast/NBCU. On February 20, Judge Leon rejected AT&T’s discovery request related to its political interference argument. In the Order, Judge Leon dropped a footnote saying he “declined the invitation” to compare this transaction to Comcast/NBCU, and that AT&T would have lots of opportunity “to prove ‘how weak this case is’ during the upcoming trial.” As I read this, Judge Leon is making it clear to anyone paying attention that he plans to focus on the merits of the current case and ignore Comcast/NBCU as irrelevant.


Issue #6: AT&T’s Proposed Remedy: Do You Get To Chose Your Own Remedy? Does The Government Have To Consider It?


AT&T touts its offer of an irrevocable agreement for the next 7 years to engage in “baseball-style” arbitration with any purchasers of Time Warner programming that think AT&T is abusing its vertical integration. This mirrors the primary condition from Comcast/NBCU with regard to access by competing MVPDs to the vertically integrated programming. AT&T argues that this basically takes care of any potential problem with the deal.


DoJ argues two things. First, the government has discretion to chose whether to accept a proposed behavioral remedy or not. Just because AT&T thinks this should solve everyone’s potential concerns does not mean that AT&T gets to make that decision. DoJ argues that it, as the prosecuting agency, gets to decide whether or not to offer a behavioral remedy. Additionally, DoJ has a list of reasons why it thinks AT&T’s proposal does not come close to addressing all their concerns.


What to watch for: DoJ has its own motion for the court to exclude consideration AT&T’s proposed remedy. Even if the Court denies the motion that doesn’t tell us whether or not Judge Leon will decide that AT&T’s proposed remedy is relevant. In a standard criminal case, the government is not required to consider my offer of a plea bargain, no matter how much I might like to make it or how fair I think it is. Even in a standard civil case, I can offer to settle, but nothing requires to other side to accept – or even consider – my settlement offer.


AT&T basically invites the court to substitute its evaluation of what would serve the public interest rather than have DoJ prove its case. A requirement that DoJ (or FTC, for that matter) must consider whatever remedy merging parties offer, and that the court will evaluate whether or not the DoJ was reasonable to refuse the proffered remedy, would be a major change in the law and would seriously hamstring the ability of the antitrust agencies to bring cases blocking mergers. The antitrust agencies would be constantly second guessing themselves on whether they adequately considered a proffered remedy and whether the remedy “adequately” protects the public from the harms to competition.


Issue #7: The Return of the Coordinated Effects Doctrine?


Legal theories around antitrust go into and out of fashion, even if they don’t get officially overturned. One of these is the idea of “coordinated effects.”


First, remember the statutory language. The question is whether the merger “may have the effect of substantially lessening competition.” Usually, people think of that as giving the applicants sufficient power in the market to impose a SSNIP, or otherwise mess with competition on its own. But another way a merger can have the effect of reducing competition is by dropping the number of competitors in the market to a small enough number of like-minded players that they can coordinate with each other without explicitly conspiring to fix prices or otherwise act in an anticompetitive fashion.


How does that work? I’m old enough to try to explain to people about the card game bridge, which works on the idea that I and my partner know each other well enough that we can communicate lots of information using a relatively small set of permissible signals. But no one plays bridge anymore, so I will try something else.


Assume that you and I would like to work together, but we can’t actually talk to each other. We have to figure out what the other person will do based on what we see that person doing. Meanwhile, the other person is trying to figure out the same thing for us. If the two of us know each other fairly well, we can work this out over time. If I point to the dishes, you know to put them away. If you point to the table, I know to set the table for dinner. We don’t have to get together to agree on the signals. Context and our knowledge of each other lets us figure it out.


Long ago, a mathematician named Nash figured out that if people who know each other play the same game with each other over time, they learn all kinds of things about each other. They can then test certain strategies. If we are playing a high-low poker game, and I know you always go high, I can go low and we split the pot. Or, if you can tell from my exposed cards that I am the only player who could win the high hand, the real contest is between the players competing for the low hand and I can jack up the pot with no worries.


Companies work the same way. They try to figure out how they will each react to particular moves in the market. Because they all know lots of information about each other, and know lots of information about how the market works, they can decide to divide the market between them without communicating. This is sometimes called “conscious parallelism.” It’s why you never see major cable companies compete with each other, and why the landline telephone companies never overbuilt each other’s territory. Each one knows that if they overbuild the other one, then the other provider will overbuild their own market and they will have to compete. However, as long as I don’t build in your territory, you won’t build in mine. We don’t have to get together to have an explicit agreement not to compete. It just makes good business sense. Furthermore, because it takes a long time to overbuild, I will have plenty of warning if you do try to compete directly and can plan all sorts of retaliatory tricks to cause you big short term loses. So the combination of few players and a particular market structure make it possible for us to coordinate with each in a way that is good for us, but bad for competition and consumers.


DoJ claims that if AT&T buys out Time Warner, it will basically become enough like Comcast/NBCU that the two of them can jointly use their combined horizontal and vertical integration to crush disruptive competition. On the one hand, it takes a particular combination of market factors and a high enough level of concentration to make coordinated effects possible. On the other hand, that is pretty much exactly how the traditional cable market has operated for most of its history. AT&T, consistent with its claim that the current market structure is broad and dynamic, argues that it could not possibly engage in coordinated effects with Comcast/NBCU. DoJ, arguing that the market structure is still similar enough to what it has always been to make the traditional antitrust concerns real, argues that the market remains susceptible to coordinated effects if MVPDs can once again achieve high enough levels of vertical integration.


What to watch for: Just as this transaction tests whether we care about vertical integration, it also provides a good test case about whether coordinated effects remains a serious doctrine. Given that we have a significant number of industries where critics claim we see coordinated effects in action, to the detriment of competition and consumers, this is a major deal. For example, critics of the airline industry argue that because we have two massive airlines, one major discount airline (Southwest), and one lesser discount airline (JetBlue), we have coordinated effects in the airline industry that allow airlines to keep prices up and fees high. So if the court reaches the coordinated effects issue, and decides in favor of DoJ, we could see this moving to other industries such as the pharmaceutical industry or airline industry.


Epilogue: Political Interference — The Dog That Didn’t Bark (But Maybe On Appeal).


AT&T has invested a bunch of time trying to make this case about whether Trump pushed DoJ to block the merger as revenge for unfavorable CNN coverage. As a legal argument however, this faced a fairly high hurdle. On February 20, Judge Richard Leon (the presiding judge) issued an Order expressing extreme skepticism about the vitality of this “selective enforcement” argument. Technically, the Order merely rejected AT&T’s discovery request for access to communications from the White House with DoJ which they claimed would show whether or not political interference occurred. But the language of the Order made it very clear Judge Leon simply did not buy the idea that this case represented such a departure from the norms of enforcement that AT&T deserved to examine normally privileged communications and documents relating to enforcement. Since then, AT&T has dropped Antitrust Division Head Makan Delrahim from the list of potential witnesses and did not mention its “selective enforcement” argument in its pre-trial brief.


Judge Leon’s rather forceful rejection underscores that he does not consider DoJ’s arguments as meritless. In a separate footnote, Judge Leon also indicated that he did not consider comparisons with Comcast/NBCU particularly relevant and made it clear he planned to judge the case based on the evidence presented at trial rather than with reference to the DoJ’s willingness to offer Comcast/NBCU behavioral remedies in a consent decree. Still, while a victory for DoJ, it does not do to read too much into this. The standard for a selective enforcement defense, as Judge Leon pointed out, is extremely high. Additionally, while rejecting the idea that AT&T is entitled to the same remedy as Comcast, the footnote and the language of the Order make it clear that Judge Leon plans to make this a very fact specific inquiry. So while AT&T can’t argue that it is entitled to the same treatment as Comcast, DoJ can’t argue that the fact that it forced Comcast to accept a consent decree (and Judge Leon, as presiding judge, accepted the consent decree) demonstrates that such a vertical merger automatically violates the antitrust laws either.


So has AT&T entirely given up on its selective enforcement claim? Not necessarily. Several former DoJ officials filed an amicus brief arguing the court should consider the selective enforcement defense. While it is unclear that AT&T would appeal an adverse decision from the district court, it can certainly preserve the selective enforcement argument for appeal. But for now, at least, this line of argument looks like it will drop out of the district court case.




United States v. AT&T and Time Warner would be exciting in its own right, it has potential major repercussions for antitrust enforcement and the future of media regulation. Unfortunately, it is a lot easier to describe the flashpoints and issues than to guess how this case will come out.


Stay tuned . . . .


  1. A Harold Feld sausage factory classic. 5 stars. In the sequel perhaps he could run down the main witnesses.

  2. This is a tough case. Let’s just wait and see for the result.

Comments are closed