Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 1 of 76 UNITED STATES DISTRICT COURT FOR THE DISTRICT OF COLUMBIA UNITED STATES OF AMERICA, Plaintiff, v. AT&T INC., DIRECTV GROUP HOLDINGS, LLC, and TIME WARNER INC., Defendants. ) ) ) ) ) ) Case Number 1:17-cv-02511-RJL ) ) ) ) ) ) PRETRIAL BRIEF OF DEFENDANTS AT&T INC., DIRECTV GROUP HOLDINGS, LLC, AND TIME WARNER INC. March 9, 2018 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 2 of 76 TABLE OF CONTENTS Page TABLE OF AUTHORITIES ......................................................................................................... iv INTRODUCTION AND SUMMARY ............................................................................................1 THE RATIONALE OF THE MERGER .......................................................................................11 A. Overview of the Merging Parties and the Video Programming Industry ..............12 1. Time Warner: Video Programming Creation and Aggregation ................12 2. AT&T: Video Distribution .......................................................................14 B. The Video Marketplace Is Transitioning from the Television Era to an Intensely Competitive Internet Video Era.........................................................15 C. The Merger Addresses Specific Competitive Challenges Each Entity Faces in the New Video Marketplace ....................................................................19 D. The Merger Will Create Significant Pro-Competitive Synergies That Will Result in Significant Consumer Benefits ..............................................23 LEGAL STANDARDS .................................................................................................................27 A. The Government Must Prove That the Merger Will Cause “Probable” and “Substantial” Harm to Competition ................................................................27 B. The Government Bears the Burden of Proof on All Issues and Does Not Benefit from Any Presumption of Harm ........................................................28 C. The Government Must Define Relevant Markets, Including a Programming Market, and Prove Substantial Market Power in All Relevant Markets .............................................................................................29 ARGUMENT .................................................................................................................................30 I. THE GOVERNMENT’S THEORY OF TURNER PRICE INCREASES FOR RIVAL DISTRIBUTORS FAILS ON BOTH THE FACTS AND THE LAW ................31 A. The Government Will Not Prove Even the Trivial Price Increase Its Bargaining Model Ostensibly Predicts ..................................................................32 1. The government’s model does not reflect the realities of bargaining in this industry .........................................................................32 i Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 3 of 76 B. C. II. 2. The arbitration/no-blackout mechanism and the FCC’s “Program Access Rules” defeat the government’s priceincrease theory ...........................................................................................36 3. Correcting the data inputs both eliminates any predicted price increase and shows the unreliability of the model .....................................41 4. The government’s model does not adequately account for effects of the merger that will drive prices down ......................................45 The Government’s Predictions of Minor Price Increases Are Insufficient as a Matter of Law To Constitute “Substantially Lessened Competition” ..........................................................................................46 1. The projected price increases are tiny, highly sensitive to minor input variations, and competitively immaterial .........................................46 2. The government will not show that the merger will stunt MVPD rivals and thereby enable DIRECTV to raise its own retail prices .........................................................................................47 3. The government also will not meet its burden of defining a relevant programming market and proving that Turner has substantial market power ...........................................................................50 The Negligible Predicted Price Increases Are Swamped by the Merger’s Consumer Benefits .................................................................................52 THE GOVERNMENT’S THEORY THAT THE MERGED ENTITY WILL LIKELY COORDINATE WITH COMCAST TO WITHHOLD PROGRAMMING FROM VIRTUAL MVPDS ALSO FAILS ........................................54 A. The Government Must Prove That the Firms Would Have Both the Incentive and the Ability To Coordinate in Depriving Virtual MVPDs of Their Content .....................................................................................................55 B. Even the Government’s Expert Does Not Opine That Coordinated Conduct Is Probable or Likely To Cause Harm .....................................................56 C. The Merged Firm Would Have No Incentive To Coordinate with Comcast To Thwart Virtual MVPD Growth .........................................................57 D. Multiple Obstacles Would Prevent the Parties from Successfully Coordinating To Withhold Content from Virtual MVPDs ....................................58 1. The arbitration/no-blackout mechanism precludes any ability to coordinate...............................................................................................59 ii Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 4 of 76 2. E. III. Staggered multi-year contracts preclude enforcing coordination ...............................................................................................60 Past Information Sharing on Distinct Issues Does Not Establish an Incentive or Ability To Coordinate with Comcast To Deny Content to Virtual MVPDs ......................................................................................................60 THE GOVERNMENT’S THEORY THAT THE COMBINED COMPANY WILL LIMIT HBO’S USE AS A “PROMOTIONAL TOOL” IS ILLOGICAL AND CONTRARY TO THE EVIDENCE .................................................62 CONCLUSION ..............................................................................................................................64 iii Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 5 of 76 TABLE OF AUTHORITIES * CASES Page AD/SAT, Div. of Skylight, Inc. v. Associated Press, 181 F.3d 216 (2d Cir. 1999).........................30 Alberta Gas Chems. Ltd. v. E.I. Du Pont de Nemours & Co., 826 F.2d 1235 (3d Cir. 1987) ...............................................................................................................29, 47 * Ambrosini v. Labarraque, 101 F.3d 129 (D.C. Cir. 1996) ............................................................33 AT&S Transp., LLC v. Odyssey Logistics & Tech. Corp., 22 A.D.3d 750 (N.Y. App. Div. 2005) .......................................................................................................36 * Auburn News Co. v. Providence Journal Co., 659 F.2d 273 (1st Cir. 1981) ..........................30, 50 Ball Mem’l Hosp., Inc. v. Mutual Hosp. Ins., Inc., 784 F.2d 1325 (7th Cir. 1986) .......................29 * Brooke Grp. Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209 (1993)..........................55 Brown Shoe Co. v. United States, 370 U.S. 294 (1962) ..........................................................49, 50 Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477 (1977) ............................................47 * Comcast Cable Commc’ns, LLC v. FCC, 717 F.3d 982 (D.C. Cir. 2013)...............................28, 47 Ford Motor Co. v. United States, 405 U.S. 562 (1972) .................................................................50 * Fruehauf Corp. v. FTC, 603 F.2d 345 (2d Cir. 1979) .............................................................48, 49 * FTC v. Arch Coal, Inc., 329 F. Supp. 2d 109 (D.D.C. 2004) ..................................................27, 61 * FTC v. Cardinal Health, Inc., 12 F. Supp. 2d 34 (D.D.C. 1998) ............................................29, 50 FTC v. Indiana Fed’n of Dentists, 476 U.S. 447 (1986) ...............................................................51 FTC v. Tenet Health Care Corp., 186 F.3d 1045 (8th Cir. 1999) .................................................28 Hermanowski v. Acton Corp., 580 F. Supp. 140 (E.D.N.Y. 1983), aff ’d in part and rev’d and remanded in part, 729 F.2d 921 (2d Cir. 1984) .........................................36 Hospital Corp. of Am. v. FTC, 807 F.2d 1381 (7th Cir. 1986) ..........................................49, 56, 59 * Jefferson Parish Hosp. Dist. No. 2 v. Hyde, 466 U.S. 2 (1984) ..............................................48, 51 * Authorities upon which we chiefly rely are marked with an asterisk. iv Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 6 of 76 * McWane, Inc. v. FTC, 783 F.3d 814 (11th Cir. 2015), cert. denied, 136 S. Ct. 1452 92016) ......................................................................................................48 Monsanto Co. v. Spray-Rite Serv. Corp., 465 U.S. 752 (1984) .....................................................61 National Fuel Gas Supply Corp. v. FERC, 468 F.3d 831 (D.C. Cir. 2006) ............................28, 47 * Oracle Am., Inc. v. Google Inc., 798 F. Supp. 2d 1111 (N.D. Cal. 2011) .....................................33 * ProMedica Health Sys., Inc. v. FTC, 749 F.3d 559 (6th Cir. 2014) ..............................................29 * Republic Tobacco Co. v. North Atl. Trading Co., 381 F.3d 717 (7th Cir. 2004) ....................28, 51 Schurz Comm’ns, Inc. v. FCC, 982 F.2d 1043 (7th Cir. 1992)........................................................2 Silverstein v. United Cerebral Palsy Ass’n of Westchester Cnty., Inc., 17 A.D.2d 160 (N.Y. App. Div. 1962) ..............................................................................36 Suture Express, Inc. v. Owens & Minor Distrib., Inc., 851 F.3d 1029 (10th Cir.), cert. denied, 138 S. Ct. 146 (2017) ....................................................................................29 * United States v. Baker Hughes, Inc., 908 F.2d 981 (D.C. Cir. 1990) ................................27, 49, 56 United States v. E.I. du Pont de Nemours & Co., 353 U.S. 586 (1957) ............................30, 49, 50 United States v. Engelhard Corp., 126 F.3d 1302 (11th Cir. 1997) ..............................................30 United States v. Marine Bancorp., 418 U.S. 602 (1974) ...................................................27, 30, 50 * United States v. Microsoft Corp., 253 F.3d 34 (D.C. Cir. 2001) ...................................................48 United States v. Oracle Corp., 331 F. Supp. 2d 1098 (N.D. Cal. 2004)..................................27, 56 United States v. Philadelphia Nat’l Bank, 374 U.S. 321 (1963) ...................................................48 * VirnetX, Inc. v. Cisco Sys., Inc., 767 F.3d 1308 (Fed. Cir. 2014) ..................................................33 Westman Comm’n Co. v. Hobart Int’l, Inc., 796 F.2d 1216 (10th Cir. 1986) ...............................62 ADMINISTRATIVE DECISIONS B.F. Goodrich Co., In re, 110 F.T.C. 207 (1988) ....................................................................55, 56 Eighteenth Report, Annual Assessment of the Status of Competition in the Market for the Delivery of Video Programming, 32 FCC Rcd 568 (2017) ..............................14, 15 Fourteenth Report, Annual Assessment of the Status of Competition in the Market for the Delivery of Video Programming, 27 FCC Rcd 8610 (2012) ..................................13 v Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 7 of 76 Memorandum Opinion and Order, Applications of AT&T Inc. and DIRECTV for Consent to Assign or Transfer Control of Licenses and Authorizations, 30 FCC Rcd 9131 (2015) ...................................................................................................40 Memorandum Opinion and Order, Applications of Charter Communications, Inc., Time Warner Cable Inc., and Advance/Newhouse Partnership, for Consent to Assign or Transfer Control of Licenses and Authorizations, 31 FCC Rcd 6327 (2016) ................................................................................................................40 Memorandum Opinion and Order, Applications of Comcast Corp., General Elec. Co. & NBC Universal, Inc. for Consent to Assign Licenses and Transfer Control of Licenses, 26 FCC Rcd 4238 (2011)......................................................26, 38, 39 Memorandum Opinion and Order, General Motors Corp. and Hughes Electronics Corp., Transferors, and The News Corp. Ltd., Transferee, 19 FCC Rcd 473 (2004) ..........................................................................................................................39 Memorandum Opinion and Order, News Corp. and The DIRECTV Group, Inc., Transferors, and Liberty Media Corp. Ltd., Transferee, 23 FCC Rcd 3265 (2008) .................................................................................................................................39 Tenth Annual Report, Annual Assessment of the Status of Competition in the Market for the Delivery of Video Programming, 19 FCC Rcd 1606 (2004) .....................13 STATUTES AND REGULATIONS Clayton Act, 15 U.S.C. § 12 et seq. ......................................................................27, 41, 46, 47, 48 § 3, 15 U.S.C. § 14 .............................................................................................................48 § 7, 15 U.S.C. § 18 .....................................................................................27, 41, 46, 47, 48 47 U.S.C. § 548(c)(2)(A-B) ...........................................................................................................40 47 U.S.C. § 548(j) ..........................................................................................................................40 47 C.F.R.: § 76.1002(a-b) ....................................................................................................................40 § 76.1003(h) .......................................................................................................................40 § 76.1004............................................................................................................................40 OTHER MATERIALS Phillip E. Areeda & Herbert Hovenkamp, Antitrust Law: Vol. IIIB (3d ed. 2008) ......................................................................................................28 vi Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 8 of 76 Vol. IV (4th ed. 2016) ........................................................................................................55 Vol. IVA (4th ed. 2016) .....................................................................................................62 Vol. VII (4th ed. 2017) ......................................................................................................49 Robert H. Bork, The Antitrust Paradox (1978) .......................................................................29, 47 Comcast-NBCU Annual Compliance Reports to the FCC 2012-2016, http://corporate.comcast.com/news-information/nbcuniversal-transaction .......................38 Compl., United States v. Charter Commc’ns, Inc., No. 1:16-cv-00759-RCL, (D.D.C. filed Apr. 25, 2016), ECF No. 1 ...........................................................................11 Consumer Intelligence Research Partners, LLC Press Release, Amazon Prime Hits 90 Million US Members (Oct. 18, 2017)..............................................................................5 Scott Galloway, Silicon Valley’s Tax-Avoiding, Job-Killing, Soul-Sucking Machine, Esquire (Feb. 8, 2018)..........................................................................................4 Douglas H. Ginsburg, Vertical Restraints: De Facto Legality Under the Rule of Reason, 60 Antitrust L.J. 67 (1991) .............................................................................29, 47 Alex Heath, Facebook and Google completely dominate the digital ad industry, Bus. Insider (Apr. 26, 2017) ................................................................................................9 Hulu Press Release, Hulu Grows to Over 17 Million Subscribers in 2017, Emerges As Powerful Pay TV Alternative Combining Live Television with the Largest SVOD TV Library in the U.S. (Jan. 9, 2018).............................................4 Interview by Jon Steinberg, Cheddar TV, with Charlie Ergen, CEO, DISH Network (Sept. 27, 2017) ........................................................................................52 David Lieberman, Dish Network’s Charlie Ergen Calls Loss of CNN and Turner Nets a “Non-Event,” Deadline.com (Nov. 4, 2014) ..........................................................52 Tripp Mickle, Apple Readies $1 Billion War Chest for Hollywood Programming, Wall St. J. (Aug. 16, 2017) ..................................................................................................4 Rani Molla, Netflix Spends More on Content Than Anyone Else on the Internet – and Many TV Networks Too, Recode (Feb. 26, 2018) .........................................................4 Netflix Letter to Shareholders (Jan. 22, 2018).................................................................................4 NFLX – Q4 2017 Netflix Inc Earnings Video Interview, Thomson Reuters StreetEvents, Edited Transcript (Jan. 22, 2018) ..................................................................4 Aaron Pressman, Cord Cutting Hits Another Record, Bashing Cable and Telecom Stocks, Fortune (Mar. 1, 2018)...........................................................................................17 vii Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 9 of 76 Shalini Ramachandran & Keach Hagey, Dropping Viacom Costs Suddenlink Few Subscribers, Wall St. J. (Feb. 24, 2015) ............................................................................42 Roundtable on Vertical Mergers – Note by the U.S. Delegation to the Organization for Economic Co-operation and Development, Competition Committee (Feb. 15, 2007) ............................................................................................1, 29 Deepa Seetharaman, Facebook is Willing to Spend Big in Video Push, Wall St. J. (Sept. 8, 2017) ......................................................................................................................4 SNL Kagan: Profile: Amazon Prime Video (US) (Jan. 30, 2017)............................................................4 Profile: Hulu (Feb. 22, 2017) .............................................................................................4 Profile: Netflix (U.S.) (Mar. 6, 2017)..................................................................................4 Joan E. Solsman, Netflix Streamed to Almost Half a Billion Devices Last Month, CNET (Mar. 6, 2018) ...........................................................................................................4 Todd Spangler, Netflix Eyeing Total of About 700 Original Series in 2018, Variety (Feb. 27, 2018) ........................................................................................................4 George J. Stigler, A Theory of Oligopoly, 72 J. Pol. Econ. 44 (1964) ...........................................55 Thomson Reuters StreetEvents, CMCSA – Q4 2017 Comcast Corp Earnings Call, Edited Transcript (Jan. 24, 2018).......................................................................................57 U.S. Dep’t of Justice & Fed. Trade Comm’n, Horizontal Merger Guidelines (2010) .................................................................................................................................56 United States v. Comcast Corp., No. 1:11-cv-00106-RJL (D.D.C.): First Annual Rep. (filed Sept. 4, 2012), ECF No. 31 .........................................................38 Suppl. Stmt. of U.S. in Supp. of Entry of the Final J. (filed Aug. 5, 2011), ECF No. 26 ........................................................................................................................38 Tr. of Fairness Hrg. (filed July 27, 2011), ECF No. 38 .....................................................38 U.S. Dep’t of Justice, Competitive Impact Statement (filed Jan. 18, 2011), ECF No. 4 ..........................................................................................................................49 U.S. Resp. to Pub. Comments (filed June 6, 2011), ECF No. 23 ......................................38 viii Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 10 of 76 INTRODUCTION AND SUMMARY The transaction before the Court is a vertical merger of two companies that operate in highly competitive environments, but do not compete against each other. Time Warner sells television content through networks such as HBO, TNT, TBS, and CNN. AT&T pays Time Warner for the right to package those networks, along with hundreds more from an array of suppliers, in AT&T television products, including satellite-based DIRECTV and the lower-cost, mobile version of the same service called DIRECTV NOW. AT&T is the distributor with the consumer relationship; Time Warner is one of many suppliers for AT&T’s products. Modern antitrust law recognizes that mergers between suppliers, such as Time Warner, and distributors, such as AT&T, almost always create efficiencies and synergies that lead to lower consumer prices and greater innovation. For these reasons, as the government itself has explained, such vertical mergers “should be allowed to proceed except in those few cases where convincing, fact-based evidence relating to the specific circumstances of the vertical merger indicates likely competitive harm.” 1 This was never such a case, as discovery has confirmed. There is no fact-based evidence that this merger will harm competition. Nothing will be withheld from competitors; consumer prices will not go up. To the contrary, the government now concedes it would not be profitable for the new company to withhold its television networks from pay-TV distributors and that the new company’s prices to its own television customers will go down. As a result, the government’s suit to block this merger is not only baseless in fact, but it is affirmatively contrary 1 Roundtable on Vertical Mergers – Note by the U.S. Delegation to the Organization for Economic Co-operation and Development, Competition Committee 2, 10 (Feb. 15, 2007). 1 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 11 of 76 to consumer welfare, making it difficult for the government even to allege a viable antitrust claim, much less prove one. * * * * * When it filed its Complaint, the government’s principal theory of harm was that, after the merger, the combined firm could “more credibly threaten to withhold” Time Warner’s Turner television networks (e.g., TNT, TBS, and CNN) from pay-TV distributors like Comcast and DISH, forcing them to pay more for Turner programming and raise their consumer prices, in turn enabling AT&T’s traditional pay-TV business, DIRECTV, “to raise its own [consumer] prices relative to what it could have” charged. Compl. ¶¶ 35, 38. Since then, the government’s lead economic expert has conclusively refuted that theory. Among other things, he has opined that the merged entity will not withhold the Turner networks because the licensing and advertising revenues it would lose would vastly exceed any gains it might theoretically receive from denying rivals access to that programming. He also has opined that, because the merger will achieve otherwise unattainable efficiencies, the combined firm will reduce its own consumer prices for DIRECTV compared to what it would have charged. Now, what remains of the government’s case, “like a Persian cat with its fur shaved, is alarmingly pale and thin.” Schurz Comm’ns, Inc. v. FCC, 982 F.2d 1043, 1050 (7th Cir. 1992). Today, the government claims that the merger will enable Turner to charge its pay-TV providers higher prices for Turner programming than it otherwise could have and that the providers will pass on some portion of that price increase to their customers. The essential premise of the theory is that the post-merger Turner—now backed by AT&T—will be more likely to drive harder bargains with AT&T’s distribution rivals, forcing them to pay higher prices to avoid losing access to Turner programming. 2 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 12 of 76 The flaws in this theory are manifest. First, as the government itself now recognizes, withholding programming is not an outcome Turner can tolerate. To Turner, withholding programming means immediate, catastrophic losses in licensing and advertising revenue that can never be recovered. Because those losses will not be any more sustainable after the merger than before, any threat to withhold content will be just as hollow as it was before. And everyone will know it. For this reason and others, Time Warner executives will explain at trial that the merger will not—and realistically could not—affect their negotiations with video distributors, as confirmed by experience with multiple comparable transactions in this industry. Second, and even more fundamentally, this merger has never been about making Time Warner programs less accessible or more expensive. Just the opposite: it is about making Time Warner and AT&T more competitive during a revolutionary transformation that is occurring in the video programming marketplace. As will be demonstrated at trial, the new video revolution is defined by the spectacular rise of Netflix, Amazon, Google, and other vertically integrated, direct-to-consumer technology companies as market leaders in both video programming and video distribution. For these massive Internet platforms, vertical integration means a new model of delivering high-quality programming “over the top” of existing wired and wireless broadband networks directly to consumers’ televisions, phones, and tablets—a “flywheel effect” where video content drives deeper engagement on their platforms, creating more sales and advertising revenue from participating retailers and advertisers, subsidizing the cost of the service to consumers, putting downward pressure on consumer prices, and supporting additional investment in video content. The advantages of this model explain why Google now offers 50+ channels of premium content on YouTube TV for $35 a month, why Amazon and Netflix create award-winning 3 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 13 of 76 original programs exclusively for their customers, and why Apple and Facebook will each spend $1 billion on video content this year alone. 2 In the television industry, the disruptive impact of these new competitors could hardly be more dramatic: • Subscribers: As of the end of 2017, Netflix had approximately 118 million global subscribers and nearly 55 million U.S. subscribers, more than the top six U.S. cable companies combined. 3 Last month alone, Netflix subscribers streamed video to 450 million televisions, phones, tablets, and other devices. 4 Meanwhile, Amazon Prime’s video product now reaches an estimated 90 million U.S. subscribers, and Hulu boasts 54 million unique viewers. 5 • Content Spend: In 2017, Amazon, Netflix, and Hulu together spent more than $13 billion on content, roughly 21% of all program expenditures by all major broadcast, basic cable, and premium cable networks. 6 Netflix alone spent more than $6 billion in 2017. This year, it will spend $8 billion and have about 700 original TV series worldwide. 7 • Market Capitalization: As Esquire reported last month: “You could merge the world’s top five advertising agencies (WPP, Omnicom, Publicis, IPG, and 2 See Tripp Mickle, Apple Readies $1 Billion War Chest for Hollywood Programming, Wall St. J. (Aug. 16, 2017) (Apple “budget of roughly $1 billion to procure and produce original content over the next year”); Deepa Seetharaman, Facebook is Willing to Spend Big in Video Push, Wall St. J. (Sept. 8, 2017) (Facebook “willing to spend as much as $1 billion to cultivate original shows for its platform”). 3 See Netflix Letter to Shareholders (Jan. 22, 2018) (54.8M domestic memberships, of which 52.8M were paid, and 62.8M international memberships, of which 57.8M were paid). 4 See Joan E. Solsman, Netflix Streamed to Almost Half a Billion Devices Last Month, CNET (Mar. 6, 2018). 5 See Consumer Intelligence Research Partners, LLC Press Release, Amazon Prime Hits 90 Million US Members (Oct. 18, 2017) (estimating 90M Amazon Prime members in the U.S.); Hulu Press Release, Hulu Grows to Over 17 Million Subscribers in 2017, Emerges As Powerful Pay TV Alternative Combining Live Television with the Largest SVOD TV Library in the U.S. (Jan. 9, 2018). 6 See SNL Kagan, Profile: Netflix (U.S.) (Mar. 6, 2017); SNL Kagan, Profile: Hulu (Feb. 22, 2017); SNL Kagan, Profile: Amazon Prime Video (US) (Jan. 30, 2017); Rani Molla, Netflix Spends More on Content Than Anyone Else on the Internet – and Many TV Networks, Too, Recode (Feb. 26, 2018). 7 See Todd Spangler, Netflix Eyeing Total of About 700 Original Series in 2018, Variety (Feb. 27, 2018); NFLX – Q4 2017 Netflix Inc Earnings Video Interview, Thomson Reuters StreetEvents, Edited Transcript, at 7 (Jan. 22, 2018) (statement of Reed Hastings, Founder, Chairman of the Board, CEO & President, Netflix, Inc.). 4 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 14 of 76 Dentsu) with five major media companies (Disney, Time Warner, 21st Century Fox, CBS, and Viacom) and still need to add five major communications companies (AT&T, Verizon, Comcast, Charter, and Dish) to get only 90 percent of what Google and Facebook are worth together.” 8 In only a few years, this phenomenon of “over the top” premium video has irreversibly reshaped the landscape for the creation and delivery of television content, pushing all players in the market to respond in numerous ways that benefit consumers. This merger is one of those responses. * * * * * Faced with these real-world market facts, the government retreats to a purely theoretical realm. Ignoring all contrary evidence in the merger documents and party testimony, it relies on a hypothetical “bargaining model” to predict a price increase for the Turner networks to certain distributors. But this simplistic model is a moving target that cannot begin to replicate the complex, combative, and constant negotiations between television programmers and distributors. For example, in his initial report, the government’s expert claimed, with startling and implausible precision, that the merger will cause consumer pay-TV prices to rise by a monthly total of 27 cents per subscriber, or less than 0.2% of a consumer’s average monthly bill. Just a few weeks later, after fiddling with some input dials, the expert managed to almost double that insubstantial result to a still-insubstantial 45-cent monthly increase, all of 0.4% per bill, which is where the government currently stakes its case. The ease with which the government manipulated its model to get a different, but still strikingly small, alleged increase says almost everything the Court needs to know about its reliability. The trial evidence will say everything else. As defendants’ expert will show, the government’s modeled price increase is so negligible that, given the inherent uncertainty in the 8 Scott Galloway, Silicon Valley’s Tax-Avoiding, Job-Killing, Soul-Sucking Machine, Esquire (Feb. 8, 2018). 5 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 15 of 76 model’s predictive inputs, the claimed increase is statistically indistinguishable from zero. And the government’s expert’s data inputs—both his initial inputs and his “adjusted” inputs—are demonstrably incorrect in several critical respects. Once those errors are corrected, the government’s own model projects an industry-wide consumer price reduction, confirming that this vertical combination, like so many before it, benefits consumers. And this is before the Court considers any of the billions of dollars of synergies that flow from this merger. Just yesterday, these failures of proof were on full display during the deposition of the government’s lead economic expert, who revealed that (i) he never tested whether his bargaining model accurately predicts real-world bargaining outcomes in this industry and (ii) his model cannot predict specific price increases that will occur in actual negotiations. The government’s priceincrease theory—tenuous to start—thus ultimately collapses on itself. But even assuming the government could factually establish the modest price increase it predicts—which it cannot—such a showing still would not satisfy the government’s legal burden. The purpose of the antitrust laws is to protect competition, not competitors. Because vertical integration makes companies more efficient, it increases competitive pressure on the merged firm’s rivals and forces them to respond with competitive initiatives of their own, all to the benefit of consumers. Vertical integration raises antitrust concerns only in the rare case where the government can prove that the merger will hobble rivals’ ability to check the merged firm’s pricing conduct, thereby allowing the merged firm to raise its own prices above competitive levels. Here, the government does not even attempt to show that rival firms will be unable to compete in response to an alleged 45-cent price increase to consumers. And, by admitting that the merged firm will, in fact, lower its own consumer prices, the government effectively 6 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 16 of 76 concedes that this transaction strengthens, rather than harms, competition. Had the government reviewed its own historical pricing data on other combinations of content and distribution, it likely would have reached the same conclusion, but, as it represented to the Court when defendants brought the data to the Court’s attention, the government did not consider the data “relevant.” At trial, defendants will show that these real-world data are not only relevant, but entirely inconsistent with the government’s case. The government’s secondary theory of harm fares no better. When it filed its Complaint, the government alleged that the combined company would withhold its programming from some online providers. Compl. ¶ 40. Since then, however, the government has abandoned that claim and focused exclusively on the theory that, instead of acting unilaterally, the merged firm will coordinate with its chief rival Comcast to withhold programming from a small subset of online providers. But at his deposition, the government’s own expert essentially discarded that theory too when he admitted that he could not opine that post-merger coordination was more likely than not to occur. Indeed, he refused even to say whether the probability of coordination after the merger exceeded just one percent, which should be the end of the matter. * * * * * Contrary to the government’s superficial theories of harm, the evidence will show that this combination will make both Time Warner and AT&T stronger competitors and promote consumer welfare. For its entire history, Time Warner has been effectively trapped as a wholesaler of video content—it produces and aggregates valuable programming, but it traditionally has had to rely on other companies to distribute its programming to consumers. With that wholesaler position comes two competitive challenges that Time Warner must overcome to continue to compete effectively. 7 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 17 of 76 First, unlike Google and Facebook, Time Warner has no access to meaningful data about its customers and their needs, interests, and preferences. In most cases, Time Warner does not even know its viewers’ names. This data gap impedes its ability to compete with Google, Facebook, and other digital companies in advertising sales, which are critical to Turner’s viability, and which allow Turner to keep subscription fees much lower than they otherwise would be. Whereas digital companies have the data and the technology to deliver advertisements that are both specifically addressed (shown) to a particular viewer and tailored to that viewer’s specific needs and interests, Time Warner cannot target its television advertising in those ways, creating an increasing competitive disadvantage for the company. The data gap also gives online video programmers a competitive advantage in the production and aggregation of content based on extensive data about the content preferences of their viewers. Second, all of the new digital titans are able to offer their products and services directly to consumers. By doing so, they not only avoid the costs associated with a distribution affiliate, but they are also able to market directly to consumers and tailor their offers directly to the individual consumer. It is therefore increasingly important that Time Warner be able to go “direct to consumer” to compete head on with these very large new entrants with new and innovative offerings not controlled by distributors. But Time Warner has never had the access, assets, or expertise required to support a direct-to-consumer business. AT&T’s assets and capabilities fit hand in glove with Time Warner’s competitive needs. Through its DIRECTV set-top box, its broadband infrastructure, and its nationwide cellular network, AT&T has precisely the sort of first-party customer data that Time Warner needs to remain competitive with the digital companies both in advertising sales and in understanding consumer viewing preferences. In addition, AT&T will provide tens of millions of consumer 8 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 18 of 76 touchpoints and an extensive retail-service infrastructure, securing for Time Warner the assets and expertise it needs to compete effectively in an increasingly direct-to-consumer world. From AT&T’s perspective, on the other hand, the online video revolution has created conditions that constitute competitive opportunity as much as they do competitive challenge. Unlike traditional video distribution rivals like Comcast and DISH, AT&T operates a nationwide wireless network, which means AT&T benefits when people watch television on their mobile devices. As a result, AT&T is merging with Time Warner not to thwart online viewing, but to advance it, by enabling AT&T to introduce new video products better suited to mobile viewing. AT&T has already begun to develop such products on its own, including its popular DIRECTV NOW service, but it has been unable to obtain sufficient distribution rights from unaffiliated programmers to achieve its vision for the next wave of products and packages, including lowercost, ad-supported services. The merger will solve that problem and allow AT&T to integrate Time Warner programming with AT&T’s mobile video platform to connect with viewers where they increasingly are today: on their phones, tablets, and other mobile devices. And for the combined company, the merger will also create the combined scale necessary to develop the first industry-wide platform for bringing targeted advertising to premium video, offering advertisers a viable alternative to the digital advertising duopoly of Google and Facebook. Today, Google and Facebook have a stranglehold on digital advertising through their automated, data-driven advertising platforms. As one headline put it, “Facebook and Google completely dominate the digital ad industry,” with a combined 99% of growth. 9 By applying AT&T’s data to Time Warner’s advertising inventory, this merger will not only increase the 9 Alex Heath, Facebook and Google completely dominate the digital ad industry, Bus. Insider (Apr. 26, 2017). 9 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 19 of 76 effectiveness of Time Warner’s existing advertising business, but it will justify the investments necessary to build an industry-wide, fully-automated advertising platform directly opposite Google and Facebook. This again means more competition, not less. * * * * * All told, AT&T projects merger synergies in the billions—more than $2.5 billion in annual synergies by 2020 and more than $25 billion in total synergies on a net present value basis. These are verifiable, quantifiable, and pro-competitive synergies that will unquestionably benefit consumers, but the government rejects them out of hand in favor of its hypothetical predictions of harm. It may be enough for the government that an easily manipulated theoretical model can be made to estimate a pocket-change increase in monthly pay-TV bills, but it should not be enough for this Court. AT&T and Time Warner are not seeking to hold back the tide, as the government asserts, but to ride an irresistible wave. By integrating their almost perfectly complementary services in a new mobile-centric video marketplace, merging AT&T and Time Warner will lead to new products, better services, more innovation, ever-fiercer competition, and lower consumer prices. But there is only one way the merger will succeed, and both companies know it: the combined entity must increase the quality, quantity, and breadth of its programming distribution to a new generation of online, mobile viewers. Demanding new increases in Turner content prices or denying online viewers access to that content would be contrary to this imperative. To erase any doubt on that score, Turner sent all its distributors written, irrevocable offers to submit disputes over the price and other terms of Turner programming to binding arbitration and to guarantee continued access to that programming—i.e., Turner cannot withhold its networks (“go dark”) while any arbitration is pending. This irrevocable “arbitration/no- 10 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 20 of 76 blackout” mechanism is self-executing, can be invoked solely by the distributor (not Turner), requires no action from this Court or the government, and applies to both traditional and online distributors. By itself, this offer defeats both unilateral and coordinated theories of harm, and thus should conclusively resolve this case. This Court approved the Comcast/NBCU merger on the government’s representation that a virtually identical arbitration/no-blackout mechanism would prevent the combined entity from inflating prices or withholding content. The government was right. The trial evidence will show that the Comcast/NBCU merger resulted in no harm to competition whatsoever. Nothing is materially different here. In the end, the central point is this: AT&T and Time Warner do not compete against each other today, and this merger does not remove any competitor from the television marketplace. The number of upstream suppliers selling television content, which are many and growing, remains the same; the number of downstream distributors providing television content to consumers, which are many and growing, remains the same. But, by integrating Time Warner’s video programming with AT&T’s video distribution assets—particularly its wireless network— this merger will enable the combined entity to address competitive challenges and opportunities created by sweeping changes in the video marketplace, enhancing competition and benefiting consumers in ways that naturally flow from the vertical combination of complementary assets. Blocking the transaction would deny consumers these benefits and shield large, vertically integrated firms such as Comcast/NBCU, Netflix, Google, Amazon, and Facebook from new competition on their own turf. THE RATIONALE OF THE MERGER The video marketplace is undergoing a profound transformation that is, in the government’s words, “revolutioniz[ing] the way Americans receive and experience video 11 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 21 of 76 content.” 10 Although the government posits that Time Warner and AT&T seek to thwart this revolution, the evidence will show that the opposite is true. These companies are merging to meet transformative change head on. A. Overview of the Merging Parties and the Video Programming Industry. Time Warner and AT&T operate at distinct—but highly complementary—levels of the video programming industry. Time Warner is engaged both in creating content (through its Warner Bros. studio) and in aggregating content into networks (the Turner networks and HBO) that are then distributed to consumers through myriad third parties. AT&T delivers video content to consumers over its various networks, including its satellite-based DIRECTV service, its telephone-network-based U-verse service, and its Internet-based DIRECTV NOW service. 1. Time Warner: Video Programming Creation and Aggregation. Although it was once a much larger company, today Time Warner is purely a video programming creator and aggregator, organized into three business units: Warner Bros. operates a Hollywood studio that creates movies, television programs, and other video content that are licensed to third parties. Warner Bros. creates and licenses content both for Time Warner’s businesses and for other parties. Netflix, for example, is the single largest customer of Warner Bros. content. Turner operates 10 cable networks that televise scheduled video programming around the clock. 11 Turner does not distribute these networks to consumers directly, but instead relies on and contracts with distributors—such as cable operators—to do so. The distributors sell the 10 Compl. ¶ 1, United States v. Charter Commc’ns, Inc., No. 1:16-cv-00759-RCL (D.D.C. filed Apr. 25, 2016), ECF No. 1. 11 Those 10 networks are: TNT, TBS, CNN, CNN Español, CNN International, Cartoon Network/Adult Swim, TruTV, TCM, Boomerang, and HLN. 12 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 22 of 76 Turner channels to consumers in bundles with hundreds of other channels for a fixed monthly fee. There are more than 800 other cable networks competing today—up from about 308 in 2002—in addition to the big four national broadcast networks: ABC, CBS, NBC, and Fox. 12 The Turner channels, like other cable networks, typically earn revenues in two ways. First, they charge distributors a per-subscriber “affiliate fee” for the right to carry their content. It is thus in Turner’s interest to be distributed to as many video programming subscribers as possible. Second, Turner channels sell advertising time to third parties, retaining the right to sell 14 minutes of each programming hour to their own advertising partners. (They cede two additional minutes of each hour to distributors to sell at their discretion.) Because advertising rates are tied to viewership, it is again in Turner’s interest to maximize distribution of its programming. HBO is a premium subscription-based service that offers movies and television shows, including a significant amount of original content. Unlike the Turner networks, HBO is not included in traditional programming bundles and has no advertising. HBO is sold as an add-on service; only about one third of consumers who subscribe to traditional video programming services also subscribe to HBO. In addition to obtaining it from distributors, consumers may also obtain access to HBO programming directly, through the Internet, by subscribing to the HBO NOW online service. 12 See Fourteenth Report, Annual Assessment of the Status of Competition in the Market for the Delivery of Video Programming, 27 FCC Rcd 8610, ¶ 43 & n.96 (2012); Tenth Annual Report, Annual Assessment of the Status of Competition in the Market for the Delivery of Video Programming, 19 FCC Rcd 1606, 1691, Tbl. 8 (2004). 13 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 23 of 76 2. AT&T: Video Distribution. AT&T is a distribution company. It owns landline telephone, mobile and fixed broadband, and satellite TV distribution assets. It is not a creator of any significant television or movie content. 13 AT&T’s telephone and broadband operations are concentrated in portions of 21 States; its mobile broadband network is national in scope. As a video distributor, AT&T is one of many companies known in industry jargon as a Multi-Channel Video Programming Distributor, or “MVPD.” AT&T offers two distinct MVPD services. First, it operates DIRECTV, which is a satellite-based MVPD service. Satellite TV services distribute video service nationwide by transmitting programming from satellites located more than 20,000 miles above the earth to rooftop dishes installed at the customers’ homes. Second, AT&T operates U-verse, which is provided over AT&T’s local wireline telephone network. AT&T—and other telephone companies such as Verizon and CenturyLink—began upgrading their local telephone networks to fiber-optic technology in the mid-2000s, which has allowed them to offer cable-like video services. 14 Both DIRECTV and U-verse are competitive entrants in the video marketplace. Cable operators such as Comcast, Charter, Cox, and Altice are the incumbents in this marketplace and began with monopolies granted by state or local franchise authorities. Although there is typically only one incumbent cable operator in any given geographic region, cable operators collectively offer service to more than 99% of U.S. homes. 15 The local cable incumbent is also 13 AT&T operates four regional sports networks: AT&T SportsNet Southwest, AT&T SportsNet Pittsburgh, AT&T SportsNet Rocky Mountain, and ROOT Sports Northwest. 14 Telephone companies have deployed video services to approximately 39% of U.S. homes. See Eighteenth Report, Annual Assessment of the Status of Competition in the Market for the Delivery of Video Programming, 32 FCC Rcd 568, 575, Tbl. III.A.1 (2017). 15 See id. ¶ 20. 14 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 24 of 76 typically the leading MVPD within its service area, often by a wide margin. Virtually all consumers, however, can choose from among at least three traditional pay-TV distributors, and many can choose from among at least four. 16 AT&T also recently began offering an online video distribution service known as DIRECTV NOW. This enables consumers to obtain a service that is substantially similar to the satellite-based DIRECTV service, except it is delivered over the Internet rather than via a rooftop dish and offers “skinnier” channel packages at lower price points. Because DIRECTV NOW and similar services appear to consumers much like a traditional MVPD service, they have become known as virtual Multi-Channel Video Programming Distribution (or “virtual MVPD”) services. Competing virtual MVPDs (or just “virtuals”) include DISH’s Sling, Sony PlayStation Vue, YouTube TV, Hulu Live, FuboTV, and Philo. B. The Video Marketplace Is Transitioning from the Television Era to an Intensely Competitive Internet Video Era. For decades, people watched video programs on a television, at a pre-scheduled time, as part of a standard subscription package. Traditionally, the supply of video content was largely the province of the “big four” broadcast networks (ABC, CBS, NBC, and Fox) and cable networks owned by Disney, Fox, NBCUniversal (“NBCU”), Time Warner, Viacom, Discovery, A&E, and others. And, traditionally, consumers received their video programming from MVPDs. All that is rapidly changing, as the government concedes. The recent widespread deployment of broadband wireline and wireless networks allows consumers to obtain high-quality video programming from a range of online providers. Consumers no longer need a television connected to an antenna, cable wire, or satellite dish to 16 See id. at 577, Tbl. III.A.2. 15 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 25 of 76 receive video programming. Using a wide array of devices from smart phones to tablets to laptops to smart TVs, they can use broadband connections to access video programming over the Internet. Similarly, companies that seek to distribute video programming do not need a broadcast tower, cable network, or satellite to reach consumers. They can deliver video over the Internet, giving them instant (and inexpensive) access to customers across the nation and around the world. This transformation has enabled companies like Netflix, Amazon, and Hulu—which are known in industry jargon as Subscription Video on Demand (“SVOD”) providers 17—to begin delivering video over the Internet and gain massive scale in a short period of time. For a monthly fee, SVODs provide access to large libraries of video programming, which consumers can stream (or download) to their computers, mobile devices, and TV set-top boxes at any time. The virtual MVPDs described above also have been offering online alternatives structured more like traditional pay-TV, with linear programming from traditional network-based content providers like Turner. But it is mainly SVODs that have driven the revolution in online video, making enormous investments in programming and giving rise to a new golden age of video content. When Netflix, Amazon, and other SVODs first entered the market, they generally offered content acquired from production companies like Warner Bros., such as previously run television shows and movies. This gave consumers access to extensive libraries of content and allowed them to “binge watch” entire seasons of old television shows, rather than watching them at prescheduled times. As these new entrants gained scale and experience, however, they seized the 17 SVODs, together with virtual MVPDs, are also sometimes identified under the broader heading of online video distributors (“OVDs”) or “over the top” providers (“OTTs”). 16 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 26 of 76 opportunity to produce exclusive original content such as House of Cards and Transparent, thereby integrating their distribution business into the programming business. These vertically integrated content creators and distributors, including Netflix, Amazon, and Hulu, are no longer upstarts. They are industry giants, and they have quickly become some of the largest spenders on video programming content. In fact, in just a few years’ time, Netflix, Amazon, Google, and others have harnessed the power of vertical integration to disrupt and displace content suppliers and distributors alike. These vertically integrated firms have succeeded by using their direct customer relationships to better provide customers the programming that most interests them. Netflix, Amazon, and Hulu all use anonymized customer data—such as viewing preferences and habits and demographic characteristics—to craft content and services that better appeal to consumers. Google, Facebook, and other tech behemoths with access to customer data have also begun providing Internet-based video distribution services or have announced plans to do so. And many have announced plans to invest heavily in original content. The success of these SVOD and virtual MVPD platforms has caused an increasing shift away from traditional MVPD service. As of the end of 2017, more than 22 million subscribers have dropped traditional MVPD services—i.e., they “cut the cord.” And many other consumers have “shaved the cord,” relying increasingly on virtual MVPDs and SVODs for video entertainment while spending less on MVPD services—for example, by selecting a less expensive cable package with fewer channels. Just last week, under the headline “Cord Cutting Hits Another Record, Bashing Cable and Telecom Stocks,” Fortune reported: The cable bundle has become increasingly unappealing as consumers have turned in droves to more flexible and less expensive video offerings, from services like 17 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 27 of 76 Netflix and Hulu that feature traditional TV and movie formats, to shorter programming from YouTube, Facebook, and a host of others. 18 As a result of the extensive and growing competition, AT&T/DIRECTV’s position as a video distributor is declining. At the end of the third quarter of 2017, AT&T/DIRECTV had approximately 24 million subscribers to its MVPD services, down from more than 26 million in 2014. Over the course of 2017 alone, AT&T lost approximately 1.2 million DIRECTV and U-verse subscribers. The company accounts for about 26% of nationwide subscribers to traditional video distribution services. That share is substantially smaller than the nation’s cable companies’ 53% collective share. Each of those cable companies typically has the greatest number of subscribers within its local footprint. Thus, in the overwhelming majority of geographic areas, AT&T/DIRECTV runs second or third behind the incumbent cable company, and also faces growing competition from dozens of Internet-based distributors. Time Warner likewise faces greater competition than ever before. Since 2010, the total number of scripted television shows has more than doubled (from 216 to 487). Most of that growth is due to Internet-based video distributors, which now account for approximately 24% of original series, up from less than 2% in 2010. The Turner networks account for only about 6% of combined viewership on traditional and Internet-based video distribution services, down from 9% in 2010. Even when Internet-based viewership is excluded, the Turner networks account for only 8% of television viewership. And during primetime hours (8 p.m. to 11 p.m. ET), none of the Turner networks ranks in the top 10, only two rank in the top 20, and eight rank in the top 100. 19 18 Aaron Pressman, Cord Cutting Hits Another Record, Bashing Cable and Telecom Stocks, Fortune (Mar. 1, 2018). 19 By comparison, at least four other programmers—NBCU, Viacom, Discovery, and Fox—have at least eight networks in the top 100. 18 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 28 of 76 Turner’s sports programming, while valuable, is also limited. Turner televises (i) no NFL, NCAA football, NHL, NASCAR, or Olympic events; (ii) less than 8% of NBA games; (iii) about 1% of professional baseball games; and (iv) well below 3% of NCAA men’s basketball games. In 2017, Turner networks accounted for only 12% of the 500 highest-rated sports telecasts, with its highest-rated event ranked 161. Turner’s most popular sporting event is the NCAA men’s college basketball tournament (“March Madness”), which has lower ratings than many other sporting events and television series (including an average episode of Dancing with the Stars). Time Warner’s other major programming network, HBO, likewise faces robust competition. HBO currently has approximately 20 original scripted series showing new episodes. Only one (Game of Thrones) ranks among the top 200, and that series is ending after its next season. HBO competes not only with several other premium networks (including Showtime and the Movie Channel, owned by CBS, and Starz, owned by Lionsgate), but also with other cable networks (such as FX, A&E, and AMC), as well as online distributors such as Netflix and Amazon. Last year, Netflix’s $6 billion content budget and Amazon’s $4.5 billion content budget dwarfed HBO’s $2 billion content budget. C. The Merger Addresses Specific Competitive Challenges Each Entity Faces in the New Video Marketplace. This revolution has presented unprecedented challenges to traditional video distributors like AT&T and traditional video programmers like Time Warner. Consumers are migrating from AT&T’s traditional video distribution networks to the Internet. As they do so, consumers are shifting from watching Time Warner’s programming over those traditional networks to watching the programming of Internet-based competitors like Netflix and Amazon. Both AT&T and Time Warner recognize that, to meet consumer needs, they must quickly adapt to the new video era. 19 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 29 of 76 For Time Warner, the changing video environment is increasingly isolating the company as a stranded wholesaler of video content. Until recently, being a wholesaler of video content was a perfectly sensible business model for Time Warner’s HBO and Turner, which make their money by (i) sharing in the subscription fee that distributors charge consumers for their network bundles, and (ii) in the case of Turner, selling commercial time on its networks to advertisers. In today’s television landscape, operating as a wholesaler without direct consumer contacts creates two serious problems that are central to understanding the pro-competitive reasons for this merger. First, Time Warner’s lack of consumer contacts means a lack of meaningful information about its customers, i.e., their demographic characteristics, interests, and viewing preferences. Indeed, except in limited cases, neither Turner nor HBO knows even the names, e-mails, or IP (Internet protocol) addresses of its viewers. As a result, Turner networks are increasingly disadvantaged in competing against dominant digital advertisers such as Facebook and Google for advertising revenues, which are essential to Turner’s dual-revenue-stream business model and essential to preventing additional upward pricing pressure on consumer subscription prices. In traditional television advertising, each ad is delivered to every viewer of a particular show, including many viewers not interested in the advertised product. Digital advertisers, by contrast, can target customized ads to specific individuals based on their demonstrated interests. An advertisement for North Face parkas, for example, can be delivered directly to a Facebook user who “likes” a mountain resort website or shares a hiking vacation experience. Digital advertising is so effective that it surpassed television advertising in 2016, and the gap is growing. The lack of customer data also creates a market disadvantage against vertically integrated video content suppliers like Amazon, Netflix, and Comcast/NBCU. Because of their direct customer 20 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 30 of 76 relationships, these firms can use data about viewers’ preferences to optimize their decisions about what programming to develop, when to schedule it, and how to market it—all to the benefit of consumers. Second, because of its lack of consumer contacts, Time Warner has never had the incentive or means to develop the infrastructure and expertise needed to sell video content directly to consumers. It does not have customer call centers. It does not have consumer billing systems. It does not have retail outlets, consumer marketing experts, or technical experience with the mechanisms used to reach today’s consumers directly. Time Warner currently has no choice but to rely on unaffiliated third parties to distribute and market its video programming. To remain competitive, Time Warner must fill in the gaps in its assets and expertise. Time Warner was trying to do exactly that when it was approached by AT&T. But it was far from clear to Time Warner executives or its board members that unilateral changes could be implemented at the scale and velocity needed to keep pace with the rapid transformation of the video programming industry. And even if it were possible in theory, the costs would have been staggering, necessitating an increase in its content prices, just when consumers are increasingly resisting such increases. Merging with AT&T represented a sound practical solution that reduced the risks in filling these critical gaps and allowed immediate access to AT&T’s distribution assets and rich consumer data. Merging would enable Time Warner to connect directly with its viewers, improve advertising by linking it to consumer interests, track and respond to viewer preferences, and experiment with innovative programming. AT&T faces a different but equally fundamental set of challenges. AT&T provides traditional MVPD services through its DIRECTV satellite service and its U-verse fiber-based service. These services are extremely costly to operate and maintain, yet they increasingly are 21 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 31 of 76 losing customers to Internet-based distributors that have no underlying network facilities costs and that allow consumers to access video through a wide range of fixed and mobile devices. For AT&T, the future of video lies in its wireless network, and the future of its wireless network lies in video. As noted above, AT&T does not produce any meaningful video content of its own. As a result, its fortunes are not tied to the creation of content in competition with Time Warner. Rather, AT&T’s overriding economic objective is to encourage consumers to use its networks, no matter whose programs they watch, and AT&T is unique among other video distributors in that it operates a large-scale national wireless network. AT&T earns revenues from use of its wireless service both by consumers and—increasingly in the future—by advertisers. In this way, AT&T stands apart from incumbent cable companies (e.g., Comcast, Cox, and Charter), pure satellite operators (DISH), and all over-the-top video providers (e.g., Google, Amazon, Netflix, Hulu), none of which has the pro-consumer incentives associated with operating a leading nationwide wireless network. Core to AT&T’s business strategies is its conviction that as more television viewing takes place over wireless networks—and as those networks are upgraded to ultra-fast 5G technologies—content companies like Time Warner will enjoy new demand for their programming, and traditional distributors will begin to sell more of the customized, lower-cost, mobile-first video services that consumers will want. Merging with Time Warner will allow AT&T to participate in both sides of the equation, all to consumers’ benefit. Today, AT&T’s primary video product, DIRECTV, lags behind Comcast and other cable incumbents in most regions, even as it competes nationally with Netflix, Amazon, Google, Hulu, and other well-funded platforms. To meet these competitive challenges, AT&T’s strategies call for a variety of new mobile video offerings, building on the success of its first such product, 22 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 32 of 76 DIRECTV NOW. To date, however, AT&T has not been able to obtain the distribution rights it needs from programmers (AT&T’s suppliers) to deliver the video innovations that consumers crave. This merger will address that challenge by allowing AT&T and Time Warner to overcome bargaining friction that has stifled innovation in this industry and better serve consumers where they increasingly are today: on their phones, tablets, and other mobile devices. Consumers win. Not only is AT&T’s traditional MVPD business in decline, but, as a stand-alone distributor, its ability to adapt fully to the Internet video era is also constrained. AT&T cannot offer new programming packages and services unless it obtains the programming for those services, and AT&T’s experience has been that it is extremely difficult to obtain the rights to develop and provide new services through contracts with independent programmers. Furthermore, although AT&T has direct access to customer data, absent the much larger advertising inventory of a programmer, it does not have the scale necessary to realize the full value of that information. Without this scale, AT&T is not able to innovate and bring new competition to the advertising industry. D. The Merger Will Create Significant Pro-Competitive Synergies That Will Result in Significant Consumer Benefits. By directly addressing each company’s distinct but related competitive challenges in the new video marketplace, the merger will create significant synergies that will make the combined entity a more effective competitor and benefit consumers. In total, AT&T projects merger synergies of more than $2.5 billion in annual synergies by 2020 and more than $25 billion in total synergies on a net present value basis. First, the merger will enable the combined firm to innovate and compete more aggressively in advertising. The combined firm will use AT&T’s data to make the advertising in 23 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 33 of 76 Time Warner’s programming more customized to its viewers’ needs and interests. Targeted ads benefit consumers both because they are more relevant to them, and because they help boost advertising revenues and thereby reduce the pressure to increase subscription fees to cover steadily rising content costs. Time Warner—unlike major online competitors like Netflix and Amazon—relies heavily on advertising revenues, which make up approximately half of Turner’s revenues. At the same time, Turner’s other source of revenue—monthly per-subscriber license fees—is tied heavily to the large bundles of channels that MVPDs have traditionally sold but that customers are increasingly abandoning. It is therefore critical for Time Warner to improve its advertising to remain competitive with the vertically integrated Internet giants that have direct relationships with their customers and immediate access to massive customer data. The merger also will give AT&T the scale in advertising inventory necessary to further develop its platform for premium video advertising. This merger presents the industry’s only foreseeable opportunity to create a video advertising platform with scale adequate to give advertisers a viable alternative to the duopoly of Google and Facebook. Today, Google and Facebook exercise near-total control of digital advertising through their automated, programmatic advertising platforms for search engines and social media. This merger will inject a new competitor for the benefit of advertisers and consumers alike. Today, AT&T is an emerging player in placing data-driven commercials within premium video. Unfortunately, as a stand-alone distributor, AT&T’s DIRECTV has only two minutes of advertising time per hour to sell. Turner, on the other hand, owns substantial television ad inventory—typically 14 minutes of ad time for every hour of programming it produces on each of its 10 channels—as well as popular ad-supported digital sites like CNN.com and bleacherreport.com. This merger therefore combines AT&T’s expertise in targeted video 24 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 34 of 76 advertising with a much larger pool of ad inventory that AT&T can use to grow advertising revenue and justify the investments necessary to build a fully automated, programmatic platform for the pay-TV and digital ecosystems at large. On this platform, video distributors and programmers will be able to contribute their inventory, and advertisers will be able to launch state-of-the-art, targeted ad campaigns to wide audiences of television viewers. This merger synergy alone is projected to yield billions of dollars. For consumers, better ad-supported video products mean (i) fewer advertisements, (ii) more useful advertisements, and (iii) advertisers, not subscribers, bearing more of the cost of the service. Advertisers gain a compelling alternative to Google and Facebook, and consumers win again. Second, the merger will enable AT&T to transform the mobile video marketplace by combining Time Warner’s content assets with its wireless platform to develop new and more valuable services especially for mobile video devices. For example: • AT&T would launch a new service with Turner and a small number of popular cable networks, which would be made available for free to AT&T’s wireless customers on unlimited data plans and for a nominal price to anyone else. • AT&T could make other specialized bundles of video available to wireless customers. • AT&T would also be able to integrate video content with social media applications on a smart phone, allowing consumers to clip and share videos they are watching with their friends. AT&T has been stymied in developing innovations because of its inability to obtain the necessary rights from content owners on flexible and affordable terms and in a timely manner. The changes transforming the video marketplace have created enormous uncertainty among programmers about how to realize the value of their content—uncertainty that has impeded AT&T’s efforts to develop the products and services consumers want. The merger will resolve 25 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 35 of 76 that problem by allowing AT&T to use Turner content to drive these new innovations. Once AT&T demonstrates their viability, other content providers are likely to follow. Third, the merger will enable the combined firm to better market Time Warner content in at least two ways. The merged firm can use AT&T’s consumer touchpoints and wireless business to reach new audiences, making Time Warner a stronger competitor to programmers like Disney and Comcast/NBCU that enjoy consumer connections Time Warner lacks. For example, AT&T can use its thousands of nationwide retail locations, which sell wireless and traditional video distribution services, to launch extensive and innovative promotions for Time Warner films, video games, merchandise, and television programming. The combined company will also be able to bundle Time Warner content with AT&T’s wireless services, offering consumers new packages and greater value. Separately, the merger will enable Time Warner to use AT&T’s customer data to improve Time Warner’s ability to market content to the customers most likely to value it. Time Warner can use this “content intelligence” to schedule programming to maximize its viewership or to provide customized guides featuring programming that customers are most likely to watch. Time Warner cannot accomplish this on its own—it is extremely difficult for programmers to contract with distributors for the sharing of data, as the Federal Communications Commission (“FCC”) recognized in the Comcast/NBCU Merger Order. 20 Fourth, the combined company will be able to realize substantial cost savings by eliminating redundant services and through scale economies. A majority of these savings involves variable costs that will enable AT&T to provide consumers greater value. Other 20 See Memorandum Opinion and Order, Applications of Comcast Corp., General Elec. Co. & NBC Universal, Inc. for Consent to Assign Licenses and Transfer Control of Licenses, 26 FCC Rcd 4238, ¶ 231 (2011) (“Comcast/NBCU Merger Order”). 26 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 36 of 76 savings involve fixed costs, which will allow the merged company to invest in improved content, innovations, and other consumer benefits. As the documents and testimony of both companies’ most senior executives will show, Time Warner and AT&T decided to merge to embrace the unstoppable and transformative changes redefining the video marketplace. By combining their assets, Time Warner and AT&T have the best chance of confronting the enormous challenges their businesses face. And, by leading the charge in innovation and competition, AT&T and Time Warner will drive other firms to respond in kind, triggering a “virtuous circle” of investment and innovation. All of this will greatly benefit consumers. The law encourages that result; the Court should not condemn it. LEGAL STANDARDS A. The Government Must Prove That the Merger Will Cause “Probable” and “Substantial” Harm to Competition. Section 7 of the Clayton Act prohibits a merger only if its “probab[le]” effect, United States v. Baker Hughes, Inc., 908 F.2d 981, 984 (D.C. Cir. 1990) (emphasis omitted), is “substantially to lessen competition, or to tend to create a monopoly,” 15 U.S.C. § 18. Although Section 7 requires predictive judgments, it “involves probabilities, not . . . possibilities.” Baker Hughes, 908 F.2d at 984 (Thomas, J.). The government must therefore prove that “the challenged acquisition [is] likely substantially to lessen competition.” FTC v. Arch Coal, Inc., 329 F. Supp. 2d 109, 115 (D.D.C. 2004) (emphasis added); see United States v. Marine Bancorp., 418 U.S. 602, 623 n.22 (1974) (alleged future harm to competition must be “sufficiently probable and imminent” to warrant relief ); United States v. Oracle Corp., 331 F. Supp. 2d 1098, 1109 (N.D. Cal. 2004) (rejecting merger challenge because government failed to prove the “merger will likely lead to a substantial lessening of competition”) (emphasis added); 27 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 37 of 76 see also FTC v. Tenet Health Care Corp., 186 F.3d 1045, 1051 (8th Cir. 1999) (“Section 7 deals in probabilities not ephemeral possibilities.”). B. The Government Bears the Burden of Proof on All Issues and Does Not Benefit from Any Presumption of Harm. In any merger challenge, “the ultimate burden of persuasion . . . remains with the government at all times.” Baker Hughes, 908 F.2d at 983. When challenging a horizontal merger, however, a presumption of competitive harm arises if the merger substantially increases concentration in an already concentrated market. See id. at 982. The merging parties must then produce evidence rebutting this structural presumption of harm. See id. The government does not benefit from any such presumption when challenging a vertical merger. See Republic Tobacco Co. v. North Atl. Trading Co., 381 F.3d 717, 737 (7th Cir. 2004) (“As horizontal agreements are generally more suspect than vertical agreements, we must be cautious about importing relaxed standards of proof from horizontal agreement cases into vertical agreement cases. To do so might harm competition and frustrate the very goals that antitrust law seeks to achieve.”). Vertical mergers preserve the same number of competitors in each market and thus never trigger any structural presumption of harm. Modern antitrust law in fact recognizes that vertical mergers are affirmatively “procompetitive” because, as Judge Kavanaugh has explained, they “encourage product innovation, lower costs for businesses, and create efficiencies—and thus reduce prices and lead to better goods and services for consumers.” Comcast Cable Commc’ns, LLC v. FCC, 717 F.3d 982, 990 (D.C. Cir. 2013) (Kavanaugh, J., concurring); see National Fuel Gas Supply Corp. v. FERC, 468 F.3d 831, 840 (D.C. Cir. 2006) (Kavanaugh, J.) (“[V]ertical integration creates efficiencies for consumers.”); IIIB Phillip E. Areeda & Herbert Hovenkamp, Antitrust Law ¶ 756a, at 9 (3d ed. 2008) (vertical integration “is 28 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 38 of 76 either competitively neutral or affirmatively desirable because it promotes efficiency”). 21 For these reasons, in the government’s words, vertical mergers “should be allowed to proceed except in those few cases where convincing, fact-based evidence relating to the specific circumstances of the vertical merger indicates likely competitive harm.” Roundtable on Vertical Mergers – Note by the U.S. Delegation to the Organization for Economic Co-operation and Development, Competition Committee 2 (Feb. 15, 2007); see id. at 10. Because vertical mergers are presumed pro-competitive, defendants bear no burden of proof on any issue. The government instead bears the burden of proving all facts needed to establish consumer harm, which includes disproving facts showing that the merger will make consumers better off rather than worse off. As to such facts, defendants bear at most a burden of producing evidence. See, e.g., Suture Express, Inc. v. Owens & Minor Distrib., Inc., 851 F.3d 1029, 1038 (10th Cir.), cert. denied, 138 S. Ct. 146 (2017). C. The Government Must Define Relevant Markets, Including a Programming Market, and Prove Substantial Market Power in All Relevant Markets. The legality of a merger “almost always depends upon the market power of the parties involved.” FTC v. Cardinal Health, Inc., 12 F. Supp. 2d 34, 45 (D.D.C. 1998); see ProMedica Health Sys., Inc. v. FTC, 749 F.3d 559, 570 (6th Cir. 2014) (the “ ‘ultimate inquiry in merger analysis’ is . . . ‘whether the merger is likely to create or enhance market power or facilitate its exercise’ ”) (quoting Carl Shapiro, the government’s lead economic witness here); Ball Mem’l Hosp., Inc. v. Mutual Hosp. Ins., Inc., 784 F.2d 1325, 1335 (7th Cir. 1986) (“Firms without 21 See also Alberta Gas Chems. Ltd. v. E.I. Du Pont de Nemours & Co., 826 F.2d 1235, 1244 (3d Cir. 1987) (“respected scholars question the anticompetitive effects of vertical mergers in general” because such “ ‘mergers are [a] means of creating efficiency, not of injuring competition’ ”) (quoting Robert H. Bork, The Antitrust Paradox 226 (1978)); Douglas H. Ginsburg, Vertical Restraints: De Facto Legality Under the Rule of Reason, 60 Antitrust L.J. 67, 76 (1991). 29 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 39 of 76 [market] power bear no burden of justification.”). And because determining market power requires identifying the market in which the power is created, defining a relevant market is a “necessary predicate to deciding whether a merger contravenes the Clayton Act.” Marine Bancorp., 418 U.S. at 618; see United States v. E.I. du Pont de Nemours & Co., 353 U.S. 586, 593 (1957); United States v. Engelhard Corp., 126 F.3d 1302, 1305 (11th Cir. 1997). A vertical merger, in particular, “will not have an anticompetitive effect” where “substantial market power is absent at any one product or distribution level.” Auburn News Co. v. Providence Journal Co., 659 F.2d 273, 278 (1st Cir. 1981); see AD/SAT, Div. of Skylight, Inc. v. Associated Press, 181 F.3d 216, 228 (2d Cir. 1999) (per curiam) (“[A]ntitrust law is concerned only with ‘substantial’ market power.”). ARGUMENT The government cannot establish any legal or factual grounds for blocking this merger. Its theories of harm contradict modern antitrust doctrine and rest on premises at odds with market realities and the merging companies’ business strategies. Defendants have underscored that point by irrevocably offering distribution rivals a self-executing arbitration/no-blackout mechanism, which would preclude the combined firm both from inflating prices to distributors and from coordinating with Comcast to withhold content from virtual MVPDs. The balance of this brief proceeds in three main sections. Section I refutes the government’s primary theory of harm, viz., that the merger will increase Turner’s “bargaining leverage” and thus enable it to charge distributors more for Turner channels. Compl. ¶¶ 32-38. Section II then refutes the government’s claim that the merger will likely harm competition because the combined company will coordinate with Comcast to withhold programming from virtual MVPDs. Id. ¶ 41. Finally, Section III refutes the government’s claim that the merged company will use HBO to disadvantage other MVPDs. Id. ¶ 39. 30 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 40 of 76 I. THE GOVERNMENT’S THEORY OF TURNER PRICE INCREASES FOR RIVAL DISTRIBUTORS FAILS ON BOTH THE FACTS AND THE LAW. The government and defendants agree on at least one thing: the merger will immediately benefit consumers by reducing AT&T’s costs for Turner content, leading AT&T to reduce prices to consumers. The government has abandoned its claim (Compl. ¶ 38) that the merger will cause the combined firm to increase its own prices. It is now trying to block the merger despite its own conclusion that AT&T’s prices will go down. The recognition that AT&T will achieve reduced costs that will be passed on to consumers means that the government starts its case needing to climb a multi-hundred-million-dollar mountain before it can even begin to show that the merger will harm consumers. And the mountain becomes far steeper once the many synergies created by the merger are properly credited. The government also no longer claims (Compl. ¶ 35) that the combined company will harm competition by unilaterally withholding Turner programming from rival distributors. The government’s own expert concedes that such a withholding strategy would make no economic sense because the company would suffer enormous losses in the licensing and advertising revenues it would receive from Turner content. Instead, the government tries to find adverse consumer impact through a theoretical model, called a “Nash bargaining model.” According to the government, its model obviates the need for real-world price-increase evidence because the model itself can accurately predict specific post-merger bargaining outcomes between Turner and MVPDs. That claim is as implausible as it sounds. In fact, the evidence will show (Part I.A) that, as the government uses it here, the bargaining model ignores most of what matters in negotiations in this industry and yields predictions demonstrably at odds with real-world conduct and outcomes. The model also ignores the arbitration/no-blackout mechanism, which categorically negates the model’s critical 31 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 41 of 76 assumption that the merger will give Turner an enhanced incentive and ability to walk away from bargaining without a deal. In addition, the model relies on demonstrably flawed data inputs, resulting in predictably flawed outcomes. In fact, with correct inputs, the model actually projects a consumer price decrease. As a matter of law, even accepting the government’s view that rival distributors would pay marginally more for Turner content after the merger, the law still would not condemn the transaction (Part I.B). That is because the government cannot show that distributors would be disabled from constraining the merged firm’s prices, which will actually go down after the merger. Especially taking into account the myriad other consumer benefits of the merger (Part I.C), the government’s theoretical price increase for Turner provides no legal or factual basis for blocking this pro-competitive transaction. A. The Government Will Not Prove Even the Trivial Price Increase Its Bargaining Model Ostensibly Predicts. 1. The government’s model does not reflect the realities of bargaining in this industry. To predict that overall prices will go up—even though AT&T’s own prices will go down—the government asserts that Turner will charge higher prices to other distributors. But Turner already has every incentive today to charge as much as distributors will pay for its networks. That incentive will not change after the merger. And distributors, obviously, will want to pay as little as they can, just as they do today. They are not going to reach deeper into their pockets just because Turner is affiliated with their rival—indeed, they may well want to pay less. To argue otherwise, the government invokes a largely untested “Nash bargaining” model. As applied here, the theory goes like this: In negotiations with distributors, Turner has a minimum price it is willing to accept for its content. The distributor, meanwhile, has a 32 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 42 of 76 maximum price it is willing to pay. Each party’s minimum/maximum price depends on the best alternative it could obtain if a deal were not reached. The model assumes that bargaining parties will always agree to a price that exactly splits the difference equally between their minimum and maximum prices. Exactly. Every time. After the merger, the theory goes, that bargained-for price will increase, because the failure to reach a bargain will become less costly for Turner than before the merger: if Turner withholds its networks from the distributor, some number of the distributor’s subscribers will switch to other distributors that carry the Turner networks, and some number of those switchers in turn will sign up with AT&T’s DIRECTV service, partially (but only partially) offsetting Turner’s lost licensing and advertising revenues. The model then assumes that, because of the prospect of that offset, Turner will increase its minimum price. The model then re-applies its key assumption that the parties will split the difference equally—but this time, because Turner’s minimum price is somewhat higher, splitting the difference results in a somewhat higher price. That model rests on two assumptions—the 50/50 bargaining split and the increased leverage from the threat of withholding programming—with no proven connection to the facts of this case. See VirnetX, Inc. v. Cisco Sys., Inc., 767 F.3d 1308, 1333 (Fed. Cir. 2014) (rejecting use of Nash model because “the use of a 50/50” bargaining split as “starting point” was “unjustified by evidence about the particular facts” of case); Oracle Am., Inc. v. Google Inc., 798 F. Supp. 2d 1111, 1119 (N.D. Cal. 2011) (excluding Nash model where there was “no anchor . . . in the record of actual transactions” for the 50/50 bargaining-split assumption); see generally Ambrosini v. Labarraque, 101 F.3d 129, 134 (D.C. Cir. 1996) (expert opinion must be “sufficiently tied to the facts of the case”). 33 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 43 of 76 Most importantly, the bargaining model assumes that Turner will demand more for its networks post-merger because failing to reach a deal and removing its channels from a given distributor’s package will be slightly less costly to Turner than it was before the merger. But in the real world, what the government’s model deems “less costly” is simply “slightly less catastrophic, but still intolerable.” Withholding content imposes devastating, unrecoverable costs on programmers, and that will continue to be the case for Turner after the merger—which is why the government’s own expert openly concedes that after the merger it would still be irrational for Turner to withhold content from any given distributor. Were Turner to withhold programming, it would suffer massive losses in revenues from affiliate fees and advertising— losses outstripping the distributor’s lost subscription revenues. In addition, Turner would suffer even greater medium- and long-term adverse consequences, because its ability to attract top talent (including top directors and actors) and license valuable content (especially live sports programming) depends on its reputation for encouraging wide distribution of its programming. None of those severe adverse effects will be ameliorated by the merger—nobody contends that Turner would make up for its blackout losses through new customers lured to DIRECTV. None of this will be news to any pay-TV executive negotiating with Turner for a new carriage contract. Any negotiating executive will recognize—as the government now does—that Turner could not realistically withhold content after the merger any more easily than it can today. At most, Turner might be buried five feet down, rather than six—but either way it would be dead. Knowing exactly that, the pay-TV negotiator will find any threat to withhold Turner content just as empty as it was before the merger, eliminating any real-world prospect of a change in bargaining dynamics or outcomes. 34 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 44 of 76 At its very best, the government’s bargaining model only posits some purely theoretical amount of upward pressure on prices. It is far too abstracted from reality to reliably predict the specific outcome of any particular negotiation—much less to reliably quantify how much a given price would go up in any one negotiation. As anyone who has ever haggled over a car knows, many factors can influence the outcome of an individual negotiation. Those factors must be multiplied a thousand-fold to approach the complexity and contentiousness of negotiations over multi-year carriage agreements in the video programming industry. As the government’s expert admitted in his deposition, the model does not even begin to explain whether—or, more to the point, by exactly how much—individual negotiating executives would be influenced by the programmer’s affiliation with a rival distributor. Actual negotiating executives, however, will answer that question at trial—and they will say it has no effect at all. Other direct evidence of industry experience confirms that the bargaining model does not accurately predict the price effects of vertical integration in the video programming industry. Three specific transactions show that, contrary to the model, distributors do not pay more for content where the content provider is affiliated with a competing distributor: (1) the 2011 merger of Comcast and NBCU; (2) News Corp./Fox’s control of DIRECTV from 2004 through 2009; and (3) Time Warner Inc.’s 2009 divestiture of control over Time Warner Cable. The price effects of these transactions can be assessed by essentially comparing the programming prices charged before and after the transaction (controlling for relevant variables). For each transaction, the evidence will show that vertical integration of these companies was not associated with higher programming prices. Indeed, it appears the government does not argue otherwise. To this end, the government did not even consider the price effects of those transactions, even though it had the relevant pricing data in its files. These real-world 35 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 45 of 76 experiences refute the government’s theory-based claim that Turner’s affiliation with DIRECTV will scare rival distributors into paying higher prices for Turner content. 2. The arbitration/no-blackout mechanism and the FCC’s “Program Access Rules” defeat the government’s price-increase theory. In addition to ignoring the real-world outcomes of business negotiations in this industry, the government’s theoretical model ignores two real-world mechanisms that will preclude Turner from inducing distributors to pay higher prices for its content: (1) the arbitration/no-blackout mechanism; and (2) the FCC’s “Program Access Rules.” Modeled on a provision of the Comcast/NBCU consent decree the government itself endorsed, the arbitration/no-blackout mechanism allows a distributor (MVPD or virtual MVPD) to demand all Turner networks or any bundle of them licensed to AT&T or another distributor. See Arb. Procedures ¶¶ A.2, A.6, D.14. If the parties cannot agree on terms, the distributor can elect “baseball” arbitration, id. ¶¶ A.1, A.3, A.4, where each party submits its best offer and an arbitrator decides which one “most closely approximates the fair market value of the Turner Networks at issue,” id. ¶¶ A.4, A.5, C.7. If a distributor chooses arbitration, Turner cannot go dark—it must provide the networks to the distributor on existing terms, subject to a true-up based on the arbitrator’s award. Id. ¶¶ B.1, B.2. 22 This arbitration/no-blackout mechanism disposes entirely of the government’s claims in this case related to Turner (leaving only a minor issue related to HBO, discussed later). As explained above, the Nash model’s essential assumption is that Turner will raise its minimum 22 This offer requires no action by this Court. Under governing New York law, the offer is irrevocable, see Silverstein v. United Cerebral Palsy Ass’n of Westchester Cnty., Inc., 17 A.D.2d 160, 163 (N.Y. App. Div. 1962); Hermanowski v. Acton Corp., 580 F. Supp. 140, 143 (E.D.N.Y. 1983), aff ’d in part and rev’d and remanded in part, 729 F.2d 921 (2d Cir. 1984), and will bind the merged company if the merger is consummated, see AT&S Transp., LLC v. Odyssey Logistics & Tech. Corp., 22 A.D.3d 750, 752 (N.Y. App. Div. 2005). 36 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 46 of 76 price because failing to reach a deal and going dark will be less costly than before the merger. The arbitration/no-blackout mechanism definitively negates that key assumption: failing to reach a deal will not mean going dark, because Turner cannot go dark and an arbitrator will decide the disputed terms of the deal. The government, in other words, has modeled a bargain that will never actually occur while the mechanism is in place. As one of the government’s own experts explicitly conceded, the availability of the mechanism changes the relevant “threat point” in bargaining and must be taken into account. And the government’s lead economic expert agreed that an arbitration mechanism will affect negotiations, but then admitted that he did not take it into account in his model at all—leaving the model’s predictions inapplicable to the actual negotiations and deals that will take place after the merger. The government advances makeweight objections to the arbitration/no-blackout mechanism, but none responds to the critical point that the mechanism negates the model’s underlying assumption that the alternative to “no deal” is a Turner programming blackout. 23 The government’s principal criticism is that baseball arbitration does not perfectly replicate outcomes that would occur but for the merger. The mechanism need not be perfect, however, or even anywhere close to perfect, to prevent the bargaining harms the government posits. The evidence will show, for example, that, even if the mechanism were not fully effective in keeping prices of all distributors at the levels that would exist “but for” the merger, the harms predicted by the government’s model would disappear. 23 The government’s description of the arbitration/no-blackout mechanism as a “conduct” remedy is therefore inaccurate. By eliminating Turner’s blackout threat, the mechanism alters the merged firm’s incentives in negotiations and is thus akin to a structural remedy—it does not require the merged firm to act contrary to its interests to comply with a legal restriction. 37 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 47 of 76 The government also criticizes the arbitration/no-blackout mechanism on grounds that were equally applicable to the comparable mechanism in the Comcast/NBCU merger, which showed no price effect. 24 In that merger, both the Department of Justice (“DOJ”) and the FCC explicitly represented that a baseball-style arbitration/no-blackout provision would “prevent harms from integrating content and distribution market power.” Comcast/NBCU Merger Order ¶ 4; see Suppl. Stmt. of U.S. in Supp. of Entry of the Final J. at 7, United States v. Comcast Corp., No. 1:11-cv-00106-RJL (D.D.C. filed Aug. 5, 2011), ECF No. 26 (arbitration/no-blackout mechanism would “resolve the competitive problems raised by” merger); Tr. of Fairness Hr’g at 14-15, United States v. Comcast Corp., No. 1:11-cv-00106-RJL (D.D.C. July 27, 2011), ECF No. 38; U.S. Resp. to Pub. Comments at 12-13, United States v. Comcast Corp., Case 1:11-cv00106-RJL (D.D.C. filed June 6, 2011), ECF No. 23. In approving the Comcast/NBCU consent decree, this Court required the government to periodically report to it about the arbitration procedure. Over the ensuing seven years, the government has never told this Court that the arbitration mechanism was failing to protect competition. Nor could it credibly advance such a claim. Between 2011 and 2016, Comcast sent notices to more than 3,500 distributors in response to new requests for NBCU programming or to provide notice that an agreement was expiring. 25 Yet the government has reported only a single 24 The government claims, for example, that the merged firm will have an unfair advantage in arbitrations because it will know more about other programming agreements than MVPDs. Not so. Distributors have information about their own agreements that Turner does not have and, because of most-favored-nation provisions in their contracts (“MFNs”), they have significant information about terms in other programming agreements as well. See Pl’s. Objs. and Resps. to Defs.’ First Set of Interrogs. at 36 n.49 (Jan. 4, 2018). 25 See Comcast-NBCU Annual Compliance Reports to the FCC 2012-2016, http://corporate.comcast.com/news-information/nbcuniversal-transaction. 38 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 48 of 76 dispute that was arbitrated to a decision before the FCC. 26 Moreover, post-merger prices for NBCU programming did not rise compared to industry benchmarks. In fact, the evidence will show that, if anything, NBCU prices have risen more slowly than fees for other content over the same period. At the same time, competition has flourished as several online providers that were small or nonexistent in 2011 have become powerful competitors in providing video content and distribution, including Netflix, Amazon, Hulu, Google, and Facebook. It is not surprising, therefore, that the government, in answering defendants’ interrogatories, stated it takes no position regarding the efficacy of the Comcast/NBCU arbitration mechanism. See Pl’s. Objs. and Resps. to Defs.’ First Set of Interrogs. at 28. The government identifies no meaningful distinctions between the Comcast/NBCU remedy and the arbitration/no-blackout mechanism here. It says that, because Comcast provides only regional service, an arbitrator could consider NBCU contracts in non-Comcast markets as a “benchmark” to identify the fair-market value of content absent the merger—a benchmark that, it claims, is unavailable here because DIRECTV offers service nationwide. But an arbitrator here would still have highly relevant benchmarks, including Turner contracts before the merger and MVPD contracts with programmers similar to Turner. See Arb. Procedures ¶¶ C.4.a, C.4.b.ii-iii. The FCC has consistently recognized that both types of contracts can provide valid benchmarks 26 See First Annual Rep. ¶ 1, United States v. Comcast Corp., No. 1:11-cv-00106-RJL (D.D.C. filed Sept. 4, 2012), ECF No. 31. The government claims this may be because distributors were afraid to invoke arbitration for fear that NBCU would make a final offer that, if accepted by the arbitrator, would be too costly or could even destroy their distribution service. But that theory ignores that arbitration is also risky for a vertically integrated programmer like Comcast/NBCU or AT&T/Turner. If it presses for too high a price, it increases the risk that the arbitrator will choose the distributor’s lower price, which could trigger MFNs in other contracts and force Turner to lower prices to other distributors as well—a disastrous result for Turner. The arbitration mechanism thus gives both sides the incentive to be reasonable and to reach an agreement on terms that approximate fair-market value without incurring the risks and expense of arbitration. 39 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 49 of 76 in the case of a programming arbitration. See Memorandum Opinion and Order, General Motors Corp. and Hughes Electronics Corp., Transferors, and The News Corp. Ltd., Transferee, 19 FCC Rcd 473, ¶¶ 147, 174 (2004); Comcast/NBCU Merger Order, 26 FCC Rcd at 4366-67, App. A, § VII.B.5-6, .8; Memorandum Opinion and Order, News Corp. and The DIRECTV Group, Inc., Transferors, and Liberty Media Corp. Ltd., Transferee, 23 FCC Rcd 3265, 3343, App. B, § IV.B.4 (2008). The government also observes that government agencies oversaw arbitration and the merged entity under the Comcast decree. But the government cannot show that privately enforceable arbitration that relies on arbitrators who are experts in this field and provides for judicial review (see Arb. Procedures ¶ C.2) is inadequate. The government certainly does not identify defects that would be addressed by adding layers of costly and time-consuming agency review. In all events, there is, in fact, effective FCC oversight here. Federal “Program Access Rules” mandated by Congress and promulgated by the FCC prohibit a vertically integrated company from using programming as a weapon to harm rival MVPDs. These rules will require the merged firm to provide its affiliated programming to all MVPDs on nondiscriminatory terms, 27 and will prohibit it from improperly influencing affiliated programmers in their dealings with unaffiliated MVPDs. 28 Every post-merger negotiation with an MVPD (traditional or 27 See Memorandum Opinion and Order, Applications of Charter Communications, Inc., Time Warner Cable Inc., and Advance/Newhouse Partnership, for Consent to Assign or Transfer Control of Licenses and Authorizations, 31 FCC Rcd 6327, ¶ 175 (2016); Memorandum Opinion and Order, Applications of AT&T Inc. and DIRECTV for Consent to Assign or Transfer Control of Licenses and Authorizations, 30 FCC Rcd 9131, ¶¶ 168-170 (2015). 28 See 47 U.S.C. § 548(c)(2)(A-B); 47 C.F.R. § 76.1002(a-b); see also 47 U.S.C. § 548(j) (extending rules to common carriers and their affiliates); 47 C.F.R. § 76.1004 (same). Monetary relief is available for violations of these rules. See 47 C.F.R. § 76.1003(h). 40 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 50 of 76 virtual) thus will be conducted in the shadow of a potential complaint that Turner’s price demand violated these Rules. Parties bargaining in the real world obviously would account for that outcome, as the government’s own expert admitted. Although the government complains about the efficacy of its own Rules, it cannot deny the relevant point: the potential for an FCCdetermined outcome will be present in every MVPD negotiation. By ignoring that potential, the government has again failed to model bargains as they actually occur in the real world. In any event, the Rules reflect a regulatory judgment by Congress and the FCC about the extent to which the government should constrain private bargains. Merger review under Clayton Act § 7 should respect policymakers’ regulatory judgments, not supplant them. 3. Correcting the data inputs both eliminates any predicted price increase and shows the unreliability of the model. In addition to contravening real-world experience and ignoring the arbitration/noblackout mechanism and FCC “Program Access Rules,” the government’s bargaining model relies on badly flawed data inputs. To start with one especially glaring example, the model relies on data that do not account for the timing of the actual contracts between Turner and its distributors. Comcast, for instance, just signed a comprehensive long-term contract with Turner, meaning there will be no bargaining with the nation’s largest cable company for at least four years (at which point prices may be renegotiated in some circumstances). Yet the government includes a price increase for Turner programming for Comcast in its model. Excluding Comcast from receiving a Turner affiliate fee increase—even leaving all other data inputs untouched—reduces the predicted consumer price increase by more than half, to just 20 cents per month. Yet other data inputs cannot remain untouched because they, too, are flawed. As discussed above, the model assumes that Turner will increase its minimum price in bargaining 41 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 51 of 76 because, after the merger, the losses it would suffer from failing to reach agreement would be partially offset by revenues DIRECTV would receive from switching subscribers. To identify any price effect, the model requires the government to quantify at least three variables: the number of subscribers the rival distributor would lose (“departure rate”); the proportion of those subscribers who would sign up with DIRECTV (“diversion rate”); and the added profit margin for each new DIRECTV subscriber (“margin”). The government grossly exaggerates all three inputs. Departure rate: The government assumes a departure rate of 12%, but that figure is inconsistent with all of the most relevant empirical evidence. For example, both sides’ economists considered what would happen if a distributor dropped Turner by looking to what did happen when MVPD Suddenlink dropped all of Viacom’s 24 channels in 2014. The government states that Suddenlink lost 12% of its subscribers. But that conclusion is incorrect. The evidence will show that Suddenlink actually lost fewer than 5% of its subscribers over the course of nearly three years. Other MVPDs analyzed the Suddenlink/Viacom event in the ordinary course of business, and they all reached the same conclusion of a drop rate lower than 5%. Suddenlink itself stated that it lost only 2-2.5% from this incident. 29 The government here reached a dramatically higher rate only because it inexplicably failed to control for overall market trends in the decline in Suddenlink’s subscriber numbers (in contrast to analyses undertaken by other MVPDs, which control for those trends in finding a much smaller impact). Correcting for the government’s inflated departure rate—and making no other change—eliminates all predicted consumer harm. 29 See Shalini Ramachandran & Keach Hagey, Dropping Viacom Costs Suddenlink Few Subscribers, Wall St. J. (Feb. 24, 2015). 42 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 52 of 76 Other key empirical evidence ignored by the government further confirms that its 12% figure is way off. In late 2013, Cable One dropped the Turner networks for 23 days, and during that time the departure rate proved negligible—less than 1%, a fraction of the departure rate the government relies upon for its model. The Cable One experience is the one time a distributor dropped all of the Turner networks, 30 and yet the government excluded the event from its analysis. It turns out the omission has a remarkable explanation—the government’s expert testified that he did not know about this widely reported event, and thus failed to consider it. The government’s departure analysis also relied on a departure-rate study of a number of different networks performed by a third-party consultant for one of AT&T’s rival MVPDs, which provided the study to the government. But before giving it to the government, and after consulting with the MVPD that commissioned it, the consultant changed the number for Turner—and only Turner—to produce a departure rate for Turner programming higher than the consultant had initially reported. The government’s economist testified that, when he chose to rely on that study for his analysis, he was entirely unaware of the last-minute, Turner-only switch in the data—which makes reliance on the study all the more disqualifying. Diversion rate: The government calculates a diversion rate by relying on current MVPD market shares, essentially assuming that switching subscribers will distribute themselves proportionally in accordance with today’s market shares. But those market shares reflect in most instances the choice of providers that subscribers made years ago. To calculate a diversion rate, what matters is where switching subscribers are likely to go today, which requires consideration of the undeniable trend away from traditional MVPDs and toward new online alternatives (both 30 DISH dropped a subset of the Turner networks (all but TNT and TBS) in 2014, and the departure rate was inconsequential—less than 0.1%. 43 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 53 of 76 virtual MVPDs like Sling and YouTube TV, and non-Turner options like Netflix). The government’s diversion-rate calculation badly underestimates this cord-cutting trend—even though it is obviously easier to switch to a service requiring only an Internet connection than to a service that requires a customer to wait for someone to install rooftop hardware. The evidence will show that, when cord-cutting is properly taken into account, the government’s model shows marketwide consumer prices going down, not up. Margins: The government’s profit-margin figure is based on AT&T’s estimate of DIRECTV margins from 2016. But DIRECTV’s “churn”—the proportion of subscribers it loses each month—has been going up, its programming costs have been increasing, and its revenues have not kept pace. As a result, DIRECTV’s more recent margins are sharply lower than they were in early 2016. Making that one correction to the model largely wipes out the government’s prediction of a consumer price increase. In addition to identifying the specific flaws in the government’s inputs to its model, the foregoing analysis exposes the fundamental unreliability of the model itself, at least in the context of the minuscule price effects claimed here. Nobody can seriously contest that the model’s inputs are estimates, and such estimates are inherently subject to significant uncertainty. The government’s expert himself illustrated this point when, in his rebuttal report, he justified his new 45-cent number by “updating” certain estimates he had used in his initial report, where he had projected only a 27-cent increase. But if current departure and diversion rates are that difficult to estimate accurately, the difficulties are multiplied exponentially for estimating future departure and diversion rates. Departure and diversion rates will almost certainly fall in the future as online options become more ubiquitous and content options continue to expand, but nobody can predict the speed and magnitude of those changes. Yet the government offers a 44 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 54 of 76 projection of consumer prices calculated to the penny. That projection on its face claims a precision for the model that it cannot possibly sustain. Given the high risk of error inherent in the model, it makes little sense to rely on the artificially precise outcomes it generates to justify blocking a vertical merger with the potential for transformational consumer benefits, particularly in a highly volatile market that may change beyond recognition in a matter of years. 4. The government’s model does not adequately account for effects of the merger that will drive prices down. The variables discussed above are not, of course, the only ones affected by the merger that could influence bargaining. There are other effects of the merger that will put downward pressure on prices under the government’s model. For one, because AT&T will no longer bear a real economic cost for Turner programming, the merged firm will have a stronger incentive than AT&T would pre-merger to reduce prices to attract new subscribers. The government’s expert acknowledges this effect, but understates it by failing to account for the industry-wide cordcutting trend. Moreover, the benefits of this cost reduction are felt immediately—unlike any supposed increase in Turner prices, which would not be experienced until current contracts are renegotiated years after the merger closes. Furthermore, in positing that Turner will demand higher prices for its content because of its supposedly greater willingness to go dark, the government has ignored entirely the combined firm’s ability to realize substantial gains on advertising revenues. See supra pp. 23-25. Because Turner will make more money from advertising, rival MVPDs obtain a bargaining advantage— they know that Turner is even more eager to have its programming distributed as widely as possible. Under the government’s own theory, that effect will put downward pressure on Turner’s rates. The government’s failure to account for the demonstrable advertising synergy 45 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 55 of 76 that will result from the merger further illustrates the malleability of the model and its unreliability as a meaningful predictor of specific price effects. B. The Government’s Predictions of Minor Price Increases Are Insufficient as a Matter of Law To Constitute “Substantially Lessened Competition.” Even taken at face value, the government’s prediction that an average pay-TV bill will increase by 0.4% does not establish “substantial” harm to competition for three independent reasons. First, the government’s projected price increases are so small and so sensitive to minor changes in assumptions that they are competitively immaterial in this rapidly changing environment. Second, in any event, the government admits that it cannot show, as it must, that the merged firm will raise its own retail prices in response to price increases by its supposedly disadvantaged rivals. And third, the government cannot show that Turner has market power in a well-defined programming market, as it also must. 1. The projected price increases are tiny, highly sensitive to minor input variations, and competitively immaterial. The Clayton Act condemns mergers only if they are likely to lessen competition “substantially.” 15 U.S.C. § 18. The government’s case here essentially reads that word out of the Act. Its lead economist concedes that his projected 45-cent increase in retail MVPD prices is “a very small percentage” of monthly cable bills—specifically, “0.4%.” The problem for the government is not only that this projected increase is “very small”—but also that it is so sensitive to minor variations in assumptions about this rapidly changing marketplace, as discussed above, see supra pp. 43-45, that it cannot plausibly be considered “substantial.” In fact, as defendants’ expert will explain, the government’s modeled price increase is so negligible that, given the inherent uncertainty in that predictive exercise, it is not meaningfully distinguishable from zero. 46 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 56 of 76 2. The government will not show that the merger will stunt MVPD rivals and thereby enable DIRECTV to raise its own retail prices. The government’s “price increase” claim also fails as a matter of law because the government concedes that the merger will not lead DIRECTV to raise its own retail prices. Yet showing that the defendant will raise its prices above competitive levels is an essential element of proof that a vertical deal “substantially . . . lessen[s] competition.” 15 U.S.C. § 18. “The antitrust laws . . . were enacted for the protection of competition not competitors.” Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 488 (1977). That distinction is crucial in vertical cases because vertical integration normally disadvantages rivals by making the merged firm a more efficient and thus stronger competitor. In response, its competitors will have to compete harder by cutting their own prices or offering more innovative products. This pro-competitive dynamic is the natural and expected benefit of vertical integration, whether through contract or merger. In general, “vertical integration and vertical contracts are procompetitive” because they “encourage product innovation, lower costs for businesses, and create efficiencies—and thus reduce prices and lead to better goods and services for consumers.” Comcast Cable Commc’ns, LLC v. FCC, 717 F.3d 982, 990 (D.C. Cir. 2013) (Kavanaugh, J., concurring). 31 Modern antitrust law has thus created a bright-line rule to create business certainty and preserve incentives for vertical integration. Under that rule, vertical integration raises antitrust 31 See National Fuel Gas Supply Corp. v. FERC, 468 F.3d 831, 840 (D.C. Cir. 2006) (“[V]ertical integration creates efficiencies for consumers.”); Alberta Gas Chems. Ltd. v. E.I. Du Pont de Nemours & Co., 826 F.2d 1235, 1244 (3d Cir. 1987) (“respected scholars question the anticompetitive effects of vertical mergers in general” because such “ ‘mergers are [a] means of creating efficiency, not of injuring competition’ ”) (quoting Bork, Antitrust Paradox 226); Douglas H. Ginsburg, Vertical Restraints: De Facto Legality Under the Rule of Reason, 60 Antitrust L.J. 67, 76 (1991). 47 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 57 of 76 concerns only where it leaves rivals “stunted” as competitors, prevents them from “provid[ing] meaningful price competition,” and thereby enables the defendants to raise their own prices above competitive levels. McWane, Inc. v. FTC, 783 F.3d 814, 838-39 (11th Cir. 2015), cert. denied, 136 S. Ct. 1452 (2016); see United States v. Microsoft Corp., 253 F.3d 34, 71 (D.C. Cir. 2001) (en banc) (per curiam) (issue is whether exclusive dealing keeps competitors “below the critical level necessary . . . to pose a real threat” to defendant’s market power); see also Jefferson Parish Hosp. Dist. No. 2 v. Hyde, 466 U.S. 2, 45 (1984) (O’Connor, J., concurring in the judgment). To date, courts have applied that bright-line principle in the context of vertical contracts rather than vertical mergers, but only because the government has not taken a vertical merger to trial in four decades. The same principle logically applies whether vertical integration occurs by merger or contract. See Fruehauf Corp. v. FTC, 603 F.2d 345, 352 n.9 (2d Cir. 1979) (vertical mergers and vertical exclusive-dealing arrangements present “similar issue[s]” under the Clayton Act). 32 That principle is fatal to the government’s case here. The government does not contend that the modest price increases it alleges will “stunt” any MVPD or deprive it of competitive responses. And it affirmatively concedes that the post-merger AT&T will reduce its retail prices 32 For example, in an “exclusive dealing” contract, a supplier agrees to sell all of its goods through one distributor (or vice versa), thereby aligning the parties’ interests in ways similar to a vertical merger. Because this contractual form of vertical integration typically creates efficiencies, it is favored under modern antitrust doctrine, except in the rare case where it allows one of the contracting firms to raise its own prices above competitive levels. See, e.g., McWane, 783 F.3d at 838-39. Notably, many exclusive-dealing cases arise under Section 3 of the Clayton Act, 15 U.S.C. § 14, whose text is identical to Section 7 in all relevant respects (“may . . . substantially lessen competition”) and is thus read in pari materia with that provision. See Fruehauf, 603 F.2d at 352 n.9; see also United States v. Philadelphia Nat’l Bank, 374 U.S. 321, 365-66 (1963) (“[T]he tests of illegality under amended § 7 are intended to be similar to those which the courts have applied in interpreting the same language as used in other sections of the Clayton Act.”). 48 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 58 of 76 below what it would charge in the absence of the merger. Because the government cannot show that the merger will disable rivals from constraining the merged entity’s ability to raise its prices, it cannot prove a substantial lessening of competition under modern antitrust doctrine. Tellingly, the government did allege these elements of a vertical merger challenge in Comcast/NBCU. The government there alleged that, without an arbitration remedy, Comcast/NBCU would “hobble” distribution rivals by “denying or raising the price” of programming and that such “higher licensing fees will reduce pricing pressure on Comcast’s MVPD business and increase its ability to raise prices to its subscribers.” U.S. Dep’t of Justice, Competitive Impact Statement at 20, 24, United States v. Comcast Corp., No. 1:11-cv-00106-RJL (D.D.C. filed Jan. 18, 2011), ECF No. 4. The government has abandoned such a claim here because it cannot make any such claim here. Under older decisions, a vertical merger could be invalidated merely because it caused the merging parties to buy or sell exclusively to or from each other if such exclusive dealing “foreclosed” their rivals from “a substantial share of [the] market.” United States v. E.I. du Pont de Nemours & Co., 353 U.S. 586, 595 (1957) (alterations omitted); see, e.g., Brown Shoe Co. v. United States, 370 U.S. 294 (1962); Fruehauf, 603 F.2d at 352 & n.9. As then-Judge Thomas observed, however, “ ‘[t]he most important developments that cast doubt on the continued vitality of such cases as Brown Shoe . . . are found in other cases, where the Supreme Court . . . has said repeatedly that the economic concept of competition, rather than any desire to preserve rivals as such, is the lodestar that shall guide the contemporary application of the antitrust laws, not excluding the Clayton Act.’ ” United States v. Baker Hughes, Inc., 908 F.2d 981, 990 n.12 (D.C. Cir. 1990) (Thomas, J.) (quoting Hospital Corp. of Am. v. FTC, 807 F.2d 1381, 1386 (7th 49 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 59 of 76 Cir. 1986)). 33 Courts have now made clear that vertical integration raises antitrust concerns only where foreclosure is so extensive that it renders rivals ineffective as competitors. In any event, because the government concedes that the merged entity will not withhold Turner programming from rival MVPDs, let alone “foreclose” them from “a substantial share of [the programming] market,” du Pont, 353 U.S. at 595 (alterations omitted), the government’s case fails even under traditional “foreclosure” analysis. 3. The government also will not meet its burden of defining a relevant programming market and proving that Turner has substantial market power. Because the legality of a merger “almost always depends upon the market power of the parties involved,” FTC v. Cardinal Health, Inc., 12 F. Supp. 2d 34, 45 (D.D.C. 1998), market definition is “a necessary predicate to deciding whether a merger contravenes the Clayton Act,” United States v. Marine Bancorp., 418 U.S. 602, 618 (1974). Here, the government must prove not only that DIRECTV has market power in relevant distribution markets, but also that Time Warner has substantial market power in a relevant programming market because, “[w]here substantial market power is absent at any one product or distribution level, vertical integration will not have an anticompetitive effect.” Auburn News Co. v. Providence Journal Co., 659 F.2d 273, 278 (1st Cir. 1981). Put simply, if Turner lacks substantial market power in a relevant market for the sale of programming to distributors, the merged firm will be unable to use Turner as an effective weapon to harm competition from those distributors. 33 See also Comcast, 717 F.3d at 990 (Kavanaugh, J., concurring) (“[b]eginning in the 1970s,” the Supreme Court shifted to a favorable view of “vertical integration and vertical contracts”); VII Areeda, Antitrust Law ¶ 1511e2, at 517-18 (4th ed. 2017) (explaining that Ford Motor Co. v. United States, 405 U.S. 562 (1972), and Brown Shoe addressed vertical mergers “under an aggressive standard that struck down many mergers that would never be challenged today” because “our theory and most of our law of vertical integration have changed very considerably since that time”). 50 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 60 of 76 The government has not even attempted to define a relevant programming market in which it claims that Time Warner has such power. 34 The government’s failure to define such a market is no surprise because Turner channels account for a modest and shrinking portion of the rapidly expanding market for television programming. Turner’s networks together account for only about 6% of television viewership, down from 9% in 2010, and for only 3.4% of the top 500 television series, down from 6.8% in 2014. See supra p. 18. And despite the government’s focus on Turner’s rights to various sports events, Turner’s share of sports programming is small as well. See id. at 19. Turner accounts for only 12% of the top-500 highest-rated sports telecasts, and its highest-rated sports event ranked 161 out of those 500. See id. Overall, Turner has rights to only 1% of Major League Baseball games, less than 8% of NBA games, well less than 3% of NCAA men’s basketball games—and no rights at all to football (NFL or college) or hockey games. See id. In short, Turner programming accounts for far less than 30% of the programming market, the minimum threshold needed for finding “the kind of market power” sufficient to trigger antitrust concerns. Jefferson Parish, 466 U.S. at 27. Moreover, any claim that the combined company would exercise sufficient market power to use Turner as a competitive weapon crashes headlong into the new realities of the video marketplace, as the government essentially confirms by asserting such a trivial consumer price increase. The last few years have seen explosive growth in video programming created by an 34 The government argues that “a structural analysis based on market definition and market share is but one way to establish market power.” Pl’s. Objs. and Resps. to Defs.’ First Set of Interrogs. at 15. But in doing so it relies on cases involving a “finding of actual, sustained adverse effects on competition.” FTC v. Indiana Fed’n of Dentists, 476 U.S. 447, 461 (1986). Here, the government merely (and implausibly) predicts such effects. Because it can show no actual effects, the government must allege and prove a properly defined market in which such effects are likely to result from power in that market, as the alternative “direct effects” proof is generally not “used outside the context of horizontal agreements.” Republic Tobacco Co. v. North Atl. Trading Co., 381 F.3d 717, 737 (7th Cir. 2004). 51 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 61 of 76 expanding array of content sources. See supra pp. 16-18. In this environment, it is implausible to assert that Turner programming, while valuable, is genuinely essential for any given distributor. As DISH CEO Charlie Ergen explained in September 2017: In today’s market [a pay-TV provider] could take ESPN down and survive it. Five years ago that would not have been an option; that would have been suicidal. But today they could do it. [It would] be painful but they could do it. There’s not any channel I go to sleep at night that I say if the price is too high for our consumers, based on our analysis, that we wouldn’t be willing to take down. 35 Indeed, when DISH failed to reach a renewal agreement to carry most Turner channels in 2014, Ergen explained that Turner is “one of the easy ones to take down” and that channels “like CNN are not quite the product they used to be.” 36 Real-world data confirm Ergen’s observations. During the rare episodes in which distributors have failed to reach agreement with programmers similar to Turner, the distributors not only survived the ensuing “blackouts” but emerged as strong competitors after taking measures to adjust to the programming loss. C. The Negligible Predicted Price Increases Are Swamped by the Merger’s Consumer Benefits. The government also cannot carry its burden to prove that there will be any net harm to consumers once the natural and expected benefits of vertical integration are considered. In fact, the evidence will show large and varied benefits resulting from this transaction. As to all of those benefits (except for the elimination of “double marginalization,” discussed below), the government erroneously assumes that consumers will receive no benefits whatsoever. That conclusion conflicts with the established economic understanding of the benefits of vertical transactions and with the facts of this case. 35 Interview by Jon Steinberg, Cheddar TV, with Charlie Ergen, CEO, DISH Network, at 6:48 (Sept. 27, 2017). 36 David Lieberman, Dish Network’s Charlie Ergen Calls Loss of CNN and Turner Nets a “Non-Event,” Deadline.com (Nov. 4, 2014). 52 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 62 of 76 As summarized above, the basic point of the merger is to synthesize AT&T’s distribution networks, relationships with customers, and technology, on the one hand, with Time Warner’s high-quality content and advertising inventory, on the other. The combination of these highly complementary assets would necessarily facilitate new consumer products and services that the companies, as a practical matter, cannot provide as separate entities. As detailed above in the Rationale of the Merger, the merger will give rise to the following benefits and more: • The combined firm can innovate and compete more aggressively in advertising, both by using AT&T’s data to make the advertising in Time Warner’s programming more targeted to its viewers and by leveraging AT&T’s advertising technology with Time Warner’s programming to offer a new advertising platform that competes with Facebook and Google for both television and digital advertising. • Time Warner can use AT&T’s customer data to improve Time Warner’s ability to deliver content to customers that best meets their needs. • AT&T can combine its strong mobile network and expertise with Time Warner’s content assets to create new mobile video products and services. • The combined company can extend Time Warner content to new audiences through marketing campaigns drawing on AT&T’s extensive consumer touchpoints and wireless business. • The combined company can realize cost savings of approximately $1.5 billion annually by eliminating redundant services and through scale economies. In total, this transaction projects merger synergies in the billions—more than $2.5 billion a year in synergies by 2020 and more than $25 billion on a net present value basis. To sustain the pretense that there will be a negligible price increase, the government is forced to maintain that none of these innovations will benefit consumers at all. The imbalance between the alleged theoretical harm and verifiable real-world benefits of this transaction is even more striking when one considers their timing: many of the benefits will accrue in the short term, while much of the predicted price increase would not occur (if at all) for 53 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 63 of 76 years. For example, one major benefit of the merger, which the government itself concedes, is the “elimination of double marginalization”—i.e., the combination of each company’s profit margin into one—which will immediately result in the merged firm charging lower consumer prices. By contrast, the trivially higher consumer prices the government predicts would occur only years later, after one or more bargaining cycles for each distributor. As the Comcast example discussed above shows, there will be zero price increase for the largest cable provider in the nation for at least four years. To avoid the wholly theoretical (and factually unsupportable) risk of a higher price for Comcast and other providers in the future, the government would deny consumers the immediate concrete benefit of a price reduction in DIRECTV services, as well as the competitive responses that reduction would necessarily provoke, not to mention all the other benefits the government cannot seriously contest. That outcome makes no logical, legal, or economic sense. II. THE GOVERNMENT’S THEORY THAT THE MERGED ENTITY WILL LIKELY COORDINATE WITH COMCAST TO WITHHOLD PROGRAMMING FROM VIRTUAL MVPDS ALSO FAILS. The government has abandoned its claim that the combined firm would unilaterally “withhold . . . programming entirely from some virtual MVPDs” that provide multichannel TV packages over the Internet. Compl. ¶ 40. For good reason: virtual MVPDs are growing quickly in both numbers and popularity, and the merged firm will be unable to do anything to stop that inexorable trend. See supra pp. 16-18, 51-52. Nor would it even want to do so. Turner receives substantial affiliate fees and advertising revenue from these virtual distributors. And Turner would have rapidly diminishing relevance and a grim long-term outlook if millennials and other cord-cutters did not see Turner channels in the programming arrays offered by online providers. Indeed, it is far better from Turner’s perspective that consumers choose virtual MVPDs that carry its networks than that they sign up only for non-network-based services like Netflix. For 54 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 64 of 76 its part, AT&T benefits from virtual MVPDs because they encourage use of AT&T’s nationwide wireless broadband network, which increases revenue for AT&T and discourages consumers from “churning” away from its own broadband products. Indeed, as AT&T CEO Randall Stephenson will explain, AT&T’s core strategy is to embrace the mobile distribution of premium video, exploiting the major competitive advantage its nationwide wireless broadband network provides over Comcast and other cable providers. The government instead argues that, even though post-merger AT&T would not withhold Turner content from virtual MVPDs unilaterally, the merged firm would coordinate with Comcast to do so, thereby ostensibly achieving benefits neither firm could achieve on its own. The government will not be able to prove that speculative claim at trial. A. The Government Must Prove That the Firms Would Have Both the Incentive and the Ability To Coordinate in Depriving Virtual MVPDs of Their Content. Coordination, sometimes referred to as “tacit collusion,” “oligopolistic price coordination,” or “conscious parallelism,” Brooke Grp. Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209, 227 (1993), does not involve an express agreement between rivals. Rather, coordination “describes the process, not in itself unlawful, by which firms in a concentrated market might in effect share monopoly power . . . by recognizing their shared economic interests and their interdependence with respect to price and output decisions.” Id. Coordination between rivals can occur only if the firms can solve what economists call “cartel problems,” i.e., the difficulties of maintaining a consensus to take actions that would not be in each company’s individual interest absent coordination. See George J. Stigler, A Theory of Oligopoly, 72 J. Pol. Econ. 44, 44-46 (1964). These problems make the “anticompetitive minuet” of tacit coordination “most difficult to compose and to perform.” Brooke Grp., 509 U.S. at 227-28. 55 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 65 of 76 The first step—establishing a consensus—“requires harmonizing the incentives of participating firms and mitigating firm uncertainty concerning rival firms, so that they can effectively coordinate their behavior.” In re B.F. Goodrich Co., 110 F.T.C. 207, 295 (1988); see IV Areeda, Antitrust Law ¶ 919, at 171 (4th ed. 2016) (“Coordinated effects analysis of mergers requires a prediction about how a merger might change firms’ incentives to coordinate their behavior.”). The second step—enforcing the consensus—is equally critical, because without “mutual trust and forbearance . . . an informal collusive arrangement is unlikely to overcome the temptation to steal a march on a fellow colluder by undercutting him slightly.” Hospital Corp., 807 F.2d at 1388-89. Consequently, firms will not coordinate unless they can “retaliate effectively if and when cheating occurs.” Goodrich, 110 F.T.C. at 295. In the government’s words, the “ability of rival firms to engage in coordinated conduct depends on the strength and predictability of rivals’ responses to a price change or other competitive initiative.” DOJ & FTC, Horizontal Merger Guidelines § 7, at 25 (2010). B. Even the Government’s Expert Does Not Opine That Coordinated Conduct Is Probable or Likely To Cause Harm. The government must prove that harmful post-merger coordination is probable. See Baker Hughes, 908 F.2d at 984; see also United States v. Oracle Corp., 331 F. Supp. 2d 1098, 1109 (N.D. Cal. 2004) (rejecting government claim where it had not proved that defendants “would likely engage in coordinated interaction”) (emphasis added). The government’s expert did not even attempt that showing. As a result, he explicitly testified that he could not say whether it was likely that the merged firm would coordinate with Comcast to withhold content from virtual MVPDs. He refused even to say whether such coordination was 1% probable. The problem, he has explained, is that withholding from virtual MVPDs would sacrifice affiliate fees and advertising revenues, which would be contrary to each firm’s interest. Coordinated 56 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 66 of 76 withholding thus could be profitable for the merged firm and for Comcast (and thus would likely be pursued) only if they both would receive profits from new subscribers or retained existing subscribers that exceeded their lost affiliate fees and advertising revenues. But because he admittedly never undertook that critical calculation, he has no idea whether coordinated withholding would be profitable for both firms, and thus no idea whether it would be likely to occur. The Court need go no further to reject this claim. C. The Merged Firm Would Have No Incentive To Coordinate with Comcast To Thwart Virtual MVPD Growth. In addition, the evidence will thoroughly refute the government’s contention that the merged firm and Comcast/NBCU will share a common interest in withholding content from virtual MVPDs—the core predicate to its coordination theory. Post-merger, AT&T will have no interest whatsoever in trying to impede the success of virtual MVPD services. AT&T wants growth in these services because more online video viewership increases usage of its wireless broadband network, where AT&T has a decided market advantage over Comcast. AT&T has more than 100 million wireless connections in the U.S.; Comcast has almost no wireless customers (380,000 by its own account 37). Consumers who watch video on AT&T mobile devices will benefit the merged firm even if they use an over-the-top provider other than DIRECTV NOW: they will increase their use of AT&T’s broadband network; they will expand the number of devices to which AT&T can send targeted advertising, the most valuable form of advertising in the industry; and they will become less likely to switch (“churn”) to other wireless providers. They will also become more likely over time to treat their mobile broadband 37 See Thomson Reuters StreetEvents, CMCSA – Q4 2017 Comcast Corp Earnings Call, Edited Transcript at 3 (Jan. 24, 2018) (statement of Comcast Corp. Chairman & CEO Brian L. Roberts). 57 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 67 of 76 connection as their default video viewing mechanism, again benefiting the merged firm over Comcast. Beyond that, as discussed above, AT&T will want Turner content to be included in this growing sector of the video ecosystem, both to obtain affiliate fees and advertising revenue and to encourage consumers to continue to use network-based distribution options rather than turning to Netflix and similar alternatives. Distributing Turner programming through virtual MVPDs will be essential to the continued relevance and value of that programming. Because Comcast has almost no mobile broadband customers, its incentives are distinctly different. But they clearly would not include an incentive to withhold content from virtual MVPDs in conjunction with the merged firm. First, coordinating with the merged firm to deny content to virtual MVPDs would require Comcast to make a massively disproportionate sacrifice in profits. NBCU’s broadcast and cable content is significantly more valuable than Time Warner’s, so withholding that content will always inflict greater economic pain on NBCU in the form of lost affiliate and advertising revenue than withholding Turner content would inflict on AT&T. Second, while Comcast offers service in 40 States, there are areas of the country it does not serve. If Comcast withheld content from nationwide virtual MVPDs, it would be denying content to many millions of consumers who could not even switch to Comcast service. In other words, Comcast would be forgoing revenue for a wide swath of consumers with no hope of obtaining revenues from them to compensate for those losses. The government’s expert does not explain why one firm would agree to work with its chief rival on such asymmetrical terms. D. Multiple Obstacles Would Prevent the Parties from Successfully Coordinating To Withhold Content from Virtual MVPDs. Even assuming, despite the contrary evidence, Comcast and the merged firm actually shared an incentive to impede the growth of virtual MVPDs and the customers they would bring to AT&T’s wireless service, they could not successfully coordinate to implement such a strategy. 58 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 68 of 76 1. The arbitration/no-blackout mechanism precludes any ability to coordinate. The arbitration/no-blackout mechanism would categorically preclude the merged entity from coordinating with Comcast over the terms of virtual MVPD distribution deals and is, by itself, a complete response to the government’s claim. Because of this mechanism, the combined firm could not fulfill its end of any implicit coordination deal with Comcast/NBCU. For one thing, Turner would be unable to withhold content from virtual MVPDs. For another, an arbitrator, not Turner, would decide the terms on which Turner must sell programming to virtual MVPDs in the event of an impasse. These mechanisms would make the merged firm an extremely unreliable coordination partner for Comcast, eliminating the “mutual trust and forbearance” that coordinated conduct requires. Hospital Corp., 807 F.2d at 1388-89. 38 The arbitration/no-blackout mechanism also precludes Turner from withholding from any MVPD (traditional and virtual, as well as bona fide new entrants), which in turn ensures that there will be at least 11 or 12 distributors with access to Turner content in any given area. In that context, the government cannot come close to showing harm to competition based on this theory. 38 The arbitration/no-blackout mechanism also prevents Turner from forcing a competing virtual MVPD to take more channels than it currently has and thus undermining the economics of the “skinny bundle.” Section A.2 of the “Arbitration Procedures” permits a virtual MVPD (the arbitration “Claimant”) to demand a “standalone offer” for a bundle or bundles of Turner networks “that Turner has licensed to AT&T, the Claimant, or another Video distributor for distribution to consumers on or after October 22, 2014.” In a renewal context (provided the prior carriage agreement was made “on or after October 22, 2014”), that means the Claimant could demand a stand-alone offer for the same “skinny bundle” of networks the Claimant currently carries (or that another distributor currently carries). If the parties do not reach agreement, the Claimant may then submit a final offer to arbitration. Arb. Procedures § A.4. Turner then is obliged to respond with its own final offer, which must be limited to “the Turner Networks requested by the Claimant.” Id. § A.6 (emphasis added). 59 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 69 of 76 2. Staggered multi-year contracts preclude enforcing coordination. Independently, the structure of the contracts in this industry precludes the kind of competitive response necessary to enforce coordination. As the evidence will show, the programming market is defined by staggered multi-year agreements, under which Comcast’s and Turner’s contracts with providers frequently expire years apart. That fact makes it difficult—or even impossible, depending on the contract timing—to respond quickly enough to rivals’ decisions. If, for example, Comcast decided to forgo significant revenues by withholding content from a given virtual MVPD, on the expectation that the merged entity would eventually do the same during its own contracting cycle, it would not actually know for months or even years whether the merged firm would live up to its end of the tacit bargain (even leaving aside the arbitration obligation just discussed). Throughout all that time, it would be giving up revenue without any assurance that Turner would follow its lead. Because neither party can trust that the other will withhold months or years later, it will have little incentive to forgo the profits associated with virtual MVPD deals. 39 E. Past Information Sharing on Distinct Issues Does Not Establish an Incentive or Ability To Coordinate with Comcast To Deny Content to Virtual MVPDs. The government at trial may try to cite past episodes where it claims AT&T and Time Warner employees sought to share information among industry participants concerning specific issues. One involved discussions among distributors about working together to benefit consumers by negotiating rights to multiple seasons of television series, referred to as “stacking 39 Nor would it be sufficient for Comcast to withhold from one virtual MVPD and Turner to withhold from another. Under the government’s theory, the benefits of coordination depend on both companies withholding from a given virtual MVPD. If each company withholds from a different virtual MVPD, each would suffer immediate losses without obtaining any benefit and would need to rely on the other company to do the same at some future date, with no mechanism to enforce that expectation. 60 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 70 of 76 rights” (because seasons are “stacked” one after the other). The other concerned a civil antitrust lawsuit brought by the government against DIRECTV involving alleged discussions in 2014 regarding a regional sports network that broadcast Los Angeles Dodgers games (SportsNet LA or “SNLA”). The evidence will show that neither AT&T nor Time Warner actually coordinated with any rival on stacking rights (where discussions never led to any agreement) or on the decision whether or not to carry SNLA at the price being demanded (where, after DIRECTV’s motion to dismiss, the government settled its “information sharing” case, with no admission of liability or proof of the government’s disputed allegations). These incidents are irrelevant, yet the government persists in pursuing them, over defendants’ objection. Even assuming they show an ability to share information, that is not at issue here, because nobody disputes that each firm will readily know from publicly available information whether the other agreed to provide content to a virtual MVPD. 40 And the ability to share information is nowhere near enough to satisfy the government’s burden to establish a likelihood of anticompetitive coordination between two bitter rivals. See FTC v. Arch Coal, Inc., 329 F. Supp. 2d 109, 138 (D.D.C. 2004) (even where there is mutual access to “[k]ey market information,” “whether anticompetitive coordination is likely requires closer examination” of critical market features); see also Monsanto Co. v. Spray-Rite Serv. Corp., 465 U.S. 752, 762 (1984) (“[T]he fact that a manufacturer and its distributors are in constant communication about prices and marketing strategy does not alone show that the distributors are not making independent pricing decisions.”). Rather, the government must prove that both the merged firm and Comcast will have the incentive to thwart virtual MVPD growth; that it would be profitable 40 The “stacking rights” discussions are irrelevant for an additional reason as well: they relate to whether SVODs like Netflix will have exclusive access to reruns of particular shows, not to whether virtual MVPDs will have access to cable channels such as TBS and TNT. 61 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 71 of 76 for them to jointly withhold from virtual MVPDs; that they could successfully execute a coordinated strategy to do so notwithstanding the divergent interests and impediments; and that any such coordination would significantly lessen competition. For the reasons discussed above, the government cannot show any of those things—and in most respects it does not even try. 41 III. THE GOVERNMENT’S THEORY THAT THE COMBINED COMPANY WILL LIMIT HBO’S USE AS A “PROMOTIONAL TOOL” IS ILLOGICAL AND CONTRARY TO THE EVIDENCE. With respect to HBO, the government alleges that the merged company will gain the incentive and ability to prevent rival distributors from using HBO as a promotional tool to attract and retain subscribers and that such conduct will substantially lessen competition. Compl. ¶ 39. To prevail on such a claim, the government must show that there is no effective substitute for HBO’s use as a promotional tool, so that depriving competitors of that tool would substantially impede their ability to compete with AT&T/DIRECTV in the distribution market. See Westman Comm’n Co. v. Hobart Int’l, Inc., 796 F.2d 1216, 1220-21 (10th Cir. 1986); IVA Areeda, Antitrust Law ¶¶ 1004, 1008 (4th ed. 2016). The evidence will show the opposite—that HBO needs these promotions far more than MVPDs need to promote it, and so HBO typically structures its contracts with MVPDs to encourage the widest distribution possible; that HBO is not an essential or unique promotional tool for any MVPD; that the combined firm would have no incentive to restrict distributors from using HBO as a promotional tool; and that HBO’s existing agreements allow distributors to make many promotional offers without the need for 41 Even further afield are AT&T documents describing the existing MVPD business as a “cash cow” or using similar colorful phrases. Those documents reflect the reality that the traditional MVPD business is a mature, low-growth one. AT&T CEO Randall Stephenson, who establishes the company’s strategy, has long made clear that AT&T seeks to embrace the movement of video to broadband and mobile. 62 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 72 of 76 additional approval, so HBO could not prevent them from using HBO to promote their programming even if it wanted to do so. It is true that MVPDs use HBO (among many other benefits) in promotions to gain or maintain customers, but this is not because they need to do so. Rather, it is HBO that needs the promotions and incentivizes distributors to actively promote it. HBO’s agreements with its distributors are specifically designed to encourage broad distribution, by providing a volumediscount pricing structure—the per-subscriber rate an MVPD pays is reduced as subscribership increases. Furthermore, HBO negotiates for guarantees that MVPDs will use HBO in their promotions and provides substantial funds, sometimes millions of dollars, to support and subsidize these promotion efforts. Along the same lines, HBO will frequently waive fees or contractual requirements in order to induce a distributor to launch a special HBO-related promotion. The evidence will show that the government has it exactly backwards: it is HBO that needs its distributors to promote its programming—not vice versa—and nothing in this merger will affect those fundamental business imperatives. HBO faces greater competition than ever from other premium channels like Showtime and Starz, and from subscriber video on-demand services like Netflix. Distributors can—and do—use all of these other services in promotions. Distributors also rely on a wide range of promotions that use no premium channels at all, including free equipment and—most importantly—gift cards and other price reductions. Furthermore, in order to attract the content and talent it needs to compete with the exploding universe of other high-quality video programming, HBO requires the broadest distribution possible. The vast majority (more than two-thirds) of MVPD customers do not currently subscribe to HBO. Without assurances of wide distribution, content creators would 63 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 73 of 76 leave HBO and go elsewhere. HBO will continue to need to rely on MVPDs post-merger to distribute its programming, and it will have the same incentives to encourage these distributors to package and promote HBO as widely as possible. Finally, HBO’s existing contracts provide distributors with substantial rights to use HBO to promote their own services without the need for additional approvals. So even if HBO postmerger were to seek to limit distributors from using HBO as a promotional tool, the agreements already in place provide distributors with substantial rights—and indeed, obligations—to promote HBO to attract and retain subscribers. The government has no evidence showing otherwise. Its lead expert testified that he did not even calculate whether it would be economically profitable for the merged firm to restrict HBO promotions, even while conceding that AT&T would do so only if it were profitable. He also did not quantify a single dollar of potential consumer harm associated with this theory. In the end, the evidence will not show that HBO is essential to MVPDs’ ability to compete for subscribers. The fact that some MVPDs already compete effectively without using HBO as a promotional tool means that the government cannot meet its burden of proving that competition would likely suffer substantial harm. And most distributors already have the right to promote HBO under their existing contracts, so even if HBO’s entire business strategy were to change after the merger, those rights and obligations remain unaffected. CONCLUSION This merger will provide substantial benefits to consumers and, contrary to the government’s contentions, will strengthen rather than harm competition in these markets. For those reasons, and based on the evidence to be adduced at trial, this Court should grant judgment for defendants and permit this merger to proceed. 64 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 74 of 76 Date: March 9, 2018 Respectfully submitted, By: /s/ Daniel M. Petrocelli Daniel M. Petrocelli (pro hac vice) dpetrocelli@omm.com M. Randall Oppenheimer (pro hac vice) roppenheimer@omm.com O’MELVENY & MYERS LLP 1999 Avenue of the Stars Los Angeles, CA 90067 Telephone: (310) 553-6700 Facsimile: (310) 246-6779 Katrina M. Robson (D.C. Bar No. 989341) krobson@omm.com O’MELVENY & MYERS LLP 1625 Eye Street, N.W. Washington, D.C. 20006 Telephone: (202) 383-5300 Facsimile: (202) 383-5414 Counsel for AT&T Inc., DIRECTV Group Holdings, LLC, and Time Warner Inc. Peter T. Barbur (pro hac vice) pbarbur@cravath.com Kevin J. Orsini (pro hac vice) korsini@cravath.com Margaret Segall D’Amico (pro hac vice) mdamico@cravath.com CRAVATH, SWAINE & MOORE LLP 825 8th Avenue New York, NY 10019 Telephone: (212) 474-1140 Facsimile: (212) 474-3700 Counsel for Time Warner Inc. 65 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 75 of 76 Robert C. Walters (pro hac vice) rwalters@gibsondunn.com Michael L. Raiff (pro hac vice) mraiff@gibsondunn.com GIBSON DUNN & CRUTCHER 2100 McKinney Avenue Suite 1100 Dallas, TX 75201 Telephone: (214) 698-3114 Facsimile: (214) 571-2932 Counsel for AT&T Inc. and DIRECTV Group Holdings, LLC 66 Case 1:17-cv-02511-RJL Document 77 Filed 03/09/18 Page 76 of 76 CERTIFICATE OF SERVICE I hereby certify that, on March 9, 2018, I caused the foregoing to be electronically filed with the Clerk of the Court using CM/ECF, which will send notification of such filing to all registered participants. By: /s/ Daniel M. Petrocelli Daniel M. Petrocelli (pro hac vice) O’MELVENY & MYERS LLP 1999 Avenue of the Stars Los Angeles, CA 90067 Telephone: (310) 553-6700 Facsimile: (310) 246-6779 dpetrocelli@omm.com