Federal Communications Commission FCC 15-94 Before the Federal Communications Commission Washington, D.C. 20554 In the Matter of Applications of AT&T Inc. and DIRECTV For Consent to Assign or Transfer Control of Licenses and Authorizations ) ) ) ) ) ) ) ) MB Docket No. 14-90 MEMORANDUM OPINION AND ORDER Adopted: July 24, 2015 Released: July 28, 2015 By the Commission: Chairman Wheeler and Commissioners Clyburn and Rosenworcel issuing separate statements; Commissioner Pai approving in part, dissenting in part and issuing a statement; Commissioner O’Rielly approving in part, concurring in part and issuing a statement. TABLE OF CONTENTS Heading Paragraph # I.  EXECUTIVE SUMMARY .................................................................................................................... 1  II.  DESCRIPTION OF THE APPLICANTS ............................................................................................ 11  A.  AT&T ............................................................................................................................................. 11  B.  DIRECTV ...................................................................................................................................... 13  III. THE PROPOSED TRANSACTION .................................................................................................... 15  A.  Description ..................................................................................................................................... 15  B.  Application and Review Process ................................................................................................... 16  IV. STANDARD OF REVIEW AND PUBLIC INTEREST FRAMEWORK .......................................... 18  V.  QUALIFICATIONS OF APPLICANTS ............................................................................................. 24  A.  Background .................................................................................................................................... 24  B.  DIRECTV ...................................................................................................................................... 25  C.  AT&T ............................................................................................................................................. 26  1.  Minority Cellular Partners Coalition Comments ..................................................................... 27  a.  Standing ............................................................................................................................ 31  b.  Pro Forma Transactions ................................................................................................... 32  c.  Notice of Apparent Liability ............................................................................................. 39  d.  CALEA ............................................................................................................................. 41  2.  New Networks Institute & Teletruth Petition .......................................................................... 45  VI.    COMPLIANCE WITH COMMUNICATIONS ACT AND FCC RULES AND POLICIES ............ 52  VII.  BACKGROUND ON VIDEO PROGRAMMING DISTRIBUTORS ............................................... 53  VIII. INCREASED MARKET CONCENTRATION IN VIDEO DISTRIBUTION SERVICES ............. 59  IX.    HORIZONTAL EFFECTS ANALYSIS ............................................................................................ 82  A.  Evaluation of Potential Unilateral Effects Using Economic Analysis ........................................... 83  1.  Merger Simulation ................................................................................................................... 86  2.  BH Simulation Structure and the “Modified Simulation” ....................................................... 91  3.  Effects of the Transaction on Consumers .............................................................................. 105  4.  Competitive Effects of Integrated Bundles ........................................................................... 111  5.  Reduction of Competition in Video Distribution .................................................................. 127  6.  Standalone Broadband ........................................................................................................... 134  Federal Communications Commission FCC 15-94 2 B.  Documentary and Other Record Evidence of Competition between AT&T and DIRECTV and the Need for Bundles ............................................................................................................. 146  C.  Conclusion ................................................................................................................................... 160  X.  ADDITIONAL COMPETITIVE EFFECTS AND PUBLIC INTEREST HARMS RAISED IN THE RECORD .............................................................................................................................. 161  A.  Limits on Competitors’ Access to Programming ......................................................................... 162  1.  Limiting Access to RSNs and Other Affiliated Programming .............................................. 163  2.  Exclusive Programming Agreements .................................................................................... 177  3.  Restricting Access to Online Video Content ......................................................................... 185  4.  Forcing Competitors to Compensate Programmers for Reduced Payments from the Applicants .............................................................................................................................. 188  B.  Lack of Regulatory Parity ............................................................................................................ 193  C.  Potential Harm to Subscribers’ Access to OVD Services ............................................................ 198  1.  Increased Incentive to Discriminate Against Unaffiliated OVDs ......................................... 201  2.  Potential Levers for Discrimination Against Unaffiliated OVDs ......................................... 206  a.  Data Caps ........................................................................................................................ 208  b.  Interconnection ............................................................................................................... 214  D.  Harm to Supply, Quality, and Diversity in Video Programming ................................................. 220  1.  Background on Video Programming ..................................................................................... 222  2.  Potential Competitive Harms ................................................................................................ 225  a.  Increased Leverage of Combined Entity in Programming Negotiations ........................ 225  b.  PEG Channels ................................................................................................................. 239  c.  Local Broadcast Television Stations ............................................................................... 245  E.  Video Device Market ................................................................................................................... 249  F.  Potential Loss of DIRECTV as a Partner for MDU Broadband Entrants .................................... 255  G.  Increased Incentive of Combined Entity to Hinder Competition for Broadband in MDUs ........ 260  H.  Increased Incentive of Combined Entity to Hinder Competition in Mobile Wireless Sector ........................................................................................................................................... 265  I.  Increased Incentive and Ability of Combined Entity to Shift Wired Subscribers to FWLL ....... 268  J.  Use of Orbit and Spectrum Resources ......................................................................................... 271  XI. ANALYSIS OF POTENTIAL BENEFITS ........................................................................................ 273  A.  Analytical Framework ................................................................................................................. 273  B.  Improved Bundles ........................................................................................................................ 278  C.  Reduced Payments for Programming and Bundling Efficiencies ................................................ 283  1.  Reduced Payments for Content Acquisition .......................................................................... 283  2.  Other Cost Savings and Efficiencies ..................................................................................... 292  3.  Innovation in Video Services ................................................................................................ 295  a.  Traditional Video ............................................................................................................ 296  b.  OVD ................................................................................................................................ 303  c.  Improved Advertising Capabilities ................................................................................. 305  D.  Video Programming Market ........................................................................................................ 307  E.  Video Device Market ................................................................................................................... 311  F.  Expanded Deployment of Fiber to the Premises .......................................................................... 315  1.  Analysis of the FIM ............................................................................................................... 327  a.  Modification of FIM ....................................................................................................... 328  b.  Results from Modifications............................................................................................. 338  2.  Non-FIM Factors ................................................................................................................... 342  3.  Conclusion ............................................................................................................................. 344  G.  Evaluation of Applicants’ Claimed Fixed Wireless Local Loop Benefits ................................... 346  1.  Introduction ........................................................................................................................... 346  2.  FWLL Coverage and Performance Claims ........................................................................... 348  3.  Claims that FWLL Would Benefit 13 Million Rural Customers .......................................... 355  4.  Competitive Standalone FWLL and DIRECTV Integrated Bundles Would Be a Federal Communications Commission FCC 15-94 3 Benefit of the Transaction ..................................................................................................... 358  5.  FWLL Deployment is Transaction Specific .......................................................................... 365  6.  Conclusion ............................................................................................................................. 376  H.  Other Potential Public Interest Benefits ....................................................................................... 378  1.  Cybersecurity ........................................................................................................................ 378  2.  Diversity Practices ................................................................................................................. 389  3.  Labor Practices ...................................................................................................................... 390  XII. REMEDIES ....................................................................................................................................... 392  A.  Introduction .................................................................................................................................. 392  B.  Fiber to the Premises Deployment Commitment ......................................................................... 394  C.  Non-Discriminatory Usage-Based Practices ................................................................................ 395  D.  Internet Interconnection Disclosure Requirement ....................................................................... 396  E.  Discounted Broadband Services for Low-Income Subscribers ................................................... 397  F.  Reporting and Outside Compliance Officer ................................................................................. 398  XIII. BALANCING POTENTIAL PUBLIC INTEREST HARMS AND BENEFITS ............................ 399  XIV. CONCLUSION ................................................................................................................................ 400  XV.  ORDERING CLAUSES ................................................................................................................... 401  APPENDIX A – List of Licenses to be Transferred APPENDIX B – Conditions APPENDIX C – Analysis of Merger Simulation Models APPENDIX D – Analysis of AT&T’s FWLL Coverage and Performance Claims and Claimed Rural Benefits Federal Communications Commission FCC 15-94 I. EXECUTIVE SUMMARY 1. In this proceeding, we approve, subject to conditions, the applications of AT&T Inc. (“AT&T”) and DIRECTV (collectively, the “Applicants”) for Commission consent to the transfer of control of various Commission licenses and other authorizations from DIRECTV to AT&T pursuant to Section 310(d) of the Communications Act of 1934, as amended (the “Act”).1 2. Our consent to transfer these licenses is based on a careful review of the economic, documentary, and other record evidence. We engaged in a rigorous analysis of the potential harms and benefits to ensure that the proposed transaction serves the public interest, convenience, and necessity.2 Based on this review, we have concluded that, with the adoption of certain conditions designed to address specific harms and confirm certain benefits that would result from the transaction, the license transfer is in the public interest. 3. Our record supports the Applicants’ claim that the newly combined entity will be a more effective multichannel video programming distributor (“MVPD”) competitor, offering consumers greater choice at lower prices. As standalone companies, neither has the full set of assets necessary to compete against the dominant providers of video service. Although DIRECTV has approximately 20 million video subscribers, it lacks broadband3 capabilities. This limits DIRECTV in its ability to offer video-on- demand (“VOD”) and other interactive viewing experiences that consumers increasingly seek. In addition, DIRECTV’s current partnerships with broadband providers to sell third-party bundles of broadband and DIRECTV satellite video cannot match the convenience and lower prices associated with bundles of broadband and video offered by a single provider. AT&T offers bundles of its own broadband and U-verse video where it has deployed fiber to the node (“FTTN”) or fiber to the premises (“FTTP”) technologies – however, it too faces significant competitive challenges. With fewer than 6 million subscribers, AT&T’s video product is hampered by higher costs of procuring programming – limiting its ability to both offer lower consumer prices and expand its high-speed broadband footprint. 4. We find that the combined AT&T-DIRECTV will increase competition for bundles of video and broadband, which, in turn, will stimulate lower prices, not only for the Applicants’ bundles, but also for competitors’ bundled products – benefiting consumers and serving the public interest. We also expect that this improved business model will spur, in the long term, AT&T’s investment in high-speed broadband networks, driving more competition and thus expanding consumer access and choice. This is, in other words, a bet on competition. 5. However, the transaction also creates the potential for competitive harms, which we impose conditions to address. 6. First, the Applicants’ claim that a benefit of this transaction is that it would increase AT&T’s incentive to deploy FTTP. To the contrary, we find that the transaction creates, at least in the short term, a disincentive to deploy faster broadband because an FTTP buildout would potentially “cannibalize” profits from AT&T’s newly acquired DIRECTV subscribers and revenue. To address this harm, we impose as a condition that the combined entity deploy FTTP to 12.5 million locations within four years, to capture all of AT&T’s pre-transaction planned deployment, its projected deployment absent the transaction, and the deployment that the record suggests is profitable as a result of the transaction. In addition, and because it is important that competition with cable also reach public institutions, AT&T is required to offer to schools and libraries where it deploys FTTP, which is about 6,000 institutions, the ability to purchase 1 gigabit E-rate services from AT&T. 1 See 47 U.S.C. § 310(d); Application of AT&T Inc. and DIRECTV, Description of Transaction, Public Interest Showing, and Related Demonstrations, transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 10 (filed June 11, 2014) (“Application”). 2 47 U.S.C. § 310(d). 3 Unless the context indicates otherwise, this Order uses the term “broadband” colloquially to refer to the Internet services AT&T provides other than dial-up Internet services. Federal Communications Commission FCC 15-94 5 7. Second, Applicants have stated that as a result of this transaction they will be able to offer their own new, flexible, and innovative online video products, which increases the risk that the combined entity will use its broadband services to hamper competition from online video content or online video distribution services. We also note that AT&T is alone among the large Internet Service Providers (“ISPs”) in applying fixed data caps across its broadband services. Thus we impose as a condition certain restrictions on the use of discriminatory usage-based allowances. We also impose certain disclosure requirements for interconnection agreements and interconnection metrics, which will help the Commission address any future concerns about the nature of AT&T’s exchange of Internet traffic and the potential impact of congestion upon consumers. Coupled with the FTTP buildout requirements, these conditions improve the ability of alternative video distribution methods to replace the loss of a horizontal MVPD competitor within AT&T’s video footprint that results from this merger. 8. Third, while we acknowledge that a benefit of the transaction is the Applicants’ ability to be a more effective competitor to cable providers, we are concerned that the Applicants’ efforts to expand consumer choice for bundles might prove to be an obstacle for low-income populations who desire standalone broadband. Thus we impose as a condition a requirement that the Applicants offer discounted broadband Internet access to eligible consumers. 9. In addition to addressing potential harms and confirming potential benefits, these conditions as a group create the opportunity for more robust broadband and video distribution competition in a variety of respects. To ensure that the goals of these conditions are achieved, we require as a condition of this transaction that the Applicants employ an independent, outside officer responsible for monitoring and reporting to the Commission any failure to comply with the conditions imposed by this Order. 10. In general, these conditions will run for four years from the consummation of the transaction and, with them in place, we find that this combination is in the public interest. II. DESCRIPTION OF THE APPLICANTS A. AT&T 11. AT&T provides Internet, video, local and long distance voice, mobile wireless voice and broadband, and Wi-Fi services in the United States.4 In addition, AT&T offers worldwide wireless service and Internet Protocol (“IP”)-based business communications services.5 Within the United States, AT&T’s wireline footprint covers portions of 22 states, while its Long Term Evolution (“LTE”) wireless network covers approximately 300 million people.6 AT&T offers bundles of high-speed broadband, video, and Voice over Internet Protocol (“VoIP”) services under its U-verse brand within portions of its wireline footprint.7 Through its Project Velocity IP (“Project VIP”), AT&T has stated that it has begun an expansion of its U-verse services to reach approximately 57 million customer locations or 75 percent of its wireline footprint.8 Of these 57 million customer locations, AT&T states that it plans to deploy FTTN or FTTP technologies to deliver U-verse video, high-speed broadband, and VoIP services to 33 million customer locations.9 For the remaining 24 million customer locations where U-verse services are 4 See Application at 10. 5 Id. 6 Id. 7 Id. 8 Id. at 10-11. 9 Id. at 10. According to the Application, AT&T currently uses FTTN architecture in most of the U-verse video footprint. Id. at 11. Under this approach, AT&T deploys fiber to neighborhood nodes. Individual customer locations are connected to the network via existing copper plant using very-high-bit-rate digital subscriber line (“VDSL”) technology. Id. U-verse FTTN offers speeds of up to 45 megabits per second (“Mbps”). Id. At the time (continued….) Federal Communications Commission FCC 15-94 6 or will be available, AT&T’s IPDSLAM (“IPDSL”) technology will deliver U-verse high-speed broadband and VoIP services, but not video services.10 12. AT&T currently provides broadband Internet access service to approximately 14.5 million residential subscribers of which 6.5 million subscribers receive broadband Internet access service at download speeds above 10 Mbps.11 AT&T provides MVPD services to approximately 6 million subscribers.12 AT&T estimates that more than 97 percent of its U-verse video subscribers purchase at least one other U-verse product and about two-thirds of U-verse video subscribers bundle three or four services from AT&T.13 Post-transaction, IPDSL customers, which currently are not offered U-verse video service, could purchase DIRECTV satellite video from the combined entity. B. DIRECTV 13. DIRECTV offers direct-to-home satellite digital television services to consumers nationwide.14 According to the Application, DIRECTV is a “pure-play” satellite video provider with approximately 20 million U.S. subscribers.15 Currently, DIRECTV does not provide any broadband or voice services of its own.16 It does offer synthetic service17 bundles of DIRECTV satellite video service and broadband and/or voice services provided by various third-party telecommunications, cable, and satellite partners, including AT&T.18 14. DIRECTV owns and operates two regional sports networks (“RSNs”), Root Sports Pittsburgh and Root Sports Rocky Mountain, and holds a minority interest in, and manages, the Seattle- based RSN, Root Sports Northwest.19 In a recent joint venture, DIRECTV and AT&T have also acquired majority ownership of a Houston-area RSN (“CSN Houston”) out of bankruptcy and relaunched it as (Continued from previous page) of the Application, AT&T was using in Austin, Texas, FTTP architecture in which fiber extends all the way to a customer’s location. Id. AT&T provides “U-verse with GigaPower” service over this FTTP architecture, and it plans to offer Internet speeds of up to 1 gigabit per second (“Gbps”). Id. Prior to the DIRECTV transaction, AT&T announced plans to bring its FTTP deployment and U-verse with GigaPower service to Dallas; Raleigh-Durham, N.C.; and Winston-Salem, N.C.; and to expand further, to as many as 21 other major metropolitan areas, including Atlanta; Chicago; Charlotte, N.C.; San Francisco; and Houston. Id. at 11-12. 10 Id. at 11. According to the Application, IPDSL provides high-speed broadband over copper wires at speeds up to 18 Mbps, but it is not suitable for delivering U-verse video services. Id. at 12. 11 See AT&T Inc. Updated Response to Sept. 9, 2014, Information and Discovery Requests, transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Vanessa Lemmé, Media Bureau, FCC, MB Docket No. 14-90, Exhibit 5.b.1 – updated (Oct. 20, 2014) (“AT&T Updated Response to Sept. 9, 2014, Information Request”). 12 See id. 13 Application at 12. Where AT&T U-verse FTTP, FTTN, or IPDSL are not available within AT&T’s wireline footprint, AT&T sells legacy digital subscriber line (“DSL”) Internet service, providing speeds of up to 6 Mbps, and does not sell its own video component. Id. at 12 n.14. 14 Id. at 13. 15 Id. DIRECTV also holds interests in entities with approximately 18 million video subscribers in Latin America. Id. at 13, 15. 16 Id. at 13-14. 17 A synthetic bundle is a bundle of services offered by two different companies. See infra ¶ 56 (discussing synthetic bundles). 18 Application at 14. DIRECTV has arm’s length agreements to provide these synthetic bundles with CenturyLink, AT&T, Verizon, Exede, Cincinnati Bell, HughesNet, Windstream, and Mediacom, among others. Id. 19 Id. Federal Communications Commission FCC 15-94 7 Root Sports Southwest.20 DIRECTV also has a 42 percent non-controlling interest in the Game Show Network, and smaller, minority interests in the MLB Network, the NHL Network, and a handful of other networks.21 III. THE PROPOSED TRANSACTION A. Description 15. AT&T has entered into an agreement with DIRECTV whereby AT&T will acquire DIRECTV in a stock-and-cash transaction.22 Under the terms of the agreement, each share of DIRECTV common stock will be converted into $28.50 in cash plus the right to receive between 1.724 and 1.905 shares of AT&T common stock, depending on AT&T’s stock price prior to closing.23 At closing, DIRECTV will merge with and into a wholly owned subsidiary of AT&T, Steam Merger Sub LLC, which will be the surviving entity and will be renamed “DIRECTV.”24 The new DIRECTV will own the stock of the subsidiaries of the pre-transaction DIRECTV, and these subsidiaries will continue to hold the Commission licenses and other authorizations they held prior to the transaction.25 B. Application and Review Process 16. On June 11, 2014, AT&T and DIRECTV filed the Application.26 On August 7, 2014, the Commission released the Public Notice accepting the Application for filing and establishing a pleading 20 Joint Opposition of AT&T Inc. and DIRECTV to Petitions to Deny and Condition and Reply to Comments, transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 55 (filed Oct. 16, 2014) (“Joint Opposition”). See also In re Houston Reg’l Sports Network, LP, 514 B. R. 211 (Bankr. S.D. Tex. 2014) (“In re Houston”). 21 Application at 14. 22 Id. at 16. 23 Id. 24 Id. 25 Id. at 16-17. 26 See supra n.1. Subsequent to filing the Application and prior to release of the Public Notice accepting the Application for filing, the Applicants submitted additional information. See Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (Aug. 6, 2014) (submitting merger simulation and supporting data relied upon in the Declaration of Michael L. Katz); Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (July 28, 2014) (submitting paper prepared by Compass Lexecon entitled “Additional Detail on the Demand Estimation, Merger Simulation, and Investment Model Analysis Performed by Professor Katz” and associated files); Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (July 28, 2014) (submitting paper entitled “Overview of AT&T FTTP Investment Model” and associated files); Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (July 24, 2014) (submitting copy of presentation made by Professors Berry and Haile); Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (July 21, 2014) (submitting additional materials supporting the merger simulation described in the presentation prepared by Professors Berry and Haile); Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (July 17, 2014) (submitting (1) presentation prepared by Professors Steve Berry and Phil Haile of Yale University on behalf of the Applicants; and (2) data supporting the merger simulation described in the presentation); Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (July 17, 2014) (submitting additional data supporting the merger simulation relied upon in the Declaration of Michael L. Katz); Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (July 8, 2014) (submitting additional materials relied upon in the Declaration of Michael L. Katz); Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (July 7, 2014) (submitting additional details on the number of (1) customer locations and subscribers for FTTP and FTTN technologies; and (2) consumer and business customer subscribers); Letter from (continued….) Federal Communications Commission FCC 15-94 8 cycle.27 Eight petitions to deny and thousands of public comments and other filings were received in this proceeding.28 In addition to building its record through public comment, the Commission requested additional information from the Applicants29 and other entities.30 The responses to those requests are included in the record,31 subject to the protections of the Protective Order issued in this proceeding.32 (Continued from previous page) Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (June 30, 2014) (submitting additional materials relied upon in the Declaration of Michael L. Katz); Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (June 27, 2014) (describing the (1) number of customer locations that AT&T serves by the following technology, and (2) the number of subscribers by service to each technology); Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (June 25, 2014) (submitting the FTTP model and supporting data relied upon in the Declaration of Michael L. Katz); Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (June 23, 2014) (submitting additional data supporting the merger simulation relied upon in the Declaration of Michael L. Katz); Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (June 20, 2014) (submitting additional materials relied upon in the Declaration of Michael L. Katz); Letter from William M. Wiltshire, Counsel for DIRECTV, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (June 19, 2014) (submitting materials considered by Michael L. Katz in preparing his Declaration); Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (June 17, 2014) (submitting merger simulation and supporting data relied upon in the Declaration of Michael L. Katz). 27 See Commission Seeks Comment on Applications of AT&T Inc. and DIRECTV to Transfer Control of FCC Licenses and Other Authorizations, MB Docket No. 14-90, Public Notice, DA 14-1129, 29 FCC Rcd 9464 (MB 2014) (“Public Notice”). The Public Notice established September 16, 2014, as the deadline for filing comments or petitions to deny, and October 16, 2014, as the deadline for responses to comments or oppositions to petitions to deny. See id. On August 28, 2014, the Media Bureau denied a request to extend the filing deadline for initial comments and petitions to deny. See Applications of AT&T Inc. and DIRECTV to Transfer Control of FCC Licenses, MB Docket No. 14-90, Order, DA 14-1253, 29 FCC Rcd 10318 (MB 2014). The deadline for filing replies was extended to January 7, 2015. See Commission Restarts Clock in Comcast-Time Warner Cable and AT&T-DIRECTV Merger Proceedings and Establishes Dates for Respective Pleading Cycles, MB Docket No. 14- 90, Public Notice, DA 14-1739, 29 FCC Rcd 14491 (MB 2014) (“Notice of Merger Pleading Cycle Restarts”). 28 Petitions to Deny or to Impose Conditions were filed by: Alliance for Community Media, the Alliance for Communications Democracy, and Common Cause; Cox Communications, Inc.; DISH Network Corporation; The Greenlining Institute; Free Press; Public Knowledge and Institute for Local Self-Reliance; and Writers Guild of America, West, Inc. 29 See Letter to Robert W. Quinn, Jr., Senior Vice President – Federal Regulatory and Chief Privacy Officer, AT&T, from William T. Lake, Chief, Media Bureau, FCC, MB Docket No. 14-90, 2014 WL 4460323 (Sept. 9, 2014) (“Sept. 9, 2014, Information Request to AT&T”); Letter to Stacy Fuller, Vice President, Regulatory Affairs, DIRECTV, from William T. Lake, Chief, Media Bureau, MB Docket No. 14-90, 2014 WL 4460324 (Sept. 9, 2014) (“Sept. 9, 2014, Information Request to DIRECTV”). Additional information was also sought from the Applicants later in the Commission’s review. See Letter to Robert W. Quinn, Jr., Senior Vice President – Federal Regulatory and Chief Privacy Officer, AT&T, from Jamillia Ferris, Office of the General Counsel, FCC, MB Docket No. 14-90, 2014 WL 6070716 (Nov. 14, 2014) (“Nov. 14, 2014, Information Request to AT&T”); Letter to Robert W. Quinn, Jr., Senior Vice President – Federal Regulatory and Chief Privacy Officer, AT&T, from Jamillia Ferris, Office of the General Counsel, FCC, MB Docket No. 14-90, 2014 WL 7172163 (Dec. 15, 2014) (“Dec. 15, 2014, Information Request to AT&T”). 30 See Letter to Catherine Bohigian, Executive Vice President – Government Affairs, Charter Communications, Inc., from William T. Lake, Chief, Media Bureau, FCC, MB Docket No. 14-90, 2015 WL 128688 (Jan. 8, 2015); Letter to Steven Teplitz, Senior Vice President – Government Relations, Time Warner Cable Inc., from William T. Lake, Chief, Media Bureau, FCC, MB Docket No. 14-90, 2015 WL 128689 (Jan. 8, 2015); Letter to Kathryn Zachem, Senior Vice President – Regulatory and State Legislative Affairs, Comcast Corporation, from William T. Lake, Chief, Media Bureau, FCC, MB Docket No. 14-90, 2015 WL 223239 (Jan. 8, 2015). 31 See AT&T Inc. Response to Sept. 9, 2014, Information and Discovery Requests, transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Vanessa Lemmé, Media Bureau, FCC, MB Docket No. 14-90 (Oct. 7, (continued….) Federal Communications Commission FCC 15-94 9 17. In addition to Commission review, the proposed transaction is subject to review by the United States Department of Justice (“DOJ”) pursuant to its concurrent authority in Section 7 of the Clayton Act.33 IV. STANDARD OF REVIEW AND PUBLIC INTEREST FRAMEWORK 18. Pursuant to Section 310(d) of the Act, we must determine whether the Applicants have demonstrated that the proposed transfer of control of licenses and authorizations will serve the public interest, convenience, and necessity.34 In making this determination, we assess whether the proposed transaction complies with the specific provisions of the Act,35 other applicable statutes, and the (Continued from previous page) 2014) (“AT&T Response to Sept. 9, 2014, Information Request”); AT&T Updated Response to Sept. 9, 2014, Information Request; DIRECTV Response to Sept. 9, 2014, Information and Discovery Requests, transmitted by letter from William M. Wiltshire, Counsel for DIRECTV, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (Oct. 7, 2014) (“DIRECTV Response to Sept. 9, 2014, Information Request”); AT&T Inc. Response to Nov. 14, 2014, Information and Discovery Requests, transmitted by letter Robert W. Quinn, Jr., Senior Vice President – Federal Regulatory and Chief Privacy Officer, AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (Nov. 25, 2014) (“AT&T Response to Nov. 14, 2014, Information Request”); AT&T Inc. Response to Sept. 9, 2014, and Dec. 15, 2014, Information and Discovery Requests, transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (Dec. 19, 2014) (“AT&T Response to Dec. 15, 2014, Information Request”); Comcast Corporation Response to Jan. 8, 2015, Information and Data Request, transmitted by letter from Kathyrn A. Zachem, Senior Vice President – Regulatory and State Legislative Affairs, Comcast Corporation, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (Jan. 23, 2015) (“Comcast Response to Jan. 8, 2015, Information Request”); Time Warner Cable Inc. Response to Jan. 8, 2015, Information and Data Request, transmitted by letter from Matthew A. Brill, Counsel for Time Warner Cable, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (Jan. 23, 2015) (“Time Warner Cable Response to Jan. 8, 2015, Information Request”); Charter Communications, Inc. Response to Jan. 8, 2015, Information and Data Request, transmitted by letter from John L. Flynn, Counsel for Charter, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (Jan. 20, 2015) (“Charter Response to Jan. 8, 2015, Information Request”). 32 The Media Bureau adopted a Protective Order to (i) limit access to proprietary or confidential information filed in this proceeding and (ii) more strictly limit access to certain particularly competitively sensitive information. See Applications of AT&T Inc. and DIRECTV to Transfer Control of FCC Licenses, MB Docket No. 14-90, Joint Protective Order, DA 14-804, 29 FCC Rcd 6047 (MB 2014), modified by DA 14-1465, 29 FCC Rcd 11883 (MB 2014), amended by DA 14-1602, 29 FCC Rcd 13616 (MB 2014), amended by DA 14-1640, 29 FCC Rcd 13810 (MB 2014) (“Protective Order”). In this Order, Highly Confidential Information, as defined in the Protective Order, will be marked by the terms “[BEGIN HIGHLY CONF. INFO.]” and “[END HIGHLY CONF. INFO.],” or “[BEGIN VIDEO PROG. CONF. INFO.]” and “[END VIDEO PROG. CONF. INFO.]” as appropriate. In this Order, Confidential Information, as defined in the Protective Order, will be marked by the terms “[BEGIN CONF. INFO.]” and “[END CONF. INFO.]” as appropriate. Such information will be redacted from the publicly available version of the Order. The unredacted information will be available upon request to persons qualified to view it under the Protective Order. 33 15 U.S.C. § 18. 34 47 U.S.C. § 310(d); 47 C.F.R. § 25.119. 35 Section 310(d) requires that we consider applications as if the proposed transferee were applying for the licenses directly. 47 U.S.C. § 310(d). See Applications of Comcast Corporation, General Electric Company, and NBC Universal, Inc. for Consent to Assign Licenses and Transfer Control of Licensees, MB Docket No. 10-56, Memorandum Opinion and Order, 26 FCC Rcd 4238, 4247, ¶ 22 n.42 (2011) (“Comcast-NBCU Order”); Applications for Consent to the Transfer of Control of Licenses, XM Satellite Radio Holdings Inc., Transferor, to Sirius Satellite Radio Inc., Transferee, MB Docket No. 07-57, Memorandum Opinion and Order and Report and Order, 23 FCC Rcd 12348, 12363, ¶ 30 n.114 (2008) (“Sirius-XM Order”); News Corp. and DIRECTV Group, Inc. and Liberty Media Corp. for Authority to Transfer Control, Memorandum Opinion and Order, MB Docket No. 07- 18, 23 FCC Rcd 3265, 3276, ¶ 22 n.72 (2008) (“Liberty Media-DIRECTV Order”); Application of EchoStar Communications Corporation, General Motors Corporation, and Hughes Electronics Corporation (Transferors) (continued….) Federal Communications Commission FCC 15-94 10 Commission’s rules.36 If the transaction does not violate a statute or rule, we consider whether the transaction could result in public interest harms by substantially frustrating or impairing the objectives or implementation of the Act or related statutes.37 We then employ a balancing test weighing any potential public interest harms of the proposed transaction against any potential public interest benefits.38 The Applicants bear the burden of proving, by a preponderance of the evidence, that the proposed transaction, on balance, serves the public interest.39 If we are unable to find that the proposed transaction serves the public interest for any reason, or if the record presents a substantial and material question of fact, we must designate the Application for hearing.40 19. Our public interest evaluation necessarily encompasses the “broad aims of the Communications Act,” which include, among other things, a deeply rooted preference for preserving and enhancing competition, accelerating private sector deployment of advanced services, promoting a diversity of information sources and services to the public, and generally managing the spectrum in the public interest.41 Our public interest analysis also entails assessing whether the proposed transaction would affect the quality of communications services or result in the provision of new or additional services to consumers.42 In conducting this analysis, we may consider technological and market changes, and the nature, complexity, and speed of change of, as well as trends within, the communications industry.43 20. Our competitive analysis, which forms an important part of the public interest evaluation, is informed by, but not limited to, traditional antitrust principles.44 The Commission and the DOJ each has independent authority to examine the competitive impacts of proposed communications mergers and transactions involving transfers of Commission licenses, but the standards governing the Commission’s competitive review differ somewhat from those applied by the DOJ.45 The Commission, like the DOJ, considers how a transaction would affect competition by defining a relevant market, looking at the market (Continued from previous page) and EchoStar Communications Corporation (Transferee), MB Docket No. 01-348, Hearing Designation Order, 17 FCC Rcd 20559, 20574, ¶ 25 n.102 (2002) (“EchoStar-DIRECTV HDO”). 36 See Comcast-NBCU Order, 26 FCC Rcd at 4247, ¶ 22; Sirius-XM Order, 23 FCC Rcd at 12363-64, ¶ 30; Liberty Media-DIRECTV Order, 23 FCC Rcd at 3276-77, ¶ 22; EchoStar-DIRECTV HDO, 17 FCC Rcd at 20574, ¶ 25. 37 See id. 38 See id.; General Motors Corp. and Hughes Electronics Corp., Transferors, and the News Corporation, Transferee, MB Docket No. 03-124, Memorandum Opinion and Order, 19 FCC Rcd 473, 483, ¶ 15 (2004) (“News Corp.-Hughes Order”). 39 See id. 40 See 47 U.S.C. § 309(e); see also Comcast-NBCU Order, 26 FCC Rcd at 4247-48, ¶ 22; Sirius-XM Order, 23 FCC Rcd at 12364, ¶ 30; Liberty Media-DIRECTV Order, 23 FCC Rcd at 3277, ¶ 22; News Corp.-Hughes Order, 19 FCC Rcd at 483, ¶ 15 n.49; EchoStar-DIRECTV HDO, 17 FCC Rcd at 20574, ¶ 25. 41 See Comcast-NBCU Order, 26 FCC Rcd at 4248, ¶ 23; Sirius-XM Order, 23 FCC Rcd at 12364, ¶ 31; Liberty Media-DIRECTV Order, 23 FCC Rcd at 3277-78, ¶ 23; News Corp.-Hughes Order, 19 FCC Rcd at 483-84, ¶ 16; EchoStar-DIRECTV HDO, 17 FCC Rcd at 20575, ¶ 26. 42 See Comcast-NBCU Order, 26 FCC Rcd at 4248, ¶ 23; Sirius-XM Order, 23 FCC Rcd at 12365, ¶ 31; Liberty Media-DIRECTV Order, 23 FCC Rcd at 3277-78, ¶ 23; EchoStar-DIRECTV HDO, 17 FCC Rcd at 20575, ¶ 26. 43 See id. 44 See Comcast-NBCU Order, 26 FCC Rcd at 4248, ¶ 24; Sirius-XM Order, 23 FCC Rcd at 12365, ¶ 32; Liberty Media-DIRECTV Order, 23 FCC Rcd at 3278, ¶ 24; News Corp.-Hughes Order, 19 FCC Rcd at 484, ¶ 17; EchoStar-DIRECTV HDO, 17 FCC Rcd at 20575, ¶ 27. 45 See, e.g., id. Federal Communications Commission FCC 15-94 11 power of incumbent competitors, and analyzing barriers to entry, potential competition, and the efficiencies, if any, that may result from the transaction.46 21. The DOJ, however, reviews telecommunications mergers pursuant to Section 7 of the Clayton Act, and if it sues to enjoin a merger, it must demonstrate to a court that the merger may substantially lessen competition or tend to create a monopoly.47 The DOJ review is consequently limited solely to an examination of the competitive effects of the acquisition, without reference to diversity, localism, or other public interest considerations.48 Moreover, the Commission’s competitive analysis under the public interest standard is broader. For example, the Commission considers whether a transaction would enhance, rather than merely preserve, existing competition, and often takes a more expansive view of potential and future competition in analyzing that issue.49 22. Finally, our public interest authority enables us, where appropriate, to impose and enforce transaction-related conditions that ensure that the public interest is served by the transaction.50 Specifically, Section 303(r) of the Communications Act authorizes the Commission to prescribe restrictions or conditions not inconsistent with law that may be necessary to carry out the provisions of the Act.51 Indeed, our extensive regulatory and enforcement experience enables us, under this public interest authority, to impose and enforce conditions to ensure that the transaction will yield overall public interest benefits.52 In exercising this authority to carry out our responsibilities under the Act and related statutes, we have imposed conditions to confirm specific benefits or remedy specific harms likely to arise from transactions.53 46 See Sirius-XM Order, 23 FCC Rcd at 12365, ¶ 32; see also Applications of Sprint Nextel Corp. and SoftBank Corp. and Starburst II, Inc. for Consent to Transfer Control of Licenses and Authorizations, IB Docket No. 12-343, Memorandum Opinion and Order, Declaratory Ruling, and Order on Reconsideration, 28 FCC Rcd 9642, 9652, ¶ 25 (2013) (“SoftBank-Sprint Order”). 47 15 U.S.C. § 18; see also Comcast-NBCU Order, 26 FCC Rcd at 4248, ¶ 24; Sirius-XM Order, 23 FCC Rcd at 12365, ¶ 32; Liberty Media-DIRECTV Order, 23 FCC Rcd at 3278, ¶ 24; News Corp.-Hughes Order, 19 FCC Rcd at 484, ¶ 17; EchoStar-DIRECTV HDO, 17 FCC Rcd at 20575, ¶ 27. 48 See SoftBank-Sprint Order, 28 FCC Rcd at 9652, ¶ 25; Sirius-XM Order, 23 FCC Rcd at 12365, ¶ 32; Liberty Media-DIRECTV Order, 23 FCC Rcd at 3278, ¶ 24. 49 See Comcast-NBCU Order, 26 FCC Rcd at 4248, ¶ 24; Sirius-XM Order, 23 FCC Rcd at 12365-66, ¶ 32; see also Liberty Media-DIRECTV Order, 23 FCC Rcd at 3278-79, ¶ 25; EchoStar-DIRECTV HDO, 17 FCC Rcd at 20575, ¶ 27. 50 See Comcast-NBCU Order, 26 FCC Rcd at 4249, ¶ 25; Sirius-XM Order, 23 FCC Rcd at 12366, ¶ 33; Liberty Media-DIRECTV Order, 23 FCC Rcd at 3279, ¶ 26; see also Application of WorldCom, Inc. and MCI Communications Corporation for Transfer of Control of MCI Communications Corporation to WorldCom, Inc., CC Docket No. 97-211, Memorandum Opinion and Order, 13 FCC Rcd 18025, 18032, ¶ 10 (1998) (“WorldCom-MCI Order”) (stating that the Commission may attach conditions to the transfers). 51 47 U.S.C. § 303(r). See Comcast-NBCU Order, 26 FCC Rcd at 4249, ¶ 25; Sirius-XM Order, 23 FCC Rcd at 12366, ¶ 33; Liberty Media-DIRECTV Order, 23 FCC Rcd at 3279, ¶ 26; WorldCom-MCI Order, 13 FCC Rcd at 18032, ¶ 10 (citing FCC v. Nat’l Citizens Comm. for Broad., 436 U.S. 775 (1978) (upholding broadcast-newspaper cross-ownership rules adopted pursuant to Section 303(r))); United States v. Southwestern Cable Co., 392 U.S. 157, 178 (1968) (holding that Section 303(r) permits the Commission to order a cable company not to carry broadcast signal beyond station’s primary market); United Video, Inc. v. FCC, 890 F.2d 1173, 1182-83 (D.C. Cir. 1989) (affirming syndicated exclusivity rules adopted pursuant to Section 303(r) authority). 52 See Comcast-NBCU Order, 26 FCC Rcd at 4249, ¶ 25; Sirius-XM Order, 23 FCC Rcd at 12366, ¶ 33; Liberty Media-DIRECTV Order, 23 FCC Rcd at 3279, ¶ 26. 53 See Comcast-NBCU Order, 26 FCC Rcd at 4249, ¶ 25; Liberty Media-DIRECTV Order, 23 FCC Rcd at 3279, ¶ 26. Federal Communications Commission FCC 15-94 12 23. This Order examines the proposed transaction as follows. First, we examine whether the transaction complies with the Act, other applicable statutes, and the Commission’s rules and policies, and we assess the qualifications of the Applicants. Second, we consider the potential harms and purported public interest benefits resulting from the transaction. Then, we consider and, where appropriate, impose conditions to ameliorate the harms or confirm the benefits. Finally, we balance the public interest harms posed by and the benefits to be gained from the transaction. V. QUALIFICATIONS OF APPLICANTS A. Background 24. Section 310(d) of the Act requires that we make a determination as to whether the Applicants have the requisite qualifications to hold Commission licenses.54 Among the factors the Commission considers in its public interest review is whether the applicant for a license has the requisite “citizenship, character, and financial, technical, and other qualifications.”55 As a threshold matter, the Commission must determine whether the Applicants to the proposed transaction – both the transferee and the transferor – meet the requisite qualifications and requirements to hold and transfer licenses under Section 310(d) and the Commission’s rules.56 With respect to Commission-related conduct, the Commission has stated that all violations of the Act, or of the Commission’s rules or policies, are predictive of an applicant’s future truthfulness and reliability and thus have a bearing on an applicant’s character qualifications.57 The Commission has previously determined that in its review of character issues, it will also consider certain types of adjudicated, non-Commission-related misconduct, specifically: (1) felony convictions; (2) fraudulent misrepresentation to governmental units; and (3) violations of antitrust or other laws protecting competition.58 For the reasons discussed below, we find that the Applicants have the requisite character qualifications to hold Commission licenses. B. DIRECTV 25. No parties have raised issues with respect to the basic qualifications of the transferor, DIRECTV. The Commission generally does not reevaluate the qualifications of transferors unless issues related to basic qualifications have been sufficiently raised in petitions to warrant designation for hearing.59 We find that there is no reason to reevaluate the requisite citizenship, character, financial, technical, or other basic qualifications under the Communications Act and our rules, regulations, and policies, of DIRECTV. 54 47 U.S.C. § 310(d). 55 47 U.S.C. §§ 308, 310(d); see Comcast-NBCU Order, 26 FCC Rcd at 4349, ¶ 276; News Corp.-Hughes Order, 19 FCC Rcd at 485, ¶ 18; EchoStar-DIRECTV HDO, 17 FCC Rcd at 20576, ¶ 28. 56 See 47 U.S.C. § 310(d); see also, e.g., Comcast-NBCU Order, 26 FCC Rcd at 4349, ¶ 276; News Corp.-Hughes Order, 19 FCC Rcd at 485, ¶ 18; EchoStar-DIRECTV HDO, 17 FCC Rcd at 20576, ¶ 28. 57 See Applications of Cellco Partnership d/b/a Verizon Wireless and Atlantis Holdings LLC for Consent to Transfer Control of Licenses, Authorizations, and Spectrum Manager and de facto Transfer Leasing Arrangements, WT Docket No. 08-95, Memorandum Opinion and Order and Declaratory Ruling, 23 FCC Rcd 17444, 17464, ¶ 32 (2008) (citations omitted) (“Verizon Wireless-ALLTEL Order”). 58 See Applications of AT&T Wireless Services, Inc. and Cingular Wireless Corporation for Consent To Transfer Control of Licenses and Authorizations, Memorandum Opinion and Order, 19 FCC Rcd 21522, 21548, ¶ 47 (2004) (“AT&T-Cingular Order”). 59 See, e.g., SoftBank-Sprint Order, 28 FCC Rcd at 9653, ¶ 27. Federal Communications Commission FCC 15-94 13 C. AT&T 26. Two parties, Minority Cellular Partners Coalition (“MCPC”) and New Networks Institute & Teletruth (“New Networks”), have raised issues with respect to the basic qualifications of the transferee, AT&T.60 These issues are discussed more fully below. 1. Minority Cellular Partners Coalition Comments 27. The Commission has determined on numerous occasions that AT&T was qualified to acquire Commission licenses.61 Here, MCPC62 questions those qualifications. MCPC first alleges that AT&T defrauded its minority partners and engaged in serious Commission-related misconduct, including forcing out minority partners in certain cellular license partnerships, and that AT&T engaged in related violations of the Commission’s license transfer rules.63 It states that the Commission, through an evidentiary hearing or otherwise, should elicit the facts necessary to resolve the question of whether AT&T intentionally violated Section 310(d) of the Act and Sections 1.17 and 1.948 of the Rules.64 In 60 On May 4, 2015, Erik Underwood, Founder and CEO of My24HourNews.Com, Inc., belatedly informed the Commission that he opposes the transaction and requested that the Commission suspend its review until the completion of civil litigation recently initiated by My24HourNews.Com concerning a contract dispute between My24HourNews.Com and AT&T. Letter from Erik M. Underwood, Founder/CEO My24HourNews.Com, Inc., to Thomas Wheeler, Chairman, FCC, MB Docket No. 14-90 (May 4, 2015) (“Underwood Letter”); see also Email from Erik M. Underwood to Thomas Wheeler, MB Docket No. 14-90 (June 11, 2015) (attaching Complaint and Civil Cover Sheet in My24HourNews.Com, Inc. v. AT&T Corp., No. 1:15-cv-01210 (D. Colo. filed June 9, 2015)). Although Underwood asserts that “AT&T has violated anti-competition and telecommunication laws of this country,” Underwood Letter at 1, he does not identify any specific Commission rule or policy or any provision of the Communications Act that he believes is at issue. Underwood also claims that he attached confidential emails to his letter as evidence of AT&T’s anticompetitive conduct. We have been unable to ascertain that we received the emails and note that Underwood did not file them according to the procedures specified for the filing of confidential information. See Protective Order, 29 FCC Rcd at 13816, ¶ 14. The pending litigation at issue concerns claims for breach of contract, misappropriation of intellectual property, and other alleged wrongdoing arising from a pre- existing dispute not involving the Commission’s rules or the Communications Act. Underwood does not claim that his request to suspend Commission review of the Application is necessary to address a transaction-specific harm or benefit or that AT&T lacks the necessary character qualifications to hold Commission licenses. Moreover, except in limited circumstances that do not apply here, the Commission does not consider unadjudicated claims of non- Commission misconduct. Policy Regarding Character Qualifications in Broadcast Licensing Amendment of Rules of Broadcast Practice and Procedure Relating to Written Responses to Commission Inquiries and the Making of Misrepresentations to the Commission by Permittees & Licensees, Gen. Docket No. 81-500, Report, Order and Policy Statement, 102 FCC 2d 1179, 1204-06, ¶ 48 (1986), recon. granted in part, denied in part, 1 FCC Rcd 421 (1986), appeal dismissed sub nom. Nat’l Ass’n for Better Broad. v. FCC, No. 86–1179 (D.C. Cir. June 11, 1987) (“1986 Character Policy Statement”). Thus, we deny the request. 61 See, e.g., Applications of AT&T Mobility Spectrum LLC, New Cingular Wireless PCS, LLC, Comcast Corporation, Horizon Wi-Com, LLC, NextWave Wireless, Inc., and San Diego Gas & Electric Company for Consent to Assign and Transfer Licenses, WT Docket No. 12-240, Memorandum Opinion and Order, 27 FCC Rcd 16459, 16466-67, ¶ 19 (2012) (“AT&T-WCS Order”); see also Applications of Cricket License Company, LLC, Leap Wireless International, Inc., and AT&T Inc. for Consent to Transfer Control and Assignment of Authorizations, WT Docket No. 13-193, Memorandum Opinion and Order, DA 14-349, 29 FCC Rcd 2735, 2745, ¶ 19 (WTB, IB 2014) (“AT&T-Leap Wireless Order”); AT&T Inc., Cellco Partnership d/b/a Verizon Wireless, Grain Spectrum, LLC, and Grain Spectrum II, LLC, WT Docket No. 13-56, Memorandum Opinion and Order, 28 FCC Rcd 12878, 12885, ¶ 17 (2013). 62 MCPC explains that its members were minority partners in partnerships with AT&T that held 11 cellular licenses. Comments of Minority Cellular Partners Coalition, MB Docket 14-90, at 1 (filed Sept. 16, 2014) (“MCPC Comments”). 63 See id. at 9-12, 15-17. 64 See id. at 21. Federal Communications Commission FCC 15-94 14 particular, MCPC maintains that AT&T divested MCPC’s members of their interests in cellular licenses held in partnership with AT&T by mischaracterizing certain transactions as pro forma license assignments that require only notification to the Commission, rather than transfers of control that require prior Commission approval.65 MCPC notes that, while the Commission’s findings may or may not be disqualifying, they may work to ensure AT&T’s future compliance with Section 310(d) and be of substantial aid to the Chancery Court in Delaware, where MCPC’s claims are currently being litigated.66 28. The Applicants respond that the claims raised by MCPC relate to matters of Delaware Corporate and Partnership Law that are unrelated to the current proceeding.67 They note that MCPC’s claims are already being heard in Delaware Chancery Court, which is the appropriate forum.68 Further, they state that MCPC should have raised the allegations about transfer of control years ago.69 Finally, AT&T contends it properly filed the pro forma notifications challenged by MCPC in full compliance with the requirements of Section 310(d).70 29. MCPC also notes that the Commission issued a notice of apparent liability (“NAL”) in January 2015, finding that AT&T appeared to have operated a number of wireless stations at variance with their licensed parameters.71 While MCPC observes that the unadjudicated NAL it cites cannot be used against AT&T in this proceeding,72 it recommends that the Commission examine the facts underlying the NAL to determine whether AT&T is engaging in a pattern of noncompliant behavior.73 AT&T responds that the Commission has not yet determined what sanction, if any, is appropriate in that enforcement proceeding.74 AT&T argues further that the Commission has concluded in the 1986 Character Policy Statement that an unadjudicated NAL is an inappropriate ground for a finding of unfitness.75 30. Finally, MCPC claims that AT&T violated the Communications Assistance for Law Enforcement Act (“CALEA”)76 and the Commission’s CALEA rules77 by participating in the President’s 65 Id. at 15-17. 66 Id. at 21. 67 See Joint Opposition at 73. 68 See id. 69 See id. at 74. 70 See id. at 74-75. 71 Letter from Russell D. Lukas, Counsel for MCPC, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 1-2 (March 4, 2015) (“MCPC March 4, 2015, Ex Parte Letter”) (citing AT&T Inc., Parent Company of New Cingular Wireless PCS, LLC and AT&T Mobility Puerto Rico, Inc., Notice of Apparent Liability for Forfeiture, 30 FCC Rcd 856 (2015) (“AT&T Mobility Puerto Rico NAL”)). 72 MCPC March 4, 2015, Ex Parte Letter at 2 (citing 47 U.S.C. § 504(c)). 73 Id. (citing The Commission’s Forfeiture Policy Statement and Amendment of Section 1.80 of the Rules to Incorporate the Forfeiture Guidelines, CI Docket No. 95-6, Memorandum Opinion and Order, 15 FCC Rcd 303, 304, ¶¶ 3-4 (1999) (“1999 Forfeiture Guidelines”); Infinity Radio Operations, Inc., Order on Review, 22 FCC Rcd 9824, 9827, ¶ 9 (2007) (“Infinity Forfeiture Review Order”). 74 Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14- 90, at 4-5 (March 11, 2015) (“AT&T March 11, 2015, Ex Parte Letter”). 75 Id. at 5 (citing 1986 Character Policy Statement, 102 FCC 2d at 1204-06, ¶ 48). 76 Section 103(a)(1) of CALEA, 47 U.S.C. § 1002(a)(1), requires telecommunications carriers to establish the capability of providing to law enforcement agencies (“LEAs”) call content information, pursuant to a court order or other lawful authorization. Section 103(a)(2) of CALEA, 47 U.S.C. § 1002(a)(2), requires telecommunications carriers to establish the capability of providing to LEAs reasonably available call-identifying information (“CII”), pursuant to a court order or other lawful authorization. Section 105 of CALEA, 47 U.S.C. § 1004, requires (continued….) Federal Communications Commission FCC 15-94 15 Surveillance Program (“PSP”).78 According to MCPC, under this program, AT&T permitted the National Security Agency (“NSA”) to intercept communications or to have access to call-identifying information without the lawful authorization required by CALEA and the CALEA rules for the 33-month period ending on July 15, 2004.79 While MCPC recognizes that AT&T has been granted immunity from civil damages claims arising from its participation, if any, in the PSP,80 and that the statute of limitations for any forfeiture penalty under Section 503(b)(6)(B) of the Communications Act has passed,81 MCPC nevertheless advocates that the Commission investigate this matter in the context of this transaction proceeding for purposes of evaluating AT&T’s qualifications to hold the licenses currently held by DIRECTV.82 MCPC also argues that an investigation into AT&T’s involvement with the PSP is warranted because such an investigation would help restore the public’s confidence in the privacy of their communications.83 In addition, MCPC recommends conditioning this transaction on AT&T submitting its systems security and integrity (“SSI”) plan to the Commission, subject to notice and comment.84 MCPC asserts that the public’s confidence in the privacy of individuals’ communications has been “shaken” by AT&T’s participation in the PSP and that a “modicum” of that confidence could be restored if the Commission were to impose the condition MCPC recommends.85 a. Standing 31. As an initial matter, although MCPC frames its comments as raising the question of whether AT&T has the requisite character qualifications to hold Commission licenses, MCPC’s participation in this proceeding appears to be motivated by its ongoing business dispute with AT&T, which is wholly unrelated to the transaction. To establish party-in-interest standing to challenge an application, a petitioner must allege facts sufficient to demonstrate that grant of the application would (Continued from previous page) telecommunications carriers to ensure that “any interception of communications or access to call-identifying information effected within its switching premises can be activated only in accordance with a court order or other lawful authorization.” 77 See 47 C.F.R. §§ 1.20000-08 (“CALEA rules”). 78 MCPC March 4, 2015, Ex Parte Letter at 5. 79 Id. at 8-11. For background on the PSP, see In re NSA Telecomm. Records Litig., 633 F. Supp. 2d 949, 955-957 (N.D. Cal. 2009). 80 MCPC March 4, 2015, Ex Parte Letter at 11. See also 50 U.S.C. § 1885a (statutory provision in which Congress granted telecommunications carriers immunity from civil suits to the extent they participated in the PSP under certain circumstances); In re NSA Telecomm. Records Litig., 671 F.3d 881 (9th Cir. 2011) (finding this grant of immunity to be Constitutional). 81 MCPC March 4, 2015, Ex Parte Letter at 11 (citing 47 U.S.C. § 503(b)(6)(B)). 82 Id. In response, AT&T points out that substantially similar allegations were raised in a 2006 transfer of control proceeding. AT&T notes that the Commission did not investigate these issues in the context of the 2006 proceeding, finding that these issues were outside the scope of its investigative powers. AT&T argues that the Commission should follow that precedent here. AT&T March 11, 2015, Ex Parte Letter at 2-4 (citing AT&T Inc. and BellSouth Corporation, Application for Transfer of Control, WC Docket No. 06-74, Memorandum Opinion and Order, 22 FCC Rcd 5662, 5757, ¶ 192 (2007) (“AT&T-BellSouth Merger Order”)). MCPC responded that the factors underlying the Commission’s decision in the AT&T-BellSouth Merger Order not to investigate these issues do not apply in this proceeding. Letter from Russell D. Lukas, Counsel for MCPC, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 1 (March 19, 2015) (“MCPC March 19, 2015, Ex Parte Letter”). We need not address this issue here because we find for other reasons that this issue does not warrant investigation in this transaction review proceeding. 83 MCPC March 19, 2015, Ex Parte Letter at 2. 84 MCPC March 4, 2015, Ex Parte Letter at 12; MCPC March 19, 2015, Ex Parte Letter at 2. 85 MCPC March 19, 2015, Ex Parte Letter at 2. Federal Communications Commission FCC 15-94 16 cause it to suffer a direct injury.86 In addition, petitioners must demonstrate a causal link between the claimed injury and the challenged action.87 To demonstrate a causal link, petitioners must establish that the injury can be traced to the challenged action and that the injury would be prevented or redressed by the relief requested.88 MCPC has not articulated any theory by which the Commission’s disposition of the Application would redress an injury to MCPC. Moreover, MCPC does not allege that its members are competitors or viewers of AT&T’s or DIRECTV’s programming.89 In other words, MCPC does not allege that its members currently compete with AT&T in the video programming or video distribution market, or indeed, in any market. Accordingly, we conclude that MCPC’s dispute with AT&T does not give it standing to object to the transfer of control of DIRECTV to AT&T.90 b. Pro Forma Transactions 32. At the outset, we note that MCPC’s allegations regarding violations under Delaware law are being adjudicated by the Chancery Court in Delaware.91 They do not involve alleged violations of the Communications Act or Commission rules, and there has been no adjudicated finding of wrongdoing. Thus, they are outside the scope of our character qualifications inquiry.92 33. Furthermore, we note that MCPC has not offered any evidence to support its allegations. Section 309(d)(1) of the Communications Act requires parties filing petitions to deny applications to support their allegations of fact with an affidavit of a person or persons with personal knowledge 86 See, e.g., AT&T-WCS Order, 27 FCC Rcd at 16465, ¶ 16; Touchtel Corporation, Assignor, Penryn Corporation, Assignee, Order on Reconsideration, 29 FCC Rcd 16249, 16250-51, ¶ 7 (2014) (“Touchtel Order”). See also AT&T- Cingular Order, 19 FCC Rcd at 21547, ¶ 46 n.196 (the Commission had “doubts” regarding petitioner’s standing when there was no demonstration that it would be directly affected by the order); Applications of Nextel Communications, Inc. and Sprint Corporation For Consent to Transfer Control of Licenses and Authorizations, WT Docket No. 05-63, Memorandum Opinion and Order, 20 FCC Rcd 13967, 14021, ¶ 150 n.335 (2005) (“Sprint- Nextel Order”) (same). 87 See Touchtel Order, 29 FCC Rcd at 16250-51, ¶ 7 (and sources cited therein). 88 See id. (and sources cited therein). 89 Sunburst Media-Louisiana, LLC, Memorandum Opinion and Order, 29 FCC Rcd 9777, 9778, ¶ 5 (2014) (Generally, to establish standing in the broadcast regulatory context, a petitioner must show that it is: (1) a competitor in the market suffering signal interference; (2) a competitor in the market suffering economic harm; or (3) a resident of the station’s service area or a regular listener of the station.). 90 We recognize that an informal objection may be filed pursuant to Section 1.41 of the Commission’s rules without demonstrating standing. 47 C.F.R. § 1.41. The Commission has discretion whether to consider an informal objection. AT&T-Cingular Order, 19 FCC Rcd at 21547, ¶ 46 n.196; Sprint-Nextel Order, 20 FCC Rcd at 14021, ¶ 150 n.335; Touchtel Order, 29 FCC Rcd at 16251, ¶ 8. In this case, we find that the public interest warrants considering MCPC’s contentions as informal objections. However, for the reasons discussed below, we conclude that they do not present a substantial and material question of fact warranting further inquiry into AT&T’s character qualifications. For these reasons, we also find that MCPC’s proposed condition is unnecessary. 91 The matters being considered in the Chancery Court in Delaware address private contractual disputes, for which the Commission had repeatedly stated it is not the appropriate forum for resolution. See, e.g., AT&T-Cingular Order, 19 FCC Rcd at 21551, ¶ 56 n.222 (rejecting argument that transfer should be denied on grounds that it violated partnership agreements; “these are private contractual disputes that are not relevant to our public interest analysis and are best resolved in courts of competent jurisdiction”). 92 1986 Character Policy Statement, 102 FCC 2d at 1204-06, ¶ 48 (1986); Policy Regarding Character Qualifications in Broadcast Licensing, Amendment of Part 1, the Rules of Practice and Procedure, Relating to Written Responses to Commission Inquiries and the Making of Misrepresentations to the Commission by Applicants, Permittees and Licensees, and the Reporting of Information Regarding Character Qualifications, Memorandum Opinion and Order, 7 FCC Rcd 6564, 6566, ¶ 9 (1992) (pending litigation involving alleged non-Commission misconduct is presumptively not relevant to a licensee’s character qualifications). Federal Communications Commission FCC 15-94 17 thereof.93 MCPC did not submit any such affidavit. Accordingly, even if MCPC did have standing to raise its allegations, MCPC has not satisfied the evidentiary threshold of Section 309(d)(1) to show that grant of the Application would be prima facie inconsistent with the public interest so as to warrant further inquiry. 34. As a separate and independent basis for rejecting MCPC’s allegations, we find that the allegations are unpersuasive. MCPC claims that AT&T mischaracterized and possibly misrepresented substantive transfer of control transactions as pro forma transactions that do not require prior Commission approval, thereby violating a Commission rule.94 Specifically, as mentioned above, MCPC maintains that the 11 transactions at issue divested MCPC’s members of their interests in cellular licenses held in partnership with AT&T and that AT&T mischaracterized those transactions as pro forma.95 In addition, MCPC observes that, in 1998, the Commission adopted the Section 310(d) Forbearance Order, identifying six kinds of transactions as warranting pro forma treatment, and contends that the 11 transactions at issue here do not fit into any of those six categories.96 35. MCPC is mistaken in asserting that the 11 transactions it cites were not pro forma transactions. The first type of transaction identified in the Section 310(d) Forbearance Order as warranting pro forma treatment is an “assignment from an individual or individuals (including partnerships) to a corporation owned or controlled by such individuals or partnerships without any substantial change in their relative interests.”97 In each of the 11 transactions at issue here, AT&T assigned a license from a partnership to a corporation. These transactions did not result in any substantial change in ownership because AT&T had de jure and de facto control of each assigning partnership prior to the assignment and it held such control after the assignment.98 Therefore, the 11 transactions fit perfectly within the first category of pro forma transactions identified in the Section 310(d) Forbearance Order.99 93 47 U.S.C. § 309(d)(1). 94 MCPC Comments at 16. See also 47 C.F.R. § 1.948(c)(1) (exempting pro forma transactions from the requirement that licensees obtain approval from the Commission prior to transferring or assigning their licenses). 95 MCPC Comments at 15-17. 96 Id. at 15-16 (citing FCBA’s Petition for Forbearance from Section 310(d) of the Communications Act, Memorandum Opinion and Order, 13 FCC Rcd 6293 (1998) (“Section 310(d) Forbearance Order”)). In that order, the Commission decided to forbear from enforcing the requirement in Section 310(d) of the Communications Act that parties obtain prior approval for transfers of licenses, provided that the license was issued by the Wireless Telecommunications Bureau, the licensee notified the Bureau prior to completing the transaction, and the transaction fell into one of the six categories listed in that order. Section 310(d) Forbearance Order, 13 FCC Rcd at 6299, ¶ 9. 97 Section 310(d) Forbearance Order, 13 FCC Rcd at 6298-99, ¶ 8 (quoted in MCPC Comments at 15). 98 See Joint Opposition at 74. See also Section 310(d) Forbearance Order, 13 FCC Rcd at 6297, ¶ 7 (noting that a pro forma transaction is one in which there is no substantial change in de jure or de facto control). 99 In its reply, MCPC also states that, “[b]y the time the public was notified of AT&T’s actions, the issue of AT&T’s compliance with § 1.948(c) of the Rules was largely moot.” Reply Comments of Minority Cellular Partners Coalition, MB Docket 14-90, at 5 (filed Nov. 5, 2014) (“MCPC Reply”). However, the Commission’s rules permit a party-in-interest to file a petition for reconsideration of the grant of a pro forma application within 30 days of the date of public notice of the acceptance of the pro forma notification. 47 C.F.R. § 1.106(f). Thus, MCPC has no basis for contending that it did not have adequate notice of AT&T’s actions. Significantly, with respect to the 11 applications specifically identified by MCPC, no petition for reconsideration was ever filed following the release of the public notices reflecting Commission acceptance of such applications. As a result, MCPC’s contentions here are in effect late-filed petitions for reconsideration of AT&T’s pro forma transfer applications, and so they do not warrant further consideration in this proceeding. Federal Communications Commission FCC 15-94 18 36. MCPC maintains that the 11 transactions resulted in a substantial change in ownership and therefore could not be treated as pro forma transactions, because, according to MCPC, the Commission treats “any general partnership interest” as a “controlling interest.”100 However, the rules that MCPC cites for this proposition do not govern the determination of whether the pro forma transaction procedures are applicable to a proposed transaction, but rather how to attribute interests in spectrum holdings for purposes of a competitive analysis of the spectrum aggregation that would result from a transaction.101 37. On the other hand, the standard established by the Commission for determining whether the pro forma transaction procedures are applicable is whether there is a “substantial change in ownership.” Generally, a “substantial change in ownership” requires a transfer of 50 percent or more of a licensee’s stock.102 A “substantial change in ownership” requires a change in de facto or de jure control, and so it requires more than simply determining whether there is “any general partnership interest,” as MCPC suggests.103 Because AT&T exercised control both before and after the transactions, there was not a substantial change in ownership.104 38. Further, MCPC has not attempted to substantiate its claims of misrepresentation, the critical element of which is intent to deceive.105 MCPC claims that AT&T conceded the truth of this allegation by not refuting it.106 To the contrary, however, AT&T disagreed with MCPC’s underlying premise that the pro forma transactions were improper,107 and, for the reasons stated above, we agree. Thus, MCPC has not established a prima facie case that AT&T intended to deceive the Commission. 100 MCPC Comments at 15 (citing 47 C.F.R. § 20.22(b)(1) (adopted in Policies Regarding Mobile Spectrum Holdings, WT Docket No. 12-269, Report and Order, 29 FCC Rcd 6133, 6248 (2014) (“Spectrum Holdings Report and Order”); 47 C.F.R. §§ 1.919(c)(ii)(A), 20.6(d)(1)). 101 Spectrum Holdings Report and Order, 29 FCC Rcd at 6245, ¶ 302. Since 2004, the Commission has applied a test, called an “initial spectrum screen,” to help identify for case-by-case review local markets where changes in spectrum holdings resulting from the transaction may be of particular concern. AT&T-Cingular Order, 19 FCC Rcd at 21525, ¶ 4 (cited in Spectrum Holdings Report and Order, 29 FCC Rcd at 6140-41, ¶ 13). 102 See, e.g., SoftBank-Sprint Order, 28 FCC Rcd at 9700-01, ¶ 142. In the SoftBank-Sprint Order, the Commission noted that there may be other factors in particular cases that may affect de facto control, but nothing in the record in this proceeding suggests that any such factors were relevant in the 11 assignments cited by MCPC. 103 In the Section 310(d) Forbearance Order, the Commission listed several factors that may be relevant to a finding of de facto control: (1) power to constitute or appoint more than 50 percent of the board of directors or partnership management committee; (2) authority to appoint, promote, demote, and fire senior executives that control the day- to-day activities of the licensee; (3) ability to play an integral role in major management decisions of the licensee; (3) authority to pay financial obligations, including expenses arising out of operating; (4) ability to receive monies and profits from the facility’s operations; and (5) unfettered use of all facilities and equipment. Section 310(d) Forbearance Order, 13 FCC Rcd at 6297-98, ¶ 7 (citing, e.g., Intermountain Microwave, 24 Rad. Reg. (P & F) 983, 984 (1963)). 104 Joint Opposition at 74 (stating that AT&T filed pro forma notices consistent with 47 C.F.R. § 1.948(c) because it controlled the entity both before and after the transaction). 105 Swan Creek Commc’ns, Inc. v. FCC, 39 F.3d 1217, 1222 (D.C. Cir. 1994); Citadel Broadcasting Co., Memorandum Opinion and Order, Notice of Apparent Liability, 22 FCC Rcd 7083, 7090, ¶ 14 (2007) (“Citadel Broadcasting NAL”). 106 MCPC Reply at 3 (“[I]t is particularly for the Commission to adjudicate MCPC’s undisputed allegation that AT&T misrepresented facts in a series of filings that purported to notify the Commission of the transactions which effected the ouster of its partners.”) (emphasis in original). 107 Joint Opposition at 74 (stating that “MCPC’s arguments are incorrect” and that AT&T properly filed pro forma notices). Federal Communications Commission FCC 15-94 19 c. Notice of Apparent Liability 39. As noted above, MCPC recommends that the Commission examine the facts underlying an NAL recently issued to AT&T108 to determine whether AT&T has engaged in a pattern of noncompliant behavior.109 As both MCPC and AT&T observe, Section 504(c) prohibits the Commission from using the mere issuance of an NAL to the prejudice of AT&T in this or any other proceeding.110 MCPC is correct that, as a general matter, the Commission can consider the facts underlying an NAL, when appropriate, to determine whether a party is engaging in a pattern of noncompliant behavior.111 40. In this case, we find that MCPC has not provided any basis for finding that the recently issued NAL evidences a pattern of noncompliant behavior sufficient to call into question AT&T’s reliability as a licensee. MCPC merely alleges, without more, that the NAL, together with orders adopting consent decrees “that forced AT&T to settle for $126.75 million,” establishes a “pattern of flagrant disregard” for the Communications Act and the Commission’s rules.112 The Commission has fully investigated the conduct underlying the NAL and determined that the appropriate sanction is forfeiture, not revocation or another remedy premised on a conclusion that AT&T lacks the necessary character qualifications to hold a license. The orders adopting consent decrees that MCPC cites prohibit the Commission from using the facts developed in the subject investigations in any other proceeding, and the adopting orders concluded that there was no basis for an adverse character finding as a result of the investigations.113 Finally, MCPC has provided no evidence to suggest that AT&T has engaged in misrepresentation or lack of candor, and nothing in the record otherwise presents a substantial and material question of fact as to AT&T’s proclivity to deal truthfully with the Commission.114 d. CALEA 41. In addition to our conclusion that MCPC generally lacks standing to participate in this proceeding, to the extent that MCPC raises concerns regarding public confidence in the privacy of individuals’ communications, we disagree that it has standing to raise such concerns in this proceeding. Generally, courts will grant a third party standing to assert the rights of another only if the third party asserting the right has a close relationship with the person who possesses the right and if there is a hindrance to the possessor’s ability to protect his own interests.115 MCPC has not demonstrated that any of its members were subscribers of AT&T whose communications or call-identifying information was inappropriately disclosed to NSA, nor has it identified any hindrance limiting AT&T subscribers’ ability 108 AT&T Mobility Puerto Rico NAL, 30 FCC Rcd at 856, ¶¶ 1-2 (proposing $640,000 forfeiture based on a finding that AT&T willfully and repeatedly operated 59 of its common carrier fixed point-to-point microwave stations at variance from their authorized parameters in violation of Section 301 of the Communications Act, 47 U.S.C. § 301, and Sections 1.903(a) and 1.947(a)-(b) of the Commission’s rules, 47 C.F.R. §§ 1.903(a), 1.947(a)-(b)). 109 MCPC March 4, 2015, Ex Parte Letter at 1-2 (citing 1999 Forfeiture Guidelines, 15 FCC Rcd at 304, ¶¶ 3-4; Infinity Forfeiture Review Order, 22 FCC Rcd at 9827, ¶ 9). 110 47 U.S.C. § 504(c). 111 1999 Forfeiture Guidelines, 15 FCC Rcd at 304, ¶ 3. AT&T appears to have misread the 1986 Character Policy Statement because in that order the Commission found that allegations of non-Commission misconduct prior to adjudication by another agency or court are inappropriate grounds for a finding of unfitness and was not referring to unadjudicated Commission NALs. 1986 Character Policy Statement, 102 FCC 2d at 1204-06, ¶ 48. 112 See MCPC March 4, 2015, Ex Parte Letter at 1-2; MCPC Reply at 2 (citing AT&T Mobility LLC, Order, DA 14- 1457, 29 FCC Rcd 11803 (EB 2014) (“AT&T Mobility LLC”); AT&T Inc., Order, DA 13-594, 28 FCC Rcd 5994 (EB 2013) (“AT&T Inc. Order”)). 113 AT&T Mobility LLC, 29 FCC Rcd 11803; AT&T Inc. Order, 28 FCC Rcd 5994. 114 See 1986 Character Policy Statement, 102 FCC 2d at 1209, ¶ 55 (relevant character traits are truthfulness and reliability). 115 Kowalski v. Tesmer, 543 U.S. 125, 129-130 (2004) (citing Powers v. Ohio, 499 U.S. 400, 411 (1991)). Federal Communications Commission FCC 15-94 20 to protect their own interests. Thus, MCPC has not shown that it can meet either prong of the Supreme Court’s legal test. 42. Further, even if MCPC had standing to raise its CALEA concern, we find that there are procedural grounds for not considering MCPC’s contentions. MCPC raised its concern for the first time in its reply.116 As was emphasized in the Public Notice, to allow the Commission to consider fully all substantive issues regarding the Application in as timely and efficient a manner as possible, petitioners and commenters must raise all issues in their initial filings. New issues may not be raised in responses or replies.117 A party or interested person seeking to raise a new issue after the pleading cycle has closed must show good cause why it was not possible for it to have raised the issue previously.118 MCPC’s allegations are untimely, and MCPC has offered no explanation for its failure to raise them earlier. The existence of the surveillance program was widely known well before this proceeding was initiated, and the sources MCPC cites in support of its allegations that AT&T participated in the program all were publicly available in advance of this proceeding. 43. In addition, MCPC states that the alleged CALEA rule violations ceased in July 2004.119 One of the factors that the Commission considers in making character determinations is the passage of time since the conduct. In the 1986 Character Policy Statement, the Commission determined that, even as to consideration of past conduct indicating “a flagrant disregard of the Commission’s regulations and policies,” a 10-year limitation should apply, given the “inherent inequity and practical difficulty” involved in requiring applicants to respond to allegations of greater age.120 As the AT&T conduct that MCPC maintains resulted in rule violations ended more than 10 years ago, the limitation adopted in the 1986 Character Policy Statement applies to that conduct. 44. Finally, assuming for the sake of argument that MCPC had raised this argument in a timely manner, we conclude that the argument does not provide an adequate basis for investigating AT&T’s character qualifications or imposing a remedial condition. MCPC asserts not only that AT&T participated in the PSP surveillance program but that it knew that its participation violated CALEA and the Commission’s rules.121 However, Congress adopted legislation in 2008 to give electronic communications service providers, including AT&T, immunity from civil suit for their involvement in the PSP surveillance program under certain circumstances.122 Prior to adopting this legislation, the Senate Select Committee on Intelligence determined that, contrary to MCPC’s assertion, the PSP participants acted on a good faith belief that the program and their assistance were lawful.123 Therefore, we do not believe any facts that might be developed in connection with this proceeding are necessary to make a determination regarding AT&T’s proclivity to deal truthfully with the Commission and to comply with the Communications Act and Commission rules, which is the essence of the character qualifications inquiry. 116 MCPC Reply at 4-5. 117 Public Notice, 29 FCC Rcd at 9469 (citing 47 C.F.R. § 1.45(c)). 118 Id. 119 MCPC March 4, 2015, Ex Parte Letter at 11. 120 1986 Character Policy Statement, 102 FCC 2d at 1229, ¶ 105 (quoting Kaye-Smith Enterprises, 71 FCC 2d 1402, 1406-07, ¶ 10 (1979), recon. denied, 46 Rad. Reg. 2d (P & F) 1583 (1980)). 121 MCPC March 4, 2015, Ex Parte Letter at 9-11. 122 See 50 U.S.C. § 1885a (immunity legislation), S. REP. NO. 110-209, at 7 (2007). 123 S. REP. NO. 110-209, at 7. Federal Communications Commission FCC 15-94 21 2. New Networks Institute & Teletruth Petition 45. On May 12, 2015, New Networks filed a Petition for Investigation opposing the transaction and requesting that the Commission investigate whether AT&T engaged in misrepresentation, in violation of Section 1.17 of the Commission’s rules, in its representations to the Commission of its compliance with the AT&T-BellSouth Merger Order124 broadband deployment condition.125 New Networks alleges that AT&T did not meet the AT&T-BellSouth Merger Order broadband condition based on statements AT&T has made since 2008 regarding its plans for its wireline and wireless broadband deployment.126 AT&T asserts that it complied with the condition.127 46. Consistent with the Public Notice establishing the pleading cycle in this proceeding, we dismiss these untimely new allegations.128 Under the pleading schedule established for this proceeding, comments and petitions to deny were due over nine months ago.129 New Networks had ample time to submit its petition during the established pleading cycle, but it failed to do so and offers no justification for its late submission. Nor does New Networks seek leave to file belatedly. We note, however, that New Networks may pursue its allegations by filing a complaint with the Enforcement Bureau.130 47. We further conclude, as an alternative, independent basis for rejecting the petition, that the record does not present a substantial and material question of fact regarding New Networks’ claim 124 AT&T-BellSouth Merger Order, 22 FCC Rcd 5662. 125 New Networks Institute & Teletruth Petition for Investigation into Whether AT&T Committed Perjury in its Representations to the FCC Regarding its Deployment of Broadband and Request to Delay Action on the AT&T- DIRECTV Merger Pending Investigation, MB Docket No. 14-90 (filed May 14, 2015) (“New Networks Petition”); see 47 C.F.R. § 1.17. New Networks also alleges that promises for fiber to the curb and fiber to the home (“FTTH”) were made to the Commission in 2004 and have not been fulfilled. See New Networks Petition at 4-5, 14-17. 126 See New Networks Petition at 1-5. 127 Joint Opposition at 23 n.67; AT&T Response to Sept. 9, 2014, Information Request at 222-223. 128 To the extent New Networks seeks to have its petition treated as a petition to deny, it is dismissed as untimely for the reasons stated above. Further, New Networks lacks standing to file a petition to deny because it has not established that any individual member would be harmed by the transaction. See Applications for Consent to the Assignment and/or Transfer of Control of Licenses Adelphia Communications Corporation (and Subsidiaries, Debtors-In-Possession), Assignors, to Time Warner Cable Inc. (Subsidiaries), Assignees, Adelphia Communications Corporation, (and Subsidiaries, Debtors-In-Possession), Assignors and Transferors, to Comcast Corporation (Subsidiaries), Assignees and Transferees, MB Docket No. 05-192, Memorandum Opinion and Order, 21 FCC Rcd 8203, 8215-16, ¶¶ 18-20 & n.73 (2006) (“Adelphia Order”) (standing of organization established by evidence that members resided in service area of MVPD applicants). Our dismissal is without prejudice to New Networks’ ability to raise its allegations outside of this proceeding in a complaint filed with the Enforcement Bureau. 129 Public Notice, 29 FCC Rcd at 9464 (The Public Notice accepted the Application for filing on August 7, 2014, and it established September 16, 2014, as the deadline for filing comments or petitions to deny, October 16, 2014, as the deadline for responses to comments or oppositions to petitions to deny, and November 5, 2014, as the deadline for replies to responses or oppositions). The Public Notice stated that any “party or interested person seeking to raise a new issue after the pleading cycle has closed must show good cause why it was not possible for it to have raised the issue previously . . . . Absent such a showing of good cause, any issues not timely raised may be disregarded by the Commission.” Id. at 9469. The pleading cycle was temporarily suspended by the Media Bureau on October 22, 2014, and subsequently, the deadline for replies to responses or oppositions was extended until January 7, 2015. Applications of Comcast Corp. and Time Warner Cable Inc. for Consent to Assign or Transfer Control of Licenses and Authorizations, and AT&T, Inc. and DIRECTV for Consent to Assign or Transfer Control of Licenses and Authorizations, MB Docket Nos. 14-57, 14-90, Order, DA 14-1523, 29 FCC Rcd 12715 (MB 2014); Notice of Merger Pleading Cycle Restarts, 29 FCC Rcd 14491. 130 47 U.S.C. §§ 208, 503. See Softbank-Sprint Order, 28 FCC Rcd at 9676, ¶ 85 (finding that allegations of prior conduct by the licensee were more appropriately resolved through the Commission’s complaint process under Section 208 of the Act). Federal Communications Commission FCC 15-94 22 that AT&T violated the Commission’s rules and AT&T intended to deceive the Commission with respect to its broadband deployment.131 The critical representation, according to New Networks, pertained to residential broadband deployment as of the end of 2007, defined as a specific speed of service in a defined geographic area.132 48. Specifically, the AT&T-BellSouth Merger Order broadband condition required that, by December 31, 2007, AT&T-BellSouth would offer Internet access service at speeds in excess of 200 kilobits per second (“kbps”) in at least one direction to 100 percent of the residential living units in the AT&T-BellSouth in-region territory, using wireline broadband for at least 85 percent of such living units, and using “alternative technologies and operating arrangements including but not limited to satellite service and [Worldwide Interoperability for Microwave Access (“Wi-Max”)] fixed wireless technologies” for the remaining 15 percent of living units.133 Further, at least 30 percent of the new deployment needed to be to rural or low-income units.134 New Networks’ petition alleges that AT&T neither deployed Internet service to 100 percent of the residential living units in the in-region territory, nor achieved the 30 percent incremental deployment to rural or low-income units. 49. New Networks’ position appears to be that AT&T did not satisfy the broadband deployment condition because it failed to cover 100 percent of its footprint with its wireline U-verse service, a technology that provides download speeds of up to 45 Mbps or greater.135 However, New Networks neglects to consider that only 85 percent of living units in the AT&T-BellSouth in-region territory needed to be covered by wireline technologies and that technologies other than U-verse, including digital subscriber line (“DSL”) and IPDSL, would also meet the defined level of service. In addition, the deployment totals that New Networks provides are based upon AT&T’s current footprint and not AT&T’s footprint as it existed on December 31, 2007, when the condition applied.136 New Networks further neglects to consider that 15 percent of living units in the footprint could be covered using alternative technologies or operating arrangements, such as satellite service and Wi-Max fixed wireless service.137 131 See Swan Creek Commc’ns, Inc. v. FCC, 39 F.3d at 1222 (intent to deceive is a critical element of misrepresentation); Citadel Broadcasting NAL, 22 FCC Rcd at 7090, ¶ 14 (“[C]arelessness, exaggeration, or slipshoddiness, which lack [the] necessary element [of intent to deceive], do not constitute misrepresentation.”). 132 New Networks Petition at 17-19. See also AT&T-BellSouth Merger Order, 22 FCC Rcd at 5807, Appendix F. 133 AT&T-BellSouth Merger Order, 22 FCC Rcd at 5807, Appendix F. 134 Id. 135 See supra n.9 (noting that AT&T currently offers U-verse primarily via FTTN architecture, which offers speeds of up to 45 Mbps). 136 Free Press also alleges that AT&T failed to fulfill the AT&T-BellSouth Merger Order condition, citing a press report from 2012 indicating that some individuals at that time had been unable to obtain broadband service from AT&T. Free Press Petition at 32-33. The Commission has previously determined that news articles are not sufficient to establish a prima facie showing under Section 309(d) of the Communications Act, 47 U.S.C. § 309(d)(1). See Application of the Pikes Peak Broad. Co., Memorandum Opinion and Order and Notice of Apparent Liability, 12 FCC Rcd 4626, 4630, ¶ 14 (1997) (“Pikes Peak Broad. Order”) (“[A] newspaper article is not an acceptable substitute for the requirement of Section 309(d) of the Communications Act that allegations in a petition to deny be supported by the affidavit of a person with personal knowledge of the facts alleged.”). Moreover, the merger condition applied to a period that ended several years before the period at issue in the cited news article. Free Press does not claim that its allegations call into question AT&T’s character qualifications, and for the reasons explained here, we conclude that they do not. 137 For example, Wideband Code Division Multiple Access (“WCDMA”) technology and High Speed Data Packet Access (“HSDPA”) technology both were capable of delivering Internet service that would have met the definition of Internet service under the AT&T-BellSouth Merger Order condition. WCDMA provides speeds up to 2 Mbps and average user speeds of 220-320 kbps. See Implementation of Section 6002(b) of the Omnibus Budget Reconciliation Act of 1993, Annual Report and Analysis of Competitive Market Conditions with Respect to Commercial Mobile (continued….) Federal Communications Commission FCC 15-94 23 50. New Networks also argues that the Applicants’ statements regarding the deployment of an additional 15 million locations in connection with the instant proceeding is further evidence that AT&T does not have 100 percent broadband deployment in its footprint and has therefore not satisfied its 2006 condition to achieve 100 percent deployment.138 However, New Networks fails to consider the extent to which deployment of such additional service might be to areas located outside of AT&T’s incumbent local exchange carrier (“ILEC”) territory.139 Further, as discussed above, it fails to consider the extent to which AT&T had deployed wireline broadband technologies other than U-verse or wireless technologies. 51. As a result, we cannot conclude that the facts alleged by New Networks, if true, would demonstrate that AT&T falsely certified compliance with the AT&T-BellSouth Merger Order conditions or intended to deceive the Commission.140 Thus, we find that New Networks has not made out even a prima facie case of a rule violation or misrepresentation.141 Even assuming New Networks has made a prima facie case, we conclude, for the reasons explained above, that the record does not present a substantial and material question of fact on this issue warranting further inquiry in this proceeding.142 (Continued from previous page) Services, WT Docket No. 07-71, Twelfth Report, 23 FCC Rcd 2241, 2300, ¶ 130 (2008) (“12th Annual CMRS Competition Report”). HSDPA allows average download speeds of 400-700 kbps. Id. at ¶ 130. In 2005, AT&T (Cingular) began its launch of WCDMA. See Implementation of Section 6002(b) of the Omnibus Budget Reconciliation Act of 1993, Annual Report and Analysis of Competitive Market Conditions with Respect to Commercial Mobile Services, WT Docket No. 06-17, Eleventh Report, 21 FCC Rcd 10947, 10999, ¶ 130 (2006). AT&T continued to deploy WCDMA/HSDPA on its network in 2006, and it planned to continue to deploy WCDMA/HSDPA throughout a majority of its network. See 12th Annual CMRS Competition Report, 23 FCC Rcd at 2302, ¶ 137. See also AT&T Inc., AT&T Significantly Expands Broadband Service to Rural Consumers Across Traditional 22-State Territory (press release), May 9, 2007, available at http://www.prnewswire.com/news- releases/att-significantly-expands-broadband-service-to-rural-consumers-across-traditional-22-state-territory- 58052722.html (visited June 12, 2015) (announcing the expansion of AT&T’s satellite broadband service). 138 New Networks Petition at 3. In addition, New Networks relies on alleged facts contained in third-party sources such as trade press and analyst reports, id. at 20-22, 24, Items 9, 10, 13, that are not supported by an affidavit of an individual with personal knowledge of the facts alleged, as required by Section 309(d)(1) of the Communications Act. 47 U.S.C. § 309(d)(1); Pikes Peak Broad. Order, 12 FCC Rcd at 4630, ¶ 14. 139 The 15 million locations in connection with the instant proceeding include 13 million FWLL locations. Our analysis verifies that the majority of the FWLL deployment would be outside of AT&T’s wireline territory. See infra ¶ 270. 140 See Gencom Inc. v. FCC, 832 F.2d 171, 181 (D.C. Cir. 1987) (“The Commission’s inquiry at this level is much like that performed by a trial judge considering a motion for a directed verdict: if all the supporting facts alleged in the affidavits were true, could a reasonable factfinder conclude that the ultimate fact in dispute had been established.” (citing Citizens for Jazz on WRVR, Inc. v. FCC, 775 F.2d 392, 397 (D.C. Cir. 1985))). 141 Id. at 180 (“Initially, the Commission must determine whether the applicant seeking a hearing has set forth ‘specific allegations of fact sufficient to show that . . . a grant of the application would be prima facie inconsistent with [the public interest, convenience, and necessity].’” (quoting Citizens for Jazz on WRVR, Inc. v. FCC, 775 F.2d at 394); Shareholders of Hispanic Broad. Corp. (Transferor) and Univision Communications, Inc. (Transferee) for Transfer of Control of Hispanic Broad. Corp., and Certain Subsidiaries, Licensees of KGBT9AM), Harlingen, TX et al., MB Docket No. 02-235, Memorandum Opinion and Order, 18 FCC Rcd 18834, 18845, ¶ 29 (2003) (petitions that offer conclusory statements or allegations not supported by record evidence, affidavits, or sworn statements are not sufficient to meet the pleading requirements of Section 309(d)(1) of the Act, 47 U.S.C. § 309(d)(1)). 142 47 U.S.C. § 309(d)(2), (e) (if the record presents a substantial and material question of fact as to whether an application is in the public interest, the Commission shall designate the matter for hearing). Federal Communications Commission FCC 15-94 24 VI. COMPLIANCE WITH COMMUNICATIONS ACT AND FCC RULES AND POLICIES 52. As noted above, for the proposed transaction to be in the public interest, it must be in compliance with the Communications Act, other applicable statutes, and the Commission’s rules.143 We find that the proposed transaction will not violate any statutory provision or Commission rule.144 VII. BACKGROUND ON VIDEO PROGRAMMING DISTRIBUTORS 53. This transaction involves the combination of two entities that deliver video programming to consumers. As background for our analysis of that combination, below we provide an overview of the video programming distribution industry. 54. Today there are primarily three types of entities that deliver video programming to consumers – broadcast television stations, MVPDs, and online video distributors (“OVDs”).145 We focus our industry description on MVPDs because our analysis of the public interest benefits and harms considers in substantial part the competitive effects of the transaction on those services. We also consider the evolution of OVD services. Related to our analysis are bundles that combine video services with Internet access and/or voice telephony as part of a “double-” or “triple-play” bundle.146 55. MVPDs include cable operators (both incumbent cable operators and cable “overbuilders”),147 direct broadcast satellite (“DBS”) providers, such as DIRECTV, and telephone companies (“telephone MVPDs”), such as AT&T and Verizon, that provide MVPD service in some of the areas where they offer landline telephone service.148 MVPDs bundle linear programming networks into groups of channels or “tiers”149 and sell this programming to consumers, deriving revenues from subscription fees and the sale of advertising time they receive through their carriage agreements.150 MVPDs primarily deliver video programming services using their own facilities.151 As part of an MVPD subscription, MVPDs also typically offer VOD and TV Everywhere services, which allow subscribers to 143 See, e.g., Comcast-NBCU Order, 26 FCC Rcd at 4247, ¶ 22; Sirius-XM Order, 23 FCC Rcd at 12363-64, ¶ 30; Liberty Media-DIRECTV Order, 23 FCC Rcd at 3276-77, ¶ 22; EchoStar-DIRECTV HDO, 17 FCC Rcd at 20574, ¶ 25. 144 The Applicants state that the transaction would not violate any law or rule. Application at 18. Moreover, no party has alleged that the transaction would violate any provision of the Communications Act, related statutes, or the Commission’s rules. 145 See Annual Assessment of the Status of Competition in the Market for the Delivery of Video Programming, MB Docket No. 14-16, Sixteenth Annual Report, 30 FCC Rcd 3253, 3256-59, ¶¶ 2-11 (2015) (“Sixteenth Annual Report”). 146 See id. at 3261, ¶ 21, 3285-86, ¶¶ 74, 77, 3288, ¶ 81. 147 Overbuilders are generally defined as companies that build additional cable systems “over” one that already exists and offer customers a competitive alternative. See Sixteenth Annual Report, 30 FCC Rcd at 3265 n.59. 148 The two largest telephone MVPDs are Verizon and AT&T. See generally id. at 3263-64, ¶ 27. 149 Linear television channels are streams of programming that offer video programs on a specific channel at a specific time of day. See id. at 3260, ¶ 18; see also Promoting Innovation and Competition in the Provision of Multichannel Video Programming Distribution Services, MB Docket No. 14-261, Notice of Proposed Rulemaking, 29 FCC Rcd 15995 (2014) (“MVPD Definition NPRM”) (seeking comment on a proposal to define “linear video” as a “stream of video programming that is prescheduled by the programmer”). 150 Liberty Media-DIRECTV Order, 23 FCC Rcd at 3280, ¶ 30. We note that the Commission is currently seeking comment on the interpretation of the statutory definition of “MVPD” and on whether that definition includes certain Internet-based distributors of video programming. See MVPD Definition NPRM, 29 FCC Rcd 15995. 151 See Sixteenth Annual Report, 30 FCC Rcd at 3260-61, ¶¶ 19, 21. Federal Communications Commission FCC 15-94 25 access a selection of programming at a time of their choosing and on a variety of in-home and mobile, Internet-connected devices.152 56. Where capable, MVPDs may offer their subscribers such video services as part of a bundle that may include Internet access/and or voice telephony.153 Bundles are one way that MVPDs attempt to differentiate their services from the services of their rivals.154 These bundles are usually offered at a discount to purchasing the parts of the bundle separately.155 An MVPD that also has Internet access facilities is able to offer a bundle (an “integrated bundle”).156 Other MVPDs (such as DBS providers) that do not have the facilities to provide bundles of services may enter into cooperative agreements with telephone companies, cable operators, or satellite Internet providers so that they can market bundles to their video subscribers (“synthetic bundles”).157 57. Subscribers are increasingly buying video services as part of a bundle. From 2012 to 2013, revenues from bundles increased for several cable providers from a little more than 1 percent (Time Warner Cable) to almost 10 percent (Charter).158 152 See id. at 3260, ¶ 18, 3294-96, ¶¶ 95-100. In contrast to cable systems, DBS systems have less bandwidth and use a one-way technology, which puts DBS at a disadvantage when it comes to providing two-way video services like VOD. DIRECTV and DISH deliver some VOD content over broadband, which requires their subscribers to also subscribe to a broadband service provided by a separate entity. Id. at 3300-01, ¶¶ 111-114; see Annual Assessment of the Status of Competition in the Market for the Delivery of Video Programming, MB Docket No. 12- 203, Fifteenth Annual Report, 28 FCC Rcd 10496, 10546-57, ¶¶ 112-114 (2013) (“Fifteenth Annual Report”); see also DIRECTV, SEC Form 10-K for the Year Ended Dec. 31, 2013, at 3 (“We also provide video-on-demand, or VOD, by ‘pushing’ top-rated movies onto customers’ digital video recorders, or DVRs, for instant viewing, as well as via broadband to our subscribers who have connected their set-top receiver to their broadband service.”); Application, Declaration of Patrick T. Doyle, Executive Vice President and Chief Financial Officer, DIRECTV, transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, ¶¶ 19-21 (filed June 11, 2014) (“Doyle Decl.”) (discussing the importance of a two-way connection using broadband and the fact that DIRECTV’s ability to provide VOD services has been hampered by the lack of such a connection). 153 Broadband Internet access service means a mass-market retail service by wire or radio that provides the capability to transmit data to and receive data from all or substantially all Internet endpoints, including any capabilities that are incidental to and enable the operation of the communications service, but excluding dial-up Internet access service. Protecting and Promoting the Open Internet, GN Docket No. 14-28, Report and Order on Remand, Declaratory Ruling, and Order, 30 FCC Rcd 5601, 5682, ¶ 187 (2015) (“2015 Open Internet Order”); see also Sixteenth Annual Report, 30 FCC Rcd at 3372, ¶ 258 (“Access to high-speed data pipelines capable of delivering a high quality video signal is critical for OVD entrants. In some offerings, OVDs require sufficient Internet capacity to transmit their programming, and consumers need sufficient broadband service to access OVDs’ content.”) (citations omitted). 154 See Sixteenth Annual Report, 30 FCC Rcd at 3288, ¶ 81, 3297, ¶ 101. 155 See id. at 3288, ¶ 81. 156 Application, Declaration of Paul Guyardo, Executive Vice President and Chief Revenue and Marketing Officer, DIRECTV, transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, ¶ 7 (filed June 11, 2014) (“Guyardo Decl.”); AT&T and DIRECTV, White Paper, Additional Evidence that Video and Broadband are Complements, at 2, 26 (“Video and Broadband Complementarity White Paper”), transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (filed Nov. 12, 2014). 157 Guyardo Decl. ¶ 7; see also Sixteenth Annual Report, 30 FCC Rcd at 3261, ¶ 21 n.24, 3286, ¶ 77. 158 Time Warner Cable Inc., SEC Form 10-K for the Year Ended Dec. 31, 2013, at 39; Comcast Corp., SEC Form 10-K for the Year Ended Dec. 31, 2013, at 53; Charter Communications, Inc., SEC Form 10-K for the Year Ended Dec. 31, 2013, at 40. Federal Communications Commission FCC 15-94 26 58. OVDs offer consumers choices that may either complement the consumer’s MVPD services or compete directly with at least some of the services provided by MVPDs.159 Most OVDs today do not offer a substantial amount of the most popular video programming that is provided by MVPDs, including live sports programming and local broadcast programming, nor do most OVDs offer bundles of linear programming such as those offered by traditional MVPDs. The number and types of OVDs have grown significantly over the last few years and include programmers, content producers and owners, affiliates of online services, retailers, manufacturers, and MVPDs.160 The types of services that OVDs offer vary widely and include, but are not limited to, linear programming, on-demand programming, and combinations of original programming and full length movies and television programs.161 Further, several online video services have been announced or launched recently that promise to offer access to popular linear networks in a manner similar to MVPD services.162 Consumers can access OVD services through an Internet connection on their computers, tablets, and mobile wireless devices and, within the last few years, using a range of devices that allow consumers to view OVD services on their 159 See Sixteenth Annual Report, 30 FCC Rcd at 3352-53, ¶ 215. 160 See id. at 3353-63, ¶¶ 216-235. Further, some devices that access OVDs, such as Roku and Amazon Fire TV, function as aggregators. See id. at 3362, ¶ 233. It is difficult at this time to determine to what extent individual OVDs have grown because rating/viewing information is non-standard and limited. See id. at 3365-66, ¶ 242. However, Netflix publicly reports its subscriber and revenue figures for its online streaming service. Netflix had 20 million streaming subscribers in the United States at the end of 2011 and its subscribers increased to approximately 32 million at the end of 2013. Netflix, Inc., SEC Form 10-Q/A for the Quarterly Period Ended Sept. 30, 2012, at 3- 4; Netflix, Inc., SEC Form 10-K for the Year Ended Dec. 31, 2013, at 19. 161 See Application, “An Economic Assessment of AT&T’s Proposed Acquisition of DIRECTV,” Declaration of Michael L. Katz, transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, ¶ 48 (filed June 11, 2014) (“Katz Decl.”). An example of a linear OVD is Pluto.TV. See Pluto.TV, http://pluto.tv/ (visited June 18, 2015). An example of an OVD providing on-demand programming is Hulu Plus, which includes programs that originally aired the previous day on broadcast and cable television. See Hulu, http://www.hulu.com/tv (visited June 18, 2015). Examples of OVDs offering combinations of original programming and full length movies and television shows are Netflix and Amazon. See Sixteenth Annual Report, 30 FCC Rcd at 3359, ¶ 229, 3361-62, ¶ 232; Amazon.com, Inc., Amazon Original Series Alpha House and Betas to Premier This Month (press release), Nov. 4, 2013, available at http://phx.corporate- ir.net/phoenix.zhtml?c=176060&p=irol-newsArticle&ID=1871791 (visited June 15, 2015). 162 See, e.g., DISH Network Corp., Sling TV to Launch Live, Over-the-Top Service for $20 Per Month; Watch on TVs, Tablets, Computers, Smartphones, Game Consoles (press release), Jan. 5, 2015, available at http://about.dish.com/press-release/products-and-services/sling-tv-launch-live-over-top-service-20-month-watch- tvs-tablets (visited June 18, 2015); Sony Corp., Sony Network Entertainment International and Sony Computer Entertainment Unveil PlayStationTM Vue, A New Cloud-Based TV Service That Pioneers the Future of Television (press release), Nov. 13, 2014, available at http://www.sony.com/en_us/SCA/company-news/press-releases/sony- corporation-of-america/2014/sony-network-entertainment-international-and-sony-.html?icid=pr-newswire-feed (visited June 24, 2015); Joe Flint, CBS Launches Online Subscription Video Service, WALL ST. J., Oct. 16, 2014, available at http://www.wsj.com/articles/cbs-launches-online-subscription-video-service-1413465013 (visited June 18, 2015); Ryan Knutson, Verizon Eyes Digital Video Service by Mid-2015, WALL ST. J., Sept. 14, 2014, available at http://www.wsj.com/articles/verizon-ceo-eyes-digital-video-service-by-mid-2015-1410467151 (visited June 18, 2015). Federal Communications Commission FCC 15-94 27 televisions.163 Although the number of customers who are relying only on OVD services to access video programming is growing, it is still a small fraction of the consumers purchasing video services.164 VIII. INCREASED MARKET CONCENTRATION IN VIDEO DISTRIBUTION SERVICES 59. AT&T and DIRECTV both provide MVPD service to consumers in the United States. AT&T offers MVPD service under its U-verse brand within portions of its 22-state wireline footprint, while DIRECTV offers direct-to-home satellite digital television service to consumers nationwide.165 The record includes concerns raised by commenters that the transaction reduces the number of competitors that provide MVPD services and, as a result, harms competition for distribution of video programming. 60. In many geographic areas, the combination of AT&T and DIRECTV would result in the loss of an MVPD and provider of bundled video and broadband services. The Applicants maintain that this geographic overlap is minimal.166 In addition, the Applicants allege that even within those areas where both AT&T and DIRECTV offer video service, consumers would continue to have numerous competitive video options following the transaction.167 The Applicants contend that the current combination of Internet broadband and video offered by AT&T and DIRECTV in collaboration is a synthetic bundle and inferior to the integrated bundle that AT&T and cable providers each offer independently.168 163 See Sixteenth Annual Report, 30 FCC Rcd at 3353-59, ¶¶ 217-228; Katz Decl. ¶ 45; Matt Burnes, Google Launches the $35 Chromecast Streaming Device to Bring Chrome to the Living Room, TECHCRUNCH, July 24, 2013, available at http://techcrunch.com/2013/07/24/google-chromecast/ (visited June 18, 2015); Molly Wood, Where the Amazon Fire TV Fits Into the Living Room, N.Y. TIMES, April 2, 2014, available at http://bits.blogs.nytimes.com/2014/04/02/where-the-amazon-fire-tv-fits-into-the-living-room/?_r=0 (visited June 18, 2015). 164 See, e.g., Number of Cable “Cord Cutters” Continues to Rise, RT, April 19, 2014, available at http://rt.com/usa/cable-cutters-flock-netflix-hulu-532/ (visited June 18, 2015) (citing a study published by Experian Marketing Service and stating that from 2010 to 2013 the number of customers, with high-speed Internet access, who have never subscribed to cable (“cord nevers”) or stopped subscribing (“cord cutters”) increased by 44 percent from 5.1 to 7.6 million households). In contrast there are approximately 101 million MVPD subscribers. See Sixteenth Annual Report, 30 FCC Rcd at 3256, ¶ 2. 165 See Application at 10, 13. 166 See id. at 53, 68; see also Comments of the Free State Foundation, MB Docket 14-90, at 20 (filed Sept. 16, 2014) (“Free State Comments”). 167 See Application at 72-79. But see Petition to Deny of Free Press, MB Docket 14-90, at 16-19 (filed Sept. 16, 2014) (“Free Press Petition”) (arguing that the Applicants overstate the level of local MVPD competition when the Applicants point to: (1) Google Fiber, which has extremely limited coverage; (2) overbuilders, without providing any evidence of the extent of competition from overbuilders; and (3) OVD services, which rely on a broadband connection such as that provided by AT&T and which are not in the same product market as MVPD services). 168 Application, Declaration of Lori M. Lee, Senior Executive Vice President – Home Solutions, AT&T, transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14- 90, ¶¶ 4, 53-58 (filed June 11, 2014) (“Lee Decl.”); see also Application at 20, 52; Doyle Decl. ¶¶ 24-25; Guyardo Decl. ¶¶ 7, 21, 41-45; Katz Decl. ¶¶ 26-27, 29-32, 68-71, 97-106; Application, Declaration of John T. Stankey, Group President and Chief Strategy Officer, AT&T, transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, ¶¶ 26-29 (filed June 11, 2014) (“Stankey Decl.”); AT&T and DIRECTV, White Paper, Why the “Double Moral Hazard” Problem Cannot be Resolved by Contract, at 17-19 (“Double Moral Hazard White Paper”), transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (filed Nov. 12, 2014). The Applicants also note that competition “for video/broadband bundles occurs primarily between the competitors offering integrated bundles.” Application at 57-62. Federal Communications Commission FCC 15-94 28 61. The Applicants also explain that AT&T’s primary business is its broadband product and bundled services.169 Thus AT&T views cable providers, not DIRECTV, as its closest competitors.170 The Applicants assert that DIRECTV focuses on its standalone video product, which AT&T views as a complement to AT&T’s broadband offering.171 62. Several commenters express concern that the transaction would result in the loss of a video provider competitor in the areas where AT&T currently offers video service.172 Free Press and Writers Guild of America, West, Inc. (“WGAW”) also argue that the transaction would harm consumers by reducing the Applicants’ incentives to invest more in direct competition through expanded and improved broadband access, as well as by foreclosing innovation in MVPD services.173 Free Press challenges the Applicants’ assertion that several competitive options would remain for standalone video customers post-transaction.174 63. In contrast, Communications Workers of America (“CWA”) and the Free State Foundation (“Free State”) contend the transaction raises few antitrust concerns, noting that the transaction is primarily a non-horizontal merger of complementary services.175 Free State argues the transaction would potentially benefit consumers by expanding consumer choice for broadband and MVPD services.176 Further, Free State contends that AT&T and DIRECTV combined would have only 24 percent of the national market for video subscribers.177 64. To analyze these concerns about the loss of competition between AT&T and DIRECTV, we first consider the relevant product and geographic market definitions for the proposed transaction and evaluate any resulting concentration, and then we turn to a direct estimate of the competitive effects of the transaction. 169 Application at 52-53. 170 Id. at 56-58. 171 See id. at 7-8, 57, 69-71; see also AT&T and DIRECTV, White Paper, Additional Evidence that AT&T and DIRECTV are not Particularly Close Substitutes, at 2, 18-39 (“Competition White Paper”), transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (filed Nov. 12, 2014). 172 See, e.g., Free Press Petition at 6-19; Petition to Deny of Public Knowledge and Institute for Local Self-Reliance, MB Docket 14-90, at 5-8 (filed Sept. 16, 2014) (“Public Knowledge-ILSR Petition”); MCPC Comments at 19-21; Comments of Senator Al Franken, MB Docket 14-90, at 7 (filed Sept. 16, 2014) (“Franken Comments”); Petition to Condition Consent of Cox Communications, Inc., MB Docket 14-90, at 3 (filed Sept. 16, 2014) (“Cox Petition”); Comments of National Association of Broadcasters, MB Docket 14-90, at 2-3 (filed Sept. 16, 2014) (“NAB Comments”); Joint Petition to Deny of the Alliance for Community Media, the Alliance for Communications Democracy, and Common Cause, MB Docket 14-90, at 5-8 (filed Sept. 16, 2014) (“ACM et al. Petition”); Reply Comments of the Alliance for Community Media, the Alliance for Communications Democracy, and Common Cause, MB Docket 14-90, at 1-2 (filed Jan. 7, 2015) (“ACM et al. Reply”); Petition to Deny of Writers Guild of America, West, Inc., MB Docket 14-90, at 4, 9 (filed Sept. 16, 2014) (“WGAW Petition”); Reply Comments of Writers Guild of America, West, Inc. to Opposition, MB Docket 14-90, at 3-5 (filed Jan. 7, 2015) (“WGAW Reply”). 173 Free Press Petition at 6-7; WGAW Petition at 14-19. 174 Free Press Petition at 16-19. 175 Comments of Communications Workers of America, MB Docket 14-90, at 14-15 (filed Sept. 16, 2014) (“CWA Comments”); Free State Comments at 20-22. 176 Free State Comments at 16-17. 177 Id. at 21-22. Federal Communications Commission FCC 15-94 29 Product Market 65. We consider the relevant product market consistent with Commission precedent and the analytical framework and principles outlined by the 2010 DOJ/FTC Horizontal Merger Guidelines.178 In previous transactions involving video providers, the Commission defined the relevant product market as “multichannel video programming service” as offered by all MVPDs.179 66. Positions of the Parties. Several commenters argue that there is growing competition to MVPDs from online video.180 The Applicants note that online video offerings have grown exponentially and are becoming formidable alternatives to MVPD services.181 The Applicants further argue that although online video services have been considered complementary to traditional MVPD service, OVDs are becoming increasingly important competitors.182 As a result, the Applicants assert that traditional MVPDs are adjusting product offerings and marketing strategies in response to the mounting competition from these online video services.183 67. Free Press dismisses claims that online video services are substitutes for MVPD service and argues that OVD services should not be included in the same product market as MVPD service.184 As support for its position, Free Press claims that only a small number of consumers leave MVPD services for online video service and that, despite the expansion of online video offerings, the MVPD market continues to increase in both subscribers and profits.185 68. Discussion. Consistent with Commission findings in prior transactions, we conclude that the relevant product market for evaluating the record on market concentration is “multichannel video programming service” as offered by all MVPDs.186 Consistent with the Commission’s decision in the 178 DOJ and the Federal Trade Commission (“FTC”) Horizontal Merger Guidelines define the relevant product market as the smallest group of competing products for which a hypothetical monopoly provider of the products could profitably impose at least a “small but significant and non-transitory price increase,” presuming no change in the terms of sale of other products. U.S. Department of Justice and the Federal Trade Commission Horizontal Merger Guidelines, August 19, 2010, § 4.1.1 at 9 (“2010 DOJ/FTC Horizontal Merger Guidelines”). In other words, when one product is a reasonable substitute for the other in the eyes of a sufficiently large number of consumers, it is included in the relevant product market even though the products themselves are not identical. Thus, the relevant product market includes all products “reasonably interchangeable by consumers for the same purposes.” United States v. E.I. du Pont de Nemours & Co., 351 U.S. 377, 395 (1956); see also United States v. Microsoft Corp., 253 F.3d 34, 52 (D.C. Cir. 2001), cert. denied, 122 S. Ct. 350 (2001) (“[T]he relevant market must include all products ‘reasonably interchangeable by consumers for the same purposes.’” (quoting E.I. du Pont de Nemours, 351 U.S. at 395)). 179 See Comcast-NBCU Order, 26 FCC Rcd at 4255-56, ¶ 40; Liberty Media-DIRECTV Order, 23 FCC Rcd at 3280- 81, ¶ 31; Adelphia Order, 21 FCC Rcd at 8235, ¶ 63; News Corp.-Hughes Order, 19 FCC Rcd at 501, ¶ 53. 180 See, e.g., Comments of Netflix, Inc., MB Docket 14-90, at 12-13 (filed Sept. 16, 2014) (“Netflix Comments”); Comments of Cogent Communications Group, Inc., MB Docket 14-90, at 2, 9-10 (filed Sept. 16, 2014) (“Cogent Comments”); Franken Comments at 1-2; WGAW Petition at 14-15, 18-20; WGAW Reply at 14-16, 24. 181 Application at 75-76; Katz Decl. ¶¶ 45-54. 182 Katz Decl. ¶¶ 50, 52-54. 183 Application at 76-79; Katz Decl. ¶¶ 50-51; see also WGAW Petition at 14-15, 18, 20; WGAW Reply at 14-16, 24 (agreeing that OVD services are complementary to MVPD services, but stating that OVDs are increasingly becoming more viable alternatives to MVPD services). 184 Free Press Petition at 16-19. Free Press asserts that online video can be accessed only through a broadband connection available from a facilities-based broadband provider, such as AT&T, and that, as a result, the future viability of online video competition is in doubt. Id. at 17. 185 Id. at 19. 186 See Comcast-NBCU Order, 26 FCC Rcd at 4255-56, ¶ 40; Liberty Media-DIRECTV Order, 23 FCC Rcd at 3280- 81, ¶ 31; Adelphia Order, 21 FCC Rcd at 8235, ¶ 63; News Corp.-Hughes Order, 19 FCC Rcd at 501, ¶ 53. Federal Communications Commission FCC 15-94 30 Comcast-NBCU Order, we find that, for most consumers today, OVD services are not substitutes for MVPD services.187 Rather, as we note in our description of current industry conditions discussed above,188 OVDs typically offer consumers choices that may either complement their MVPD services or compete with some portion of the services MVPDs offer, such as VOD. Indeed, despite the increased number of OVDs and increased use by consumers of OVD services, we do not have evidence on the record that any OVD would be, in the near term, a disciplining force if the combined entity were to increase price or decrease quality. However, given the development of additional and new OVD services and the proliferation of new technologies and devices that allow consumers to view video programming sold by OVDs on their computers, phones, and televisions,189 we acknowledge that OVDs have the potential to become substitutes for MVPD services with a market presence that is sufficient to counter effectively an increase in price or decrease in quality by the combined entity.190 Therefore, as we analyze the competitive effects of the transaction, we consider any potential competitive harms that may arise from the transaction that would delay or minimize entry of OVDs into the market. Geographic Market 69. We consider the relevant geographic market consistent with Commission precedent and the analytical framework and principles outlined by the 2010 DOJ/FTC Horizontal Merger Guidelines. 191 In previous transactions, the Commission defined the relevant market as local.192 187 See Comcast-NBCU Order, 26 FCC Rcd at 4269, ¶ 79; see also United States v. Microsoft, 253 F.3d at 52-54 (excluding “middleware” software from the definition of the relevant product market because of its present non- interchangeability with Windows, notwithstanding its long-term future potential). In the Comcast-NBCU Order, the Commission found that instances of consumers replacing MVPD services with OVD services were “relatively infrequent.” Comcast-NBCU Order, 26 FCC Rcd at 4269, ¶ 79. Consumers may choose to cancel their MVPD service (“cord cutters”), reduce their MVPD spending (“cord shavers”), or forego subscribing to an MVPD service in the first place (“cord nevers”). While observers differ on the degree to which these behaviors are occurring today, estimates continue to be relatively small. See Sixteenth Annual Report, 30 FCC Rcd at 3395-97, ¶¶ 301-304. Moreover, the vast majority of consumers who watch video programming from OVDs also subscribe to an MVPD video service, indicating that consumers consider the services to be complements, rather than substitutes. Sixteenth Annual Report, 30 FCC Rcd at 3289, ¶ 84, 3352-53, ¶ 215; see also Fifteenth Annual Report, 28 FCC Rcd at 10557- 58, ¶ 132 (noting that Netflix, the largest online video subscription service, has reported that the overwhelming majority of its subscribers also subscribe to an MVPD service and view the products as complementary). 188 See supra ¶ 58. 189 See supra ¶ 58. 190 We note further that the Commission is currently seeking comment on the interpretation of the statutory definition of “MVPD” and on whether that definition includes certain Internet-based distributors of video programming. See MVPD Definition NPRM, 29 FCC Rcd 15995. As the Commission has stated, however, that proceeding will not define or opine on which services or providers are in the same relevant product market as a service designated as an MVPD. Id. at 16002, ¶ 15 n.33. 191 See, e.g., Comcast-NBCU Order, 26 FCC Rcd at 4256-57, ¶ 42; Liberty Media-DIRECTV Order, 23 FCC Rcd at 3281, ¶ 32; Adelphia Order, 21 FCC Rcd at 8235-36, ¶ 64; News Corp.-Hughes Order, 19 FCC Rcd at 505, ¶ 62; EchoStar-DIRECTV HDO, 17 FCC Rcd at 20610, ¶ 119. The 2010 DOJ/FTC Horizontal Merger Guidelines define a relevant geographic market as the region where a hypothetical monopolist that is the only producer of the relevant product in the region could profitably impose at least a “small but significant and non-transitory” increase in the price of the relevant product, assuming that the prices of all products provided elsewhere do not change. See 2010 DOJ/FTC Horizontal Merger Guidelines § 4.2.1 at 13. In cases where price discrimination based on customer location is feasible, the 2010 DOJ/FTC Horizontal Merger Guidelines state that geographic markets may be defined based on the locations of customers, rather than the locations of suppliers. Id. § 4.2.2 at 14-15. 192 See, e.g., Comcast-NBCU Order, 26 FCC Rcd at 4256-57, ¶ 42; Liberty Media-DIRECTV Order, 23 FCC Rcd at 3281, ¶ 32; Adelphia Order, 21 FCC Rcd at 8235-36, ¶ 64; News Corp.-Hughes Order, 19 FCC Rcd at 505, ¶ 62; EchoStar-DIRECTV HDO, 17 FCC Rcd at 20610, ¶ 119. Federal Communications Commission FCC 15-94 31 70. Positions of the Parties. Free Press, Public Knowledge and Institute for Local Self- Reliance (“Public Knowledge-ILSR”), and WGAW argue that the relevant geographic market for this transaction is local for MVPD services.193 Free Press also contends that the appropriate geographic market is smaller than the Designated Market Area (“DMA”)194 level,195 which the Commission has used to identify local markets in prior transactions.196 Free Press argues that analysis of the transaction at the DMA level understates the competitive effect of the instant transaction because U-verse video typically is not offered throughout an entire DMA; rather AT&T typically serves only portions of certain DMAs.197 71. Discussion. Consistent with past practice and the record before us, for the purposes of analyzing market concentration issues raised by commenters in the record, we define the relevant geographic market as a local market where consumers face similar choices for MVPD services. Consumers make decisions based on the MVPD services available to them at their residences, as they are unlikely to move in order to change providers.198 In previous transactions involving MVPDs, the Commission defined the relevant geographic market for MVPD services as the franchise area of the local cable operator.199 In this transaction, neither Applicant is tied to traditional cable franchise areas. DIRECTV’s satellite network has a nationwide footprint. AT&T’s U-verse video service is available in discrete geographic areas and overlaps areas served by incumbent cable operators, but unlike such cable providers, AT&T is not registered with the Commission as a cable system and therefore does not serve prescribed franchise areas.200 Accordingly, we do not find the franchise area of the local cable operator to be a useful approximation of the local market for the purposes of evaluating market concentration that may result from this transaction. Moreover, our analysis and the record evidence confirm that AT&T does not serve entire DMAs and indeed faces different competitors in different parts of any given DMA.201 Therefore, a DMA is not the ideal geographic area for analyzing changes in concentration that result from this transaction. We find that for this transaction the relevant local geographic area is where 193 Free Press Petition at 9-16; Public Knowledge-ILSR Petition at 5-6; WGAW Reply at 4-5. 194 A DMA is a Nielsen-defined television market consisting of a unique group of counties. The United States is divided into 210 DMA markets. Nielsen identifies television markets by placing each U.S. county (except for certain counties in Alaska) in a market based on measured viewing patterns and by MVPD distribution. See Sixteenth Annual Report, 30 FCC Rcd at 3274-75, ¶ 45 n.122. 195 Free Press Petition at 9-11, 14-16. 196 See, e.g., News Corp.-Hughes Order, 19 FCC Rcd at 506, ¶ 65 (“[W]e conclude that in the case of broadcast television programming, it is reasonable to use DMAs to define the relevant geographic market for each individual broadcast station.”). 197 Free Press Petition at 9-11, 14-16. 198 See, e.g., Comcast-NBCU Order, 26 FCC Rcd at 4256-57, ¶ 42; Liberty Media-DIRECTV Order, 23 FCC Rcd at 3281, ¶ 32; Adelphia Order, 21 FCC Rcd at 8235-36, ¶ 64; News Corp.-Hughes Order, 19 FCC Rcd at 505, ¶ 62; EchoStar-DIRECTV HDO, 17 FCC Rcd at 20610, ¶ 119. 199 See, e.g., id. 200 See Sixteenth Annual Report, 30 FCC Rcd at 3263-64, ¶ 27; Fifteenth Annual Report, 28 FCC Rcd at 10507, ¶ 28. A majority of AT&T’s U-verse video footprint falls within the states that have statewide video franchising laws. Compare Fifteenth Annual Report, 28 FCC Rcd at 10516-17, ¶ 41 n.100 and Sixteenth Annual Report, 30 FCC Rcd at 3270, ¶ 36 n.90 (identifying states that had adopted statewide video franchising laws), with AT&T Response to Sept. 9, 2014, Information Request, Exhibit 5.a.1 (identifying DMAs in which AT&T provides MVPD service). 201 There are [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] DMAs in which the U-verse footprint covers [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See AT&T Response to Sept. 9, 2014, Information Request, Exhibit 5.a.1; SNL Kagan, 2014Q3 TV Household (Projections), January 2015 (“SNL Kagan 2014Q3 TV Household Projections”). Federal Communications Commission FCC 15-94 32 AT&T has deployed its U-verse network and customers have similar choices of MVPD providers.202 We rely on county level data as a proxy of these local areas because, of the data that are available to the Commission, county data offer a reasonable approximation. Market Participants 72. Positions of the Parties. The Applicants claim that, in addition to traditional MVPDs, other sources of video competition continue to emerge, primarily OVDs.203 Free Press argues the Applicants overstate the level of local MVPD competition from online video services.204 As noted above, Free Press dismisses the Applicants’ claim that online video services are substitutes for MVPD service.205 73. Discussion. Within the relevant product market of multichannel video programming service distributed by MVPDs, we conclude that the market participants are cable operators, DBS providers, telephone MVPDs, and overbuilders.206 The record indicates that the primary providers of MVPD services in the vast majority of the U-verse video footprint are the two DBS operators, DIRECTV and DISH, and an incumbent franchised cable operator.207 In portions of the U-verse video footprint, a cable overbuilder or other wireline provider may also offer MVPD service.208 We find that these other MVPDs compete in at least some relevant geographic markets. As discussed above, for the purposes of evaluating the record related to market concentration, we do not include OVDs in the market, but recognize that they are growing in competitive importance.209 We also note that there is no evidence in the record that suggests there are other MVPDs that would likely enter the market and obtain sufficient share to counteract the competitive effects of the transaction.210 While entry may be possible, we find that there appear to be several significant barriers to rapid entry in the MVPD market.211 Increase in Concentration 74. Concentration in the relevant markets can be one indicator of the likely competitive effects of a proposed merger. Consistent with Commission precedent, as well as the 2010 DOJ/FTC Horizontal Merger Guidelines, in order to respond to comments filed in this proceeding, we examine the 202 The Applicants’ merger simulation, discussed in our competitive effects analysis, Section IX.A.5, relies on a DMA-level analysis, and so we rely on DMAs when analyzing that economic data. 203 Application at 74-79. 204 Free Press Petition at 17-19. 205 Id. 206 See supra ¶¶ 55, 68. 207 Application at 72-73. 208 Id. at 73-74. For example, AT&T states that it faces competition from Grande Communications in Texas and RCN in New York and Chicago. Id. at 73-74 n.262. 209 See supra ¶ 68. 210 The 2010 DOJ/FTC Horizontal Merger Guidelines include as market participants firms that are considered “rapid entrants.” Rapid entrants are firms that are not current producers in a relevant market but that are very likely to provide a rapid response to a price increase as a result of a merger, without incurring significant sunk costs. 2010 DOJ/FTC Horizontal Merger Guidelines § 5.1 at 15-16. 211 See, e.g., Sixteenth Annual Report, 30 FCC Rcd at 3269-83, ¶¶ 35-64 (discussing entry conditions for MVPDs). See also AT&T Response to Sept. 9, 2014, Information Request at 115-120; MVPD Definition NPRM, 29 FCC Rcd 15995. Federal Communications Commission FCC 15-94 33 post-transaction market concentration and the change in market concentration that is likely to result from the transaction.212 75. Market concentration is often measured by the Herfindahl-Hirschman Index (“HHI”).213 Under the 2010 DOJ/FTC Horizontal Merger Guidelines, the DOJ and Federal Trade Commission (“FTC”) consider a market with an HHI that exceeds 2,500 to be “Highly Concentrated.”214 Moreover, the 2010 DOJ/FTC Horizontal Merger Guidelines indicate that a merger resulting in a Highly Concentrated market (i.e., a post-merger HHI that exceeds 2,500) and producing an increase in the HHI of more than 200 points “will be presumed to be likely to enhance market power.”215 “The presumption may be rebutted by persuasive evidence showing that the merger is unlikely to enhance market power.”216 However, market concentration measures are the beginning, not the end, of the competitive analysis.217 76. Positions of the Parties. Free Press and Public Knowledge-ILSR argue that the proposed transaction would harm video consumers in the geographic areas where AT&T’s U-verse video product and DIRECTV’s satellite video service overlap by increasing concentration well beyond levels that raise antitrust concerns.218 They claim that the proposed transaction would violate the presumption articulated in the 2010 DOJ/FTC Horizontal Merger Guidelines in 64 DMAs, where approximately 99 percent of AT&T’s video subscribers and half of the U.S. population reside, with an average HHI increase of 450 points and an average HHI above 3,300.219 Furthermore, Free Press contends that an HHI analysis at the DMA level understates the competitive effect of the proposed transaction because U-verse video is not offered throughout an entire DMA that AT&T serves.220 Free Press claims that when the scope of review 212 See EchoStar-DIRECTV HDO, 17 FCC Rcd at 20614, ¶ 133; 2010 DOJ/FTC Horizontal Merger Guidelines § 5.3 at 18 (“In evaluating market concentration, the Agencies consider both the post-merger level of market concentration and the change in concentration resulting from a merger.”). 213 The HHI is calculated as the sum of the squares of the market shares of each firm participating in a relevant market. The HHI can range from nearly zero in the case of an atomistic market to 10,000 in the case of a pure monopoly. Because the HHI is based on the squares of the market shares of the participants, it gives proportionately greater weight to market participants with larger market shares. Changes in market concentration are measured by the change in the HHI. See 2010 DOJ/FTC Horizontal Merger Guidelines § 5.3 at 18-19. To assess whether the increase in horizontal market concentration is significant or not, we consider the absolute level of the post- transaction HHI, a widely utilized measure of market concentration, as well as the change in the HHI. See infra ¶¶ 78-81. 214 2010 DOJ/FTC Horizontal Merger Guidelines § 5.3 at 19. 215 Id. 216 Id. 217 See, e.g., WorldCom-MCI Order, 13 FCC Rcd at 18049-50, 18100-01, ¶¶ 39, 135 (stating that an HHI analysis alone is not conclusive, but it provides guidance on potential anticompetitive effects of a merger); see also 2010 DOJ/FTC Horizontal Merger Guidelines § 5.3 at 18 (“Market shares may not fully reflect the competitive significance of firms in the market or the impact of a merger. They are used in conjunction with other evidence of competitive effects.”). Such measures may potentially identify mergers that may raise competitive concerns and, therefore, warrant more in-depth analysis of the potential anticompetitive effects. 2010 DOJ/FTC Horizontal Merger Guidelines § 5.3 at 18-19. 218 Free Press Petition at 6; Public Knowledge-ILSR Petition at 5-6; see also WGAW Reply at 8 (citing the arguments made by Free Press and Public Knowledge-ILSR). 219 Free Press Petition at 9-14; see also Public Knowledge-ILSR Petition at 5-6. Free Press asserts that in 61 DMAs, the transaction would increase the market’s HHI value by more than 200 points and result in a post-transaction HHI above 2,500. In the remaining three DMAs, the HHI would increase by between 100 and 200 points, resulting in a post-transaction HHI above 2,500. Free Press Petition at 12-13 n.23. 220 Free Press Petition at 9-11, 14-15. Federal Communications Commission FCC 15-94 34 is narrowed to just the U-verse footprint, the HHI increases to nearly 4,000.221 Free Press argues that for almost a quarter of the country, the proposed transaction would reduce the number of pay-TV competitors from four to three and the level of market concentration post-transaction would be higher in this case than after consummation of AT&T’s bid for T-Mobile USA, Inc. (“T-Mobile”) in 2011, which the DOJ filed suit to block.222 77. The Applicants state that opponents have not provided any evidence to rebut the economic evidence that they have submitted; instead, opponents rely only on “primitive” HHI concentration calculations using MVPD market shares.223 The Applicants dismiss these calculations, noting that market shares and HHI calculations are merely a starting point for any competitive analysis224 and cannot overcome the Applicants’ detailed econometric analyses submitted in the record.225 Additionally, because this transaction involves complementary products, the Applicants argue that an MVPD concentration analysis does not accurately predict the competitive effects of the transaction.226 Instead, the Applicants argue that the transaction necessitates a more in-depth analysis to predict whether the transaction would harm consumers.227 78. Discussion. As our analysis below indicates, using the thresholds from the 2010 DOJ/FTC Horizontal Merger Guidelines, we find that 58 DMAs and 1,109 counties meet the Highly Concentrated definition.228 However, as noted in previous Commission Orders as well as the 2010 DOJ/FTC Horizontal Merger Guidelines, calculations of market concentration measures may not fully reflect the competitive significance of the merging firms or the impact of the transaction. Thus, HHIs are often used in conjunction with other analyses to evaluate potential anticompetitive harms of a merger and we consider them here.229 221 Id. at 15. Free Press’s calculations are based on local geographies within U-verse’s footprint that are smaller than the DMA, but larger than the household level. However, Free Press clarifies that this local geography is not referring to specific local franchise areas. Id. at 15 n.24. 222 Id. at 15-16. In 2011, DOJ filed an antitrust lawsuit to block the merger of AT&T and T-Mobile. Complaint, United States v. AT&T, Inc., No. 1:11-cv-01560 (D.D.C. filed Aug. 31, 2011), available at http://www.justice.gov/sites/default/files/opa/legacy/2011/08/31/Justice-ATT-TMobile-Complaint.pdf (visited June 24, 2015). 223 Joint Opposition at 28-30; Joint Opposition, “An Economic Assessment of AT&T’s Proposed Acquisition of DIRECTV,” Reply Declaration of Michael L. Katz, transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, ¶¶ 51, 59 (filed Oct. 16, 2014) (“Katz Reply Decl.”). 224 Joint Opposition at 28-29. 225 Id. at 30 (citing Katz Reply Decl. ¶¶ 52-59). 226 Id. at 29-30. 227 Id. at 30-31. 228 We note that Free Press uses DMA-level data to calculate market concentration. See Free Press Petition at 11-14. Accordingly, to more accurately compare our calculations with those of Free Press, we find it appropriate to begin our initial HHI analysis by using DMA-level data. In the second part of our analysis, we use county-level data, which reflect the relevant geographic market defined for this transaction. In addition, our analysis of the economic data considers DMA-level data because the Applicants used a DMA-level analysis in the merger simulations submitted in the record. See Appendix C, Analysis of Merger Simulation Models, ¶ 9 (“Appendix C”). 229 2010 DOJ/FTC Horizontal Merger Guidelines § 5.3 at 18. See also Applications of GCI Communication Corp., ACS Wireless License Sub, Inc., ACS of Anchorage License Sub, Inc., And Unicom, Inc. for Consent To Assign Licenses To The Alaska Wireless Network, LLC, WT Docket No. 12-187, Memorandum Opinion and Order and Declaratory Ruling, 28 FCC Rcd 10433, 10450-51, ¶¶ 42-43 (2013) (conducting a two-part analysis, an initial screen using HHI and a case-by-case market analysis, to determine the competitive effects of a proposed joint venture of two wireless providers); Applications of AT&T Inc. and Cellco Partnership d/b/a/ Verizon Wireless, WT (continued….) Federal Communications Commission FCC 15-94 35 79. Using data available to the Commission, we are able to calculate230 the post-transaction HHI and the change in the HHI for 74 DMAs within the U-verse video footprint.231 Under the thresholds from the 2010 DOJ/FTC Horizontal Merger Guidelines, 58 of the 74 DMAs meet the Highly Concentrated definition.232 80. However, as discussed above, because AT&T often offers service to only part of a DMA, a DMA is too large a geographic market in which to measure market concentration. Accordingly, we rely on county level data, which is a reasonable approximation of the local geographic market. According to our analysis, AT&T’s U-verse network covers at least part of 1,159 counties in parts of 74 DMAs.233 The post-transaction HHI for these counties ranges from 1,971 to 7,827 and the change in the HHI ranges from 0.362 to 3,745.234 There are 1,109 counties that meet the Highly Concentrated definition.235 81. Our analysis indicates that the proposed transaction would increase concentration in the video distribution market in certain geographic areas affecting 24 percent of U.S. households.236 (Continued from previous page) Docket No. 09-104, Memorandum Opinion and Order, 25 FCC Rcd 8704, 8724-27, ¶¶ 42-48 (2010) (“AT&T- Verizon Wireless Order”) (similarly conducting a two-part analysis, an initial screen using HHI and a case-by-case market analysis, to determine the competitive effects of the proposed transaction). 230 To calculate market share in DMAs and in the counties within those DMAs, we utilized data from SNL Kagan’s MediaCensus. MediaCensus provides subscriber data for third-party information, and it is commonly used by the Commission and outside parties for media analysis. Therefore, we find it reasonable to rely on data from MediaCensus for our analysis herein. 231 The HHI for these DMAs range from 1,811 to 4,705, and the change in the HHI ranges from 0.46 to 1,019. To identify DMAs within the U-verse video footprint, we relied on data submitted by the Applicants. See AT&T Response to Sept. 9, 2014, Information Request, Exhibit 5.a.1. To conduct the HHI analysis, we retrieved subscriber counts for all U.S. video providers on a DMA level using MediaCensus for the DMAs AT&T identified as being within the U-verse video footprint. We aggregated the total video subscribers on a DMA level by adding the subscriber counts per provider in each unique DMA. We then calculated individual provider market share by dividing each provider’s subscriber count, for each market, by the total DMA video subscriber count. According to the Applicants, AT&T video’s service overlaps with DIRECTV’s in 77 DMAs. See AT&T Response to Sept. 9, 2014, Information Request, Exhibit 5.a.1. MediaCensus did not report any U-verse subscriber projections for [BEGIN CONF. INFO.] [END CONF. INFO.] . However, our analysis finds that the number of homes passed by U-verse video in these DMAs is negligible and that AT&T no longer provides U-verse service in the Hartford-New Haven, CT, DMA. Therefore, our findings are consistent with MediaCensus’s report of no U-verse subscriber projections in these DMAs. See AT&T Response to Sept. 9, 2014, Information Request, Exhibit 5.a.1; SNL Kagan 2014Q3 TV Household Projections; AT&T Inc., AT&T Services, AT&T U-verse, Availability, U-Verse in Connecticut, http://www.att-services.net/att-u-verse/availability/uverse-connecticut.html (visited June 18, 2015) (noting that effective Oct. 25, 2014, the state of Connecticut is no longer part of the AT&T local service territory). Accordingly, our analysis utilizes data for 74 DMAs and excludes the three DMAs for which MediaCensus did not have U-verse subscriber projections. SNL Kagan, MediaCensus (Projections) 2014Q3, January 2015 (“MediaCensus 2014Q3”). 232 See MediaCensus 2014Q3. See also 2010 DOJ/FTC Horizontal Merger Guidelines § 5.3 at 19. 233 See MediaCensus 2014Q3. To conduct the HHI analysis at the county level, we retrieved subscriber counts for all U.S. video providers on a county level using MediaCensus. We aggregated the total video subscribers on a county level by adding the subscriber counts per provider in each unique county. We then calculated individual provider market share by dividing each provider’s subscriber count, for each county, by the total county video subscriber count. We analyzed only the same 74 DMAs previously used in our HHI DMA analysis above. See supra n.231. 234 See MediaCensus 2014Q3. 235 See id.; see also 2010 DOJ/FTC Horizontal Merger Guidelines § 5.3 at 19. 236 See SNL Kagan, MediaCensus (Projections) 2014Q4, April 2015 (“MediaCensus 2014Q4”); Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, Attachment (continued….) Federal Communications Commission FCC 15-94 36 However, given the complexities of the video services industry and the evidence in the record, in the subsequent section we analyze the competitive effects of the transaction using economic analysis and other record evidence. IX. HORIZONTAL EFFECTS ANALYSIS 82. In the sections below, we discuss the economic analysis submitted by the Applicants and documentary evidence related to the harm from a loss of competition between AT&T and DIRECTV. A. Evaluation of Potential Unilateral Effects Using Economic Analysis 83. We first evaluate the potential and magnitude of any unilateral effects. Horizontal transactions such as the combination of AT&T’s and DIRECTV’s MVPD services, in which rival firms are combining, raise potential competitive concerns when the combined entity has the incentive and the ability to raise prices, lower quality, or otherwise harm competition in a relevant market.237 If consumers consider products offered by firms other than those offered by the merging entities to be close substitutes, the combined firm is unlikely to substantially raise its product’s price because consumers would respond by switching to a substitute product.238 84. Unilateral effects arise when firms, regardless of the anticipated actions or responses of other firms, find it profitable to raise prices or otherwise exercise market power following a horizontal merger.239 When a merger combines two firms selling products that substitute for each other, there is an increased incentive for the combined entity to unilaterally raise the price of one or both firms’ offerings above the pre-merger level.240 The degree of direct competition or substitution between the merging parties and whether there are non-merging parties that are close substitutes (or that could quickly reposition themselves to be close substitutes) are important factors in determining the likelihood and magnitude of any potential price increase as a result of such unilateral effects.241 85. The Applicants maintain that AT&T’s broadband product and DIRECTV’s standalone video service are complementary products.242 According to the Applicants, the transaction would allow (Continued from previous page) (July 7, 2014) (submitting additional details on the number of (1) customer locations and subscribers for FTTP and FTTN technologies and (2) consumer and business customer subscribers). In addition, while we note that WGAW asserts that “consumers in 129 designated market areas (‘DMAs’) will lose a competing MVPD if this merger is approved” (WGAW Petition at 4; see also WGAW Reply at 1, 3), based on SNL Kagan data, the information submitted by the Applicants, and our own analysis, we find that AT&T and DIRECTV both offer MVPD service in portions of only 74 DMAs, as detailed above. 237 See 2010 DOJ/FTC Horizontal Merger Guidelines § 1 at 1-2, § 6 at 20-24. 238 See Liberty Media-DIRECTV Order, 23 FCC Rcd at 3280, ¶ 28; Adelphia Order, 21 FCC Rcd at 8234, ¶ 59; Applications for Consent to the Transfer of Control of Licenses from Comcast Corporation and AT&T Corp., Transferors, to AT&T Comcast Corporation, Transferee, Memorandum Opinion and Order, 17 FCC Rcd 23246, 23260, ¶ 41 (2002) (“Comcast-AT&T Order”); 2010 DOJ/FTC Horizontal Merger Guidelines § 6.1 at 22. 239 See EchoStar-DIRECTV HDO, 17 FCC Rcd at 20619, ¶ 152; 2010 DOJ/FTC Horizontal Merger Guidelines § 1 at 2, § 6 at 20-24. 240 See EchoStar-DIRECTV Order, 17 FCC Rcd at 20624, ¶ 169; 2010 DOJ/FTC Horizontal Merger Guidelines § 6.1 at 20; Joseph Farrell & Carl Shapiro, Antitrust Analysis of Horizontal Mergers: An Economic Alternative to Market Definition, 10 B.E. J. OF THEORETICAL ECON. 1 (Policies and Perspectives), Art. 9, at 6 (2010). 241 See Comcast-NBCU Order, 26 FCC Rcd at 4286-87, ¶ 119 n.287; 2010 DOJ/FTC Horizontal Merger Guidelines § 6.1 at 20-22; see also United States v. H&R Block, 833 F. Supp. 2d 36, 81 (D.D.C. 2011) (stating that unilateral effects in a differentiated product market are likely to be profitable where the products controlled by the merging firms are close substitutes, products offered by non-merging firms are sufficiently different to make a small but significant and non-transitory price increase profitable for the merging firms, and non-merging firms are unlikely to reposition their products to offer close substitutes for the products offered by the merging firms). 242 Joint Opposition at 27-30. Federal Communications Commission FCC 15-94 37 the combined entity to offer consumers better, integrated bundles of these complementary products.243 The Applicants argue that the transaction would intensify, not harm, MVPD competition.244 Furthermore, the Applicants contend that their detailed economic analysis supports the assertion that consumers would benefit from the transaction, including a simulation of the effect of this transaction.245 In the section below, we present our findings on that economic analysis.246 1. Merger Simulation 86. Merger simulations may be used to evaluate potential unilateral effects, that is, the effects based primarily upon the elimination of competition between the merging parties following the transaction.247 Generally, merger simulations evaluate the degree to which, following a merger of competitors, the merging parties and their rivals would increase their prices and whether potential cost savings would preclude any price increases. Merger simulation methods need not rely on market definition, but they often include independent price responses by non-merging parties.248 The question posed in any merger simulation is essentially: “Assuming that all industry participants’ product offerings remain the same, what price changes arise from the changed pricing incentives created by the proposed transaction?” 87. The Applicants submitted two merger simulations – a simulation submitted by Dr. Michael Katz (the “Katz Simulation”) and a simulation submitted by Professors Steve Berry and Phil Haile (the “BH Simulation”).249 The Applicants assert that these merger simulations demonstrate that the 243 Id. at 28; Lee Decl. ¶ 25. 244 Joint Opposition at 27-32. 245 Id. at 28-32. 246 For a more detailed discussion of our economic analysis, see generally Appendix C. 247 2010 DOJ/FTC Horizontal Merger Guidelines § 6.1 at 21; EchoStar-DIRECTV Order, 17 FCC Rcd at 20621-24, ¶¶ 158-169 and Appendix E, “Merger Simulations of the EchoStar-DIRECTV Merger”; see United States v. H&R Block, 833 F. Supp. 2d at 88. 248 2010 DOJ/FTC Horizontal Merger Guidelines § 6.1 at 21. 249 Katz Decl. ¶¶ 85-95; Katz Reply Decl. ¶¶ 52-55; Compass Lexecon, “Additional Detail on the Demand Estimation, Merger Simulation, and Investment Model Analysis Performed by Professor Katz” ¶¶ 5-46 (“Katz Additional Detail”), transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (filed July 28, 2014); Steve Berry & Phil Haile, “Quantitative Analysis of an AT&T-DIRECTV Merger” (“Berry-Haile Analysis”), transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (filed July 17, 2014); Steve Berry & Phil Haile, “Quantitative Analysis of an AT&T-DIRECTV Merger: Updated Results” (“Berry-Haile Updated Results”), transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (filed Sept. 23, 2014); Steve Berry & Phil Haile, “Quantitative Analysis of an AT&T-DIRECTV Merger: Additional Discussion of Modeling Choices, Data, and Results” (“Berry-Haile Additional Discussion”), transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (filed Sept. 23, 2014). Dr. Katz compares his simulation with the BH Simulation, noting the improvements that Professors Berry and Haile made. See Katz Reply Decl. ¶¶ 56-57. Federal Communications Commission FCC 15-94 38 proposed transaction would have a net positive effect on overall consumer surplus,250 even without considering the transaction’s projected operational efficiencies and other benefits.251 88. The Applicants claim that the two merger simulations are closely related,252 and after examining both, we agree. Our evaluation of the Applicants’ merger simulations focuses on the BH Simulation because it is more detailed and economically rigorous. In doing so, we make a number of additions and corrections to the BH Simulation. We base our findings on this “Modified Simulation.” We note that no party has submitted comments on either of the Applicants’ merger simulations.253 89. The BH Simulation considers the post-transaction changes in the prices different providers charge for their standalone video, a video/broadband bundle, and standalone broadband services.254 This framework reflects that currently: (1) AT&T and DIRECTV each offer to subscribers standalone video services and a bundle of video and Internet service (AT&T with an integrated or synthetic bundle and DIRECTV with a synthetic bundle) and (2) AT&T offers a standalone broadband service. 90. We conduct an economic analysis of whether the AT&T-DIRECTV transaction is likely to result in a unilateral price increase post-transaction based on the BH Simulation. Below we discuss briefly the structure of the BH Simulation, our modifications, and the results for standalone video services, video and broadband bundles, and standalone broadband services. The merger simulation quantifies the net result of three primary effects of the transaction that may influence the prices of the services offered by the firms: (1) the “horizontal effect” from the loss of a competitor in the markets where AT&T and DIRECTV both offer video services, (2) a “bundle effect” that results from the synthetic bundle of AT&T broadband and DIRECTV video being priced jointly by a single firm, and (3) the effect of claimed reductions in the payments made by AT&T to video programmers. We discuss the relative magnitude of these three effects and their impact on consumer surplus as evidenced by the merger simulations. 2. BH Simulation Structure and the “Modified Simulation” 91. The BH Simulation combines an estimation of consumer demand for each of a variety of video-broadband bundles, standalone broadband, and standalone video products, with assumptions about firms’ pricing decisions of all of these products to predict the effects of the transaction on prices and consumer welfare.255 It is important to note at the outset that this merger simulation differs from typical merger simulations in that it includes a variety of services, not all of which compete with each other 250 The change in consumer surplus can be interpreted as the additional amount of money that each consumer would have to pay each month following the merger to make him or her indifferent between the merger occurring and not occurring. Thus, a positive change in consumer surplus – which implies that each consumer would be indifferent between the status quo and a post-merger world in which he or she has to pay an additional positive amount – indicates that the merger leads to an increase in consumer welfare. For a more detailed explanation, see Appendix C n.118. 251 Katz Decl. ¶¶ 85, 94; Katz Additional Detail ¶ 1; Joint Opposition at 30-32; Katz Reply Decl. ¶¶ 53-55; Berry- Haile Additional Discussion at 3. 252 Berry-Haile Analysis at 11; Joint Opposition at 31; Katz Reply Decl. ¶¶ 2, 24, 56-58 (The merger simulation performed by Professors Berry and Haile confirms Dr. Katz’s merger simulation results). 253 The Applicants submitted the merger simulations in the record and they were available to those parties authorized to view material under our Protective Order. Protective Order, 29 FCC Rcd at 6050-51, ¶ 7. None of the information relied on by the Applicants’ merger simulations is the subject of the court order that precludes access to certain information in the record. See CBS Corp. v. FCC, 785 F.3d 699, 700 (D.C. Cir. 2015). 254 See Katz Decl. ¶¶ 87-88; Katz Additional Detail Table A-16; Berry-Haile Analysis at 99-120; Berry-Haile Updated Results at 100a-118. 255 Joint Opposition at 30-32. For further details on the structure of the BH Simulation, see generally, Appendix C. Federal Communications Commission FCC 15-94 39 directly, and some of which may complement each other. Specifically, the BH Simulation considers that consumers can purchase broadband and video services as standalone products or bundled together.256 92. The BH Simulation makes a series of computations: (1) obtaining subscribership shares and a measure of price from marketplace data; (2) modeling and estimating demand; (3) estimating marginal costs consistent with the assumption that observed prices are those that each seller would separately choose to maximize its own profits given the prices chosen by others; and (4) simulating post- transaction prices offered by all firms when the combining parties maximize profits jointly, rather than separately, given those marginal costs.257 The inputs into these computations include data on prices and product characteristics of offered services (i.e., video, broadband, or a combination of the two). The output of these computations is a set of prices based on the new post-transaction pricing incentives.258 93. The BH Simulation uses data on price, subscriber counts, consumer demographics, and component characteristics.259 The BH Simulation constructs subscriber shares and a price measure for each product.260 These shares and prices, along with product characteristics and consumer demographics, are used to estimate a demand model of consumer willingness to pay for the video, broadband, integrated bundles, and synthetic bundles.261 The demand model assumes that each consumer chooses one product, which can include bundles of video and broadband, and that consumers choose the product that offers the most benefit after taking into account price and product characteristics.262 94. The BH Simulation then estimates the post-transaction prices at which each firm would sell its services.263 Underlying this estimation is an assumption that each firm sets prices that maximize its profits given its marginal costs taking the prices of other firms as given.264 To estimate the post- transaction prices, the BH Simulation first infers the marginal costs for which the previously derived price measure would maximize current profits for each firm, with the assumption that each firm is setting prices independently of the others.265 Using these marginal costs, the BH Simulation estimates the profit- 256 See Appendix C ¶ 11 and Section III.C. The BH Simulation defines products from the point of view of the consumer, and with providers offering components to consumers. Components include standalone video, standalone broadband, and integrated bundles and are best thought of as the services “priced” by a provider. From the consumer’s perspective, a product can be a standalone video service from a single provider, a standalone broadband service from a single provider, an integrated bundle from a single provider – all of which are “components” – or a synthetic bundle of video and broadband components from two separate providers (the price of the latter being the sum of the component prices of the two providers, possibly including a discount if the components are marketed jointly). 257 See id. Section II. 258 See id. Section II.E. 259 See id. ¶¶ 12-13. The BH Simulation estimation is performed at the DMA level. See id. ¶ 9. 260 See id. Section II.B. The BH Simulation relies on a single price measure per product – a quality adjusted price index for each firm and component (which includes integrated bundles). See id. Section II.B.2. 261 See id. Section II.C. 262 See id. ¶ 22. In addition to price and product characteristics, the demand model also controls for average consumer demographics in the DMA such as age, income, and education. The demand model also makes use of what are called nests. Nesting is an important feature that allows substitution patterns between products to be more flexible and enables the data to better guide these substitution patterns. The BH Simulation uses a nested logit model to estimate demand. The BH Simulation uses four nests (bundles, video only, broadband only, and nothing - the outside good) to model demand. See id. ¶¶ 24-25. 263 See id. ¶¶ 29-32. 264 See id. 265 See id. ¶ 29. Federal Communications Commission FCC 15-94 40 maximizing prices for all marketplace participants following the transaction.266 The BH Simulation restricts the price of products (and therefore the prices of underlying components) to be constant across a DMA, as well as accounting for the pre-transaction discount on the synthetic bundles DIRECTV offers with AT&T and other telephone MVPDs and cable providers.267 By maximizing the joint AT&T and DIRECTV profits, the BH Simulation captures the changes in the Applicants’ pricing incentives as a result of the transaction, and it allows prices by all other firms to respond to the changed ownership.268 95. In evaluating the BH Simulation, we found small coding errors in the computer programs submitted by the Applicants.269 We corrected these coding errors and found that these corrections did not significantly change the BH Simulation results, though the post-transaction changes in consumer surplus were slightly lower than those generated by the BH Simulation.270 Further, the BH Simulation relied on third-party price data collected using web scraping (i.e., collected from third-parties’ web advertised pricing, where available). We replaced the web-scraped data for Comcast, Time Warner Cable, and Charter with actual pricing data provided by these companies, which better reflect the prices offered to consumers.271 The variant of the BH Simulation that includes the corrections to the coding errors as well as actual third-party pricing data is referred to as the “Modified Simulation.” We use the Modified Simulation to conduct an economic analysis of the competitive effects of the transaction.272 96. Also important for the analysis are the effects of the BH Simulation’s inclusion of an estimation of programming payment reductions that result from the transaction. These programming payment reductions are the only claimed transaction efficiencies accounted for in the BH Simulation.273 The Applicants estimate that by combining with DIRECTV, AT&T could lower its programming payments by at least [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .274 The BH 266 See id. 267 See id. ¶¶ 30-31. 268 See id. ¶ 32. 269 See id. Section IV.A. 270 See id. ¶ 49 n.95. Correcting these coding errors resulted in consumers being slightly more price sensitive. See id. ¶ 49. 271 See Comcast Response to Jan. 8, 2015, Information Request; Time Warner Cable Response to Jan. 8, 2015, Information Request; Charter Response to Jan. 8, 2015, Information Request; see also Appendix C Section IV.B. 272 See Appendix C ¶ 54. For output from the “Corrected Simulation” and the original BH Simulation, see generally Appendix C. As noted above, no commenters challenged the BH Simulation, which was made available to third parties that signed the Protective Order in this proceeding. Protective Order, 29 FCC Rcd at 6050-51, ¶ 7. While commenters did not have access to certain information that we used to analyze the BH Simulation, see CBS Corp., 785 F.3d at 701-03, our analysis yielded more conservative results than the Applicants’ analysis and so commenters are not disadvantaged by their lack of access. See infra nn.278-279 (discussing the Applicants’ payments for programming). In addition, because our use of this information yielded results less favorable for the Applicants, were we to rely only on information that was available to third parties, we would still find that the transaction is unlikely to cause significant anticompetitive unilateral effects, given the potential for lower prices for consumers as a result of reduced programming payments. Thus, while the additional information informed our analysis, we would reach the same conclusion without it. 273 These efficiencies are marginal cost reductions and do not include other potential public interest benefits. 274 See Application at 36; Katz Decl. ¶ 115; Joint Opposition at 16. AT&T Response to Sept. 9, 2014, Information Request at 244. Dr. Katz estimates that DIRECTV’s programming payments are approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent less than AT&T’s. See Katz Decl. ¶ 115. Federal Communications Commission FCC 15-94 41 Simulation projects that [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent of the reductions in programming payments would be passed on to consumers.275 97. In the BH Simulation, the programming payment reduction is modeled as a flat dollar reduction [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] in estimated marginal costs for products that include the AT&T U-verse video product after the Applicants combine.276 These reductions in programming payments achieved post-transaction are a significant assumption in the BH Simulation and have a large influence on the claimed benefits that are estimated to flow to consumers.277 98. We evaluated the Applicants’ claimed programming payment reductions using data submitted by the Applicants. 278 We calculated programming payments per subscriber per month for AT&T and DIRECTV in 2014.279 Using data submitted in the record, we find that there is almost no difference between (1) our calculation of AT&T’s and DIRECTV’s programming payments and (2) the programming payments calculated by the Applicants to derive their claimed programming payment reductions in the BH Simulation.280 99. However, the approximate [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent difference between AT&T’s and DIRECTV’s per subscriber costs for video programming appears to result from a combination of the following three factors: (1) AT&T’s programming payments on individual channels are typically higher per subscriber per month than DIRECTV’s; (2) AT&T offers more channels on its video service than DIRECTV, which inflates AT&T’s per subscriber payments relative to DIRECTV’s; and (3) the difference between AT&T’s and DIRECTV’s distribution of subscribers across the full range of offered channels, which may depend on 275 AT&T and DIRECTV, White Paper, Content Cost Savings Will Result in Both Improved Profitability and Pass Through to Consumers, at 11-12 (“Content Cost White Paper”), transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (filed Nov. 12, 2014); Katz Reply Decl. ¶ 24. 276 See Appendix C ¶ 37. 277 In the BH Simulation, the consumer surplus typically increases by around $1.00 per household per month if the full programming payment reduction is credited. See id. Section V.A. 278 See AT&T Response to Sept. 9, 2014, Information Request, Exhibit 40 a-c; DIRECTV Response to Sept. 9, 2014, Information Request, Schedule 37. See supra n.272 (discussing access to certain confidential information). 279 Our analysis also compares AT&T’s and DIRECTV’s programming payments with Comcast’s programming payments. For a detailed discussion, see Appendix C Section IV.C., Table 1. See also supra n.272 (discussing access to certain confidential information). 280 Our modifications to the Applicants’ analysis yield a reduction of [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent per subscriber per month, while the Applicants found a reduction of [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent per subscriber per month. See AT&T Response to Sept. 9, 2014, Information Request, Exhibit 40 a-c; DIRECTV Response to Sept. 9, 2014, Information Request, Schedule 37. See also AT&T Response to Sept. 9, 2014, Information Request, Exhibit 69.c.2; ATT-FCC- 01741325, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ; ATT-FCC-01645622, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . These estimates are the differences in the Applicants’ payments in 2014, while their filings state that the difference would be up to [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent in future years. AT&T also claims elsewhere in its filings that it expects the difference between AT&T and DIRECTV programming payments to be [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent prior to the projected closing date of the transaction. See AT&T Response to Sept. 9, 2014, Information Request at 243. We measure the programming payment reduction in the merger simulation [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] relative to the marginal costs coming out of the merger simulation, and not the “actual” costs of video programming. When compared to the marginal costs in the BH Simulation, the programming payment reduction translates to a reduction of approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent, similar to the amount that the Applicants claim would be realized in the long run. Federal Communications Commission FCC 15-94 42 whether a channel is included in a “basic” or “premium” tier.281 The effects of these three factors are considered in our adjustment to programming payment reductions in the BH Simulation.282 100. In determining the appropriate post-transaction reduction in programming payments to build into the merger simulation, one consideration is the implicit assumption in the simulation that the characteristics of the pre- and post-transaction video products offered by each company do not change.283 Programming payment reductions that are the result of changes in the channel lineup offered to subscribers (i.e., dropping channels or placing channels on a higher service tier) could potentially reduce the quality of the video product and lead to ambiguous effects on consumers. In general, consumers would benefit if the reduced programming payments from altered channel lineups were passed through as lower prices, but some consumers could be harmed if they were to lose channels that were once offered or if they are forced to pay higher prices to receive those channels on a higher service tier. Regardless of whether such a change would benefit or harm consumers, the record does not provide any evidence of whether the Applicants intend to change the channel lineup to reduce programming payments. Indeed, the BH Simulation implicitly assumes the characteristics of the products are held constant when considering the effect of programming payment reductions on prices and consumer surplus. 101. Therefore, we estimate the consumer surplus effect in the Modified Simulation under two separate programming payment reductions. We first credit only those reductions that are a result of channel-by-channel [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] in the Modified Simulation (the “Low Programming Payment Reduction” or “Low PPR”). However, we also acknowledge that there may be additional potential programming payment reductions if the channel lineup is changed – that is if the Applicants decide, for reasons not evident or established on the record, to offer a different selection of channels – and therefore we also estimate the consumer effect using the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] difference in programming payment reductions claimed by the Applicants as an upper bound of the potential, consumer surplus effects holding other characteristics constant (the “High Programming Payment Reduction” or “High PPR”).284 102. We estimate that the difference in programming payments that are due to the channel-by- channel difference in licensing fees paid by AT&T and DIRECTV is approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent difference in the overall payments between the two companies. Therefore, in evaluating the effects of the programming payment reductions in the Modified Simulation, we consider programming payment reductions of both the High Programming Payment Reduction [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] and the Low Programming Payment Reduction [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] , the former representing the full difference between AT&T and DIRECTV programming payments,285 and the latter representing only those payments that are not due to channel-by-channel differences in license fees.286 281 See Appendix C ¶¶ 59-67. 282 See supra n.272 (discussing access to certain confidential information). 283 Although the input prices in a merger simulation are adjusted for characteristics of the products, to estimate how prices change as a result of the transaction, the characteristics of each service must be the same in the pre- and post- transaction scenarios. 284 That is, the High Programming Payment Reduction would be the appropriate measure if (1) consumers do not view changes in the channel lineup as changes in product “quality” that reduce surplus and (2) the reductions achieved by changing the channel lineup are passed through to consumers in the form of lower prices. 285 See Appendix C ¶¶ 75-76. 286 [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] See id. ¶ 76. Federal Communications Commission FCC 15-94 43 103. Finally, we acknowledge that the “nested logit” structure used in the BH Simulation is known to pass through a higher percent of transaction efficiencies relative to other merger simulation structures.287 The BH Simulation pass-through rate – the percent of the programming payment reductions that would be passed on to consumers through lower prices – is [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent.288 It is not possible to alter the pass-through rate without adopting a different merger simulation structure that may not capture the significant benefits of the nested logit structure in determining the substitution between products and/or the discrete choices made by subscribers.289 104. However, the Applicants also claim that the BH Simulation estimates demonstrate that there is no harm from the transaction, regardless of whether there are off-setting programming payment reductions that are passed through to consumers, and that any pass-through of programming payment reductions is over and above the benefits that would accrue.290 To support this claim, they rely on the BH Simulation without cost savings to be passed through. Thus, when considering consumer surplus and consumer benefits of the transaction, we also analyze the results from the Modified Simulation assuming no programming payment reductions to compare with the relevant BH Simulation results, as well as to assess potential competitive harms when reductions in programming payments are not passed through to consumers. 3. Effects of the Transaction on Consumers 105. Our economic analysis of the Applicants’ merger simulation finds that the net effect of the transaction on consumers is positive.291 Our finding considers the elimination of competition between U-verse video and DIRECTV’s video service, the benefits of the pricing complementarity of AT&T broadband and DIRECTV video, and the pass-through to subscribers of a certain percentage of programming payment reductions.292 We find that the economic analysis in our record demonstrates that the consumer surplus would increase slightly without accounting for programming payment reductions and would increase more substantially when programming payment reductions are included. 106. Positions of the Parties. The Applicants claim that the transaction would result in a positive net effect on consumer surplus, inside and outside of AT&T’s footprint.293 The Applicants assert 287 See Simon P. Anderson, Andre de Palma & Brent Kreider, Tax Incidence in Differentiated Product Oligopoly, 81 J. PUB. ECON. 173, 173-192 (2001). 288 Content Cost White Paper at 11-12; Katz Reply Decl. ¶¶ 24, 53. 289 Changing the pass-through rate in the BH Simulation would require changing the demand model used in the simulation to another form (e.g., linear demand, Constant Elasticity of Substitution (“CES”) demand, etc.). However, given the structure of the industry, we find that the nested logit model is appropriate, in spite of the higher pass-through rate, because it permits consumers to purchase only a single product from a single firm. In particular, we find that the nested logit demand model appropriately captures the likelihood that most households will subscribe to only a single broadband service. In addition, the nested logit demand model reflects the fact that consumers are more likely to substitute within nests (or services) that share product characteristics (e.g., they will choose among broadband and video services, only video services, or only broadband services versus substituting from video services to broadband services). 290 See Content Cost White Paper at 10-11. For further discussion of the Berry and Haile merger simulation and pass-through of programing payment reductions, see Appendix C Section V.B. 291 The consumer surplus changes estimated by the BH Simulation are limited to those that would arise from unilateral quality-adjusted price changes; the simulation does not account for potential surplus changes from price coordination among providers or from non-price factors, such as improved customer service, installation, VOD integration, and improved video user interfaces. 292 See Appendix C Section V. 293 Application at 83-84; Katz Decl. ¶ 85; Katz Additional Detail ¶ 1, Tables A-16, A-19; Joint Opposition at 31; Katz Reply Decl. ¶¶ 53, 55. Federal Communications Commission FCC 15-94 44 this would be the case, even without accounting for any programming payment reductions.294 The results from the Katz Simulation indicate that, based solely on bundling benefits outweighing increased horizontal concentration,295 the population-weighted average effect of the transaction in AT&T’s U-verse footprint would increase consumer surplus by $0.85 per consumer per month,296 and outside the U-verse footprint would increase consumer surplus by $1.82 per consumer per month.297 107. The Applicants claim that the BH Simulation confirms the Katz Simulation’s consumer surplus estimates.298 The BH Simulation estimates that the transaction would have an annual net positive effect between $940 million and $1.44 billion on consumer surplus, assuming a conservative estimate of programming payment reductions.299 108. Discussion. Based on our analysis of the Modified Simulation, we find that even were there to be no reductions in programming payments, consumer surplus would increase slightly as a result of the transaction, though the effect does not appear to be very significant. The Modified Simulation’s monthly per subscriber consumer surplus without programming payment reductions is estimated to be $0.02, approximately $0.83 less and $0.07 less than the Katz Simulation estimates300 and the BH Simulation estimates, respectively.301 The Modified Simulation indicates that the harm from the increased horizontal concentration of video in DMAs where AT&T video and DIRECTV overlap is approximately offset by the benefits of reduced prices that result from the newly integrated AT&T broadband/DIRECTV video bundle. 109. Including the reduction in payments from the Low Programming Payment Reduction increases the per subscriber per month consumer surplus to $0.51, although this is less than half of the per month consumer surplus of $1.11 that results from using the High Programming Payment Reduction.302 Comparing the results using the High Programming Payment Reduction in the Modified Simulation with results from the BH Simulation with programming payment reductions of the same magnitude, our estimate of the increase in consumer surplus is slightly higher than the change reported in the BH Simulation.303 110. We acknowledge that the change in consumer surplus would be small were the transaction not to lead to programming payment reductions and that merger simulations lack measures with which to test whether this number is statistically different from zero. 294 Katz Decl. ¶ 85; Joint Opposition at 31; Katz Reply Decl. ¶¶ 53, 55. 295 Katz Decl. ¶¶ 89; Katz Reply Decl. ¶ 55. 296 Katz Decl. ¶ 90; Katz Additional Detail Table A-16. 297 Katz Decl. ¶ 92; Katz Additional Detail Table A-19. 298 Joint Opposition at 31; Katz Additional Detail ¶ 4; Katz Reply Decl. ¶¶ 56, 58. 299 Joint Opposition at 31-32; Katz Reply Decl. ¶¶ 56, 58; Berry-Haile Analysis at 118-119; Berry-Haile Additional Discussion at 3, 8-9, 44. 300 See Appendix C Table 3; Katz Decl. ¶ 90; Katz Additional Detail Table A-16. 301 See Berry-Haile Updated Results at 18. Professors Berry and Haile report results for three nesting parameters and one nesting parameter. The comparison between the BH Simulation and the Modified Simulation are for the three nesting parameters and for the case of “All DMAs w/ AT&T Availability.” 302 See Appendix C Table 3. 303 See Berry-Haile Updated Results at 18 (estimating consumer surplus to be $0.94. The comparison between the BH Simulation and the Modified Simulation is for the three nesting parameters and for the case of “All DMAs w/ AT&T Availability.” Federal Communications Commission FCC 15-94 45 4. Competitive Effects of Integrated Bundles 111. Our analysis of the merger simulation also supports a finding that the transaction has a positive effect on the price of bundled products and, as a consequence, offers consumers a more competitive alternative to the cable bundle offered by cable operators. In addition, we find that other operational efficiencies beyond the pricing complementarities estimated in the merger simulation may potentially benefit consumers, including by offering a stronger competitor for bundled services. 112. Positions of the Parties. The Applicants assert that a customer purchasing a synthetic bundle generally pays significantly more than the customer would pay for a comparable integrated bundle.304 The Applicants argue that combining AT&T broadband services and DIRECTV video would benefit consumers by creating an integrated bundle offering that is a better alternative to the cable operators’ bundles.305 113. The Applicants argue that the BH Simulation demonstrates that the transaction would produce downward pressure on the price of the bundle of AT&T’s broadband and DIRECTV’s video and on the price of competing products (e.g., cable bundles and standalone broadband and video products offered by cable operators).306 The Applicants rely on the BH Simulation to predict that the price of an AT&T broadband and DIRECTV video bundle would fall by $8.27 and $7.32 per month under the cases of a zero programming payment reduction and a High PPR, respectively.307 The change to the AT&T integrated bundle would result in an increase in price of $4.19 with no programming payment reductions and decline by $4.48 in the case of a High PPR.308 They also claim that because of the significant downward pressure on the price of cable providers’ video and broadband bundles, those prices would fall by $0.13 and $0.79 per month under the cases of a zero programming payment reduction and a High PPR, respectively.309 114. The Applicants argue that the economic theory of “double marginalization” supports their finding that the synthetic bundle price is higher than the price would be for an integrated bundle.310 The Applicants note that double marginalization occurs when “each independent firm in the supply chain applies its own mark-up” to the product.311 In this case, the Applicants assert that both AT&T and DIRECTV “mark up” the price of their respective components of the synthetic broadband and video bundle.312 In comparison, a bundle of broadband and video offered by a single cable operator has only one price mark-up.313 As a result, the Applicants argue that the prices offered by cable companies are more competitive than the Applicants are able to offer with their synthetic bundle. 304 See Lee Decl. ¶¶ 54-56; Guyardo Decl. ¶¶ 27-29. 305 See Application at 7, 23, 32-33, 63. 306 See Joint Opposition at 31; Berry-Haile Additional Discussion at 3. The Katz Simulation also predicts that this transaction creates downward pressure on prices, which in turn, would lead to a competitive response by rival providers of bundled services. See Application at 6-7, 54-55, 64-65, 83-84; Katz Decl. ¶¶ 87-89, 92. 307 Berry-Haile Updated Results at 101, 110. 308 Id. at 102, 111. 309 Id. 310 See Application at 65-67; Katz Decl. ¶¶ 4, 67-71, 73-77, 82-83; Katz Reply Decl. ¶ 12; Double Moral Hazard White Paper at 5, 14-16. 311 See Double Moral Hazard White Paper at 5. 312 See Application at 66; Guyardo Decl. ¶ 27; see also Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, Attachment (“Bundles Ex Parte Presentation”) at 6-7 (April 21, 2015) (submitting written ex parte presentation on bundled services). 313 See Double Moral Hazard White Paper at 5. Federal Communications Commission FCC 15-94 46 115. In addition, the Applicants assert that the integrated bundle would be more competitive in ways that cannot be quantified in the merger simulation. The Applicants state that currently, synthetic bundle customers must schedule two separate installation appointments and then each company must send out a separate installer.314 The Applicants report that in the vast majority of cases, those installations occur on different days.315 The Applicants also explain that AT&T-DIRECTV synthetic bundle subscribers receive separate bills, which the Applicants assert makes it difficult for consumers to determine whether appropriate bundle discounts have been applied.316 They report that it can take [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] for the bundle discount to appear on the customer’s bill because of the AT&T and DIRECTV internal activation confirmation process.317 The Applicants assert that post-transaction they can consolidate all of these services.318 116. CWA agrees with the Applicants and argues that the proposed transaction would “benefit consumers by exerting pressure to constrain rate increases and offer new and better services.”319 Other commenters support the transaction, contending that it would increase competition for the provision of bundled services.320 Free State believes that consumers would benefit because “bundled offerings by an integrated provider enjoy greater cost efficiencies, enabling integrated providers to offer more competitive pricing.”321 A number of other commenters argue that the transaction would enable services to be bundled in a way that could benefit consumers without a price increase.322 314 See Application at 38; Katz Decl. ¶¶ 36, 104; Bundles Ex Parte Presentation at 3-4. 315 See Joint Opposition at 12 & n.24; DIRECTV Response to Sept. 9, 2014, Information Request at 81; Katz Decl. ¶ 36; Guyardo Decl. ¶ 32; Doyle Decl. ¶ 20; Lee Decl. ¶ 57. The Applicants note that, currently, only [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent of new customers get video and Internet/phone installed on the same day. They also assert that DIRECTV typically schedules installation first, which means that subscribers cannot have their Internet-enabled set-top box connected (and thus the non-linear offerings supported) until the Internet access service is installed by AT&T. See DIRECTV Response to Sept. 9, 2014, Information Request at 81; Guyardo Decl. ¶ 32. In many cases, the task of connecting the set-top box to the Internet falls on the Internet access provider’s technician, who does not have the expertise to install the set-top box. See DIRECTV Response to Sept. 9, 2014, Information Request at 81; Guyardo Decl. ¶ 32; Doyle Decl. ¶ 20; Katz Decl. ¶ 36. See also infra ¶ 150. 316 See Guyardo Decl. ¶ 33; Lee Decl. ¶ 57; Katz Decl. ¶ 36; Joint Opposition at 12; DIRECTV Response to Sept. 9, 2014, Information Request at 82. 317 See Guyardo Decl. ¶ 33; Lee Decl. ¶ 57; see DIRECTV Response to Sept. 9, 2014, Information Request at 82. 318 See Joint Opposition at 15. 319 See CWA Comments at 9. 320 See, e.g., Reply Comments of the Information Technology and Innovation Foundation, MB Docket 14-90, at 2-3 (filed Oct. 16, 2014) (“ITIF Reply”); Letter from Todd Maisch, President, Illinois Chamber of Commerce, to FCC, MB Docket No. 14-90, at 1 (Sept. 12, 2014); Letter from A. Richard Heffron, President, Delaware State Chamber of Commerce, to FCC, MB Docket No. 14-90, at 1 (Sept. 15, 2014); Letter from New Jersey State Chamber of Commerce to FCC, MB Docket No. 14-90, at 1 (Sept. 12, 2014); Letter from Catherine Glover, President, Tennessee Chamber of Commerce and Industry, to FCC, MB Docket No. 14-90, at 1 (Oct. 8, 2014); Letter from Kevin Brinegar, President, Indiana Chamber of Commerce, to Thomas Wheeler, Chairman, FCC, MB Docket No. 14-90, at 1 (Oct. 9, 2014). 321 See Free State Comments at 17. 322 See, e.g., Letter from Brent Wilkes, Executive Director, League of United Latin American Citizens et al., to Thomas Wheeler, Chairman, FCC, MB Docket No. 14-90, at 2 (Oct. 16, 2014); Letter from Mike Beebe, Governor of Arkansas, to Thomas Wheeler, Chairman, FCC, MB Docket No. 14-90, at 1 (Sept. 11, 2014); Letter from Evelyn Lugo, President, South Carolina Hispanic Chamber of Commerce, to FCC, MB Docket No. 14-90, at 1 (Sept. 9, 2014). Federal Communications Commission FCC 15-94 47 117. Conversely, WGAW argues that the new bundles would not be competitive because they contain the “least attractive technologies” and that the benefits from lower prices are only “theoretical … [and] not verifiable.”323 WGAW also expresses concern that “[b]undles also have the effect of tying a consumer to a single provider, which discourages innovation and competition between providers offering Internet and video service.”324 Free Press asserts that the Applicants do not need an integrated bundle to be competitive. Free Press argues that the synthetic bundles are less competitive today because AT&T’s introductory price for broadband is 133 percent higher for DIRECTV’s synthetic bundle than it is for AT&T’s integrated bundle ($34.95 versus $14.95).325 Without the transaction, Free Press maintains that the parties would adjust and offer synthetic bundles that are more attractively priced than the synthetic bundles they currently offer.326 118. In addition, several commenters assert that the Applicants are trying to divert the Commission’s attention from the transaction’s competitive harms by focusing on bundling opportunities.327 Commenters dismiss the Applicants’ claims regarding the benefits of integrated bundles, arguing that there is a market for standalone video service and consumers should have the choice to purchase bundled services from one provider or standalone services from one or multiple providers.328 119. Discussion. Our examination of this merger simulation concludes that the Applicants have adequately demonstrated that the transaction will likely increase the competitiveness of the Applicants’ integrated bundles as an alternative to the cable providers’ bundles.329 Although there may be certain differences between the integrated bundle of AT&T broadband and DIRECTV video compared to an integrated bundle of broadband and video offered by cable operators,330 some consumers may view the integrated bundle of AT&T-DIRECTV to be a better alternative to the integrated bundle offered by a cable provider.331 323 See WGAW Petition at 27-28. Public Knowledge-ILSR argue that the Commission should only consider a future integrated fixed wireless/DBS bundle to be an adequate alternative to cable if it is fully substitutable for the integrated bundle of broadband and video offered by cable operators, a point that they maintain the Applicants have not adequately demonstrated. In addition, Public Knowledge-ILSR argue that the Applicants have not made specific or verifiable pricing commitments with respect to the integrated fixed wireless/DBS bundle. See Public Knowledge- ILSR Petition at 17. 324 See WGAW Petition at 21. 325 See Free Press Petition at 20-21. 326 See id. at 22-23, 34-35. 327 See id. at 19-20; ACM et al. Petition at 5-7; ACM et al. Reply at 1-2; WGAW Reply at 2. 328 See ACM et al. Petition at 6; WGAW Reply at 2, 24. For a description of the Applicants’ claims regarding the potential efficiencies and benefits of bundled offerings, see infra Section XI.B. 329 See Appendix C ¶ 3. Using the BH Simulation and post-transaction prices, the subscriber diversion rates from the cable bundles to the integrated bundles of the combined firm would be [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] under the cases of no PPR, Low PPR and High PPR, respectively. Using the BH Simulation and pre-transaction prices, the subscriber diversion rate from cable bundles to the integrated AT&T bundle and the synthetic AT&T-DIRECTV bundle is [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] , indicating that the bundles of the combined firm are slightly more attractive post-transaction than they were before the transaction and that the attractiveness increases with the reduction in programming payments. The diversion rate is the fraction of subscribers leaving the cable bundle that would subscribe to an integrated bundle of the combined firm, were the price of the cable bundle to increase. The diversions rates are generated as output by the BH and Modified Simulations. 330 See Guyardo Decl. ¶¶ 22-26; WGAW Petition at 28. 331 See Katz Reply Decl. ¶ 17. Federal Communications Commission FCC 15-94 48 120. We also find that the economic analysis submitted by the Applicants supports their assertion that the transaction will likely result in downward pressure on the prices of AT&T-DIRECTV broadband and video bundles, and to a lesser extent, the prices of cable bundles. In the Modified Simulation, the price of the AT&T broadband/DIRECTV video bundle decreases by $2.74 with no programming payment reductions and decreases by $2.20 and $1.38 for Low PPR and High PPR, respectively. This change represents a decrease of 2.2 percent relative to the pre-transaction prices in the simulation when no programming payment reductions are considered, and decreases of approximately 1.8 and 1.1 percent relative to pre-transaction prices when using the Low PPR and the High PPR, respectively. 121. For AT&T’s U-verse bundle, the Modified Simulation calculates a price increase of $1.31 with no programming payment reductions and decrease by $2.93 and $6.70 with the Low PPR and the High PPR, respectively.332 This change represents an increase of 1.1 percent relative to the prices in the simulation with no programming payment reductions and decreases of approximately 2.4 and 5.5 percent with Low PPR and High PPR, respectively.333 For the cable bundle, the Modified Simulation calculates a price decrease of $0.08 with no programming payment reductions and decreases of $0.52 and $0.66 for Low PPR and High PPR, respectively. This change represents an increase of 0.05 percent relative to the prices in the simulation with no programming payment reductions and decreases of approximately 0.34 and 0.66 percent with Low PPR and High PPR, respectively. 122. In the Modified Simulation, as the programming payment reduction increases, the decrease in the price of the AT&T broadband/DIRECTV video bundle post-transaction becomes smaller. On the other hand, the AT&T U-verse bundle price rises post-transaction in the case of no programming payment reductions and then declines as programming payment reductions are introduced. When reductions are set at the Low PPR or the High PPR, the post-transaction price falls below the pre- transaction price.334 123. The differential impact of the programming payment reductions on the post-transaction prices of the bundles of the combined firm is a result of AT&T’s incentive to maximize its profits over all of its products – bundles and standalone products.335 In the merger simulation, programming payment reductions reduce [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . The results of the merger simulation indicate that it is more profitable for the combined firm to slightly increase the 332 The direction of the price changes for the AT&T-DIRECTV bundle and the cable bundle do not differ between the BH Simulation and the Modified Simulation, but there are differences in the magnitude. See Berry-Haile Updated Results at 101-102, 110-111; Appendix C ¶¶ 89-94, Tables 7-9. 333 Notably, all percentage increases in prices reported here are changes relative to the “recentered” hedonic prices in the merger simulation (i.e., they are not increases over an established price). Percentage price changes are only informative to the extent that the simulation price levels roughly correspond to notional real marketplace prices, and they have only limited value, as the simulation is able to identify only level changes in prices. See Appendix C ¶¶ 19-21, 85. As an example, AT&T currently offers an Internet and video bundle on its website (which requires a 12-month commitment and includes the U450 package with HBO) at an introductory price of $109. See AT&T Inc., Bundles, https://www.att.com/shop/bundles.html (visited June 18, 2015). However, the pre-transaction “recentered” hedonic price of the AT&T integrated Internet and video bundle is approximately $121 in both the BH Simulation and the Modified Simulation. The percentage price change predicted by the simulation would differ slightly depending on which price is used as the base price. 334 For the AT&T-DIRECTV bundle, the BH Simulation price falls by $8.27 and $7.32 per month under the cases of no PPR and High PPR, respectively; for the Modified Simulation, for the same two PPR scenarios, the price falls by $2.74 and $1.38. For the AT&T integrated bundle, the BH Simulation price increases by $3.76 and decreases by $4.88 per month under the cases of no PPR and High PPR, respectively; for the Modified Simulation, for the same two PPR scenarios, the price increases by $1.31 and declines by $6.70. 335 For a discussion of the effect of programming payment reductions on standalone video services, see infra ¶¶ 142- 143. Federal Communications Commission FCC 15-94 49 price of the AT&T broadband/DIRECTV video bundle and significantly decrease the price of the bundle with the AT&T video product as the programming payments fall. The increased profits from those that continue to purchase the higher-priced DIRECTV video products and the increased profit from those that purchase the lower-cost AT&T video products (due to less than 100 percent pass-through) may outweigh the loss in profit from those who leave the DIRECTV video products. The transaction also allows the combined entity to partially recapture the profit lost from those subscribers that are diverted from DIRECTV products to AT&T’s video products when prices on those products rise. The combined effects of all price changes results in higher profit for the combined firm in the merger simulation. 124. In addition to the increased consumer surplus, we also find that there are certain other quality improvement benefits of the newly integrated bundle that cannot be captured by a merger simulation. Specifically, we agree with the Applicants that integrated bundles have the potential to improve the quality and competitiveness of the Applicants’ bundle by allowing for single installation, billing, and customer service. 125. We disagree with commenters that argue that the new integrated bundles are not a benefit of the transaction. As discussed above, Free Press argues that the parties do not need to merge to compete in the provision of bundles and that instead AT&T could lower its introductory broadband synthetic bundle price to match the price of AT&T’s own integrated bundle.336 We find that the evidence in the record supports the conclusion that AT&T is more likely to offer discounts for integrated bundles than synthetic bundles that include DIRECTV video, including because AT&T offers better discounts of U- verse video today.337 In addition, economic theory supports our finding that AT&T is unlikely to match the price of broadband offered in a synthetic bundle with the price of broadband offered with its integrated bundle because AT&T captures all of the benefit of increased video sales from lower broadband prices offered with its integrated bundle.338 Conversely, if AT&T lowered its broadband price for the synthetic bundle, then DIRECTV would capture the benefit from the increased video sales instead of AT&T. 126. Further, one of the primary reasons that AT&T and DIRECTV are entering into this transaction is to compete more effectively against the integrated bundle of broadband and video offered by cable operators.339 We find that the Applicants’ provision of integrated bundles will increase competition for bundled services. 5. Reduction of Competition in Video Distribution 127. We recognize that because AT&T and DIRECTV both offer video services, post- transaction, there will be a loss of a video provider within the U-verse video footprint. However, as explained below, we find that this very limited potential for competitive harm, when balanced against the benefits of the transaction, does not require a condition. In addition, we impose certain conditions to create a pathway for new competition from online video providers. These include the deployment of 336 Free Press Petition at 20-23. 337 See ATT-FCC-00729772, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . The Applicants state that there is often a price disparity between the AT&T-DIRECTV synthetic bundle and the integrated bundle offered by a single company. See Lee Decl. ¶ 55; see also Double Moral Hazard White Paper at 14. The Applicants claim that discounts for integrated bundles are often greater such that the price of an “integrated U-verse bundle with a comparable AT&T U-verse IPTV package” for the first 24 months of service “is approximately $575 less than the comparable synthetic AT&T/DIRECTV bundle.” Similarly, the cost of a Comcast integrated bundle for 24 months is “over $375 less than the AT&T/DIRECTV synthetic bundle.” See Lee Decl. ¶ 56. 338 See Katz Reply Decl. ¶¶ 14-16. In addition, the customer experience for integrated bundles is often superior to synthetic bundles. See infra Section XI.B. 339 See Application at 55-68. Federal Communications Commission FCC 15-94 50 more and faster high-speed broadband, as well as imposing certain conditions that ensure AT&T does not favor its own video services through the use of discriminatory usage allowances or other retail terms and conditions. 128. Positions of the Parties. The Applicants acknowledge that both AT&T and DIRECTV offer standalone video services. However, AT&T claims that it does not offer its U-verse service in most of the country and also that, with fewer than 6 million video subscribers, it focuses its video marketing efforts almost exclusively on bundles and is not a significant provider of standalone video.340 Therefore, according to the Applicants, there is little head-to-head competition between AT&T and DIRECTV in standalone video services.341 As support, AT&T notes that more than 97 percent of its video customers buy U-verse video as a bundle with broadband or other services.342 Therefore, the Applicants conclude that there would be little competitive impact from the transaction because: (1) AT&T is not a significant standalone video competitor; (2) AT&T does not provide significant competition to DIRECTV; and (3) consumers would continue to have “numerous” standalone video options in the U-verse region.343 129. The Applicants consider the potential increase in standalone video prices using the BH Simulation. According to the Applicants, the BH Simulation predicts a small price increase for DIRECTV’s standalone video subscribers ranging from less than 2 percent to less than 5 percent depending on the model specifications, and the Applicants argue that, in any case, such increase would be offset by a “standard cost efficiency.”344 The Applicants argue that even this small price increase is overstated because the BH Simulation does not capture important synergies that would put downward pressure on the combined firm’s pricing of DIRECTV video services, nor does it account for likely cost savings and quality improvements that would benefit DIRECTV’s subscribers.345 The BH Simulation reports an average post-transaction price increase of $0.07 for all video subscribers, which the Applicants argue is a negligible change of [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent.346 130. Several commenters express concern that the transaction would result in the loss of a competitor in the markets where AT&T currently offers video service, leading to fewer MVPD choices and higher prices.347 Commenters dismiss the Applicants’ claims regarding the benefits of integrated bundles, arguing that consumers should have the choice to purchase bundled services from one provider 340 See id. at 24, 68-72. AT&T claims that it focuses on providing bundled services because U-verse video service is [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See Lee Decl. ¶¶ 11, 16. 341 See Application at 69-71. 342 See id. at 70; Lee Decl. ¶ 12. AT&T claims that it has approximately 138,000 standalone video customers. See Lee Decl. ¶ 12; Katz Decl. ¶ 80. 343 See Application at 71-74. 344 See AT&T and DIRECTV, White Paper, There is No Basis for Competitive Concern About the Impact on DIRECTV Stand-Alone Video Customers, at 4-5 (“Standalone Video White Paper”), transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (filed Nov. 12, 2014). 345 See Standalone Video White Paper at 5-12. 346 See id. at 15 (stating that “[g]iven that average monthly charges for video service are typically in the neighborhood of [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] , a predicted seven-cent price increase represents a negligible change – [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ”). 347 See, e.g., Free Press Petition at 6-19; Public Knowledge-ILSR Petition at 5-8; MCPC Comments at 19-21; Franken Comments at 7; Cox Petition at 3; NAB Comments at 2-3; ACM et al. Petition at 5-8; ACM et al. Reply at 1-2; WGAW Petition at 4, 9; WGAW Reply at 3-5. Federal Communications Commission FCC 15-94 51 or standalone services from one or multiple providers.348 Free Press also criticizes the Applicants for their assertion that several competitive options would remain post-transaction for standalone MVPD customers.349 In contrast, Free State discounts the harms to competition on the basis that the combined entity would have only 24 percent of the national market for video subscribers.350 131. Discussion. Based on our analysis of the Modified Simulation, we recognize that there is some potential upward pricing pressure on the price for standalone video within AT&T’s U-verse footprint, with the price effect being somewhat dependent on the extent of the programming payment reductions. As described in more detail below, our economic analysis shows that if the programming payment reductions on AT&T’s video service increase, the post-transaction price on DIRECTV standalone video also increases and the post-transaction price for AT&T standalone video decreases. This result is similar to the effect we found on the AT&T broadband and DIRECTV video integrated bundles discussed above. Specifically, the merger simulation demonstrates that it is more profitable for the combined firm to raise the prices on DIRECTV video services in those areas where DIRECTV and U- verse video overlap (and lower prices on AT&T video services) as AT&T’s programming payments fall. This is due, in part, to the fact that a portion of the subscribers who leave DIRECTV video products after the price increases will subscribe to AT&T video products at the now lowered prices (which also have higher margins given less than 100 percent of the reduction in programming payments is passed through to consumers). The profits from subscribers diverted from DIRECTV to AT&T are internalized by the combined firm after the transaction. 132. To inform our assessment of the effects of the reduction in competition between AT&T’s and DIRECTV’s video services, we analyzed the Modified Simulation351 with no programming payment reductions, and we found the decrease in consumer surplus to be insignificant. The Modified Simulation calculates that the monthly per subscriber price would increase for DIRECTV’s standalone video customers by $0.80, roughly less than 1 percent over pre-transaction levels.352 For AT&T’s standalone video customers, the price increase would be $1.57 per month, an increase of approximately 2 percent.353 When accounting for Low Programming Payment Reductions, the price for DIRECTV’s standalone video service increases by $0.97 per month, roughly 1.2 percent over the pre-transaction levels whereas the price for AT&T’s U-verse standalone video decreases by $2.85 per month, approximately 3.8 percent less 348 See ACM et al. Petition at 6; WGAW Reply at 2, 24; see Reply Comments of Cox Communications, Inc. to Joint Opposition, MB Docket 14-90, at v (filed Nov. 5, 2014) (“Cox Reply”). For a description of the Applicants’ claims regarding the potential efficiencies and benefits of bundled offerings, see infra Section XI.B. 349 Free Press Petition at 16-19. 350 Free State Comments at 21-22. 351 The direction of the price changes for the AT&T-DIRECTV standalone video service does not differ between the BH Simulation and the Modified Simulation with and without programming payment reductions, but there are differences in the magnitude. See Berry-Haile Updated Results at 101-102, 110-111; Appendix C ¶¶ 89-94, Tables 7-9. The original BH tables do not show the standalone AT&T video price changes, but examination of the BH Simulation results indicates that the price changes are directionally the same as those of the Modified Simulation. 352 See supra n.333. As an example, in terms of current rate plan prices, the price changes from the Modified Simulation would reflect an approximate 0.9 to 1.4 percent price increase for a DIRECTV Premier plan depending on the amount of the programming payment reduction. See DIRECTV, DIRECTV Entertainment Package Selection, https://www.directv.com/DTVAPP/pepod/configure.jsp#package-section (visited June 24, 2015) (providing prices for DIRECTV packages upon input of zip code). 353 See DIRECTV, DIRECTV Entertainment Package Selection, https://www.directv.com/DTVAPP/pepod/configure.jsp#package-section (visited June 24, 2015). As an example, in terms of current rate plan prices, the price changes from the Modified Simulation would reflect an approximate 11.7 percent price decrease to a 2.7 percent price increase for an AT&T U300 plan depending on the amount of the programming payment reduction. See AT&T Inc., AT&T U-verse, https://www.att.com/shop/en/u-verse.html (visited June 15, 2015) (providing prices for U-verse bundles). Federal Communications Commission FCC 15-94 52 than pre-transaction levels.354 Finally, when the Modified Simulation incorporates the High PPR, the price for DIRECTV’s standalone video service increases by $1.24 per month, approximately 1.5 percent more than the pre-transaction price whereas the price for AT&T’s U-verse standalone video decreases by $6.92 per month, roughly 9.1 percent less than pre-transaction levels.355 133. Given this very small potential increase, compared to the overall benefits of the transaction, and our findings based on the other evidence in the record, we decline to impose a condition that would regulate the prices of standalone DIRECTV video.356 6. Standalone Broadband 134. Currently, customers may purchase AT&T’s broadband Internet access service as a standalone product (i.e., without purchasing AT&T video or phone services). Approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] of AT&T’s residential U-verse broadband customers purchase standalone residential broadband service.357 AT&T offers standalone broadband to customers at prices ranging from $29.95 a month for downstream speeds up to 3 Mbps, $34.95 a month for downstream speeds up to 6 Mbps, $44.95 a month for downstream speeds up to 18 Mbps, and $64.95 a month for downstream speeds up to 45 Mbps.358 In Austin, Texas, Nashville, Tennessee, and Atlanta, where AT&T has launched FTTP with GigaPower, AT&T offers standalone broadband at speeds between 75 Mbps to 1 Gbps per second for prices ranging from $70 a month to $120 a month.359 DIRECTV does not offer its own broadband service.360 135. Positions of the Parties. Commenters express concern that the combined entity would increase the price for standalone broadband in order to encourage consumers to purchase its bundled services.361 Netflix, Inc. (“Netflix”) argues that standalone broadband service would not remain a 354 See Appendix C Table 8. 355 See id. Table 9. 356 The Applicants claim that the number of DIRECTV subscribers with no broadband service is small [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See Standalone Video Customers White Paper at 12-13. 357 See AT&T Response to Sept. 9, 2014, Information Request, Exhibit 5.g.2; Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (April 21, 2015) (submitting AT&T’s Form 477 data filings for the periods ended June 30, 2013, and December 31, 2013) (“AT&T April 21, 2015, Form 477 Data Filing”). 358 AT&T Inc., AT&T U-verse High Speed Internet, http://www.att.com/shop/internet/u-verse-internet.html (visited June 18, 2015) (providing U-verse Internet bundle prices). AT&T offers promotions in certain geographic areas. 359 AT&T Inc., Shop U-verse, U-verse with AT&T GigaPower, https://www.att.com/shop/u-verse/gigapower.html (visited June 1, 2015) (the pricing data for each city was determined by clicking on “Enter ZIP” at the top of the page and entering the zip codes for Austin, Texas, Nashville, Tenn., and Atlanta). 360 Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14- 90, Attachment (“OVD Ex Parte Presentation”) at 2 (April 21, 2015) (submitting written ex parte presentation on OVD services). 361 See Public Knowledge-ILSR Petition at 6 (“Even AT&T admits that this merger could exert upward pressure on ‘the price of standalone video or broadband.’ Suggesting only that this could be offset by cheaper bundles (though it is not committing to offering cheaper bundles).”); WGAW Petition at 21 (“In the Comcast-NBCU Order, the Commission found that a provider offering both video and Internet services could use bundling to hinder competition by requiring cable and Internet to be purchased together, or by making it economically impractical to purchase standalone broadband. Even Applicants acknowledge this harm, writing ‘in theory, there may be a potential incentive for the combined company to raise prices for standalone broadband in order to incentivize customers to purchase the bundle of services.’”). But see Application at 80-81; Joint Opposition at 37. Federal Communications Commission FCC 15-94 53 competitive option for the combined entity in the long term.362 The Greenlining Institute (“Greenlining”) argues that the proposed transaction would reduce the availability of standalone broadband services to low-income communities whose residents cannot afford bundled services.363 WGAW argues that the combined entity’s bundles would harm competition by tying customers to a single provider and that, although the Applicants claim that the ability to bundle video and Internet services would be a benefit of the transaction, bundling is also an effective strategy to discourage the substitution of OVD services for the MVPD services that the combined entity would provide as part of a bundle of integrated broadband and video services.364 Commenters argue that the Commission has long shown a preference for protecting standalone telecommunications services in order to safeguard consumer choice365 and that the Commission should adopt conditions to ensure that the combined entity offers reasonably priced standalone broadband.366 136. Several parties state that AT&T should commit to adopting an affordable, low-cost broadband Internet service.367 The National Association of Telecommunications Officers and Advisors (“NATOA”) argues that AT&T’s commitment to provide high-speed Internet to underserved rural areas comes up short when compared to what Comcast had announced as a commitment in connection with the formerly contemplated merger with Time Warner Cable.368 NATOA states that AT&T should voluntarily commit to offering a low-cost broadband adoption service similar to Comcast’s Internet Essentials broadband adoption program.369 The California Emerging Technology Fund (“CETF”) recommends that the Commission require AT&T to offer standalone broadband at an affordable rate for low-income individuals and likewise notes that the $9.95 per month rate offered by Comcast as part of its Internet Essentials program has worked well in recent years.370 Several public interest groups also suggest that the 362 Netflix Comments at 28 (stating that if the Applicants are correct that standalone high-speed broadband service is not competitive in the long term, there is unlikely to be significant competitive constraints in the future on the combined entity that would prevent it from harming OVDs to protect its own video offering). 363 Petition to Deny of the Greenlining Institute, MB Docket 14-90, at 4-6 (filed Sept. 16, 2014) (“Greenlining Petition”). 364 WGAW Petition at 20-21, 28 (citing Comcast-NBCU Order, 26 FCC Rcd at 4279, ¶ 102) (noting that tying a consumer to a single provider discourages innovation and competition between providers offering Internet and video service). 365 Id. at 22-24 (citing Comcast-NBCU Order, 26 FCC Rcd at 4279, ¶¶ 101-102); Joint Statement of Chairman Kevin J. Martin and Commissioner Deborah Taylor Tate, Concurring Statement of Commissioner Michael J. Copps, and Concurring Statement of Commissioner Jonathan S. Adelstein, AT&T-BellSouth Merger Order, 22 FCC Rcd at 5827, 5831, 5838. 366 See Franken Comments at 8; WGAW Reply at 35 (asserting that AT&T should be required to offer standalone broadband service of at least 10 Mbps down and 3 Mbps up for no more than $25 a month and AT&T should further agree that fixed wireless broadband will be made available as a standalone service). 367 See Comments of the National Association of Telecommunications Officers and Advisors, MB Docket 14-90, at 4-5 (filed Sept. 16, 2014) (“NATOA Comments”); Letter from John Bergmayer et al., Public Knowledge et al., to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 1-2 (May 28, 2015) (“Public Interest Advocates May 28, 2015, Ex Parte Letter”); Comments of California Emerging Technology Fund, MB Docket 14-90, at 10 (filed Oct. 14, 2014) (“CETF Comments”). We note that CETF’s comments were filed after the deadline established by our pleading cycle and shall therefore be treated as an informal request. 47 C.F.R. § 1.41. 368 NATOA Comments at 4-5 (addressing low-cost broadband service commitment announced by Comcast in connection with its proposed merger with Time Warner Cable). 369 Id. at 5; see also Application of Comcast Corp. and Time Warner Cable Inc., Description of Transaction, Public Interest Showing, and Related Demonstrations, MB Docket No. 14-57, at 59-66, 106-26 (filed April 8, 2014). 370 CETF Comments at 10. CETF recommends that the low rate broadband service be available to the general public for three years but suggests that, if the Commission wishes to limit the rate to specific underserved groups, the rate (continued….) Federal Communications Commission FCC 15-94 54 Commission require that AT&T make affordable broadband service available to all low-income customers and underserved communities as a condition of any approval of the Application.371 The public interest groups argue that such a condition is necessary “to ensure that all individuals within the merged entity’s service territory may share in the alleged, merger-specific broadband deployment public interest benefits of the transaction.”372 137. The Applicants assert that the combined entity would continue to have strong incentives to compete for standalone broadband customers.373 While the Applicants acknowledge a post-transaction incentive to potentially raise the price of their standalone broadband product to induce customers to purchase bundled service, they claim that the overall effect on consumer welfare from that incentive is minimal as it is counterbalanced by lower prices for the integrated bundle relative to the synthetic bundle currently offered374 as well as by lower prices of cable bundles, cable standalone video, and cable standalone broadband.375 Thus, AT&T contends that “the overall effect on current purchasers of standalone Internet access services may be beneficial.”376 138. The Applicants claim that AT&T’s broadband speed disadvantage, relative to broadband offerings of cable, other telephone MVPDs, and Internet service providers that are building high-speed broadband networks, also provides a strong incentive to keep standalone prices competitive.377 Further, AT&T states the projected increase in the number of households that receive all of their video from online services378 also would maintain its incentive to compete for standalone broadband customers.379 (Continued from previous page) should be available to low-income persons, seniors, and people with disabilities. CETF adds that returning veterans should also be considered for eligibility. Id. CETF also recommends that the Commission establish verifiable performance goals and compliance oversight for AT&T, as well as require AT&T to provide non-onerous service terms and adequate notice of availability for the low rate broadband service. Id. 371 Public Interest Advocates May 28, 2015, Ex Parte Letter at 1-2. 372 Id. 373 Application at 80. The Applicants also assert that “DIRECTV currently does not have broadband service offerings and has no reasonable prospect of developing such offerings organically.” Id.; see also Doyle Decl. ¶¶ 5- 6, 14, 25. Thus, the Applicants claim that “there is no current or potential horizontal broadband competition between the merging parties.” Application at 80; see also Katz Decl. ¶¶ 78, 84. 374 Application at 80; see also Katz Decl. ¶ 84. 375 Application at 80-81; see also Katz Decl. ¶¶ 4, 84, 88-89. The Applicants assert that the econometric analysis “shows that the drop in the price of the AT&T/DIRECTV bundle creates downward pressure on the range of cable broadband offerings, i.e., cable video/broadband bundles and cable standalone broadband, as well as cable standalone video.” Application at 81; see also Katz Decl. ¶ 88, Table 2, ¶ 92, Table 3. The Applicants believe that the “overall effect of any theoretical increase in prices for the relatively small number of AT&T standalone broadband customers would be more than offset by the expected decrease in prices for the much larger group of consumers purchasing the combined AT&T/DIRECTV broadband/video bundle, cable bundles, or cable standalone broadband.” Application at 81; see also Katz Decl. ¶¶ 84-85, 88, 90, 92, 94. 376 Application at 81 (quoting Katz Decl. ¶ 84). The Applicants contend that the overall projected impact on consumer welfare is positive and any cost efficiencies unaccounted for in Dr. Katz’s merger simulation would tend to put further downward pressure on standalone prices. Joint Opposition at 31 n.94; see also Application, Declaration of Rick L. Moore, Senior Vice President, AT&T, transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, ¶ 20 (filed June 11, 2014) (“Moore Decl.”) (asserting that the transaction would generate cost savings that would lead to improved standalone services for consumers). 377 Application at 81-82; see also Lee Decl. ¶ 25. 378 Application at 82. 379 Id. Federal Communications Commission FCC 15-94 55 139. The Applicants commit that, for three years after closing, they will “offer standalone wireline broadband service at reasonable market-based prices, including a service with speeds of at least 6 Mbps down (where feasible) at a 12-month price no greater than $34.95 per month (provided that the price can be increased by no more than any increase in the Consumer Price Index for All Urban Consumers (CPI-U) for Communications every 12 months starting 12 months following deal close).”380 The Applicants assert that this commitment would provide additional protections and guaranteed benefits to consumers who want to purchase only broadband service.381 140. Commenters argue that the Applicants’ commitment is too vague and that the time limitation of three years is insufficient to protect low-income consumers.382 Some commenters request that the Applicants be required to offer standalone broadband service at speeds of at least 25 Mbps where it is technically able to do so at a price not greater than $29.95 for seven years, and that the minimum speed offered as a standalone broadband service should be increased as the Commission’s benchmark broadband speed increases. Further, commenters request that any broadband speed offered as part of a bundled package must also be made available on a standalone basis, on equivalent usage terms and conditions as the bundled offering, at a reasonable price that takes into account the cost savings of not having to provide bundled services.383 Applicants contend that their proposed commitment to offer “standalone broadband service for three years at reasonable market-based prices” is “designed to assure customers that, after the merger, they will continue to enjoy a rich selection of standalone options throughout AT&T’s wireline broadband footprint.”384 141. In addition, the Applicants state that their commitment to offer a 6 Mbps service where feasible ensures that “it would be more widely available than a service of higher speeds.” 385 They state 380 Id. at 50; Joint Opposition at 37; see also Letter from Debbie Goldman, Telecommunications Policy Director, Communications Workers of America, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 1 (Feb. 3, 2015) (“CWA Feb. 3, 2015, Ex Parte Letter”) (noting that AT&T’s voluntary commitment to offer standalone broadband services for the next three years is a substantial public interest benefit); Letter from Robert W. Quinn Jr., Senior Vice President – Federal Regulatory and Chief Privacy Officer, AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 1 (Nov. 5, 2014) (noting AT&T’s commitment to keep standalone broadband services available for those customers who do not desire a bundle of services); Letter from Robert W. Quinn Jr., Senior Vice President – Federal Regulatory and Chief Privacy Officer, AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 1 (June 25, 2014) (“AT&T June 25, 2014, Ex Parte Letter”) (noting AT&T’s commitment to keep standalone broadband services available for those customers who do not desire a bundle of services). According to the Applicants, this specific commitment to maintain affordable standalone broadband service is clear evidence of the importance it places on broadband, thereby refuting Greenlining’s concerns that the transaction would reduce the availability of standalone broadband. Joint Opposition at 37 n.120. The Applicants further argue that the specific terms offered by AT&T also rebut Greenlining’s suggestion that AT&T’s commitment is at “whatever speed and price AT&T designs to offer.” Joint Opposition at 37 n.120 (quoting Greenlining Petition at 10). 381 See Application at 50, 82. 382 Greenlining Petition at 9-10 (stating that AT&T’s commitment is nothing more than a promise that AT&T provide standalone broadband service at “whatever speed and price AT&T designs to offer”); see also WGAW Petition at 21 (noting that the Applicants’ standalone broadband protections are offered for only three years). 383 Letter from Robert M. Cooper, Counsel for Cogent Communications, Inc.; Matt Wood, Free Press; John Bergmayer, Public Knowledge; Jeffrey Blum, DISH Network Corporation; and Josh Stager, New America’s Open Technology Institute, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 2-3 (May 12, 2015) (“Cogent et al. May 12, 2015, Ex Parte Letter”). 384 Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14- 90, Attachment (“Conditions Ex Parte Presentation”) at 9 (May 26, 2015) (submitting written ex parte in response to conditions proposed by Cogent, DISH, Cox, and other parties). 385 Conditions Ex Parte Presentation at 9. Federal Communications Commission FCC 15-94 56 that the commenters’ proposal for higher-speed standalone service would “benefit far fewer people, as only about [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent of AT&T customers subscribe to speed tiers of 25 Mbps or above.”386 Applicants also state that the commenters’ proposed seven-year duration fails to recognize the potential harm such long-term restrictions can cause in a market known for rapid technological advances, and that the proposed three-year term follows the precedent set in the Comcast-NBCU transaction.387 142. Discussion. The Modified Simulation predicts very little change in the price of AT&T’s standalone broadband post-transaction.388 The Modified Simulation finds that with no programming payment reductions the price would decrease by $0.33, which is roughly 0.73 percent less than pre- transaction levels.389 The Modified Simulation also finds, with a Low Programming Payment Reduction, that the price of AT&T’s standalone broadband service falls by $0.12, which is approximately 0.27 percent less than pre-transaction levels. For a High Programming Payment Reduction, the price of standalone broadband increases by $0.21 per month, which is roughly an increase of 0.46 percent over pre-transaction levels. 143. In addition, the Modified Simulation shows little change (positive or negative) in the prices of standalone broadband offered by other cable and telephone MVPD providers. The largest change in standalone broadband prices (in absolute magnitude) in the Modified Simulation is a $0.64 decrease in the cable standalone broadband price when there is a High PPR. This change represents a decrease of 1.2 percent relative to the prices in the simulation, and all other standalone broadband prices have changes of less than 1 percent in magnitude. Thus, we decline to impose as a condition the Applicants’ standalone broadband pricing commitment. 144. However, we have stated previously that it is in the public’s interest to ensure that a bundled option is not the consumer’s only competitive choice,390 and this protection may be particularly important for low-income subscribers who may not be able to afford bundled services.391 145. Thus, we adopt, as a condition to granting the Application, a requirement that AT&T implement and offer a discounted broadband services program to eligible consumers, as set forth in Appendix B. 386 Id. 387 Id. at 10. 388 See Appendix C Tables 7-9. The direction of the price changes for the AT&T-DIRECTV standalone broadband service does not differ between the BH Simulation and the Modified Simulation without programming payment reductions, but there is a difference in the magnitude. For the High PPR scenario, the direction and magnitude are not the same. See Berry-Haile Updated Results at 101-102, 110-111; Appendix C ¶¶ 89-94, Tables 7-9. 389 As an example, in terms of current rate plan prices, the price changes from the Modified Simulation would reflect an approximate 0.73 percent price decrease to a 0.36 percent price increase for an AT&T Max Plus plan depending on the amount of the programming payment reduction. See AT&T Inc., AT&T U-verse High Speed Internet, http://www.att.com/shop/internet/u-verse-internet.html (visited June 18, 2015) (providing U-verse Internet bundle prices). 390 Comcast-NBCU Order, 26 FCC Rcd at 4279, ¶¶ 101-103 (the Commission determined that the standalone broadband condition could help mitigate Comcast’s ability to use its vertical properties to harm competing video distributors). WGAW contends that the Commission’s preference for unbundled services promotes competition among distributors, allowing consumers to choose the services and the providers that best meet their needs, and that the emergence of OVDs increases the importance of access to affordable high-speed standalone broadband. See WGAW Reply at 24. 391 Greenlining Petition at 4-5. Federal Communications Commission FCC 15-94 57 B. Documentary and Other Record Evidence of Competition between AT&T and DIRECTV and the Need for Bundles 146. We also examined documentary and other evidence in the record to determine whether the transaction would harm competition. Such evidence is an important part of our analysis of the potential competitive effects of the loss of competition between AT&T and DIRECTV.392 Although the record shows that the proposed transaction results in some loss of competition between AT&T and DIRECTV, the record also supports the Applicants’ assertion that they have focused their marketing efforts on customers of the cable companies and view the cable companies as their primary competitors. In addition, the record supports the Applicants’ assertion that they are significantly limited by their inability to offer a competitive integrated bundle and that those limitations cannot be overcome by offering a synthetic bundle of AT&T broadband and DIRECTV satellite video. Thus, we find the Applicants’ post-transaction ability to offer a competitive integrated bundle of broadband and video, which may also have the effect of reducing competitors’ prices, to be a significant consumer benefit that outweighs the limited reduction in competition. This benefit, along with the other conditions we impose, outweighs the risk from the limited reduction in competition. 147. The Applicants claim that the transaction would promote competition because AT&T and DIRECTV’s video services are not “particularly close substitutes.”393 As previously noted, AT&T regards itself as primarily a wireline provider, and as such, it focuses its marketing efforts on its broadband product and bundled services.394 AT&T explains that its video footprint covers only one- quarter of U.S. households and, within that footprint, more than 97 percent of AT&T U-verse video customers buy that service bundled with broadband or other services.395 AT&T notes that only 138,000 of its customers buy U-verse video on a standalone basis.396 148. AT&T also asserts that its primary competitors are cable operators and cable overbuilders, citing evidence that when AT&T’s bundle customers switch to another provider, they largely choose cable and not satellite service.397 In contrast, AT&T considers DIRECTV’s satellite video service to be a complement to its broadband service.398 149. DIRECTV states that it focuses its marketing efforts on the customers of other MVPD providers, including cable operators and DISH, and on its standalone video subscribers.399 DIRECTV also asserts that it has had limited success in targeting broadband or bundled services customers,400 including because its satellite technology does not offer a two-way connection for non-linear video.401 DIRECTV explains the competitive importance of a two-way connection, stating that “the rise in non- 392 2010 DOJ/FTC Horizontal Merger Guidelines § 2.2 at 4-6. 393 See generally Competition White Paper. See also Katz Decl. ¶ 81 (arguing that DBS MVPDs and wireline MVPDs offer differentiated products primarily because wireline MVPDs provide integrated bundles). 394 Lee Decl. ¶¶ 28, 38-41. 395 See Application at 7; Lee Decl. ¶¶ 7, 12. 396 Application at 7; Lee Decl. ¶ 12; Katz Decl. ¶ 80. 397 See Application at 57-58; Lee Decl. ¶¶ 23-25, 28, 30, 38-41; Competition White Paper at 2-18. 398 See Application at 57-58; Competition White Paper at 2, 18-19. 399 See Application at 7-8, 57, 69-71; Competition White Paper at 2, 19-39. 400 See Application at 62-63 (“[T]he DIRECTV synthetic bundle is severely disadvantaged both in terms of speed and price, in comparison to the integrated bundles offered by cable operators, AT&T, and Verizon. It also provides an inferior customer experience, such as requiring two separate installation visits. … When customers leave DIRECTV, the lack of an integrated bundle is often a key reason they give.”). 401 Doyle Decl. ¶ 19. Federal Communications Commission FCC 15-94 58 linear viewing – outside of the scheduled broadcast, such as VOD and DVR content – has . . . increased the importance of offering a two-way connection using broadband.”402 DIRECTV maintains that the absence of a two-way connection “hampers its ability to integrate traditional linear video with on-demand and OTT [over-the-top] services in ways that create the richer, more flexible, and increasingly ubiquitous video experience demanded by consumers.”403 DIRECTV states that “cable companies, which offer broadband bundles that organically provide a two-way connection, have capitalized on this advantage by offering innovative features and services such as remote digital video recorders and VOD programming stored in the ‘cloud.’”404 150. DIRECTV also explains its efforts to overcome this inherent satellite disadvantage, for example by offering connected set-top boxes to allow its subscribers to access services over the Internet.405 DIRECTV notes that this strategy requires subscribers to separately arrange for and maintain their own broadband service,406 which makes it difficult for DIRECTV to get the DIRECTV set-top boxes connected to other providers’ Internet service.407 DIRECTV cites data suggesting that “ [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .”408 DIRECTV also notes that DIRECTV’s most advanced boxes permit customers to record roughly 100 hours of content while DIRECTV “manages” another 100 hours of content.409 DIRECTV asserts that broadband-enabled cable operators, with essentially unlimited cloud storage capacity, face no such difficulty.410 151. The Applicants maintain that DIRECTV’s growth has stagnated because even “a high- quality standalone video product may not be sufficient to compete with providers that offer their own integrated bundles of video and broadband services.”411 They note that in 2013 approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent of new DIRECTV video subscribers also activated broadband purchased in a synthetic bundle sold by DIRECTV.412 Similarly, AT&T sales of DIRECTV video [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .413 152. The Applicants also claim that, consistent with the economic analysis described above, their synthetic bundles are not competitive with bundles offered by cable operators “because they cannot match either the discounts on price or the seamless customer service offered by cable companies.”414 The Applicants state that neither AT&T nor DIRECTV has enough incentive to resolve the problems associated with synthetic bundles because the solutions “require considerable capital investment and 402 Id. 403 Id. 404 Id. 405 Id. ¶ 20. 406 Id. 407 Id. 408 Id. 409 Id. ¶ 21. 410 Id. 411 Id. ¶ 14. 412 Guyardo Decl. ¶ 20. 413 Lee Decl. ¶ 58. 414 Id. ¶¶ 4, 53-58. See also Application at 20, 52; Doyle Decl. ¶¶ 24-25; Guyardo Decl. ¶¶ 7, 21, 41-45; Katz Decl. ¶¶ 26-27, 29-32, 68-71, 97-106; Stankey Decl. ¶¶ 26-29; Double Moral Hazard White Paper at 17-19. The Applicants also note that competition “for video/broadband bundles occurs primarily between the competitors offering integrated bundles.” Application at 57-62. Federal Communications Commission FCC 15-94 59 effort by each party, but there is no business case for such outlays because the anticipated additional revenues and profits the projects might generate do not justify the anticipated costs.”415 153. The Applicants contend that they cannot improve the competitiveness of the synthetic bundles.416 Indeed, the Applicants claim that efforts to resolve the shortcomings of the synthetic bundles have been largely unsuccessful.417 The Applicants argue that “it is impractical, if not impossible” to overcome the synthetic bundle’s shortcomings because any action by one party (e.g., AT&T) to increase demand for its own product creates a benefit for the other party (e.g., DIRECTV), which is not internalized by the acting party.418 The Applicants have attempted to resolve these issues by modifying the joint marketing arrangement but report that these efforts have not been successful.419 As a result, neither party has an incentive to invest in the synthetic bundle at the optimal level.420 As noted above, the Applicants describe this situation as the commonly known “double moral hazard problem,” which occurs when independent firms contract to supply complementary goods jointly.421 The Applicants explain that the firms have misaligned incentives, which results in less investment in the joint offering than is optimal because neither party takes into account the benefits that accrue to the other party from their investment.422 The Applicants state that post-transaction, the double moral hazard problem would be eliminated and the customer experience would be greatly simplified because there would be a single sales process, a single installation appointment, a single bill, and “one call” resolution for billing and service issues.423 154. As stated above, the Applicants also argue that the synthetic bundle customer experience is inferior to the integrated bundle customer experience.424 For example, the synthetic bundle customer must make two different installation appointments and receives two different bills.425 The Applicants note that synthetic bundlers are further disadvantaged because they are likely to incur additional fees that fully integrated providers typically waive for their own bundled customers.426 The Applicants also assert that the synthetic bundle price discount given to the customer might not be applied for several months, which results in customer confusion and complaints.427 They note that if customers have a complaint, they often need to contact both AT&T and DIRECTV instead of being able to resolve all of their issues 415 Double Moral Hazard White Paper at 21; see also Katz Reply Decl. ¶ 16. 416 Joint Opposition at 15. 417 Guyardo Decl. ¶¶ 36-38; Lee Decl. ¶ 59. 418 Double Moral Hazard White Paper at 2-5. See also Application at 67. 419 Application at 65; Joint Opposition at 13-15; Guyardo Decl. ¶¶ 36-38; Katz Decl. ¶¶ 74-76, 100-103; Lee Decl. ¶ 59. The Applicants do note that they have had some limited success resolving the customer service issues and improving the incentives for each side to promote and sell the service. Guyardo Decl. ¶¶ 36, 38. See also Double Moral Hazard White Paper at 8-21. 420 Application at 66; Katz Decl. ¶¶ 4, 69. 421 Application at 66; Katz Decl. ¶ 97. 422 Double Moral Hazard White Paper at 4-5; Application at 66-67. 423 Joint Opposition at 15; see also Doyle Decl. ¶ 27; Katz Decl. ¶¶ 61, 63; Katz Reply Decl. ¶ 12. 424 Application at 20; Guyardo Decl. ¶¶ 31-38; Katz Decl. ¶¶ 104-105; Lee Decl. ¶¶ 4, 13. 425 Katz Decl. ¶ 36; Lee Decl. ¶ 57. They could also be required to undergo multiple credit checks. Guyardo Decl. ¶ 31. 426 Guyardo Decl. ¶ 29 ( [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ). Guyardo Decl. ¶ 29. See also Katz Decl. ¶ 75; Double Moral Hazard White Paper at 11. 427 Guyardo Decl. ¶ 33; Lee Decl. ¶ 57. Federal Communications Commission FCC 15-94 60 through a single call because AT&T has to resolve broadband service issues while DIRECTV has to resolve video problems.428 155. Discussion. We find that the record reflects that AT&T primarily focuses its competitive strategy, marketing, and pricing on other providers of broadband and bundled services, namely incumbent cable operators and cable overbuilders.429 Although AT&T tracks pricing strategies and service offerings of all MVPDs, including DIRECTV, the record also supports AT&T’s claim that its primary competition is the bundle offered by cable operators.430 156. The record also supports DIRECTV’s position that it is at a competitive disadvantage without broadband or bundled services. Like AT&T, DIRECTV tracks pricing strategies and service offerings of all MVPDs, but it identifies cable operators as its primary competitors.431 To compete with cable operators and their bundled services, DIRECTV offers synthetic bundles by partnering with broadband providers including AT&T.432 However, as explained, DIRECTV has had limited success with its synthetic bundles. DIRECTV also has attempted to differentiate its standalone video product from other video providers, including by emphasizing DIRECTV’s exclusive content, and providing consumer upgrades.433 157. Additionally, the record supports the Applicants’ position that bundles of broadband and video are more attractive to consumers. An analysis prepared for AT&T by Frost and Sullivan found that [BEGIN CONF. INFO.] [END CONF. INFO.] .434 The report also indicated that [BEGIN CONF. INFO.] [END CONF. INFO.] .435 The Applicants note that 78 percent of basic cable video subscribers purchase a bundle of services436 and that “more than 97 percent of AT&T’s 5.7 million video customers subscribe to bundled services.”437 The Applicants disclose that in the first quarter of 2014 “approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent of the subscribers leaving DIRECTV reported that they will purchase a bundle of video and broadband services from their new provider,” which they assert is an increase from the level reported three years earlier.438 AT&T’s internal documents state that [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .439 In 428 Guyardo Decl. ¶ 34; Katz Decl. ¶ 36; Lee Decl. ¶ 57. 429 The Commission has previously noted that telephone MVPDs and cable operators differentiate their services by highlighting the benefits of their bundled services, in contrast to DBS providers, which focus their marketing efforts on video service. See Sixteenth Annual Report, 30 FCC Rcd at 3288, ¶ 81. 430 For example, documents discussing AT&T’s U-verse subscriber churn rates demonstrate that [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] See, e.g., ATT-FCC-03227145, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ; ATT-FCC-03226408, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See also Lee Decl. ¶ 30. 431 Documents discussing DIRECTV’s subscriber churn rates evidence that [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See DTVFCC-00645037, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See, also, e.g., DTVFCC-01748388, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ; DTVFCC-00244051, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 432 See, e.g., DTVFCC-01748388, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 433 See, e.g., id.; see also Competition White Paper at 2, 19-39. 434 ATT-FCC-00140646, [BEGIN CONF. INFO.] [END CONF. INFO.] . 435 Id. 436 Application at 21; Doyle Decl. ¶ 16. 437 Application at 2. See also Application at 70; Lee Decl. ¶ 12; Competition White Paper at 6. 438 Guyardo Decl. ¶ 11. See also Application at 63; Katz Decl. ¶ 31. 439 Video and Broadband Complementarity White Paper at 7. Federal Communications Commission FCC 15-94 61 addition, the Applicants’ documents discussing the AT&T-DIRECTV synthetic bundle show that [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .440 In addition, AT&T found that having “ [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .”441 DIRECTV internal documents also support [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .442 158. Based on our review of the record, we also agree that the customer experience for synthetic bundles is often inferior to the customer experience for integrated bundles. DIRECTV’s documents demonstrate that [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .443 Additionally, DIRECTV’s documents state that “ [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .”444 DIRECTV’s customer satisfaction data support this viewpoint because customers of DIRECTV’s synthetic bundle report having a dramatically inferior experience compared to DIRECTV’s video-only customers.445 Similarly AT&T’s documents state that synthetic bundles “ [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .”446 159. Thus, we conclude that the Applicants’ ability to provide an integrated bundle of AT&T broadband and DIRECTV video throughout AT&T’s broadband footprint is a significant positive competitive effect of the transaction. C. Conclusion 160. We find that our competitive effects analysis does not support a finding that the transaction creates overall public interest harm. X. ADDITIONAL COMPETITIVE EFFECTS AND PUBLIC INTEREST HARMS RAISED IN THE RECORD 161. In addition to our analysis of the competitive effects described above, we also considered other public interest harms that were raised in the record. Our findings on each of these potential public interest harms are described below. A. Limits on Competitors’ Access to Programming 162. Commenters identified concerns that the combined entity may have an increased incentive, and potentially increased ability, to impede competitors’ or potential competitors’ access to affiliated and unaffiliated programming. We find that to the extent that the issues identified by commenters are transaction specific, the Commission’s existing rules afford adequate protections. 440 See, e.g., ATT-FCC-01463960, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 441 ATT-FCC-00428824, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See also Video and Broadband Complementarity White Paper at 7; Competition White Paper at 5. 442 ATT-FCC-03371650, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See also DTVFCC-00935963, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 443 See DTVFCC-00691428, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See also DTVFCC-01008205, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ; DTVFCC-02626878, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 444 See DTVFCC-02626878, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ; Guyardo Decl. ¶ 32. 445 Joint Opposition at 13 (At AT&T, in particular, sales of the synthetic bundle of AT&T’s broadband along with DIRECTV’s video [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ). See also Guyardo Decl. ¶ 35. 446 ATT-FCC-03149985, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See also Double Moral Hazard White Paper at 17-21. Federal Communications Commission FCC 15-94 62 1. Limiting Access to RSNs and Other Affiliated Programming 163. DIRECTV owns and operates two RSNs, Root Sports Pittsburgh and Root Sports Rocky Mountain, and holds a minority interest in, and manages, the Seattle-based RSN, Root Sports Northwest.447 DIRECTV also has a 42 percent non-controlling interest in the Game Show Network and smaller, minority interests in the MLB Network, the NHL Network, and a handful of other networks, including the Tennis Channel.448 DIRECTV and AT&T also recently acquired majority ownership of a Houston-area RSN (“CSN Houston”) out of bankruptcy and relaunched it as Root Sports Southwest.449 As a result of the transaction, AT&T would acquire DIRECTV’s interests in these programming content holdings. 164. Positions of the Parties. Some commenters express concern that the combined entity would have the incentive and ability to restrict access to vertically owned programming, especially sports programming.450 These commenters argue that the combined entity could foreclose and unfairly disadvantage rival MVPDs by withholding or artificially raising costs for these programming assets.451 Cequel Communications, LLC d/b/a Suddenlink Communications (“Suddenlink”) notes that some subscribers are willing to change MVPD providers just to gain access to sports content such as DIRECTV’s RSNs452 and that an MVPD’s access to sports programming, and the price and terms of such access, are important to an MVPD’s ability to compete.453 165. To mitigate the potential harm that the combined entity could exploit its ownership of programming, including RSNs, against competing MVPDs, some commenters propose that the Commission impose certain conditions for approving the transaction.454 Commenters ask the Commission to affirm that the program access rules would apply to all video services provided by the combined entity, regardless of the facilities used to provide that service.455 Several commenters also ask the Commission to impose arbitration conditions, as it has done in some previous transactions, to mitigate 447 Application at 14; Doyle Decl. ¶ 9. 448 Id. 449 Joint Opposition at 55. See also In re Houston, 514 B.R. 211. We note that AT&T, DIRECTV, Comcast, and several smaller MVPDs are currently carrying Root Sports Southwest, but DISH and Suddenlink have not yet entered into agreements to carry the RSN. See Root Sports, Channel Finder, http://southwest.rootsports.com/channel-finder.com/channel-finder/ (visited June 18, 2015). 450 Franken Comments at 6-7; Comments of Cequel Communications, LLC d/b/a Suddenlink Communications, MB Docket 14-90, at 4-5 (filed Sept. 16, 2014) (“Suddenlink Comments”); Reply Comments of American Cable Association, MB Docket 14-90, at 3-9 (filed Jan. 7, 2015) (“ACA Reply”); see also Letter from Bridget Watkins et al., All West Commc’ns et al., to Thomas Wheeler, Chairman, FCC, MB Docket No. 14-90, at 1-2 (July 9, 2015) (submitting the concerns of small MVPDs that currently purchase carriage rights for RSNs owned or controlled by the Applicants). 451 Id. 452 Suddenlink Comments at 4. 453 Id. 454 Id. at 4-5, 8-14. 455 Franken Comments at 7; Cox Petition at 18-21; Cox Reply at 6-7; Letter from Jason E. Rademacher, Cooley LLP, Counsel for Cox Communications, Inc., to Marlene Dortch, Secretary, FCC, MB Docket No. 14-90, at 1 (Nov. 7, 2014) (“Cox Nov. 7, 2014, Ex Parte Letter”); Letter from Jason E. Rademacher, Cooley LLP, Counsel for Cox Communications, Inc., to Marlene Dortch, Secretary, FCC, MB Docket No. 14-90, at 3 (Dec. 4, 2014) (“Cox Dec. 4, 2014, Ex Parte Letter”); Letter from Jason E. Rademacher, Cooley LLP, Counsel for Cox Communications, Inc., to Marlene Dortch, Secretary, FCC, MB Docket No. 14-90, at 3 (Dec. 22, 2014) (“Cox Dec. 22, 2014, Ex Parte Letter”). Federal Communications Commission FCC 15-94 63 potential harms from vertical integration between video programmers and distributors.456 Commenters note that, until recently, DIRECTV’s RSNs were subject to arbitration conditions under the Liberty Media-DIRECTV Order but that the conditions have since expired.457 American Cable Association (“ACA”) argues that, despite the expiration of these conditions, DIRECTV has the same incentives to charge competing MVPDs higher prices for its RSN programming.458 Other commenters agree, pointing to AT&T’s and DIRECTV’s prior conduct and arguing that increased opportunity costs and bargaining power as a result of the transaction create a greater incentive to harm rivals.459 Commenters propose that the Commission adopt various other related conditions to address these concerns.460 166. The Applicants respond by observing that neither AT&T nor DIRECTV has substantial content holdings.461 AT&T’s U-verse service is not available in the areas served by DIRECTV’s existing RSNs. The Applicants contend the transaction does not increase the incentive to withhold DIRECTV’s RSNs because AT&T’s U-verse service is not available in the areas served by those RSNs.462 The Applicants also note that Root Sports Southwest was unsuccessful prior to their acquisition because it lacked wide distribution. Although Root Sports Southwest operates in an area where U-verse is available, the Applicants argue that they would have strong incentives to seek carriage to strengthen the RSN’s economic viability.463 Finally, the Applicants also note that AT&T is subject to the Commission’s existing program access rules and that DIRECTV remains subject to program access conditions for its RSN programming.464 Therefore, the Applicants conclude there are no transaction-specific harms that warrant additional program access conditions.465 456 Suddenlink Comments at 5-6, 8; ACA Reply at 15-22, 54; Reply Comments of WaveDivision Holdings, LLC, MB Docket 14-90, at 3, 6 (filed Jan. 7, 2015) (“Wave Reply”). 457 Comments of American Cable Association, MB Docket 14-90, at 12-14 (filed Sept. 16, 2014) (“ACA Comments”); Suddenlink Comments at 5-6; Wave Reply at 3. 458 ACA Comments at 13-14. For example, Wave asserts that following the recent expiration of DIRECTV’s arbitration conditions, DIRECTV has failed to negotiate meaningfully for Root Sports Northwest. Wave Reply at 4- 5. Wave states that its agreement with DIRECTV to carry Root Sports Northwest expired on December 31, 2014, but that DIRECTV has granted Wave an extension of this agreement to allow negotiations to continue. Id. at 4 n.9. 459 Cox Petition at 7, 9; Suddenlink Comments at 6-7; ACA Comments at 12-15. For example, Suddenlink argues that the recent program access arbitration between Armstrong Utilities, Inc. and DIRECTV demonstrates that DIRECTV has a history of imposing unfair rates on competing MVPDs. Suddenlink Comments at 6-7. See DIRECTV Sports Net Pittsburgh, LLC v. Armstrong Utilities, Inc., Memorandum Opinion and Order, 29 FCC Rcd 8624, 8629-30, ¶ 14 (2014) (affirming an arbitration decision in favor of Armstrong Utilities in a dispute concerning the fair market value of an RSN owned by DIRECTV Sports Net Pittsburgh). 460 ACA proposes several conditions, including a non-discriminatory access condition; various modifications of the program access complaint process for MVPDs that file complaints under the non-discriminatory access condition; certain conditions designed to prevent the combined entity’s increased size from harming MVPDs in their negotiations with other programmers; and a requirement that any program access condition remain in effect for at least nine years. See ACA Reply at 43-59. Suddenlink proposes a prohibition on tying carriage of one RSN to carriage of any other RSN(s) and an a la carte condition (or alternatively, an MFN condition) on DIRECTV’s RSN programming. See Suddenlink Comments at 8-13. Wave proposes a requirement that the rates, terms, and conditions charged to terrestrial facilities-based providers for distribution of DIRECTV-affiliated RSNs be no more than those made available to competitors of those providers. See Wave Reply at 6. 461 Joint Opposition at 54. 462 Id. 463 Id. at 55. 464 Id. at 55-56. 465 Id. at 56. Federal Communications Commission FCC 15-94 64 167. Discussion. On the basis of this record, we do not find harms relating to access to affiliated programming that require us to impose upon the combined entity company-specific conditions in addition to our existing program access rules. With the exception of the recent joint venture in Houston, AT&T’s U-verse service is not available in any of the markets served by DIRECTV’s existing affiliated RSNs (Root Sports Pittsburgh, Root Sports Rocky Mountain, and Root Sports Northwest).466 Accordingly, because the combined entity’s concentration in these markets would not increase as a result of the transaction, the transaction would not affect the combined entity’s incentive to make these RSNs available to its rivals.467 In addition, AT&T and DIRECTV already jointly control the Houston RSN through a joint venture, with AT&T owning a 40 percent interest and DIRECTV owning a 60 percent interest.468 Because the Houston RSN is already vertically integrated as to AT&T and DIRECTV separately, the transaction does not change the combined entity’s ability or incentives to limit access of the RSN to other MVPDs. As a joint venture, the Houston RSN can be expected to maximize the profits of both AT&T and DIRECTV when negotiating with a rival MVPD.469 168. Congress enacted the program access provisions of Section 628 of the Act as part of the 1992 Cable Act to address concerns that vertically integrated program suppliers have the incentive and ability to favor their affiliated cable operators over unaffiliated MVPDs.470 Specifically, Congress was concerned that a vertically integrated programmer may exclude rival distributors, including new entrants and new technologies, from access to its programming, or raise programming prices to harm competition in the video distribution market.471 Thus, the program access rules adopted by the Commission pursuant to Section 628 provide several important protections to unaffiliated MVPDs.472 169. We confirm that the combined entity will be subject to the program access rules with respect to all affiliated programming, including RSN programming and non-RSN programming, offered on both its U-verse video and DBS service. Section 628(j) of the Act extends the program access rules to common carriers, such as AT&T, as well as affiliates of common carriers, that provide video programming by any means to subscribers.473 As a DBS service, DIRECTV is not currently subject to the 466 Id. at 54. 467 See Adelphia Order, 21 FCC Rcd at 8261-62, ¶ 128 (“It is the combination of RSN ownership and MVPD market share that makes anticompetitive strategies possible.”); Implementation of the Cable Television Consumer Protection and Competition Act of 1992; Development of Competition and Diversity in Video Programming Distribution: Section 628(c)(5) of the Communications Act Sunset of Exclusive Contract Prohibition, CS Docket No. 01-290, Report and Order, 17 FCC Rcd 12124, 12140, ¶ 38 (2002) (“2002 Extension Order”) (“The number of subscribers that a vertically integrated cable programmer serves is of particular importance in calculating the benefits of withholding programming from rival MVPDs. … Other things being equal, then, as the number of subscribers rises, so does the likelihood that withholding would be profitable.”). 468 See DIRECTV Response to Sept. 9, 2014, Information Request at 36; Derek Baine, Cable Network M&A Market Awakens From Quiet Period, SNL KAGAN, Nov. 28, 2014. 469 See ACA Comments, Exhibit A, Statement of Professor Gary Biglaiser (“Biglaiser Statement”) at 10 n.9. 470 See 1992 Cable Act § 2(a)(5), 47 U.S.C. § 521 (2012). 471 See id.; see also H.R. REP. NO. 102-862, at 93 (1992) (Conf. Rep.); S. REP. NO. 102-92, at 28 (1991). 472 See 47 C.F.R. §§ 76.1000-1004. 473 47 U.S.C. § 548(j) (“Any provision that applies to a cable operator under this section shall apply to a common carrier or its affiliate that provides video programming by any means directly to subscribers.”); see also 47 C.F.R. § 76.1004 (“Any provision that applies to a cable operator under §§ 76.1000 through 76.1003 shall also apply to a common carrier or its affiliate that provides video programming by any means directly to subscribers.”). The Applicants have acknowledged that AT&T is subject to these existing program access rules. Joint Opposition at 55- 56. We note that AT&T has contested previously whether its U-verse video service is a “cable service.” See, e.g., Comments of AT&T Services, MB Docket No. 13-140, at 5 (filed June 19, 2013) (“AT&T Services MB 13-140 Comments”) (contesting regulatory classification as a cable system for regulatory fee purposes); see also Fifteenth Annual Report, 28 FCC Rcd at 10508, ¶ 28 (noting that AT&T U-verse has not registered with the Commission as a (continued….) Federal Communications Commission FCC 15-94 65 program access rules under Section 628, although DIRECTV’s affiliated RSN programming remains subject to program access conditions under the Liberty Media-DIRECTV Order.474 Following the transaction, however, DIRECTV will be a wholly owned subsidiary, and hence an affiliate, of AT&T.475 Accordingly, we conclude that following the transaction, DIRECTV, as an affiliate of a common carrier that provides video programming to subscribers, will be subject to the program access rules under Section 628(j). 170. We further conclude that it is unnecessary to impose arbitration or any other additional program access conditions to address concerns regarding access to RSNs or other programming owned or controlled by the combined entity. We find that the transaction would not increase the combined entity’s incentives or ability to withhold RSN programming from its rivals. For the combined entity to foreclose RSN programming from its rivals, the transaction would have to change the opportunity cost of doing so. The combined entity would have to weigh the loss in revenue from selling its affiliated programming to rivals against the gains in revenue from subscribers leaving the rival provider as a result of RSN programming being withheld and becoming a subscriber of the combined entity. Based on our analysis of the facts described herein, we do not believe that as a result of this transaction the gains from withholding RSN programming from its rivals would outweigh the losses from providing the programming, nor has any party submitted evidence to the contrary. Thus, on the basis of this record we find the existing program access rules address the potential for anticompetitive conduct with respect to the combined entity’s RSNs and other vertically integrated programming.476 171. Furthermore, we find that there is insufficient evidence of harm from the combined entity’s ownership of the Houston RSN to warrant imposition of arbitration or other program access conditions. The combined entity’s [BEGIN CONF. INFO.] [END CONF. INFO.] percent market share in the Houston DMA is not [BEGIN CONF. INFO.] [END CONF. INFO.] percent market share in this DMA.477 We also note that, following AT&T’s and DIRECTV’s acquisition of this RSN out of bankruptcy, Root Sports Southwest has already reached agreements with most of the MVPDs that serve the Houston DMA to distribute the RSN.478 172. We do not have a basis on the record to conclude that the combined entity would have an incentive to foreclose access to RSN programming in DMAs where its combined market share would increase significantly post-transaction, such as Dallas and St. Louis, where the Applicants do not (Continued from previous page) cable system). As AT&T concedes that it is subject to the program access requirements pursuant to Section 628(j), we need not address the broader issue of whether U-verse video constitutes a cable service. See Joint Opposition at 55-56. 474 See Liberty Media-DIRECTV Order, 23 FCC Rcd at 3303, ¶ 83. 475 Application at 16. As a wholly owned subsidiary of AT&T post-transaction, DIRECTV will be an “affiliate” of a common carrier pursuant to Section 3 of the Act. See 47 U.S.C. § 153(2) (defining “affiliate” as “a person that (directly or indirectly) owns or controls, is owned or controlled by, or is under common ownership or control with, another person,” with “own” meaning “to own an equity interest (or the equivalent thereof) of more than 10 percent”). Both Section 628(j) and the definition of affiliate were added to the Act by the Telecommunications Act of 1996. Telecommunications Act of 1996, Pub. L. No. 104-104, §§ 3, 301(h), 110 Stat. 56, 58, 117 (1996). 476 We acknowledge the concerns raised by parties regarding the effectiveness of the Commission’s existing program access and program carriage rules; however, we believe that such concerns are industry-wide, not merger specific, and therefore are better addressed in a separate proceeding. See, e.g., ACA Reply at 10-11; Cox Petition at 6-18; Petition to Impose Conditions of DISH Network Corp., MB Docket 14-90, at 12 (filed Sept. 16, 2014) (“DISH Petition”). 477 See MediaCensus 2014Q3. The Houston RSN was formerly affiliated with Comcast, one of the largest MVPDs in the Houston market. Despite that affiliation, however, the RSN struggled and eventually sought bankruptcy protection, after which the Applicants acquired majority control of the RSN in late 2014. See supra ¶¶ 163, 167. 478 See supra n.449. Federal Communications Commission FCC 15-94 66 currently own RSNs.479 Nothing in the record suggests that AT&T or DIRECTV has explored RSN opportunities in these DMAs or that such opportunities are likely to be available in these DMAs in the near future.480 173. Furthermore, we find unpersuasive the analysis by ACA’s economics expert Professor Gary Biglaiser, and consequently, we do not change our findings based on that submission. Professor Biglaiser uses the Nash bargaining framework the Commission relied on in the Comcast-NBCU Order to analyze potential RSN related harms.481 According to Professor Biglaiser, given that AT&T does not provide U-verse service in the same areas as DIRECTV’s three RSNs and that the Applicants already have a RSN joint venture in Houston, the only variable that would change is the profit.482 He claims that the efficiencies and increased bargaining power in buying programming achieved by combining AT&T with DIRECTV’s distribution and programming assets would increase the profitability per video subscriber of the DIRECTV service.483 This higher profit margin, Professor Biglaiser avers, would increase the combined entity’s opportunity cost of selling affiliated programming to rival MVPDs, leading to higher prices for rivals and ultimately consumers.484 174. Professor Biglaiser, however, does not provide estimates for the other variables in the Nash bargaining model, which he acknowledges would not change as a result of the transaction.485 Nor does he quantify the extent to which profits would increase. Instead, he simply claims that given the Applicants’ claimed efficiencies, profits would increase, and increase such that the Applicants would have an incentive to withhold RSN programming from their rivals. ACA’s economics expert does not provide evidence that the combined entity’s profits would increase to such a degree that it would provide the combined entity with the incentive to foreclose rival MVPDs from its RSN programming. Thus, we find speculative the assertion that the combined entity’s profit margin would increase to the degree he suggests, thereby increasing its opportunity cost of selling affiliated programming to rivals. 175. Further, some of the efficiencies on which ACA’s economics expert relies to support his argument that the combined entity’s profits would increase such that it would have an incentive to 479 The combined market share in the Dallas and in the St. Louis DMAs are approximately [BEGIN CONF. INFO.] [END CONF. INFO.] respectively. See MediaCensus 2014Q3. 480 DIRECTV, since 2012, has explored potential RSN opportunities in the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] DMAs. See DIRECTV Response to Sept. 9, 2014, Information Request at 36. AT&T does not provide U-verse service in the Phoenix DMA, but it does provide U-verse service in the Chicago DMA. The Applicants’ combined market share in the [BEGIN CONF. INFO.] [END CONF. INFO.] DMA is less than [BEGIN CONF. INFO.] [END CONF. INFO.] percent and in the [BEGIN CONF. INFO.] [END CONF. INFO.] DMA is slightly greater than [BEGIN CONF. INFO.] [END CONF. INFO.] percent. See MediaCensus 2014Q3. Given the current contractual situations of sports teams in both DMAs, it is unlikely that opportunities to acquire RSNs in these DMAs would be available in the near future. See DIRECTV Response to Sept. 9, 2014, Information Request at 36. 481 See Biglaiser Statement at 7. The Nash bargaining model relied on in the Comcast-NBCU model was: C = ?d? where C is the opportunity cost for an MVPD to sell its programming assets to a rival MVPD, ? is the diversion rate, d is the departure rate if programming is withheld, and ? is the monthly profit. See Comcast-NBCU Order, 26 FCC Rcd at 4258, ¶ 46 and Appendix B, “Technical Appendix,” Section 1.B (describing the vertical price increases in the Comcast-NBCU merger). 482 See Biglaiser Statement at 10. 483 See id. at 11-14; see also ACA Comments at 14-16; ACA Reply at 4. 484 See Biglaiser Statement at 9-11, 14; see also ACA Comments at 14-17; ACA Reply at 4-5. 485 These other variables are the diversion rate (the rate at which subscribers of one provider would switch to another) and the departure rate (the estimated rate that a subscriber would leave a rival provider if an MVPD withheld affiliated programming). Professor Biglaiser states that neither of these variables would change in the DMAs where the combined entity would have a RSN. See Biglaiser Statement at 9-10. Federal Communications Commission FCC 15-94 67 withhold RSN programming may not be as large as claimed. For example, as discussed above, we find that it is likely that at least part of the programming payment reductions would be passed onto consumers in the form of lower prices, and therefore its profits may not be as great.486 Additionally, Professor Biglaiser claims that benefits from AT&T’s ability to bundle its Fixed Wireless Local Loop service (“FWLL”) service with DIRECTV’s video service would raise the opportunity cost of selling its RSN programming to rivals in the four relevant DMAs.487 However, the record provides no evidence that the penetration rate of FWLL in any of these DMAs would be high. Further, even if the penetration rate were high, there is no evidence on the record that customer uptake of a FWLL and DIRECTV bundle would be large enough to raise the opportunity cost of selling RSN programming to rivals in these DMAs.488 176. Finally, we find that this transaction is distinguishable from previous transactions where the Commission imposed arbitration or other program access conditions. In each such prior transaction, the Commission found that competitive harm would likely result from the vertical integration of significant programming interests (including RSNs or other programming) that could not be addressed by the Commission’s program access rules.489 That is not the case here. As discussed above, the record does not support a finding that the transaction would increase the combined entity’s ability or incentive to implement foreclosure or price raising strategies with respect to its RSN programming or other affiliated programming. Accordingly, based on all of these factors, we conclude that adoption of arbitration or other program access conditions is not warranted. 2. Exclusive Programming Agreements 177. Positions of the Parties. Commenters express concern about the Applicants’ ability to obtain exclusive programming agreements. Cox Communications, Inc. (“Cox”) argues that the transaction increases the risk that the combined entity would secure exclusive contracts for programming content, particularly sports programming, and that such contracts would inhibit competition from smaller MVPDs.490 Cox asserts that DIRECTV’s exclusive NFL Sunday Ticket package, which allows subscribers to view every out-of-market live NFL game, provides DIRECTV with a significant competitive advantage and that permitting the combined entity to maintain and expand exclusive access to NFL Sunday Ticket would harm competition.491 In addition, Cox states that allowing the combined entity to obtain other exclusive programming agreements would give it an unfair advantage that would be nearly 486 See supra ¶ 110. As discussed below, while AT&T is likely to achieve some programming payment reductions, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] which may similarly affect the amount of the reduction and thereby AT&T’s profits. See infra ¶ 289. 487 See Biglaiser Statement at 12-13 (using AT&T’s LTE network); ACA Comments at 16. 488 As discussed below, we find that there is uncertainty about the actual user speed and pricing that would affect the profitability of the FWLL deployment. See infra ¶¶ 370-374. 489 In the Liberty Media-DIRECTV Order and the News Corp.-Hughes Order, the Commission concluded that the combined entities in those cases would have an increased incentive to adopt a temporary foreclosure strategy in order to increase fees for RSN programming. See Liberty Media-DIRECTV Order, 23 FCC Rcd at 3306, ¶ 88; News Corp.-Hughes Order, 19 FCC Rcd at 543-548, ¶¶ 147-162. In the Adelphia Order, the Commission found that the transactions would increase the combined entity’s incentive and ability to adopt a uniform price strategy for RSN programming. See Adelphia Order, 21 FCC Rcd at 8267-73, ¶¶ 140-154. In the Comcast-NBCU Order, the Commission found that Comcast-NBCU would have an increased ability and power to implement exclusionary (either complete foreclosure or price raising) strategies with respect to RSN programming, local broadcast programming, and national cable programming. See Comcast-NBCU Order, 26 FCC Rcd at 4254-58, ¶¶ 36-44. 490 Cox Petition at 7; Cox Nov. 7, 2014, Ex Parte Letter at 1-2; Cox Dec. 4, 2014, Ex Parte Letter at 3; Cox Dec. 22, 2014, Ex Parte Letter at 3. 491 Cox Petition at 9; Cox Nov. 7, 2014, Ex Parte Letter at 1-2; Cox Dec. 4, 2014, Ex Parte Letter at 3; Cox Dec. 22, 2014, Ex Parte Letter at 3. Federal Communications Commission FCC 15-94 68 impossible for smaller rivals to overcome.492 Cox suggests that AT&T’s nationwide, multi-platform reach gives it the ability to outbid rival MVPDs for other exclusive programming rights.493 Other commenters also raise concerns about the combination of AT&T’s wireless service with DIRECTV’s exclusive programming content.494 Cox proposes a condition prohibiting the combined entity from entering into or continuing any existing exclusive programming contracts or, at a minimum, exclusive contracts for major sports programming.495 178. The Applicants note that the Commission recently declined to extend the prohibition on exclusive arrangements between cable operators and affiliated programmers, finding that exclusive contracts may have the procompetitive benefit of allowing MVPDs to differentiate their programming.496 The Applicants further state that there is no reason that the combined entity should be the only MVPD unable to enter into exclusive arrangements with unaffiliated programmers.497 The Applicants observe that the Commission rejected requests for conditions prohibiting DIRECTV from entering into exclusive agreements for NFL Sunday Ticket in prior transaction reviews.498 In its reply, Cox asserts that previous transaction decisions approving DIRECTV’s access to exclusive programming to encourage competition are irrelevant given that this transaction would result in the loss of a DIRECTV competitor and the combined entity would have unprecedented size and customer reach.499 Cox also notes that there is no precedent for allowing exclusive agreements with a company that can offer a nationwide bundle of video, voice, and data services.500 Cox asserts that DIRECTV is the only large MVPD that has exclusive programming arrangements for national or competitively significant programming.501 179. Discussion. We recognize that, depending on the factual circumstances, exclusive contracts may be procompetitive or anticompetitive. In connection with this transaction, we do not find potential harms relating to exclusive contracts that are not already addressed by our existing program access rules. Thus, we decline to impose a condition prohibiting the combined entity from continuing or entering into any exclusive programming contracts, whether for affiliated or unaffiliated programming. 180. With respect to affiliated programming, the Commission in 2012 allowed the prohibition on exclusive contracts involving satellite-delivered, cable-affiliated programming to expire, concluding that such a prohibition was no longer necessary to protect and preserve competition and diversity in the 492 Cox Petition at 9-10. Cox also cites reports that renewal of DIRECTV’s exclusive agreement for NFL Sunday Ticket is a condition for completion of the transaction as evidence that the combined entity intends to pursue an exclusive programming strategy. Cox Petition at 7; see also DISH Petition at 17-18 (arguing that because AT&T has conditioned its transaction agreement on retaining exclusive rights for NFL Sunday Ticket, the Commission should assess the competitive effects of such an important exclusive deal). 493 Cox Petition at 7. 494 AMC et al. Petition at 10; DISH Petition at 17-18. 495 Cox Petition at 11-13; Cox Nov. 7, 2014, Ex Parte Letter at 2; Cox Dec. 4, 2014, Ex Parte Letter at 3; Cox Dec. 22, 2014, Ex Parte Letter at 3. 496 Joint Opposition at 56-57; see also Revision of the Commission’s Program Access Rules et al., MB Docket No. 12-68, Report and Order, Order on Reconsideration, and Further Notice of Proposed Rulemaking, 27 FCC Rcd 12605, 12631, ¶ 37 (2012) (“2012 Program Access Order”). 497 Joint Opposition at 58. 498 Id. at 56-57. 499 Cox Reply at 4. 500 Id. at 5. 501 Id. Federal Communications Commission FCC 15-94 69 distribution of video programming.502 The Commission may still assess the effect of exclusive contracts on a case-by-case basis, pursuant to which it considers whether an exclusive contract involving affiliated programming may be a prohibited “unfair” act under Section 628(b) of the Act.503 The Commission’s case-by-case approach allows for consideration of the potential procompetitive benefits of individual exclusive contracts, such as promoting investment in new programming, particularly local programming, and permitting MVPDs to differentiate their service offerings.504 Thus, any exclusive contract involving affiliated programming entered into by the combined entity may be challenged on a case-by-case basis as an “unfair” act under Section 628(b) of the Act. 181. We conclude that this case-by-case complaint process will be sufficient to remedy potential competitive concerns regarding exclusive contracts with vertically integrated programmers that may result from this transaction. Neither AT&T nor DIRECTV currently has any exclusive contracts involving affiliated programming. Indeed, as discussed above, neither AT&T nor DIRECTV has a significant amount of vertically integrated programming.505 Cox nevertheless contends that the combined entity intends to pursue an exclusive programming strategy, citing reports that the transaction was conditioned on the renewal of DIRECTV’s exclusive agreement for NFL Sunday Ticket.506 We find Cox’s contention, based in part on the fact that the Applicants conditioned the transaction on the renewal of DIRECTV’s exclusive agreement for NFL Sunday Ticket, to be speculative and unsupported by the record.507 Moreover, there is nothing in the record that demonstrates that competing MVPDs would be 502 2012 Program Access Order, 27 FCC Rcd at 12625, ¶ 31. The exclusive contract prohibition generally banned exclusive contracts for satellite cable programming or satellite broadcast programming between any cable operator and any cable-affiliated programming vendor in areas served by a cable operator. See 2012 Program Access Order, 27 FCC Rcd at 12607, ¶ 1; see also 47 U.S.C. § 548(c)(2)(D). The exclusive contract prohibition in Section 628(c)(2)(D) applied only to “satellite cable programming” and “satellite broadcast programming.” See 47 U.S.C. § 548(c)(2)(D). However, in 2010, the Commission adopted rules providing for the processing of complaints alleging that an “unfair act,” including an exclusive contract, involving terrestrially delivered, cable-affiliated programming violates Section 628(b) of the Act. See Review of the Commission’s Program Access Rules and Examination of Programming Tying Arrangements, MB Docket No. 07-198, First Report and Order, 25 FCC Rcd 746 (2010) (“2010 Program Access Order”), affirmed in part and vacated in part sub nom. Cablevision Sys. Corp. et al. v. FCC, 649 F.3d 695 (D.C. Cir. 2011) (“Cablevision II”). 503 See 47 U.S.C. § 548(b) (prohibiting a cable operator, a satellite cable programming vendor in which a cable operator has an attributable interest, or a satellite broadcast programming vendor from engaging in “unfair” acts, the purpose or effect of which is to hinder significantly or to prevent any MVPD from providing satellite cable programming or satellite broadcast programming to subscribers or consumers); 47 C.F.R. § 76.1001(a) (same). The same case-by-case process applies to Section 628(b) complaints challenging exclusive contracts involving satellite- delivered, cable-affiliated programming and exclusive contracts involving terrestrially delivered, cable-affiliated programming. See 2012 Program Access Order, 27 FCC Rcd at 12640-45, ¶¶ 53-58; 2010 Program Access Order, 25 FCC Rcd at 777-788, ¶¶ 46-61. 504 See 2012 Program Access Order, 27 FCC Rcd at 12608, 12629-31, ¶¶ 2, 35-37. 505 See supra ¶¶ 166-167. 506 Cox Petition at 7. Cox also emphasizes that AT&T has argued previously that exclusive video programming contracts with an affiliated network are anticompetitive. Id. at 8 (citing AT&T 2012 Program Access Order Comments, MB Docket No. 12-68, at 20-22 (filed Dec. 14, 2012) (emphasis added); Cox Nov. 7, 2014, Ex Parte Letter at 2; Cox Dec. 4, 2014, Ex Parte Letter at 3; Cox Dec. 22, 2014, Ex Parte Letter at 3. We note, however, that it has not been alleged, nor is it reflected in the record, that either AT&T or DIRECTV holds an exclusive contract with any of the limited number of affiliated networks involved in this proceeding. 507 To the extent that Cox bases this argument on reports that the transaction was conditioned on the renewal of DIRECTV’s exclusive agreement for its unaffiliated NFL Sunday Ticket package, we note that the Commission has previously found no evidence that exclusive arrangements for unaffiliated programming, including DIRECTV’s exclusive agreements for NFL Sunday Ticket, have harmed competition. See infra ¶ 183. Federal Communications Commission FCC 15-94 70 unable to address concerns regarding any exclusive contracts involving affiliated programming through the existing case-by-case process. 182. With respect to unaffiliated programming, the record does not support a condition barring the combined entity from entering into exclusive contracts. As the Commission has previously stated, exclusive agreements between MVPDs and non-vertically integrated programmers are not captured by the underlying rationale for the program access rules.508 Congress enacted the exclusive contract prohibition of the program access rules based on its determination that combining MVPD market power with ownership of programming resulted in an imbalance of power, which limited the development of competition among MVPDs and restricted consumer choice.509 Congress found that programming networks affiliated with cable operators have an incentive and ability to discriminate against MVPDs with which their affiliated cable operators compete.510 In contrast, unaffiliated programmers may have a lesser incentive to favor one MVPD over another in order to achieve particular competitive outcomes in the video distribution market.511 In general, an unaffiliated programmer is concerned with obtaining the highest price it can for its programming.512 183. The Commission has previously found that there was no evidence to conclude that exclusive arrangements involving unaffiliated programmers have harmed competition in the video distribution market.513 In addition, in prior transactions involving DIRECTV, the Commission has specifically rejected arguments that DIRECTV should be prohibited from entering into exclusive contracts with respect to unaffiliated programming, such as NFL Sunday Ticket.514 There is nothing in the record of this proceeding that would cause us to reach a different conclusion. We recognize that NFL Sunday Ticket is valuable programming and that the combined entity’s size and reach would increase as a result of the transaction. Nevertheless, commenters have offered no evidence that DIRECTV’s exclusive contract for NFL Sunday Ticket has harmed competition in the video distribution market or would harm competition post-transaction. Moreover, we note that competing MVPDs would continue to have access to the NFL games that are aired on the broadcast television stations that they carry, as well as the NFL games aired on ESPN and NFL Network. 184. We note further that, while some commenters raise concerns about the competitive impact of combining AT&T’s wireless service with DIRECTV’s exclusive NFL Sunday Ticket package, Verizon Wireless currently holds exclusive distribution rights for NFL games on mobile devices through the 2017 season.515 There is also no evidence in the record that the transaction would confer an unfair 508 See Liberty Media-DIRECTV Order, 23 FCC Rcd at 3319, ¶ 117; see also 2002 Extension Order, 17 FCC Rcd at 12126-7, ¶¶ 6-7. 509 See 2002 Extension Order, 17 FCC Rcd at 12127, ¶ 7. 510 See id. 511 See Liberty Media-DIRECTV Order, 23 FCC Rcd at 3319, ¶ 117. 512 See id. 513 See Implementation of the Cable Television Consumer Protection and Competition Act of 1992 – Development of Competition and Diversity in Video Programming Distribution: Section 628(c)(5) of the Communications Act: Sunset of Exclusive Contract Prohibition, MB Docket No. 07-26, Report and Order, 22 FCC Rcd 17791, 17843, ¶ 77 (2007) (“2007 Extension Order”), aff’d sub nom. Cablevision Sys. Corp. et al. v. FCC, 597 F.3d 1306 (D.C. Cir. 2010) (“Cablevision I”). See also 2012 Program Access Order, 27 FCC Rcd at 12625, ¶ 37 (noting that even with respect to programming affiliated with an MVPD, the “Commission has recognized that exclusive contracts may result in the procompetitive benefit of allowing MVPDs to differentiate their service offerings”). 514 See Liberty Media-DIRECTV Order, 23 FCC Rcd at 3319-20, ¶ 118 (finding that the record did not support a condition forbidding DIRECTV from entering into exclusive distribution agreements involving unaffiliated programming, such as NFL Sunday Ticket); News Corp.-Hughes Order, 19 FCC Rcd at 600-601, ¶¶ 291-293 (same). 515 See Adam Swanson, Verizon Wireless Extends Streaming-Content Deal with NFL, SNL KAGAN, June 7, 2013. Federal Communications Commission FCC 15-94 71 advantage on the combined entity in obtaining exclusive distribution agreements for unaffiliated programming in the future. Thus, we conclude that the record does not support a condition barring the combined entity from entering into exclusive contracts with unaffiliated programmers. Furthermore, to the extent any competitor believes the combined firm has unfairly entered into an exclusive agreement with an unaffiliated programmer, we note that the Commission has not foreclosed a challenge under Section 628(b) to an exclusive contract involving unaffiliated programming.516 3. Restricting Access to Online Video Content 185. Positions of Parties. DISH expresses concern that the addition of approximately 20 million new video subscribers from DIRECTV would give AT&T an increased incentive to preserve and grow its subscribers by diminishing the attractiveness of rival MVPD and OVD services.517 DISH suggests that the combined entity may accomplish this by using its newly enlarged negotiating leverage to pressure third-party programmers to grant online video rights to the combined entity and to withhold these same rights from rival MVPDs and OVDs.518 To address this potential harm, DISH proposes that the Commission adopt a condition prohibiting the combined entity from entering into or enforcing any contractual provisions with programmers, including any most favored nation (“MFN”) provisions, which restrict the online rights of third-party distributors.519 In addition, ACA calls generally for conditions to prevent the combined entity from interfering with a third-party programmer’s ability to provide any prices, terms, or conditions to another MVPD.520 186. The Applicants argue that programmers have far greater bargaining power than video distributors and that the transaction would not alter relative bargaining power in such a way as to reduce the quantity and variety of programming content that providers have to offer.521 The Applicants assert that the combined entity would not have the leverage to require third-party programmers to withhold or restrict access to programming by competitors, either traditional MVPDs or OVDs.522 Therefore, the Applicants do not believe that any conditions are necessary.523 187. Discussion. We find DISH’s generalized assertion that the combined entity would have an increased incentive and ability to force third-party programmers into withholding online video rights from rival MVPDs and OVDs to be unsupported by the record. Based on our review of the totality of the 516 See Implementation of Section 302 of the Telecommunications Act of 1996, Open Video Systems, CS Docket No. 96-46, Second Report and Order, 11 FCC Rcd 18223, 18319, ¶ 184 (1996) (“[C]able operators, common carriers providing video programming directly to subscribers and open video system operators are not generally restricted from entering into exclusive contracts with non-vertically integrated programmers. Nonetheless, as we found in the DBS Order, our finding herein does not preclude an aggrieved party from seeking relief in an appropriate case under other provisions of Section 628 and the Commission’s rules thereunder.” (citing Implementation of the Cable Television Consumer Protection and Competition Act of 1992, Memorandum Opinion and Order on Reconsideration of the First Report and Order, 10 FCC Rcd 3105, 3121, ¶ 33, 3126-27, ¶ 40 (1994))); see also 2010 Program Access Order, 25 FCC Rcd at 779 n.191 (“We do not reach any conclusions in this Order, nor do we foreclose potential complaints, regarding other acts that may be ‘unfair methods of competition or unfair acts or practices’ under Section 628(b). For example, the rules established by this Order do not address exclusive contracts between a cable operator and a non-cable-affiliated programmer.”). 517 DISH Petition at 14. 518 Id. at 14-16; see also WGAW Reply at 16-17 (quoting DISH’s Petition and noting potential for AT&T to disadvantage competing OVDs by negotiating restrictive distribution agreements). 519 DISH Petition at 30. 520 ACA Reply at 57-58. 521 Joint Opposition at 50-51. 522 Id. at 52 n.188. 523 Id. at 49. Federal Communications Commission FCC 15-94 72 record we cannot find that DIRECTV has been able to limit consumers’ access to distribution of video programming online or that with an additional approximately 6 million U-verse video subscribers the combined entity would be better positioned to impede the ability of other MVPDs or OVDs to attract and retain subscribers. Nor does the record contain evidence that AT&T has pursued or, post-transaction, intends to pursue such a strategy with respect to programming contracts. Therefore, we find DISH’s argument to be unsubstantiated by the record and will not impose the conditions requested by DISH and ACA.524 4. Forcing Competitors to Compensate Programmers for Reduced Payments from the Applicants 188. Positions of the Parties. Several commenters assert that the combined entity’s increased size and market power would give it the ability to negotiate substantial volume discounts for programming, leading programmers to offset these discounts by charging smaller MVPDs higher prices.525 These commenters assert that this offset would competitively disadvantage the combined entity’s rivals and result in higher prices for their customers.526 Cox proposes a condition prohibiting the combined entity from entering into programming agreements that result in the combined entity receiving an “unreasonable discount” for programming, on a per subscriber basis, as compared to the rates a programmer charges to any other MVPD.527 Cox suggests that the Commission establish a percentage discount that would be presumptively unreasonable and enforce such a condition by requiring the combined entity to include in its new programming contracts a provision in which both the combined entity and the programmer certify compliance with the discount limitation.528 Cox maintains that such a condition would allow the combined entity to secure programming deals only at a reasonable discount compared to what mid-sized and smaller MVPDs are forced to pay.529 189. The Applicants respond that there is no support for the assertion that the savings realized by the combined entity would increase competitors’ costs.530 They state that programmers already negotiate aggressively to obtain the best possible rates from each individual MVPD based on its size and the value that it brings to the programmer.531 The Applicants further state that programmers would not need to recoup “lost” revenues through higher prices from other MVPDs because the transaction would create new revenue opportunities for programmers by, for example, expanding the demand for content 524 We acknowledge that commenters did not have access to all of the record related to this argument. See CBS Corp., 785 F.3d at 701-03. While we would have preferred for the commenters to have full access to the record, given the unusual delays that resulted from litigation related to access to certain documents and the unique circumstances before us, we determine that the public interest is best served by completing our review of the Application. We reach this conclusion after balancing the commenters’ ability to have full access to the record against the Commission’s careful review of the record, our interest in bringing the benefits of this transaction to the public, and fairness to the Applicants. 525 ACA Comments at 18-19; Biglaiser Statement at 14-16; Cox Petition at 13-14; DISH Petition at 10-13; WGAW Petition at 14; Reply Comments of COMPTEL, MB Docket 14-90, at 6 (filed Jan. 7, 2015) (“COMPTEL Reply”); Cox Reply at 7-9; WGAW Reply at 11-12; Cox Nov. 7, 2014, Ex Parte Letter at 2. 526 ACA Comments at 19-20; Cox Petition at 16; DISH Petition at 13-14; WGAW Petition at 14; ACA Reply at 11- 13; COMPTEL Reply at 6; Cox Reply at 8; WGAW Reply at 12; Cox Nov. 7, 2014, Ex Parte Letter at 2. 527 Cox Petition at 17; Cox Nov. 7, 2014, Ex Parte Letter at 2; Cox Dec. 4, 2014, Ex Parte Letter at 3; Cox Dec. 22, 2014, Ex Parte Letter at 3. 528 Cox Petition at 17. 529 Id. 530 Joint Opposition at 53; Katz Reply Decl. ¶ 29; Conditions Ex Parte Presentation at 12. 531 Joint Opposition at 53; Katz Reply Decl. ¶ 29. Federal Communications Commission FCC 15-94 73 and increasing programmers’ advertising reach.532 The Applicants state that Cox’s proposal on volume discounts would prevent the combined entity from fully realizing substantial cost savings that would benefit consumers and that there is no justification to prevent the combined entity from achieving these “pro-competitive” consumer benefits.533 Cox counters that advertising revenues for basic cable networks advertising to the Applicants’ customers would have to increase by 25 percent or more to offset the reduction in programmers’ revenues from AT&T’s claimed content cost savings.534 190. Discussion. We conclude that commenters have failed to substantiate a harm that requires a condition establishing a limit on the volume discounts that the combined entity may negotiate for programming. For example, ACA claims that programmers would have to offset volume discounts to the combined entity with increased prices to smaller MVPDs in order to ensure that financial market expectations are realized.535 Commenters have failed to offer evidence, however, to show that this behavior has occurred when other MVPDs, including DIRECTV, have received volume discounts. 191. Further, Cox asserts that the structure of the video distribution marketplace has caused a severe imbalance between the prices paid for programming by the largest MVPDs and those paid by mid- sized and small MVPDs, citing comments it filed in a pending program access rulemaking proceeding.536 To the extent that there potentially is an industry-wide public interest harm associated with volume discounts as such, it has not been established on the record before us, and it would be beyond the scope of this proceeding in any event as it is not transaction specific.537 192. Based on all of these factors, we conclude that the record does not support a finding that programmers would offset any volume discounts received by AT&T post-transaction by increasing programming costs for smaller MVPDs such that we could justify adoption of the proposed condition.538 B. Lack of Regulatory Parity 193. Positions of the Parties. Cox observes that the basic service tier requirements that apply to cable operators do not apply to DIRECTV’s satellite service and that AT&T has contested whether such “cable operator” obligations apply to its U-verse video service.539 These basic tier requirements, it 532 Joint Opposition at 53-54; Katz Reply Decl. ¶ 33. 533 Conditions Ex Parte Presentation at 12. 534 Cox Reply at 8; Cox Nov. 7, 2014, Ex Parte Letter at 2. See infra n.924. 535 ACA Comments at 20; Biglaiser Statement at 15-16. 536 Cox Petition at 13, nn.26-27; Cox Reply at 8. See Revision of the Commission’s Program Access Rules et al., MB Docket No. 12-68, Notice of Proposed Rulemaking, 27 FCC Rcd 3413, 3466-68, ¶¶ 98-100 (2012) (“2012 Program Access NPRM”) (seeking comment on whether the program access rules adequately address potentially discriminatory volume discounts and, if not, how the rules should be revised to address these concerns); see also Cox 2012 Program Access NPRM Comments, MB Docket No. 12-68, at 3-7 (filed June 22, 2012); Cox 2012 Program Access NPRM Reply, MB Docket No. 12-68, at 2 (filed July 23, 2012); Cox 2012 Program Access Order Comments, MB Docket No. 12-68, at 1-5 (filed Dec. 14, 2012); Cox 2012 Program Access Order Reply, MB Docket No. 12-68, at 1-2 (filed Jan. 14, 2013). 537 We also note that the Commission’s program access rules contemplate that a complaint may be filed challenging volume-based pricing in certain circumstances. On the filing of such a complaint, a cable-affiliated programmer may be required “to demonstrate that such volume discounts are reasonably related to direct and legitimate economic benefits reasonably attributable to the number of subscribers . . . but may also identify non-cost economic benefits related to increased viewership.” 47 C.F.R. § 76.1002(b)(3) note. 538 See Comcast-AT&T Order, 17 FCC Rcd at 23269, ¶ 65 (rejecting arguments that programmers would offset volume discounts realized by the combined firm by raising rival MVPDs’ costs). 539 Cox Petition at 26 n.65 (citing AT&T Services MB 13-140 Comments at 5 (“AT&T’s U-verse TV service is an IP-based MVPD service, and not a ‘cable service.’”)). Under Section 623 of the Act, cable operators are required to offer an entry-level basic service which includes: (1) all commercial and noncommercial local broadcast stations (continued….) Federal Communications Commission FCC 15-94 74 states, impose costs on Cox and other cable operators and reduce the choices that they are able to offer customers.540 Cox asserts that without these requirements, the combined entity would have unfair cost, pricing, and packaging advantages over mid-sized cable operators to the detriment of consumers.541 Accordingly, Cox proposes a condition requiring the combined entity to provide all television broadcast stations to its video customers regardless of whether it is distributing programming via its U-verse video platform, its satellite platform, or any wireless video platform the new company develops.542 194. Cox further argues that the transaction would create opportunities and incentives for the combined entity to harm competition for Multiple Dwelling Unit (“MDU”) services.543 Cox notes that cable operators and common carriers or their affiliates are expressly prohibited under the Commission’s 2007 MDU Order from entering into exclusive agreements to serve MDUs, but satellite providers are not.544 Cox asks the Commission to confirm that any video programming services offered by the combined entity to MDUs – whether furnished through AT&T’s broadband service or satellite – are covered by the restrictions on exclusive contracts imposed by the 2007 MDU Order.545 195. The Applicants respond that the basic tier issues are unrelated to the transaction.546 The Applicants also assert that the MDU issue is an open question in a pending rulemaking proceeding and is properly addressed in that proceeding.547 196. Discussion. We decline to impose a condition requiring the combined entity to comply with the basic service tier requirements. Cox’s concerns regarding these requirements are unrelated to the proposed transaction and, rather, are focused on a broader concern about the applicability of the basic tiering requirements. Thus, we find that this proceeding is not the appropriate venue to address basic tier issues. 197. We also conclude that the combined entity will be subject to the prohibition on exclusive contracts to serve MDUs imposed in the 2007 MDU Order.548 The 2007 MDU Order made clear that common carriers or their affiliates that provide video programming directly to subscribers under Section 628(j) of the Act are prohibited from executing or enforcing contracts that give them the exclusive right to provide video programming services to MDUs.549 As discussed above, AT&T is a common carrier and, following the transaction, DIRECTV will be a wholly owned subsidiary of AT&T. Accordingly, the (Continued from previous page) entitled to carriage under the Communication Act’s must-carry provisions; (2) any public, educational, and governmental access channels; and (3) any other local broadcast station provided to any subscriber. See 47 U.S.C. § 543(b)(7)(A). Cable operators may also offer additional non-broadcast channels on their basic service tiers. See id. § 543(b)(7)(B). 540 Cox Petition at 26. 541 Id. at 27; Cox Reply at 7. 542 Cox Petition at 27; Cox Reply at 6-7. 543 Cox Petition at 28; Cox Reply at 12. 544 Id.; see Exclusive Service Contracts for Provision of Video Services in Multiple Dwelling Units and Other Real Estate Developments, MB Docket No. 07-51, Report and Order and Further Notice of Proposed Rulemaking, 22 FCC Rcd 20235, ¶ 51 (2007) (“2007 MDU Order”), aff’d, NCTA v. FCC, 567 F.3d 659 (D.C. Cir. 2009); 47 C.F.R. § 76.2000. 545 Cox Petition at 28; Cox Reply at 12-13. 546 Joint Opposition at 59 n.216. 547 Id. at 71-72. 548 See also supra Section X.F. (addressing arguments that the transaction would result in the loss of DIRECTV as a potential partner for MDU broadband entrants). 549 See 2007 MDU Order, 22 FCC Rcd at 20254, ¶ 40; see also 47 C.F.R. § 76.2000. Federal Communications Commission FCC 15-94 75 combined entity will be prohibited from entering into or enforcing exclusive agreements to serve MDUs.550 C. Potential Harm to Subscribers’ Access to OVD Services 198. We consider whether the transaction would harm the public interest by increasing the ability or incentive of the combined firm to use its broadband network to limit competition from unaffiliated OVDs. 199. AT&T currently provides broadband Internet access service to approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] subscribers551 and MVPD services to approximately 6 million subscribers.552 Of AT&T’s residential broadband subscribers, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] receive broadband Internet access service at download speeds of 10 Mbps or greater, representing [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent of all subscribers nationally who receive service at those speeds.553 Today, AT&T generally views its MVPD offering as ancillary to these broadband services.554 In contrast, DIRECTV, which has approximately 20 million video subscribers, does not offer facilities-based broadband services.555 Thus, because the proposed transaction would not increase AT&T’s number of broadband subscribers or the geographic footprint of its broadband networks, the transaction would not increase AT&T’s ability to discriminate against OVDs. 200. However, commenters have expressed concern that the transaction would increase AT&T’s incentive to discriminate against unaffiliated OVDs in order to favor its own enhanced bundle of MVPD and OVD services, particularly through the use of discriminatory data allowances or “caps,” and degradation of interconnection with AT&T’s broadband network. In addition, the Applicants have stated that a benefit of the transaction is the combined entity’s ability to develop its own online video offerings.556 As discussed below, we find that the transaction may increase the Applicants’ incentive and ability to use data allowances to discriminate in favor of their own, affiliated online offerings, and we impose a condition to prevent the combined entity from discriminatory data practices. We also impose 550 See Exclusive Service Contracts for Provision of Video Services in Multiple Dwelling Units and Other Real Estate Developments, Second Report and Order, MB Docket No. 07-51, 25 FCC Rcd 2460, 2474-75, ¶¶ 40-41 (2010) (“2010 MDU Order”) (concluding that the prohibition on exclusive contracts to serve MDUs applied to a private cable operator that was an affiliate of a common carrier). As a wholly owned subsidiary of AT&T post- transaction, DIRECTV will be an “affiliate” of a common carrier pursuant to Section 3 of the Act. See 47 U.S.C. § 153(2) (defining “affiliate” as “a person that (directly or indirectly) owns or controls, is owned or controlled by, or is under common ownership or control with, another person,” with “own” meaning “to own an equity interest (or the equivalent thereof) of more than 10 percent”). Thus, post-transaction, both AT&T and DIRECTV will be subject to the prohibition on exclusive contracts to serve MDUs pursuant to the 2010 MDU Order. Accordingly, we need not address the broader issue of whether AT&T’s U-verse service constitutes a cable service. 551 See AT&T April 21, 2015, Form 477 Data Filing; AT&T Updated Response to Sept. 9, 2014, Information Request, Exhibit 5.b.1 – updated. 552 See AT&T Updated Response to Sept. 9, 2014, Information Request, Exhibit 5.b.1 – updated. 553 As of December 2013, AT&T had fewer than [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] subscribers with broadband service with speeds of 25 Mbps or greater. See AT&T April 21, 2015, Form 477 Data Filing. In a filing submitted in this proceeding, AT&T reports that as of February 2015 it had fewer than [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] subscribers of broadband service with speeds of 25 Mbps or greater. OVD Ex Parte Presentation at 2. 554 See Section IX.B. 555 See Application at 13. 556 See id. at 48-49. Federal Communications Commission FCC 15-94 76 conditions to enable the Commission to monitor the Applicants’ future interconnection arrangements and certain performance metrics. 1. Increased Incentive to Discriminate Against Unaffiliated OVDs 201. Positions of the Parties. Several commenters express concern that the transaction would increase AT&T’s incentive to discriminate against unaffiliated OVDs.557 They note that post-transaction, AT&T would become the nation’s second largest MVPD558 and, as a consequence, AT&T would have a greater incentive to protect its video revenues from future OVD competition.559 WGAW notes that the bundling of video and broadband is a key rationale given for the transaction but that such bundles would remain attractive only if OVDs are not competitive substitutes for the combined entity’s video service, which gives the combined entity increased incentive to hinder OVD competition.560 Cogent Communications Group, Inc. (“Cogent”) acknowledges that OVD services are complementary to AT&T’s standalone broadband, but it asserts that OVD services are a threat to DIRECTV’s video services.561 202. Commenters maintain that the Applicants’ post-transaction incentives are not limited by the possibility that consumers would switch from AT&T broadband to another provider in the event their OVD service is degraded. Instead, commenters argue that when OVD services are subjected to discriminatory treatment, an OVD subscriber is more likely to switch to an alternative OVD than to an alternative broadband provider because of: (i) the significantly lower costs associated with switching OVDs as compared to broadband providers; (ii) the substantial number of alternative OVDs with overlapping content; and (iii) the lack of alternative broadband providers.562 203. The Applicants reject the assertion that they have an incentive to discriminate against unaffiliated OVDs. The Applicants assert that OVD services are a complement to AT&T’s broadband business and that the combined entity’s overriding incentive would be to grow that business.563 The Applicants also assert that AT&T has made substantial investments in its broadband network, which would remain competitive only if it provides customers with seamless and high-quality access to the full range of online video services.564 Thus, the Applicants state that rather than discriminating against online 557 See Cogent Comments at 6-11; DISH Petition at 3, 25-27; Franken Comments at 1-2; Netflix Comments at 11- 14; Public Knowledge-ILSR Petition at 5-6, 8; WGAW Petition at 4, 19-25. 558 See Cogent Comments at 6-7; WGAW Petition at 4, 19; Reply Comments of Cogent Communications Group, Inc., MB Docket 14-90, at 3 (filed Jan. 7, 2015) (“Cogent Reply”); WGAW Reply at 14, 18. 559 See Cogent Comments at 6-7; DISH Petition at 26; Franken Comments at 1-2; Netflix Comments at 13, 28; Public Knowledge-ILSR Petition at 5-6, 8; WGAW Petition at 4; Cogent Reply at 3-4; Reply Comments of DISH Network Corp., MB Docket 14-90, at 5-6 (filed Jan. 7, 2015) (“DISH Reply”); Reply Comments of Netflix, Inc., MB Docket 14-90, at 3 (filed Jan. 7, 2015) (“Netflix Reply”); Reply Comments of Public Knowledge to Opposition to Petition to Deny, MB Docket 14-90, at 5-6 (filed Jan. 7, 2015) (“Public Knowledge Reply”); WGAW Reply at 2, 14, 18. 560 See WGAW Reply at 15. 561 Cogent Reply at 5; cf. WGAW Reply at 14 (“Online video offerings currently serve as a complement to MVPD service but as OVDs invest in high-budget original content, their growth could facilitate a decline in MVPD subscribers.”). 562 See Cogent Comments at 15 n.48; Netflix Comments at 17-18, 29-32; Cogent Reply at 9-11; Public Knowledge Reply at 5; WGAW Reply at 17. 563 See Application at 79; Joint Opposition at 4-5, 34-37; Katz Reply Decl. ¶¶ 72-78. 564 See Joint Opposition at 4-5 (“To drive broadband adoption, AT&T has long supported unfettered access to OTT services, and it has made significant investments to ensure that customers enjoy all that the Internet has to offer. With this transaction, those investments will only increase . . . .”); id. at 34-37; Katz Reply Decl. ¶¶ 72-75; see also Application at 82. Federal Communications Commission FCC 15-94 77 video, they are encouraging the growth of online video services and providers, including by investing in their own OVD services.565 204. In the Applicants’ view, any restrictions that AT&T attempts to impose on the ability of its broadband customers to access online video services would likely reduce the quality of its broadband Internet access service.566 The Applicants point out that reducing the quality of AT&T’s broadband Internet access service would, in turn, reduce the total sales of that service, both as part of bundles and on a standalone basis.567 According to the Applicants, a strategy of degrading broadband would be profitable only if it has the effect of encouraging consumers to purchase more video services from the combined entity.568 The Applicants contend, however, that the most likely response by AT&T’s standalone broadband customers would be to seek an alternative access provider.569 In addition, the Applicants assert that the BH Simulation demonstrates that a degradation of AT&T’s broadband service would also reduce AT&T’s video sales.570 Accordingly, the Applicants conclude that a strategy of discrimination against OVDs would be unprofitable and, thus, that the combined entity would not have an incentive to foreclose OVD competition.571 205. Discussion. We conclude that post-transaction AT&T has an increased incentive to discriminate against unaffiliated OVDs.572 As we have found in other proceedings, “broadband providers have incentives to interfere with and disadvantage the operation of third-party Internet-based services that compete with their own services.”573 We disagree that the Applicants’ incentive to attract and retain broadband subscribers precludes any incentives to engage in conduct that hinders consumers’ access to unaffiliated OVDs.574 While we acknowledge that the documentary evidence supports the Applicants’ 565 See Application at 79; Joint Opposition at 37. 566 See Katz Reply Decl. ¶ 74. 567 See id. ¶ 77. The Applicants also state that if the combined entity were to degrade its customers’ online video experience, it would lose not only the profits from its broadband service, but also the profits from the video and voice components of its double-play and triple-play bundles. See id. ¶ 75; see also OVD Ex Parte Presentation at 7- 8 (stating that degradation of OVD access would decrease demand especially for AT&T DSL and DSL service bundles). 568 See Katz Reply Decl. ¶ 77. 569 See id. 570 See id. ¶ 78. Dr. Katz models broadband degradation as a unilateral increase in the quality-adjusted price of the service. We were able to reproduce his reported findings with our modified version (described in Appendix C) of the BH Simulation and to verify their robustness under different modeling assumptions regarding pricing behavior. We note, however, that the BH Simulation – which does not capture dynamic incentives and also does not allow either granular adjustments to product quality or a precise identification of consumers’ substitution patterns between video services provided by OVDs and MVPDs – is of limited use in evaluating the profitability of a foreclosure strategy. 571 See id. 572 The Commission has previously recognized the incentive of Internet access providers such as AT&T to discriminate against unaffiliated OVDs. See Annual Assessment of the Status of Competition in the Market for the Delivery of Video Programming, MB Docket No. 07-269, Fourteenth Report, 27 FCC Rcd 8610, 8731, 8733, ¶¶ 271, 274 (2012) (“MVPDs have the ability and incentive to degrade the broadband service available to unaffiliated OVDs.”) (“Fourteenth Annual Report”); Preserving the Open Internet, GN Docket No. 09-191, Report and Order, 25 FCC Rcd 17905, 17916, ¶ 22 (“2010 Open Internet Order“) (“[B]roadband providers have incentives to interfere with the operation of third-party Internet-based services that compete with the providers’ revenue-generating telephony and/or pay-television services.”). 573 2015 Open Internet Order, 30 FCC Rcd at 5662, ¶ 140. 574 While the Applicants use the BH Simulation to predict that a strategy of degrading broadband would reduce broadband and video sales, their analysis ignores some important aspects of OVD degradation that distinguish it (continued….) Federal Communications Commission FCC 15-94 78 position that AT&T has traditionally viewed its video product as ancillary to its more profitable broadband services,575 post-transaction AT&T would have different incentives to use strategies that limit consumers’ access to OVD services in order to favor the DIRECTV video product or the combined entity’s online video products. Again, as we stated in prior proceedings, many end users may have limited choice among broadband providers576 and switching costs can be a significant impediment to the ability of consumers to change broadband providers.577 2. Potential Levers for Discrimination Against Unaffiliated OVDs 206. Given our finding that the transaction has the potential to increase the Applicants’ incentive to discriminate against OVDs, we analyze the potential levers that would enable the Applicants to engage in such discrimination and consider whether any conditions are warranted to address potential public interest harms. 207. As noted above, currently AT&T provides broadband Internet access services to 14.5 million subscribers. Commenters assert that post-transaction the Applicants would use these services to harm OVDs in two principal ways: (1) usage-based data restrictions or data caps; and (2) interconnection fees.578 a. Data Caps 208. Positions of the Parties. Commenters contend that generally applied data caps can be a significant problem for broadband subscribers who rely on OVDs for video entertainment.579 For example, Netflix explains that some subscribers could consume their entire monthly data allotment after (Continued from previous page) from quality reductions and price increases. See supra ¶ 204. We conclude that the Applicants’ analysis understates AT&T’s incentive to engage in a strategy of OVD degradation, particularly over the long term. 575 ATT-FCC-00460915, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ; ATT-FCC- 01154928, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ; ATT-FCC-00518651, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 576 See 2010 Open Internet Order, 25 FCC Rcd at 17921, ¶ 25; see also id. at 17923, ¶ 32 (“most residential end users today have only one or two choices for wireline broadband Internet access service”); see also 2015 Open Internet Order, 30 FCC Rcd at 5631, ¶ 81 n.134 (“… data suggests that meaningful alternative broadband options may be largely unavailable to many Americans, further limiting the ability to switch providers. Based on the submissions from various commenters, it appears that between 65% and 70% of households have at most two options for high speed Internet access.”). 577 See Netflix Comments at 17-18, 29-32 (stating that consumers face significant switching costs when changing ISPs, including early termination fees; the inconvenience of ordering, installing and setup, and associated deposits or fees; the difficulty in returning equipment and the cost of replacing incompatible customer-owned equipment; the risk of temporarily losing service; problems learning how to use the new service; and the loss of a provider specific email address or website); see also Cogent Comments at 15 n.48; Cogent Reply at 9-11; Public Knowledge Reply at 5; WGAW Reply at 17. The Commission has also previously recognized the significant costs consumers face in switching broadband providers. See 2010 Open Internet Order, 25 FCC Rcd at 17924-25, ¶ 34; see also 2015 Open Internet Order, 30 FCC Rcd at 5631-62, ¶ 81 (“[t]he broadband provider’s position as gatekeeper is strengthened by the high switching costs consumers face when seeking a new service. … These costs may limit consumers’ willingness and ability to switch carriers, if such a choice is indeed available.”). AT&T’s internal documents confirm this view. When Netflix was suffering from congestion in interconnecting to AT&T’s last-mile network, AT&T’s Chief Technology Officer stated [BEGIN HIGHLY CONF. INFO] [END HIGHLY CONF. INFO] See ATT-FCC-02459548, [BEGIN HIGHLY CONF. INFO] [END HIGHLY CONF. INFO] . 578 See generally DISH Petition at 26-27; Public Knowledge-ILSR Petition at 8; Netflix Comments at 14-25, 27; WGAW Petition at 20, 22-24; Cogent Comments at 11-16; COMPTEL Reply at 7; WGAW Reply at 17-20. 579 See Netflix Comments at 27. Federal Communications Commission FCC 15-94 79 just a single weekend of “binge watching” online video content.580 Commenters also argue that the combined entity could impose discriminatory data caps that exempt its own, or affiliated, online video services, which would discourage consumers from accessing unaffiliated OVD services.581 Commenters request that, to the extent that the combined entity utilizes usage-based billing for its broadband Internet service, it should be prohibited from exempting any video service from such usage-based billing.582 209. The Applicants respond that data caps ensure that the lightest users of broadband do not subsidize the heaviest users.583 The Applicants also state that, to the extent that AT&T has implemented data caps, it has done so at levels sufficient to accommodate the broadband needs of most customers.584 AT&T notes that it has been transparent about its data allowance policies and practices, in compliance with the 2010 Open Internet Order.585 AT&T also states that it would voluntarily commit to abide by the 2010 Open Internet Order for three years after closing586 and that such commitment empowers the Commission to enforce the rule prohibiting unreasonable discrimination.587 Further, the Applicants stress that the Commission declined to impose across-the-board prohibitions of usage-based billing when it issued the 2015 Open Internet Order and that the Commission instead opted to look at data allowance policies on a case-by-case basis.588 210. Discussion. As discussed below, the record establishes that the Applicants seek to improve the combined entity’s ability to offer its own, or affiliated, online video content. We note that today AT&T imposes usage-based data caps on its wireline broadband customers and is alone among the 580 See id. 581 See DISH Petition at 26-27 (“AT&T could impose restrictive data caps for data that travels over the public Internet portion of its pipe, while exempting any AT&T/DIRECTV online services from those caps. If DISH online video services are subject to a low monthly data cap, this could depress consumer interest in accessing those services, while at the same time driving consumers to use AT&T/DIRECTV services because they are exempted from the data cap.”); Public Knowledge-ILSR Petition at 8 (expressing concern that AT&T could “discriminate in transmitting lawful network traffic over a consumer’s Internet connection, prioritize its own video services, give its own video services preferential treatment with respect to caps, tiers, metering, or other usage-based pricing, or measure, count, or otherwise treat its own video services differently than other over-the-top video services”); Netflix Comments at 25, 27; WGAW Petition at 20, 22-23. 582 Cogent et al. May 12, 2015, Ex Parte Letter at 6. 583 See Joint Opposition at 38-39; Katz Reply Decl. ¶ 68 (“It can also be an efficient mechanism for recovering common network costs by having the users who derive the greatest benefit from the network make the greatest contribution toward those costs.”). 584 See Joint Opposition at 39. 585 Id.; see also Conditions Ex Parte Presentation at 6 (stating that AT&T’s usage-based pricing for wireline broadband includes data allowances that accommodate a majority of customers and that are similar to those offered by other broadband providers). 586 Application at 51. 587 See Joint Opposition at 39 n.126; Katz Reply Decl. ¶ 70. Several commenters claim that the combined entity’s commitment to abide by the Commission’s 2010 Open Internet Order protections is not sufficient to address the concerns with discriminatory data caps. See, e.g., Netflix Comments at 34 (stating that the combined entity should also be prohibited from exempting its affiliated services from any data cap applicable to any of its services); DISH Petition at 27-30 (contending that the Commission should require the combined entity to provide its U-verse broadband Internet access service at reasonable and non-discriminatory wholesale rates and stating that this condition would, among other things, “reduce AT&T’s incentive to block or degrade competing online content traveling on AT&T’s U-verse residential broadband network (such as DISH’s IPVOD), because consumers could switch to another broadband access provider that did not engage in such practices”). 588 Conditions Ex Parte Presentation at 5-6. Federal Communications Commission FCC 15-94 80 major ISPs in imposing such data allowances for all subscribers.589 We find that as the combined entity expands its online offerings, it will have an increased incentive to limit subscriber demand for competitors’ online video content, including through data caps that discriminate against third-party content by exempting its own content from the data cap. Indeed, AT&T’s internal documents indicate that [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .590 211. AT&T and DIRECTV have each developed online video programming offerings and, prior to announcing this transaction, stated that they will expand such offerings.591 For example, in 2014, AT&T announced that it will join The Chernin Group, which manages and invests in media businesses, to invest over $500 million to acquire and launch online video services.592 In addition, DIRECTV’s internal documents describe [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .593 In December 2014, DIRECTV launched Yaveo, its first Internet-only subscription video service, aimed at U.S. Hispanic consumers, in conjunction with partners including Univision Communications and Viacom. When the acquisition was announced, DIRECTV’s Chief Revenue and Marketing Officer, Paul Guyardo, stated, “Yaveo gets DirecTV into the OTT business, and we’re excited to start with a compelling Spanish- language service targeted to the Hispanic community,” and “[w]e’ll learn a great deal, use the findings to grow and improve the Yaveo platform and expand our OTT offering over time.”594 212. AT&T also has stated that its acquisition of DIRECTV would improve its ability to introduce OVD services, offering a number of reasons why the transaction would improve its online video capabilities.595 First, the Applicants claim that, post-transaction, the combined entity would be a more desirable partner for development of innovative online video arrangements due to its combination of assets, including a nationwide base of video subscribers, a nationwide state-of-the-art wireless network, a 21-state wireline broadband network, and DIRECTV’s expertise in customer interfaces for video 589 Id. at 6. 590 See ATT-FCC-00159899, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 591 ATT-FCC-03405559, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ; DTVFCC- 03726077, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 592 The Chernin Group and AT&T Create New Venture to Acquire, Invest In and Launch Online Video Businesses, AT&T NEWSROOM, April 22, 2014, available at http://about.att.com/story/the_chernin_group_and_att_create_new_venture_to_acquire_invest_in_and_launch_onlin e_video_businesses.html (visited June 24, 2015); see also Application at 77; Lee Decl. ¶ 48 (noting that AT&T recently entered into an arrangement with The Chernin Group to acquire, invest in, and launch online video services, which will “generate additional capabilities to develop and market innovative OTT services that can be delivered through all types of wired and wireless devices,” and that “[t]he programming for those services will likely include ad-supported and subscription-based video-on-demand channels, as well as streaming services”). Netflix contends that AT&T’s $500 million joint venture with The Chernin Group to invest in online video services “provides further incentive for AT&T to favor affiliated OTT services over those provided by third parties.” Netflix Reply at 3 n.6. See Cogent Reply at 6-7. 593 DTVFCC-00031845, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 594 See Todd Spangler, DirecTV Launches First Over-the-Top Video Service, Yaveo, for U.S. Hispanic Audiences, VARIETY.COM, Dec. 22, 2014, available at http://variety.com/2014/digital/news/directv-launches-first-over-the-top- video-service-yaveo-for-u-s-hispanic-audiences-1201385186/ (visited June 24, 2015). 595 See Stankey Decl. ¶ 59 (“The transaction will dramatically improve AT&T’s ability to develop OTT services in a number of important ways.”); Guyardo Decl. ¶ 15 (noting DIRECTV’s online video developments and stating that “[w]e hope through this transaction to be able to combine our efforts with those of AT&T, and that the combination will in turn result in a more comprehensive OTT response”); AT&T Response to Sept. 9, 2014, Information Request at 225 (“the transaction will significantly enhance AT&T’s ability to promote the development of over-the-top (‘OTT’) services”). Federal Communications Commission FCC 15-94 81 services.596 Second, they argue that the combined entity’s aggregate spending on content would increase its attractiveness to content providers and allow it to secure more innovative content rights arrangements.597 Third, the Applicants note that DIRECTV’s in-house development team of engineers has substantial expertise in encoding digital content and developing interfaces for consumers to interact with online video and has already deployed this technology to facilitate live, real-time online streaming of linear content.598 Post-transaction, the Applicants claim that the combined entity “ [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .”599 Fourth, noting that DIRECTV has production facilities and efforts underway to produce original programming, the Applicants state that the combined entity, with its increased scale, would be better positioned both to launch and to market original programming and to fund more investment in new programming ventures.600 Fifth, from an operational standpoint, the Applicants note that [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] and “facilitat[ing] the development of innovative OTT services.”601 213. Although the increased online offerings that may arise as a result of the transaction have the potential to benefit consumers, they also will add to the Applicants’ incentive to shield affiliated content from competition. We find that the transaction will increase the combined entity’s incentive to discriminate against unaffiliated OVDs602 and online video programming to protect both its traditional video services and its OVD services. Therefore, we impose as a condition of this transaction conditions that prohibit certain discriminatory usage-based allowances. b. Interconnection 214. Positions of the Parties. Commenters contend that AT&T has the ability to discriminate against unaffiliated OVDs by allowing congestion to build up at the interconnection points with AT&T’s last-mile network and charging OVDs for access to the network, thereby raising its rivals’ costs of conducting business.603 Netflix claims that recently, AT&T allowed its interconnection points with Netflix to become congested, and Cogent claims that AT&T in the past has refused to upgrade capacity at interconnection points between AT&T’s last-mile broadband network and Netflix’s transit provider (Cogent),604 which had a significant detrimental effect on the ability of AT&T’s DSL and U-verse 596 See Stankey Decl. ¶ 59 (“Because AT&T has both wireline and wireless broadband networks to complement its MVPD offerings, it is especially well-positioned to offer content providers a coordinated set of platforms through which to reach their potential viewers, wherever those viewers want to be.”). 597 See id. ¶ 61. 598 See id. ¶ 62. 599 Id. See also AT&T Response to Sept. 9, 2014, Information Request, Exhibit 68.e.1, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 600 See Stankey Decl. ¶ 63. 601 Application at 37-38. See Moore Decl. ¶ 23; DIRECTV Response to Sept. 9, 2014, Information Request at 72- 73. 602 The Commission has previously recognized the incentive of Internet access providers such as AT&T to discriminate against OVDs. See Fourteenth Annual Report, 27 FCC Rcd at 8731, ¶ 271, 8733, ¶ 274 (“MVPDs have the ability and incentive to degrade the broadband service available to unaffiliated OVDs.”); 2010 Open Internet Order, 25 FCC Rcd at 17916, ¶ 22 (“[B]roadband providers have incentives to interfere with the operation of third-party Internet-based services that compete with the providers’ revenue-generating telephony and/or pay- television services.”). 603 See Cogent Comments at 11-16; Netflix Comments at 14-25; WGAW Petition at 24; COMPTEL Reply at 7; WGAW Reply at 17-20. 604 See Netflix Comments at 23; see also Cogent Comments at 12-13; Cogent Reply at 14. Cogent states that AT&T and Cogent peered historically on a settlement-free basis. After Cogent started carrying Netflix traffic, however, (continued….) Federal Communications Commission FCC 15-94 82 customers to access the Netflix OVD service.605 Netflix asserts that AT&T refused to allow Netflix’s Open Connect content delivery network (“CDN”) platform to interconnect without payment to AT&T.606 Netflix also asserts that it paid AT&T to directly connect to avoid degradation of its service.607 Netflix and Cogent claim that AT&T’s strategy is to refrain from augmenting capacity on settlement-free routes in order to force traffic onto paid routes.608 They also maintain that AT&T is one of only four broadband Internet access service providers that can extract such interconnection fees.609 215. The Applicants dispute the claim that AT&T intentionally degraded Netflix’s service by allowing congestion to build up at the interconnection points with AT&T’s last mile network.610 As (Continued from previous page) Cogent claims that AT&T suggested that they reassess the settlement-free nature of their relationship based on “the purported significance of traffic ratios.” Cogent Comments at 13. 605 See Netflix Comments at 23-25; see also Cogent Comments at 13-14. According to Netflix, its service using DSL and U-verse declined to 1.0 Mbps and 1.5 Mbps, respectively, at their lowest points, far below Netflix’s recommended 3 Mbps for DVD quality video. See Netflix Comments at 23. 606 See Netflix Comments at 22-23; see also Cogent Comments at 13. Netflix states that Open Connect allows Netflix content to be stored at interconnection exchange points or at any location a terminating access network requests. See Netflix Comments at 19-20 (“By placing popular Netflix content closer to those [broadband Internet access service] provider’s subscribers who are seeking access to it (either through embedded cache servers or by interconnecting at public Internet exchange points) Netflix can help terminating access networks avoid creating unnecessary traffic ‘up the chain’ – either over the middle-mile or at the [broadband Internet access service] provider’s interconnection points.”). 607 See Netflix Comments at 24; see also Cogent Comments at 14. Netflix states that, after the agreement was implemented, AT&T subscribers were able to access Netflix content at bit rates approximately 63 percent (for DSL) or 85 percent (for U-verse) higher than at its lowest ebb. See Netflix Comments at 25. The Applicants contend that the direct interconnection agreement between AT&T and Netflix is very favorable to Netflix and is working well for its customers. See Joint Opposition at 44; Joint Opposition, Declaration of Scott Mair, Senior Vice President of Technology Planning and Engineering, AT&T Services, transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, ¶¶ 7, 25-30 (filed Oct. 16, 2014) (“Mair Decl.”) (“AT&T is rapidly closing in on the fastest average download speeds among all large U.S. ISPs for Netflix traffic. This contract thus demonstrates both that AT&T is interested in ensuring that Netflix receives the level of Internet access necessary to provide a good experience for AT&T customers that use Netflix’s service, and that Netflix has a means of recourse if AT&T failed to do so.”). 608 See Netflix Reply at 9; Cogent Reply at 14-16; see also Cogent Reply at 16 (“The AT&T-Netflix agreement highlights the manner in which a combined AT&T/DIRECTV can be expected to use a technical problem of its own making – congestion at interconnection points – as a lever to force edge and/or transit providers to pay for access to the merged firm’s broadband subscribers.”). 609 See Netflix Comments at 14, 24 (“AT&T’s substantial broadband footprint (approximately 10 million subscribers) and its status as a Tier 1 network operator give it the ability to demand terminating access fees from edge providers such as Netflix.”); see also Cogent Comments at 12, 14. On June 23, 2015, New America’s Open Technology Institute (“OTI”) expressed its concern to FCC staff that millions of Americans experienced persistent degradation of their Internet connections over the past month, according to the results of the Internet Health Test conducted by M-Lab. OTI noted that congestion was particularly acute at interconnection points with transit providers GTT Communications, Inc. (“GTT”) and Tata Communications. Letter from Joshua Stager, Policy Counsel, New America’s Open Technology Institute, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 2 (filed June 24, 2015). AT&T responded that it recently reached long-term interconnection agreements with Level 3, Cogent, and GTT and that to “address consumers’ needs today and in the future, AT&T remains open to negotiating with any similarly situated provider and reaching similar commercial agreements that properly align the incentives of all parties for the benefit of end users.” Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 2 (filed June 25, 2015). 610 See Joint Opposition at 39-49; Mair Decl. ¶ 3. The Applicants state that the Commission has decided previously not to regulate backbone interconnection and peering and that any change in that approach should be handled in an (continued….) Federal Communications Commission FCC 15-94 83 evidence, the Applicants assert that AT&T has peering arrangements with 23 partners as well as direct connections with CDNs and other entities and that Netflix and any other OVD can choose any combination of these services to send traffic to AT&T.611 The Applicants argue that the only way a “degradation by congestion” strategy could work would be for AT&T to congest or block all of its interconnection points, which would have the effect of degrading all Internet traffic headed to AT&T customers and would thereby harm the performance of AT&T’s broadband service.612 Even if AT&T blocked only Cogent, AT&T states that Netflix could switch to another backbone provider that delivers traffic to AT&T.613 Thus, the Applicants argue, Netflix’s prior experience with congestion could have been avoided if it used a different mix of backbone providers.614 Netflix disputes the Applicants’ claims, arguing that there are only six competitive options for transit to high-bandwidth customers in the United States and that Netflix used all options reasonably available.615 216. Commenters contend that the combined entity’s commitment to abide by the 2010 Open Internet Order protections for three years would not address these concerns because such protections do not extend to “existing arrangements for network interconnection, including existing paid peering arrangements.”616 Accordingly, Netflix617 and Cogent618 suggest conditions to address their concerns, (Continued from previous page) industry-wide proceeding. See Joint Opposition at 40-41 (citing 2010 Open Internet Order, 25 FCC Rcd at 17944, ¶ 67 n.209). 611 See Joint Opposition at 40; Katz Reply Decl. ¶ 63; Mair Decl. ¶¶ 4, 10, 15, 19-20, 23. 612 See Joint Opposition at 42; Katz Reply Decl. ¶ 64 (“[B]ecause transit providers and CDNs generally deliver the content of many different edge providers, including those that do not compete with AT&T and DIRECTV’s video services (and may even be complementary to them), the degradation strategy would harm the quality of AT&T’s broadband services broadly.”); Mair Decl. ¶¶ 5, 24. 613 See Joint Opposition at 42; Katz Reply Decl. ¶ 64; Mair Decl. ¶¶ 4, 19-20. 614 See Joint Opposition at 42-43; Mair Decl. ¶¶ 4, 19-21 (“Just as Netflix has the flexibility to shift its traffic to different providers to avoid congestion, if it concentrates its traffic on too few links, it can create congestion. That is precisely what Netflix has done. Over the past several years, Netflix has concentrated the traffic it sends to AT&T’s network among a relatively small number of AT&T’s peers, and thus among a small number of ingress points into AT&T’s network. These interconnection points were not designed to handle the very large traffic volumes Netflix was attempting to send through them and, predictably, congestion began to occur.”) (emphasis in original). 615 See Netflix Reply at 6-7. 616 See Cogent Comments at 16-17 (citing 2010 Open Internet Order, 25 FCC Rcd at 17944, ¶ 67 n.209 (“We do not intend our rules to affect existing arrangements for network interconnection, including existing paid peering arrangements.”)); Franken Comments at 4; WGAW Petition at 24, 27-28; Cogent Reply at 19-20. Indeed, commenters contend that Comcast was involved in a similar dispute with Netflix despite Comcast’s commitment to abide by the 2010 Open Internet Order. See Cogent Comments at 16 n.51 (“Comcast, itself bound by the Open Internet Order as a result of conditions associated with its acquisition of NBC Universal, engaged in precisely the same sort of interconnection strategy as AT&T.”). 617 See Netflix Comments at 34 (“the combined entity should be prohibited from charging a content provider a terminating access fee to interconnect”). But see Joint Opposition at 49 (stating that this condition would be “contrary to industry norms and would be extremely costly to AT&T and its customers”); Mair Decl. ¶ 50. 618 See Cogent Comments at 20-22; Cogent Reply at 20-25 (stating that: (i) AT&T-DIRECTV should be subject to enhanced transparency requirements to aid in detection of discriminatory behavior; (ii) if any interconnection point between AT&T-DIRECTV and another network reaches 70 percent capacity, then AT&T-DIRECTV should upgrade capacity to avoid congestion; (iii) for a period of seven years following consummation of the transaction, AT&T-DIRECTV should be required to maintain settlement-free peering relationships with any network with whom AT&T had such a relationship as of May 18, 2014, the date AT&T and DIRECTV announced the proposed transaction; and (iv) AT&T-DIRECTV should not be permitted to engage in unreasonable network management practices with respect to interconnection); see also WGAW Reply at 36. But see Joint Opposition at 48-49 (stating (continued….) Federal Communications Commission FCC 15-94 84 including that the Applicants be required to interconnect on a “bill-and-keep” basis with other network operators or edge providers for the exchange of Internet traffic between the Applicants’ broadband customers and the other network or the edge provider, provided that the party requesting interconnection agrees to localize the exchange of traffic, and that if any interconnection port reaches 70 percent capacity average utilization for more than three hours in any 24-hour period on more than five separate occasions within any 30-day period, then the Applicants must upgrade the ports and augment capacity to avoid congestion, with the interconnecting parties each bearing the costs associated with upgrades to their own networks.619 The Applicants counter that, today, private commercial negotiations balance the interests of broadband access and edge providers and that commenters seek to shift a disproportionate share, if not the entire share, of the burden of maintaining broadband access upon the Applicants and that it would be unprecedented and unjustified to require that the Applicants provide free backbone service to other backbone carriers and edge services.620 In addition, the Applicants assert that while a 70 percent utilization was often the trigger point for augmentation, provisioning processes have now become more streamlined, and software has become more sophisticated, so that it is reasonable to implement augmentation at 85 percent utilization.621 217. Discussion. As stated in the 2015 Open Internet Order, “consumers bear the harm when they experience degraded access to the applications and services of their choosing due to a dispute between a large broadband provider and an interconnecting party.”622 Also, because OVD subscribers expect high-quality video, OVDs are vulnerable to degradation at the interconnection point with a broadband Internet access service provider’s last mile network.623 Thus, as stated in the 2015 Open Internet Order, we find that “broadband Internet access providers have the ability to use terms of interconnection to disadvantage edge providers and that consumers’ ability to respond to unjust or unreasonable broadband provider practices are limited by switching costs.”624 We appreciate commenters’ concerns in this area. 218. We believe that in this particular case the protections in the 2015 Open Internet Order, coupled with certain conditions we impose today, will best address any potential for anticompetitive activity by the combined entity in its interconnection practices that affects OVDs. In the 2015 Open Internet Order, the Commission decided to take a case-by-case approach to considering whether an ISP’s interconnection practices constitute unjust, unreasonable, or unjustly discriminatory practices under Sections 201 and/or 202 of the Act.625 Parties such as interconnecting OVDs that believe that an ISP’s interconnection practices are not consistent with Sections 201 or 202 are able to file complaints under Section 208. Thus, if any interconnection concerns with the combined entity rise to that level, OVDs (Continued from previous page) that the settlement-free peering condition would require AT&T to subsidize Cogent, and other members of a closed group of current peers, for years on end, regardless of potentially drastic changes in traffic balances, the terms of those parties’ freely negotiated agreements, and the existence of numerous alternative backbone providers, and that the free upgrade condition would be contrary to industry norms, encourage inefficient routing, and impose enormous uncompensated costs on AT&T); Mair Decl. ¶¶ 44-49, 51-52. 619 Cogent et al. May 12, 2015, Ex Parte Letter at 5. 620 Conditions Ex Parte Presentation at 3-4. 621 Id. at 4. 622 2015 Open Internet Order, 30 FCC Rcd at 5689-90, ¶ 199. 623 See Netflix Comments at 17-18 (citing a study that found that viewers of streaming video begin to abandon a video if it takes more than two seconds to start, with each incremental delay resulting in a 5.8 percent increase in the abandonment rate); Cogent Reply at 16. 624 2015 Open Internet Order, 30 FCC Rcd at 5694-95, ¶ 205. 625 Id. at 5693, ¶ 203. Federal Communications Commission FCC 15-94 85 have this vehicle available to have their specific concerns adjudicated by the Commission.626 We also note that AT&T entered into long-term interconnection agreements with Cogent and Level 3 to provide added capacity and new interconnection points for their IP networks.627 In addition, on June 18, 2015, GTT Communications, Inc. (“GTT”) and AT&T entered into a long-term interconnection agreement to provide added capacity and new interconnection points for their IP networks.628 219. We have not identified in the record any evidence that would support blanket restrictions on all interconnection arrangements between the Applicants and OVDs.629 However, given our heightened concern where ISPs compete with third-party Internet-based services (i.e., OVDs), we impose additional conditions that require the combined entity to file all interconnection agreements with the Commission and to provide the Commission with certain interconnection performance metrics, which we will use in combination to monitor the terms and effects of such interconnection arrangements. To the extent we determine that the interconnection performance metrics are not competitively sensitive or that they should be released in the public interest, we will make them publicly available. D. Harm to Supply, Quality, and Diversity in Video Programming 220. In this section, we consider whether the transaction would increase the Applicants’ incentive or ability to engage in behavior that is likely to reduce the supply, quality, or diversity of video programming. After considering structural factors affecting the likelihood of competitive harms to programmers, we examine how the combined entity’s increased size may confer greater leverage when negotiating with programmers and what effects that might have on consumer welfare. We then address concerns raised by two specific types of programmers: public, educational, and governmental (“PEG”) channels and local broadcast stations. We do not find that any increased bargaining power of the combined entity is likely to harm consumer welfare, nor do we find any transaction-specific harms that necessitate the adoption of conditions related to program carriage, PEG channels, or the carriage of local broadcast stations. 1. Background on Video Programming 221. Many of the potential public interest harms raised in the record relate to the video programming industry. We provide this background that informs our analysis of those harms. 222. Content providers, including cable networks and over-the-air broadcasters, may both produce their own programming and acquire programming produced by others.630 They package and sell 626 Id. at 5638-39, ¶¶ 93-94. 627 See Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (filed June 10, 2015); see also Cogent Communications, Inc., Cogent and AT&T Enter Into Interconnection Agreement (press release), June 10, 2015, available at http://www.cogentco.com/en/news/press-releases/741-cogent- and-at-t-enter%20-into-interconnection-agreement (visited June 18, 2015) (Cogent’s CEO stated “[b]oth Cogent and AT&T’s customers will benefit from this agreement for years to come. We are putting customer needs at the forefront by enabling an expanded, secure and resilient interconnection environment.”); Level 3 Communications, Level 3 and AT&T Enter Into Interconnection Agreement (press release), May 11, 2015, available at http://level3.mediaroom.com/index.php?s=23600&item=137034 (visited June 18, 2015) (Level 3’s Chief Marketing Officer stated “[t]his agreement will benefit Level 3’s and AT&T’s customers for years to come. With customer needs at the forefront, you enable a growing, secure and resilient interconnection environment.”). 628 See Letter from Maureen R Jeffreys, Counsel to AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 1 & Exhibit 75.2.28 (filed June 24, 2015). 629 See 2010 Open Internet Order, 25 FCC Rcd at 17944, ¶ 67 n.209 (“We do not intend our rules to affect existing arrangements for network interconnection, including existing paid peering arrangements.”). 630 Liberty Media-DIRECTV Order, 23 FCC Rcd at 3281, ¶ 33; Adelphia Order, 21 FCC Rcd at 8236, ¶ 65; News Corp.-Hughes Order, 19 FCC Rcd at 502, ¶ 54; EchoStar-DIRECTV HDO, 17 FCC Rcd at 20653, ¶ 248. Federal Communications Commission FCC 15-94 86 this programming as a network or networks to MVPDs for distribution to consumers.631 To provide multichannel video services to subscribers, MVPDs combine broadcast television signals and (non-over- the-air) programming networks for distribution on their cable, satellite, fiber, or wireless systems.632 MVPDs compensate owners of programming networks through license fees that are based in part on the number of subscribers that receive the programming from the MVPD.633 Similarly, broadcast station owners may receive retransmission consent fees from MVPDs. Discounts of such fees often are negotiated based on the number of MVPD subscribers and on other factors, such as placement of the network on a particular programming tier.634 Most programming networks, broadcasters, and MVPDs also derive revenue by selling advertising time during the programming.635 223. Video programming comes in a wide variety of characteristics, focus, and subject matter.636 Programming is offered by over-the-air broadcast stations; RSNs; national program networks, including news, entertainment and hobby networks; and various non-sports regional networks.637 Moreover, the Commission has found that at least a proportion of MVPD subscribers views certain types of programming as so vital or desirable that they are willing to switch to a different MVPD in order to gain or retain access to that programming.638 These findings inform our current analysis. 224. When considering potential harms involving video programming in prior transactions, the Commission considered the geographic area in which the program owner licenses the programming, and we do so here.639 For national programming networks, such areas are at least national in scope.640 Such networks generally are licensed to MVPDs nationwide, and, in some cases, they are licensed internationally.641 In contrast, with respect to RSNs and other regional networks, the Commission considered the effects in regional markets.642 Contracts between sports teams and RSNs generally are limited to the “distribution footprint” established by the owner of the programming.643 631 Id. 632 Id. 633 Liberty Media-DIRECTV Order, 23 FCC Rcd at 3281-82, ¶ 34; Adelphia Order, 21 FCC Rcd at 8236, ¶ 65; News Corp.-Hughes Order, 19 FCC Rcd at 502, ¶ 55; EchoStar-DIRECTV HDO, 17 FCC Rcd at 20654, ¶ 249. 634 Id. 635 Id. 636 Liberty Media-DIRECTV Order, 23 FCC Rcd at 3282, ¶ 35; Adelphia Order, 21 FCC Rcd at 8236, ¶ 66; News Corp.-Hughes Order, 19 FCC Rcd at 504, ¶ 59. 637 Id. 638 Comcast-NBCU Order, 26 FCC Rcd at 4285, ¶ 117; Liberty Media-DIRECTV Order, 23 FCC Rcd at 3282, ¶ 35; Adelphia Order, 21 FCC Rcd at 8236-37, ¶ 66; News Corp.-Hughes Order, 19 FCC Rcd at 504, ¶ 59; see also Adelphia Order, 21 FCC Rcd at 8270-71, ¶ 146. 639 Liberty Media-DIRECTV Order, 23 FCC Rcd at 3282, ¶ 37; Adelphia Order, 21 FCC Rcd at 8237, ¶ 68; News Corp.-Hughes Order, 19 FCC Rcd at 506, ¶ 64. 640 Liberty Media-DIRECTV Order, 23 FCC Rcd at 3282, ¶ 37; Adelphia Order, 21 FCC Rcd at 8237, ¶ 68; News Corp.-Hughes Order, 19 FCC Rcd at 506, ¶ 66. 641 Id. 642 Id. 643 Id. In the case of broadcast television programming, we have found DMAs to define the relevant geographic market for each individual broadcast station. See Liberty Media-DIRECTV Order, 23 FCC Rcd at 3283, ¶ 37 n.116; News Corp.-Hughes Order, 19 FCC Rcd at 506, ¶ 65. Federal Communications Commission FCC 15-94 87 2. Potential Competitive Harms a. Increased Leverage of Combined Entity in Programming Negotiations 225. As discussed below, several commenters contend that the combined entity’s subscriber base would afford the Applicants increased leverage in negotiations with programmers, leading to several harms to programmers, including reduced carriage opportunities, below-market licensing fees, anticompetitive contractual conditions, and less investment in programming. The Applicants disagree, contending that the transaction would not alter their bargaining power with programmers. 226. Positions of the Parties. Several commenters claim that the Applicants would gain substantial leverage in their negotiations with programmers due to the size of their combined share of the video distribution market.644 ACA observes that AT&T’s expansion from approximately 6 million to 26 million video subscribers would give it a national reach approaching the Commission’s previous 30 percent cable horizontal ownership limit.645 Other commenters contend that a national programmer would not survive without the combined entity’s distribution regardless of whether its subscribership remains below 30 percent of pay-TV households nationwide.646 644 See, e.g., DISH Petition at 12 (arguing that the combined entity could use its enhanced position as a “‘must-have’ distribution outlet” to extract unfavorable prices, terms, and conditions from programmers); Franken Comments at 5-6 (expressing concern that the increased market power of the combined entity would enable it to negotiate below- market programming rates); NAB Comments at 3-7 (arguing that the proposed transaction would strengthen the Applicants’ bargaining power in retransmission consent negotiations with broadcasters and urging the Commission to redress the imbalance, in part, by relaxing its media ownership rules); WGAW Petition at 7, 9-14 (asserting that the combined entity would have greater leverage to reduce license and retransmission consent fees below competitive levels); WGAW Reply at 4-10 (arguing that consolidation in the video distribution market would give the Applicants greater leverage over programmers and local broadcasters). 645 ACA Comments at 17-20; see also ACA Reply at 10 (noting that the Applicants’ economist acknowledges that the combined entity would be able to exert greater leverage over programmers than either Applicant can on its own). Consistent with the 1992 Cable Act, in 1993, the Commission set the cable horizontal ownership limit at 30 percent of all cable homes passed nationwide, which it subsequently modified in 1999 to limit a cable operator from serving more than 30 percent of all MVPD subscribers. See Implementation of Sections 11 and 13 of the Cable Television Consumer Protection and Competition Act of 1992, Horizontal and Vertical Ownership Limits, MM Docket No. 92- 264, Second Report and Order, 8 FCC Rcd 8565 (1993); Implementation of Section 11(c) of the Cable Television Consumer Protection and Competition Act of 1992, Horizontal Ownership Limits, MM Docket No. 92-264, Third Report and Order, 14 FCC Rcd 19098 (1999); 47 U.S.C. § 533(f). In 2001, the D.C. Circuit remanded the limit for inadequate support. Time Warner Entm’t Co. v. FCC, 240 F.3d 1126 (D.C. Cir. 2001). The Commission reestablished the horizontal ownership limit at 30 percent in 2008. The Commission’s Cable Horizontal and Vertical Ownership Limits, MM Docket No. 92-264, Fourth Report and Order and Further Notice of Proposed Rulemaking, 23 FCC Rcd 2134 (2008) (“Cable Horizontal and Vertical Ownership Limits Order”). However, the D.C. Circuit vacated that decision the following year, and the Commission has not reinstated the horizontal limit. See Comcast Corp. v. FCC, 579 F.3d 1 (D.C. Cir. 2009) (vacating the limit as arbitrary and capricious). 646 DISH Petition at 12 (asserting that the transaction would make the combined entity a “‘must-have’ distribution outlet” for any programmer seeking national coverage); Letter from F. William LeBeau, Holland & Knight LLP, Counsel for ReelzChannel, LLC, to Marlene Dortch, Secretary, FCC, MB Docket No. 14-90, at 2 (July 30, 2014) (“ReelzChannel July 30, 2014, Ex Parte Letter”) (claiming that a national programming network cannot be viable today unless it is carried on each of the top-four MVPDs, including DIRECTV); Reply Comments of Hubbard Broadcasting, Inc., MB Docket 14-90, at 4 (filed Jan. 7, 2015) (“Hubbard Reply”) (stating that programming carriage under reasonable terms and conditions on the nation’s largest four or five MVPDs already is essential for success); WGAW Reply at 7-8 (arguing that the survival of national broadcast and programming networks would depend on distribution by the combined entity, which could become the nation’s largest MVPD). Federal Communications Commission FCC 15-94 88 227. Some commenters claim that the increased consolidation in the video distribution market resulting from the transaction would make it more difficult for programmers to obtain carriage.647 In addition, commenters contend that the combined entity’s bargaining power would lead to lower license fees and retransmission consent fees than what the Applicants separately pay today to programmers and broadcasters, which in turn would result in less investment in programming in the future.648 WGAW disputes that the Applicants’ projected 20 percent decrease in programming costs would stem from volume discounts that programmers would willingly offer the combined entity due to its larger subscriber base.649 WGAW argues that the Applicants’ projected programming cost savings instead would derive from the monopsony power the combined entity would hold post-transaction.650 WGAW provides annual data from 2008 to 2013 purporting to show that license fees have not increased at a higher rate than programmers’ investment in content.651 WGAW concludes that programmers do not have an expanding profit margin that can accommodate a volume discount and that any reduction in license fees would result in less revenue to invest in content production, to the ultimate detriment of the consumer.652 In an effort to reduce the risk that the combined entity would extract below-market rates, WGAW asks the Commission to impose a condition on the transaction that would require binding arbitration when the combined entity and a programmer fail to reach a carriage agreement.653 228. Several commenters contend that the combined entity’s increased leverage in programming negotiations would strengthen the Applicants’ ability to demand anticompetitive carriage terms. In particular, commenters express concern with MFN contract clauses,654 which give the MVPD 647 See, e.g., Franken Comments at 5-6 (warning that the removal of a video distributor from the market would reduce opportunities for independent programmers to obtain MVPD carriage); Letter from Burt A. Braverman, Davis Wright Tremaine LLP, Counsel for INSP, LLC, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14- 90, at 1 (Nov. 20, 2014) (“INSP Nov. 20, 2014, Ex Parte Letter”) (claiming that consolidation exacerbates the disadvantages independent programmers already face in vying for limited channel capacity in competition with large content producers); Comments of RFD-TV, MB Docket 14-90, at 1-2, 7-9, 10-12 (filed Sept. 16, 2014) (“RFD-TV Comments”) (arguing that consolidation in the MVPD industry reduces carriage options for independent programmers while acknowledging that AT&T and DIRECTV have been supportive of independent rural programming). 648 See Franken Comments at 5-6 (asserting that the combined entity could exert its enhanced market power to force unaffiliated content providers to accept artificially low programming rates, which would harm the development of new and independent content); WGAW Petition at 7, 9-14 (warning that the increased incentive and ability of the combined entity to reduce license and retransmission consent fees below competitive levels could diminish an essential revenue stream that supports the production of original programming); WGAW Reply at 8-10 (concerned that local broadcasters, which continue to offer the most popular content, would be forced to accept lower retransmission consent fees); see also WGAW Reply at 6-8; DISH Petition at 12. 649 WGAW Petition at 9-10. WGAW claims that programming cost savings generally are not linked to an MVPD’s volume of subscribers because content production costs are the same regardless of the number of MVPD subscribers and because transmission costs are paid by the MVPDs. See id. 650 Id.; see also WGAW Reply at 6-7 (positing that the Applicants’ intended goal is “to cut AT&T’s costs below competitively negotiated rates”); DISH Petition at 12 (suggesting that the combined entity would gain monopsony power over programmers). 651 WGAW Petition at 11, 13; see also WGAW Reply at 6-7 (claiming that “the growth in content spending has outpaced growth in [license] fees”). 652 WGAW Petition at 12-14; see also WGAW Reply at 6 (noting that the Applicants “offer no evidence to suggest that programming fees overvalue content”). 653 WGAW Reply at 33-34. 654 See, e.g., Franken Comments at 6; DISH Petition at 12; ReelzChannel July 30, 2014, Ex Parte Letter at 2; INSP Nov. 20, 2014, Ex Parte Letter at 1; ACA Reply at 10-11; Letter from Mark J. Coleman, Senior Partner and General Counsel, InterMedia Partners, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 2 (Sept. 25, 2014) (“InterMedia Partners Sept. 25, 2014, Ex Parte Letter”). Federal Communications Commission FCC 15-94 89 the right to modify the programming agreement to incorporate more favorable rates, terms, or conditions that the programmer subsequently may negotiate with another MVPD.655 Typically, an MFN clause specifies the type of term that may be substituted, such as a more favorable pricing term. ReelzChannel claims, however, that independent program networks increasingly are forced to comply with more demanding MFN provisions, including unconditional clauses that allow the MVPD to import any term from any other contract that the programmer has with another MVPD, without being bound by the conditions agreed to by that MVPD.656 ReelzChannel argues that this type of “cherry picking” MFN provision hampers competition, innovation, content diversity, and viewpoint diversity by allowing carriage agreements to be composed of all the worst terms the programmer has been forced to accept from any MVPD.657 ReelzChannel suggests that, as a condition to transaction approval, the Commission prohibit the Applicants from subjecting independent program networks to unreasonable MFN provisions.658 Similarly, INSP argues that the Commission’s current rules do not protect independent programmers sufficiently from an MVPD’s enormous leverage and asks the Commission to consider measures to ensure the fair and reasonable treatment of independent program networks.659 229. In response to commenters’ arguments, the Applicants state that the combined entity’s subscribership share would be “considerably less” than the 30 percent permitted under the Commission’s previous cable horizontal ownership limit.660 Further, the Applicants assert that the D.C. Circuit, when striking down that limit, suggested that an MVPD’s subscribership share even above that level is not in itself indicative of excessive market power.661 The Applicants also point to antitrust precedent finding that a market share below 30 percent is presumed not to confer market power.662 In addition, DIRECTV maintains that there is no one particular MVPD or OVD on which carriage is required to ensure the viability of a new programming network.663 230. The Applicants dispute the notion that the transaction would alter relative bargaining power to an extent that would affect the quantity and variety of programming.664 They note that 655 MFN rights can be conditional or unconditional. A conditional MFN provision entitles a distributor to certain contractual rights that the programmer has granted to another distributor, subject to the acceptance of related terms and conditions contained in that other distributor’s agreement. An unconditional MFN provision, by contrast, contains no such requirement that the distributor entitled to MFN rights accept related terms and conditions. 656 ReelzChannel July 30, 2014, Ex Parte Letter at 2. 657 Id. 658 Id.; see also ACA Reply at 57-58 (arguing that the Commission should prohibit the combined entity from using its increased leverage in ways that interfere with a programmer’s ability to negotiate more favorable terms with other MVPDs). ReelzChannel subsequently submitted a letter supporting the proposed transaction and arguing that it would cause no material harm to the marketplace. See Letter from Charles R. Naftalin, Holland & Knight LLP, Attorney for ReelzChannel, LLC, to Marlene Dortch, Secretary, FCC, MB Docket No. 14-90, at 1 (Feb. 13, 2015) (“ReelzChannel Feb. 13, 2015, Ex Parte Letter”). 659 INSP Nov. 20, 2014, Ex Parte Letter at 1; see also Letter from Mark DeVitre, Executive Vice President and General Counsel, Entertainment Studios, Inc., to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 1 (March 30, 2015) (detailing ex parte meeting with the Office of the Chairman regarding a complaint filed by the National Association of African American Owned Media against AT&T and DIRECTV for alleged racial discrimination in contracting under Section 1981 of the Civil Rights Act with regard to the companies’ “unilateral refusal to contract with [Entertainment Studios Networks] for cable channel carriage, license fees, and advertising”). 660 Joint Opposition at 51. 661 Id. at 51-52 n.185 (citing Comcast Corp. v. FCC, 579 F.3d at 8). 662 Id. (citing Commercial Data Servers, Inc. v. Int’l Bus. Machs. Corp., 262 F. Supp. 2d 50, 74-75 (S.D.N.Y. 2003)). 663 DIRECTV Response to Sept. 9, 2014, Information Request at 47. 664 Joint Opposition at 51-53. Federal Communications Commission FCC 15-94 90 programmers control a critical input to MVPD services and contend that ownership of the most popular content is highly concentrated.665 They argue that programmers therefore enjoy far more bargaining power than video distributors, as evidenced by rapidly rising programming rates.666 231. The Applicants also disagree with commenters that the transaction would result in less investment in programming. According to the Applicants’ economist, Dr. Katz, AT&T’s anticipated programming cost savings of 20 percent, while significant from AT&T’s perspective, would not have a material effect on programming investment because the corresponding reduction in content providers’ revenues, if achieved, would represent a very small share, approximately [BEGIN CONF. INFO.] [END CONF. INFO.] percent, of programmers’ overall programming revenues.667 The Applicants assert that content providers’ revenues instead would increase due to the greater scale and efficiencies stemming from the proposed transaction that would enable both content providers and the combined entity “to offer more competitive, cost-effective, and integrated services to consumers.”668 The Applicants conclude that, as a result, the proposed transaction would spur investment in programming.669 232. In addition, the Applicants deny that their projected programming cost savings would result from an exercise of monopsony power, and they argue that a monopsony model does not apply to the bargaining context in which content providers sell programming to MVPDs.670 Free State argues further that a monopsony threat is unlikely given the number of other video programming purchasers, such as cable MVPDs, DISH, telephone MVPDs (Verizon, CenturyLink, Frontier), broadband service providers such as Google Fiber, and an increasing number of Internet-based providers.671 Similarly, the Applicants point to increasing competition from OVDs, particularly Netflix, as a constraint on the bargaining power of MVPDs in programming negotiations.672 233. Several programmers agree with the Applicants that the proposed transaction would not lessen supply or diversity in the video programming market. RFD-TV praises both AT&T and DIRECTV for their carriage of its network and support of rural programming.673 Likewise, Herring Networks, Inc., an independent owner of two national programming networks, commends AT&T for its history of fair dealing with independent programmers and claims that expanding AT&T’s reach would 665 Id. at 50. 666 Id. 667 Katz Reply Decl. ¶ 32. 668 Joint Opposition at 52-53. 669 Id. 670 Id. at 51-52 (pointing to the Comcast-AT&T transaction as an example where the Commission found that the merger of the nation’s then-largest and third largest cable operators would not gain monopsony power over the programming market); see also Katz Decl. ¶ 111 n.195 (arguing that carriage negotiations should be analyzed under the bargaining theory rather than under the standard theory of monopsony because an MVPD has an incentive to increase the amount of programming purchased when it succeeds in negotiating a lower price, as opposed to the output-reducing effects of a monopsony); Katz Reply Decl. ¶ 31 (asserting that the standard monopsony model does not apply to carriage negotiations). 671 Free State Comments at 23-24. 672 Joint Opposition at 51-52. 673 RFD-TV Comments at 10-12. In its Reply, however, RFD-TV reports that DIRECTV recently declined to carry RFD-TV HD or RURAL TV/FamilyNet. In addition, RFD-TV is concerned that DIRECTV may decide to move RFD-TV SD from its basic channel tier to a more expensive tier with a smaller subscribership. RFD-TV asks the Commission to be vigilant in protecting the interests of independent rural programmers. Reply Comments of RFD- TV, MB Docket 14-90, at 2-4 (filed Jan. 7, 2015) (“RFD-TV Reply”); see also InterMedia Partners Sept. 25, 2014, Ex Parte Letter at 2 (noting the particular vulnerabilities of independent programmers, including being subject to re- tiering without good cause). Federal Communications Commission FCC 15-94 91 promote competition and benefit consumers and independent programmers.674 Similarly, NUVOtv lauds AT&T for distributing NUVOtv’s independent programming network serving the Latino community and for dealing with the network fairly.675 In addition, BabyFirst, an independently owned network focused on child development programming, supports the transaction and applauds both AT&T and DIRECTV for their “strong commitment to ensuring consumer access to diverse programming.”676 Despite ReelzChannel’s concerns about the leverage of large MVPDs to impose unreasonable MFN provisions,677 Hubbard Broadcasting, Inc. (“Hubbard”), the owner and operator of ReelzChannel, supports the proposed transaction, claiming that it would promote competition, innovation, and rural broadband access.678 Although it encourages the Commission to take appropriate steps to redress the market imbalance between small independent programming networks and large MVPDs, at the same time Hubbard argues that the proposed transaction has little import for programmers given that a programming network’s success already requires carriage on the largest four or five MVPDs.679 234. Discussion. Given the Commission’s interest in promoting supply and quality of programming, we consider carefully the potential harms that could arise from the loss of a potential video programming distributor, including the harms raised on the record relating to the increase in leverage that the combined entity may gain in contract negotiations. We are particularly attentive to whether the transaction would decrease consumer welfare by reducing the output or quality of programming. However, we find that the record does not establish that the proposed transaction is likely to result in such a consumer welfare decrease. We decline to impose a condition specifically to address the potential for a reduction in programming rates or the use of MFN provisions, which some commenters allege would flow from the combined entity’s increased leverage. 235. To support the argument that the combined entity would coerce programmers and broadcasters into accepting lower license fees and lower retransmission consent fees, programmers offer only generalized assertions of harm to their business operations without sufficiently demonstrating how such rate reduction would harm consumer welfare through the decrease in output or quality of programming.680 Commenters have not provided adequate empirical evidence to show that the reduction 674 Comments of Herring Networks, Inc., MB Docket 14-90, at 1-3 (filed Sept. 16, 2014) (“Herring Comments”); see also id. 2, 3-5 (stating that AT&T is an “important counterweight” to vertically integrated MVPDs, such as Comcast and Time Warner Cable, and that the transaction could cause vertically integrated cable providers “to rethink their discriminatory practices” toward independent programmers). 675 Letter from Michael Schwimmer, CEO, NUVOtv, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 1, 3-4 (Oct. 8, 2014). 676 Letter from Sharon Rechter, Co-Founder and EVP, BabyFirst, to FCC Chairman and Commissioners, MB Docket No. 14-90, at 1-2 (Jan. 7, 2015) (touting its “strong partnership” with AT&T, which it calls “forward thinking, visionary, and fair in [its] negotiations” and noting that DIRECTV was the first major distributor to carry BabyFirst’s programming). 677 But see ReelzChannel Feb. 13, 2015, Ex Parte Letter at 1 (supporting the proposed transaction). 678 Hubbard Reply at 2-3, 4-5 (lauding the Applicants, particularly DIRECTV, for their support of its program networks). 679 Id. at 3-4. 680 We note that Northwest Broadcasters, et al. (collectively, “The TV Station Group”) recently filed an Informal Objection, more than five months after the close of the pleading cycle, see supra n.27 (initial comments and petitions to deny due September 16, 2014; replies due January 7, 2015), asking the Commission to deny the Application or suspend its review until it has resolved a retransmission complaint that they recently filed against DIRECTV. Northwest Broadcasting, L.P., et al. Informal Objection and Request to Hold Applications in Abeyance, MB Docket No. 14-90 (filed June 12, 2015) (“The TV Station Group Informal Objection and Request”). These broadcasters have already availed themselves of the appropriate process for resolving disputes about retransmission consent by filing a complaint. Further, for the reasons discussed in this section, we find that the record as a whole regarding potential effects on retransmission consent negotiations does not present a substantial and material (continued….) Federal Communications Commission FCC 15-94 92 in programming rates that the combined entity might achieve would curtail investment in content production.681 Thus, we find that the record here does not allow us to conclude that a decrease in programming rates would have the net effect of lowering the quality or quantity of programming. 236. Similarly, commenters’ unsupported argument that the Applicants would gain monopsony power in the video programming market post-transaction fails to establish that competitive harm would occur as a result of this transaction.682 In addition, the Commission has determined previously that, for several reasons, the monopsony model is not useful in analyzing the impact of an MVPD’s market power on programming rates and that “the usual incentive for a firm to exercise monopsony power does not occur in this market.”683 237. With respect to MFN provisions, the record does not establish that the proposed transaction would result in programming contract provisions that harm competition. The MFN provisions in the record cover both price and non-price terms and involve programmers of all types,684 but we do not have a record that establishes the competitive impact of MFNs sufficient to support a general condition restricting the use of such contractual provisions. In addition, the record does not establish that the proposed transaction is likely to limit the ability of new programmers to enter the market, particularly given the relatively recent rollout of AT&T’s U-verse video service within the past decade and the fact that it has fewer than 6 million subscribers. To the extent that some commenters allege that carriage on DIRECTV already is essential today, the transaction does not change that conclusion.685 Thus, based on our record, we cannot find that the subscribership of the combined entity is likely to prevent the entry of a (Continued from previous page) question of fact as to whether the Application will serve the public interest. See Applications of Capital Cities/ABC, Inc. (Transferor) & the Walt Disney Co. (Transferee), et al., 11 FCC Rcd 5841, 5862, ¶ 31 (1996) (separate enforcement proceeding was the appropriate forum for resolution of allegations that applicant violated certain broadcasting rules). Thus, we deny The TV Station Group Informal Objection and Request. 681 Although WGAW offers some annual data regarding license fees and programming costs, we find that the data do not show that lower license fees will harm programming investment. WGAW Petition at 10-14. We note further that the Applicants estimate that content providers’ revenues would decrease by only [BEGIN CONF. INFO.] [END CONF. INFO.] percent, assuming that the Applicants’ anticipated programming cost savings are accurate and actually achieved. See Katz Reply Decl. at ¶ 32. We acknowledge that our review of the evidence included data that was not available to third-parties. However, we also find that none of the commenters in this proceeding was prejudiced by the unavailability of this data. The Applicants’ estimate of their reduced programming costs – the amount of revenues the content providers will lose – is an extremely small percentage of the programmers’ total revenues. Moreover, we find that the Applicants have overestimated their cost savings from reduced programming costs – and hence overestimated the amount of revenues the programmers would lose. For these reasons, as well as the fact that we independently reviewed the record in this respect, we do not find that commenters’ review of the programming contracts and negotiation materials would provide them with any support for a plausible argument that the difference in programming costs that could be achieved as a result of this transaction would be substantial enough to have a non-insignificant effect on programming companies’ investment. 682 See WGAW Petition at 10; DISH Petition at 12. 683 Cable Horizontal and Vertical Ownership Limits Order, 23 FCC Rcd at 2150, ¶¶ 32-33; see also Comcast-AT&T Order, 17 FCC Rcd at 23265, ¶ 53 n.126 (finding that the complexity and confidentiality of programming rate information available at that time made it impossible to determine “a unique point at which a firm’s subscriber reach allows the firm to exercise monopsony power over programmers”). 684 See DIRECTV Response to Sept. 9, 2014, Information Request, Schedule 44(d) (providing an extensive list of sample MFN provisions from its programming contracts). 685 See, e.g., ReelzChannel July 30, 2014, Ex Parte Letter at 2 (claiming that a national programming network cannot be viable today unless it is carried on each of the top-four MVPDs, including DIRECTV); Hubbard Reply at 3-4. Federal Communications Commission FCC 15-94 93 new programming network or, for that matter, to cause the exit of an existing programming network.686 We also note that the Commission currently is exploring whether to define MVPD in a technology-neutral way that could provide expanded opportunities for new programmers by promoting competition between incumbent video providers and Internet-based providers of multiple linear video programming streams.687 238. We do not find based on the record before us that the Commission should impose company-specific program carriage conditions. The program carriage rules prohibit an MVPD from exerting its leverage as a distributor to require a financial interest in, or exclusive rights to, any program service as a condition for carriage.688 The program carriage rules also proscribe an MVPD from engaging in conduct that unreasonably restrains the ability of unaffiliated video programming providers to compete by discriminating in the distribution of programming based on the programmer’s affiliation or non- affiliation with the MVPD.689 Moreover, we note that the instant transaction raises few, if any, vertical integration concerns that the combined entity would discriminate against unaffiliated programmers. Given that AT&T and DIRECTV hold only minor programming interests, they have few affiliated programming interests to protect. Thus, we conclude that the existing program carriage rules address the concerns raised in this record.690 b. PEG Channels 239. Background. Cable systems are subject to special carriage requirements for PEG channels under the Communications Act.691 Congress afforded PEG channels special status in order to promote localism and diversity.692 Congress has noted that “PEG channels serve a substantial and compelling government interest in diversity, a free market of [ideas], and an informed and well-educated citizenry.”693 Accordingly, a cable operator is required to allocate channel capacity to PEG channels in its local market if a local franchising authority requests carriage pursuant to a franchising agreement.694 686 The Applicants also note that their combined MVPD subscribership of approximately 26 million would be less than 30 percent of the country’s total MVPD subscribers, a measurement that has guided the Commission in past transactions in determining the potential impact on the flow of video programming from the programmer to the consumer. Joint Opposition at 51. See, e.g., Comcast-AT&T Order, 17 FCC Rcd at 23263-69, ¶¶ 48-65; Adelphia Order, 21 FCC Rcd at 8283-84, ¶ 179. 687 See MVPD Definition NPRM, 29 FCC Rcd at 15996-97, ¶¶ 1, 5 (noting that permitting certain Internet-based providers to qualify as MVPDs, and granting them rights and responsibilities attendant to that status, may encourage new investment and entry in online video programming distribution). 688 47 C.F.R. § 76.1301(a)-(b). 689 Id. § 76.1301(c). 690 In previous transactions, the Commission has, on occasion, adopted conditions, or accepted the applicants’ voluntary commitments, to supplement the Commission’s existing program carriage rules. Typically, such conditions were prompted by a concern that the vertical integration of programming and distribution assets would increase the new entity’s incentives or abilities to discriminate against unaffiliated programmers. See, e.g., Comcast-NBCU, 26 FCC Rcd at 4284-89, ¶¶ 116-124 (imposing additional program carriage obligations as a condition of approval of the transaction resulting in vertical integration of two companies that, combined, owned 26 broadcast television stations, two broadcast television networks, more than 20 non-broadcast programming networks, a variety of regional sports and news networks, and the nation’s largest cable operator). 691 See 47 U.S.C. § 531. 692 See, e.g., id. §§ 531, 543(b)(7); H.R. REP. NO. 102-628, at 183 (1992) (“Making over-the-air broadcast and PEG access channels available on a separate tier promotes the time-honored principle of localism.”). 693 H.R. REP. NO. 102-628, at 85. 694 47 U.S.C. § 531. AT&T does not consider itself a cable operator with PEG obligations, but AT&T nevertheless provides PEG programming to its video subscribers. See Joint Opposition at 60-62 & n.222 (referencing, inter alia, Comments of AT&T, MB Docket No. 09-13, at 19-21 (filed March 9, 2009)). Federal Communications Commission FCC 15-94 94 DBS providers are not subject to PEG requirements, but they are subject to a carriage obligation for “noncommercial, educational, state public affairs, and informational programming.”695 240. Positions of the Parties. Several commenters contend that the proposed transaction would harm PEG channel programming.696 Noting that DBS providers are not subject to PEG requirements, they contend that AT&T might replace its U-verse video service with DIRECTV’s video service and thereby escape its PEG obligations.697 Commenters argue that the proposed transaction could result in the elimination of PEG programming from all markets where AT&T currently offers its U-verse video service.698 Alliance for Community Media, the Alliance for Communications Democracy, and Common Cause (“ACM et al.”) further question AT&T’s commitment to PEG programming in light of its past conduct and ask that the Commission deny the Application.699 ACM et al. argue in the alternative that, before acting on the AT&T-DIRECTV Application, the Commission should act on their petition seeking a declaratory ruling regarding whether AT&T’s “Channel 99 PEG Product” violates the Cable Act and Commission rules.700 NATOA also states that AT&T should voluntarily commit to remove restrictions on PEG operators’ use of PEG funds.701 241. The Applicants represent that the proposed transaction would have no adverse effect on PEG programming.702 AT&T states that it has supported and will continue to support PEG programming.703 AT&T reiterates that the combined entity would operate both U-verse and satellite video services and would comply with the respective regulatory obligations of each service.704 The Applicants characterize as baseless the suggestion that AT&T would offload U-verse video services to DIRECTV so as to escape PEG obligations.705 ACM et al. respond that the Applicants have not provided a minimum time frame for how long the Applicants plan to offer both U-verse video and DIRECTV after 695 47 U.S.C. § 335; see also 47 § C.F.R. 25.701 (listing DBS providers’ requirements for political and noncommercial programming). Because of technological differences between satellite and cable, the public interest programming carried by satellite providers is national in focus as opposed to the local focus of PEG channels. See STEVEN WALDMAN AND THE WORKING GROUP ON INFORMATION NEEDS OF COMMUNITIES, FCC, THE INFORMATION NEEDS OF COMMUNITIES: THE CHANGING MEDIA LANDSCAPE IN A BROADBAND AGE 180 (2011), available at http://transition.fcc.gov/osp/inc-report/The_Information_Needs_of_Communities.pdf (visited June 18, 2015). This difference results in satellite public interest programming being an imperfect substitute for PEG programming. 696 See ACM et al. Petition at 19-25; NATOA Comments at 2-3; Letter from Tillman L. Lay, Counsel for the Alliance for Community Media, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, Attachment at 2-3 (Feb. 24, 2015) (“ACM Feb. 24, 2015, Ex Parte Letter”); see also Public Knowledge-ILSR Petition at 11 (“[T]o ensure that viewers have access to diverse voices, [the Commission] should ensure that AT&T and DIRECTV give adequate access to Political, Educational, and Government (PEG) programming.”). 697 ACM et al. Petition at 24-25; NATOA Comments at 2-3; ACM Feb. 24, 2015, Ex Parte Letter, Attachment at 2. 698 See ACM et al. Petition at 24; ACM Feb. 24, 2015, Ex Parte Letter, Attachment at 2; see also NATOA Comments at 2-3 (“Because DIRECTV is under no obligation to carry or financially support PEG programming, any potential efforts by AT&T to eliminate its U-verse video product and replace it with satellite video service would necessarily undercut these local services and is problematic for local governments.”). 699 ACM et al. Petition at 19-26. ACM et al. state that AT&T’s “Channel 99 PEG Product” on U-verse makes accessing PEG channels more difficult and burdensome than accessing other linear programming. Id. at 19-20. 700 Id. at 20. 701 NATOA Comments at 3-4. 702 Joint Opposition at 60-62. 703 Id. at 60. 704 Id. 705 Id. Federal Communications Commission FCC 15-94 95 closing the proposed transaction.706 ACM et al. state that nothing would prevent the Applicants from enticing their subscribers to switch from U-verse to DIRECTV and then terminating U-verse video.707 242. Despite the Applicants’ refutation that they would migrate viewers from U-verse video to DIRECTV in order to limit or avoid PEG obligations, ACM et al. nonetheless contend that the Applicants do not deny the observations that AT&T has strong incentives to do so.708 ACM et al. state that should the Commission decide to grant the Application, the Commission should impose a condition that would require AT&T to continue offering U-verse video with PEG programming, and inform subscribers of the availability of U-verse video, for at least five years after closing the transaction.709 ACM et al. also request conditions requiring AT&T to carry PEG channels in the same manner with the same functionalities and accessibility as other linear local broadcast channels.710 243. Discussion. While we reaffirm the importance of PEG programming, the record does not establish a transaction-related harm to PEG programming, and we find it unnecessary to impose PEG programming-related conditions on the proposed transaction. We recognize that PEG channels serve important public interest objectives by providing a platform for causes and organizations that might otherwise not receive carriage on cable systems.711 Among other things, PEG channels educate the local electorate by providing opportunities for local candidates to address the public during local elections.712 Further, we acknowledge the argument of ACM et al. that the programming provided on PEG channels is unique and would likely be limited or nonexistent on commercial television channels.713 244. To the extent that the harm alleged by the commenters results from a dissatisfaction with the disparate public interest programming obligations imposed on different types of MVPDs, rather than with a public interest harm resulting from the proposed transaction, we do not find a transaction-specific harm.714 Furthermore, we note that, to the extent that commenters take issue with the manner in which AT&T currently offers PEG programming, such concerns are properly addressed in the docket specifically designated to that issue.715 Finally, while commenters have alleged that AT&T could terminate its entire U-verse video service in an effort to avoid providing PEG channels, AT&T has stated 706 ACM et al. Reply at 4. 707 Id. Moreover, several parties, including the Applicants themselves, point out that AT&T would be able to free up more capacity for its broadband service by shutting off its U-verse video product and relying solely on DIRECTV for video. See id. at 4; ATT-FCC-03377455, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ; DTVFCC-01487673, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ; see also NATOA Comments at 2-3 (discussing “potential efforts by AT&T to eliminate its U-verse video product and replace it with satellite video service”); ACM Feb. 24, 2015, Ex Parte Letter, Attachment at 2-3 (stating that AT&T has “obvious economic incentives” to migrate U-verse video subscribers to DIRECTV’s DBS service and that reduced MVPD capacity demands on AT&T’s landline U-verse network would reduce AT&T’s incentive to invest in a higher-capacity landline broadband network). 708 ACM et al. Reply at 4-5; ACM Feb. 24, 2015, Ex Parte Letter, Attachment at 2. 709 Letter from Tillman L. Lay, Counsel for the Alliance for Community Media, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 3 (March 3, 2015). 710 Id. at 2. 711 ACM et al. Petition at 14. 712 Id.; see also H.R. REP. NO. 102-628, at 85. 713 ACM et al. Petition at 14-18. 714 After the transaction, the combined entity will remain subject to the same carriage obligations for PEG and public interest programming on its U-verse video and DBS video services, respectively, as AT&T and DIRECTV were subject to prior to the transaction. 715 We do not agree with ACM et al. that the MB Docket No. 09-13 proceeding should be resolved before we take action on the instant transaction. See ACM et al. Petition at 20. Federal Communications Commission FCC 15-94 96 clearly in its Application that it intends to continue offering its U-verse video service (including PEG channels) alongside DIRECTV after closing the transaction.716 Similarly, we have found no evidence in the record that AT&T is planning to modify how it provides PEG programming on U-verse video following the proposed transaction. c. Local Broadcast Television Stations 245. Background. The Communications Act requires satellite providers to carry all broadcast television stations in a local television market, defined by the Commission’s rules using Nielsen’s DMA determinations, if they carry one local television signal in that market under the compulsory copyright license.717 With the exception of Alaska and Hawaii, a satellite carrier is not required to carry local television signals in a market where it carries no other local television signals.718 246. Positions of the Parties. One commenter, the National Association of Broadcasters (“NAB”), states that DIRECTV does not carry local broadcast stations in 13 of the DMAs in which it operates.719 NAB states that, as part of its decision on the proposed transaction, the Commission should require the Applicants to commit to expanding local television offerings to all 210 of the DMAs where the combined entity would operate.720 The Applicants respond that this request is unrelated to the proposed transaction because DIRECTV’s incentives and capability to carry local television signals would not be affected by the transaction.721 They also note that broadcasters attempted to impose a similar requirement in a previous transaction involving DIRECTV.722 247. The Commission has previously declined to do so because such a requirement did not 716 See Application at 72; Joint Opposition at 60-62. 717 See 47 U.S.C. § 338. This obligation is often referred to as the “carry-one, carry-all” requirement. Satellite video providers may choose to carry a television station’s signal pursuant to a compulsory copyright license, in which case the satellite distributor makes statutorily calculated payments to the U.S. Register of Copyrights, which then distributes payments to the individual copyright holder. See 17 U.S.C. §§ 119(b), 122. 718 See 47 U.S.C. § 338(a)(4) (mandating satellite carriers to retransmit the analog and digital signals of each television station in local markets in Alaska and Hawaii); Implementation of Section 210 of the Satellite Home Viewer Extension and Reauthorization Act of 2004 to Amend Section 338 of the Communications Act, MB Docket No. 05-181, Report and Order, 20 FCC Rcd 14242 (2005). 719 NAB Comments at 7. In addition, on June 18, 2015, several television affiliates (the “Joint Television Network Affiliates”) jointly filed a brief ex parte letter in support of NAB’s Comments. Letter from John R. Feore et al., Counsel for FBC Television Affiliates Association et al., to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 1-3 (June 18, 2015). The Joint Television Network Affiliates’ belated filing asserts generally that the proposed transaction will reduce competition and “create new threats to localism” such that the Commission should require DIRECTV to carry local broadcast television stations in all 210 DMAs to offset the harms to competition and localism. Id. at 2. The Joint Television Network Broadcast Affiliates’ general assertions that the transaction will harm local television and “exaggerate [DIRECTV]’s disinterest in providing local service to markets throughout the country” are ultimately vague and unsupported. See id. Furthermore, as discussed below, we do not find the requested condition to be transaction specific nor necessary to remedy a substantiated harm. Because we find that issues relating to DIRECTV’s carriage of local broadcast television stations are not transaction specific, we decline to provide the remedies requested by broadcasters through their ex parte filings in this proceeding. See id. at 1; see also Letter from Francisco M. Montero, Counsel for ZGS Commc’ns Inc. (“ZGS”), to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (July 16, 2015) (requesting Commission assistance in a matter relating to DIRECTV’s refusal to carry ZGS’s local television stations) (“ZGS July 16, 2015, Ex Parte Letter”). We also note that ZGS does not allege that DIRECTV has violated the Communications Act or any Commission rule, and ZGS takes no position on whether the Application should be granted. ZGS July 16, 2015, Ex Parte Letter at 5. 720 NAB Comments at 7-8. 721 Joint Opposition at 66. 722 See id. at 65-66 (citing Liberty Media-DIRECTV Order, 23 FCC Rcd at 3330, ¶ 137). Federal Communications Commission FCC 15-94 97 remedy any transaction-specific harm.723 The Applicants ask the Commission to reach the same result here.724 The Applicants note further that the Commission has previously recognized that limits in satellite capacity and infrastructure costs are challenges for DBS providers for providing local broadcast signals.725 The Applicants state that DIRECTV provides local-into-local service in the vast majority of DMAs in the United States despite these challenges and the absence of any statutory requirement to do so.726 In its reply comments, NAB attempts to distinguish the proposed transaction from Commission precedent by noting the difference in the scale of the transactions and the alleged harms.727 248. Discussion. We do not find that the proposed transaction gives rise to a harm that would require us to expand DIRECTV’s local television signal carriage obligations.728 We agree with the Applicants that nothing in the record indicates that the transaction would affect DIRECTV’s incentives or capability to carry local broadcast channels.729 Therefore we agree with the Applicants that the issue of DIRECTV’s carriage of local television signals is unrelated to the proposed transaction. E. Video Device Market 249. Background. Consistent with Section 629(a) of the Act, the Commission adopted regulations in 1998 requiring MVPDs to allow unaffiliated navigation device manufacturers access to their video programming and security systems.730 These “set-top box” rules were later augmented by a 2003 Memorandum of Understanding between the cable industry and the consumer electronics industry.731 When the set-top box rules were adopted, the Commission concluded that DBS operators, 723 See Liberty Media-DIRECTV Order, 23 FCC Rcd at 3330, ¶ 137. 724 Joint Opposition at 65-66. 725 Id. at 66. 726 Id. 727 Reply Comments of National Association of Broadcasters, MB Docket 14-90, at 3-5 (filed Jan. 7, 2015) (“NAB Reply”) (distinguishing the proposed transaction from the Liberty Media-DIRECTV transaction). 728 Although NAB states that its proposed requirement is necessary to “mitigate” a transaction-specific harm, the harm cited by NAB – that the transaction would result in competitive advantages and greater negotiating leverage for the combined entity to the detriment of local broadcast stations – would not be remedied by NAB’s proposed condition to expand DIRECTV’s local television signal carriage obligations. See NAB Comments at 4-6; NAB Reply at 3, 5-6. Furthermore, the alleged harm of greater negotiating leverage by the combined entity in retransmission consent negotiations is addressed above. See Section X.D.2.a. 729 The Applicants state that “DIRECTV’s satellite capacity” and the “infrastructure costs of collecting and uplinking broadcast signals” are the key factors affecting DIRECTV’s local signal carriage decisions. Joint Opposition at 66. The Applicants maintain that the proposed transaction would not affect either factor. Id. NAB does not refute this claim. 730 Implementation of Section 304 of the Telecommunications Act of 1996: Commercial Availability of Navigation Devices, CS Docket No. 97-80, Report and Order, 13 FCC Rcd 14775 (1998) (“First Navigation Device Report and Order”); see also Section 629(a) of the Communications Act of 1934, as amended, 47 U.S.C. § 549(a) (requiring the Commission to “adopt regulations to assure the commercial availability, to consumers of multichannel video programming and other services offered over multichannel video programming systems, of converter boxes, interactive communications equipment, and other equipment used by consumers to access multichannel video programming and other services offered over multichannel video programming systems, from manufacturers, retailers, and other vendors not affiliated with any multichannel video programming distributor”); see also Telecommunications Act of 1996, Pub. L. No. 104-104, § 304, 110 Stat. 56, 125-126 (1996). 731 Implementation of Section 304 of the Telecommunications Act of 1996: Commercial Availability of Navigation Devices, CS Docket No. 97-80, Second Report and Order and Second Further Notice of Proposed Rulemaking, 18 FCC Rcd 20885, 20886-87, ¶ 2 (2003) (“Second Navigation Device Report and Order”). The D.C. Circuit vacated the Second Navigation Device Report and Order in EchoStar Satellite L.L.C. v. FCC, 704 F.3d 992 (D.C. Cir. 2013), on the grounds that DBS operators were not party to the Memorandum of Understanding. Regulations (continued….) Federal Communications Commission FCC 15-94 98 such as DIRECTV, should be subject to the navigation device regulations. However, the Commission exempted from the rules MVPDs – such as DBS providers at the time – that supported nationally operable navigation devices made available at retail throughout the United States from vendors unaffiliated with the MVPD.732 At the time the rules were adopted, major MVPDs using novel network technologies different from cable systems, such as AT&T U-verse, did not exist, and the Commission has never addressed directly if, and how, such MVPDs should comply with these rules. 250. Positions of the Parties. Commenters have raised questions regarding the Applicants’ responsibilities under the Commission’s set-top box rules in light of the proposed transaction. TiVo Inc. (“TiVo”) argues that the transaction would create the largest MVPD in the country and that the Commission’s earlier rationale for exempting DBS operators from the navigation device regulations no longer applies.733 TiVo further contends that AT&T should be required to comply fully with the rules.734 Accordingly, TiVo requests that the Commission impose conditions that would require the Applicants to grant TiVo and other retail device manufacturers access to the Applicants’ conditional access systems, subject to certain expressly limited conditions.735 Specifically, TiVo requests that the Commission require the Applicants to comply with Section 629 of the Act and Sections 76.1201, 76.1203, and 76.1205 of the Commission’s rules.736 251. In addition, Public Knowledge-ILSR argue that the combined entity would have greater incentive to discriminate against competing video providers.737 Public Knowledge-ILSR assert that the Commission should ensure that consumers can access OVDs through operator-leased set-top boxes to encourage OVD development because set-top boxes are still the consumers’ default video platform.738 Public Knowledge-ILSR further argue that because MVPD-provided content remains indispensable to most consumers, third-party device manufacturers must have access to a video distributor’s signals and security system in order to ensure consumers have meaningful choices for video devices and their attendant user interfaces and features.739 252. The Applicants respond that the set-top box conditions proposed by TiVo and Public Knowledge-ILSR are unnecessary and not transaction specific.740 The Applicants argue that the TV Everywhere model, under which they authenticate applications provided by either the video programmers or by themselves, offers the kind of device competition envisioned by Congress and the Commission, and (Continued from previous page) adopted in the First Navigation Device Report and Order, however, are unaffected by the ruling and remain in effect. 732 First Navigation Device Report and Order, 13 FCC Rcd at 14783, ¶ 22; see also 47 C.F.R. § 76.1204(a)(2)(ii). The Commission effectively granted an exemption from Section 629 regulations to DBS operators for as long as they offer navigation devices at retail. 733 Comments of TiVo Inc., MB Docket 14-90, at 2-6 (filed Sept. 16, 2014) (“TiVo Comments”). 734 Id. at 5-7; Letter from Henry Goldberg and Devendra T. Kumar, Attorneys for TiVo Inc., to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 1 (Nov. 19, 2014) (“TiVo Nov. 19, 2014, Ex Parte Letter”). 735 See TiVo Comments at 7-9; TiVo Nov. 19, 2014, Ex Parte Letter at 2. Consistent with longstanding Commission policy, TiVo acknowledges the need to prevent the attachment of devices that may physically harm the network or facilitate theft of service. TiVo Comments at 8-9; see also 47 C.F.R. § 76.1201. 736 TiVo Comments at 7-9. 737 Public Knowledge-ILSR Petition at 5. 738 Id. at 9-11. 739 Id. 740 Joint Opposition at 62-64. Federal Communications Commission FCC 15-94 99 thus complies with the Commission’s rules.741 DIRECTV also argues that its RVU technology742 provides the kind of device competition envisioned by Congress.743 TiVo replies, however, that the Applicants merely provide the ability to watch video over multiple screens but do not demonstrate competition in retail consumer devices used to navigate video programming – that is, devices that allow a manufacturer to innovate in the way consumers view, search for, and discover programming – as the rules originally intended.744 Public Knowledge clarifies that it does not ask the Commission to impose conditions in lieu of its Section 629 rulemaking authority, but it requests conditions to protect device manufacturers – including those not covered by Section 629 – from possible anticompetitive behavior arising from the Applicants’ increased incentive to discriminate against rival video providers.745 253. Discussion. We agree with the Applicants that the transaction does not create a public interest harm with respect to set-top boxes. Rather we find that commenters raise broader regulatory policy questions that are more appropriately addressed in the rulemaking context. In addition, there is nothing about the record in this transaction to demonstrate a change in the Applicants’ incentive or ability to limit competition in the market for navigation devices. Given the lack of a transaction-related harm, and noting the Commission’s other activities related to these rules, discussed further below, we decline to adopt the conditions requested by the commenters or to take other action in this context. 254. Congress recently enacted the STELA Reauthorization Act of 2014,746 which directed the Commission to host an advisory committee comprised of representatives from MVPDs, content providers, and the consumer electronics industry, as well as related public interest organizations and academics, “to identify, report, and recommend performance objectives, technical capabilities, and technical standards of a not unduly burdensome, uniform, and technology- and platform-neutral software- based downloadable security system” to promote the competitive availability of navigation devices in furtherance of Section 629 of the Communications Act.747 The newly created Downloadable Security Technology Advisory Committee (“DSTAC”) must file a report with the Commission by September 4, 2015, to detail its findings and recommendations.748 Accordingly, many of the regulatory policy issues raised by the parties, which we find do not raise transaction-specific public interest harms, will likely be addressed in a broader, industry-wide context, both through the DSTAC and possible future Commission proceedings. F. Potential Loss of DIRECTV as a Partner for MDU Broadband Entrants 255. Positions of the Parties. The Independent Multifamily Communications Council (“IMCC”) expresses concern that the proposed transaction would result in the loss of DIRECTV as a 741 Id. Additionally, we note that AT&T has asserted that its U-verse video service is not a cable system, but it maintains that it is in compliance with the navigation device regulations. See Letter from Robert W. Quinn, Jr., Senior Vice President – Federal Regulatory, AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket 09-13, Attachment at 4 (Sept. 17, 2009); Joint Opposition at 62-64. 742 RVU technology allows consumers to access DIRECTV’s user interface and programming on smart televisions and other devices, as long as they have a DIRECTV gateway set-top box in their home. See, DIRECTV, What is DIRECTV Ready TV and How Does it Work, available at https://support.directv.com/app/answers/detail/a_id/3992/~/what-is-a-directv-ready-tv-and-how-does-it-work%3F (visited June 18, 2015). 743 Joint Opposition at 63-64. 744 Reply Comments of TiVo Inc., MB Docket 14-90, at 3 (filed Jan. 7, 2015) (“TiVo Reply”). 745 Public Knowledge Reply at 6. 746 H.R. 5728, 113th Cong., § 106 (2014) (STELAR). 747 See Downloadable Security Technology Advisory Committee, www.fcc.gov/DSTAC (visited June 18, 2015). 748 See id. Federal Communications Commission FCC 15-94 100 potential partner for broadband providers.749 Specifically, IMCC raises concerns about the potential impact of the transaction on private cable operators (“PCOs”) that offer voice, video, and data services to residents of MDUs and compete with AT&T in the provision of voice and data services to MDUs.750 IMCC argues that the proposed transaction would make it difficult for PCOs to acquire programming, which, as acknowledged by the Applicants, is necessary for broadband deployment. 256. IMCC states that PCOs can acquire video programming content from only DIRECTV and DISH and that the PCO industry requires at least two sources of video programming content to effectively serve MDUs.751 IMCC also emphasizes that PCOs increasingly can provide high-speed Internet access services to MDU customers,752 offering a competitive alternative to franchised cable operators and telephone MVPDs, particularly at connectivity speeds greater than 20 Mbps.753 However, IMCC indicates that there are significant barriers to entry that undermine a PCO’s ability to build out its Internet access services.754 IMCC explains that PCOs are developing the infrastructure required to deliver video, Internet, and telephone services to MDU communities. It is concerned that if the combined entity terminates DIRECTV’s contracts with PCOs such that they can no longer access DIRECTV programming, this outcome would impair the PCOs’ ability to accelerate deployment of advanced services in the MDU marketplace.755 IMCC argues that AT&T, unlike DIRECTV, would view the PCOs as rivals.756 IMCC contends that, post-transaction, AT&T might exercise its right to terminate programming contracts with PCOs, or it might refuse to allow existing programming contracts to renew, in an attempt to force PCOs out of MDUs.757 257. IMCC asserts that the elimination of PCOs would “leave many MDU residents with inadequate Internet services (often, maximum speeds at between 3 and 6 Mbps) and in many cases, few or no options for access” to indispensable broadband services.758 IMCC accordingly requests that the Commission impose conditions to modify existing contracts between DIRECTV and PCOs, including provisions relating to contract period, termination of rights, compensation, and protection of confidential and subscriber information.759 749 See Comments of Independent Multifamily Communications Council, MB Docket 14-90, at 6 (filed Sept. 12, 2014) (“IMCC Comments”); see also supra ¶¶ 194-195, 197; infra ¶¶ 260-261. 750 IMCC Comments at 3-4. PCOs were formerly known as satellite master antenna, or SMATV, systems. PCOs use a satellite master antenna to distribute video programming throughout a property (e.g., an apartment building, hotel, hospital, or commercial property with multiple tenants) from a single satellite feed. 751 Id. at 5-6. IMCC explains that many PCOs have existing agreements with DIRECTV to obtain video programming to serve their properties. These PCOs also compete with AT&T and other providers at such properties. Id. at 6; Letter from Gary I. Resnick, Counsel for Independent Multi-Family Communications Council, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 2 (Nov. 25, 2014). 752 Letter from Valerie Sargent, Managing Director, Independent Multi-Family Communications Council, to Staff, FCC, MB Docket No. 14-90, at 3 (Dec. 16, 2014) (“IMCC Dec. 16, 2014, Ex Parte Letter”) (“particularly [PCOs offer Internet service to MDUs] in less desirable communities that do not attract significant investment from the franchised cable operators and telecommunications providers”). 753 Id. at 3-4. 754 Id. at 4. 755 Id. 756 IMCC Comments at 6; IMCC Dec. 16, 2014, Ex Parte Letter at 3. 757 IMCC Comments at 7. 758 IMCC Dec. 16, 2014, Ex Parte Letter at 4. 759 IMCC Comments at 12-13; IMCC Dec. 16, 2014, Ex Parte Letter at 4-9. Federal Communications Commission FCC 15-94 101 258. The Applicants argue that the concerns raised by IMCC are not related to the transaction and urge that the Commission reject the proposed conditions.760 They contend that IMCC merely seeks to amend the terms of existing contracts between DIRECTV and PCOs.761 The Applicants assert that the contracts were privately negotiated with DIRECTV and their terms would govern post-transaction.762 259. Discussion. The Commission has previously determined that Commission proceedings are not the proper forum for the adjudication of private contractual disputes or the modification of contract terms between parties,763 and therefore, we decline to adopt IMCC’s proposed conditions, which request that the Commission modify terms of the existing agreements in effect between DIRECTV and various PCOs. Although we acknowledge that the PCO business model relies on satellite antennas to receive video programming from a DBS provider, we find no basis for prohibiting AT&T from competing for the business of providing its broadband service to MDUs, nor do we have a basis to conclude that PCOs would be unable to acquire programming as a result of the transaction. The record does not support a finding that AT&T intends to preclude DIRECTV from bundling its services with those of an unaffiliated broadband service provider or private cable operator. G. Increased Incentive of Combined Entity to Hinder Competition for Broadband in MDUs 260. Positions of the Parties. Cox claims that, prior to the announcement of this transaction, DIRECTV engaged in conduct that interfered with Cox’s deployment of broadband services to MDUs and that the combined entity would have an increased incentive to engage in this conduct.764 Cox’s concern relates to subscribers who switch to DIRECTV for video services but retain their existing cable operator for broadband.765 In that situation, Cox claims that DIRECTV insists on attaching diplexers 760 See Joint Opposition at 70. 761 Id. 762 Id. at 71. 763 See Verizon Communications, Inc., Transferor, and América Móvil, S.A. DE C.V., Transferee, Application for Authority to Transfer Control of Telecomunicaciones de Puerto Rico, Inc. (TELPRI), WT Docket No. 06-113, Memorandum Opinion and Order and Declaratory Ruling, 22 FCC Rcd 6195, 6208-09, ¶ 29 (2007) (“Moreover, to the extent that certain conditions are being negotiated in ongoing contract disputes … we decline, as we did in the Verizon/MCI Order, to address them in this proceeding.”); Telefuture Fresno LLC v. Echostar Communications Corp., Memorandum Opinion and Order, DA 03-3509, 18 FCC Rcd 22940, 22944, ¶ 12 (MB 2003) (“[W]e will not interject ourselves into specific arguments concerning private agreements between broadcasters and satellite carriers. … Contractual issues are to be resolved by the parties or by courts of proper jurisdiction.”); AirTouch Paging, Inc., Application for Consent to Partial Assignment of Station KCC485 to Schuylkill Mobile Fone, Inc., Order, DA 99- 1175, 14 FCC Rcd 9658, 9660, ¶ 5 (WTB 1999) (“The Commission … is not the proper forum for the adjudication of private contractual disputes.” (citing Listener’s Guild, Inc. v. FCC, 813 F.2d 465, 469 (D.C. Cir. 1987); Application of WorldCom, Inc. and MCI Communications Corporation for Transfer of Control of MCI Communications Corporation to WorldCom, Inc., Memorandum Opinion and Order, 13 FCC Rcd 18025, 18148, ¶ 214 (1998); PCS 2000, L.P., Memorandum Opinion and Order, 12 FCC Rcd 1681, 1691, ¶ 23 (1997); Applications of Centel Corporation and Sprint Corporation for Consent to the Transfer of Control of Authorizations in the Domestic Public Cellular Radio Telecommunications Service and Other Common Carrier Services, Memorandum Opinion and Order, 8 FCC Rcd 1829, 1831, ¶ 10 (CCB 1993))). See also SoftBank-Sprint Order, 28 FCC Rcd at 9676, ¶ 85 (“We agree with Sprint that these intercarrier compensation disputes are not merger specific, are based on arguments about prior conduct by Sprint, and are more appropriately resolved through the contractual provisions between the parties or through the Commission’s complaint process under section 208 of the Act. As the Commission has repeatedly held, we will generally not impose conditions to remedy pre-existing harms or harms that are unrelated to the transaction at issue.”) (citations omitted). 764 See Cox Petition at 29-34; Cox Reply 13-14; see also Cox Dec. 22, 2014, Ex Parte Letter, Attachment at 2; Cox Dec. 4, 2014, Ex Parte Letter, Attachment at 2; Cox Nov. 7, 2014, Ex Parte Letter at 2-3. 765 See Cox Petition at 29-30. The Commission’s inside wiring rules pertain to the disposition of inside wiring installed by the cable operator within the premises of the subscriber after a subscriber voluntarily terminates cable (continued….) Federal Communications Commission FCC 15-94 102 (splitters) to the MDU’s inside wiring that cause harmful interference to Cox’s DOCSIS 3.0 broadband signals.766 Cox contends that DIRECTV has insisted that Cox install a second MDU wire if it wants to provide interference-free broadband service.767 Cox states that, post-transaction, the combined entity would have an increased incentive to use DIRECTV video service installations at MDUs as a means of assuming control over internal MDU wiring and thereby would hinder competition to the combined entity’s bundled offering.768 261. Cox proposes that the Commission adopt conditions to address its concerns.769 In particular, Cox seeks restrictions on the combined entity’s ability to “commandeer” cable wiring in MDUs and to employ technologies that preclude competitors from using that wire to provide standalone broadband services.770 Cox also asks the Commission to adopt conditions that would require the combined entity to either employ technologies that can successfully coexist with existing inside wiring or to run their own wiring to new customer units.771 262. The Applicants oppose Cox’s requests for conditions and assert that the transaction would enhance competition for consumers who reside in MDUs.772 The Applicants assert that Cox’s concerns are not transaction-related, but are issues of general industry concern.773 The Applicants also assert that any “interference” between the diplexers installed by DIRECTV and Cox’s equipment is the fault of Cox, not DIRECTV, and that Cox has refused all of DIRECTV’s efforts to cooperate to avoid disruption of service.774 The Applicants state that Cox and other incumbent cable operators continue to dominate the MDU market in most areas, “[d]espite efforts by Congress and the Commission to remove barriers to video competition for [MDUs], and despite DIRECTV’s diligent attempts to enter the MDU (Continued from previous page) service. 47 C.F.R. §§ 76.800-06. Cox claims, however, that there is “no specific FCC rule” governing the treatment of cable operator-deployed MDU wiring when an MDU resident switches to DIRECTV for video but retains the cable operator for broadband. Cox Petition at 30. 766 See Cox Petition at 29. Cox indicates that these diplexers would also interfere with DOCSIS 3.1 signals. Id. at 30; Cox Reply at 13-14 (noting that “[b]ecause DIRECTV has been unwilling to consider reasonable solutions to the interference problems caused by its diplexers on shared MDU internal wiring in which both sides bear burdens in order to preserve consumer choice, Cox must either decline to serve units that wish to receive [Cox’s] DOCSIS 3.0 cable modem service or incur the costs of running a second wire” and that this is “not a sustainable method of providing DOCSIS 3.0 at MDUs”). 767 See Cox Petition at 30; Cox Reply at 14 n.39. 768 See Cox Petition at 32-33; Cox Reply at 14. Cox asserts that because AT&T offers a broadband Internet access service, it “will therefore have even less incentive to act reasonably in circumstances where MDU tenants wish to retain Cox as their broadband provider. Those circumstances may be particularly prevalent in areas where AT&T’s broadband offering consists only of DSL service, given the disparity in service quality between DOCSIS 3.0 broadband and convention and DSL speeds.” Cox Reply at 14. 769 See Cox Petition 34-35; see also Cox Dec. 22, 2014, Ex Parte Letter, Attachment at 2; Cox Dec. 4, 2014, Ex Parte Letter, Attachment at 2; Cox Nov. 7, 2014, Ex Parte Letter at 3. 770 Cox Dec. 22, 2014, Ex Parte Letter, Attachment at 2; Cox Dec. 4, 2014, Ex Parte Letter, Attachment at 2; Cox Nov. 7, 2014, Ex Parte Letter at 2-3; see also Cox Petition at 35; Cox Reply at 13. 771 Cox Nov. 7, 2014, Ex Parte Letter at 3; see also Cox Petition at 34-35. Cox also proposes that any disputes arising between AT&T-DIRECTV and any service provider covered by these conditions should initially be presented to the Commission or its designee for mediation. If the parties are unable to resolve the dispute via mediation, either party should be permitted to seek review of the dispute by the Media Bureau, subject to procedures established by the Commission. Cox Petition at 35. 772 See Joint Opposition at 69-70. 773 See id. 774 See id. at 72 n.265; Conditions Ex Parte Presentation at 11 (footnote omitted). Federal Communications Commission FCC 15-94 103 market.”775 According to the Applicants, Cox’s proposed condition is an attempt to hinder competition from a new rival in the MDU space, rather than an attempt to promote competition.776 263. Moreover, the Applicants state, DIRECTV designed its equipment to share inside wiring based on the assumption that cable operators such as Cox would use an industry-standard frequency range for their broadband service.777 The Applicants assert that DIRECTV shared its diplexer design specifications with Cox before beginning installation and that Cox “unilaterally” changed the frequency range used for its broadband service after DIRECTV began installing these devices in MDUs.778 The Applicants state that DIRECTV’s efforts to work with Cox to avoid disruption to either service “met with little cooperation,” leading DIRECTV to transfer rights of entry to the MDUs in which Cox provided broadband and focus on other properties where it did not face a similar issue.779 The Applicants argue that adopting Cox’s proposed condition simply would add another disincentive to competitive MVPD entry.780 The Applicants therefore urge the Commission to reject Cox’s proposed conditions.781 264. Discussion. We acknowledge that Cox’s allegations that DIRECTV is intentionally causing harmful interference to Cox’s broadband service, if established, would be an anticompetitive harm. However, any such claims, which arise from a pre-existing dispute with DIRECTV and are not established on the record in this proceeding, are not related to this transaction. The Commission does not impose conditions to remedy pre-existing disputes between parties that are unrelated to the transaction at issue,782 and therefore, we decline to adopt Cox’s proposed conditions regarding internal wiring in MDUs.783 If any pre-existing disputes give rise to violations of the Commission’s rules, those could be addressed through complaints to or investigations by the Enforcement Bureau. 775 Conditions Ex Parte Presentation at 11. 776 Id. 777 Id. at 12. 778 Id. 779 Id. 780 Id. 781 Joint Opposition at 70; Conditions Ex Parte Presentation at 12. 782 See, e.g., AT&T-Leap Wireless Order, 29 FCC Rcd at 2767, 2805, ¶¶ 74, 171 (rejecting proposed conditions as they were not narrowly tailored or necessary to remedy any purported harms arising out of the transaction); Applications of AT&T Inc. and Atlantic Tele-Network, Inc. for Consent to Transfer Control of and Assign Licenses and Authorizations, WT Docket No. 13-54, Memorandum Opinion and Order, DA 13-1940, 28 FCC Rcd 13670, 13704, ¶¶ 62-63 (WTB, IB 2013) (“AT&T-ATN Order”) (rejecting request to place interoperability conditions on AT&T because alleged harms were not transaction specific); Applications of Cellco Partnership d/b/a Verizon Wireless and SpectrumCo LLC and Cox TMI, LLC for Consent to Assign AWS-1 Licenses, WT Docket No. 12-4, Memorandum Opinion and Order and Declaratory Ruling, 27 FCC Rcd 10698, 10734, ¶ 94 (2013) (“Verizon Wireless-SpectrumCo Order”); Application of AT&T Inc. and Qualcomm Incorporated for Consent to Assign Licenses and Authorizations, WT Docket No. 11-418, Order, 26 FCC Rcd 17589, 17622, ¶ 79 (2011); Verizon Wireless-ALLTEL Order, 23 FCC Rcd at 17463, ¶ 29; Sprint Nextel Corporation and Clearwire Corporation, Applications for Consent to Transfer Control of Licenses, Leases, and Authorizations, WT Docket No. 08-94, Memorandum Opinion and Order, 23 FCC Rcd 17570, 17582, ¶ 22 (2008) (“Sprint Nextel-Clearwire Order”); AT&T-Cingular Order, 19 FCC Rcd at 21545-46, ¶ 43. 783 Cox also argues that the Commission should require AT&T to commit to compliance with all Section 251 and 252 interconnection obligations as AT&T transitions from delivering telephone traffic over its existing time-division multiplexed (“TDM”) networks to all-IP networks. Cox Petition at 21-25; Cox Nov. 7, 2014, Ex Parte Letter at 2. Cox’s concerns regarding these requirements do not raise any transaction-specific harms and are not appropriately addressed in this proceeding. See, e.g., Applications for Consent to the Transfer of Control of Licenses from Comcast Corporation and AT&T Corp., Transferors, to AT&T Comcast Corporation, Transferee, WT Docket No. 02-70, Order, 17 FCC Rcd 22633, 22637, ¶ 11 (2002) (“Comcast-AT&T Broadband Order”) (“merger review is (continued….) Federal Communications Commission FCC 15-94 104 H. Increased Incentive of Combined Entity to Hinder Competition in Mobile Wireless Sector 265. Positions of the Parties. DISH claims that the transaction would allow AT&T- DIRECTV to bundle fixed and mobile video, data, and voice in direct competition to DISH and that such bundling could give the combined entity the incentive to stifle wireless competition by refusing to enter into data roaming agreements with DISH.784 DISH argues that the Data Roaming Order is not adequate to protect new entrants and asserts that the Commission instead should require AT&T-DIRECTV to provide low-cost data roaming on the company’s networks to mobile providers, including DISH, with unbuilt facilities.785 ACM et al. argue that the transaction would eliminate any incentive DIRECTV might otherwise have to participate in upcoming Commission spectrum auctions.786 Further, ACM et al. argue that a bundle of AT&T’s wireless broadband service with DIRECTV’s video service would strengthen AT&T’s dominant wireless position and further entrench the Verizon Wireless and AT&T wireless duopoly.787 266. In response, the Applicants state that “the Commission’s data roaming rules already make certain that entities like DISH can obtain data roaming agreements on commercially reasonable terms and conditions.”788 The Applicants argue that DISH’s claim that the combined entity would see DISH as a competitive threat and seek to thwart it is speculative, as it is based on potential services DISH may offer in the future.789 267. Discussion. We find that DISH’s position that the transaction would limit competition for bundling wireless services with broadband and video products and that such bundling could give the combined entity the incentive to stifle wireless competition by refusing to enter into data roaming agreements with DISH is not supported on the basis of the record in front of us.790 We note that the Commission’s general roaming policies and rules are intended to enable entities to obtain roaming agreements on reasonable terms and conditions.791 Further, if an entity were to face difficulties in (Continued from previous page) limited to consideration of merger-specific effects”); Verizon Wireless-ALLTEL Order, 23 FCC Rcd at 17463, ¶ 29 (The Commission “will not impose conditions to remedy pre-existing harms or harms that are unrelated to the transaction.”); Applications of AT&T Inc. and Centennial Communications Corporation for Consent to Transfer Control of Licenses, Authorizations, and Spectrum Leasing Arrangements, WT Docket No. 08-246, Memorandum Opinion and Order, 24 FCC Rcd 13915, 13969, ¶ 133 (2009) (“AT&T-Centennial Order”) (The Commission will “impose conditions only to remedy harms that arise from the transaction (i.e., transaction-specific harms) ….”). 784 DISH Petition at 30-31. 785 Id. 786 See ACM et al. Petition at 7-8. 787 Id. at 10. 788 Joint Opposition at 67. 789 Id. 790 In contrast, the Commission has previously conditioned consent in certain cases on the ability of service providers to have access, on behalf of their customers, to roaming services in the areas affected by the transaction in order to assure an orderly transition where service providers intend to change network technology (e.g., a change from Code Division Multiple Access (“CDMA”) to Global System for Mobile Communications (“GSM”) technology) because the assignee might not sufficiently preserve or maintain the assignor’s original network technology. See, e.g., AT&T-Verizon Wireless Order, 25 FCC Rcd at 8742-43, 8746, ¶¶ 89, 96; Verizon Wireless- ALLTEL Order, 23 FCC Rcd at 17501, 17254, ¶¶ 126-127, 178. See also AT&T-Leap Wireless Order, 29 FCC Rcd at 2783-85, ¶¶ 105, 108; AT&T-ATN Order, 28 FCC Rcd at 13703, ¶ 60. 791 See Reexamination of Roaming Obligations of Commercial Mobile Radio Service Providers and Other Providers of Mobile Data Services, WT Docket No. 05-265, Second Report and Order, 26 FCC Rcd 5411, 5451, ¶ 81 (2011), aff’d sub nom. Cellco P’ship v. FCC, 700 F.3d 534 (D.C. Cir. 2012); Reexamination of Roaming Obligations of (continued….) Federal Communications Commission FCC 15-94 105 negotiating roaming agreements, such entity may rely on the protections afforded by the Commission’s general roaming policies and rules as well as on the availability of relief under the Commission’s complaint procedures. While ACM et al. assert that the transaction would eliminate any incentive DIRECTV might have to participate in spectrum auctions, we find this concern speculative. Neither ACM et al. nor any other party offered evidence in the record suggesting that DIRECTV is likely to be a participant in any of the Commission’s upcoming auctions. I. Increased Incentive and Ability of Combined Entity to Shift Wired Subscribers to FWLL 268. Positions of the Parties. Public Knowledge-ILSR claim that AT&T may have an increased incentive as a result of the transaction to shift its wired subscribers to its FWLL network post- transaction and that this action would harm consumers by limiting their options for Internet service and forcing them onto a technology that does not best suit their needs.792 Public Knowledge-ILSR argue that the Commission should impose conditions to alleviate these concerns. Specifically, Public Knowledge- ILSR request that the Commission impose the following conditions within AT&T’s wireline service territory: (1) an adequate process for handling complaints about the quality of service of both copper and wireless service; (2) copper repair deadlines; (3) public reports on complaints; (4) assurance that a person who finds that a wireless product is unsuitable can get wired service back; (5) public reporting on the results of IP transition trials; and (6) clarity about the future of wired service for businesses and the interconnection rights of competitive carriers.793 269. The Applicants respond that the vast majority of AT&T’s planned FWLL deployment is outside AT&T’s wireline region.794 Further, the Applicants argue that the Commission has open dockets to address Public Knowledge-ILSR’s concerns and that those dockets are the appropriate place to address any such concerns.795 270. Discussion. We find that the evidence does not support a finding that AT&T would have an increased incentive or ability to shift wired subscribers to its planned FWLL deployment. Specifically, our analysis verifies that the majority of the FWLL deployment will be outside of AT&T’s wireline territory.796 Finally, the Commission has open dockets on these issues that are the appropriate forum for the issues raised by Public Knowledge-ILSR.797 Thus, we decline to impose additional conditions. J. Use of Orbit and Spectrum Resources 271. Background. DISH asks that we condition any consent to the proposed transaction on DIRECTV divesting, to DISH, its three DBS channels at the 110° W.L. orbital location.798 DISH argues (Continued from previous page) Commercial Mobile Radio Service Providers and Other Providers of Mobile Data Services, WT Docket No. 05-265, Declaratory Ruling, DA 14-1865, 29 FCC Rcd 15483, 15486, ¶ 8 (WTB 2014); 47 C.F.R. § 20.12(e)(1). 792 Public Knowledge-ILSR Petition at 11-12; Public Knowledge Reply at 4. 793 Public Knowledge-ILSR Petition at 12. 794 Joint Opposition at 72 (claiming that 85 percent of the combined entity’s FWLL deployment would be outside its wireline region). 795 Id. at 73. See Pleading Cycle Established on AT&T and NTCA Petitions, GN Docket No. 12-353, Public Notice, DA 12-1999, 27 FCC Rcd 15766 (WCB 2013) (“AT&T and NTCA Petitions PN”); Technology Transitions Policy Task Force Seeks Comment on Potential Trial, GN Docket No. 13-5, Public Notice, DA 13-1016, 28 FCC Rcd 6346 (2013) (“Technology Transitions PN”). 796 See infra Section XI.G. 797 See AT&T and NTCA Petitions PN, 27 FCC Rcd 15766; Technology Transitions PN, 28 FCC Rcd 6346. 798 DISH Petition at 27; see also DISH Reply at 5. Eight orbital locations are assigned to the United States under an international plan for the provision of broadcasting-satellite service, known as DBS service domestically, found in (continued….) Federal Communications Commission FCC 15-94 106 that DIRECTV has underutilized its three licensed DBS channels at this location because they are used to provide service to only Puerto Rico.799 DISH claims that it could put the three DBS channels to more productive use by adding them to the 29 DBS channels it already uses at this orbital location.800 DISH asserts that DIRECTV’s use of these channels is inconsistent with Commission policies on efficient spectrum use, and it alleges that their main value to DIRECTV is to block DISH’s access to the full 32- channel assignment at this orbital location.801 DISH argues that DIRECTV already enjoys other orbital resources that are superior to DISH’s own.802 The Applicants, in response, state that DIRECTV has operated continuously at the 110° W.L. orbital location since 1999, that the three channels are fully integrated into its long-term plans, and that DISH’s claims are unrelated to the proposed transaction and designed solely to benefit DISH.803 272. Discussion. We decline to impose the condition requested by DISH. The potential concern DISH describes relates to DIRECTV’s ongoing orbital location use and an authorization that has (Continued from previous page) Appendices 30 and 30A of the International Telecommunication Union Radio Regulations. The 110° W.L. orbital location is one of three of these orbital locations capable of providing coverage of the 48 contiguous United States. See Amendment of the Commission’s Policies and Rules for Processing Applications in the Direct Broadcast Satellite Service, Feasibility of Reduced Orbital Spacing for Provision of Direct Broadcast Satellite Service in the United States, IB Docket No. 06-160, Notice of Proposed Rulemaking, 21 FCC Rcd 9443, 9444-46, ¶¶ 3-6 (2006). The license to operate the DIRECTV 5 DBS space station at the 110.1° W.L. orbital location, within the 110° W.L. “orbital cluster,” is held by DIRECTV Enterprises, LLC, a wholly owned subsidiary of DIRECTV. See DIRECTV Enterprises, LLC, IBFS File No. SAT-MOD-20131114-00133 (granted Feb. 20, 2014). 799 DISH Petition at 21-22; DISH Reply at 3-5. DBS space stations transmit to subscriber antennas in the 12.2-12.7 GHz frequency band. The spectrum used for DBS service is divided into 32 frequency channels, each nominally 24 megahertz in bandwidth. DIRECTV Enterprises, LLC is authorized to operate on DBS channels 28, 30, and 32 at the 110° W.L. orbital location. See United States Satellite Broadcasting Co., Inc., Transferor, and DIRECTV Enterprises, Inc., Transferee, for Consent to Transfer of Control of the USSB II, Inc. Authorization to Operate a Direct Broadcast Satellite System Using Five Channels at the 101° W.L. Orbital Location; Authorization to Construct, Launch, and Operate a Direct Broadcast Satellite System Using Three Channels at the 110° W.L. Orbital Location; and the Related Earth Registration (Call Sign E930437), Order and Authorization, DA 99-633, 14 FCC Rcd 4585 (IB 1999); DIRECTV Enterprises, LLC, Application for Authorization to Operate DIRECTV 5, a Direct Broadcast Satellite, at the 109.8° W.L. Orbital Location, Order and Authorization, DA 05-2654, 20 FCC Rcd 15778 (IB 2005). Each DBS channel at a given orbital location can transmit multiple channels of video programming to customers, depending on the bandwidth required for each program. (High Definition (“HD”) program channels require more bandwidth than Standard Definition (“SD”) channels.) For example, DIRECTV states that the DIRECTV 5 satellite at 110.1° W.L. uses the three DBS channels to transmit 30 channels of programming. See Joint Opposition at 69. 800 DISH Petition at 22-23; id., Attachment A, Confidential Declaration of Vivek Khemka (“Khemka Statement”) at 1-2. DISH Operating L.L.C., a wholly owned subsidiary of DISH, operates on DBS channels 1-27, 29, and 31 at the 110° W.L. orbital location using the EchoStar 10 and EchoStar 11 satellites. See EchoStar Satellite Corp., Application for Modification to Direct Broadcast Satellite Authorization and for Operation Authority, Memorandum Opinion and Order, DA 99-1758, 15 FCC Rcd 6727, 6730, ¶ 7 (IB 1999); DISH Operating L.L.C., IBFS File No. SAT-MOD-20090909-00095 (granted Nov. 1, 2012) (EchoStar 10); EchoStar Satellite Operating Corporation, IBFS File No. SAT-LOA-20070622-00085 (granted Jan. 11, 2008) (EchoStar 11). 801 DISH Petition at 21-22. DISH also argues that the combined entity would have a reduced incentive to make “efficient” use of the three channels. DISH Reply at 4. 802 DISH Petition at 2, 18-20; DISH Reply at 4. 803 Joint Opposition at 68-69. Federal Communications Commission FCC 15-94 107 existed for nearly 25 years.804 There is no evidence in the record to support DISH’s contention that the transaction would harm operations at the 110° W.L. orbital location.805 XI. ANALYSIS OF POTENTIAL BENEFITS A. Analytical Framework 273. In determining whether approval of a transaction is in the public interest, the Commission evaluates whether the transaction is likely to produce public interest benefits. The Commission applies several criteria in deciding whether a claimed benefit is cognizable. First, the claimed benefit must be transaction specific. That is, the claimed benefit must be likely to occur as a result of the transaction but unlikely to be realized by other practical means having less anticompetitive effect.806 274. Second, the claimed benefit must be verifiable.807 Because much of the information relating to the potential benefits of a transaction is in the sole possession of the Applicants, they have the burden of providing sufficient evidence to support each claimed benefit to enable the Commission to verify its likelihood and magnitude.808 The Commission will discount or dismiss speculative benefits that it cannot verify.809 As the Commission explained in the EchoStar-DIRECTV HDO, “benefits that are to occur only in the distant future may be discounted or dismissed because, among other things, predictions about the more distant future are inherently more speculative than predictions about events that are expected to occur closer to the present.”810 275. Third, the Commission calculates the magnitude of benefits net of the cost of achieving them.811 Fourth, benefits must flow through to consumers, and not inure solely to the benefit of the company.812 For example, the Commission will more likely find marginal cost reductions to be 804 See Application of United States Satellite Broadcasting Company, Inc. for Modification of Construction Permit for Direct Broadcast Satellite System, Memorandum Opinion and Order, 5 FCC Rcd 7576, 7577, ¶ 6 (1990) (creating the three-DBS-channel authorization at the 110° W.L. orbital location later transferred to DIRECTV). 805 We also note that we may not take into account in this proceeding whether the public interest would be better served by the transfer of the three-channel DBS license to DISH, rather than to AT&T. See 47 U.S.C. § 310(d) (In acting on a transfer application, “the Commission may not consider whether the public interest, convenience, and necessity might be served by the transfer, assignment, or disposal of the permit or license to a person other than the proposed transferee or assignee.”). 806 See Comcast-NBCU Order, 26 FCC Rcd at 4330-31, ¶ 226; Sirius-XM Order, 23 FCC Rcd at 12383, ¶ 75; Liberty Media-DIRECTV Order, 23 FCC Rcd at 3330, ¶ 140; News Corp.-Hughes Order, 19 FCC Rcd at 610, ¶ 317; EchoStar-DIRECTV HDO, 17 FCC Rcd at 20630, ¶ 189. 807 See Comcast-NBCU Order, 26 FCC Rcd at 4331, ¶ 226; Sirius-XM Order, 23 FCC Rcd at 12383, ¶ 75; Liberty Media-DIRECTV Order, 23 FCC Rcd at 3331, ¶ 140; News Corp.-Hughes Order, 19 FCC Rcd at 610, ¶ 317; EchoStar-DIRECTV HDO, 17 FCC Rcd at 20630, ¶ 190. 808 See id. 809 See Sirius-XM Order, 23 FCC Rcd at 12383, ¶ 75; Liberty Media-DIRECTV Order, 23 FCC Rcd at 3331, ¶ 140; News Corp.-Hughes Order, 19 FCC Rcd at 611, ¶ 317; EchoStar-DIRECTV HDO, 17 FCC Rcd at 20630, ¶ 190. 810 EchoStar-DIRECTV HDO, 17 FCC Rcd at 20630-31, ¶ 190. See Comcast-NBCU Order, 26 FCC Rcd at 4331, ¶ 226; Sirius-XM Order, 23 FCC Rcd at 12383, ¶ 75; Liberty Media-DIRECTV Order, 23 FCC Rcd at 3331, ¶ 140; News Corp.-Hughes Order, 19 FCC Rcd at 611, ¶ 317. 811 See Comcast-NBCU Order, 26 FCC Rcd at 4331, ¶ 226; Sirius-XM Order, 23 FCC Rcd at 12383, ¶ 75; Liberty Media-DIRECTV Order, 23 FCC Rcd at 3331, ¶ 140; News Corp.-Hughes Order, 19 FCC Rcd at 610-11, ¶ 317; EchoStar-DIRECTV HDO, 17 FCC Rcd at 20630, ¶ 190. 812 See Comcast-NBCU Order, 26 FCC Rcd at 4331, ¶ 226; Sirius-XM Order, 23 FCC Rcd at 12383, ¶ 75; Liberty Media-DIRECTV Order, 23 FCC Rcd at 3331, ¶ 140. Federal Communications Commission FCC 15-94 108 cognizable than reductions in fixed cost because reductions in marginal cost are more likely to result in lower prices for consumers.813 276. The Commission applies a “sliding scale approach” to evaluating benefit claims.814 Under this sliding scale approach, where potential harms appear both substantial and likely, the Applicants’ demonstration of claimed benefits must reveal a higher degree of magnitude and likelihood than the Commission would otherwise demand.815 On the other hand, where potential harms appear less likely and less substantial, the Commission will accept a lesser showing.816 277. As discussed below, we recognize that the transaction offers certain benefits as a result of efficiencies associated with improved bundling opportunities. This conclusion is supported by the economic analysis and documentary evidence discussed above. We also acknowledge that the transaction likely will result in some amount of programming payment reductions, although these constitute a public interest benefit only to the extent the reductions are passed on to consumers in the form of lower prices. Other claimed benefits that result from efficiencies such as improved video programming and user interfaces are unquantified and not verified, and therefore are only minimally credited as a public interest benefit. Furthermore, as discussed in more detail below, we find that the transaction reduces the Applicants’ incentives to increase fiber deployment and, therefore, we impose a condition that AT&T build out FTTP to 12.5 million customer locations to mitigate this reduced incentive. B. Improved Bundles 278. As noted in our analysis of the competitive effects of the transaction above, the Applicants state that one of the significant public interest benefits from the transaction is that the new entity would be able to offer new and more competitive bundled services817 that can offer consumers an alternative to cable broadband and video bundles.818 The Applicants argue that consumers prefer to buy broadband and video services as a bundle, which limits the Applicants’ ability to compete because neither company can provide both services to all their customers. In addition, and as analyzed above,819 the Applicants assert that the transaction would result in lower prices and better broadband and video bundles. 279. AT&T currently offers an integrated bundle of AT&T broadband and U-verse video service only in locations within its 22-state footprint where AT&T has deployed FTTN or FTTP technologies,820 and to fewer than one-quarter of all U.S. households.821 DIRECTV does not have any 813 See News Corp.-Hughes Order, 19 FCC Rcd at 611, ¶ 317; EchoStar-DIRECTV HDO, 17 FCC Rcd at 20631, ¶ 191. 814 See Comcast-NBCU Order, 26 FCC Rcd at 4331, ¶ 227; Sirius-XM Order, 23 FCC Rcd at 12384, ¶ 76; Liberty Media-DIRECTV Order, 23 FCC Rcd at 3331, ¶ 141; News Corp.-Hughes Order, 19 FCC Rcd at 611, ¶ 318; EchoStar-DIRECTV HDO, 17 FCC Rcd at 20631, ¶ 192. 815 See id. 816 See Comcast-NBCU Order, 26 FCC Rcd at 4331, ¶ 227; Sirius-XM Order, 23 FCC Rcd at 12384, ¶ 76; Liberty Media-DIRECTV Order, 23 FCC Rcd at 3331, ¶ 141. 817 See supra ¶ 56 (describing the difference between “synthetic” and “integrated” bundles). This section analyzes bundles that include wireline broadband. Wireless broadband bundles are discussed in Section XI.G.4. 818 See, e.g., Application at 4. 819 See supra Section IX.A.4. 820 Application at 3, 53. In addition, AT&T offers the synthetic bundle of AT&T broadband and DIRECTV video within this service territory. See Lee Decl. ¶ 50. See also Competition White Paper at 18-19. 821 Application at 3. In a couple of markets where it offers integrated bundles, AT&T also bundles its “Mobile Share Value” mobile broadband plans with its wireline broadband and video services. Lee Decl. ¶ 17. Federal Communications Commission FCC 15-94 109 broadband capabilities of its own, so it does not offer an integrated bundle of DIRECTV video and broadband.822 To compensate for their inability to offer integrated bundles in a substantial number of locations, the Applicants have contracted to offer a synthetic bundle of AT&T’s broadband service and DIRECTV’s video service.823 280. Positions of the Parties. In support of their position, the Applicants argue that consumers prefer to purchase integrated bundles of video and broadband.824 They note that, among the basic cable subscribers served by six of the nation’s largest cable operators, 78 percent purchase a bundle of services825 and that “more than 97 percent of AT&T’s 5.7 million video customers subscribe to bundled services.”826 The Applicants disclose that “approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent of the subscribers leaving DIRECTV reported that they will purchase a bundle of video and broadband services from their new provider, which they assert is “ a marked increase” from the level reported three years earlier.827 AT&T also states that “[f]ocusing on bundles is also an efficient way for us to do business. Bundles allow U-verse to deliver [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] video services by recovering the content-acquisition costs from a larger revenue base.”828 In particular, the “ [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .”829 281. As explained in more detail elsewhere,830 the Applicants claim that their synthetic bundles are not competitive “because they cannot match either the discounts on price or the seamless customer service offered by cable companies.”831 In addition to satisfying consumer preferences and improving the overall customer experience, the Applicants state that another benefit of the post- transaction integrated bundles is that customers “will be able to purchase these bundled products in more places . . . [because] AT&T will use DIRECTV’s retail channels to offer these new bundled products” in addition to its own stores and authorized dealers and agents.832 Furthermore, more integrated bundles 822 Application at 1, 13-14, 62-63; Doyle Decl. ¶ 5; Guyardo Decl. ¶¶ 10-11. 823 The Applicants have also created synthetic bundles with other companies. See, e.g., Application at 14, 23-26; Guyardo Decl. ¶ 19; Katz Decl. ¶ 65; Lee Decl. ¶¶ 50-52; Moore Decl. ¶ 27; Video and Broadband Complementarity White Paper at 14; Double Moral Hazard White Paper at 22-26. 824 See, e.g., Application at 55; Doyle Decl. ¶ 4; Katz Decl. ¶ 28; Moore Decl. ¶¶ 27-29; Bundles Ex Parte Presentation at 4-6, 7-9. The Applicants also note that “[i]ncreasingly, the selection of a broadband provider also determines consumers’ choices of video services.” Application at 18; Doyle Decl. ¶ 17. The Applicants argue that “although consumers value broadband as their highest priority, ‘a compelling [v]ideo product is needed to preserve and expand [a company’s] broadband sales.’” Video and Broadband Complementarity White Paper at 10. 825 Application at 21; Doyle Decl. ¶ 16. In addition, two recent studies commissioned by AT&T found that consumers prefer to purchase video and broadband services in a single, integrated bundle. One study, conducted by Professor Ravi Dhar, found that “all other things being equal, over 80 percent of respondents preferred an integrated bundle to a synthetic one.” Bundles Ex Parte Presentation at 2. A second study found that consumers of video and broadband “strongly prefer more integrated offers.” Id. at 9. 826 Application at 2. See also Application at 70; Lee Decl. ¶ 12; Competition White Paper at 6. 827 Guyardo Decl. ¶ 11. See also Application at 63; Katz Decl. ¶ 31. 828 Lee Decl. ¶ 15. 829 Lee Decl. ¶ 15. 830 See supra ¶¶ 112-115. 831 Lee Decl. ¶¶ 4, 53-58. See also Application at 20, 52; Doyle Decl. ¶¶ 24-25; Guyardo Decl. ¶¶ 7, 21, 41-45; Katz Decl. ¶¶ 26-27, 29-32, 68-71, 97-106; Stankey Decl. ¶¶ 26-29; Double Moral Hazard White Paper at 17-21. The Applicants also note that competition “for video/broadband bundles occurs primarily between the competitors offering integrated bundles.” Application at 56-62. 832 Stankey Decl. ¶ 31. See also Application at 52; Katz Decl. ¶ 63; Moore Decl. ¶ 29. Federal Communications Commission FCC 15-94 110 may be available because AT&T expects to offer integrated bundles of DIRECTV video and AT&T wireless service after the transaction is consummated.833 282. Discussion. As discussed in detail in our competitive effects analysis, we find that the Applicants’ ability to offer an integrated bundle is an important public interest benefit of the transaction, and we accord it significant weight in our balancing of the public interest harms and benefits. C. Reduced Payments for Programming and Bundling Efficiencies 1. Reduced Payments for Content Acquisition 283. Positions of the Parties. According to the Applicants, video programming payments are largely a function of scale, and, typically, larger MVPDs are able to obtain lower per subscriber rates from video content providers.834 The Applicants claim that AT&T’s programming payments account for a substantial portion of U-verse video recurring costs and consume a substantial portion of video subscriber revenues.835 The Applicants also claim that AT&T pays significantly more for programming content than video distributors with larger subscriber bases.836 The Applicants contend that AT&T’s per subscriber payments for programming have been growing at a rate of [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent per year from 2011 to 2014 and project that these payments would increase by [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent over 2013 levels by 2023.837 The Applicants argue that, because programming payments are closely tied to scale, AT&T needs to expand its customer base significantly to reduce these payments.838 284. The Applicants estimate that the transaction enables AT&T to lower its programming payments by 20 to [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent within [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] years by [BEGIN HIGHLY 833 Application at 33; Joint Opposition at 11; Katz Decl. ¶ 107; Stankey Decl. ¶ 30. 834 Application at 21; Moore Decl. ¶ 14; Stankey Decl. ¶¶ 6, 23-24; Lee Decl. ¶ 19; Katz Decl. ¶ 114. The Applicants note that Time Warner Cable claimed that its programming costs fell after its merger with Insight Communications. See Application at 21-22 n.50. Dr. Katz notes that some theoretical literature identifies circumstances under which large video service providers may have less bargaining power and pay higher content fees. However, industry participants and analyst reports support the conclusion that large video distributors generally pay less for content. See Katz Decl. ¶ 114 & nn.200-201. 835 Application at 22; Moore Decl. ¶ 6; Stankey Decl. ¶ 15; Lee Decl. ¶ 18; Katz Decl. ¶ 110 & n.194; Katz Reply Decl. ¶ 18 n.35. The Applicants claim that in 2014 AT&T’s programming payments accounted for approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent of its video subscriber revenues and [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent of U-verse video variable recurring costs. Lee Decl. ¶ 18. 836 See Application at 22; Stankey Decl. ¶ 15; Lee Decl. ¶ 20. The Applicants estimate that AT&T’s payment per video subscriber is more than [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent and [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent higher than Comcast’s and Time Warner Cable’s, respectively. Lee Decl. ¶ 20. Dr. Katz provides programming payments per subscriber per month for several different MVPDs. He estimates that Comcast’s average programming payments per subscriber per month in 2013 were about [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent lower than AT&T’s. Katz Decl. ¶ 114, Figure 1. Comparing the seven non-premium channel license agreements that make up the largest share of AT&T’s programming payments, the Applicants claim that DIRECTV’s per subscriber rates are approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent lower than AT&T’s rates for those seven agreements. See Katz Decl. ¶ 115 n.206. 837 Lee Decl. ¶ 18. 838 Application at 25; Joint Opposition at 17-18. According to the Applicants, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] due to high and rising programming payments. See Application at 24; Stankey Decl. ¶ 42; Lee Decl. ¶ 15. Federal Communications Commission FCC 15-94 111 CONF. INFO.] [END HIGHLY CONF. INFO.] .839 The Applicants claim that the reduction in programming payments would begin in [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .840 The Applicants assert that the savings would exceed [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .841 The Applicants estimate that AT&T’s programming payment per video subscriber at closing would be [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .842 The Applicants claim that these estimated programming payment reductions are conservative because they are based on [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .843 285. The Applicants argue that they would pass on a portion of the programming payment reductions to consumers, and also use those savings to improve the profitability of their video product.844 The Applicants contend that, because programming payments are marginal costs, some of the savings would be passed through to consumers in the form of lower prices.845 The Applicants also note that the BH Simulation projects that more than [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent of programming payment reductions would be passed on to consumers.846 Further, the Applicants claim that the programming payment reductions and other cost savings would allow them to deploy FTTP to an additional 2 million customer locations than could be justified financially absent the transaction.847 286. Some commenters agree with the Applicants that a larger subscriber base would allow AT&T to lower its programming payments.848 Commenters argue, however, that lower programming payments should be considered a benefit of the transaction only if the payment reductions are passed on 839 Application at 36; Katz Decl. ¶ 115; Joint Opposition at 16; Katz Reply Decl. ¶ 32. AT&T Response to Sept. 9, 2014, Information Request at 244. Dr. Katz estimates that DIRECTV’s average programming payments per subscriber per month in 2013 were approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent less than AT&T’s. Katz Decl. ¶ 115. 840 Application at 36. 841 Id.; Moore Decl. ¶ 16; Stankey Decl. ¶ 22; Joint Opposition at 16; AT&T Response to Sept. 9, 2014, Information Request at 226, 231. AT&T estimates that the per subscriber savings would be over [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . These savings are estimated by [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See Moore Decl. ¶¶ 16-17; AT&T Response to Sept. 9, 2014, Information Request at 231, 244. 842 Moore Decl. ¶ 18; AT&T Response to Sept. 9, 2014, Information Request at 231, 243. AT&T expects that [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See AT&T Response to Sept. 9, 2014, Information Request at 243-244. 843 Application at 36-37; Katz Decl. ¶ 115; Moore Decl. ¶ 15; Stankey Decl. ¶¶ 23-24; Joint Opposition at 17; AT&T Response to Sept. 9, 2014, Information Request at 231-232, 241-242, 244-245. Programming payment reductions are based on only the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See Moore Decl. ¶¶ 15, 17; AT&T Response to Sept. 9, 2014, Information Request at 241. 844 Content Cost White Paper at 8; Joint Opposition at 18-19; Katz Reply Decl. ¶¶ 20-26. 845 Katz Reply Decl. ¶¶ 21-23; Content Cost White Paper at 9-10. 846 Content Cost White Paper at 8, 11-14; Katz Reply Decl. ¶ 24. The Applicants claim that the Katz and Berry and Haile merger simulations show that there is no consumer harm to offset and that the pass-through of reduced programming payments is over and above the benefits that would accrue because of the complementarities in the products supplied by the combined firms. See Content Cost White Paper at 10-11. For further discussion of the Berry and Haile merger simulation and pass-through of content costs, see Appendix C ¶¶ 87-88, 94. 847 Application at 41; Stankey Decl. ¶¶ 38-46; Joint Opposition at 22-23; AT&T Response to Sept. 9, 2014, Information Request at 217-218, 221, 232-234. See also Section XI.F. 848 ACA Comments at 17-19; Biglaiser Statement at 14-15; CWA Comments at 6-7; Cox Petition at 16; DISH Petition at 11-12; ACA Reply at 10; COMPTEL Reply at 6; see also Wave Reply at 2. Federal Communications Commission FCC 15-94 112 to consumers and that AT&T has not sufficiently proven such pass-through.849 In addition, Free Press argues that recent research suggests that rising payments for sports programming, broadcast retransmission consent, and digital rights may erode the benefit achieved in reducing programming payments through scale.850 287. Discussion. We find that AT&T’s programming payments may be reduced as a result of the proposed transaction, but we decline to attribute this possibility as a benefit unless the savings accrue to consumers. Our analysis of the record indicates that AT&T currently pays approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] on a per subscriber basis. However, our analysis also finds that [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] is due to differences in the number of channels offered by the two providers and the relative distribution of subscribers over those channels, i.e., channel tiering.851 288. The Applicants’ calculations of potential programming payment reductions use an average per subscriber per month programming payment for both AT&T and DIRECTV. The Applicants calculated the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .852 The Applicants’ estimated monthly per subscriber programming payment reductions range from [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .853 The Applicants estimate that the reductions in programming payments would be phased in over several years, starting at [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .854 The record does not provide details on how the phase-in percentages were calculated but notes that they are based on [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .855 Additionally, while AT&T has indicated that it is interested in providing content on mobile devices, it does not appear that the Applicants have accounted for the provision of such new services in calculating AT&T’s estimated programming payments post- transaction.856 289. Using data submitted by the Applicants, we calculated programming payments per subscriber per month for AT&T and DIRECTV in 2014.857 As noted elsewhere, based on our analysis of 849 WGAW Petition at 27; WGAW Reply at 12-13; Free Press Petition at 26-28; ACM et al. Petition at 12-13; DISH Petition at 12; Franken Comments at 6. 850 Free Press Petition at 24-25. 851 For a discussion of these differences, see supra ¶¶ 99-102; Appendix C Section IV.C. 852 AT&T Response to Sept. 9, 2014, Information Request, Exhibit 69.c.2, tab “Content Cost (output)”; ATT-FCC- 01645622, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 853 AT&T Response to Sept. 9, 2014, Information Request, Exhibit 69.c.2, tab “Content Cost (output)”; Content Cost White Paper at 12-13; see also ATT-FCC-01645622, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 854 AT&T Response to Sept. 9, 2014, Information Request, Exhibit 69.c.2, tab “Content Cost (output)”; ATT-FCC- 01645622, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ; AT&T Response to Sept. 9, 2014, Information Request, Exhibit 68.e.1 at 52, Exhibit 69.c.1. This analysis excludes programming payments for NFL Sunday Ticket. 855 See AT&T Response to Sept. 9, 2014, Information Request, Exhibit 68.e.1 at 52, Exhibit 69.c.1. Further, the documents estimate that the annual increase in programming payments [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See AT&T Response to Sept. 9, 2014, Information Request, Exhibit 68.e.1 at 52, Exhibit 69.c.1; id., Exhibit 69.c.2, tab “Content Cost (output)”; ATT-FCC-01645622, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 856 See Application at 33; Katz Decl. ¶ 107; Stankey Decl. ¶ 30; Joint Opposition at 11; AT&T Response to Sept. 9, 2014, Information Request, Exhibit 68.e; ATT-FCC-01070922, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 857 Our analysis also compares AT&T’s and DIRECTV’s programming payments with Comcast. For a detailed discussion of our programming payment analysis, see Appendix C Section IV.C. Federal Communications Commission FCC 15-94 113 the data submitted by AT&T and DIRECTV, we estimate that the programming payment per subscriber per month for [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .858 Our analysis also indicates that to fully achieve the claimed programming payment reductions, certain adjustments such as re-tiering of channels may need to occur. Without such adjustments, the per subscriber [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .859 There is no evidence in our record that the Applicants have considered these types of adjustments in calculating the programming payment reductions. Therefore, while we find that AT&T is likely to achieve some programming payment reductions, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] which may similarly affect the amount of the reduction. 290. As noted above, in analyzing the potential competitive effects arising from the transaction, we considered the effect of programming payment reductions and the pass-through of such reductions in the merger simulation.860 We also considered programming payment reductions that are not passed through to consumers when analyzing the FTTP buildout, which the Applicants claim is a benefit of the transaction.861 We find it likely that some of the programming payment reductions will be passed through to subscribers and, as discussed below, that some portion of such reductions may help in funding FTTP expansion. 291. We credit as a benefit only the portion of any programming payment reduction that is passed on to consumers and not any programming payment reduction that is retained by the Applicants. As the Commission has found previously, to the extent a change in video programming costs of this nature is a transfer of surplus between video programmers and video distributors, it generally is not a public interest benefit.862 There is no evidence in this record that would lead us to a different conclusion. The record also does not establish that the reduction in AT&T’s programming costs would have a negative effect on consumer welfare such that we would consider such reductions to be a public interest harm of this transaction. In particular, we do not have evidence that the programming payment reductions would result in increased programming payments by other MVPDs or reduced investment in the quality or quantity of programming.863 2. Other Cost Savings and Efficiencies 292. Positions of the Parties. In addition to content cost savings, the Applicants assert that the transaction would result in other cost savings of more than [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .864 These savings result from consolidating the two companies’ installation and service operations, among other efficiencies.865 For example, if the Applicants are able to perform multiple installation services, such as DBS service, broadband, and voice, with a single truck roll, they estimate savings of approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] 858 For a detailed discussion of this analysis, see supra ¶¶ 98-102; see also Appendix C ¶ 58, Table 1. 859 See supra ¶ 102. 860 See supra ¶¶ 96-103; Appendix C Section V.B. 861 See, e.g., Application at 41; Stankey Decl. ¶¶ 38-46; Joint Opposition at 22-23. See also infra ¶¶ 332-334. 862 EchoStar-DIRECTV HDO, 17 FCC Rcd at 20637, ¶ 211. 863 See supra Section X.A.4 and Section X.D.2.a. 864 See Moore Decl. ¶ 20; AT&T Response to Sept. 9, 2014, Information Request at 226. These estimates include savings from the adoption of DIRECTV’s set-top box technology for U-verse customers, which is discussed below. See infra Section XI.E. 865 See Application at 38; Moore Decl. ¶ 24; Katz Decl. ¶¶ 104-106; AT&T Response to Sept. 9, 2014, Information Request at 238; DIRECTV Response to Sept. 9, 2014, Information Request at 73. Federal Communications Commission FCC 15-94 114 .866 They also intend to integrate AT&T’s IP distribution network and DIRECTV’s satellite network, consolidate the two companies’ broadcast centers, and achieve additional cost savings in operation of their super hub offices,867 all of which they estimate would result in cost savings of approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .868 293. The combined entity also plans to [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .869 They further claim cost savings in customer call center operations, integration of information technology (“IT”) systems, and other general administrative and headquarter functions.870 The Applicants estimate that the elimination of duplicative systems and operations would reduce expenses by [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .871 Moreover, the Applicants explain that the transaction would provide them with additional marketing and sales channels.872 Specifically, AT&T would be able to market AT&T Mobility products to existing DIRECTV subscribers and also to use DIRECTV’s retail distribution network to market those services.873 Similarly, DIRECTV would be able to utilize AT&T retail distribution channels for DIRECTV video products.874 294. Discussion. While the transaction may result in certain cost savings and efficiencies alleged by the Applicants, we ascribe minimal weight to these benefits. As discussed above, because much of the information relating to the potential benefits of a transaction is in the sole possession of the Applicants, they are required to provide sufficient evidence supporting each claimed benefit to enable the Commission to verify its likelihood and magnitude.875 The Applicants have failed to do so for the efficiencies listed above.876 For example, the Applicants have failed to distinguish claimed cost savings that would result in a reduction in marginal cost from cost savings that would result in a reduction in fixed 866 See AT&T Response to Sept. 9, 2014, Information Request at 238; see also Application at 38; Moore Decl. ¶ 24; Katz Decl. ¶ 104. 867 The Applicants explain that “‘[s]uper hub offices’ are where video programming is gathered and redistributed to network facilities for delivery to subscribers.” Application at 38 n.114. 868 See id. at 38; Moore Decl. ¶ 25; AT&T Response to Sept. 9, 2014, Information Request at 238; DIRECTV Response to Sept. 9, 2014, Information Request at 73. 869 See Application at 37-38; Moore Decl. ¶ 23; AT&T Response to Sept. 9, 2014, Information Request at 238-239; DIRECTV Response to Sept. 9, 2014, Information Request at 72-73. 870 See Application at 38; Moore Decl. ¶ 25; AT&T Response to Sept. 9, 2014, Information Request at 237-238; DIRECTV Response to Sept. 9, 2014, Information Request at 73. 871 See AT&T Response to Sept. 9, 2014, Information Request at 238. 872 See Application at 39; Stankey Decl. ¶ 31; Moore Decl. ¶ 29; Katz Decl. ¶ 108; AT&T Response to Sept. 9, 2014, Information Request at 228-229. 873 See id. 874 See id. 875 See supra ¶ 274. 876 As discussed above, the Commission has stated previously that it will discount or dismiss speculative benefits that it cannot verify. See supra ¶ 274 & n.809; see also Sirius-XM Order, 23 FCC Rcd at 12383, ¶ 75; Liberty Media-DIRECTV Order, 23 FCC Rcd at 3331, ¶ 140; News Corp.-Hughes Order, 19 FCC Rcd at 611, ¶ 317; EchoStar-DIRECTV HDO, 17 FCC Rcd at 20630, ¶ 190. Here, AT&T’s calculations for many of the purported savings note a number of “key considerations” or “key assumptions” that call into question the extent to which certain purported savings or synergies would be achieved. See AT&T Response to Sept. 9, 2014, Information Request, Exhibit 68.e.1, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] at 57, 61-62. Federal Communications Commission FCC 15-94 115 cost.877 As noted above, we generally find reductions in fixed cost to be less cognizable than reductions in marginal costs because the former are less likely to result in lower prices for consumers.878 3. Innovation in Video Services 295. The Applicants claim that the transaction would enable them to combine their research and development efforts and to spread the cost of those efforts over a more extensive customer base.879 Moreover, the Applicants state that the transaction would combine DIRECTV’s expertise in acquiring content and assembling programming packages, as well as its video engineering talent and resources, with AT&T Labs’ innovative technology leadership.880 The Applicants believe that this combination would equip the combined entity with the tools to respond to changing consumer demand and to develop new services and product features.881 a. Traditional Video 296. Positions of the Parties. The Applicants state that the proposed transaction would utilize the complementary nature of their separate video services to create a higher quality video service than either AT&T or DIRECTV could offer independently.882 The Applicants note that DIRECTV today must rely on third parties for the delivery of most of its VOD content.883 The Applicants claim that, after the transaction, “DIRECTV will be able to use AT&T’s Internet backbone and broadband infrastructure to provide higher-quality service at reduced cost, through measures such as more efficient use of caching to store content closer to the customer.”884 297. Similarly, the Applicants state that DIRECTV has been developing Ultra HD (“UHD”) video with its focus on new compression technologies that allow for the highest video quality with the least amount of bandwidth.885 They contend that the proposed transaction would enable AT&T to take advantage of DIRECTV’s expertise and to deploy UHD faster than would otherwise have been possible.886 877 The Sept. 9, 2014, Information Request to AT&T specifically asked AT&T to state, for each purported cost savings, whether it is a fixed cost savings or a variable cost savings and to explain the reasoning. See Sept. 9, 2014, Information Request to AT&T at Question 68.d. AT&T responded that “[d]etail regarding a break-down of estimated cost savings between fixed and variable cost is not available at this time beyond what is set forth in [Exhibit 68.d].” See AT&T Response to Sept. 9, 2014, Information Request at 240. Exhibit 68.d, however, does not break down cost savings between fixed costs and variable costs. 878 See supra ¶ 275 & n.813. 879 Application at 46; Stankey Decl. ¶¶ 6, 20 (“The combined company’s increased scale will provide a broader customer base across which to spread the fixed costs associated with developing [new] services and features. As a result of these and other synergies, the combined company will offer consumers greater value, better and more flexible programming packages, and enhanced interactivity.”). 880 Stankey Decl. ¶¶ 6, 32 (“[W]ith technology leadership from both DIRECTV and AT&T Labs, the combined company will be well-situated to devote the resources necessary to innovate in additional ways that cannot even be foreseen today.”). 881 See id. 882 Katz Decl. ¶ 122 (“By allowing complementarities between the parties’ video services to be realized, the proposed transaction will create a higher-quality video offering than either firm could provide alone.”). 883 Stankey Decl. ¶ 21. 884 Id.; Application at 31. 885 Stankey Decl. ¶ 32 (“[W]ith its focus on new compression technologies that yield the highest quality video with the least bandwidth, DIRECTV is leading the development of ‘ultra high definition’ (‘UHD’) television.”). 886 Id. Federal Communications Commission FCC 15-94 116 298. The Applicants also state that the proposed transaction would enable DIRECTV to extend its programming lineup to AT&T U-verse video customers.887 AT&T would also use DIRECTV’s experience to build on the lineup and to design a broader range of best-in-class video programming packages tailored to a variety of tastes and price levels.888 299. The Applicants also argue that the proposed transaction would allow the combined entity to develop more sophisticated interactive services and a more advanced user interface than either company would alone.889 They state that a migration to a single video platform would create a common development and operating environment as well as a more uniform consumer experience.890 The Applicants believe that a combined entity would also be better positioned to bear the fixed costs associated with the development of interactive user interfaces and other capabilities.891 300. Several commenters assert that the Applicants could have achieved the purported benefit of improved video services absent the transaction through internal organic growth.892 Free Press argues that the resources AT&T has devoted to acquiring DIRECTV could have been used for internal improvement.893 ACM et al. argue that allowing AT&T to acquire DIRECTV eliminates AT&T’s incentives to expand and enhance its U-verse footprint.894 301. Discussion. We ascribe minimal weight to the purported benefits of the transaction for traditional video services. The Commission has stated in previous transaction review proceedings that the deployment of improved video services is a recognized public interest benefit.895 We find that the proposed transaction will enable the Applicants to combine their resources and share their expertise to improve their MVPD offerings to the benefit of consumers.896 We also find that DIRECTV’s programming lineup will likely be extended to AT&T’s video service897 and may improve AT&T’s U- verse video programming. The question before us is whether such benefits could have been achieved absent the transaction through a means less harmful to competition.898 We note that the extent to which 887 Application at 30; Stankey Decl. ¶ 19. 888 Application at 30; Katz Decl. ¶ 122 (“[DIRECTV has] greater experience in providing content packaging services, as AT&T was a relatively recent entrant into packaging and has small scale. AT&T will benefit from DIRECTV’s superior software and, more generally, greater experience in providing content packaging services.”). 889 Katz Decl. ¶ 123. 890 Id. (“Migration to a single TV platform also will allow the achievement of a common development and operating environment, and uniform customer experience.”). 891 Id. (“Because there are fixed costs associated with the development of interactive capabilities and the user interface, the combined company will be better positioned to develop more sophisticated interactive services and a more advanced user interface than would either company alone.”). 892 See ACM et al. Petition at 7-9; see also WGAW Petition at 17-19; Free Press Petition at 37-40; ACM Feb. 24, 2015, Attachment at 1-2. 893 See Free Press Petition at 37-40. 894 See ACM et al. Petition at 8-10; ACM Feb. 24, 2015, Ex Parte Letter, Attachment at 1-2. 895 See Adelphia Order, 21 FCC Rcd at 8312, ¶ 256; Comcast-AT&T Order, 17 FCC Rcd at 23316-17, ¶¶ 182-185; Applications for Consent to the Transfer of Control of Licenses and Section 214 Authorizations from MediaOne Group, Inc., Transferor, to AT&T Corp., Transferee, CS Docket No. 99-251, Memorandum Opinion and Order, 15 FCC Rcd 9816, 9886, ¶ 160 (2000); News Corp.-Hughes Order, 19 FCC Rcd at 614-615, ¶ 327. 896 See DTVFCC-03667643, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 897 See DTVFCC-03667643, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 898 Liberty Media-DIRECTV Order, 23 FCC Rcd at 3330-31, ¶ 140; News Corp.-Hughes Order, 19 FCC Rcd at 610, ¶ 317; EchoStar-DIRECTV HDO, 17 FCC Rcd at 20630, ¶ 189; Applications of NYNEX Corp., Transferor, and Bell Atlantic Corp., Transferee, for Consent to Transfer Control of NYNEX Corp. and Its Subsidiaries, Memorandum (continued….) Federal Communications Commission FCC 15-94 117 such benefits could have resulted from organic growth is limited given AT&T’s competitive position relative to the larger cable MVPDs and its relatively limited MVPD assets.899 DIRECTV also lacks the infrastructure for broadband delivery of its VOD content and is unlikely to have developed broadband on its own without partnering with a broadband provider.900 302. While we recognize that the proposed transaction could enable innovation in traditional video services greater than what either Applicant could have achieved independently, we find any benefit difficult to verify or quantify.901 Furthermore, we question the degree to which these innovations would be driven by the proposed transaction rather than by market forces.902 For example, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .903 The record indicates that industry conditions were already pushing Applicants to innovate and improve their video services. Therefore, although we accept, in theory, that benefits might result from the Applicants combining their video innovation efforts, we are unable to determine the extent to which their combined efforts would have outpaced the market- influenced efforts of each Applicant separately. Accordingly, we ascribe minimal weight to the purported benefit.904 b. OVD 303. Positions of the Parties. The Applicants state that the proposed transaction creates potential for innovation to result in enhanced video options across all screens, including an OVD offering.905 The Applicants claim that the improved cost structure and large video subscriber base would justify more risky investments in the infrastructure necessary to create an OVD product.906 The Applicants also state that the proposed transaction would allow AT&T to obtain more attractive terms for digital content rights because the increase in scale makes AT&T more attractive as a partner for online video providers.907 The Applicants also claim that the proposed transaction would allow the combined (Continued from previous page) Opinion and Order, 12 FCC Rcd 19985, 20063-64, ¶ 158 (1997); Comcast-AT&T Order, 17 FCC Rcd at 23313, ¶ 173. 899 See ATT-FCC-02898059, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 900 See DTVFCC-00254236, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ; DTVFCC- 03677960, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 901 See Comcast-NBCU Order, 26 FCC Rcd at 4333, ¶ 231 (stating that the benefit of cooperation between applicants was “difficult to quantify aside from specific commitments and contexts”). 902 See ATT-FCC-02898059, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 903 See ATT-FCC-01816620, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 904 See Comcast-NBCU Order, 26 FCC Rcd at 4333, ¶ 231; News Corp.-Hughes Order, 19 FCC Rcd at 619, ¶ 344. 905 Application at 46-49; Stankey Decl. ¶¶ 9, 56-64. 906 Stankey Decl. ¶ 58 (“The improved cost structure and much larger video subscriber base enabled by this transaction will allow us to justify the more risky investments in software, platforms and service development necessary to create a world-class OTT customer experience.”). But see Free Press Petition at 35-37 (claiming that the proposed transaction would decrease the incentives of the Applicants to invest in OVD services). Free Press also argues that absent the transaction both AT&T and DIRECTV would have entered the online video market as competitors and that the ensuing competition would have strengthened the viability of online video services in the video distribution market. Free Press Petition at 35-36. We find that such concerns have been properly addressed by our discussion of potential harm to competition in the MVPD market. See supra Section VIII and Section IX.A.5. 907 Stankey Decl. ¶ 58 (“At the same time, the increase in video scale will make AT&T a much more attractive OTT partner for content providers and thus allow AT&T to obtain more attractive terms for the new types of digital content rights necessary to provide innovative OTT offerings.”); see also Stankey Decl. ¶ 61 (“AT&T projects that its annual spending with content providers will increase from approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] today to nearly [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. (continued….) Federal Communications Commission FCC 15-94 118 entity to build on DIRECTV’s longstanding relationships with content providers to negotiate flexible digital rights for new OVD services.908 304. Discussion. While we accept that the proposed transaction may allow the Applicants to improve their ability to launch OVD services by pooling their assets and efforts,909 we do not find that the transaction creates a significant, quantifiable public interest benefit, and as noted above, we are concerned that any such improvement in the Applicants’ OVD services creates an incentive to limit competition from competing OVD services. We also note that [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .910 Accordingly, we do not find that the transaction itself would necessarily or uniquely accelerate innovation in OVD services overall, and we ascribe only minimal weight to this benefit. c. Improved Advertising Capabilities 305. Positions of the Parties. AT&T states that it would be able to enhance the combined entity’s ability to reach consumers through tailored, compelling advertisements by improving DIRECTV’s advertising platform.911 AT&T states that the combined entity would be better able to customize its advertisements and [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .912 AT&T predicts that the improvements to DIRECTV’s advertising platform would bring DIRECTV’s [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] due to advertisers placing more value on DIRECTV’s advertising time.913 306. Discussion. We do not find improved advertising capabilities to be a significant benefit of the transaction. The record reflects that AT&T [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] and that DIRECTV [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .914 Given that DIRECTV does not currently have a two-way connection to all of its subscribers, however, we agree with the Applicants that the proposed transaction may enable AT&T [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .915 That said, absent specific commitments and defined plans we are unable to quantify the benefit of improved advertising capabilities to video services.916 While there may be a potential efficiency from the Applicants’ combined advertising (Continued from previous page) INFO.] post-transaction. That level of spending will substantially increase AT&T’s attractiveness to content providers and allow it to secure more innovative content rights arrangements. As a much more important distributor of content to MVPD customers, AT&T will be a more attractive partner for a broader and more innovative set of content agreements to facilitate new OTT services.”); Katz Decl. ¶ 121. 908 Application at 23, 48-49, 77; see also Stankey Decl. ¶¶ 56-64. 909 See DTVFCC-03678259, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 910 See DTVFCC-03726077, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ; DTVFCC- 03726039, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ; DTVFCC-00173673, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ; ATT-FCC-00518651, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ; ATT-FCC-00356035, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ; ATT-FCC-00934828, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 911 Moore Decl. ¶ 30. 912 Id. 913 Id. 914 See ATT-FCC-01816620, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ; DTVFCC- 02051720, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 915 See Moore Decl. at ¶ 30; DTVFCC-01320042, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . Federal Communications Commission FCC 15-94 119 development efforts, we do not find that the Applicants have established how such efficiency translates into a measurable benefit to consumers. Therefore, we ascribe minimal weight to this benefit. D. Video Programming Market 307. Positions of the Parties. The Applicants claim that the proposed transaction would benefit the video programming market in two ways. First, they argue that it would expand their own opportunities to produce original video programming, which they contend is becoming an increasingly important strategy for video service providers.917 They assert that large economies of scale are needed to cover the high fixed costs of producing original content.918 They explain that a large subscriber base, by generating greater revenues to support those fixed costs, increases the net present value of investment in original content.919 The Applicants argue that the combined entity’s large scale would enable it to launch and market original programming and to invest in new programming ventures.920 They note that DIRECTV has existing production facilities that can be used to create original content, with a few projects already underway, and that AT&T plans to acquire and develop OVD services through its joint venture with The Chernin Group.921 The Applicants argue that consumers would benefit directly from the new supply of programming and indirectly from the ensuing competitive pressures on other video providers and content creators.922 308. Second, the Applicants assert that other content producers would benefit from access to the broader subscriber reach of the combined entity. They argue that the value of the program networks they carry would increase due to the programmers’ ability to offer a larger audience to advertisers.923 In addition, the Applicants state that the combined entity’s multiple video platforms would benefit content owners by providing them with “new opportunities to gain exposure for and to monetize content, while preserving the value of the core pay TV revenue stream.”924 In addition, they argue that the broadband expansion facilitated by the transaction would increase the demand for content, thereby improving its value.925 309. Discussion. We ascribe minimal weight to the purported benefits of the transaction for the video programming market. We agree with the Applicants that to the extent the combined entity (Continued from previous page) 916 See Application of Whitehall Enterprises, Inc., Assignor, and Clear Channel Broadcasting Licenses, Inc., Assignee, for Consent to Assignment of License of WAAM(AM), Ann Arbor, MI, MB Docket No. 02-284, Hearing Designation Order, 17 FCC Rcd 17509, 17525, ¶ 49 (2002) (indicating that commitments to improve advertising services may be considered public interest benefits to the extent they support the Commission’s public policy goals). 917 Katz Decl. ¶ 120. 918 Id. 919 Id. 920 Stankey Decl. ¶ 63. 921 Application at 49; Stankey Decl. ¶ 63; see also Katz Decl. ¶ 120 n.211. 922 Katz Decl. ¶ 120. 923 Stankey Decl. ¶¶ 23-24; see also Katz Decl. ¶ 112; Katz Reply Decl. ¶ 33; see also supra Section X.A.4. But see ACA Reply at 13-14 (noting that the Applicants provide no evidence that any gains to programmers in advertising revenue would be sufficient to compensate for the revenue losses from lower programming rates); Cox Reply at 8-9 (estimating that, based on licensing and advertising revenues in 2014, basic cable networks would have to boost advertising revenues by 29 percent to make up for a 20 percent decrease in programming rates); Cox Nov. 7, 2014, Ex Parte Letter at 2 (arguing that no evidence suggests that advertising revenues could fill the “revenue hole” that would be created). 924 Stankey Decl. ¶ 23; see also Katz Decl. ¶ 112; Katz Reply Decl. ¶ 33. 925 Joint Opposition at 52-53; see also Katz Reply Decl. ¶ 33. Federal Communications Commission FCC 15-94 120 becomes a new competitor in the video programming market by producing original content for the first time, consumers potentially would benefit from a greater selection of content and a boost in competition among programmers. Such a benefit, however, is not quantified in the record. We note that the Applicants make no specific commitment about the creation of original programming. Rather, they assert generally that approval of the transaction would “enhance the combined company’s ability to develop original programming.”926 We note further that the Applicants indicate that, absent the transaction, each already has plans to create original programming. Thus, in addition to being vague, it is not clear how the benefit is transaction specific. 310. Similarly, we find that the Applicants’ generalized assertion that the proposed transaction would benefit programmers by affording them access to larger audiences across the combined entity’s multiple platforms, and thus enhancing a network’s attractiveness to advertisers, is not quantified in the record. Additionally, the Applicants’ assertion that the combined entity’s broadband expansion might increase the demand for a programmer’s content is not quantified. For these reasons, we assign little weight to the purported benefit that content producers may receive from the broader subscriber reach of the combined entity.927 E. Video Device Market 311. Positions of the Parties. The Applicants state that the transaction would improve AT&T’s and DIRECTV’s set-top box offerings.928 They state that [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .929 The Applicants claim that the combined entity would integrate and enhance DIRECTV’s set-top box technology to develop a user interface with a consistent appearance and channel lineup regardless of platform or device.930 The Applicants also state that the transaction would “allow the combined company to obtain more flexible digital rights to deliver content to additional devices.”931 The Applicants assert that the expanded broadband deployment enabled by the transaction’s efficiencies would stimulate demand for new and innovative set-top devices.932 312. TiVo argues that the set-top box improvements claimed by the Applicants only benefit the Applicants.933 TiVo states that the Commission and Congress have favored competition as the preferred way to ensure innovation and improved consumer choice.934 313. Discussion. The Commission regards a robust and competitive video device market to be a benefit to consumers.935 We recognize that the transaction would potentially enable innovation in set- top boxes to the extent that the transaction facilitates the adoption of DIRECTV’s set-top box technology by AT&T.936 However, the transaction would also potentially reduce the number of participants in the video device market. Thus, we find this benefit difficult to quantify as a transaction-specific benefit. 926 Stankey Decl. ¶ 63. 927 See also supra Section X.D. (discussing potential harms to unaffiliated video programmers). 928 Joint Opposition at 63-64; see also Stankey Decl. ¶¶ 9, 20, 23, 59; Katz Decl. ¶ 121. 929 Moore Decl. ¶ 21. The Applicants estimate a one-time cost saving of [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . Id. 930 Application at 30; see also Stankey Decl. ¶ 20; Katz Decl. ¶¶ 5, 122-123. 931 Joint Opposition at 64. 932 Id. 933 TiVo Reply at 4. 934 Id. 935 See First Navigation Device Report and Order, 13 FCC Rcd at 14776, ¶ 2. 936 See DTVFCC-03677960, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . Federal Communications Commission FCC 15-94 121 Furthermore, prior to the transaction, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .937 Thus, it appears that the transaction merely would allow [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .938 Therefore, we ascribe minimal weight to this purported benefit because, while the transaction may facilitate integration between different video device technologies, the record indicates that much of the impetus for improvement occurred prior to the transaction. 314. Similarly, we do not find the potential for innovation in video devices to be a meaningful benefit of the transaction. Aside from AT&T implementing its equipment enhancement plans with DIRECTV’s technology, the Applicants have not made any measurable commitments or estimates from which we could determine how the transaction would foster innovation in the video device market.939 AT&T [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .940 While we recognize the potential for additional innovation through the Applicants’ combined set-top box development efforts, we find the benefit to the video device market from the transaction unquantified. Accordingly, we afford minimal weight to this potential benefit in our balancing of potential public interest harms and benefits.941 F. Expanded Deployment of Fiber to the Premises 315. In this section, we assess whether the Applicants’ commitment to provide FTTP wireline broadband service to “2 million more customer locations,” within four years after the closing of the transaction,942 as a result of savings and synergies associated with the transaction reflects a likely outcome of the transaction.943 We conclude that the financial model submitted by the Applicants in support of their buildout claims suggests that while the transaction may increase the net incentive to deploy FTTP, a significant amount of potential FTTP deployment may not occur because such deployment would 937 Application at 37; Moore Decl. ¶ 21; DTVFCC-00523721, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ; DTVFCC-02825583, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 938 See Application at 37; Moore Decl. ¶ 21. 939 See id.; see also Stankey Decl. ¶¶ 9, 23, 59; Katz Decl. ¶ 121. 940 See ATT-FCC-01816620, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 941 See News Corp.-Hughes Order, 19 FCC Rcd at 620, ¶ 344; see also Section X.E. 942 Application at 5, 41, 50; Stankey Decl. ¶¶ 35, 39, 44; Joint Opposition at 2, 7, 19 n.51; AT&T Response to Nov. 14, 2014, Information Request at 1-2; Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, Attachment (“FTTP Ex Parte Presentation”) at 2 n.2 (April 21, 2015) (submitting written ex parte presentation on FTTP technology); Letter from Maureen R. Jeffreys, Counsel for AT&T, and William M. Wiltshire, Counsel for DIRECTV, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 2 (April 30, 2015) (“AT&T April 30, 2015, Ex Parte Letter”); Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 2 (June 1, 2015) (“AT&T June 1, 2015, Ex Parte Letter”); Letter from Maureen R. Jeffreys, Counsel for AT&T, and William M. Wiltshire, Counsel for DIRECTV, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 2 (June 8, 2015) (“AT&T June 8, 2015, Ex Parte Letter”); Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 1, 4-5, 7 (June 15, 2015) (“AT&T June 15, 2015, Ex Parte Letter”). See also AT&T June 25, 2014, Ex Parte Letter at 1. Customer locations are the count of living units (“LUs”), consumer and business, to which AT&T is technically capable of providing service, including occupied and unoccupied locations. See AT&T Response to Nov. 14, 2014, Information Request, Attachment at 1 n.1. Customer locations are derived from [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, Attachment (“Overview of AT&T FTTP Investment Model”) at 47 (July 28, 2014) (submitting paper entitled “Overview of AT&T FTTP Investment Model” and associated files). 943 Application at 5, 41; Katz Decl. ¶ 126 n.222; Stankey Decl. ¶¶ 35, 39, 44; Joint Opposition at 8-9, 20, 22 n.62; Katz Reply Decl. ¶¶ 34, 38; FTTP Ex Parte Presentation at 2 n.2; AT&T April 30, 2015, Ex Parte Letter at 2; AT&T June 1, 2015, Ex Parte Letter at 2; AT&T June 8, 2015, Ex Parte Letter at 2; AT&T June 15, 2015, Ex Parte Letter at 4-5, 7. Federal Communications Commission FCC 15-94 122 cannibalize DIRECTV subscribers and revenue – a transaction-specific effect the Applicants acknowledge.944 In addition, any pass-through to consumers of realized reductions in programming payments would further reduce the incentives to deploy FTTP, as the profitability associated with the video product would be lower.945 The resulting reduction in the profitability of video provided over FTTP, in conjunction with the desire to avoid cannibalizing DIRECTV subscribers and revenues, could lead to a net decrease in the planned deployment of FTTP after the transaction, particularly if the quality improvements from an integrated bundle do not sufficiently increase customer retention. To address this potential decreased incentive to build out FTTP, we impose as a condition of the transaction that the Applicants deploy FTTP to 12.5 million locations, which include the prior planned deployment as well as the 2 million locations that they committed to deploy in connection with the transaction. 316. Positions of Parties. Prior to this transaction, AT&T announced certain plans to expand its fiber network. Specifically, AT&T had approved a multiyear, multibillion dollar plan, dubbed Project Velocity IP (“Project VIP”) to expand and improve its “wireless and fixed broadband networks.”946 The Applicants state that a limiting factor in AT&T’s FTTP deployment to date has been the challenging economics of AT&T’s video service, which means that broadband must bear [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] of the burden of repaying any FTTP investment.947 According to the Applicants, this transaction fundamentally changes the economics of FTTP deployment [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .948 AT&T maintains that “the improved products enabled by the transaction will translate directly into more sales, reduced churn, and improved margins” and that such changes “enhance the business case for AT&T to expand the reach and quality of its broadband networks beyond what would be possible otherwise.”949 Also, the Applicants claim that increased broadband deployment by AT&T would make OVD services available to millions of people that currently lack sufficient connectivity to enjoy these services.950 317. Pursuant to Project VIP, AT&T stated that it planned, among other things, to deploy its U-verse service using FTTP offering speeds up to 1 Gbps (GigaPower) in up to 25 metropolitan areas, 944 Katz Decl. ¶¶ 126, 129-130 & n.222; Katz Additional Detail ¶¶ 56-65. 945 Lower video revenues from increased pass-through reduce the profitability of FTTP deployment in IPDSL, DSL, and “no-broadband” distribution areas (“DAs”) (where AT&T does not currently offer video service) because the return on adding video capability to these areas is, generally, lower. Additionally, the opportunity cost of cannibalization of DIRECTV subscribers (were FTTP to be deployed) would be higher for all DAs under higher pass-through, as pass-through would result in U-verse video becoming less profitable relative to the profitability of DIRECTV. This situation is especially true in FTTN areas where AT&T already offers a video service. In those areas, there is no benefit to adding FTTP video service because AT&T’s FTTP Investment Model (“FIM”) assumes that [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] FTTN and FTTP. See Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, Attachment, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] (“FIM Programs”) (July 28, 2014) (submitting for the record files associated with the FTTP Investment Model). A DA is a predefined geography within the AT&T wireline footprint that is considered for the purposes of investment decisions. Each DA contains approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] customer locations on average. The FTTP model covers a total of [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] DAs. See Overview of AT&T FTTP Investment Model at 4 n.2, 6 n.9. 946 Application at 10-11; Lee Decl. ¶¶ 9-10. 947 Stankey Decl. ¶¶ 16, 39, 42; FTTP Ex Parte Presentation at 1-2. 948 Stankey Decl. ¶¶ 7, 39; Katz Decl. ¶ 126; FTTP Ex Parte Presentation at 2. 949 Stankey Decl. ¶ 34. 950 Application at 5, 40-41; Stankey Decl. ¶ 9; see also Moore Decl. ¶ 8. Federal Communications Commission FCC 15-94 123 including Dallas; Raleigh-Durham, N.C.; and Winston-Salem, N.C.951 Further, AT&T states that approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] customer locations had been approved for FTTP deployment prior to GigaPower expansion decisions.952 AT&T claims that the previously approved FTTP deployment and the GigaPower FTTP expansion brings planned FTTP deployment to approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] FTTP customer locations.953 318. In connection with this transaction, the Applicants state that they would commit to expand FTTP because the transaction improves the economics of AT&T’s investment in broadband.954 Applicants assert that, due to the programming payment reductions and other synergies of the transaction, AT&T would be able to deploy FTTP to at least an additional 2 million more customer locations than could be economically justified without the transaction’s synergies.955 The Applicants explicitly acknowledge predicted cannibalization of profits that would otherwise go to DIRECTV because of the expanded subscribership to U-verse’s video service that the FTTP expansion would facilitate.956 The 951 Application at 11-12; Stankey Decl. ¶ 43; Lee Decl. ¶ 8; Joint Opposition at 21; AT&T June 1, 2015, Ex Parte Letter at 2-3; AT&T June 15, 2015, Ex Parte Letter at 2-4, 7; AT&T Response to Sept. 9, 2014, Information Request at 180-182, 217-218; see also, AT&T Eyes 100 U.S. Cities and Municipalities for its Ultra-Fast Fiber Network, AT&T NEWSROOM, April 21, 2014, available at http://about.att.com/story/att_eyes_100_u_s_cities_and_municipalities_for_its_ultra_fast_fiber_network.html (visited June 18, 2015) (“AT&T NEWSROOM, AT&T Eyes 100 U.S. Cities for Ultra-Fast Fiber Network”). GigaPower is the brand name for AT&T’s U-verse service that uses FTTP technology to deliver speeds up to 1 Gbps. See AT&T June 1, 2015, Ex Parte Letter at 2; Stankey Decl. ¶ 38 n.20; Lee Decl. ¶ 8; Katz Reply Decl. ¶ 40. The Applicants used a version of the FIM that was [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See AT&T Response to Sept. 9, 2014, Information Request at 180-182. For these metropolitan areas, the FIM projected approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . AT&T June 15, 2015, Ex Parte Letter at 2 n.3. 952 AT&T June 15, 2015, Ex Parte Letter at 2, 6-7. AT&T plans to make [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See AT&T June 1, 2015, Ex Parte Letter at 3. Of these [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See AT&T April 30, 2015, Ex Parte Letter at 2 n.4. Further, an AT&T document breaks out these [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See ATT-FCC-00372418, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ; AT&T June 15, 2015, Ex Parte Letter at 2 n.4. See also Lee Decl. ¶ 8. We also note that since AT&T conducted its analysis of the profitability of FTTP deployment, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 953 AT&T June 15, 2015, Ex Parte Letter at 2-3. See also Stankey Decl. ¶ 43; Katz Reply Decl. ¶¶ 38-39; Joint Opposition at 21; FTTP Ex Parte Presentation at 2 n.2; AT&T April 30, 2015, Ex Parte Letter at 2; AT&T June 1, 2015, Ex Parte Letter at 2; AT&T June 15, 2015, Ex Parte Letter at 2, 6-7. See also AT&T Response to Sept. 9, 2014, Information Request at 180-182, 217-218. AT&T projects that the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] of these customer locations will be GigaPower locations. See AT&T June 1, 2015, Ex Parte Letter at 2-3. 954 Application at 41, 50; Stankey Decl. ¶¶ 34-35, 38-39, 43; Katz Decl. ¶ 126 n.222; Moore Decl. ¶¶ 8, 32; Joint Opposition at 8-9, 19-20; Katz Reply ¶¶ 34, 38; Katz Additional Detail ¶ 48; AT&T Response to Sept. 9, 2014, Information Request at 181; FTTP Ex Parte Presentation at 2 n.2; AT&T April 30, 2015, Ex Parte Letter at 2; AT&T June 1, 2015, Ex Parte Letter at 2; AT&T June 8, 2015, Ex Parte Letter at 2; AT&T June 15, 2015, Ex Parte Letter at 4-5, 7. 955 Application at 5, 41; Stankey Decl. ¶¶ 35, 39, 44; Katz Decl. ¶ 126 n.222; Joint Opposition at 8-9, 20, 22 n.62; Katz Reply Decl. ¶¶ 34, 38; FTTP Ex Parte Presentation at 2 n.2; AT&T April 30, 2015, Ex Parte Letter at 2; AT&T June 1, 2015, Ex Parte Letter at 2; AT&T June 8, 2015, Ex Parte Letter at 2; AT&T June 15, 2015, Ex Parte Letter at 4-5, 7. 956 See Katz Additional Detail ¶¶ 56-65; Katz Decl. ¶¶ 126, 129-130 & n.222. Federal Communications Commission FCC 15-94 124 Applicants assert that of these 2 million additional customer locations, most have access only to AT&T’s IPDSL or legacy DSL services, or no AT&T wireline broadband service at all.957 319. Further, AT&T claims that as a result of the transaction it would be profitable to deploy FTTP to an [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .958 These customer locations [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .959 As a result of the transaction, AT&T contends that these customer locations are [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .960 Therefore as a result of the transaction, AT&T claims that it would be economically viable to deploy an [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] customer locations.961 AT&T asserts that these claims capture ( [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] 962 prior planned deployment plus [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] as a result of the transaction). 320. To support these assertions, the Applicants state that they relied on the FTTP Investment Model (“FIM”), which AT&T asserts that it uses in the ordinary course of business to evaluate the profitability of FTTP.963 The Applicants’ expert, Dr. Katz, modified the FIM to quantify the incremental incentives to invest in FTTP that would result from the transaction.964 Based on his modifications, Dr. Katz concludes that the FIM predicts that approximately [BEGIN HIGHLY CONF. INFO.] [END 957 Application at 41-42; Stankey Decl. ¶ 46; Katz Reply Decl. ¶ 38; AT&T Response to Sept. 9, 2014, Information Request at 219-220; FIM Programs. 958 Joint Opposition at 22 n.62; AT&T Response to Sept. 9, 2014, Information Request at 180-181, 217-218; FTTP Ex Parte Presentation at 2 n.2; AT&T June 1, 2015, Ex Parte Letter at 3 n.6; AT&T June 15, 2015, Ex Parte Letter at 3-4, 6-7. These [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See Joint Opposition at 22 n.62; AT&T Response to Sept. 9, 2014, Information Request at 180-181; AT&T June 1, 2015, Ex Parte Letter at 3 n.6; AT&T June 15, 2015, Ex Parte Letter at 4, 6-7. The difference between [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See AT&T June 15, 2015, Ex Parte Letter at 4 n.9. 959 Joint Opposition at 22 n.62; AT&T Response to Sept. 9, 2014, Information Request at 180-181, 217-218; FTTP Ex Parte Presentation at 2 n.2; AT&T June 1, 2015, Ex Parte Letter at 3 n.6; AT&T June 15, 2015, Ex Parte Letter at 3-4, 6-7. 960 Joint Opposition at 22 n.62; AT&T Response to Sept. 9, 2014, Information Request at 180-181. 961 Joint Opposition at 22 n.62. 962 Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14- 90, at 1 (June 18, 2015) (“AT&T June 18, 2015, Ex Parte Letter”). This figure includes the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] customer locations that have already been deployed. See supra n.952. AT&T estimates that approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] of these customer locations will receive GigaPower and, for the remaining approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] customer locations, AT&T will offer speed tiers up to either [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . AT&T June 18, 2015, Ex Parte Letter at 1-2. AT&T states that FTTP without Gigabit Passive Optical Network technology and metro network upgrades can deliver [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See AT&T June 1, 2015, Ex Parte Letter at 2 n.3. 963 See Katz Decl. ¶ 127; Joint Opposition at 20-21; Katz Reply Decl. ¶ 34; Overview of AT&T FTTP Investment Model at 4; AT&T Response to Sept. 9, 2014, Information Request at 160. [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] See Overview of AT&T FTTP Investment Model at 4; AT&T Response to Sept. 9, 2014, Information Request at 160. 964 Katz Decl. ¶¶ 128-131; Katz Additional Detail ¶¶ 49-68. Dr. Katz modified the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] customer locations that were approved pre-transaction. See FIM Programs. See also supra ¶ 317. The Baseline FIM, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] customer locations that meet the profitability hurdle ( [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ), and this is the baseline number of deployments that Dr. Katz uses to calculate the effects from the transaction of FTTP deployment. See FIM Programs. See also supra ¶¶ 317, 319. Federal Communications Commission FCC 15-94 125 HIGHLY CONF. INFO.] customer locations clear AT&T’s profitability hurdle, approximately 2 million more locations than clear the hurdle absent the transaction.965 321. Several commenters support the proposed transaction, at least in part due to the proposed FTTP deployment.966 Other commenters, however, challenge whether the Applicants’ commitment to deploy FTTP is a benefit of the transaction.967 ACA states that AT&T’s commitment to expand broadband service is a public interest benefit only if AT&T does not receive Universal Service Fund (“USF”) support to serve the locations and, further, that AT&T should not receive any future legacy support and should return certain support it already received.968 ACM et al. state that the Applicants cannot promise to pass their cost savings on to consumers while also promising to use those savings to fund the FTTP expansion,969 and that AT&T could have used the money spent to acquire DIRECTV to instead expand its fiber network.970 322. Some commenters contend that AT&T’s FTTP commitment is something that AT&T was already planning to do in response to market conditions and competitive pressures.971 WGAW argues that AT&T’s MVPD service does not need to be profitable for it to expand its broadband network and that it has no shortage of funds that it could use for broadband deployment.972 WGAW also argues that because AT&T’s broadband business is larger than its video business, this “counters the notion that 965 See Katz Additional Detail ¶ 70, Table 2. Dr. Katz runs the FIM under a number of assumptions regarding AT&T video churn reductions, AT&T-DIRECTV diversion, and the timing of programming payment reductions. Depending on the specification, the FIM predicts a net increase of [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] locations that are marked as profitable for FTTP deployment after the transaction. See Katz Decl. ¶ 133; Katz Additional Detail ¶ 70, Table 2. 966 See Free State Comments at 17-18; Comments of Microsoft, MB Docket 14-90, at 2 (filed Sept. 16, 2014) (“Microsoft Comments”); CWA Comments at 12; CWA Feb. 3, 2015, Ex Parte Letter at 1; Letter from Jim Goetz, Sequoia Capital, and Ted Schlein, Kleiner Perkins Caufield & Byers, to Chairman Tom Wheeler, Commissioner Mignon Clyburn, Commissioner Jessica Rosenworcel, Commissioner Ajit Pai, and Commissioner Michael O’Reilly, MB Docket No. 14-90, at 2 (Oct. 15, 2014); Letter from Marc Andreessen, Co-Founder and General Partner, Andreessen Horowitz, to Chairman Tom Wheeler, Commissioner Mignon Clyburn, Commissioner Jessica Rosenworcel, Commissioner Ajit Pai, and Commissioner Michael O’Reilly, MB Docket No. 14-90, at 2 (Oct. 16, 2014); Letter from Karen Kerrigan, President and CEO, SBE Council, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 2 (Oct. 29, 2014). 967 Free Press Petition at 23-24, 30. See also WGAW Petition at 5, 26; Public Knowledge-ILSR Petition at 13-14; Greenlining Petition at 8-9; WGAW Reply at 2, 25-28; Letter from John Bergmayer, Senior Staff Attorney, Public Knowledge, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 1-2 (March 31, 2015) (“Public Knowledge March 31, 2015, Ex Parte Letter”); Letter from John Bergmayer, Senior Staff Attorney, Public Knowledge, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 2 (April 24, 2015). 968 ACA Comments at 29-30. ACA proposes that the Commission conduct a two-part analysis to determine whether AT&T broadband services meet the Phase I and Phase II broadband performance standards. See id. at 30-31. 969 ACM et al. Petition at 12-13; Public Knowledge March 31, 2015, Ex Parte Letter at 2; ACM Feb. 24, 2015, Ex Parte Letter, Attachment at 4. 970 ACM et al. Petition at 13. Free Press similarly states that for far less than the cost of this transaction, AT&T could extend U-verse to the remaining 24 million homes in its footprint. Free Press Petition at 38-40. See also WGAW Petition at 15-16; ACM Feb. 24, 2015, Ex Parte Letter, Attachment at 4. 971 Free Press Petition at 30-31; Public Knowledge-ILSR Petition at 14; WGAW Reply at 2, 26-28. Free Press notes that in AT&T’s Second Quarter 2014 SEC filing AT&T expects its peak investment for Project VIP to occur in 2014. See Free Press Petition at 30 n.63. 972 WGAW Reply at 30-32. Federal Communications Commission FCC 15-94 126 video revenues subsidize broadband for AT&T.”973 Further, some commenters claim that AT&T is planning only an incremental FTTP upgrade to AT&T’s existing FTTN network.974 323. ACM et al. also assert the proposed transaction actually reduces the broadband investment incentives for the Applicants. Specifically, these commenters state that but for the transaction, competition with cable would have driven AT&T to expand and aggressively market its U-verse service.975 They further state that AT&T would have less incentive to expand broadband capacity after acquiring DIRECTV.976 324. Some commenters claim that AT&T’s past claims and commitments in previous transactions should make the Commission skeptical of the Applicants’ claims of increased FTTP deployment.977 Some commenters argue that the Commission should impose a condition that requires the Applicants to abide by their FTTP commitment because AT&T has stated that it would stop investing in fiber broadband because of potential Internet regulation.978 WGAW proposes that the Commission adopt an enforceable condition and require FTTP deployment within three years of the closing date of the transaction.979 Public Knowledge argues that the Commission should condition the transaction on AT&T building out the planned pre-transaction FTTP deployment and the additional 2 million locations within four years in order to ensure that AT&T builds the additional 2 million locations.980 Public Knowledge- ILSR argue that if future broadband deployment is a public interest benefit, then it must be paired with a permanent commitment to upgrade, maintain, and invest in the network.981 325. The Applicants respond that the expansion to an additional 2 million customer locations is over and above what would be economically justified without the transaction and that claims by commenters are not based on any reasonable analysis.982 Further, the Applicants contend that capital investments may vary from year to year due to many factors and that incremental capital investment to expand FTTP deployment would begin in [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .983 The Applicants also contend that AT&T’s current upgrade strategy focuses [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] that AT&T [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] and therefore that the appropriate comparison in evaluating FTTP deployment as a benefit is the number of additional locations with and without the transaction.984 The Applicants acknowledge that, although an AT&T-DIRECTV bundle would free up more of the U- 973 Id. at 32. 974 See id. at 26; Free Press Petition at 30. Free Press also claims that AT&T’s marginal capital outlay on FTTP would be relatively minor compared to its routine investment. See Free Press Petition at 30. 975 ACM et al. Petition at 7-8; ACM Feb. 24, 2015, Ex Parte Letter, Attachment at 4. ACM et al. state that AT&T’s acquisition of DIRECTV would enable AT&T to acquire a video market share and MVPD market capacity from DIRECTV rather than gaining it through investing in and expanding its own U-verse video/broadband service throughout its ILEC footprint. ACM et al. Petition at 9. 976 ACM et al. Petition at 9-10. 977 Free Press Petition at 28-29; Public Knowledge-ILSR Petition at 14-16; Greenlining Petition at 8-9. 978 Public Knowledge Reply at 3-4; WGAW Reply at 34-35. See also infra ¶ 343. 979 WGAW Reply at 34. 980 Public Knowledge Reply at 2-4. 981 Public Knowledge-ILSR Petition at 15-16. 982 Joint Opposition at 23; Katz Reply Decl. ¶ 35. 983 Katz Reply Decl. ¶ 36. 984 Katz Decl. ¶ 126 n.219, ¶128 n.224; Katz Reply Decl. ¶ 39. The Applicants state that they are [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See Katz Decl. ¶ 128 n.224. Federal Communications Commission FCC 15-94 127 verse line capacity for broadband service, the amount of freed up capacity is not significant and would not reduce the combined entity’s incentive to deploy FTTP in favor of a slower technology.985 The Applicants also claim that the majority of incremental locations would not be upgrades from FTTN to FTTP and that even locations being upgraded from FTTN to FTTP would see considerable performance improvements.986 The Applicants assert that [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .987 Finally, the Applicants claim that they met past transaction commitments and would meet the commitments they have made in this transaction.988 326. Discussion. Our analysis of the FIM indicates that the Applicants’ claim that the transaction increases the profitability of FTTP buildout to an additional 2 million customer locations does not account for the decrease in AT&T’s incentive to deploy FTTP due to the pass-through of programming payment reductions to consumers, for example, as accounted for in the merger simulation supplied by the Applicants.989 In addition, we find that, relative to the baseline of approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] locations that the Applicants stated cleared the hurdle absent the transaction, there is a greater incentive to reduce FTTP deployment to avoid cannibalizing DIRECTV subscribers and revenues than the Applicants suggest. This incentive may not be completely offset by factors that may incentivize the combined entity to increase FTTP deployment.990 To address this potential transaction-specific harm, we impose a condition that requires the Applicants to deploy a total of 12.5 million customer locations. This increased FTTP deployment also improves the ability of alternative video distribution methods to replace the loss of a horizontal MVPD competitor. 1. Analysis of the FIM 327. The Applicants state that AT&T uses the FIM in its ordinary course of business to evaluate the profitability of expanding FTTP wireline services under various financial and industry assumptions.991 The Applicants also claim that AT&T considers other factors when deciding to deploy FTTP.992 These include [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .993 985 FTTP Ex Parte Presentation at 2-3. The Applicants estimate that, at best, offloading video to satellite could allow a [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent increase in the top speeds of FTTN broadband services, which would not move the speed much closer to the higher speeds offered by rival broadband providers. See id. at 3. 986 Katz Reply Decl. ¶¶ 38-40; AT&T Response to Sept. 9, 2014, Information Request at 219-220. See also Application at 41-42; Stankey Decl. ¶ 46. 987 AT&T Response to Sept. 9, 2014, Information Request at 220-221. AT&T states that it will not use Connect America Fund subsidies to deploy FTTP to the 2 million additional customer locations. AT&T June 1, 2015, Ex Parte Letter at 3-4; AT&T June 8, 2015, Ex Parte Letter at 2. Further, AT&T as part of its broadband deployment commitment in this transaction proposes in its deployment status reports to verify that locations built to fulfill the commitment are not funded with Connect America Fund support. See AT&T June 1, 2015, Ex Parte Letter at 3-4; AT&T June 8, 2015, Ex Parte Letter at 2. 988 Joint Opposition at 23 n.67; AT&T Response to Sept. 9, 2014, Information Request at 222-223. 989 See supra ¶ 103. 990 We agree with the Applicants’ claims that the majority of locations would not be upgrades from FTTN to FTTP. See FIM Programs. 991 See Katz Decl. ¶ 127; Overview of AT&T FTTP Investment Model at 4; Katz Additional Detail ¶ 47; Joint Opposition at 21. See also AT&T Response to Sept. 9, 2014, Information Request at 160-182. The FIM is performed separately for each DA, with the results subsequently aggregated to include AT&T’s full footprint. See Overview of AT&T FTTP Investment Model at 4-5; Katz Decl. ¶ 127. The FIM is run in two stages. In the first stage, the FIM relies on a number of assumptions about prices, expenses, capital costs, penetration rates, churn, taxes, and depreciation to calculate iteratively a number of financial metrics for each DA under both the current delivery technology and the alternative FTTP technology. Next, the FIM determines which of the DAs exceed AT&T’s internal rate of return (“IRR”). In the second stage, the FIM rolls up the DA-level analysis to the wire- (continued….) Federal Communications Commission FCC 15-94 128 a. Modification of FIM 328. Dr. Katz modified the FIM to account for the potential increase in the profitability of deploying FTTP after the transaction and focused on three primary factors: (1) improvements in the quality of the AT&T video product; (2) programming payment reductions; and (3) internalization of lost profits from those customers leaving DIRECTV and subscribing to AT&T’s FTTP video service (effects of cannibalization).994 Below, we evaluate each of Dr. Katz’s modifications and describe our further adjustments. Reduced Churn from Quality Improvements. 329. The Applicants argue that the transaction would improve the quality of AT&T’s video product through enhancements in video content and improvements to the video interface.995 Specifically, the Applicants claim that AT&T would benefit from DIRECTV’s video engineering expertise and resources as well as DIRECTV’s expertise in acquiring and assembling programming packages.996 The improvements in video quality potentially could make AT&T’s services (both video and broadband) more attractive and consequently increase the profitability of deploying FTTP.997 Dr. Katz incorporates these effects in the FIM by reducing the churn rate of AT&T’s bundled U-verse video product.998 Further, Dr. Katz contends that the transaction results in potential benefits of bundling FTTP and FTTN broadband with an improved video product, and therefore it reduces the churn rates for the underlying FTTP and FTTN broadband services (the technologies over which AT&T distributes its video service).999 330. With respect to video churn, Dr. Katz employs two separate estimates of video-churn reductions in the post-transaction FIM analysis: a 90 basis point reduction that would reduce AT&T’s churn rate to [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] and a more conservative reduction of 50 basis points.1000 Although the Applicants generally claim that the transaction (Continued from previous page) center level to account for the additional capital costs of upgrading a wire center to FTTP capability, and then the FIM determines whether the wire center exceeds the wire center IRR. For a DA to be considered a profitable investment, it must clear both IRR hurdles. See Katz Decl. ¶ 127; Overview of AT&T FTTP Investment Model at 4-5; Joint Opposition at 21. For the baseline pre-transaction case submitted by the Applicants, the FIM identifies approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] locations that would be considered profitable for FTTP expansion. See FIM Programs; supra n.965. 992 See Overview of AT&T FTTP Investment Model at 5 n.8; Katz Additional Detail ¶ 47 n.45. 993 See Overview of AT&T FTTP Investment Model at 5 n.8. 994 See Katz Decl. ¶¶ 126, 129-132; Katz Additional Detail ¶ 48. 995 Application at 29-31; Katz Decl. ¶¶ 122-124, 126, 132; Stankey Decl. ¶¶ 19-21, 34; Katz Additional Detail ¶ 52. 996 See Application 30-31; Stankey Decl. ¶¶ 18-20. 997 See Katz Decl. ¶ 126; Katz Additional Detail ¶¶ 52-53. The improved quality of the U-verse video product may also improve the profitability of AT&T’s FTTN service, which may reduce the incremental profitability of deploying FTTP in areas where FTTN is available. The FIM as modified by Dr. Katz [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See Katz Additional Detail ¶53 n.56. 998 See Katz Decl. ¶ 132; Katz Additional Detail ¶ 53. 999 See Katz Decl. ¶ 132 n.232; Katz Additional Detail ¶ 54. Dr. Katz chooses to model these effects through churn reductions. For the FTTN and FTTP churn reductions, Dr. Katz notes that [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See Katz Additional Detail ¶ 55. See also AT&T Response to Sept. 9, 2014, Information Request at 191-193. 1000 See Katz Decl. ¶ 132; Katz Additional Detail ¶ 53. Therefore, Dr. Katz assumes that AT&T’s monthly video churn would be between approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent post-transaction. See Katz Additional Detail ¶ 53. Dr. Katz also assumes that DIRECTV’s average churn (continued….) Federal Communications Commission FCC 15-94 129 would lead to enhancements in video content and the user interface, the Applicants do not value or quantify these enhancements or provide persuasive evidence that any such improvements would increase customer retention and subscribers.1001 However, we recognize there may be some reduction in churn from certain enhancements and the potential addition of NFL Sunday Ticket.1002 To allow for some reduction in churn in our adjustments, but without a basis to reduce the churn to the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] , we employ the smaller 50 basis point reduction used by the Applicants. 331. Further, to capture the effect of bundling broadband with an improved video product, Dr. Katz reduced FTTP and FTTN broadband churn in the FIM by 25 basis points each.1003 The Applicants do not provide any underlying rationale for reducing the FIM churn by 25 basis points. There is no evidence in our record that provides a different reduction, and because we acknowledge that there is some benefit to the improved bundle, we use the 25 basis point reduction in our analysis. Programming Payment Reductions. 332. The Applicants claim that the reduction of their programming payments would improve the business case for additional FTTP deployment.1004 The FIM accounts for video programming payments by building in an overall, monthly recurring expense for video service.1005 Dr. Katz models the reduced programming payment effect in the FIM by reducing this recurring expense associated with video service.1006 Dr. Katz relied on projections of programming payments from AT&T’s Corporate Financial Planning department to multiply the projected programming payments by the estimated percentage reduction in programming payments each year.1007 He then subtracted that amount from the monthly recurring video expense in the FIM.1008 (Continued from previous page) would decrease to 1.25 percent post-transaction because the combined entity would offer an enhanced fully integrated bundle. See id. 1001 See supra Section XI.C.3. (discussion of claimed benefits to video services). 1002 See ATT-FCC-01018530, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ; ATT-FCC- 01681777, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 1003 Dr. Katz lowers FTTP churn from about [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent and FTTN churn from about [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent. See Katz Additional Detail ¶ 54 n.59. 1004 See Application at 41; Stankey Decl. ¶ 45; Katz Decl. ¶¶ 126, 131; Katz Additional Detail ¶ 48. The programming payment reduction is based on AT&T’s programming payments being reduced to [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See AT&T Response to Sept. 9, 2014, Information Request, Exhibit 69.c.2, tab “Content Cost (output).” See also supra ¶ 284. 1005 See Katz Additional Detail ¶ 50. 1006 Id. 1007 See id. ¶ 50 n.50; FIM Programs. 1008 See Katz Additional Detail ¶ 50; FIM Programs. We note that the programming payment estimates that Dr. Katz uses in the FIM are lower than the estimated pre-transaction programming payments used elsewhere by the Applicants to establish that the transaction would result in a [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent reduction in programming payments overall. This inconsistency in baseline programming payments results in an understatement of the level of programming payments in both the no-transaction and post- transaction scenarios in the FIM for the years 2015 through 2024. Specifically, in evaluating the transaction, the Applicants project that [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See AT&T Response to Sept. 9, 2014, Information Request, Exhibit 69.c.2. Using these baseline projections, they estimate a [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . However, the FIM relies on a different set of AT&T programming payment projections [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See Katz Additional Detail ¶ 50 n.50. Nonetheless, because [BEGIN HIGHLY CONF. INFO.] [END (continued….) Federal Communications Commission FCC 15-94 130 333. In making adjustments to programming payments, Dr. Katz does not account for the portion of the programming payment reductions that the combined entity would pass on to consumers. This omission is very significant; the merger simulation submitted by the Applicants indicates that [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent of the programming payment reductions would be passed on to consumers.1009 For the FIM analysis, Dr. Katz assumes that the full amount of the programming payment reductions would be applied to the profitability of additional FTTP deployment.1010 334. To understand the potential impact of programming payment reduction pass-through, we evaluate the FIM under two alternative scenarios of pass-through.1011 First, we apply the full [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent programming payment reduction to the incentives to build out FTTP and assume none of the reduction is passed through in the form of lower U-verse video prices.1012 Second, in the absence of any evidence that pass-throughs in areas with new FTTP deployment would differ from those in the areas covered by the merger simulation, we apply the same full programming payment reduction but assume the prices on U-verse video fall by [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent of the reduction.1013 In order to implement this second scenario, we reduce the Average Revenue Per User (“ARPU”) on FTTP and FTTN U-verse video products by [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent of the applied programming payment reduction, which is the pass-through rate in the BH Simulation.1014 This ensures that we do not double count the programming payment reductions by allocating as a benefit both the consumer pass-through and the incentives to build out FTTP. (Continued from previous page) HIGHLY CONF. INFO.] are part of the ordinary course of business FIM, we decline to make any changes based on these programming payment reduction discrepancies. However, based on our sensitivity analysis, such a change in the cost of providing video service relative to the original AT&T and Dr. Katz modification models, in both the “no-transaction” and “post-transaction” scenarios, reduces the number of FTTP locations that would be deemed profitable using the FIM. 1009 See also supra ¶ 103. 1010 See Katz Decl. ¶ 131; Katz Additional Detail ¶¶ 49-50; FIM Programs. 1011 We tested the sensitivity of the pass-through assumption using pass-through rates ranging from 20 to 100 percent. The difference, relative to the baseline model of approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] locations, ranges from [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] additional locations for a 20 percent pass-through to approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] additional locations for a 100 percent pass-through. The sensitivity testing finds that reducing the pass-through results in fewer locations that meet the internal rate of return. 1012 See supra ¶¶ 96-102. 1013 We also adopt the full [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent pass- through from the BH Simulation in the FIM because, as a practical matter, it is more straightforward to adjust pass- through rates in the FIM than in the BH Simulation, which would require adjusting the assumed nested-logit demand model. See supra ¶ 103 & n.289. As we note in our competitive effects analysis, we recognize that AT&T may pass through less of the programming payment reduction to consumers than the BH Simulation predicts and potentially apply more of those reductions to FTTP deployment. See supra ¶ 290. 1014 Alternatively, the pass-through could be modeled by maintaining the original ARPU levels in the FIM but subtracting a smaller portion of the programming payment reductions from the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] of video service. We performed the analysis under each of these methods and found the results to be nearly identical. Consequently, in the discussion that follows, we report only the results from the specifications that adjust pass-through with adjustments to ARPU. We also note the likelihood that the combined entity could choose some other method of modeling the pass-through to consumers in the FIM after the transaction. Federal Communications Commission FCC 15-94 131 Cannibalization Effects. 335. We find that the transaction may reduce the incentives to deploy FTTP services because FTTP, by making U-verse video more attractive, would likely attract customers (and revenues) away from DIRECTV, which creates lost revenues that would be internalized by AT&T post-transaction.1015 The extent to which consumers are attracted away from DIRECTV to AT&T video service after FTTP deployment represents an opportunity cost to the combined entity because increases in profits on AT&T’s FTTP services are partially offset by losses in profits on DIRECTV services.1016 We conclude that this is a transaction-specific harm because post-transaction it could be profitable for the Applicants to reduce output below pre-transaction levels as the combined entity captures the increase in profits that DIRECTV would obtain absent that buildout. 336. Dr. Katz acknowledges this cannibalization effect in his analysis.1017 To incorporate the potential cannibalization of DIRECTV subscribers from the deployment of FTTP, Dr. Katz increases the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] of each new AT&T video subscriber by an amount equal to the lifetime value of a DIRECTV subscriber, multiplied by the projected migration rate of DIRECTV subscribers to AT&T FTTP video services.1018 We find that the estimated lifetime value used by the Applicants for DIRECTV may be too low, and as a consequence the Applicants underestimate the cost of cannibalization.1019 Although we do not change the way Dr. Katz modeled the cannibalization assumption in our analysis, we adjusted the lifetime value of DIRECTV customers to the 2014 profit margins, resulting in an increased opportunity cost of cannibalization of approximately 15 percent, which, in turn, increases the effect of cannibalization on the number of profitable FTTP locations in the FIM.1020 1015 Katz Additional Detail ¶ 56; Katz Decl. ¶ 126. 1016 Id. This situation is essentially the horizontal harm identified in the BH Simulation – that AT&T would reduce its output and raise its price because it would capture profits from the consequent boosts in demand for DIRECTV’s video service. See supra ¶¶ 59-62, 83-84, 127-133; Appendix C Section III.A, Section V.C.2. 1017 Katz Decl. ¶ 129; Katz Additional Detail ¶¶ 56-65. Also, Dr. Katz modifies the FIM to account for the increased opportunity costs of cannibalization of current AT&T IPDSL and DSL subscribers after the transaction. The baseline FIM does not account for the potential increase in attractiveness of fully integrated bundles of AT&T IPDSL and DSL broadband with DIRECTV video. To capture this potential increase in the profitability of current IPDSL and DSL subscribers, Dr. Katz reduces the churn rates of both of these services by 25 basis points from the baseline values. We do not dispute Dr. Katz’s IPDSL/DSL modification, and we use this churn reduction in our own modifications of the FIM. See Katz Decl. ¶ 130; Katz Additional Detail ¶¶ 66-68. 1018 See Katz Additional Detail ¶¶ 63-64; Katz Decl. ¶ 129. The appropriate migration rate to capture the movement of DIRECTV subscribers to AT&T video services is the AT&T-to-DIRECTV diversion rate. The diversion rate captures the following: the fraction of new AT&T video subscribers that come from DIRECTV when AT&T deploys FTTP in a specific area. See Katz Additional Detail ¶ 57. Dr. Katz runs the FIM under estimates of the lifetime value of a DIRECTV subscriber for the time period [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . The lifetime value estimate of a DIRECTV subscriber ranges from [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] in 2024. See id. ¶ 61, Table 1. 1019 Katz Additional Detail ¶ 61. The Applicants base their estimates on a 2012 calculation of DIRECTV’s subscriber lifetime value, which is based on, among other inputs, ARPU, margin, and customer acquisition costs. See id. ¶61 & n.65. However, DIRECTV’s margins have grown significantly since 2012, leading to an increase in the lifetime value estimates of their subscribers. See DIRECTV, Quarterly Results: 2014 DIRECTV Quarterly Reports (First Quarter – Fourth Quarter), http://investor.directv.com/financial-information/quarterly- results/default.aspx (visited June 24, 2015). 1020 The opportunity cost of cannibalization when the lifetime value of DIRECTV customers is updated to 2014 levels is approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] per new AT&T subscriber, relative to the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] cost assumed by Dr. Katz. See Katz Additional Detail Table 1. Federal Communications Commission FCC 15-94 132 337. Dr. Katz also uses two diversion rates, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent, to adjust the lifetime value to account for potential cannibalization.1021 The higher diversion rate is based on relative video shares between AT&T and DIRECTV.1022 Dr. Katz claims that this estimate is conservative because it is unlikely that AT&T would encourage DIRECTV customers to switch to AT&T.1023 To account for this variability, Dr. Katz reduces the diversion by five percentage points.1024 Dr. Katz states that these two diversion rates are consistent with the diversion rates based on results from his merger simulation.1025 The Modified Simulation calculates a diversion rate from AT&T to DIRECTV of just over [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent.1026 Therefore, to maintain consistency with that analysis, we use the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent diversion rate in our modified version of the FIM.1027 b. Results from Modifications 338. As noted above, we estimate the FIM under the assumption of no pass-through of programming payments as well as an [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent pass-through found in the merger simulation. We also report the post-transaction results assuming zero opportunity cost of cannibalization in order to quantify the extent to which the cannibalization effect reduces FTTP deployment incentives.1028 In all, we have reviewed three post- transaction scenarios: (1) the Katz version with investment of all programming payment reductions in FTTP deployment (“Katz Full FIM”); (2) a version with U-verse video ARPU reduced to account for pass-through of programming payment reductions to consumers and cannibalization (“Reduced Investment FIM”); and (3) a version with pass-through and no cannibalization (“No Cannibalization FIM”). 1021 See Katz Decl. ¶ 129; Katz Additional Detail ¶ 60, Table 1. 1022 Katz Decl. ¶ 129; Katz Additional Detail ¶ 58. 1023 Katz Decl. ¶ 129 n.226; Katz Additional Detail ¶ 59. 1024 Katz Decl. ¶ 129; Katz Additional Detail ¶ 60. 1025 Katz Additional Detail ¶ 60. 1026 This diversion rate is generated as output by the Modified Simulation. 1027 We tested the sensitivity of the diversion rate assumption using diversion rates of [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent. The difference, relative to the baseline model of approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] customer locations, ranges from approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] customer locations for a [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent diversion rate to approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] for a [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent diversion rate. The results from the sensitivity testing indicate that the cannibalization effect increases with an increase in the diversion rate, leading to fewer locations that clear the profitability hurdle in the FIM. 1028 Our analysis of Dr. Katz’s modifications finds that there is some nonlinearity in the interactions between the simultaneously adjusted assumptions; however, the individual effects can be approximated in an additive manner. For example, under the scenario of [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent investment of the reduction in programming payments, a 90 basis point decrease in video churn, a 25 basis point decrease in broadband churn, and cannibalization under the assumption of [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent diversion, the full model predicts an increase of approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] customer locations that clear the profitability hurdle. By comparison, simply adding the individual impacts of these changes in assumptions found in Table 1, below, results in an increase of slightly more than [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] locations that clear the hurdle. Federal Communications Commission FCC 15-94 133 Profitability With and Without Consumer Pass-through of Programming Payment Reductions. 339. In comparing the results of the Katz Full FIM (see Table below) with the results from the Reduced Investment FIM, it is evident that the pass-through of the programming payment reductions has a significant impact on FTTP investment incentives. Specifically, the Reduced Investment FIM (pass- through) projects [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] of fewer locations that would be profitable for FTTP deployment post-transaction compared to the Katz Full FIM (no pass-through) and fewer locations than were considered profitable absent the transaction.1029 340. As discussed above, we find it likely that some of the programming payment reductions would be passed on to consumers,1030 although the record does not establish the exact pass-through rate. In the Reduced Investment FIM, which considers pass-through of programming payment reductions to consumers, we find a significant reduction in the number of locations that would be profitable for investment. Therefore, we find it possible that the FIM would find fewer profitable FTTP customer locations than claimed by the Applicants as a result of the transaction, and we note that the incremental effect of the transaction on deployment may be negative given a high enough pass-through. Magnitude of Cannibalization Effect. 341. The results from the No Cannibalization FIM indicate that cannibalization of DIRECTV subscribers and revenues has a significant negative effect on AT&T’s FTTP deployment incentives. The number of locations clearing the profitability hurdle is approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] locations are profitable. The combined results indicate the impact of cannibalization extends to almost [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] locations. Table 1 [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] 2. Non-FIM Factors 342. The Applicants acknowledge that the decision to deploy FTTP services is [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .1031 Further, the Applicants claim that competition from Cable DOCSIS 3.1 and overbuilders such as Google Fiber affects FTTP deployment decisions.1032 Although the Applicants claim that the FIM is used to evaluate the profitability of FTTP deployment in the normal course of business, we are unable to determine, based on the record before us, the weight relative to other factors that AT&T would consider in its investment decision to deploy FTTP to additional locations.1033 1029 The Reduced Investment FIM projects [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . The cannibalization effect tends to dominate the post-transaction change in FTTP profitability predicted by the FIM when video ARPU is reduced by the full [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent. 1030 See supra ¶ 290. 1031 Overview of AT&T FTTP Investment Model at 5 n.8. 1032 FTTP Ex Parte Presentation at 4-5, 7; Application at 58-61; Katz Rely Decl. ¶ 39 n.77; Lee Decl. ¶¶ 24, 31-37. 1033 See supra ¶ 327. Federal Communications Commission FCC 15-94 134 343. Indeed, on November 12, 2014, Randall Stephenson, AT&T’s CEO, announced that AT&T would pause its investments to bring fiber connections to 100 cities until the Commission resolved issues related to the Open Internet proceeding.1034 Mr. Stephenson is quoted as saying, “We can’t go out and invest that kind of money deploying fiber to 100 cities not knowing under what rules those investments will be governed.”1035 On November 14, 2014, the Commission sent a letter to AT&T requesting an explanation of AT&T’s statement.1036 In response, AT&T stated that it is not limiting its FTTP deployment to 2 million homes and that, in fact, it plans to complete its previously announced FTTP expansion in 25 major metropolitan areas nationwide.1037 AT&T stated, however, that the uncertainty regarding regulatory treatment of broadband “makes it prudent to pause consideration of any further investments – beyond those discussed … AT&T simply cannot evaluate additional investment beyond its existing commitments until the regulatory treatment of broadband service is clarified.”1038 3. Conclusion 344. We find that the transaction harms the public interest to the extent that, compared to a non-merger, it creates a disincentive to build out FTTP due to the cannibalization effect from a loss of DIRECTV subscribers. This cannibalization effect is potentially larger than other factors that would incentivize AT&T to increase FTTP deployment as a result of the transaction. To address this transaction-specific harm, we impose a condition to preserve the pre-transaction FTTP buildout plans, future projections, and the transaction-specific incremental commitment.1039 1034 See Brian Fung, AT&T is Putting its Fiber Deployment on Ice Over Net Neutrality – For Now, WASH. POST, Nov. 12, 2014, available at http://www.washingtonpost.com/blogs/the-switch/wp/2014/11/12/att-is-putting-its-fiber- deployment-on-ice-over-net-neutrality-for-now/ (visited June 24, 2015) (“Fung, AT&T Putting Fiber Deployment on Ice”). In response to a Commission inquiry about this statement, AT&T affirmed that it is not limiting its FTTP deployment to 2 million homes, and in fact, it plans to complete its previously announced FTTP expansion in 25 major metropolitan areas nationwide. See AT&T Response to Nov. 14, 2014, Information Request at 2. 1035 See Fung, AT&T Putting Fiber Deployment on Ice. 1036 Nov. 14, 2014, Information Request to AT&T. 1037 AT&T Response to Nov. 14, 2014, Information Request at 1-2. AT&T had previously announced its plan to expand its GigaPower network in Raleigh-Durham, N.C., and Winston-Salem, N.C., and on April 21, 2014, it announced plans to expand its ultra-fast fiber network (GigaPower) to up to 100 candidate cities and municipalities nationwide, including 21 new major metropolitan areas and adjoining communities (Atlanta; Augusta, Ga.; Charlotte, N.C.; Chicago; Cleveland; Fort Worth, Texas; Fort Lauderdale, Fla.; Greensboro, N.C.; Houston; Jacksonville, Fla.; Kansas City; Los Angeles; Miami; Nashville, Tenn.; Oakland, Calif.; Orlando, Fla.; San Antonio; San Diego; St. Louis; San Francisco; and San Jose, Calif.). AT&T NEWSROOM, AT&T Eyes 100 U.S. Cities for Ultra-Fast Fiber Network. See also Application at 11-12. Currently, GigaPower is available in the following metropolitan areas and in some adjoining communities – Austin, Texas; Atlanta; Chicago; Dallas; Fort Worth, Texas; Houston; Kansas City; Raleigh-Durham, N.C.; San Jose, Calif.; and Winston-Salem, N.C. See AT&T Inc., U-Verse with AT&T GigaPower, http://www.att.com/att/gigapowercities/ (visited June 24, 2015). 1038 AT&T Response to Nov. 14, 2014, Information Request at 2 (emphasis added). See also FTTP Ex Parte Presentation at 5-6. 1039 The 12.5 million FTTP customer locations that will be deployed as a result of this condition include: (1) [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] locations were approved before the expansion plans, and [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] were newly approved as part of the expansion to 25 metropolitan area locations; (2) [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] that were previously determined to be profitable but were not funded as part of the GigaPower expansion; (3) 2 million additional locations that the Applicants determined were profitable as a result of this transaction; and (4) an additional [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] greenfield locations to which AT&T estimates it would deploy FTTP during the four year FTTP deployment commitment period. Federal Communications Commission FCC 15-94 135 345. Finally, we note that the deployment of FTTP may have a positive effect on broadband competition. At least one third-party report, which estimated the differences in cable market share depending on competitors’ technology, found that cable market share declines by approximately 40 percent when facing competition from FTTP instead of DSL.1040 In addition, our own analysis of data submitted in the record indicates that Comcast’s prices are [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] in areas where FTTP has been deployed compared to where DSL has been deployed.1041 These price differences for Comcast broadband plans for speeds equal to or greater than 25 Mbps range from [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] per month.1042 G. Evaluation of Applicants’ Claimed Fixed Wireless Local Loop Benefits 1. Introduction 346. The Applicants also claim that the proposed transaction would benefit consumers by enabling AT&T to deploy fixed wireless broadband services in the local loop (“Fixed Wireless Local Loop” service or “FWLL”) to 13 million households, in largely rural areas across the United States.1043 The Applicants have committed to complete the FWLL deployment within four years of the close of the transaction.1044 The Applicants state that the proposed FWLL service would offer specific coverage, speeds, and capacity and would be competitive both in areas with and without existing terrestrial broadband service, either as a standalone broadband service or as part of a bundle with wireless, video, and VoIP.1045 The Applicants also state that FWLL would allow each of the Applicants to offer integrated bundles that combine FWLL with DIRECTV video.1046 They assert that the FWLL buildout is possible only because of synergies that result from the transaction.1047 347. We find that the Applicants have not provided sufficient information in the record to enable us to determine whether any or all of the claimed benefits of the proposed FWLL are a direct result of the transaction. In addition, without challenging the assertion that FWLL deployment may provide 1040 CRAIG MOFFETT ET AL., MOFFETT NATHANSON RESEARCH, U.S. CABLE AND U.S. TELECOM: THE BROADBAND REPORT, 24, Exhibit 21 (July 8, 2014) (“MOFFETT NATHANSON BROADBAND REPORT”). The Moffett Nathanson Broadband Report estimates that Cable’s Share facing DSL is 80 percent, IPDSL is 60 percent, FTTN is 55 percent and FTTH is 40 percent. See MOFFETT NATHANSON BROADBAND REPORT at 24, Exhibit 21. 1041 Comcast Response to Jan. 8, 2015, Information Request. 1042 To derive Comcast prices when faced with FTTP or DSL service, we regressed Comcast’s monthly recurring revenue of a plan in a zip code on the percent of households in the zip code with access to competing broadband technologies and zip code demographic variables. We also included a piecewise linear spline for the percent of Comcast homes facing FTTP competition in the zip code and controlled for plan and month-year fixed effects. Reported price differences are based on the predicted prices from the regression for Comcast’s broadband plans predicted as if Comcast faced either Verizon FiOS or Verizon DSL competition at every household in the zip code. 1043 Application at 5, 44-45; Stankey Decl. ¶¶ 36, 48, 53. AT&T initially described the potential reach of FWLL as 13 million “customer locations,” which AT&T never defined. However, AT&T later determined that the correct term should be “households.” AT&T nonetheless continued to use the terms interchangeably. For the purposes of our analysis, we use the term “households.” See Appendix D, Analysis of AT&T’s FWLL Coverage and Performance Claims and Claim Rural Benefits, n.1 (“Appendix D”). 1044 Application at 5, 45. 1045 See id. at 43-45; Stankey Decl. ¶¶ 48-49, 53. 1046 Application at 44-45; Stankey Decl. ¶ 36; ATT-FCC-02210352, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 1047 See, e.g., Application at 43-45 (linking deployment of FWLL to rural customers to the financial attractiveness of deploying FWLL); Stankey Decl. ¶¶ 50-52 (stating that the transaction would allow AT&T to [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ). See also infra Section XI.G.5. Federal Communications Commission FCC 15-94 136 some benefits to consumers that have no access to terrestrial broadband or only one such option, the record establishes that they may be smaller than the Applicants have claimed. 2. FWLL Coverage and Performance Claims 348. Positions of the Parties. The Applicants assert that, post-transaction, the FWLL service would offer specific coverage and speeds in areas outside AT&T’s wireline footprint (or “franchise”) and in “areas within that footprint that currently do not receive the U-verse broadband and video bundle.”1048 349. The FWLL would use fixed LTE technology.1049 The Applicants claim that it would provide 15-20 Mbps and perform “as well as wireline broadband services advertised today.”1050 The FWLL service would be deployed in dedicated licensed spectrum already part of AT&T’s holdings on [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] in locations that meet highly specific deployment criteria.1051 Thus, deployment would occur on a cell-by-cell basis with the result that, rather than uniform coverage over large areas, coverage would be non-contiguous in many areas. Each cell site would use “the same basic equipment, spectrum and technological configuration”1052 and performance would be enhanced with professionally installed outdoor high-gain antennas (“HGAs”).1053 350. Discussion. We find limited support for the Applicants’ coverage and performance claims in the record. For example, coverage characteristics can differ between technologies and spectrum, and a network deployment would typically be closely engineered to the propagation characteristics of the spectrum being used. Here, however, the Applicants plan to deploy FWLL [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] that is uncommon and is not generally in use. They plan to use HGAs to compensate for any differences in coverage that result from using the FWLL spectrum.1054 351. To support these claims, the Applicants initially submitted lab data based on the existing service that suggested the coverage of the FWLL deployment would not be as extensive as the Applicants claimed. In response to Commission staff inquiries, the Applicants then submitted additional results from four geographically disparate FWLL field trials using a frequency band near the band they plan to use for FWLL.1055 While based on a different technology and using existing infrastructure, these data do show 1048 Application at 5. 1049 LTE is an acronym for Long Term Evolution, which is the latest mobile network technology standard set by the 3rd Generation Partnership Project (“3GPP”). See LTE Encyclopedia, https://sites.google.com/site/lteencyclopedia/home (visited June 17, 2015). See also 3GPP, The Mobile Broadband Standard, Technologies, Keywords & Acronyms, LTE, http://www.3gpp.org/technologies/keywords-acronyms/98-lte (visited June 17, 2015) (providing an overview of LTE). LTE is typically used in so-called “4G” networks. 1050 Application at 5, 43; Stankey Decl. ¶ 49; AT&T Sept. 9, 2014, Response to Information Request at 194 (stating that “even customers at the cell edge will experience speeds greater than 10 Mbps more than 90 percent of the time”). 1051 AT&T’s deployment criteria consist of four factors, each based on considerations such as location, spectrum availability, and the number of potential subscribers. See Appendix D ¶ 18. 1052 AT&T Response to Sept. 9, 2014, Information Request at 194. 1053 See id. at 194, 197, 202; AT&T Response to Dec. 15, 2014, Information Request, Exhibit 4.1 at 26; Appendix D ¶¶ 9, 20. 1054 See AT&T Response to Sept. 9, 2014, Information Request at 197, 202; AT&T Response to Dec. 15, 2014, Information Request, Exhibit 4.1 at 26; Appendix D ¶¶ 9, 20. 1055 See AT&T Response to Dec. 15, 2014, Information Request, Exhibit 4.2 at 4, 9, Exhibit 4.4 at 23, 24, 47-48, Supplemental Exhibit 58.b.1, FWLL coverage maps. See also AT&T Response to Sept. 9, 2014, Information Request at 3-4, 197 (describing lab modeling); AT&T Response to Dec. 15, 2014, Information Request at 5 (stating that AT&T has conducted “further lab modeling” since its first predictions). See generally Appendix D ¶¶ 33-34. Federal Communications Commission FCC 15-94 137 that, when compensated by other factors, the actual FWLL network coverage could be comparable to the lab-generated coverage predictions, as the Applicants have claimed. However, while these data are relevant and support the Applicants’ claims, this is a small sample in the context of a large and geographically diverse network deployment.1056 352. In addition, the record suggests that the Applicants have proposed a FWLL network design based on several assumptions that may not be realized in the actual deployed network. Specifically, Applicants have assumed that: (1) subscriber usage patterns would be consistent with a significantly lower speed tier than the Applicants propose to offer on the FWLL service; (2) they can [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .1057 The Applicants provide no evidence to explain how they would address practical, technical issues affecting performance that could plausibly arise as a result of differences between their assumptions and actual deployment and operating conditions. 353. Of particular concern is the lack of evidence to explain how the Applicants would overcome network performance challenges that could result from the difference between the conservative parameters chosen for the lab data submitted to support the FWLL performance and the more optimistic parameters at which the Applicants propose to market and operate FWLL. Similar concerns result from the evidence on FWLL capacity. The Applicants assumed that network capacity would [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] if certain technical features are incorporated during the FWLL deployment, but their own analysis does not appear to support this result.1058 In addition, the Applicants’ analysis shows that these technical features may not be realized for all sites because [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .1059 354. Network challenges could be mitigated by using additional spectrum or increased cell-site deployment. However, there is no evidence in the record that explains how AT&T might expand FWLL network capacity. Any required expansion, either through use of additional spectrum or increased cell count would increase both capital and operating expense.1060 In the absence of quantifiable financial data that support a different conclusion, we believe this expense could have a negative impact on the financial viability of the FWLL business model.1061 3. Claims that FWLL Would Benefit 13 Million Rural Customers 355. Positions of the Parties. The Applicants claim that post-transaction the FWLL deployment would reach 13 million mostly rural households.1062 AT&T claims that of these, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] would be rural, in-franchise households that do not have access to terrestrial broadband service.1063 The Applicants elsewhere claim that within 1056 See AT&T Response to Sept. 9, 2014, Information Request at 4; AT&T Response to Dec. 15, 2014, Information Request at 6. 1057 See AT&T Response to Sept. 9, 2014, Information Request at 3, 200, 202, 235, Exhibit 58.g.2 at 8; AT&T Response to Dec. 15, 2014, Information Request, Exhibit 6.1 at 4. 1058 See AT&T Response to Sept. 9, 2014, Information Request, Exhibit 58.i.1 at 13. 1059 See ATT-FCC-02208834, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 1060 See generally, AT&T Response to Sept. 9, 2014, Information Request, Exhibit 59.a.1 (showing net present values for various numerical ranges of subscribers), Exhibit 59.l.1 (showing lifetime value (“LTV”) calculations for FWLL). 1061 See AT&T Response to Sept. 9, 2014, Information Request, Exhibit 58.f.6 at 4. 1062 See supra n.1043. 1063 See ATT-FCC-02210352, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . Specifically, AT&T estimates that there are [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] households in no-broadband or “IP red” territories, of which [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. (continued….) Federal Communications Commission FCC 15-94 138 the planned FWLL deployment area, almost 20 percent of the 13 million households (i.e., 2.6 million households) have no access to terrestrial broadband and an additional 27 percent of the 13 million households (i.e., 3.5 million households) have only one terrestrial option today, “and in most instances that single option is DSL or a relatively slow cable modem service.”1064 356. Discussion. The record provides no basis to conclude either that AT&T would deploy FWLL to 13 million “largely rural” homes or that the Applicants have used the proper methodology to estimate how many of these homes are unserved by terrestrial broadband providers or have only one terrestrial broadband option.1065 AT&T’s rural coverage estimates are based on its own rural population density assumptions that we do not accept as appropriate. Applying the Commission’s standard “guideline” definition of “rural”1066 and broadband deployment estimates from the National Broadband Map data (commonly called SBI Data),1067 the Commission staff estimates that approximately 4.5 million rural households would be within AT&T’s proposed FWLL deployment area.1068 This estimate is substantially fewer than half of AT&T’s claim of 13 million “largely rural” households. 357. Using the Commission’s definition of rural, there are many fewer rural households with zero or one terrestrial broadband provider than AT&T claims in the FWLL deployment areas. According to AT&T, 2.6 million households within the projected FWLL deployment have no existing terrestrial broadband service.1069 Using the census block level SBI data for the projected FWLL deployment area, 1.5 million total households, of which 1.0 million are rural, are estimated to have no terrestrial broadband option.1070 Two million rural households would have only one terrestrial broadband option.1071 These numbers are significantly lower than AT&T’s estimates. 4. Competitive Standalone FWLL and DIRECTV Integrated Bundles Would Be a Benefit of the Transaction 358. Positions of the Parties. AT&T asserts that an additional benefit of this transaction would be allowing it to provide a competitive broadband service on a standalone basis both in rural areas where there is no terrestrial broadband option and in areas where broadband competition already (Continued from previous page) INFO.] are rural and [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] are rural and included in its FWLL deployment. Id. 1064 Application at 44. AT&T describes its methodology as follows: “ [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .” AT&T Response to Sept. 9, 2014, Information Request at 205. 1065 See Application at 44-45; Stankey Decl. ¶¶ 36, 54-55. 1066 See Facilitating the Provision of Spectrum-Based Services to Rural Areas and Promoting Opportunities for Rural Telephone Companies to Provide Spectrum-Based Services, WT Docket No. 02-381, Report and Order and Further Notice of Proposed Rulemaking, 19 FCC Rcd 19078, 19087-88, ¶ 12 (2004) (“We recognize, however, that the application of a single, comprehensive definition for ‘rural area’ may not be appropriate for all purposes. . . . Rather than establish the 100 persons per square mile or less designation as a uniform definition to be applied in all cases, we instead believe that it is more appropriate to treat this definition as a presumption that will apply for current or future Commission wireless radio service rules, policies and analyses for which the term ‘rural area’ has not been expressly defined. By doing so, we maintain continuity with respect to existing definitions of ‘rural’ that have been tailored to apply to specific policies, while also providing a practical guideline.”). We see no reason to revisit this guideline at this time. 1067 Commission staff determines broadband availability in rural areas at the census-block level. See Appendix D ¶¶ 45-49. 1068 See Appendix D ¶ 48. 1069 See Application at 44; Appendix D ¶¶ 42, 50, Table 1. 1070 See Appendix D ¶¶ 49-50, Table 1. 1071 See id. Federal Communications Commission FCC 15-94 139 exists.1072 In addition, noting consumers in these areas lack access to integrated bundles that include broadband, video, and voice services, the Applicants also claim that they would offer competitive satellite video, home broadband, and home VoIP bundles featuring FWLL service.1073 The Applicants intend to price FWLL competitively with existing comparable wireline broadband offerings.1074 In addition, AT&T plans to impose limits on the amount of data a customer may use, which could affect the competitiveness of its pricing. AT&T currently expects to offer a usage cap between [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] per month that it expects to satisfy most customers.1075 359. WGAW and Netflix challenge whether FWLL would offer significant benefits, and other commenters question whether FWLL is a public benefit at all.1076 WGAW states that FWLL would offer inferior broadband technology for an expensive price.1077 Netflix is concerned that the Applicants did not specify the usage allowance.1078 Public Knowledge-ILSR assert that unless FWLL competes with cable broadband in all material respects, it should only be considered as a partial substitute or complement to other terrestrial broadband offers.1079 Free Press states that FWLL is “the same expensive, capped, fixed 4G wireless services that it currently offers in areas where it refuses to upgrade its wired networks.”1080 360. The Applicants respond that by offering higher speeds compared to most terrestrial services available in rural areas,1081 the FWLL service would be a “dramatic improvement for customers that currently do not have access to any high-speed fixed broadband.”1082 The Applicants also assert that introduction of FWLL broadband service can “exert important competitive pressure on rival broadband providers, which can lead to lower prices and greater investments in [broadband] infrastructure. . . .”1083 361. Discussion. While the record is limited, we do not now dispute the assertion that if AT&T deploys FWLL to areas without a fixed broadband provider, FWLL would be of greater benefit to consumers than in other areas. Moreover, for customers who prefer bundles, new FWLL and DIRECTV video bundles may be more attractive than available alternative standalone services. Consequently, there is no need to dispute the assertion that there may be a benefit to deploying FWLL to customers where there is currently no terrestrial broadband service, depending on the pricing and data caps on such a service. 362. In addition, we estimate that approximately 3 million target FWLL households are served by DSL only, of which 1.6 million are in rural areas.1084 For these 3 million households, FWLL could 1072 Application at 43- 44; Stankey Decl. ¶ 55; ATT-FCC-02210352, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 1073 Application at 42-45. 1074 Id. at 43 (stating that the FWLL would be priced as a home broadband service); AT&T Response to Sept. 9, 2014, Information Request at 198-199 (“AT&T . . . intends to price [FWLL] competitively with existing comparable wireline broadband offerings.”). 1075 See ATT-FCC-01969908, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ; AT&T Response to Sept. 9, 2014, Information Request at 199. 1076 WGAW Petition at 16; Netflix Comments at 26 (citing Application at 43). 1077 WGAW Petition at 5-6. 1078 Netflix Comments at 26 (citing Application at 43). 1079 See Public Knowledge-ILSR Petition at 17. 1080 Free Press Petition at 31. 1081 Katz Reply Decl. ¶ 45. 1082 Id. ¶ 44. 1083 Id. 1084 See Appendix D n.72. Federal Communications Commission FCC 15-94 140 offer better speeds, because DSL speeds on average tend to be slower than those potentially offered by the Applicants’ FWLL service.1085 However, when factoring for the typically lower price of DSL (approximately $29.95 per month for 6 Mbps service), FWLL – which would be priced at [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] 1086 – may not be competitive. 363. Our evaluation of the Applicants’ related claims about the benefits that would flow to consumers from the ability of FWLL broadband service to compete with cable broadband in those areas where cable broadband exists is also limited. The Applicants’ merger simulation does not incorporate FWLL, and the record otherwise lacks substantial quantitative data. By contrast, the record does suggest that data rate performance is an important element of a broadband service offering.1087 The record suggests that FWLL offerings would be at [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] with a [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] usage limit.1088 And, as discussed, the Applicants may not be able to meet these speed claims due to the number of subscribers, the location of subscribers, the time of day, and other factors. With uncertainties about the actual users’ speeds on the record, and without additional information from the Applicants regarding how the price of these offerings would compare to prices of faster services from other broadband providers, we cannot determine the extent of the benefit FWLL would provide in areas where there is existing cable or FTTP competition.1089 364. We also do not have a basis to assess the potential increase in value of integrated bundles of FWLL and DIRECTV video or how it could support the funding of FWLL deployment. In areas where there is competition from cable, consumers already have access to bundled products from cable companies, which they may prefer to the integrated FWLL and DIRECTV bundle. On the other hand, the Applicants may be able to realize sufficient efficiencies to price the FWLL/DIRECTV bundle to compete effectively with cable bundles. However, pricing at those levels could jeopardize the revenue stream that the Applicants claim is a necessary input to build out FWLL.1090 The record offers no evidence; therefore we do not consider this claim in our overall analysis. 1085 DSL speeds range between 768 kbps – 8.5 Mbps and depend on multiple factors including the length of the phone wiring and the thickness of the wire. For example, AT&T advertises 768 kbps – 6 Mbps for its DSL service. See AT&T Inc., Internet, DSL High Speed Internet, Internet Speeds That Meet Your Needs, http://www.att.com/shop/internet/internet-service.html (visited June 15, 2015). See generally Bradley Mitchell, DSL Availability – DSL Lookup Services and Factors Affecting DSL Availability, http://compnetworking.about.com/od/dsldigitalsubscriberline/a/availability.htm (visited June 15, 2015) (describing technical factors that limit the coverage of DSL service); Bradley Mitchell, DSL Speed – How Fast is DSL Internet Service?, http://compnetworking.about.com/od/dsldigitalsubscriberline/f/dslspeed.htm (visited June 15, 2015) (discussing the speed of DSL service, including factors that may affect speed). 1086 ATT-FCC-01969908, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 1087 See generally Lee Decl. ¶¶ 28-37 (discussing deployments and rates). 1088 ATT-FCC-01969908, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 1089 We also are unable to assess the validity of the Applicants’ competitive claims because at the same time that the Applicants claim the speeds of FWLL are sufficient to provide broadband Internet access that is competitive with existing cable and FTTP providers, the Applicants also claim that FWLL [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . Katz Reply Decl. ¶ 34 n.64. This calls into question the ability of FWLL to provide broadband Internet access service that is robust enough to satisfy trends in consumer demand for Internet access. 1090 See infra ¶¶ 372-374. Federal Communications Commission FCC 15-94 141 5. FWLL Deployment is Transaction Specific 365. Positions of the Parties. The Applicants claim that the FWLL deployment is only feasible due to synergies resulting from the proposed transaction.1091 These include a single “truck roll” to install DBS video and FWLL services as well as marketing and sales synergies.1092 The “recovery” of FWLL deployment costs would have the effect of benefitting consumers through “expanded offerings,” including FWLL services as part of new bundles.1093 366. The Applicants also assert that the ability to offer an integrated FWLL and DIRECTV video bundle would contribute to market conditions that would allow AT&T to deploy FWLL under more favorable circumstances than those available to AT&T as a standalone company.1094 367. Several petitioners and commenters assert that the Applicants’ claimed benefits are not transaction specific. WGAW suggests that the Applicants’ offer to expand broadband Internet service through FWLL is not new and speculates that FWLL deployment would happen regardless of whether the transaction is approved.”1095 WISPA asserts AT&T’s planned roll out of FWLL is not a transaction- specific benefit but rather an incremental outgrowth of its existing spectrum and infrastructure.”1096 Free Press disputes AT&T’s claim that the transaction is necessary to allow AT&T to realize the synergies it would need to make FWLL service deployment economically feasible,1097 and it asserts that AT&T’s FWLL commitment is not transaction specific because AT&T already offers fixed LTE service nationwide.1098 368. By contrast, CWA supports the transaction, echoing the Applicants’ claim that the combined entity would have the economic incentive to increase investment in high-capacity networks and FWLL.1099 1091 Application at 45 (stating that bundling FWLL, DIRECTV video, and VoIP would attract more subscribers with lower churn outside the U-verse footprint); AT&T Response to Sept. 9, 2014, Information Request at 205 (stating that AT&T expects more subscribers and greater revenue from a broadband/MVPD/voice bundle than from a “standalone FWLL bundle”), 235 (stating that the transaction improves the business case for deploying FWLL because “[i]t brings a new revenue source (MVPD services) and a more compelling offering (a seamless broadband/MVPD/voice bundle available nationwide) that will increase per-customer revenues”). 1092 AT&T Response to Sept. 9, 2014, Information Request at 209; Joint Opposition at 12. The Applicants claim that the key impediments to deploying FWLL include the “high deployment costs.” Application at 44. These costs include the expense of installing new antennas at households and cell sites. Application at 44; see also AT&T Response to Sept. 9, 2014, Information Request at 209 (“Deploying fixed WLL requires incremental equipment at existing LTE cell sites, including new antennas, radios, and base band units.”). In its analysis, AT&T assumes FWLL deployment capital expenditures total “approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .” AT&T Response to Sept. 9, 2014, Information Request at 209. 1093 See Joint Opposition at 12 (citing Application at 29-33; Stankey Decl. ¶¶ 6, 19, 26-32; Moore Decl. ¶¶ 26-29; Katz Decl. ¶¶ 4, 98-99; Katz Reply Decl. ¶¶ 2-3, 26). 1094 AT&T Response to Sept. 9, 2014, Information Request at 205; Katz Decl. ¶ 135; Stankey Decl. ¶¶ 36, 52-53; Joint Opposition at 24-25. 1095 WGAW Petition at 25. WGAW further argues that claims about broadband deployment, including FWLL, are not transaction specific. See WGAW Reply at 24-32. 1096 Comments of the Wireless Internet Service Providers Association, MB Docket 14-90, at 4-8 (filed Sept. 16, 2014) (“WISPA Comments”). 1097 Free Press Petition at 31-32. 1098 Id. at 31. DISH also asserts that AT&T’s FWLL claims are not transaction specific. DISH Reply at 13. 1099 Reply Comments of Communications Workers of America, MB Docket 14-90, at 7-8 (filed Oct. 16, 2014) (“CWA Reply”). CWA also argues that the transaction would create good jobs, create a stronger competitor to cable in video and broadband markets, improve the economics for high-speed broadband expansion, and facilitate an (continued….) Federal Communications Commission FCC 15-94 142 369. As discussed above in connection with the FTTP deployment, ACA argues that AT&T’s commitment to deploy the FWLL “may have significant value but only if AT&T is not receiving universal service support to serve the same locations.”1100 WISPA similarly states that the Commission should condition any approval of the transaction with a requirement that AT&T not accept Connect America Phase II support.1101 370. Discussion. The Applicants have not provided sufficient quantitative financial evidence to support the assertion that the ability to deploy FWLL is a transaction-specific benefit; nor have they adequately substantiated their cost and efficiency claims in their FWLL financial deployment model. In particular, the Applicants have not provided any simulations or financial modeling that link the cost savings and bundling efficiencies to the deployment of FWLL post-transaction. Therefore, we are limited in our ability to effectively evaluate the feasibility of FWLL deployment. 371. The Applicants’ financial model analyzing FWLL deployment does not include quantitative information on its asserted post-transaction cost savings or revenue increases.1102 Key financial assumptions for the FWLL business cases both with and without the transaction appear to be the same.1103 Furthermore, AT&T has not elsewhere provided any specific quantitative assessments on how the deployment of FWLL service would become profitable with the transaction. 372. A critical element of that profitability is the uptake of their service by consumers. A central claim by the Applicants is that the bundling made possible by the transaction would improve the lifetime value (or “LTV”) of FWLL customers and would reduce churn. However, evidence used to support this claim is inconsistent with other assumptions in the record that suggest that a lower churn is not needed to support the financial feasibility of FWLL service.1104 In addition, the Applicants use different video penetration rates in the same spreadsheet to calculate post-transaction churn and lifetime value.1105 These discrepancies undermine the Applicants’ claims that these benefits are transaction specific. 373. The Applicants also argue that the ability to offer integrated bundles improves churn and penetration rates, and therefore creates additional value.1106 However, the data provided in the record do not provide any direct evidence of what customer churn or penetration rates would be post-transaction. In fact, some of the Applicants’ FWLL financial analysis assumes that churn for FWLL services would be unchanged by the transaction.1107 Thus, we cannot conclude that there is a difference in the pre- and post- transaction churn rates. As described elsewhere, we also question the Applicants’ assumption on the (Continued from previous page) open Internet through AT&T’s commitment to offer standalone retail video and broadband for three years after closing of the transaction. Id. at 1-8. 1100 ACA Comments at 29-30. See also supra n.968. 1101 WISPA Comments at 9. 1102 AT&T Response to Sept. 9, 2014, Information Request, Exhibit 59.a.1. 1103 Compare AT&T Response to Sept. 9, 2014, Information Request, Exhibit 58.g.1 [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] at 3, 6, 9 with AT&T Response to Sept. 9, 2014, Information Request at 204, 214 (narrative response assuming the same market share, churn, and average revenue per user post-transaction). 1104 AT&T Response to Sept. 9, 2014, Information Request at 215 (“ [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .”). 1105 AT&T Response to Dec. 15, 2014, Information Request at 11 n.18 (“ [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .”); id. at 10 n.16 ( [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ). 1106 Application at 45. 1107 See supra nn.1103-1104. Federal Communications Commission FCC 15-94 143 attractiveness of integrated bundles that rely on the FWLL product as the broadband alternative, especially in cable competitive areas, and hence the claimed impact on customer churn and penetration rates that would make FWLL deployment feasible. 374. As discussed in greater detail above, the Applicants have proposed to deploy a FWLL service that is designed to perform at a much lower capacity and throughput levels than they plan to deploy.1108 To the extent the Applicants are successful in marketing the FWLL service, a large number of customers attempting (or expecting) to use the network at advertised throughput and capacity limits could cause degraded network performance – making it less competitive. The financial viability of the network would necessarily be affected if the result were that the FWLL service lost customers or the Applicants were forced to upgrade the network with more spectrum and/or cell sites to improve performance. The Applicants have submitted no data to show how they would manage this potential situation, but we believe it is a possibility that could add to the other challenges to financial viability discussed above. 375. Finally, with regard to ACA’s and WISPA’s concerns about AT&T using USF or Connect America Fund support to satisfy its FWLL deployment commitment, as discussed above, the Applicants assert that [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .1109 6. Conclusion 376. The Applicants are proposing to deploy a network that would need to meet specific coverage, capacity, and throughput benchmarks to provide the benefits claimed by the Applicants. While the Applicants have provided some evidence that their FWLL coverage and performance estimates are appropriate, the record presented is limited and raises unanswered questions about potential technical issues that could adversely affect network performance. Due to the technical characteristics of the network the Applicants would deploy, any failure to meet those standards would have a direct impact on the viability of the FWLL service. In addition, the Applicants have used rural coverage estimates that do not follow the Commission’s guideline metric for identifying rural areas. More importantly, the Applicants do not adequately establish the extent of the benefits they have claimed for FWLL in rural areas. Even if there were to be some consumer benefit in areas where there is currently no terrestrial broadband or only a DSL terrestrial broadband provider, the extent of those benefits would depend on the speed-price offerings available and consumer preferences for higher speeds. The Applicants have not demonstrated that the FWLL would be a meaningful competitor to existing cable broadband providers in the proposed FWLL deployment areas. 377. Finally, there is no evidence in the record to show that the Applicants would offer competitive bundles in areas where there is existing terrestrial broadband service. For these reasons, we ascribe minimal weight to the claimed benefit of FWLL deployment in our analysis of the transaction. H. Other Potential Public Interest Benefits 1. Cybersecurity 378. Positions of the Parties. The Applicants assert that the transaction would “enable [the combined entity] to provide even better security going forward” due to smooth cybersecurity integration efforts that use the National Institute of Standards and Technology (“NIST”) Cybersecurity Framework1110 as a resource, and that expand the AT&T internal security and supply chain security protocols.1111 Through these efforts, the Applicants emphasize that the combined entity would ensure that 1108 See supra ¶¶ 352-353. 1109 AT&T Response to Sept. 9, 2014, Information Request at 220. 1110 See NATIONAL INSTITUTE OF STANDARDS AND TECHNOLOGY, FRAMEWORK FOR IMPROVING CRITICAL INFRASTRUCTURE CYBERSECURITY (Feb. 12, 2014) (“NIST CYBERSECURITY FRAMEWORK”), available at http://www.nist.gov/cyberframework/upload/cybersecurity-framework-021214-final.pdf (visited June 23, 2015). 1111 See AT&T Response to Sept. 9, 2014, Information Request at 271-273. Federal Communications Commission FCC 15-94 144 “no gaps are created through the combination of two different systems” and that the “best practices of each company are adopted to facilitate stronger security efforts going forward.”1112 In addition, AT&T asserts that the integration of the two companies would enhance the effectiveness of AT&T Security Operation Center’s (“SOC”)1113 centralized function to achieve real-time awareness of cybersecurity threats by “providing a larger footprint to draw upon in assessing overall cybersecurity risks.”1114 379. AT&T affirms that as part of the integration planning process, a high priority is the “smooth integration of cybersecurity integration efforts.”1115 First, as part of its due diligence, AT&T requested that DIRECTV identify and submit documents related to [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .1116 380. Second, AT&T asserts that the NIST Cybersecurity Framework would be an “an important resource” to draw upon once the “full-fledged integration efforts begin following the closing of the proposed [t]ransaction.”1117 Prior to the transaction, AT&T publicly supported the NIST Cybersecurity Framework and promised to review the details and assess how the document complements its existing cyber-risk management program.1118 DIRECTV states that the company “currently uses” the NIST Cybersecurity Framework.1119 381. Third, AT&T developed, and submitted for the record, the company’s security policy and program written prior to the release of the NIST Cybersecurity Framework in February of 2014.1120 AT&T asserts that it would expand this security policy and program to cover DIRECTV.1121 In addition, AT&T affirms that its prior history of incorporating acquired businesses into its risk management program includes working with the AT&T Chief Security Office.1122 382. Fourth, AT&T asserts that, as part of the AT&T security efforts, supply chain security is an “important aspect” and, furthermore, that this security process would be a “critical aspect of integration efforts with DIRECTV.”1123 This process includes: careful selection of well-established 1112 See id. at 273; see also DIRECTV Response to Sept. 9, 2014, Information Request at 84-85 (stating that “integration planning efforts are still preliminary . . . [i]mplementation of all integration plans . . . may be amended based on information gained following the closing of the transaction”). 1113 The AT&T Security Operations Center is a “centralized function that continuously monitors and analyzes traffic through AT&T’s backbone, providing near-real-time and advance notification of different types of security events across multiple devices and device types.” See AT&T Response to Sept. 9, 2014, Information Request at 273. 1114 See id. 1115 See id. 1116 DTVFCC-02406508, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ; see also ATT- FCC-01689359, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 1117 See AT&T Response to Sept. 9, 2014, Information Request at 272. 1118 ATT-FCC-03375051, “Protecting our Nation’s Critical Infrastructure,” Ed Amoroso, AT&T Senior Vice President and Chief Security Officer (Feb. 12, 2014) (blog posted upon the release of the NIST Cybersecurity Framework, with AT&T’s promise to review and “see how it best complements [AT&T’s] existing cyber-risk management program”). 1119 See DIRECTV Response to Sept. 9, 2014, Information Request at 84-85. 1120 ATT-FCC-03375006, “AT&T Information & Network Security Customer Reference Guide, v.5.1” (Feb. 2013) (providing an introduction to AT&T’s global security organization, a review of AT&T’s security roles and responsibilities, and overview of AT&T’s security policy and comprehensive programs). 1121 See AT&T Response to Sept. 9, 2014, Information Request at 273. 1122 See ATT-FCC-01259171, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 1123 See AT&T Response to Sept. 9, 2014, Information Request at 273-274. Federal Communications Commission FCC 15-94 145 infrastructure vendors, evaluation of hardware and software, equipment testing ensuring against the interception or redirection of data transfers, software testing ensuring data transmission security, examination of manufacturers’ provenance and business history, and the consultation with NIST or the Department of Commerce.1124 383. Furthermore, AT&T asserts that the combined entity would provide a “larger footprint” to assess overall cybersecurity risk, thereby enhancing the effectiveness of the AT&T SOC,1125 which monitors and analyzes traffic through AT&T’s backbone infrastructure and provides “near-real-time and advance notification of different types of security events across multiple devices and device types.”1126 384. Discussion. The Applicants, in their response to the Commission’s request for information, state that the public would benefit from increased security and from the combined entity’s increased real-time awareness of cybersecurity threats across a larger footprint. No commenter raised objections to the transaction based on cybersecurity issues, and therefore, the analysis is based on documents submitted by the Applicants. AT&T provides information about its current security and privacy program, but the record reveals little information about DIRECTV’s current programs and policies for cybersecurity and privacy. 385. The Applicants assert that the NIST Cybersecurity Framework would be a resource during the full-fledged integration efforts. The NIST Cybersecurity Framework is focused on using “business drivers to guide cybersecurity activities” and is considering cybersecurity risks as “part of the organization’s risk management processes.”1127 Part of this work includes the development of “organizational understanding to manage cybersecurity risk to systems, assets, data, and capabilities.”1128 The record shows that AT&T initiated efforts to develop an organizational understanding to manage the cybersecurity risk to DIRECTV’s systems, assets, data, and capabilities,1129 but it does not reveal additional relevant evidence to show what information DIRECTV provided, how AT&T analyzed the information, or what are the planned development efforts to follow the NIST Cybersecurity Framework. 386. AT&T asserts that the combined entity would use the best practices of both companies to facilitate “stronger security efforts going forward.”1130 Further, the Applicants state that “integration planning efforts are still preliminary,” that implementation of these plans is “subject to completion of the 1124 See id.; see also ATT-FCC-03375006, “AT&T Information & Network Security Customer Reference Guide, v.5.1” (Feb. 2013), at 03375022-03375023 (explaining that suppliers undergo background check requirements as part of supplier agreements with AT&T to ensure proper screening and to make suppliers aware of responsibilities). 1125 See ATT-FCC-03375006, ”AT&T Information & Network Security Customer Reference Guide, v.5.1” (Feb. 2013), at 03375024 (describing AT&T SOC as a centralized command and control facility that monitors and analyzes traffic through the AT&T IP backbone, providing real-time advance notifications of security events, and that produces AT&T-specific security reports and alerts). 1126 See AT&T Response to Sept. 9, 2014, Information Request at 273 (outlining the role of the AT&T SOC, including providing “alerts, situational awareness, incident response, and proactive threat vulnerability analysis to manage threats and clean harmful traffic”). 1127 NIST CYBERSECURITY FRAMEWORK at 1. The Framework Core presents the core functions: identify, protect, detect, respond, and recover, with the goal of helping enable risk management decisions, addressing threats, and improving from previous practices. Id. at 7. 1128 Id. at 8. 1129 DTVFCC-02406508, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ; see also ATT- FCC-01689359, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 1130 AT&T Response to Sept. 9, 2014, Information Request at 273. Federal Communications Commission FCC 15-94 146 regulatory review process and consummation of the transaction,”1131 and that “full-fledged integration efforts begin following [the] closing of the proposed Transaction.”1132 The record, therefore, does not reveal how DIRECTV would implement its ongoing security protocol; how the two companies would integrate with respect to cybersecurity assets and policies; or how AT&T plans to expand its security policy and program, supply chain protocols, or the reach of the AT&T SOC. 387. AT&T claims that the combined entity would assess overall cybersecurity risks over a larger footprint, which would enhance the effectiveness of the AT&T SOC.1133 While we agree that economies of scale may contribute to better cybersecurity through improved information sharing, threat detection, and response, a company’s size does not necessarily indicate the strength of its cyber defenses. We also note that DIRECTV’s technical operations includes equipment and procedures for which AT&T may not have direct experience with and absent new training, workforce, or defensive systems, AT&T may not be initially postured to defend with AT&T’s existing capabilities. The increased footprint, therefore, would not necessarily result in an increased strength in cyber defenses. A new combined entity faces interoperability and coordination challenges and may create new vulnerabilities for both systems. 388. The record does not provide sufficient information about the past security performance or independent security offered by each party in comparison to what would be offered by the combined entity to determine whether there would be a net benefit to the public in this area. Therefore, based on the currently available information, we ascribe minimal weight to the claimed cybersecurity benefit in our analysis of the transaction. 2. Diversity Practices 389. AT&T states that a diverse workforce and a commitment to inclusion in all business practices allow it to best serve its customers, suppliers, and investors.1134 AT&T states that it intends to extend its “best-in-class diversity values” to the combined entity.1135 The Applicants, however, have not demonstrated that DIRECTV does not already employ similar diversity practices or that it would not do so absent the transaction. Indeed, we note that Section 25.601 of the Commission’s rules extends certain equal employment opportunities (“EEO”) requirements to DBS providers such as DIRECTV.1136 Accordingly, we ascribe minimal weight to the claimed diversity practices benefit in our analysis of the transaction.1137 3. Labor Practices 390. AT&T states that the combined entity would continue AT&T’s practice of working responsibly with the unions representing its workforce.1138 AT&T notes that over half of its workforce is union-represented, resulting in AT&T having the largest full-time union workforce of any company in America.1139 CWA, a labor organization representing 700,000 workers, including 110,000 AT&T 1131 See DIRECTV Response to Sept. 9, 2014, Information Request at 84-85 (stating that AT&T, as the acquirer, can provide detail regarding AT&T’s plans with “respect to investment in communications security and cybersecurity technologies and practices post-closing”). 1132 See AT&T Response to Sept. 9, 2014, Information Request at 274. 1133 See id. at 273. 1134 Application at 29; Stankey Decl. ¶ 64. 1135 Application at 9, 29; Stankey Decl. ¶ 64. 1136 See 47 C.F.R. § 25.601. 1137 See News Corp.-Hughes Order, 19 FCC Rcd at 623-624, ¶ 357. 1138 Application at 9, 29; Stankey Decl. ¶ 64. 1139 Stankey Decl. ¶ 64. Federal Communications Commission FCC 15-94 147 employees, contends that the transaction would benefit the combined entity’s employees.1140 CWA notes that its collective bargaining agreements with AT&T guarantee that AT&T’s labor policies would extend to DIRECTV’s non-management employees after the transaction, providing DIRECTV employees with the opportunity to select union representation free from management intimidation, and that this opportunity would ultimately benefit the workers, “their families, communities, and consumers with the quality service provided by a skilled, trained, career workforce.”1141 391. As a general matter, we believe that labor issues are best addressed by the National Labor Relations Board.1142 Further, there is no evidence in the record that DIRECTV, too, does not have a “skilled, trained, career workforce” that provides quality service to consumers. Therefore, while AT&T asserts that it has good labor relations with its employees and CWA and that those relations will continue following the acquisition of DIRECTV, we ascribe minimal weight to these labor relations benefits in our analysis of the transaction. XII. REMEDIES A. Introduction 392. The Commission’s review of a proposed transaction entails a thorough examination of the potential public interest harms and any verifiable, transaction-specific benefits, including any commitments made by the Applicants to further the public interest. As part of this process, the Commission may impose additional remedial conditions to address potential harms likely to result from the proposed transaction or to help ensure the realization of any promised potential benefits.1143 If, on balance, after taking into consideration these additional remedial conditions, the potential benefits associated with the proposed transaction outweigh any remaining potential harms, the Commission will find that the proposed transaction serves the public interest. 393. As described above, we find that the transaction as proposed has the potential to cause public interest harms as well as public interest benefits. Under our sliding-scale approach, we cannot conclude based on this record that the potential benefits are sufficiently large, specific, and imminent to outweigh all potential harms. However, we have imposed several conditions, which as explained below 1140 See CWA Comments at 1-2, 13-14; CWA Reply at 1-3. 1141 CWA Comments at 13-14; CWA Reply at 3. See also Letter from Richard L. Trumka, President, American Federation of Labor and Congress of Industrial Organizations (AFL-CIO), to The Honorable Thomas Wheeler, Chairman, FCC, MB Docket No. 14-90, at 1-2 (Sept. 11, 2014) (transaction would benefit employees); Letter from Joshua D. Sword, Secretary-Treasurer, West Virginia AFL-CIO, to The Honorable Thomas Wheeler, Chairman, FCC, MB Docket No. 14-90, at 1-2 (Sept. 15, 2014) (same); Letter from Ken Sagar, President, Iowa Federation of Labor, to The Honorable Thomas Wheeler, Chairman, FCC, MB Docket No. 14-90 (Sept. 16, 2014) (same); Letter from Shar Knutson, President, Minnesota AFL-CIO, to FCC, MB Docket No. 14-90 (Sept. 16, 2014) (same); Letter from Mike Williams, President, Florida AFL-CIO, to The Honorable Thomas Wheeler, Chairman, FCC, MB Docket No. 14-90 (Oct. 13, 2014) (same); Letter from Bill Gerhard, President, Iowa Building and Construction Trades, to FCC, MB Docket No. 14-90 (Sept. 16, 2014) (same); Letter from Johanna Hester, President, and Gregory A. Cendana, Executive Director, Asian Pacific American Labor Alliance, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (Oct. 15, 2014) (same); Letter from Milton Rosado, President, Labor Council for Latin American Advancement, to The Honorable Thomas Wheeler, Chairman, FCC, MB Docket No. 14-90 (Oct. 16, 2014) (same). 1142 See Comcast-NBCU Order, 26 FCC Rcd at 4329, ¶ 223; Applications of AT&T Inc. and Deutsche Telekom AG for Consent to Assign or Transfer Control of Licenses and Authorizations, WT Docket No. 11-65, Order, DA 11- 1955, 26 FCC Rcd 16184, 16293, ¶ 259 (WCB 2011). 1143 Verizon Wireless-SpectrumCo Order, 27 FCC Rcd at 10739-40, ¶ 111; AT&T-Verizon Wireless Order, 25 FCC Rcd at 8717-18, ¶ 25; AT&T-Centennial Order, 24 FCC Rcd at 13929, ¶ 30. With respect to remedying harms, the Commission has held that it will impose conditions only to remedy harms that arise from the transaction (i.e., transaction-specific harms) and that are related to the Commission’s responsibilities under the Communications Act and related statutes. AT&T-Verizon Wireless Order, 25 FCC Rcd at 8747, ¶ 101. Federal Communications Commission FCC 15-94 148 allow us to find that the proposed transaction overall would be in the public interest. As discussed in detail above, we find that in light of the conditions, the public interest benefits of the proposed transaction outweigh the likelihood of significant public interest harms, such that overall, the proposed transaction is in the public interest. B. Fiber to the Premises Deployment Commitment 394. We find that the transaction reduces the combined entity’s incentive to deploy FTTP service because the resulting increase in U-verse video or other online video subscriptions will depress demand for DIRECTV’s services and thus “cannibalize” the combined entity’s profits. We therefore adopt as a condition of this transaction the Applicants’ commitment to expand FTTP broadband deployment. This condition is intended to capture all of AT&T’s pre-transaction planned deployment, projected deployment absent the transaction, and any additional deployment that the record suggests is profitable as a result of the transaction. This condition also provides an opportunity for increased competition from services that rely on fixed broadband Internet access service to deliver video by creating more customer locations that can receive broadband service at speeds capable of delivering video service. By creating pathways for alternative video distribution methods, this condition also helps to mitigate the harm of the loss of a video competitor in areas where AT&T and DIRECTV had directly competed prior to the transaction. Accordingly, as a condition of this transaction, AT&T will deploy its highest-speed fiber connections (U-verse FTTP) to at least 12.5 million more customer locations within four years of the transaction closing. In addition, to ensure that schools and libraries also benefit from the increased competition that will result from this fiber deployment, AT&T will offer gigabit service to any E-rate eligible school or library located within, or contiguous to, a distribution area in which AT&T deploys FTTP service. This commitment is estimated to include at least 6,000 E-rate eligible schools and libraries. C. Non-Discriminatory Usage-Based Practices 395. We find that the transaction increases the risk that the combined entity will use its fixed broadband Internet access service to engage in practices, such as discriminatory usage-based allowances, that favor its owned or affiliated online video content and online video services over competing online video content and OVDs. As AT&T currently imposes usage-based allowances on its broadband customers, more broadly than other large ISPs,1144 we conclude that a condition is necessary to address any increased incentive AT&T will have to use these practices to hinder the development of third-party OVDs as a competitive option to its own video offerings. Accordingly, as a condition of this transaction, we require the combined entity to refrain from discriminatory usage-based allowance practices for its fixed broadband Internet access service. D. Internet Interconnection Disclosure Requirement 396. We find that the transaction increases the risk that the combined entity will use interconnection agreements to limit competing online video content and OVDs. AT&T’s recent interconnection agreements, submitted in the record of this proceeding,1145 have greatly advanced our understanding of the current marketplace and the developments that may affect broadband providers’ interconnection practices. Therefore, we conclude that future disclosures of such agreements to the Commission are necessary to address the increased risk of anticompetitive practices by the combined entity. Accordingly, as a condition of this transaction, we require that the combined entity disclose all of its interconnection agreements to the Commission for four years after closing. This condition will enable the monitoring of the combined entity’s future interconnection agreements’ terms to determine whether the combined entity is using such agreements to deny or impede access to its networks in ways that limit competition from third-party online video content providers. In addition, this condition requires the 1144 Conditions Ex Parte Presentation at 6. 1145 See supra ¶ 218. Federal Communications Commission FCC 15-94 149 combined entity to work with an independent measurement expert to report certain Internet interconnection performance metrics, and to the extent possible, make such metrics publicly available. E. Discounted Broadband Services for Low-Income Subscribers 397. While we find that the availability of better and lower priced bundles of video and broadband service is a potential benefit of the merger, we also conclude that the public interest requires us to ensure that a bundle of video and broadband services is not the only competitive choice for low-income subscribers who may not be able to afford bundled services. Accordingly, we will require as a condition of this transaction that the combined entity make available an affordable, low-price standalone broadband service to low-income consumers in the combined entity’s wireline footprint. F. Reporting and Outside Compliance Officer 398. Some commenters contend that AT&T has failed to comply with voluntary commitments it has made in previous transaction proceedings or with conditions imposed by the Commission.1146 AT&T disputes these claims.1147 Given the important role that these conditions serve in securing the public interest benefits of this transaction, we find that compliance with the conditions must be ensured. Accordingly, to ensure that AT&T complies with the conditions of this Order, we require that AT&T retain both an internal company compliance officer and an independent, external compliance officer that will report and monitor, respectively, the combined entity’s compliance in accordance with the terms of this Order. Enforcement responsibilities remain the sole province of the Commission. XIII. BALANCING POTENTIAL PUBLIC INTEREST HARMS AND BENEFITS 399. After careful examination and analysis, we find that, while the transaction has the potential to cause some competitive harm, the economic and documentary evidence submitted to the Commission supports our conclusion that the transaction will result in greater competition for bundles of video and broadband and that this increased competition will benefit consumers, thus serving the public interest. We acknowledge that in certain overlap areas there will be a loss of an independent competitor, but we conclude that the effect of that reduction does not outweigh the benefits of a stronger combined competitor. Thus, on balance, we find that considering the imposition of the above-recited AT&T’s conditions, in conjunction with the public interest benefits that we find will likely arise from the transaction, there is sufficient evidence on this record for us to conclude that the Applicants have met their burden of demonstrating that the likely public interest benefits outweigh the likely public interest harms, such that we are able to approve the proposed transaction. XIV. CONCLUSION 400. We have reviewed the proposed transaction, the Application of AT&T and DIRECTV, and related pleadings and other submissions. We conclude that the Applicants are fully qualified and that the public interest benefits promised by the proposed transaction are sufficient to support the grant of the Application, pursuant to the public interest balancing test of Section 310(d) of the Act, subject to the conditions specified in Appendix B. XV. ORDERING CLAUSES 401. Accordingly, having reviewed the Application and the record in this matter, IT IS ORDERED, pursuant to Sections 4(i) and (j), 303(r), 214, 309, and 310(d) of the Communications Act of 1934, as amended, 47 U.S.C. §§ 154(i), 154(j), 303(r), 214, 309, 310(d), and Section 25.119 of the Commission’s rules, 47 C.F.R. § 25.119, that the Application for Consent to Assign or Transfer Control 1146 See Free Press Petition at 28-29, 32-33; Franken Comments at 8 n.32; see also Public Knowledge-ILSR Petition at 14-16 (stating that it is difficult to pin down the exact nature of AT&T’s compliance with previous merger commitments). 1147 See Joint Opposition at 23 n.67; AT&T Response to Sept. 9, 2014, Information Request at 222-223. Federal Communications Commission FCC 15-94 150 of various Commission licenses and authorizations from DIRECTV to AT&T Inc. IS GRANTED to the extent specified in this Memorandum Opinion and Order and subject to the conditions and commitments specified herein, including Appendix B. 402. IT IS FURTHER ORDERED that the above grant shall include authority for AT&T, consistent with the terms of this Memorandum Opinion and Order, to acquire control of: (a) any DIRECTV licenses and authorizations that may have been inadvertently omitted from the Application; (b) any licenses and authorizations issued to DIRECTV or its subsidiaries during the Commission’s consideration of the Application or during the period required for consummation of the transaction following approval; and (c) any applications that have been filed by DIRECTV or its subsidiaries and that are pending at the time of consummation. 403. IT IS FURTHER ORDERED that the conditions and commitments incorporated herein shall continue to apply until the conditions expire by their own terms as expressly stated, or the Commission determines that the conditions or commitments should be modified or removed. 404. IT IS FURTHER ORDERED, pursuant to Sections 4(i) and (j), 309, and 310(d) of the Communications Act of 1934, as amended, 47 U.S.C. §§ 154(i), 154(j), 309, 310(d), that the Petitions to Deny filed by Alliance for Community Media, the Alliance for Communications Democracy, and Common Cause; Cox Communications, Inc.; DISH Network Corporation; The Greenlining Institute; Free Press; Public Knowledge and Institute for Local Self-Reliance; Writers Guild of America, West, Inc. and all similar petitions ARE DENIED. 405. IT IS FURTHER ORDERED, pursuant to Sections 4(i) and (j), 309, and 310(d) of the Communications Act of 1934, as amended, 47 U.S.C. §§ 154(i), 154(j), 309, 310(d), that the requests that the Application be denied or held in abeyance pending the completion of other proceedings or investigation of allegations raised in the Petition for Investigation filed by New Networks Institute & Teletruth, the Comments filed by Minority Cellular Partners Coalition, and the Informal Objection and Request to Hold Applications in Abeyance filed by Northwest Broadcasting, L.P., et al., ARE DENIED; that all similar petitions and requests ARE DENIED; and the Petition for Investigation filed by New Networks Institute & Teletruth IS DISMISSED. 406. IT IS FURTHER ORDERED that this Memorandum Opinion and Order SHALL BE EFFECTIVE upon release, in accordance with Section 1.103 of the Commission’s rules, 47 C.F.R. § 1.103. FEDERAL COMMUNICATIONS COMMISSION Marlene H. Dortch Secretary Federal Communications Commission FCC 15-94 151 APPENDIX A List of Licenses to be Transferred Part 25 – SATELLITE COMMUNICATIONS LICENSES Satellite Space Station Licenses File No. Licensee Name/Call Sign Service Authorized SAT-T/C-20140611-00060 DIRECTV Enterprises, LLC DIRECTV 4S (S2430) DBS DIRECTV 7S (S2455) DBS DIRECTV 8 (S2632) DBS DIRECTV 9S (S2669) DBS DIRECTV 5 (S2673) DBS SAT-T/C-20140611-00061 DIRECTV Enterprises, LLC DIRECTV 8 (S2132) FSS SPACEWAY 2 (S2133) FSS SPACEWAY 1 (S2191) FSS DIRECTV 11 (S2640) FSS DIRECTV 10 (S2641) FSS DIRECTV 9S (S2689) FSS DIRECTV RB-1/14 (S2711, S2869) 17/24 GHz BSS and FSS DIRECTV RB-2/15 (S2712) 17/24 GHz BSS DIRECTV RB-2A/12 (S2796, S2797) 17/24 GHz BSS and FSS DIRECTV KU-79W (S2861) FSS DIRECTV KU-76W (S2888) FSS Satellite Space Station Licenses Granted During the Pendency of the Application File No. Licensee Name/Call Sign Service Authorized SAT-LOA-20130205-00016 DIRECTV Enterprises, LLC DIRECTV KU-45W (S2893) FSS SAT-LOA-20140825-00094 DIRECTV Enterprises, LLC DIRECTV 15 (S2930) FSS Transmit/Receive Earth Station Licenses File No. Licensee Call Sign SES-T/C-20140611-00505 DIRECTV Enterprises, LLC E020172 E020241 E020242 E030105 E030117 E050112 E050113 E050121 E050122 Federal Communications Commission FCC 15-94 152 E050255 E060014 E060236 E060441 E930229 E930304 E930485 E950349 SES-T/C-20140611-00506 DIRECTV Enterprises, LLC E010129 E010130 E010237 E050229 E050230 E050286 E060187 E060188 E060298 E060299 E070002 E070023 E070027 E070073 E070074 E070111 E070122 E070123 E080025 E080026 E080027 E080028 E080056 E080057 E090024 E090025 E090068 E090069 E090076 E090107 E090173 E100079 E100080 E100119 E100120 E100121 E100122 E110004 E120108 E120109 E120110 Federal Communications Commission FCC 15-94 153 E120148 E130081 E930191 E980285 E990159 SES-T/C-20140611-00509 California Broadcast Center, LLC E020091 Transmit/Receive Earth Station Licenses Granted During the Pendency of the Application File No. Licensee Call Sign SES-LIC-20140617-00524 DIRECTV Enterprises, LLC E140057 SES-LIC-20141106-00844 DIRECTV Enterprises, LLC E140115 SES-LIC-20141112-00848 DIRECTV Enterprises, LLC E140116 SES-LIC-20141112-00849 DIRECTV Enterprises, LLC E140117 SES-LIC-20150408-00198 DIRECTV Enterprises, LLC E150029 Transmit-Only Earth Station Licenses File No. Licensee Call Sign SES-T/C-20140611-00510 DIRECTV Enterprises, LLC E050340 Receive-Only Earth Station Registrations File No. Registration Holder Call Sign SES-T/C-20140611-00507 DIRECTV Enterprises, LLC E040179 E040180 SES-T/C-20140611-00508 DIRECTV Enterprises, LLC E980170 E980341 PART 74, 87, 90, AND 101 – PRIVATE WIRELESS LICENSES ULS File No. Licensee Call Sign 0006302429 DIRECTV Enterprises, LLC WPTZ691 WPZC401 WQHM919 WQIU946 WQTE840 0006312383 DIRECTV Sports Net Rocky Mountain, LLC WQPB424 0006302465 The DIRECTV Group, Inc. 71TV Federal Communications Commission FCC 15-94 154 APPENDIX B Conditions I. INTRODUCTION To address the potential harms posed and confirm certain benefits offered by the transaction, the Company will be subject to certain conditions imposed by the Commission. II. DEFINITIONS “Company” means AT&T and DIRECTV, both individually and collectively, including the combined entity of AT&T and DIRECTV, as well as any successor-in-interest, affiliate or subsidiary directly or indirectly controlling, controlled by, or under common control with AT&T, DIRECTV, or the combined entity of AT&T and DIRECTV. “Broadband Internet Access Service” means a mass-market retail service by wire or radio that provides the capability to transmit data to and receive data from all or substantially all Internet endpoints, including any capabilities that are incidental to and enable the operation of the communications service, but excluding dial-up Internet access service. This term also encompasses any service that the Commission determines provides a functional equivalent of the service described in the previous sentence. “Closing Date” means the date on which the acquisition of DIRECTV by AT&T occurs. “Communications Laws” means collectively, the Communications Act of 1934, as amended, the Telecommunications Act of 1996, the Rules, and the adopted and released orders and decisions of the Commission. “Company Compliance Officer” means a senior corporate manager that is part of AT&T’s Chief Compliance Office who has the requisite corporate and organizational authority to discharge the Company’s duties with respect to the conditions specified in this Appendix B and has specific knowledge of the Company’s operations referred to in these conditions in addition to general knowledge of the Communications Laws necessary to discharge his or her duties under this Order prior to assuming his or her duties. “Customer Service Representatives” means personnel at the Company’s Fixed Broadband Internet Access Service call centers and the Company’s in-region retail locations employed to provide guidance to existing and potential customers about Fixed Broadband Internet Access Service sales packages. “Fiber to the Premises” or “FTTP” means the technology for providing Broadband Internet Access Service by running fiber optic cable directly from an Internet Service Provider (“ISP”) to a subscriber’s home or business location. “Fixed Broadband Internet Access Service” means a Broadband Internet Access Service that serves end users primarily at fixed endpoints using stationary equipment. Fixed Broadband Internet Access Service includes fixed wireless services (including fixed unlicensed wireless services) and fixed satellite services. “Geocodes” means the longitude and latitude of a specific location. “Independent Compliance Officer” means an independent third party who is selected in accordance with the selection process set forth herein and engaged by the Company at its own expense to perform the duties set forth herein including an evaluation of the adequacy of the Company’s compliance with the conditions specified in this Appendix B. “Implementing Employees” means all employees and agents, including but not limited to Customer Service Representatives, of the Company who perform, supervise, oversee, or manage the performance of duties that relate to the Company’s responsibilities under the conditions specified in this Appendix B. Federal Communications Commission FCC 15-94 155 “Internet Interconnection Points” means the facilities over which traffic is exchanged between the Company’s network that carries Broadband Internet Access Service traffic and (1) peering networks or (2) customers that purchase on-net only services to deliver traffic to and from the Company’s end users over the company’s network. “Rules” means the Commission’s regulations found in Title 47 of the Code of Federal Regulations. “Standalone” means a service that is offered to subscribers not bundled with other services. “Video Programming” means programming provided by, or generally considered comparable to programming provided by, a television broadcast station or cable network, regardless of the medium or method used for distribution, and includes but is not limited to: programming prescheduled by the programming provider (also known as scheduled programming or a linear feed); programming offered to viewers on an on-demand, point-to-point basis (also known as video on demand (“VOD”), pay per view (“PPV”) or transactional video on demand (“TVOD”)); short programming segments (also known as clips); programming that includes multiple video sources (also known as feeds, including camera angles); programming that includes video in different qualities or formats (including high-definition and 3D); and feature-length motion pictures that have been theatrically released. III. FIBER TO THE PREMISES (FTTP) DEPLOYMENT 1. Introduction. As noted by the Applicants, the transaction reduces the Company’s incentive to deploy FTTP service because the resulting increase in U-verse video subscriptions will depress demand for DIRECTV’s services, thus “cannibalizing” its profits.1 The purpose of this condition is to address this competitive harm. This condition also provides an opportunity for increased competition from services that rely on Fixed Broadband Internet Access Service to deliver video by creating more customer locations that can receive broadband service at speeds capable of delivering video service. In addition, to ensure that schools and libraries also benefit from the FTTP deployment and the increased competition that will result from the FTTP deployment required by this condition, we also require the Company to offer Gigabit FTTP service to E-rate eligible schools and libraries within and contiguous to the distribution areas where the Company deploys FTTP service. 2. Condition. a. Within four (4) years, in accordance with the timing requirements set forth in subparts 2.a.(i) through 2.a.(v), the Company shall deploy FTTP-based Broadband Internet Access Service to at least 12.5 million mass-market customer locations,2 such as those occupied by residences, home offices, and very small businesses (and excluding locations solely occupied by large enterprises and institutions), of which no more than 2.9 million may be upgrades to customer locations that receive speeds of 45 Mbps or more using fiber to the node (“FTTN”) technology: (i) By December 31, 2015, the Company shall expand its FTTP coverage to at least 1.6 million of the aforementioned customer locations, including locations built as of April 15, 2015; 1 Application, “An Economic Assessment of AT&T’s Proposed Acquisition of DIRECTV,” Declaration of Michael L. Katz, transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, ¶¶ 126, 129-130 & n.222 (filed June 11, 2014). 2 Customer locations are defined as addresses to which the Company has the technical ability to provide Broadband Internet Access Service and excluding broadband-connected locations such as gates, ATMs, and elevators. Federal Communications Commission FCC 15-94 156 (ii) By December 31, 2016, the Company shall expand its FTTP coverage to at least 2.6 million of the aforementioned customer locations; (iii) By December 31, 2017, the Company shall expand its FTTP coverage to at least 5.0 million of the aforementioned customer locations; (iv) By December 31, 2018, the Company shall expand its FTTP coverage to at least 8.3 million of the aforementioned customer locations; and (v) Within four (4) years of the Closing Date the Company will complete the aforementioned FTTP deployment to all 12.5 million customer locations and the Company will offer speeds of 45 Mbps or more to at least 25.7 million customer locations. b. No more than 1.5 million greenfield locations (i.e., locations at which wire or fiber lines have not been deployed previously) may be counted towards the 12.5 million customer locations required in subsection 2.a. c. The Company may not use, receive, or request any Connect America Funds (“CAF”) for the investments required to satisfy the 12.5 million FTTP deployment transaction commitment or for operating expenses for such locations after such are deployed. Specifically, 12.5 million geocoded locations reported for purposes of this condition cannot be counted towards satisfying any CAF requirements.3 d. In addition to the 12.5 million FTTP locations required by this condition, the Company is obliged to offer 1 Gbps FTTP Service (“Gigabit FTTP Service”) to any E-rate eligible school or library located within or contiguous to a distribution area in which the Company deploys FTTP-based service, including all of the distribution areas included with the 12.5 million FTTP buildout, which includes approximately 6,000 E-rate eligible schools and libraries (“covered schools and libraries”). Provided however, the Company is not obliged to deploy Gigabit FTTP Service to schools and libraries outside of its wireline footprint. In order to satisfy this condition, the Company must offer Gigabit FTTP Service in response to a Form 470 seeking bids for Gigabit FTTP Service to any covered school or library, pursuant to the E-rate rules, and it must engage in affirmative and adequate outreach to make all covered schools and libraries aware of the opportunity to purchase its Gigabit FTTP Services. The Company shall make adjustments to its outreach efforts in response to reasonable requests from the Commission’s Office of General Counsel. 3 This would include but is not limited to any of the Connect America Fund (“CAF”) programs, as well as any other Universal Service Fund (“USF”) programs that the Commission may implement at a future date. Federal Communications Commission FCC 15-94 157 3. Reporting. a. The Company shall submit a report in accordance with the filing and service requirements set forth in Section VII.5. herein on a semi-annual basis that describes its compliance with subsections 2.a.-c. of this condition, with the first such report to be submitted six (6) months after the Closing Date, in a format similar to the report that the Company submits in connection with receiving CAF Phase I support, and is expected to submit in Phase II, which must include at least the following, in electronic format: (i) The number of new customer locations to which FTTP service has been deployed during the reporting period; (ii) A CSV file (comma separated values file) or other form approved by the Commission staff for each location to which FTTP service has been deployed in satisfaction of this deployment condition, information presented in substantially the format shown in the chart below (the same location information collected from CAF Phase I recipients);4 (iii) Any explanatory notes as required; and (iv) Any other information the Independent Compliance Officer determines is reasonably necessary to report on compliance with this condition. 4 The Company must provide LAT and LONG; however, if information for a column other than LAT/LONG is unavailable for a given location, that column may be left blank. Federal Communications Commission FCC 15-94 158 Template for FTTP Deployment Reports Field Name Description Type Example GEOID10 2010 Census Block FIPS code of the location Text "511650120001030" OCN NECA-Assigned Company Code for the operating company serving the location Text "0233" SAC Study Area Code of the location Text "190233" CLLI Telcordia-specified eight-character Common Language Location Identifier (“CLLI”) code for the ILEC wire center of the location Text "AAPKOKXA" LAT Latitude of the location to 6 decimal places Float 38.274538 LONG Longitude of the location to 6 decimal places Float -78.688942 b. The Company shall submit a report in accordance with the filing and service requirements set forth in Section VII.5. herein on an semi-annual basis that describes its compliance with subsection 2.d. of this condition, with the first such report to be submitted six (6) months after the Closing Date, which must include at least the following, in electronic format:  (i) A list of the covered schools and libraries to which the Company has provided a bid for FTTP services pursuant to this condition, the FCC Form 470 associated with each such bid, and information about the monthly recurring charges and any special construction charges associated with each such bid; (ii) The number of covered schools and libraries to which Gigabit FTTP Service has been deployed during the reporting period; (iii) A CSV file (comma separated values file) or other form approved by the Commission staff for each school or library location to which Gigabit FTTP Service has been deployed in satisfaction of this deployment condition, information presented in substantially the format shown in the chart below;5 (iv) Any explanatory notes as required; (v) A description of the Company’s outreach to covered schools and libraries to notify them of the availability of Gigabit FTTP Service; and (vi) Any other information the Independent Compliance Officer determines is reasonably necessary to report on compliance with this condition. 5 The Company must provide LAT and LONG; however, if information for a column other than LAT/LONG is unavailable for a given location, that column may be left blank. Federal Communications Commission FCC 15-94 159 Template for FTTP Schools and Libraries Deployment Reports Field Name Description Type Example GEOID10 2010 Census Block FIPS code of the location Text "511650120001030" OCN NECA-Assigned Company Code for the operating company serving the location Text "0233" SAC Study Area Code of the location Text "190233" CLLI Telcordia-specified eight-character Common Language Location Identifier (“CLLI”) code for the ILEC wire center of the location Text "AAPKOKXA" LAT Latitude of the location to 6 decimal places Float 38.274538 LONG Longitude of the location to 6 decimal places Float -78.688942 NAME Name of School or Library Text Berkeley High School DISTRICT NAME Name of School District or Library System Text Berkeley Unified School District MONTHLY CHARGE Monthly Recurring Charge for the Gigabit FTTP Service Text $ [Amount] SPECIAL CONSTRUCTION Non-Recurring Special Construction Charge for Gigabit FTTP Service Text $ [Amount] 4. Enforcement. In addition to the enforcement actions described in this Order, any material failure to comply with the conditions identified in subsections 2.a.-c. of this condition will result in extension of all of the conditions specified in this Appendix B until completion of the required buildout. Federal Communications Commission FCC 15-94 160 IV. NON-DISCRIMINATORY USAGE-BASED PRACTICES 1. Introduction. As noted by the Applicants, a benefit of this transaction is the improved ability to develop and launch the Company’s video offerings provided to customers, including through Fixed Broadband Internet Access Service. At the same time, as a result of the transaction, the Company has an increased incentive to discriminate against unaffiliated online video distribution services. Following the transaction, the Company will have additional incentives to use strategies that limit consumers’ access to online video distribution services in order to favor the U-verse or DIRECTV Video Programming product or the combined entity’s online Video Programming products. Further, the Company will have an increased incentive to limit subscriber demand for competitors’ online video content, including through discriminatory usage-based allowances, commonly known as “data caps.”6 The purpose of this condition is to address the incentive and ability to use such practices to discriminate against Video Programming services that provide content to customers through Fixed Broadband Internet Access Services by eliminating the risk that the Company will use its Fixed Broadband Internet Access Service to engage in practices, such as discriminatory usage-based allowances, that favor its owned or affiliated video content or video distribution services, which include a Company-operated online video distribution service and a Company “TV Everywhere” service (whether operated by U-verse or DIRECTV or the equivalent). 2. Condition. In the application of usage-based allowances or other retail terms and conditions for its Fixed Broadband Internet Access Service, the Company shall not discriminate in favor of its own Video Programming services, including a Company-operated online Video Programming service or any Company “TV Everywhere” service (whether operated by AT&T’s U-verse service, DIRECTV, or the equivalent), or any content or application available through its own Video Programming services, including through the exemption of one or more of its own Video Programming services from usage-based allowances. For the avoidance of doubt and consistent with such prohibition, this condition does not prohibit the Company from offering discounts for integrated bundles of the Company’s U-verse or DIRECTV satellite Video Programming service or rebranded offerings of these services with the Company’s Fixed Broadband Internet Access Services. 3. Reporting. The Company shall submit a report in accordance with the filing and service requirements set forth in Section VII.5. herein on a semi-annual basis that details its compliance with this condition, with the first such report to be submitted six (6) months after the Closing Date, which will include at least the following: a description of all terms and conditions associated with its usage-based allowances and any other information the Independent Compliance Officer determines is reasonably necessary to report as required by this condition. 6 AT&T currently imposes on its wireline broadband customers usage-based data allowances that provide for overage charges if customers exceed applicable allowances. Federal Communications Commission FCC 15-94 161 V. INTERNET INTERCONNECTION DISCLOSURE REQUIREMENTS 1. Introduction. We find that AT&T’s recently entered Internet interconnection agreements submitted in this proceeding advance our understanding of the current marketplace and developments that may affect broadband providers’ interconnection practices. We also find that the transaction increases the risk that the Company will use such agreements to limit competing online video content or online video distribution services. The purpose of this condition is to monitor the terms of the Company’s future Internet interconnection agreements to determine whether the Company is using such agreements to deny or impede access to the Company’s networks in ways that limit competition from other online video content or online video distribution services. 2. Condition. Commencing on the Closing Date and ending on the fourth anniversary of that date, absent any extension under the terms of this Appendix B, the Company shall comply with the following conditions: a. Disclosure of Internet Interconnection Agreements. Within thirty (30) days of the execution of any agreement subject to this condition, in accordance with the filing and service requirements set forth in Section VII.5. herein, the Company shall submit all agreements entered during this period for the exchange of traffic at Internet Interconnection Points located within the United States, unless the volume of traffic exchanged with the interconnecting party is less than a de minimis threshold, as specified by the Independent Measurement Expert, but no prior approval of such agreement by the Commission under any circumstances is herein required. b. Reporting Internet Interconnection Performance Metrics. Using a methodology developed and implemented as subpart 2.c., the Company must report, in accordance with the filing and service requirements set forth in Section VII.5. herein, on a schedule established by an Independent Measurement Expert, but no more than on a monthly basis, the following performance characteristics of traffic exchanged at Internet Interconnection Points located within the United States, unless the volume of traffic exchanged with the interconnecting party is less than a de minimis threshold, as specified by the Independent Measurement Expert: (i) The probability distribution of latency between the border router of the interconnecting network and the Company’s border router (“Latency”), as defined by the Independent Measurement Expert; (ii) The percentage of packets dropped at or between the border router of the interconnecting network and the Company’s border router (“Packet Loss”), as defined by the Independent Measurement Expert; and (iii) The percent usage of each Internet Interconnection Point (“Utilization”), as defined by the Independent Measurement Expert. c. Independent Measurement Expert. (i) Within sixty (60) days of the Closing Date, an Independent Measurement Expert shall be identified, whose selection is acceptable to the Company and approved by the Commission’s Office of General Counsel, in consultation with the Wireline Competition Bureau and the Chief Technologist. If the Company and the Commission’s Office of General Counsel do not agree on the selection of an Independent Measurement Expert within sixty (60) days, then the Commission’s Federal Communications Commission FCC 15-94 162 Office of General Counsel, in consultation with the Wireline Competition Bureau and the Chief Technologist, shall select the Independent Measurement Expert. (ii) The Independent Measurement Expert must be a neutral third-party organization, including with respect to sources of funding, and must have demonstrated expertise in the measurement, collection, and analysis of Internet performance data, including network performance characteristics related to Internet traffic exchange. (iii) The Company shall engage the Independent Measurement Expert at its own expense to perform the duties set forth in this condition, including the development and implementation of a methodology to measure and disclose the interconnection performance metrics set forth in subpart 2.b. of this condition. The terms of the engagement shall be subject to approval by the Commission’s Office of General Counsel, in consultation with the Wireline Competition Bureau and the Chief Technologist. (iv) Within sixty (60) days of engaging the Independent Measurement Expert, the Company, in consultation with the Independent Measurement Expert, will submit for approval by the Commission’s Office of General Counsel, in consultation with the Wireline Competition Bureau and the Chief Technologist, a report describing the Independent Measurement Expert’s proposed methodology for the measurement of the performance metrics described herein. Such report shall also be submitted to the Independent Compliance Officer. The proposed methodology should, at a minimum, address the following criteria: 1) Identification of Internet Interconnection Points, including the identity of the interconnecting parties and the location and capacity of each interconnection point; 2) Identification of a disclosure exemption threshold for a de minimis volume of traffic exchanged between the Company and interconnecting parties; 3) A definition of “Latency,” which shall include the disclosure of the probability distribution; 4) A definition of “Packet Loss”; 5) Time of measurements, which shall, at a minimum, include an identified window within peak usage periods; 6) For any performance metric contingent upon an interconnecting party’s participation in the selected measurement methodology, a process for waiving the disclosure of that metric at points of interconnection where the interconnecting party declines to participate; 7) Frequency and duration of measurements; 8) Any devices used for measurement; 9) End points of measurements; 10) Placement of any devices; and 11) Frequency of disclosures. (v) Within one hundred and twenty (120) days following the Commission’s Office of General Counsel’s approval of the proposed methodology, the Company will Federal Communications Commission FCC 15-94 163 implement the methodology and begin reporting in accordance with this condition and the filing and service requirements set forth in Section VII.5. herein. Federal Communications Commission FCC 15-94 164 VI. DISCOUNTED BROADBAND SERVICES PROGRAM 1. Introduction. We find that the availability of better and lower priced bundles of video and broadband service is a potential benefit of the transaction. However, it also is in the public interest to ensure that a bundle of video and broadband services is not the consumer’s only competitive choice, and this protection may be particularly important for low-income subscribers who may not be able to afford bundled services. Thus, the purpose of this condition is to make available an affordable, low-price Standalone broadband service to low- income consumers in the Company’s wireline footprint. 2. Condition. Within nine (9) months of the Closing Date, the Company shall establish and commence a program to substantially increase broadband adoption in low-income households throughout AT&T’s wireline footprint (the “Discounted Broadband Services Program”). a. The Company shall offer wireline Broadband Internet Access Service with download speeds of at least 10 Mbps, where technically available, to qualifying households in the Company’s wireline footprint for no more than $10 per month. If 10 Mbps wireline Broadband Internet Access Service is not technically available, the Company shall offer wireline Broadband Internet Access Service with download speeds of at least 5 Mbps, where technically available, to qualifying households in the Company’s wireline footprint for no more than $10 per month. b. Where AT&T has deployed broadband service at top speeds below 5 Mbps, the Company shall offer wireline Broadband Internet Access Service at speeds of at least 3 Mbps, where technically available, to qualifying households in the Company’s wireline footprint for no more than $5 per month. c. Qualifying households are those where at least one individual participates in the Supplemental Nutrition Assistance Program (“SNAP”), subject to annual recertification, and that do not have outstanding debt for AT&T’s Fixed Broadband Internet Access Services that was incurred within the six (6) months prior to the individual’s request for services under the Discounted Broadband Services Program or that is incurred for services provided under the Discounted Broadband Services Program and that is subject to the Company’s ordinary debt collection procedures. d. The Company shall offer the discounts set forth in this condition for at least four (4) years from the commencement of the Discounted Broadband Services Program. Qualifying households who sign up for the Discounted Broadband Services Program in the fourth year of the Discounted Broadband Services Program shall remain eligible for at least twelve (12) months. e. Qualifying households shall not be required to pay any installation or modem charges or fees in order to participate in the Discounted Broadband Services Program. f. For the period during which this condition is in effect, the Company shall clearly and conspicuously market the Discounted Broadband Services Program, including but not limited to undertaking the following actions: (i) Providing on the Company’s consumer-facing homepage a link to a webpage devoted to describing the Discounted Broadband Services Program; and (ii) Ensuring that the Company’s Customer Service Representatives are trained prior to the commencement of the program to inform consumers of the availability of Federal Communications Commission FCC 15-94 165 the Discounted Broadband Services Program offerings, including pricing, and terms and conditions as described in this condition. g. The Company shall effectively engage in targeted outreach efforts, in coordination with schools and community-based organizations serving low-income individuals and families, including, but not limited to veterans, the elderly, and those who are non- English speaking, to adequately publicize the availability of the Discounted Broadband Services Program, to ensure that qualified individuals and households are informed about and have access to the program. The Company shall make adjustments to its outreach efforts in response to reasonable requests from the Commission’s Office of General Counsel and, at a minimum, shall take the following actions during each year that the program is in effect: (i) Promote the Discounted Broadband Services Program, including through public service announcements that shall have a minimum annual value of $15 million. (ii) Distribute Discounted Broadband Services Program information to at least twenty (20) organizations that work with low-income communities on a national and local level. (iii) Coordinate with state education departments and local school districts, including requesting that all school districts within the Company’s wireline footprint include information about the Discounted Broadband Services Program with their communications to families in advance of the school year, including in each communication relating to the National School Lunch Program (“NSLP”), as feasible and appropriate, to ensure that families that qualify for the NSLP are informed about the Discounted Broadband Services Program at the beginning of the school year and have the opportunity to register. (iv) Provide appropriate promotional and collateral materials to all public school districts within the Company’s wireline footprint and requesting that the materials be included in NSLP mailings. (v) Educate school professionals about the Discounted Broadband Services Program, including by conducting outreach to various education-related associations such as parent-teacher associations and associations representing guidance counselors and social workers, in order to reach those who are most likely to work closely with students and families. h. Prospective participants shall be directed to a Company phone number dedicated to the Discounted Broadband Services Program to verify eligibility. Qualifying callers shall be transferred to a centralized order-entry center. i. The Company shall submit a report in accordance with the filing and service requirements set forth in Section VII.5. herein on a semi-annual basis that includes a description of the Company’s compliance with the condition, with the first such report to be submitted six (6) months after the Closing Date. The report shall at least include the following: (i) The total number of households participating in the Discounted Broadband Services Program; (ii) A detailed description of outreach efforts made during the reporting period to publicize the Discounted Broadband Services Program to schools and community-based organizations, including a list of the community-based Federal Communications Commission FCC 15-94 166 organizations participating, and representative examples of the promotional and collateral materials provided; and (iii) An analysis of the effectiveness of the Discounted Broadband Services Program, describing any adjustments the Company has implemented during the reporting period or plans to implement to improve its effectiveness. Federal Communications Commission FCC 15-94 167 VII. COMPLIANCE PROGRAM AND REPORTING 1. Company Compliance Officer. Within thirty (30) calendar days after the Closing Date, the Company shall designate a senior corporate manager with the requisite corporate and organizational authority to serve as a Company Compliance Officer and to discharge the Company’s duties with respect to the conditions specified in this Appendix B. The person designated as the Company Compliance Officer shall be part of AT&T’s Chief Compliance Office. In addition to the general knowledge of the Communications Laws necessary to discharge his or her duties under this Order, the Compliance Officer shall have specific knowledge of the Company’s operations referred to in these conditions prior to assuming the duties required by this Appendix B. 2. Company Implementation and Compliance Plan. The Company agrees that it shall, within sixty (60) calendar days after the Closing Date, develop and implement an Implementation and Compliance Plan designed to ensure its implementation of and compliance with the conditions specified in this Appendix B, establishing, inter alia, mechanisms to provide, on an ongoing basis, adequate notice and training to all Company personnel involved with the activities covered by the conditions in this Appendix B. This Implementation and Compliance Plan shall be provided to the Independent Compliance Officer for review upon the Independent Compliance Officer’s selection. 3. Independent Compliance Officer. a. Within ninety (90) days of the Closing Date, an Independent Compliance Officer shall be identified, whose selection is acceptable to the Company and approved by the Commission’s Office of General Counsel, in consultation with the Wireline Competition Bureau. If the Company and the Commission’s Office of General Counsel do not agree on the selection of an Independent Compliance Officer within ninety (90) days, then the Commission’s Office of General Counsel, in consultation with the Wireline Competition Bureau, shall select the Independent Compliance Officer. b. The Company shall engage the Independent Compliance Officer at its own expense to perform the duties set forth herein, including an evaluation of the adequacy of the Company’s compliance with the conditions specified in this Appendix B, and shall designate the Commission as a third-party beneficiary to the engagement. The terms of the engagement shall be subject to approval by the Commission’s Office of General Counsel, in consultation with the Wireline Competition Bureau. c. The Independent Compliance Officer and any persons retained by the Independent Compliance Officer to effectuate this Appendix B may not (i) have had any business or familial relationships with the Company within the past five (5) years; (ii) have been employed by or affiliated with any competitor of the Company within the past two (2) years; (iii) have been an employee of the Commission within the past two (2) years; (iv) have submitted any comments or otherwise participated in this transaction proceeding or have been employed by or affiliated with any entity that has submitted any comments or otherwise participated in this transaction proceeding within the past two (2) years; or (v) have any conflict of interest related to the duties of the Independent Compliance Officer that could prevent him or her from performing his or her duties in a fair and unbiased manner. In addition, for a minimum of five (5) years after the end of the Independent Compliance Officer’s engagement, the Independent Compliance Officer shall not be employed by, or have any business relationship with, the Company. Federal Communications Commission FCC 15-94 168 d. The Independent Compliance Officer shall have the power and authority to review and evaluate the Company’s Implementation and Compliance Plan and any related materials, and recommend to the Company changes to address any perceived deficiencies in the Plan. Any such recommendations shall be included in the Independent Compliance Officer’s Compliance Reports. e. The Independent Compliance Officer shall prepare and submit, in accordance with the filing and service requirements set forth in Section VII.5. herein, a Compliance Report within sixty (60) days of receiving the Company’s reports required under the conditions specified in this Appendix B. Each such Compliance Report shall include a detailed description of the Company’s efforts during the relevant period to comply with the conditions and will specifically meet the reporting requirements for the conditions set forth in this Appendix B. The Independent Compliance Officer shall provide a final copy of all Compliance Reports to the Company’s Compliance Officer at least seven (7) days before the report is submitted to the Commission, so that the Company may prepare a request for confidential treatment if necessary. f. The Company shall have thirty (30) days from submission of the Compliance Report to the Commission to comment on and/or object to the Compliance Report and must submit such comments and/or objections in accordance with the filing and service requirements set forth in Section VII.5. herein. The Company’s comments and/or objections shall be accompanied by a statement explaining the basis for its response and shall comply with Section 1.16 of the Rules and be subscribed to as true under penalty of perjury in substantially the form set forth therein.7 g. If the Independent Compliance Officer in the exercise of his or her responsibilities discovers or receives evidence that suggests to the Independent Compliance Officer that the Company is materially violating or materially failing to comply with a condition specified in this Appendix B, the Independent Compliance Officer shall promptly provide that information to the Company and the Commission’s Office of General Counsel, Wireline Competition Bureau, and Enforcement Bureau. h. The Independent Compliance Officer shall not have the authority to direct the Company to make changes to the Implementation and Compliance Plan, the Company’s efforts to comply with the conditions specified in this Appendix B, or the Company’s business practices or policies. The Commission (and its Bureaus and Offices in their delegated authority) retains all rights to determine if a violation has occurred and to take whatever action it deems appropriate. The Independent Compliance Officer also shall not have the authority to participate in the business activities or management of the Company. i. The Company shall assist the Independent Compliance Officer in the performance of the duties of the Independent Compliance Officer set forth in this Order. The Company shall take no action to interfere with or to impede the Independent Compliance Officer’s accomplishment of his or her duties. The Independent Compliance Officer, and any person retained by the Independent Compliance Officer, may, in connection with the reasonable exercise of his or her responsibilities, subject to the Company’s privilege rights, on reasonable notice to the Company during normal business hours, and in coordination with the Company Compliance Officer: (i) Interview any Company personnel for any purpose reasonably related to the Independent Compliance Officer’s duties; any such interview will be subject to 7 47 C.F.R. § 1.16. Federal Communications Commission FCC 15-94 169 the reasonable convenience of such personnel and the Company will make such personnel available; (ii) Have such access to the facilities of the Company as is reasonably required by the Independent Compliance Officer’s duties; (iii) Have full and complete access to and inspect and copy any document, email, contract, and any other information in the possession, custody, or control of the Company reasonably related to the Independent Compliance Officer’s duties; and (iv) Require the Company to provide compilations of documents, data, or other information reasonably related to the Independent Compliance Officer’s duties, and to submit reports to the Independent Compliance Officer containing such material, in such form as the Independent Compliance Officer may reasonably direct. j. Any objections by the Company to actions by the Independent Compliance Officer must be conveyed in writing to the Commission’s Office of General Counsel, Wireline Competition Bureau, Enforcement Bureau, and to the Independent Compliance Officer within thirty (30) calendar days after the action giving rise to the objection or else such objection may be considered waived at the discretion of the Commission’s Office of General Counsel. Any such objections will be resolved by the Commission’s Office of General Counsel within thirty (30) days of the date that the Company files the objection. k. The Independent Compliance Officer may hire such persons as are reasonably necessary to fulfill the Independent Compliance Officer’s duties, with prior notice and subject to the approval of the Commission’s Office of General Counsel. The Independent Compliance Officer and any persons hired to assist the Independent Compliance Officer shall serve at the cost and expense of the Company, on such terms and conditions as the Commission’s Office of General Counsel approves, and shall be subject to the execution of customary confidentiality agreements. The compensation of the Independent Compliance Officer and any persons hired to assist the Independent Compliance Officer shall be on reasonable and customary terms commensurate with the individuals’ experience and responsibilities and consistent with reasonable expense practices. The Independent Compliance Officer shall submit a monthly expense report to the Company and the Commission’s Office of General Counsel, with the first such report to be submitted within thirty (30) days after the selection of the Independent Compliance Officer, describing the total amounts expended. l. The Commission’s Office of General Counsel may at any time require the Company to replace the Independent Compliance Officer with a substitute Independent Compliance Officer selected by the same selection process as used in the initial selection. m. The Company may not refuse to pay the Independent Compliance Officer without first receiving approval of the Commission’s Office of General Counsel. If the Company determines that the Independent Compliance Officer has ceased to act or failed to act diligently or in a cost-effective manner, it may submit a request to the Commission’s Office of General Counsel proposing corrective actions to be taken by the Independent Compliance Officer or the selection of a substitute Independent Compliance Officer. n. The Independent Compliance Officer’s engagement will continue as long as one or more of the conditions in this Appendix B are in effect. 4. Company Obligation to Report Noncompliance. The Company shall report, in accordance with the filing and service requirements set forth in Section VII.5. herein, any Federal Communications Commission FCC 15-94 170 material noncompliance with the conditions specified in this Appendix B within fifteen (15) calendar days after discovery of such noncompliance. Such reports shall include a detailed explanation of: (i) each instance of material noncompliance; (ii) the steps that the Company has taken or will take to remedy such noncompliance; (iii) the schedule on which such remedial actions will be taken; and (iv) the steps that the Company has taken or will take to prevent the recurrence of any such noncompliance. 5. Confidentiality and Filing and Service Requirements. Any and all materials submitted by any party pursuant to the conditions specified in this Appendix B shall be subject to the Protective Orders already issued in this proceeding unless and until such Orders are modified or replaced, at which point the modified or replacement provisions shall apply. In addition, all such materials, unless otherwise provided in this Appendix B, shall be filed in this docket with the Commission’s Secretary’s Office with an electronic copy submitted via email to the addressees listed below. a. Wireline Competition Bureau: Christopher Sova (or his successor) With a copy submitted electronically to Christopher.Sova@fcc.gov. b. Office of General Counsel: Jamillia Ferris (or her successor) With a copy submitted electronically to Jamillia.Ferris@fcc.gov. c. Enforcement Bureau: Chief, Investigations and Hearings Division (or his successor) Enforcement Bureau With a copy submitted electronically to Jeffrey.Gee@fcc.gov or his successor. d. Independent Compliance Officer: To be selected. Federal Communications Commission FCC 15-94 171 VIII. ENFORCEMENT Any material failure to comply with the requirements identified in this Appendix B may result in: 1. An appropriate forfeiture penalty under applicable law; and/or 2. Any other appropriate remedies allowed under the Communications Laws, including, but not limited to, an award of damages for the benefit of consumers for any harm incurred, issuance of cease-and-desist orders, modification of the conditions, and issuance of an order requiring appropriate remedial action. The enforcement and compliance programs established in these conditions are intended to supplement the Commission’s usual enforcement and investigative powers, which remain fully applicable, and do not replace such powers. IX. VIOLATIONS Any violation of these conditions shall be a violation of the Order. X. TERM These conditions shall remain in effect for four (4) years beginning on the Closing Date, except as otherwise stated in this Appendix B, and if the Commission makes a formal finding that the Company has violated any conditions, in whole or in part, in this Appendix B, the Commission shall have authority to extend the terms of such conditions for two (2) additional years on its own motion. Federal Communications Commission FCC 15-94 172 APPENDIX C Analysis of Merger Simulation Models TABLE OF CONTENTS Heading Paragraph # I.  INTRODUCTION .................................................................................................................................. 1  II.  THE SIMULATION MODEL ............................................................................................................... 6  A.  Geographic Definition and Product Terminology ........................................................................... 9  B.  Deriving Shares and a Measure of Price from Data ....................................................................... 12  1.  Generating Shares .................................................................................................................... 13  2.  Creating a Simple Price Measure ............................................................................................ 17  C.  Modeling and Estimating Demand ................................................................................................ 22  1.  Modeling Demand ................................................................................................................... 22  2.  Estimating Demand Parameters .............................................................................................. 27  D.  Modeling Pricing and Determining Marginal Cost ........................................................................ 29  E.  Solving for Post-Merger Prices ...................................................................................................... 32  III. ECONOMIC EFFECTS CAPTURED BY THE SIMULATION MODEL ......................................... 33  A.  Horizontal Overlap......................................................................................................................... 34  B.  Programming Payment Reductions ................................................................................................ 36  C.  Bundling Effects ............................................................................................................................ 38  IV. MODEL CORRECTIONS AND ADJUSTMENTS ............................................................................ 42  A.  Corrections to Price Recentering Procedure .................................................................................. 45  B.  Adjustments to Third-Party Prices ................................................................................................. 50  C.  Programming Payment Reductions ................................................................................................ 55  V.  RESULTS ............................................................................................................................................. 78  A.  Consumer Surplus Effects .............................................................................................................. 79  B.  Price and Share Effects .................................................................................................................. 85  C.  DMA Specific Effects .................................................................................................................... 95  1.  Market-Level Welfare Effects ................................................................................................. 97  2.  U-verse Household Penetration and Market Outcomes ........................................................ 101  3.  Maps of Main Simulation Results ......................................................................................... 104  D.  Robustness: Outliers and Price Winsorization ............................................................................ 108  E.  Robustness: Appropriate Setting of the Pre-Merger Synthetic Bundle Discount ....................... 111  VI. COMPARISON WITH OTHER STUDIES ....................................................................................... 113  A.  Studies of the U.S. Broadband Internet Service Market .............................................................. 115  B.  Studies of the U.S. MVPD Market .............................................................................................. 121  VII. IMPLICATIONS ............................................................................................................................... 124  I. INTRODUCTION 1. This Appendix describes and presents our analysis of the merger simulation model (“BH Simulation”) that was submitted by Professors Steve Berry and Phil Haile on behalf of AT&T Inc. (“AT&T”) to evaluate the potential for unilateral effects that may arise as a result of its proposed acquisition of DIRECTV. (We refer to the two entities, collectively, as the “Applicants.”) 2. Unilateral effects, in this context, concern the incentive for AT&T to impose post- transaction price increases, particularly for video service. While the Applicants acknowledge the Federal Communications Commission FCC 15-94 173 potential for these effects,1 they claim that video and broadband services (the former of which are offered by each of the Applicants – though DIRECTV’s footprint and subscriber base are much more extensive than AT&T’s – and the latter of which are offered only by AT&T) are complements2 and, therefore, that the transaction would result in downward pricing pressure on the bundle that combines DIRECTV’s video service with AT&T’s broadband service.3 The Applicants argue that these marketplace-specific factors need to be weighed against any potential competitive harms that would arise from the transaction.4 3. In support of the above claims, the Applicants submitted two merger simulation models. The initial submission included a merger simulation model submitted by Dr. Michael Katz, which is described in the Katz Declaration.5 Subsequent submissions were based on the BH Simulation submitted by Professors Berry and Haile.6 The Applicants represented the two models as closely related.7 After examining both models, we agreed with this assessment and focused on the BH Simulation, which is the more detailed and economically rigorous variant of the commonly adopted modeling approach. We find that the BH Simulation provides a good starting point to analyze potential unilateral effects. As discussed in detail below, we find that, after some corrections and incorporation of additional data, including information on potential programming payment reductions, the quantitative prediction of the model is that the transaction is likely to produce modest benefits to consumers through downward pricing pressure on the bundle that combines DIRECTV’s video service with AT&T’s broadband service, which, in turn, puts downward pricing pressure on bundles provided by cable companies. The programming payment reductions produce further consumer welfare gains, as those reductions will independently exert downward pressure on the price of AT&T’s video service. 1 See Steve Berry & Phil Haile, “Quantitative Analysis of an AT&T-DIRECTV Merger,” transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 4 (filed July 17, 2014) (“Berry-Haile Analysis”). 2 See id. 3 See id. at 4-7. 4 See Application, “An Economic Assessment of AT&T’s Proposed Acquisition of DIRECTV,” Declaration of Michael L. Katz, transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, ¶ 4 (filed June 11, 2014) (“Katz Decl.”); Berry-Haile Analysis at 9-10; Steve Berry & Phil Haile, “Quantitative Analysis of an AT&T-DIRECTV Merger: Additional Discussion of Modeling Choices, Data, and Results,” transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 5 (filed Sept. 23, 2014) (“Berry-Haile Additional Discussion”). 5 See Katz Decl. ¶¶ 85-95, Appendix I; Joint Opposition, “An Economic Assessment of AT&T’s Proposed Acquisition of DIRECTV,” Reply Declaration of Michael L. Katz, transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, ¶¶ 52, 55 (filed Oct. 16, 2014) (“Katz Reply Decl.”); Compass Lexecon, “Additional Detail on the Demand Estimation, Merger Simulation, and Investment Model Analysis Performed by Professor Katz,” transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, ¶¶ 5-46 (filed July 28, 2014) (“Katz Additional Detail”). 6 See Berry-Haile Analysis; Steve Berry & Phil Haile, “Quantitative Analysis of an AT&T-DIRECTV Merger: Updated Results,” transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (filed Sept. 23, 2014) (“Berry-Haile Updated Results:); Berry-Haile Additional Discussion. 7 See Joint Opposition of AT&T and DIRECTV to Petitions to Deny and Condition and Reply to Comments, transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 30-32 (filed Oct. 16, 2014) (“Joint Opposition”); Katz Reply Decl. ¶¶ 2, 56-58. Federal Communications Commission FCC 15-94 174 4. It is important to note that the question posed in any merger simulation is: “Assuming that all industry participants’ product offerings remain the same, what price changes arise from the changed pricing incentives created by the proposed transaction?” The simulation predicts the transaction’s price effects, holding constant the industry product mix. The only dimension of adjustment that the BH Simulation allows is in the prices at which industry participants offer their products. In particular, while the BH Simulation captures the change in the combined firm’s pricing incentives as a result of the merger – and thus allows the price of the bundle that combines DIRECTV’s video programming service with AT&T’s broadband Internet access service to change endogenously – it does not assume that consumers will obtain any additional exogenous benefit from the post-merger integration of that (currently synthetic) bundle.8 The simulation also does not address issues involving post-merger anticompetitive incentives that AT&T and DIRECTV may have to reduce the ability of rivals to compete effectively. Rather, the BH Simulation provides a focused lens through which to consider changing pricing incentives created by the merger. 5. This Appendix proceeds as follows: In Section II, we describe the BH Simulation in detail. Section III discusses the economic effects that the model captures. In Section IV, we discuss adjustments that we made to the BH Simulation to address shortcomings of the original model. Section V presents the results of our analysis. Section VI provides a comparison of the results with results from previous studies in the economics literature. Finally, Section VII concludes. II. THE SIMULATION MODEL 6. The BH Simulation is a sophisticated application of merger simulation techniques that are well accepted in the economic literature.9 The underlying structure of the BH Simulation builds on the discrete choice modeling approach to demand estimation combined with assumptions about the structure of a firm’s pricing decisions to arrive at a computationally feasible way to derive post-merger price predictions.10 7. Given the inevitable constraints imposed on parties by limited data and the current state of the art in economic modeling, the submitted model is a very fine example of a merger simulation. This assessment is reflected in our adoption of the framework in reviewing the transaction. While we differ from the Applicants in the interpretation of certain simulation results and a few of the assumptions and 8 Because the BH Simulation ignores these possible benefits (which might include one-stop shopping, single installation, single bill, single customer-service contact, enhanced video-on-demand services, dynamic advertising, etc.), we do not discuss them in this Appendix. However, we acknowledge that, if – as the Applicants claim (see, e.g., Katz Decl. ¶ 26; Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, Exhibit 1, “Expert Report of Dr. Ravi Dhar,” ¶¶ 9-17 (April 21, 2015); Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, Exhibit 2, “2020 Value Proposition: Summary of US Consumer Quant Findings,” at 24 (April 21, 2015); Letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, Exhibit 3, “Consumer Communication Services Preferences: Enter Broadband Network Service Operators,” at 11-12 (April 21, 2015)) – such benefits exist, their presence would (all else being equal) tend to increase the transaction’s benefits for consumers. 9 See Berry-Haile Analysis at 13, 35, 49-50; Berry-Haile Additional Discussion at 4; Steven T. Berry, Estimating Discrete-Choice Models of Product Differentiation, 25 RAND J. OF ECON. 242, 242-262 (1994) (“Berry (1994)”); Steven Berry, James Levinsohn & Ariel Pakes, Automobile Prices in Market Equilibrium, 63 ECONOMETRICA 841, 841-890 (1995) (“Berry et al. (1995)”); Aviv Nevo, A Practitioner’s Guide to Estimation of Random-Coefficients Logit Models of Demand, 9 J. OF ECON. AND MGMT. STRATEGY 513, 513-548 (2000); Aviv Nevo, Mergers with Differentiated Products: The Case of the Ready-to-Eat Cereal Industry, 31 RAND J. OF ECON. 395, 395-421 (2000) (“Nevo, Mergers with Differentiated Products (2000)”). 10 See generally Berry-Haile Analysis. Federal Communications Commission FCC 15-94 175 data sources, the underlying approach is accepted as persuasive and as representing current best practice in merger simulation. 8. Merger simulation is typically a complex undertaking. The model can be thought of as a series of computations. The inputs into these computations are data on firms’ actual prices, subscriber shares, and product characteristics. The output is a set of prices for every type of product in the industry based on the new pricing incentives arising from the merger. There are several parts to the computations executed in the BH Simulation. These are: (1) deriving shares and a measure of price from data; (2) modeling and estimating demand; (3) modeling pricing; (4) estimating marginal costs; and (5) solving for post-merger prices. This section describes the key features of each of these steps. In this section, the discussion of the BH Simulation does not include any potential cost reductions that may result from the transaction. Because the simulation is somewhat complex, critical evaluation is left for subsequent sections. A. Geographic Definition and Product Terminology 9. In the simulation, the estimation is performed at the level of a Designated Market Area (“DMA”).11 Within a DMA, consumers have access to different sets of products according to the availability of providers’ services. This access to different products is at the ZIP code level. The BH Simulation aggregates product availability at the ZIP code level, with data on subscriber shares from various sources, to generate product-level shares of the number of households purchasing that product at the DMA level.12 The BH Simulation also assumes that firm pricing of a product is fixed at the DMA level.13 10. It is important to note at the outset that this merger simulation differs from others in that it includes a variety of products, not all of which compete with each other. The marketplace for broadband Internet access and video programming services allows bundling; that is, consumers can choose to purchase broadband only, video only, both from different providers, both from the same provider, or neither service (i.e., the outside good). Discussing these services requires precision in terminology. A consumer that chooses both broadband and video consumes a bundle. If the consumer purchases both from the same provider at a bundle-specific price, he or she consumes an integrated bundle at a bundle price. A synthetic bundle is purchased by a consumer that chooses two different suppliers for broadband and video services.14 An important consequence of this feature of the marketplace is that the “shares” measured here are not shares within a relevant product market, but percentages of consumers choosing a particular standalone service, synthetic bundle, or integrated bundle.15 11 A Designated Market Area (“DMA”) is a Nielsen-defined television market consisting of a unique group of counties. The United States is divided into 210 non-overlapping DMAs. 12 See Berry-Haile Analysis at 20-33, 41. 13 See id. at 74. 14 A synthetic bundle can represent a contractual relationship between two providers, but this is not a requirement in the simulation model. That is, AT&T and DIRECTV jointly market a bundle that can be purchased by a consumer, but a consumer that simply chooses to purchase standalone AT&T broadband and standalone DIRECTV video under separate agreements would also be included in the share of consumers in the model that purchase the AT&T- DIRECTV synthetic bundle. Similarly, a consumer purchase of standalone cable video and standalone Telco broadband would be counted under the cable-Telco synthetic bundle share in the BH Simulation. 15 In particular, we avoid using the term “market shares,” because not all products considered here are in the same “market” (as the term is typically used in the antitrust context). In fact, the estimates show that standalone video components and standalone broadband components combined in synthetic bundles are complements – where raising (continued….) Federal Communications Commission FCC 15-94 176 11. A product is defined from a consumer’s viewpoint. Broadband services and video programming services each may be purchased as standalone services or may be bundled together, either from a single provider or from two different providers. As noted before, consumers can choose to purchase broadband only, video only, both from different providers, both from the same provider, or neither service. That is, an integrated bundle from AT&T, a synthetic bundle that combines DIRECTV’s video service with a cable provider’s broadband service, and a standalone broadband service offered by a cable provider are all examples of products from the point of view of the consumer, as defined in the BH Simulation. Individual firms, on the other hand, provide components (e.g., video component, broadband component, integrated bundle component). A synthetic bundle is not a component but is instead a product that comprises two components from different providers.16 Firms price components (e.g., the video component) that they offer, while consumers choose products (e.g., a synthetic bundle of broadband and video). A benefit of the structure of the BH Simulation is that it allows for a comparison between the harms from a reduction in competition in the video component and the potential benefits of strengthened incentives to reduce the price of the AT&T broadband-DIRECTV video bundle following the merger. B. Deriving Shares and a Measure of Price from Data 12. The BH Simulation uses data on shares, consumer demographics, prices, and component characteristics. The data are drawn from several primary sources, including the Applicants’ information and web-scraped pricing information.17 As described below, Professors Berry and Haile use these data to estimate shares of products and also to construct a simple price measure for each product in each DMA. 1. Generating Shares 13. Professors Berry and Haile develop a methodology to construct DMA-level shares for all products. There is no single data source that provides this information. The primary data sources that Professors Berry and Haile use in constructing their DMA-level shares are: (1) the Applicants’ subscriber counts for all components; (2) aggregate video subscriber counts used for copyright payments from Cable Data Corporation; (3) Nielsen survey of video services; (4) Nielsen survey of Internet services; and (5) survey data, collected by Professor Ravi Dhar, from a large number of households on all video/broadband services, which included demographic data on households.18 Consumer demographics, other than those obtained in the Dhar survey, were collected from the U.S. Census.19 14. Each of the data sources listed above only partially captures the full picture of video and broadband services that are purchased by consumers; no data source, by itself, gives a picture of the subscriber shares for all possible products. Furthermore, shares implied by the different data sources sometimes conflict. For instance, Nielsen survey data implies a subscriber count for DISH that exceeds DISH’s reported number of subscribers.20 Additionally, national video subscriber counts for AT&T were higher than copyright payments from the Cable Data Corporation indicated.21 (Continued from previous page) the price of one reduces demand for the other – rather than substitutes in the same market, where raising the price of one increases demand for the other. 16 See Berry-Haile Analysis at 73. Though the components in a synthetic bundle are priced separately, they may be offered with a discount to consumers if they are marketed jointly by the providers. 17 As described in Section IV.B, we supplemented the pricing data submitted by the Applicants with third-party information to implement what we refer to as the “Modified Simulation.” 18 See Berry-Haile Analysis at 21. 19 See id. at 28; Katz Additional Detail ¶ 11. 20 See Berry-Haile Analysis at 32. 21 See id. Federal Communications Commission FCC 15-94 177 15. To reconcile these data sources and arrive at a coherent set of shares for each DMA, Professors Berry and Haile conduct a preliminary estimation exercise in which they infer the shares for all products implied by their data.22 This estimation procedure seeks to produce the shares that best fit the available data. It uses a weighted quadratic measure of fit across the five data sources. The weighted feature of this measure puts more weight on those data sources that appear more informative, while the quadratic feature penalizes large differences from observed data more severely than small deviations.23 Finally, Professors Berry and Haile note that the conflicts among data sources discussed above result in a differential bias, and they explore two approaches to correct for the bias.24 16. We appreciate that dealing with multiple data sources is usually problematic. Nothing in the submission suggests that the problems encountered in constructing share measures are unusual. The solutions also seem reasonable. Further, the transparency with which the challenges are discussed is notable. 2. Creating a Simple Price Measure 17. To conduct a merger simulation, it is necessary to have a single price per product.25 This requirement necessitates the construction of a simple price measure from the large array of prices, plans, add-ons, and tiers offered by each firm in the industry. For the BH Simulation, price data come from two sources: (1) prices and plans matched to subscribers from the Applicants; and (2) web-scraped data on other firms’ prices and components.26 Professors Berry and Haile note the challenges in obtaining good price data. For example, they note the great variety of components, tiers, and add-ons, as well as differences between introductory prices and prices for continuing customers.27 Further, the collection of web-scraped prices of competitors creates a number of challenges, notably from: (1) ZIP codes where prices cannot be accessed but service is provided by a firm; (2) DMAs where pricing information was collected for a provider but where coverage information is unavailable; and (3) missing data on component characteristics.28 In each case, the BH Simulation interpolates the missing information from adjacent ZIP codes.29 18. First, Professors Berry and Haile aggregate prices to the plan and ZIP code level. The AT&T and DIRECTV subscriber-level data are aggregated to the ZIP-plan level for customers in their first 12 months of service, and these data are combined with the web-scraped data collected at the ZIP- plan level for all the other firms.30 Professors Berry and Haile then regress price on quality measures (a flexible specification of broadband speed and/or the number of channels) and a set of firm, DMA, and 22 See id. at 20-33, 131-135. 23 See id. at 30. The Dhar survey is not representative of the entire population in each DMA. Hence, Professors Berry and Haile reweight the survey to make it representative. They do this by emphasizing some observations and de-emphasizing others using an estimation procedure in which they fit the survey statistics to statistics from the Consumer Price Survey conducted by the U.S. Census. See id. at 29-30. 24 See id. at 32-33; 131-135. 25 This requirement is recognized in the current approach to dealing with pricing in academic studies. See generally Gregory S. Crawford & Ali Yurukoglu, The Welfare Effects of Bundling in Multichannel Television Markets, 102 AM. ECON. REV. 643, 643-685 (2012) (“Crawford-Yurukoglu Study”). 26 See Berry-Haile Analysis at 34. 27 See id. at 34. Their web-scraping data “often reveals only ‘introductory prices.’” See id. at 34. 28 See id. at 38. 29 See id. at 38. 30 See id. at 136. Federal Communications Commission FCC 15-94 178 firm-DMA fixed effects.31 This regression was run separately for each of the component types (i.e., video-only, broadband-only, and integrated bundles).32 For each of the specifications, the regression estimates a set of firm-component-DMA fixed effects that are used as price indices.33 These price indices are intended to capture cross-market variation within each component.34 19. Due to the adjustment for quality, the constructed price index can take on an arbitrary value. As the Applicants point out, the level of the value does not reflect anything of economic interest.35 However, the price index will accurately reflect the differences in price levels across firms, and, more importantly, changes in the price index will enable calculation of the aggregate effects of the merger on consumers. However, the use of this index approach has two economically relevant limitations. 20. First, the price index does not reflect the range of prices and plans offered by a firm. A firm that offers both inexpensive, poor-quality plans and high-price, high-quality plans could well have the same price index as a firm that offers only medium quality and price plans. The price index will not reflect this difference in menu offerings, even if it is an important aspect of the competitive landscape. 21. Second, some normalization (or “recentering”) of the index is helpful to ensure that prices are similar to the range of prices observed in the actual data. When done properly,36 the normalization has no effect on price movements and welfare measures in the BH Simulation, because any change in the normalization will be fully reflected in the component, provider, and product fixed-effect estimates included throughout each step of the simulation. As a result, the parameters of the conditional indirect utilities (other than the fixed effects) that feed into the share calculations will not change based on the normalization.37 Professors Berry and Haile choose to center the index using the AT&T Average Revenue per User (“ARPU”), so average prices for AT&T components in the model resemble their real- world analogues.38 C. Modeling and Estimating Demand 1. Modeling Demand 22. The demand model used in the BH Simulation assumes that each consumer chooses the product – either standalone video service from some provider, standalone broadband service from some provider, a bundle of both services from either a single or two separate providers, or the outside good – that maximizes his or her utility (taking into account the available products’ prices) among all available 31 See id. at 35, 137. 32 See id. 33 See id. at 137. 34 See id. at 35. 35 See id. at 37. 36 As discussed in Section IV.A, the normalization procedure in the BH Simulation had a small coding error that we identified and corrected. 37 See Berry-Haile Analysis at 37, 42; Steve Berry & Phil Haile, “Price Re-Centering in the Quantitative Analysis of an AT&T-DIRECTV Merger,” transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 2-3 (filed October 1, 2014) (“Berry-Haile Price Re-Centering”). 38 See Berry-Haile Analysis at 35, 138; Berry-Haile Price Re-Centering at 1. The BH Simulation, as submitted by the Applicants, contained a small coding error in the price recentering procedure. We discuss the recentering procedure, and our correction to the coding error, in more detail in Section IV.A. Federal Communications Commission FCC 15-94 179 alternatives.39 This demand model takes a standard logit demand model and enriches it by imposing a nesting structure. We discuss the basic logit demand model first and then discuss the nesting structure that is added to it. 23. In the standard logit demand model, a consumer’s utility can be viewed as the sum of two parts: the first is common to all consumers and is called the “mean utility” (a function of the price, service quality, and other product and demographic characteristics).40 The second part is an idiosyncratic preference term, which captures the fact that individuals vary in their preferences for different products in random ways.41 This idiosyncratic component, which is unobserved by the econometrician, gives the model the flexibility to account for differences in consumer choices. 24. The model then aggregates the individuals’ choices, resulting in formulae for DMA-level shares for products that depend on prices (both the prices of the good and the prices of competitors’ goods) and product characteristics (such as broadband speed, number of video channels, and firm dummies).42 By assuming a particular distribution (which is the defining feature of the logit demand system) for the idiosyncratic component of consumer preference, the model can be fit to DMA-level data on shares, prices, product characteristics, and consumer characteristics (averaged over a DMA).43 The details of this aggregation in a logit-based merger simulation are standard in the literature.44 25. The BH Simulation enriches the standard logit model described above by imposing a nesting structure – whereby the product space is partitioned into subsets, called nests, of similar products for which a consumer’s tastes are hypothesized to be positively correlated – to arrive at a nested logit demand model. The nesting structure introduces an additional individual-specific element to consumer utility and is an important feature that shapes the interactions among products.45 Compared to the 39 This behavioral assumption is a characterizing feature of discrete choice models, of which logit and nested logit (both discussed here) are two examples. See, e.g., KENNETH E. TRAIN, DISCRETE CHOICE METHODS WITH SIMULATION (Cambridge Univ. Press 2009) (“TRAIN (2009)”). 40 See Berry-Haile Analysis at 42. 41 See TRAIN (2009); Berry-Haile Analysis at 42. 42 See Berry-Haile Analysis at 42, 50, 61; Katz Reply Decl. ¶ 57. 43 The aggregation rests heavily on the use of the extreme value type I distribution to capture the idiosyncratic preferences of consumers. The use of this distribution has economic content. The set of values that a random variable drawn from this distribution can take is unbounded. Consequently, no matter how bad a product is, there is always some small chance that a consumer will buy it. This feature can lead to implausibly high prices for relatively undesirable products, as a few consumers (who have very low idiosyncratic draws on other products and high idiosyncratic draws on an undesirable product) will still find the undesirable product to be an attractive option. Of particular relevance in this merger simulation are the implications for the pricing of bundles and the standalone versions of their constituent components. If the price of the standalone version of a component that is sold as part of a bundle exceeds the bundle’s price, then no consumer should buy the standalone version, as the bundle is a cheaper way to get access to that component. (That is, consumers have free disposal.) By contrast, the modeling of idiosyncratic preferences violates this intuition, because each consumer is allocated independent random values that represent his or her idiosyncratic preferences for the bundle and each of the two standalone components. Independence implies that, if a bundle is priced at $80 and the standalone component at $100, some consumers will still buy the component. While the above situation is a theoretical possibility given this market structure, the BH Simulation as estimated does not generate implausible pricing patterns as described above. Furthermore, on the basis of a separate set of Monte Carlo simulations that we conducted, the BH Simulation’s results do not appear to be driven by the distributional assumptions. 44 See generally Berry (1994); Berry et al. (1995); Nevo, Mergers with Differentiated Products (2000). 45 See Berry-Haile Additional Discussion at 6. Federal Communications Commission FCC 15-94 180 standard logit model, the nested logit model allows substitution patterns between products to be more flexible and for the data to better guide these substitution patterns.46 The demand model in the BH Simulation uses four nests – bundles, video only, broadband only, and neither (i.e., the outside good) – and thus requires the estimation of three nesting parameters.47 (Because choices are driven by relative utility levels, the fourth nesting parameter, for the nest that contains the outside good, is pre-set to zero.48) Thus, for example, a consumer that likes bundle A probably also likes bundle B. 26. Because product availability varies at a very local level, the demand model is formed at the individual level, aggregated to the ZIP code level (where the variation in choice sets is assumed to occur), and then further aggregated to the DMA level (where, in estimation, shares inferred by the model are matched to those generated by the data).49 These two steps of aggregation lead to a more complicated estimation procedure than would normally be required, but they allow the model to take into account local variation in product availability. 2. Estimating Demand Parameters 27. Due to the need to accommodate ZIP-code-level variation in product offerings, the estimation procedure requires non-linear numerical optimization.50 Cross-DMA price variation is determined, in part, through variation in product availability across ZIP codes within a DMA. (This variation is assumed to be exogenous).51 Absent this variation, the BH Simulation could be estimated linearly through standard (linear) instrumental variables techniques;52 indeed, apart from this variation, there is little that is unusual about the estimation procedure, which uses standard generalized method of moments (“GMM”)-based techniques.53 28. In the estimated BH Simulation, the variables that determine demand and, thus, shares of demand are: (1) the DMA price indices of the components that make up the product (including any relevant discounts on synthetic bundles); (2) maximum download speed for broadband and maximum offered channels for video; (3) factors affecting Direct Broadcast Satellite (“DBS”) quality (such as latitude measurement, urban share of a DMA, and homeowner share of a DMA); (4) firm dummies; (5) interactions between DMA demographics (age, education, median household income) and nests; and (6) product dummies.54 Price is assumed to be endogenous, and instrumental variables are used to account 46 See Katz Reply Decl. ¶ 57. The standard logit demand model, in the absence of the nesting structure, has some economically undesirable properties. Most problematic is the fact that substitution among products due to a price change becomes a function solely of market shares. See Berry et al. (1995) for an extended discussion of this feature of the logit demand model. 47 See Berry-Haile Analysis at 45, 47. Although three parameters must be estimated, a simple specification sets the nesting parameter equal across nests. 48 See id. at 47-48. 49 See id. at 41. 50 See id. at 41, 50, 52, 55-60. 51 See id. at 41. 52 See id. 53 See id. at 140. 54 See id. at 51, 61; Katz Reply Decl. ¶ 57. Federal Communications Commission FCC 15-94 181 for this endogeneity.55 The instruments used follow those used in the 1995 analysis by Steven Berry, James Levinsohn, and Ariel Pakes.56 D. Modeling Pricing and Determining Marginal Cost 29. After estimating demand for all products, the next step in the merger simulation is to use the prices at which each firm sells its components – a video-only component, a broadband-only component, or an integrated bundle – to determine the marginal costs of providing those components. The simulation assumes that each firm sets the prices that maximize its profits given its marginal costs, with the additional assumption that it takes the prices that other firms set for their components and bundles as given.57 From this the BH Simulation infers the marginal costs for which the observed prices would maximize profits for each firm.58 Knowing these marginal costs, the BH Simulation then estimates the profit-maximizing prices for all market participants following the merger, where the merging parties (AT&T and DIRECTV) now maximize profits jointly rather than separately.59 30. To make estimation feasible, the BH Simulation restricts the price for a component or bundle to be constant within a DMA.60 This means that the price responses to competitors’ footprints occur at a DMA-wide level.61 With that, a firm’s profits are determined by adding up the profit it obtains from selling any particular component in any given DMA. The firm’s profit is derived by calculating the per-unit profit – the difference between the price the firm chooses for that product in that DMA and the marginal cost of providing that component in that DMA – and multiplying that times the quantity it would sell, which is based on the demand model estimated above.62 55 See Berry-Haile Analysis at 44, 51, 140. 56 See id. at 51; see also Berry et al. (1995). 57 This is known as the Bertrand model with product differentiation. 58 See Berry-Haile Analysis at 72, 84. This technique is standard in the academic literature. See generally Berry et al. (1995); Nevo, Mergers with Differentiated Products (2000); Sofia Berto Villas-Boas, Vertical Relationships between Manufacturers and Retailers: Evidence with Limited Data, 74 REV. OF ECON. STUDIES 625, 625-652 (2007); Céline Bonnet & Pierre Dubois, Inference on Vertical Contracts between Manufacturers and Retailers Allowing for Nonlinear Pricing and Resale Price Maintenance, 41 RAND J. OF ECON. 139, 139-164 (2007). 59 See Berry-Haile Analysis at 72. 60 See id. at 74. 61 Provider penetration is assumed to be constant within a DMA between the pre- and post-merger scenarios. Penetration varies across DMAs, however, and firms may respond to higher competitor penetration by adjusting prices in that DMA. 62 See Berry-Haile Analysis at 81. A firm f’s profit in a DMA m, as a function of component prices in that DMA, is: That is, firm f’s profit is the sum over all components r offered in DMA m by firm f (with the set of components denoted by ) of the price of that component less the marginal cost ( ) of providing it, all multiplied by the share of consumers that purchase the component ( ). Finally, where relevant, half of the discount d that is offered on DIRECTV-AT&T and other DIRECTV-Telco or DIRECTV-cable synthetic bundles is subtracted and then multiplied by the share of consumers ( ) that choose options that include the discount. If a firm does not offer a component that involves a discount, the last expression is equal to zero. The discount adjusts the price that is then folded into the share calculation. Hence, it is also present on the demand side, albeit buried in this expression. Optimal prices for a firm, given other firms’ prices, are given by the solutions to the firm’s first-order conditions, (continued….) Federal Communications Commission FCC 15-94 182 31. The BH Simulation accounts for discounts on synthetic bundles that combine DIRECTV’s video component with other providers’ broadband components, both those offered with AT&T broadband and those offered with broadband from other telecommunications and cable providers. This discount is treated as fixed at [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] in the merger simulation, regardless of the firm providing the broadband service, and assumes that the discount is split equally between the two firms – DIRECTV providing video services and the company offering the broadband service.63 This is modeled as decreasing the price of each component by half of the discount [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] if bought as a bundle, and the firm’s realized expenditure on the discount is treated as contributing to marginal cost.64 E. Solving for Post-Merger Prices 32. Post-merger prices are derived by solving for the optimal prices of each firm, under the hypothetical post-merger industry structure, using the estimated marginal costs and demand system derived above, but where prices of AT&T’s and DIRECTV’s components are set to maximize joint profits after the merger.65 This changes the pricing incentives for these components and, as a result, changes the prices post-merger. In response to the change in AT&T’s and DIRECTV’s prices post- merger, prices adjust across the entire industry.66 The BH Simulation captures the changes in AT&T’s and DIRECTV’s pricing incentives and allows prices by all firms to adjust to the changed ownership. The prices ultimately derived by the BH Simulation are such that every firm is maximizing profits, taking as given the prices being offered by all other firms. III. ECONOMIC EFFECTS CAPTURED BY THE SIMULATION MODEL 33. The merger simulation allows the estimation and comparison of three different economic effects. The first two are standard parts of any merger simulation: (1) the harm from increased prices resulting from eliminating competition between the merging parties and (2) the pass-through of any cost reduction resulting from the merger to consumers in the form of lower prices. The third effect results from the pricing incentives created by the fact that products include integrated bundles or synthetic bundles of individual components offered by firms. A. Horizontal Overlap 34. A merger between two firms that supply substitute products may reduce competition by enabling the merged firm to increase joint profits by unilaterally raising the price of one or both products (Continued from previous page) which set the partial derivatives of the above expressions with respect to the firm’s components’ prices equal to zero: The full set of optimal prices, which is defined by the set of first-order conditions, does not have a closed-form expression and is determined computationally. 63 See Berry-Haile Analysis at 77. The DIRECTV synthetic bundle discount is assumed to apply to all synthetic bundles offered with Telco broadband service. The simulation also accounts for a [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] discount offered by DIRECTV with [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 64 See id. at 77. 65 See id. at 90. 66 See id. at 101-108. Federal Communications Commission FCC 15-94 183 above the pre-merger price.67 The merged firm may profit from this strategy, because some of the sales lost due to the price increase would be diverted to the product of the merger partner.68 The closer the products are as substitutes, the greater the propensity for prices to increase following a merger.69 In addition, firms that compete against the merged firm now face a competitor with less incentive to offer low prices. As a consequence, equilibrium prices across the industry may rise as well. 35. In 77 DMAs, both DIRECTV and AT&T offer video services.70 (This figure represents 36.7 percent of the 210 total DMAs.) According to the Applicants, DIRECTV’s video subscriber share in the areas in which AT&T U-verse is available to at least 90 percent of households is [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent, and AT&T’s share is approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent.71 Although AT&T and DIRECTV are not each other’s closest substitutes, the diversion rates between them are non-trivial.72 The estimated diversion rates imply that approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent of AT&T’s video subscribers that leave in response to an AT&T price increase would go to DIRECTV’s video service. Furthermore, conditional on the availability of AT&T’s video service, approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent of DIRECTV’s subscribers that leave in response to a DIRECTV price increase would go to AT&T for video service. Therefore, AT&T’s and DIRECTV’s video services are substitutes to a degree that raises the potential for the merger to reduce horizontal competition as described in the previous paragraph. B. Programming Payment Reductions 36. The Applicants claim that the transaction would result in reductions in program payments [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .73 This effect can be thought of 67 See U.S. Department of Justice and the Federal Trade Commission Horizontal Merger Guidelines, August 19, 2010, § 6.1 at 20 (“DOJ/FTC 2010 Horizontal Merger Guidelines”). 68 Whether such a strategy would be profitable depends on the relative margins of the two products. See id. 69 See id. 70 In its filings, AT&T indicated that, in September 2014, AT&T U-verse video passed at least one home in 77 unique DMAs. See AT&T Inc. Response to Sept. 9, 2014, Information and Discovery Requests, transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Vanessa Lemmé, Media Bureau, FCC, MB Docket No. 14- 90, Exhibit 5.a.1 (Oct. 7, 2014) (“AT&T Response to Sept. 9, 2014, Information Request”). Although some commenters assert that U-verse video is available in more than 77 DMAs, we rely on data submitted by the Applicants for our analysis herein. See, e.g., Petition to Deny of Writers Guild of America, West, Inc., MB Docket 14-90, at 4, 9 (filed Sept. 16, 2014); Reply Comments of Writers Guild of America, West, Inc. to Opposition, MB Docket 14-90, at 1, 3 (filed Jan. 7, 2015) (stating that U-verse is available in 129 DMAs). 71 See Katz Additional Detail ¶ 58. 72 The diversion rates presented here were generated by the BH Simulation at pre-merger prices. For purposes of comparison, the estimates at post-merger prices – based upon the Modified Simulation (as defined below) and an assumed [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] reduction (also discussed below) in the marginal cost of AT&T’s video component – are approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent for the AT&T to DIRECTV diversion and [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent for the DIRECTV to AT&T diversion. Usually a substantial unilateral post-merger price increase requires that a significant fraction of the customers purchasing one product view the other product as the next best choice. However, this significant fraction does not need to approach a majority. See DOJ/FTC 2010 Horizontal Merger Guidelines § 6.1 at 20-21. 73 See Application, Description of Transaction, Public Interest Showing, and Related Demonstrations, transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 35 (filed June 11, 2014) (“Application”); Katz Decl. ¶ 115; Joint Opposition at 16; AT&T Response to Sept. 9, 2014, Information Request at 244. Federal Communications Commission FCC 15-94 184 as a reduction in the marginal cost of providing service. The extent to which this potential cost reduction benefits consumers is a function of the extent to which the payment reductions are realized, and the extent to which the cost reduction is passed through to consumers.74 If the payment reductions are large and are substantially passed through to consumers, consumers could see lower prices after the merger, even with reduced competition. These lower prices offered by the merged entity, in turn, may induce other industry participants to reduce their prices as well. 37. The BH Simulation incorporates the effect of lower programming payments that result from the transaction by reducing the marginal cost of the AT&T video components by a flat [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .75 C. Bundling Effects 38. The last economic effect is the way the merger changes pricing incentives when integrated and synthetic bundles are available. Quantifying this effect is the novel feature of the BH Simulation. 39. Consider two firms, each of which supplies one component of a synthetic bundle. Each firm wants to set the price of its component so as to maximize its own profit. In doing so, it ignores the impact of its own price on the other firm’s revenue. More specifically, when a firm sets a high price, it ignores the fact that this high price will drive some consumers away from the other firm’s component, thus reducing the other firm’s revenue. However, if the providers of the components of the synthetic bundle merged, the combined firm would internalize the tendency of a high price on one component to reduce the revenue earned by the other component. Provided that nothing else changed, the combined firm would offer the newly integrated bundle at a price that is below the sum of the prices of the two products making up the synthetic bundle. For this reason, a merger would put downward pressure on the price of the bundle.76 40. AT&T and DIRECTV offer a synthetic bundle.77 By merging, the pricing incentives for the parties would likely be to lower the price of the newly integrated bundle. If the price of the newly integrated bundle of AT&T’s broadband service and DIRECTV’s video service falls, then this effect may induce other industry participants that also offer bundles to lower those bundles’ prices.78 41. The BH Simulation incorporates the effects of reduced horizontal competition, reduced payments for programming, and the benefits from offering an integrated instead of a synthetic bundle. 74 In a merger simulation, the pass-through rate will be heavily influenced by the choice of functional form that is used to model consumer demand, because the pass-through rate is a function of the second derivative of the demand function. 75 See Berry-Haile Analysis at 109. 76 See Application at 65-67; Katz Decl. ¶¶ 66-69, 74; Guyardo Decl. ¶ 27; Joint Opposition at 14; Katz Reply Decl. ¶ 12; AT&T and DIRECTV, White Paper, Why the “Double Moral Hazard” Problem Cannot be Resolved by Contract, at 4-6, transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (filed Nov. 12, 2014). These distortions are a version of the standard double-marginalization problem found in the economics literature on interactions between upstream and downstream firms. See LUIS M. B. CABRAL, INTRODUCTION TO INDUSTRIAL ORGANIZATION (The MIT Press 2000) for a simple explanation. 77 See Katz Decl. ¶¶ 33-34. AT&T also currently offers an integrated bundle of U-verse video and broadband. DIRECTV, however, does not provide a broadband service, and therefore is able to offer only synthetic bundles to customers that want both broadband and video services. See id. ¶¶ 15, 28-29. 78 See Application at 64; Katz Decl. ¶¶ 3-4, 89, 92. Federal Communications Commission FCC 15-94 185 The data are guiding the relative strengths of these three effects. The output of the model is the impact on consumers of the combination of these three economic factors. IV. MODEL CORRECTIONS AND ADJUSTMENTS 42. In the next section, we report the basic results from two versions of the BH Simulation. The first version, called the “Corrected Simulation,” is a modification of the version of the BH Simulation that the Applicants submitted on September 23, 2014.79 The Corrected Simulation corrects minor coding errors that remained in the recentering procedure of the hedonic pricing model in both submitted versions of the BH Simulation.80 We describe this correction more fully in the paragraphs that follow, but the primary economically important difference is that consumers are estimated to be slightly more price- sensitive in the Corrected Simulation relative to the estimates derived from the September 2014 BH Simulation. 43. The second version, called the “Modified Simulation,” uses billing data from third-party cable providers (where possible) in place of the web-scraped pricing data in the BH Simulation. These data are also described in this section. In the Modified Simulation, consumers are more price-sensitive than in the Corrected Simulation. 81 44. Last, we discuss differences between the BH Simulation and our two models in the handling of programming payment reductions. We find that the reduction in programming payments that should be modeled in the simulation is likely lower than that claimed by the Applicants. A. Corrections to Price Recentering Procedure 45. As described above, the BH Simulation requires a single price for each product in each DMA.82 Professors Berry and Haile construct a single price index for each product through a hedonic regression of firm prices on component characteristics.83 Three separate hedonic regressions are performed, one each for standalone video, standalone broadband, and integrated bundles.84 The hedonic regressions are performed at the “provider-ZIP-service plan” level and include as regressors the speed of the broadband service, the number of video channels, indicator variables for premium channels (e.g., HBO), and fixed effects for provider, DMA, and provider-DMA.85 The fixed effects from these regressions are saved, to be added later to the average prices centered on AT&T’s ARPU.86 79 See Berry-Haile Updated Results. The BH Simulation that the Applicants submitted in September 2014 corrected two errors from their initial submission: one was an error in the construction of some prices, and the second was a coding error that affected the calculation of the Generalized Method of Moments (“GMM”) weight matrices and standard errors. Professors Berry and Haile assert, in the September filing, that these corrections have no material effect on the results or conclusions of their analysis. See Berry-Haile Additional Discussion at 1 n.1. 80 See Berry-Haile Analysis; Berry-Haile Updated Results. 81 Additionally, the estimated nesting parameters imply that, relative to consumers in the Corrected Simulation, consumers in the Modified Simulation that buy bundles are more likely to substitute to other bundles following a price change, and consumers in the Modified Simulation that buy standalone products are less likely to substitute to standalone products of the same type following a price change. 82 See supra ¶¶ 17, 30. 83 See Berry-Haile Analysis at 35; Berry-Haile Price Re-Centering at 1. 84 See id. 85 See Berry-Haile Analysis at 137; Berry-Haile Price Re-Centering at 1. 86 See Berry-Haile Analysis at 35, 137–138; Berry-Haile Price Re-Centering at 1. Federal Communications Commission FCC 15-94 186 46. After each regression is estimated and the fixed effects are saved, Professors Berry and Haile calculate a weighted-average fixed effect separately for each component for five provider types (AT&T, DIRECTV, Cable, Telco, and DISH).87 Professors Berry and Haile then calculate the weighted average AT&T fixed effect by averaging the predicted values of the regression using only the AT&T observations.88 The AT&T fixed effect is then subtracted from the fixed effect of all other providers to preserve the average differences in prices across provider types.89 These differences are then added to AT&T’s estimated video, broadband, and bundle ARPUs (calculated separately from AT&T list prices), which give the new “center” for the price indices.90 This center is then added to the fixed effects from the original hedonic regression, which reintroduces the variation in product prices across DMAs.91 47. In principle, this recentering of prices does not change the results produced by the demand model estimation or merger simulation.92 The reason is that the demand estimation contains a full set of provider, DMA, and component dummies that control for the mean price level.93 Additionally, in the merger simulation, the estimates of markups and all estimated changes in prices are unaffected by the price recentering as the level of prices passes directly through to marginal cost estimates.94 However, in our analysis of the BH Simulation, we found a small coding error in the way the recentering was done for the bundles offered by non-AT&T “Telco” providers. 48. The error is due to the manner in which Professors Berry and Haile handle missing observations in the web-scraped pricing data. In some ZIP codes, a single Telco offers an integrated bundle, and the prices of those bundles were centered at the “bundle” center by the recentering procedure. Likewise, the prices for standalone broadband and standalone video products offered by Telcos in these ZIP codes were centered at the broadband and video “centers,” respectively. However, in order to allow for the possibility of a synthetic Telco-Telco bundle where one Telco provides the video component and another provides the broadband component (but neither offers an integrated bundle), Professors Berry and Haile construct a synthetic bundle by summing the prices of the standalone components. However, they create this synthetic bundle after performing the recentering procedure, which results in synthetic Telco- Telco bundles being centered at a different mean than the integrated Telco-Telco bundles. 49. With Telco bundles being centered at two different locations, the original BH Simulation results are sensitive to the choice of normalization, because Telco-Telco synthetic bundles and Telco- Telco integrated bundles are not distinguished in the demand estimation and merger simulation. Accordingly, we adjust the code to ensure all Telco-Telco bundles are recentered at the same location, 87 The weighted-average fixed effect is calculated by setting the regressors for broadband speed, number of channels, and premium channels to zero, and taking the average of the predicted values (or “intercepts”) that come out of the regression. The weights used to calculate the weighted-average fixed effect are the DMA coverage percentages of each provider. 88 Again, the fixed effect is calculated by setting the regressors for broadband speed, number of channels, and premium channels to zero, and taking the average of the predicted values that come out of the regression. As only the AT&T observations are used, only the AT&T fixed effect is set to one in the predictions; all other provider fixed effects are set to zero. The weights used to calculate the weighted average AT&T fixed effect are the total number of AT&T subscribers across DMAs. 89 See Berry-Haile Price Re-Centering at 1. 90 See id. 91 See id. 92 See Berry-Haile Analysis at 37, 138; Berry-Haile Price Re-Centering at 2-3. 93 See Berry-Haile Analysis at 37; Berry-Haile Price Re-Centering at 2-3. 94 See id. Federal Communications Commission FCC 15-94 187 and we report these results as the “Corrected Simulation” below. The results in the Corrected Simulation are not dramatically different from those from the BH Simulation, but the effect is that consumers are estimated to be slightly more price-sensitive.95 B. Adjustments to Third-Party Prices 50. The data used in the BH Simulation are described in Section II.B. Data used by Professors Berry and Haile on prices, plans, add-ons, and tiers of service at the ZIP code level for third- party providers were collected via web scraping. By contrast, information on AT&T and DIRECTV prices, plans, add-ons, and tiers was drawn from own-party subscriber billing data. These data are used to construct a price index for different components across DMAs and providers.96 51. An examination of the pricing data described above raises three issues. First, there is an extremely wide range of prices in the own-party subscriber data submitted by AT&T and DIRECTV, with some of the prices taking on unrealistic values. The own-party subscriber-plan data exhibits variation across ZIP codes, with prices that vary markedly within plan, sometimes by hundreds of dollars. In addition, observations across the entire data set take values that seem inconsistent with established industry pricing patterns. For example, individual plan prices span from less than a dollar to over $2,400 per month in some ZIP codes. Most web-scraped data did not exhibit these two features, though there is one exception: the web-scraped data for Comcast also exhibited extreme variation in some instances (both within plans and across ZIP codes). 52. A second issue is whether there exists sufficient variation in observed prices offered by providers across DMAs to generate the price index that feeds into the demand estimation. Third-party data scraped from the Web exhibit little to no fluctuations in plan prices across ZIP codes.97 Of the 59 providers in the raw data, only Comcast (among the providers with web-scraped data) and AT&T and DIRECTV (both with own-party subscriber data) showed meaningful variation in plan prices across ZIP codes. The majority of the other third-party providers with web-scraped data had prices that showed very little variation in the data which is fed into the hedonic regression. This raises the concern that the bulk of the economically meaningful variation in the pricing data used in the simulation comes from only three firms (the two Applicants and Comcast). 53. The third issue is that prices produced via web scraping were, by construction, advertised introductory prices, whereas the Applicants’ subscriber data relied on actual prices. Actual and advertised introductory prices may exhibit different patterns, and it was not immediately clear which would be the more appropriate pattern on which to base the simulation. 54. A desire to improve the underlying data feeding into the demand estimation and merger simulation led us to request ZIP-code-by-plan-level pricing data from Comcast, Time Warner Cable, and Charter.98 We replaced the web-scraped data in the original BH Simulation with the actual pricing data 95 Although the differences between the results of the BH Simulation and the Corrected Simulation are not dramatically different, the consumer surplus benefits from the transaction are found to be less in the Corrected Simulation than in the BH Simulation and are slightly negative in one specification. These results are described further in the sections that follow. 96 See supra Section II.B.2. 97 For example, even though Verizon and DISH comprise [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent of total observations, prices for each plan offered by the two providers do not vary at all across geographic areas. 98 See Comcast Corporation Response to Jan. 8, 2015, Information and Data Request, transmitted by letter from Kathyrn A. Zachem, Senior Vice President – Regulatory and State Legislative Affairs, Comcast Corporation, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (Jan. 23, 2015) (“Comcast Response to Jan. 8, 2015, Information Request”); Time Warner Cable Inc. Response to Jan. 8, 2015, Information and Data Request, (continued….) Federal Communications Commission FCC 15-94 188 submitted by the third-party cable companies.99 In the results below, the versions of the merger simulation run with the updated third-party data is referred to as the Modified Simulation. The third-party data had fewer extreme values than the web-scraped data (particularly in the case of Comcast), exhibited significantly more variation across ZIP codes than the web-scraped data (in the case of Time Warner Cable and Charter), and better matched the patterns, in a qualitative sense, seen in the subscriber level prices submitted by AT&T and DIRECTV and used in the BH Simulation. C. Programming Payment Reductions 55. The last adjustment we make to the merger simulation is the size of the programming payment reductions that accrue to, as a result of the transaction, products that include AT&T’s video service. One of the central claims by the Applicants is that, following the transaction, AT&T’s programming payments will be reduced to [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .100 The Applicants have stated that they expect that AT&T’s programming payments will be about [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent higher than DIRECTV’s prior to the date that the transaction is expected to close, and that this difference will remain at approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent into future years absent the transaction.101 Additionally, in their evaluation of the transaction, the Applicants use a model to estimate that AT&T’s programming payments will grow from [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] in 2014 to [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] by 2023, increasing at a rate of approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent per year.102 This model also estimates that DIRECTV’s programming payments (excluding [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ) will grow from [BEGIN VIDEO PROG. CONF. INFO.] [END VIDEO PROG. CONF. INFO.] in 2014 to [BEGIN VIDEO PROG. CONF. INFO.] [END VIDEO PROG. CONF. INFO.] in 2023, increasing by approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent over the next two years before leveling off at approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] to [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent growth thereafter.103 AT&T’s modeling predicts a [BEGIN HIGHLY (Continued from previous page) transmitted by letter from Matthew A. Brill, Counsel for Time Warner Cable, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (Jan. 23, 2015); Charter Communications, Inc. Response to Jan. 8, 2015, Information and Data Request, transmitted by letter from John L. Flynn, Counsel for Charter, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (Jan. 20, 2015). 99 We continue to use the web-scraped data in the estimation for all other third-party providers other than Comcast, Time Warner Cable, and Charter. 100 See Application at 36; Katz Decl. ¶ 115; Application, Declaration of Rick L. Moore, Senior Vice President, AT&T, transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, ¶ 15 (filed June 11, 2014); Joint Opposition at 16; Katz Reply Decl. ¶ 32; AT&T Response to Sept. 9, 2014, Information Request at 244. The combined firm may also enjoy increased bargaining power due to the loss of an independent video provider and the larger combined subscriber base, though the Applicants do not consider the potential impact of these in terms of efficiencies of the transaction. See Katz Decl. ¶ 115. 101 See AT&T Response to Sept. 9, 2014, Information Request at 243; AT&T Response to Sept. 9, 2014, Information Request, Exhibit 69.c.2, tab “Content Costs (Output).” 102 See AT&T Response to Sept. 9, 2014, Information Request, Exhibit 69.c.2, tab “Content Costs (Output).” 103 See id. [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] costs are projected to increase absent the transaction from [BEGIN VIDEO PROG. CONF. INFO.] [END VIDEO PROG. CONF. INFO.] per subscriber per month to [BEGIN VIDEO PROG. CONF. INFO.] [END VIDEO PROG. CONF. INFO.] over the same time frame. See id. Federal Communications Commission FCC 15-94 189 CONF. INFO.] [END HIGHLY CONF. INFO.] percent difference in 2014 payments, which is lower than the claimed [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent difference found in the Applicants’ declarations. 56. AT&T claims that, after the transaction, programming payments for AT&T U-verse video will fall to [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] , and that this reduction will phase in over a [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] period.104 The end result is that AT&T expects its programming payments to fall by approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent per subscriber per month (in 2015) to approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent per subscriber per month in 2020 and thereafter, once the companies combine.105 57. Although we are unable to assess the Applicants’ claims regarding the future growth in the payments for video programming, we can assess the extent to which AT&T’s programming payments are higher than those currently paid by DIRECTV. We requested channel-by-channel programming payments for all channels the parties offer over their MVPD services.106 We also requested similar data from Comcast, which is currently the MVPD with the largest set of subscribers.107 Using these data, we compare the actual per-channel amounts that these providers pay to acquire video programming. The results are presented in Table 1. 58. The first row in Table 1 reproduces AT&T’s own estimates of the programming payments of the three MVPD providers, as reported in AT&T’s response to specification 69 of the Commission’s Information Request.108 AT&T has the highest estimated payments at [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] per subscriber per month, which is approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] more than DIRECTV’s payments and approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] more than Comcast’s payments. In percentage terms, AT&T’s modeling shows DIRECTV’s payments being [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent lower than AT&T’s payments and that Comcast’s payments are [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent lower than AT&T’s payments. 59. The second row in Table 1 reports the programming payment per subscriber as estimated from the channel-by-channel data submitted by the Applicants and Comcast in response to the Information Requests. The totals include the payments from all channels offered by each provider, regardless of whether the channels are offered by one of the other two providers.109 The totals are also 104 The phase-in assumes that AT&T will realize [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] and thereafter. See AT&T Response to Sept. 9, 2014, Information Request, Exhibit 69.c.2, tab “Content Costs (Output).” 105 See AT&T Response to Sept. 9, 2014, Information Request, Exhibit 69.c.2, tab “Content Costs (Output).” Percentages calculated by comparing the AT&T costs per subscriber to the “rebased” AT&T costs per subscriber in the spreadsheet. 106 See id., Exhibit 40 a-c; Information and Discovery Requests, transmitted by letter from William M. Wiltshire, Counsel for DIRECTV, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, Schedule 37 (Oct. 7, 2014). We use the data from March 2014 in our analysis, which corresponds to the time period analyzed by Professors Berry and Haile in the merger simulation. 107 See Comcast Response to Jan. 8, 2015, Information Request. 108 See AT&T Response to Sept. 9, 2014, Information Request, Exhibit 69.c.2, tab “Content Costs.” 109 Retransmission consent fees paid to local broadcast television stations are excluded from these totals, as are the fees for many of the premium channels (e.g., Showtime, Encore, etc.) which do not appear to be comparable across (continued….) Federal Communications Commission FCC 15-94 190 weighted by the share of customers subscribing to each channel so that they represent average per- subscriber payments. AT&T’s estimated payment per subscriber from the channel-by-channel data is [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . Similarly, the estimated per- subscriber payment from the channel-by-channel data is [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] for DIRECTV (giving a difference of about [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] between AT&T and DIRECTV) and [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] for Comcast. In percentage terms, the difference between AT&T and DIRECTV programming payments is approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent, which is similar to the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent reduction the Applicants expect in future years under the transaction and nearly identical to the percentage difference estimated in AT&T’s model in 2014 (shown in row 1).110 It is, however, lower than the current [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent difference claimed by the Applicants elsewhere in their filings. 60. The difference between the per subscriber payments for video programming between AT&T and DIRECTV from the channel-by-channel data is likely driven by three separate factors. First, for the channels that both AT&T and DIRECTV offer to their subscribers, AT&T typically pays a higher price per subscriber for the content, likely because it has fewer subscribers (and a weaker bargaining position) than DIRECTV. This effect can be seen in the second panel of Table 1, where we report the per-channel fee that each provider pays for a subset of common national television networks.111 As an example, AT&T paid [BEGIN VIDEO PROG. CONF. INFO.] [END VIDEO PROG. CONF. INFO.] per subscriber for the USA Network, while DIRECTV paid only [BEGIN VIDEO PROG. CONF. INFO.] [END VIDEO PROG. CONF. INFO.] , and Comcast paid [BEGIN VIDEO PROG. CONF. INFO.] [END VIDEO PROG. CONF. INFO.] . 61. A second factor that may contribute to differences in per-subscriber programming payments across providers is that providers offer different sets of channels to their subscribers. All else being equal, a provider that chooses not to offer certain channels on any of its service tiers will have lower per-subscriber payments than a provider that does offer those channels. Although most providers will offer the most popular national networks to their subscribers, there is significant variation among the smaller niche and regional channels that each of the providers offer. 62. In the channel-by-channel data submitted by the providers, AT&T reports prices for 425 channels offered over its MVPD service, DIRECTV reports prices for 306 channels, and Comcast reports prices for [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] channels. Of these channels, 194 are reported by all three providers. We match the license fees and subscriber counts of these 194 common channels and calculate the monthly payment per subscriber for each of the three providers. The third row of Table 1 contains these estimates. (Continued from previous page) providers. For these reasons, the numbers presented in row 2 are likely to be lower than those presented in row 1, and focus should be placed on the percentage differences between the providers within each row. 110 See Application at 36; Katz Decl. ¶ 115; Joint Opposition at 16; Katz Reply Decl. ¶ 32; AT&T Response to Sept. 9, 2014, Information Request at 244. 111 MVPD providers will sometimes purchase a bundle of channels from a content provider rather than paying for each channel separately. For example, AT&T purchased the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] channels as a bundle, reporting a single price for all of the channels. DIRECTV and Comcast, however, reported most of these channels under separate prices. For comparison, we aggregate the prices of individual channels to the bundle level when at least one of the providers makes a bundle purchase. Federal Communications Commission FCC 15-94 191 Table 1: Programming Payments per Video Subscriber for AT&T, DIRECTV, and Comcast [BEGIN VIDEO PROG. CONF. INFO.] [END VIDEO PROG. CONF. INFO.] 63. Not surprisingly, after dropping the smaller channels that do not appear on all three services, the estimated programming payments fall quite substantially. However, the payments fall more for AT&T than they do for DIRECTV or Comcast. The monthly per-subscriber payments for AT&T for the channels that are not offered across all three providers make up [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] of the total per-subscriber costs reported in row 2. However, the incremental costs to DIRECTV and Comcast for the channels that are not offered across all three providers are only [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] and [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] , respectively. In the case of DIRECTV, this may be partly due to the fact that a higher percentage of the channels reported by DIRECTV matched with the channels reported by AT&T and Comcast. As a consequence of the higher cost of non-matched AT&T channels, the percentage differences between AT&T’s per-subscriber payments and the per- subscriber payments of the other two providers are lower than those reported in row 2 – though more so for DIRECTV than for Comcast. AT&T pays [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] per subscriber per month for these [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] matched channels, and DIRECTV pays [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] per subscriber per month for those same channels. The relevant takeaway is that, in percentage terms, the difference between the AT&T and DIRECTV payments for the channels that all three companies offer is approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent, about [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percentage points lower than the claimed [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent reduction the Applicants expect to eventually achieve, about [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] less than the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent difference in the channel-by-channel data when all channels are considered, and about half of the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent difference the Applicants claim in their filings currently exists between the two companies. 64. A third factor that contributes to the differences in the providers’ per-subscriber video programming payments concerns the distribution (i.e., tiering) of subscribers across the full range of channels offered by a provider. An MVPD that has a larger fraction of its subscribers receiving higher- cost programming will have higher overall per-subscriber costs, even if the per-subscriber fee for each channel is the same for all MVPDs. An MVPD decides the tiers, but a programmer may negotiate to have its channel carried on a lower tier (to reach more subscribers) in return for a lower payment.112 In this sense, the tiering of programming by a particular provider (i.e., the bundling of channels into service tiers 112 A programmer may also negotiate with the MVPD to have one or more of its channels placed in a given “neighborhood” as well as a specific tier. For example, a sports programmer may want to ensure that its channel is grouped with the sports channels of other programmers on the MVPD’s channel lineup. Further, broadcast stations have to be on the lowest tier and available to anyone buying cable. See generally Annual Assessment of the Status of Competition in the Market for the Delivery of Video Programming, MB Docket No. 14-16, Sixteenth Annual Report, 30 FCC Rcd 3253, 3272, ¶ 41, 3294, ¶ 94 n.311 (2015). Federal Communications Commission FCC 15-94 192 that must be purchased by the customer as a package) can have a significant impact on the per-subscriber payments for content. 65. To understand how this distributional effect affects video programming payments, we recalculate AT&T’s per-subscriber payments of the 194 matched channels using AT&T’s current subscriber-share weights, but setting the per-channel prices in this calculation to the lower of the two prices currently received by AT&T and DIRECTV. This calculation estimates the per-subscriber fees that AT&T may end up paying once it is folded under DIRECTV’s contracts after the transaction, were it to keep the distribution of subscribers over these channels at current levels. The calculation allows AT&T to benefit from the lower prices currently paid by DIRECTV but fixes the relative tiering of channels that currently exists between AT&T’s and DIRECTV’s service offerings.113 This estimate is found in row 4 of Table 1. 66. When AT&T receives the more beneficial per-channel fees (but holding constant the share of subscribers receiving each channel), AT&T’s estimated per-subscriber payments are [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] for the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] matched channels. This figure is lower than the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] that AT&T pays for these channels under its current prices by approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent, but higher than the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] that DIRECTV is currently paying for those same channels. In other words, of the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent difference between the payments of the two providers seen in row 3, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percentage points are attributable to lower per-channel prices, and [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percentage points are attributable to DIRECTV offering these channels to a smaller share of subscribers (most likely on higher service tiers). 67. As noted above, there appear to be three factors driving the estimated differences in programming payments across MVPDs: differences in the per-subscriber fees paid for individual channels, differences among the sets of channels that the MVPDs offer to their subscribers, and differences in MVPDs’ distributions of subscribers across available channels (i.e., service tiers). It is important to understand the impact of each of these factors, not only when predicting how AT&T’s programming payments may potentially change as a result of the transaction but also when deciding how to account for these changes in the merger simulation. 68. The BH Simulation implicitly assumes (as is standard in merger simulation analysis) that product characteristics remain fixed between the pre- and post-merger scenarios. Thus, reductions in programming payments that reflect reductions in programming quality should not be considered in the calculation of transaction-related differences in consumer surplus. 69. Of the three factors presented above, only the first – the channel-by-channel differences in fees – potentially represents an unambiguous benefit to consumers were it to be realized by AT&T post-transaction. If, as claimed by AT&T, its per-subscriber programming payments would be reduced to DIRECTV’s level as a result of the transaction, a portion of this savings may be passed on to consumers, which would lead to an increase in consumer surplus. 70. The second factor leading to differences in programming payments between AT&T and DIRECTV (i.e., DIRECTV offering fewer channels to consumers) has an ambiguous consumer welfare effect. Suppose that AT&T were to drop channels that DIRECTV does not offer. Consumers that do not watch these channels may be better off to the extent that the payment reductions are passed through in the 113 For the channel-by-channel data, AT&T’s programming payments are lower than DIRECTV’s for approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . Federal Communications Commission FCC 15-94 193 form of lower video prices. However, consumers that lose channels they previously were able to enjoy would be worse off if the surplus gained from lower prices does not offset the surplus lost from a smaller selection of channels. 71. Similarly, the third factor leading to differences in programming payments – the ability to negotiate placing channels on higher service tiers – also results in an ambiguous consumer surplus impact. Holding prices fixed, consumers that must subscribe to higher tiers to receive channels that they previously received on lower tiers may be made worse off. However, consumers that already subscribed to higher tiers (or that have no interest in channels moved to higher tiers) likely will be better off if the reduced payments result in lower prices. 72. The results presented in Table 1 suggest that the difference in programming payments with quality held constant are likely lower (but no larger) than those claimed by the Applicants. At one extreme, if no consumers are assumed to be harmed if AT&T chooses to drop or move channels to higher service tiers, the quality-constant reduction in programming payments may be up to [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent, assuming AT&T can close the full gap between the payments of the two providers. We note, however, that this difference is still lower than the current [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent difference that AT&T claims in its filings. 73. At the other extreme, if only the payment reductions that result in an unambiguous increase in consumer surplus are considered in the merger simulation (that is, if we completely exclude any reductions AT&T could receive by dropping or retiering channels and consider only the individual differences in prices of channels that are offered by all three MVPDs), the quality-constant payment reductions would likely be significantly smaller, as they are estimated to be only [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent in the channel-by-channel data submitted by the Applicants. This can be seen by comparing rows 3 and 4 in Table 1, which shows the estimated difference between AT&T’s current payments for the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] matched channels and AT&T’s hypothetical programming payments if it paid DIRECTV’s per-channel fees (while excluding the reductions that may arise from adjusting channel tiering). 74. The record in this proceeding provides no evidence quantifying the extent of consumer harms from potential lineup changes, and the Applicants do not provide any evidence that they would be able to achieve the full amount of payment reductions without adjusting their channel lineups. It is possible that a large number of subscribers would not perceive any difference in video service quality if AT&T drops or retiers channels, especially if the subscribers do not watch these affected channels anyway. These subscribers would likely benefit from the payment reductions AT&T achieves through adjusting their channel lineups. Yet some subscribers may perceive large quality changes if channels that they currently enjoy are dropped or moved to higher-priced tiers. The consumer surplus harms to these consumers would need to be subtracted from the consumer surplus gains that are derived from pass- through of reduced programming payments. Because we are unable to determine the changes that the merged entity may make in terms of tiers and channels for both AT&T and DIRECTV subscribers let alone the number of subscribers that would be harmed by such potential changes, we are unable to determine the exact reduction in programming payments that can be considered “quality-constant” reductions as required by the merger simulation. 75. We conclude that it is reasonable to assume that the quality-constant reduction in post- merger programming payments that AT&T could achieve if it could be folded under DIRECTV contracts is likely to lie somewhere between [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent and [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent of AT&T’s current payments. In the BH Simulation, the programming payment reduction is modeled as a flat [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] reduction in the marginal Federal Communications Commission FCC 15-94 194 costs of the AT&T video component. This [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] reduction represents approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent of the marginal costs attributable to programming payments in the model.114 Given that we find a similar [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent difference in the per-subscriber costs of the two firms when considering all channels, we continue to report the results under a [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] programming payment reduction in our analysis.115 This value can be viewed as an upper bound of the potential quality-constant payment reductions AT&T could achieve post-transaction, which would be the case if it is assumed that no consumers would be harmed from future changes in AT&T’s channel tiering. Further, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] is the appropriate programming payment reduction for comparison of our analysis with the analysis submitted by the Applicants. 76. We also report the results under the assumption of a [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] reduction in programming payments, which represents a likely lower bound of the potential quality-constant payment reduction (for purposes of the merger simulation). This lower bound is estimated by crediting only the difference between AT&T’s current per-subscriber payments and the per-subscriber payments it would receive if it were able to receive DIRECTV’s per- channel prices, but excluding payment reductions that would be achieved by dropping or retiering channels. We estimate this difference to be approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent, which is about half of the full [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent difference seen between AT&T and DIRECTV payments in rows 1 and 2 of the table. Therefore, we use half of the full marginal cost reduction that is in the BH Simulation to represent the minimum reduction that we feel that AT&T might receive if it obtained DIRECTV’s fees and held its subscriber shares fixed.116 77. Last, we report the results for no reduction in programming payments for purposes of comparison with the results reported by the Applicants using the BH Simulation, which was also 114 See Berry-Haile Analysis at 109. Additionally, AT&T’s video marginal cost in the BH Simulation is calculated to be approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] per subscriber. In the AT&T Financial Investment Model (“FIM”), the Applicants assume that payments to programmers for content comprise approximately [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent of recurring video expenses. See Katz Additional Detail ¶ 50, nn.49-51. If we apply this same percent factor to marginal costs in the merger simulation, the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] reduction in marginal costs equates to approximately a [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent reduction in programming payments, similar to the future amount claimed by the Applicants absent the transaction. See Application at 36; Katz Decl. ¶ 115; Joint Opposition at 16; Katz Reply Decl. ¶ 32; AT&T Response to Sept. 9, 2014, Information Request at 244. 115 We note that, under the Modified Simulation, the estimated marginal cost for AT&T’s video component is higher than it is under the Corrected Simulation. A reduction of [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent in the estimate of AT&T’s per-subscriber programming costs under the Modified Simulation translates to a marginal cost reduction of [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] , which is higher than the reduction of [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] that the Applicants assumed. We ran the Modified Simulation under this higher marginal cost reduction and obtained results that are very similar to the ones that we report in Section V for the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] reduction. The main difference is that estimated consumer benefits are greater under the more generous reduction in marginal cost. 116 This calculation implicitly assumes that, for the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] unmatched channels that AT&T offers but that do not match with the channels of the other two providers, AT&T is able to achieve the same average percentage reduction in its payments as it does for the 194 matched channels. Federal Communications Commission FCC 15-94 195 performed for the case of no reduction in programming payments.117 For brevity, we will refer to the three cases of no programming payment reductions, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] programming payment reductions, and [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] programming payment reductions as “No PPR,” “Low PPR,” and “High PPR,” respectively. V. RESULTS 78. In this section, we report the results of the Corrected Simulation and the Modified Simulation, each under No PPR, Low PPR, and High PPR (as defined in paragraph 77). We first report estimates of changes in the aggregate level of consumer surplus that are generated by the transaction,118 followed by estimates of changes in prices and shares for a subset of products. We then analyze how consumer surplus levels in individual DMAs are affected by the transaction, and finally we report the results of a number of robustness checks we performed to check the sensitivity of the results to certain assumptions and data quality. A. Consumer Surplus Effects 79. Tables 2 and 3 report the consumer surplus effects of various iterations of the merger simulation for the Corrected Simulation and Modified Simulation, respectively.119 The results from the 117 Reporting this value is not meant to suggest that this is the result we expect to see post-transaction. 118 The change in consumer surplus can be viewed as the additional amount of money that each consumer would have to pay each month following the merger to make him or her indifferent between the merger occurring and not occurring. Thus, a positive change in consumer surplus – which implies that each consumer would be indifferent between the status quo and a post-merger world in which he or she has to pay an additional positive amount – indicates that the merger leads to an increase in consumer welfare. Conversely, a negative change in consumer surplus indicates that the merger is detrimental to consumer welfare. As the change in consumer surplus increases in magnitude, so does the change in consumer welfare. 119 In discrete choice settings, consumer surplus is the expected utility, in monetary terms, that a representative consumer obtains from the choice situation. See, e.g., TRAIN (2009). In the nested logit demand model used here, a representative consumer’s utility from consuming a product is given by , where is a vector of product characteristics, is the product’s price, is a random variable that represents an idiosyncratic component of utility that is unobserved by the econometrician, and and are exogenous parameters that are estimated from choice data. (Note that represents the marginal utility that the consumer obtains from holding money.) The random vector is distributed according to the cumulative distribution function given by: where , the set of nests, is a partition of the product set , is a tuple of exogenous nesting parameters (which are estimated from choice data and dictate substitution patterns among products), and is Euler’s constant. Under this setup, consumer surplus is defined as: A general result due to Daniel McFadden, Modelling the Choice of Residential Location, in SPATIAL INTERACTION THEORY AND PLANNING MODELS (A. Karlqvist, L. Lundqvist, F. Snickars & J. Weibull eds., 1978) implies that expected utility (and, by extension, consumer surplus) in this model has a simple closed-form expression: (continued….) Federal Communications Commission FCC 15-94 196 Corrected Simulation are computed using the data supplied by the Applicants but based on the correction of a coding error in the Applicants’ recentering procedure. We report the results from the Corrected Simulation to provide a baseline with which to compare our results in our Modified Simulation, as they differ from the results presented by the Applicants in the BH Simulation. The consumer surplus changes are reported as weighted averages over the DMA-level results for the 85 DMAs that are included in the merger simulation. 80. The first row of Table 2 shows that, in the Corrected Simulation, relative to the pre- merger world, and under No PPR, the merger of AT&T and DIRECTV reduces consumer surplus by $0.12 per household per month. This is equivalent to an increase in industry-wide prices of slightly less than $0.12.120 The effect of reductions in programming payments is considered in the second and third rows of Table 2. If the transaction results in a [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] reduction in programming payments (i.e., Low PPR), the simulated outcome implies a consumer surplus gain of $0.30 per household per month. Naturally, under a larger [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] reduction in AT&T’s programming payments (i.e., High PPR), the simulated outcome entails a larger $0.77 increase in consumer surplus per household. 81. As described earlier, the net consumer surplus effect can be decomposed into three elements: (1) a reduction in horizontal competition; (2) pass-through from reduced programming payments; and (3) a dividend from bundling. The last three rows of Table 2 present results from decomposing these three elements. This analysis uses a model (referred to here as the “Horizontal Effects Simulation”), which we developed, in which the marginal impact of AT&T’s video service being merged with DIRECTV is evaluated relative to a hypothetical baseline in which AT&T video is a standalone business.121 This hypothetical AT&T-DIRECTV video-only merger is an indicative measure of the purely (Continued from previous page) 120 There is a tradeoff, which results in the equivalent industry-wide price change being slightly less than the change in consumer surplus. On one hand, an increase in price will induce some consumers to exit the market entirely. (This effect pushes the equivalent price change to be slightly greater than the consumer surplus change, because consumers will leave the market if, by doing so, they can limit their losses to be less than $0.12). On the other hand, the presence of synthetic bundles works to oppose this effect. This point is best demonstrated via an example. For a consumer that buys a synthetic bundle pre-merger and continues to buy it post-merger, the effect of a $0.12 across- the-board price increase on her consumer surplus will be -$0.24; that is, the consumer surplus-equivalent price increase on any one component is lower in magnitude than the consumer surplus decrease. The latter effect dominates in the estimated model. 121 The Horizontal Effects Simulation assumes an initial spinoff of AT&T’s U-verse video from the rest of AT&T. This new entity provides only video service and sets its own price, which is formally independent of the price of AT&T broadband service set by the original company (in the sense that the new entity does not take into account the effects of its pricing decision on the profit generated by AT&T’s broadband component). In the first stage of the simulation, firms and consumers respond to this divestiture until a new equilibrium of prices and market shares is established. In the second stage, the new entity merges with DIRECTV in a purely horizontal merger of two competing video components. The welfare effects of this merger are calculated relative to the post-spinoff world, which differs considerably in market structure from the pre-merger world of the original simulation. A direct comparison of the consumer surplus effects between the baseline and Horizontal Effects Simulations (using either the Corrected Simulation or the Modified Simulation) is therefore difficult. The results of the Horizontal Effects Simulation are nevertheless indicative of the relative magnitude of the harms from a reduction in video competition compared to the benefits of bundling observed in the relevant baseline simulation. Federal Communications Commission FCC 15-94 197 horizontal competitive impact of the overall transaction. The Horizontal Effects Simulation is run under the demand model used in the Corrected Simulation as well as the one used in the Modified Simulation. Table 2: Consumer Surplus Effects: Corrected Simulation Assumption Consumer surplus change ($/household/month)122 No PPR -0.12 0.30 Low PPR High PPR 0.77 Horizontal Effects: No PPR Horizontal Effects: Low PPR Horizontal Effects: High PPR -0.49 -0.22 0.07 No PPR: no reduction in programming payments for AT&T video Low PPR: [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] reduction in programming payments for AT&T video High PPR: [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] reduction in programming payments for AT&T video 82. Table 3 is analogous to Table 2, except that it shows the estimated consumer surplus effects under the Modified Simulation. In this case, the merger has a negligible consumer surplus effect under No PPR ($0.02 per household per month) but substantial positive effects under Low PPR ($0.51) and High PPR ($1.11). The Horizontal Effects Simulation returns a reduction in consumer surplus of $0.29 per household per month under No PPR and gains of $0.16 and $0.64 per household per month under Low PPR and High PPR, respectively.123 Table 3: Consumer Surplus Effects: Modified Simulation Assumption Consumer surplus change ($/household/month) No PPR Low PPR 0.02 0.51 High PPR 1.11 Horizontal Effects: No PPR Horizontal Effects: Low PPR -0.29 0.16 122 We do not report standard errors or confidence intervals for these estimated welfare effects, because, although we have computed standard errors for the estimated demand model parameters, the estimated marginal costs and optimal post-merger prices cannot be written as closed-form functions of the demand model parameters. This limitation prevents us from obtaining the consumer surplus point estimates and standard errors analytically based on the demand model parameters; indeed, the firms’ marginal costs and post-merger outcomes are obtained computationally. Monte Carlo methods provide one way to obtain standard errors or confidence intervals for the estimated consumer surplus effects based on the demand parameter estimates; in particular, one could determine a distribution of consumer surplus changes by repeatedly sampling from a multivariate distribution that is based upon the estimated demand model parameters and running the merger simulation for each draw of parameters from the distribution. The Applicants attempted such an exercise with the BH Simulation but found that the standard errors associated with the parameter estimates were too large to generate a representative sample of draws that are consistent with the nested logit demand model. See Berry-Haile Additional Discussion at 41-44. 123 That is, the marginal effect of a loss of a video competitor is unambiguously detrimental to consumer welfare. However, a reduction in the cost of providing service (in the form of either Low PPR or High PPR) more than compensates for the harm created by the reduction in competition. Federal Communications Commission FCC 15-94 198 Horizontal Effects: High PPR 0.64 No PPR: no reduction in programming payments for AT&T video Low PPR: [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] reduction in programming payments for AT&T video High PPR: [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] reduction in programming payments for AT&T video 83. We can compare the original Modified Simulation against the Horizontal Effects Simulation to explore the benefits of bundling. Though the exact size of the bundling effect is difficult to ascertain, as the baseline off of which consumer surplus is calculated differs between the original Modified Simulation and the Horizontal Effects Simulation, the difference in magnitude between the two simulations’ consumer welfare effects is still indicative of the transaction’s potential bundling benefits. In addition, examining the difference in consumer surplus in Table 3 between original and horizontal effects (0.02 – [-0.29] = 0.31) under No PPR against the analogous difference under High PPR (1.11 – 0.64 = 0.47) suggests that bundling becomes even more salient in the presence of programming payment reductions. 84. We also observe, from Table 3, that the marginal welfare effect of High PPR for the actual merger is $1.09 (1.11 – 0.02), while the marginal effect of High PPR in the hypothetical purely horizontal merger is $0.93 (0.64 – [-0.29]). This finding suggests that programming payment reductions increase consumer surplus by passing through the reduction in marginal cost of the standalone AT&T video product to consumers, as well as by reducing the cost of the integrated bundle that includes AT&T video as a component.124 B. Price and Share Effects 85. In Table 4 below, the price changes of selected products – namely, the Applicants’ products and the cable bundle, which is the largest competing product – are reported under the Corrected Simulation under the assumption of No PPR (i.e., the case considered in the top row of Table 2). These quoted prices are monthly per-subscriber prices in dollars. Because of price recentering, the prices are those used in the Corrected Simulation rather than the actual prices charged by the firms. The meaningful measures are the changes reported in dollar amounts and changes in the percentages of households purchasing a particular product. Percentage price changes are reported but are informative only to the extent that the simulation price levels roughly correspond to notional real marketplace prices. They should be viewed as only somewhat indicative at best. 86. In the No PPR case, the prices of AT&T’s standalone video service, DIRECTV’s standalone video service, and the AT&T integrated bundle are all predicted to rise. These effects are all illustrative of the merger’s potential horizontal harm, which was quantified above through the use of the Horizontal Effects Simulation. The price of the newly integrated AT&T-DIRECTV bundle, however, falls due to the fact that the combined firm internalizes the externality that the price of each component of that bundle exerts on the profit earned from sales of the other component. 124 This second cost reduction is not captured in the purely horizontal merger; therefore, the total welfare effect of a reduction in programming payments is somewhat lower. Federal Communications Commission FCC 15-94 199 Table 4: Selected Price Changes under No PPR: Corrected Simulation [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] 87. Table 5 shows the price changes and market effects under Low PPR in the Corrected Simulation for the same products as shown in the previous table. This case corresponds to the second row of Table 2. The AT&T video-only price drops by $0.66 relative to the pre-merger price. However, the price drops by $6.41 (i.e., from $81.73 to $75.32) between the post-merger scenario under No PPR and the post-merger scenario under Low PPR, implying a pass-through of reduced costs to consumers that [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent and is higher than the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent pass-through in the BH Simulation.125 Table 5: Selected Price Changes under Low PPR: Corrected Simulation [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] 88. Table 6 shows the price changes and market effects under High PPR. This case corresponds to the third row of Table 2. Again, the AT&T video only price is assumed to drop by $12.47 relative to the post-transaction scenario with no cost reductions, once again implying a pass-through of [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent. Table 6: Selected Price Changes under High PPR: Corrected Simulation [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] 89. Table 7, Table 8, and Table 9 are exactly analogous to Table 4, Table 5, and Table 6 (respectively), except that they show the results from the Modified Simulation, which employs actual third-party pricing data. 125 See AT&T and DIRECTV, White Paper, Content Cost Savings Will Result in Both Improved Profitability and Pass Through to Consumers, at 8, transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (filed Nov. 12, 2014). The pass-through is defined as the ratio of the price change to the change in marginal cost. As shown by Luke Froeb, Steven Tschantz & Gregory J. Werden, Pass-Through Rates and the Price Effects of Mergers, 23 INT’L J. OF INDUSTRIAL ORG. 703, 703-715 (2005), high estimates of pass-through rates often accompany high estimates of horizontal harm from mergers, because both are closely related to the concavity of the demand function. [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] Federal Communications Commission FCC 15-94 200 90. The results of the Modified Simulation indicated that the transaction likely would result in downward pressure on the prices of AT&T-DIRECTV broadband and video bundles and, to a lesser extent, on the prices of cable bundles. The price of the AT&T-DIRECTV bundle decreases by $2.74 with no programming payment reductions and decreases by $2.20 and $1.38 for programming payment reductions of [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] and [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] , respectively. The price of the AT&T U- verse bundle increases by $1.31 with no programming payment reductions and decreases by $2.93 and $6.70 in the Low PPR and High PPR simulations, respectively. For the cable bundle, the price decreases by $0.08 with no programming payment reductions and decreases by $0.52 and $0.66 in the Low PPR and High PPR simulations, respectively. 91. The directions of the price changes of the merging-parties’ bundles do not differ between the Corrected Simulation and the Modified Simulation, but the magnitudes are notably lower in the Modified Simulation for the AT&T-DIRECTV bundle and larger for the AT&T-AT&T bundle. However, the size of the price drop for the cable bundle is larger in the Modified Simulation than in the Corrected Simulation. 92. A comparison of Table 7 with Table 8 also shows that the post-merger price of the AT&T-DIRECTV integrated bundle is higher in the Low PPR case than in the No PPR case (although it still falls from its pre-merger level). To understand the reason behind this effect, it is helpful to temporarily view the No PPR case as a post-transaction baseline and the Low PPR case as the result of an exogenous reduction in AT&T’s cost of providing video service relative to its post-transaction baseline level (i.e., relative to the post-transaction cost of AT&T’s video service under No PPR).126 All else being equal, this reduction in programming payments leads AT&T to reduce the prices of AT&T video products – including the AT&T-AT&T bundle – which, in turn, will induce some consumers that reside in DMAs in which AT&T offers video service to switch from the AT&T-DIRECTV bundle to the AT&T-AT&T bundle. The consumers that switch away from the AT&T-DIRECTV bundle are likely those that, in the No PPR case, were nearly indifferent between the AT&T-DIRECTV bundle and some other product (possibly, but not necessarily, the AT&T-AT&T bundle) but had a slight preference for the AT&T- DIRECTV bundle. Because, after the price reduction, these nearly indifferent consumers are no longer purchasing the AT&T-DIRECTV bundle, the demand for the AT&T-DIRECTV bundle has become less elastic at its baseline price level in DMAs in which AT&T offers video service. Thus, the firm can profit by slightly raising the AT&T-DIRECTV bundle’s price from its baseline level in these DMAs. The price increase in these DMAs leads to an increase in the average price across all DMAs, which is reported in the tables above. 93. As the above discussion illustrates, the transaction allows the combined entity to partially “recapture” the profit lost from those subscribers that are diverted away from DIRECTV products after the price increase, as some of them will likely subscribe to the now lower-priced AT&T video products. The results of the Modified Simulation indicate that the increased margins on those customers that continue to purchase the higher-priced DIRECTV video products and the increased margins on those customers that purchase the lower-cost AT&T video products (due to [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent pass-through) together outweigh the loss in profit from those that leave the DIRECTV video products.127 126 Note, however, that the cost of providing DIRECTV’s video service does not change. 127 While we have provided an intuitive explanation for the increase in the price of the AT&T-DIRECTV bundle as a result of PPRs, we note that the pricing incentives in the oligopoly pricing game of this environment are fairly complex, which is reflected in our use of computational techniques in obtaining solutions. The challenges in obtaining analytic solutions, even in the simpler monopolistic screening framework, are demonstrated by, for (continued….) Federal Communications Commission FCC 15-94 201 94. The prices of standalone AT&T and DIRECTV video products are slightly higher after the transaction in the Modified Simulation with No PPR. As programming payment reductions are introduced, the price of standalone DIRECTV video tends to increase slightly, while the price of standalone AT&T video falls substantially. The Modified Simulation exhibits a pass-through rate of programming payment reductions of about [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent, which is similar to the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent pass-through of the BH Simulation but [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] than the pass-through found in the Corrected Simulation. Table 7: Selected Price Changes under No PPR: Modified Simulation [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] Table 8: Selected Price Changes under Low PPR: Modified Simulation [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] Table 9: Selected Price Changes under High PPR: Modified Simulation [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] C. DMA Specific Effects 95. In this section, we examine the merger’s effects, as predicted by the Modified Simulation, on individual DMAs. Specifically, we consider the disaggregated results to analyze whether any DMAs are disproportionately harmed by the transaction. 96. We first discuss the market-level welfare effects of the merger under the different PPR assumptions, focusing primarily on the price changes of the merging parties’ products. We then examine DMAs where AT&T has deployed U-verse video to determine whether the degree of U-verse penetration is correlated with post-merger outcomes. Finally, we present heat maps as a visual guide to the differing cross-DMA effects of the merger. 1. Market-Level Welfare Effects 97. Table 10 presents the minimum, maximum, and quartiles of the consumer surplus changes predicted by the merger simulation across DMAs. Without reductions in programming payments, 18 DMAs face a reduction in consumer surplus greater than $0.10 per household per month (Continued from previous page) example, Jean-Charles Rochet & Philippe Choné, Ironing, Sweeping, and Multidimensional Screening, 66 ECONOMETRICA 783, 783-826 (1998). Federal Communications Commission FCC 15-94 202 relative to the pre-merger status quo.128 When we assume Low PPRs or High PPRs, no markets experience a reduction in consumer surplus greater than $0.05. Markets with the highest amount of AT&T U-verse coverage benefit the most from the reduction in programming payments, as the higher availability of U-verse increases the impact of programming payment reduction pass-through. Under High PPR, the median DMA sees an increase in consumer surplus of $0.79 per household per month, which is below the household weighted average change across DMAs of $1.11 that is reported in Table 3 (and is also below an unweighted average across DMAs, which yields a consumer surplus of $0.90), indicating that there are a number of larger DMAs within the U-verse footprint that benefit disproportionately. Table 10: Consumer Surplus Effects across DMAs: Modified Simulation Assumption Consumer surplus change ($/household/month) Min Max 25th percentile 50th percentile 75th percentile No PPR -0.43 0.91 -0.06 0.01 0.12 Low PPR -0.04 1.81 0.11 0.36 0.58 High PPR -0.04 3.89 0.17 0.79 1.25 No PPR: no reduction in programming payments for AT&T video Low PPR: [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] reduction in programming payments for AT&T video High PPR: [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] reduction in programming payments for AT&T video 98. Table 11 and Table 12 report the price and share changes (respectively) across DMAs associated with the consumer surplus changes reported in the first row of Table 10, which corresponds to the case of No PPR. Price changes are reported in dollars per subscriber per month, while share changes are reported as percentage points. 99. Under No PPR, AT&T’s integrated bundle experiences a significant price increase in every DMA, with the smallest increase totaling $0.53 per month.129 Likewise, there are large price increases for AT&T’s and DIRECTV’s standalone video components in most DMAs. However, the reduction in consumer surplus from an increase in the price of AT&T’s video component is mitigated by its small share. All DMAs experience substantial price reductions (of at least $1.07 per month) for the newly integrated AT&T-DIRECTV bundle relative to the synthetic bundle that is available before the merger. 128 Though no margin of error is provided for the consumer surplus estimates, changes in consumer surplus below $0.10 are likely to be small in economic magnitude relative to pre-merger prices. 129 For AT&T’s integrated bundle as well as its standalone video component, Table 11 and Table 12 report statistics across only those DMAs in which AT&T video is available; they exclude the zero price and share changes reported by the model outside of the U-verse footprint. Federal Communications Commission FCC 15-94 203 Table 11: Selected Price Changes across DMAs under No PPR: Modified Simulation Video Broadband Price Change ($/household/month) Min Max 25th percentile 50th percentile 75th percentile DIRECTV AT&T -5.66 -1.07 -3.01 -2.70 -2.40 DIRECTV None -0.08 2.42 0.14 0.58 1.04 Cable Cable -1.12 0.25 -0.17 -0.07 0.01 AT&T AT&T 0.53 2.52 1.03 1.33 1.62 AT&T None 0.58 3.06 1.27 1.53 2.03 None AT&T -1.41 1.08 -0.36 -0.11 0.11 No PPR: no reduction in programming payments for AT&T video Table 12: Selected Share Changes across DMAs under No PPR: Modified Simulation [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] 100. Though not reported here, under Low PPR and High PPR, the post-merger price reduction on the AT&T-DIRECTV integrated bundle is smaller than under No PPR. In fact, eight DMAs experience an increase in the new integrated bundle price under High PPR (compared to zero DMAs under No PPR). However, this reduction in the bundling effect for the merging parties is more than offset by the substantially larger decrease in the price of the cable bundle, as well as the large decrease in price of the AT&T bundle within the U-verse footprint. This also suggests that, while the introduction of an integrated AT&T-DIRECTV bundle, by itself, induces a small competitive response from cable companies, the potential reduction in AT&T’s programming payments magnifies this effect.130 2. U-verse Household Penetration and Market Outcomes 101. Economic theory suggests that consumer harm from the horizontal aspect of the merger should be most pronounced in markets with higher U-verse video household penetration, as more product availability should lead to higher uptake and, therefore, to a larger negative impact of a price increase.131 However, Figure 1 below, which plots U-verse penetration against post-merger change in consumer surplus for all 85 DMAs, paints a more nuanced picture. Non-U-verse video markets (the points along the Y-axis where U-verse penetration is zero) do appear to benefit, but the overall DMA-level impact of the merger appears largely unrelated to U-verse penetration for DMAs within the U-verse footprint. Because the net welfare effect of the merger, both across and within DMAs, depends on the potential reduction in competition as well as on the gains from bundling, Figure 1 suggests either that the welfare loss from reduced competition is uncorrelated with U-verse penetration, or that the bundling effect also varies across DMAs. 130 These results are also in line with our previous observation that bundling becomes more salient in the presence of programming payment reductions. 131 Indeed, the cross-market correlation between U-verse household penetration and market share of standalone U- verse video is 0.45, while the correlation between U-verse penetration and the U-verse bundle is 0.76. Federal Communications Commission FCC 15-94 204 Figure 1: Modified Simulation, No PPR: Consumer Surplus vs. U-verse Video Household Penetration No PPR: no reduction in programming payments for AT&T video 102. In Figure 2, we repeat the above exercise for the Horizontal Effects Simulation and find that the harm from reduced video competition is in fact highly correlated with the degree of U-verse penetration. Taken together with Figure 1, this implies that the gains from bundling do vary systematically by DMA, and, in particular, that these gains are larger in DMAs with higher U-verse penetration. Federal Communications Commission FCC 15-94 205 Figure 2: Modified Simulation – Horizontal Effects Simulation, No PPR: Consumer Surplus vs. U-verse Video Household Penetration No PPR: no reduction in programming payments for AT&T video 103. Finally, in Figure 3, we plot consumer surplus versus U-verse penetration for the Modified Simulation under High PPR. The results are markedly different from the previous graphs: the largest beneficiaries of the programming payment reduction are now the markets with the highest U-verse penetration rates. As discussed above, this result is intuitive, because U-verse video needs to be widely available in a DMA in order for consumers to benefit from any pass-through of PPRs.132 132 See supra ¶ 101. Federal Communications Commission FCC 15-94 206 Figure 3: Modified Simulation, High PPR: Consumer Surplus vs. U-verse Video Household Penetration High PPR: [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] reduction in programming payments for AT&T video 3. Maps of Main Simulation Results 104. We present maps of the continental United States showing changes in consumer surplus within a DMA under various settings for the Modified Simulation. These maps help summarize much of the discussion in this section. We also present a table of the most negatively affected DMAs under the same settings to accompany the maps. 105. Under the Modified Simulation with No PPR, the weighted average post-merger change in consumer surplus is essentially zero. Figure 4 shows that this result is being driven in nearly equal parts by DMAs that suffer small welfare losses and those that enjoy small welfare gains. Under Low PPR and High PPR, the consumer surplus change is positive, at $0.51 and $1.11 respectively. Figure 5 and Figure 6 show that the PPR filters through to all DMAs within the U-verse footprint. Finally, Figure 7 maps the horizontal effects across DMAs. Note that a total of 25 DMAs are outside of the U-verse footprint and thus unaffected by the reduction in competition analyzed in the Horizontal Effects Simulation. In Figure 7, these DMAs are shown in light blue and experience no changes in consumer surplus. Federal Communications Commission FCC 15-94 207 Figure 4: Modified Simulation, No PPR: Predicted Change in Consumer Surplus by DMA No PPR: no reduction in programming payments for AT&T video Figure 5: Modified Simulation, Low PPR: Predicted Change in Consumer Surplus by DMA Low PPR: [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] reduction in programming payments for AT&T video Federal Communications Commission FCC 15-94 208 Figure 6: Modified Simulation, High PPR: Predicted Change in Consumer Surplus by DMA High PPR: [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] reduction in programming payments for AT&T video Figure 7: Modified Horizontal Effects Simulation, No PPR: Predicted Change in Consumer Surplus by DMA No PPR: no reduction in programming payments for AT&T video 106. Table 13 lists the most harmed (or least benefited) DMAs under the various specifications analyzed. As previously indicated, non-U-verse markets do not benefit from a reduction in programming payments.133 Conversely, U-verse markets with high video penetration rates are harmed the most from the horizontal aspect of the merger. 133 Though Table 13 does not present results for the Low PPR case, the markets that benefit the least in that case are identical to the ones that benefit least under High PPR (which are reported in the table). Federal Communications Commission FCC 15-94 209 Table 13: Most Harmed Markets by Simulation Specification Modified Simulation - No PPR:          State Market ID Market Name CS Change U-verse HH Penetration    TX 618 HOUSTON -0.43 0.70    AR 670 FT. SMITH-FAY-SPRNGDL-RGRS -0.35 0.54    TX 635 AUSTIN -0.31 0.71    OH 535 COLUMBUS, OH -0.26 0.57    MO 616 KANSAS CITY -0.24 0.73    IN 527 INDIANAPOLIS -0.22 0.71    TX 623 DALLAS-FT. WORTH -0.21 0.62    MO 609 ST. LOUIS -0.20 0.67    AR 693 LITTLE ROCK-PINE BLUFF -0.20 0.38    TX/NM 765 EL PASO (LAS CRUCES) -0.20 0.51 Modified Simulation - High PPR:          State Market ID Market Name CS Change U-verse HH Penetration    OR 813 MEDFORD-KLAMATH FALLS -0.04 0    IL 632 PADUCAH-CAPE GIRARD-HARSBG -0.01 0    NV 839 LAS VEGAS -0.01 0    MO 604 COLUMBIA-JEFFERSON CITY 0 0    WI 705 WAUSAU-RHINELANDER 0 0    NE 722 LINCOLN & HASTINGS-KRNY 0 0    UT 770 SALT LAKE CITY 0 0    IN 509 FT. WAYNE 0.01 0    WI 676 DULUTH-SUPERIOR 0.01 0    FL 539 TAMPA-ST. PETE (SARASOTA) 0.02 0 Horizontal Effects Simulation - No PPR:          State Market ID Market Name CS Change U-verse HH Penetration TX 618 HOUSTON -0.70 0.70 AL 630 BIRMINGHAM (ANN AND TUSC) -0.67 0.71 GA 524 ATLANTA -0.60 0.84 NV 811 RENO -0.57 0.56 TX 623 DALLAS-FT. WORTH -0.55 0.62 MO 609 ST. LOUIS -0.54 0.67 CA 862 SACRAMNTO-STKTON-MODESTO -0.52 0.68 FL 686 MOBILE-PENSACOLA (FT WALT) -0.50 0.56 CA 825 SAN DIEGO -0.50 0.94 IN 527 INDIANAPOLIS -0.49 0.71 No PPR: no reduction in programming payments for AT&T video High PPR: [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] reduction in programming payments for AT&T video 107. The DMA-specific analysis in this section suggests that the proposed merger would likely reduce video competition inside of the U-verse video footprint and, in particular, in DMAs where U-verse reaches a large proportion of TV households. The net welfare loss from these harms is generally balanced against the welfare gains from a more competitive AT&T-DIRECTV integrated bundle, though a non- Federal Communications Commission FCC 15-94 210 trivial number of DMAs do experience small reductions in consumer surplus. Under High PPR, the Modified Simulation predicts significant welfare gains for all DMAs within the U-verse video footprint. These gains outweigh the horizontal harms. The Horizontal Effects Simulation likewise predicts a non- trivial increase in consumer surplus when incorporating PPRs. Given that the benefits of bundling are even stronger when PPRs are introduced into the simulation, the predicted effect of the merger across DMAs depends largely on the magnitude of the realized PPR. D. Robustness: Outliers and Price Winsorization 108. As noted previously, the AT&T and DIRECTV subscriber price data that is fed into the simulations shows a very high degree of variability within each plan, including extreme values. To help determine if the results were being driven by the presence of extreme values, two robustness checks – both involving winsorization, a form of data censoring in which extreme values below a predetermined lower bound are replaced with the lower bound and extreme values above a predetermined upper bound are replaced with the upper bound – were performed.134 In the first robustness check, prices across all plan types were winsorized using the first and 99th percentile values as the lower and upper bounds, respectively. The second robustness check was motivated by the concern that censoring prices may affect bundled plans (which are, on average, more expensive than video-only and broadband-only plans) disproportionately; thus, in the second robustness check, prices within each nest (video-only plans, broadband-only plans, and bundles) were winsorized using the nest’s first and 99th percentile values. In both robustness checks, all other simulation processes were left unchanged. We refer to the simulations performed under these two robustness checks as “Winsorized” and “Nest-Winsorized,” respectively. 109. As shown in Table 14, we find that the simulation results are sensitive to alterations in the underlying pricing data. In the Corrected Simulation, consumer surplus is slightly negative, at -$0.12 under No PPR, and positive, at $0.77, under High PPR. Compared to this baseline, winsorizing over plan types improves consumer surplus post-merger. Consumer surplus rises to $0.15 under No PPR and to $0.97 under High PPR. Winsorization within product nests does not qualitatively change the results, as consumer surplus increases slightly to $0.20 and $1.01, under No PPR and High PPR, respectively. Table 14: Consumer Surplus Estimates with Price Winsorization Simulation and Data Consumer surplus change ($/household/month), Consumer surplus change ($/household/month), No PPR High PPR Corrected Simulation - Original -0.12 0.77 Corrected Simulation - Winsorized 0.15 0.97 Corrected Simulation - Nest-Winsorized 0.20 1.01 Modified Simulation - Original 0.02 1.11 Modified Simulation - Winsorized 0.67 1.47 Modified Simulation - Nest-Winsorized 0.56 1.38 No PPR: no reduction in programming payments for AT&T video High PPR: [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] reduction in programming payments for AT&T video 110. We also performed the price-winsorization robustness checks in the Modified Simulation. Running the same analysis using the updated third-party pricing data in the Modified Simulation, we find 134 We also investigated simply dropping (omitting) the extreme observations. This was found to interact with the data handling structures employed by the simulation code in problematic ways, leading to computational and convergence challenges. Federal Communications Commission FCC 15-94 211 that the results continue to be sensitive to the underlying data. Recall that, under the Modified Simulation, consumers are more price-sensitive than in the Corrected Simulation. When prices are winsorized across all plan types, consumer surplus rises to $0.67 and $1.47 under No PPR and High PPR, respectively. Winsorization within product nests slightly dampens the increase in overall consumer surplus. We find that consumer surplus falls to $0.56 and $1.38 under No PPR and High PPR, respectively. E. Robustness: Appropriate Setting of the Pre-Merger Synthetic Bundle Discount 111. The BH Simulation compares post-merger outcomes against the status quo and includes fixed [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] price discounts that are offered on synthetic bundles that combine DIRECTV video service with broadband service offered by AT&T, Telco providers, and one Cable provider. (Each discount is borne equally by both of the firms whose components compose the associated bundle.) These discounts are exogenous, and, in principle, it is possible that an alternative discount level exists for the synthetic bundle of DIRECTV video service and AT&T broadband service that would increase the merging parties’ individual pre-merger profits while also enhancing consumer surplus. In this case, one might argue that a counterfactual scenario, in which the parties remain unmerged but coordinate in modifying their synthetic bundle discount to a level that improves each of their individual profits as well as consumer welfare, might provide a more appropriate benchmark for welfare comparisons than the status quo does. 112. We explored this possibility by reprogramming the setting of the discount within the simulation under two scenarios. In the first scenario, the merging parties coordinate in setting a discount level, but all other prices (including those of other providers) remain fixed at status quo levels. In the second scenario, the merging parties first coordinate in setting and announcing a discount level, and then all firms (including the merging parties) play a Nash equilibrium in pricing their components. (In both scenarios, the only synthetic bundle discount that might change is the one that combines DIRECTV’s video service with AT&T’s broadband service; all other discounts remain fixed at [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .) The results from these exercises did not indicate that an alternative discount level would improve both of the merging parties’ profits while also substantially enhancing consumer welfare. Thus, we believe that the current [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] discount level is unlikely to hide any realizable efficiency that significantly skews the results in the Applicants’ favor, and therefore the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] discount is retained for assessing post-merger welfare gains. VI. COMPARISON WITH OTHER STUDIES 113. While we are not aware of any empirical analyses of the same product set that has been analyzed here, there are several comparable studies, which we discuss below, that look at subsets of this product set. In this section, we demonstrate that the demand model estimated in the Modified Simulation comports fairly well with the demand models that have been estimated in these studies. 114. Comparing studies that consider distinct (though partly overlapping) product sets and that use different model specifications is not straightforward; because of such differences, we are limited to comparisons of quantities that are reported (or that can be inferred from reported quantities) in the Modified Simulation and other analyses and that carry the same interpretations across models. Given these considerations, we focus on comparisons of own-price demand elasticities (and, when they can be computed, own-price demand semielasticities as well).135 135 Let be a set of products, and, for each , let denote the demand (as a function of the vector of product prices) for product . The own-price demand elasticity for product at prices can be interpreted as the ratio of the percentage change in product ’s demand to the percentage change in its price from initial prices given by . It is defined as: (continued….) Federal Communications Commission FCC 15-94 212 A. Studies of the U.S. Broadband Internet Service Market 115. Table 15 summarizes the own-price demand elasticities and semielasticities of broadband Internet access service products from various studies, including the Modified Simulation. Table 15: Comparison of Own-Price Demand Elasticity and Semielasticity Estimates for Broadband Internet Service Products from Different Studies Provider or Technology Study Years Elasticity Semielasticity All Modified Simulation 2013-14 -0.66 -0.0145 Carare et al. (2015)136 2011 -0.62 Not available Dutz et al. (2009)137 2005 -1.53 -0.0375 2006 -1.17 -0.0299 2007 -0.88 -0.0238 2008 -0.69 -0.0195 Goolsbee (2006)138 1998 -2.75 -0.0688 (Continued from previous page) For example, if , then, starting from a price vector of , a 1 percent increase in would lead the demand for product to fall by approximately 0.3 percent. (Note that, because the elasticity captures an instantaneous change, this figure is a first-order approximation.) As mentioned in Section II.B.2, one consequence of the method of estimation (and, in particular, of the construction of price indices) is that only the differences in prices across products are identified. Hence, the price levels themselves (and, by extension, the percentage changes in prices) are not economically meaningful. While the recentering procedure makes the constructed price indices more “price-like,” the caveat remains that, due to the invariance of consumer choice behavior to different price normalizations, the estimated elasticities may not carry their usual interpretations. To address this problem, wherever possible, we compute the own-price demand semielasticity for product at prices , which can be interpreted as the ratio of the instantaneous proportional change in product ’s demand to the instantaneous level change in its price from initial prices given by : For example, if , then, starting from a price vector of , an increase of $0.01 (or, more generally, of 0.01 units) in would lead the demand for product to fall by approximately 0.2 percent. 136 Octavian Carare, Chris McGovern, Raquel Noriega & Jay A. Schwarz, The Willingness to Pay for Broadband of Non-Adopters in the U.S.: Estimates from a Multi-State Survey, 30 INFO. ECON. AND POL’Y 19, 19-35 (2015) (“Carare et al. Study”). 137 Mark Dutz, Jonathan Orszag & Robert Willig, The Substantial Consumer Benefits of Broadband Connectivity for U.S. Households, Internet Innovation Alliance Report (2009), available at http://internetinnovation.org/files/special- reports/CONSUMER_BENEFITS_OF_BROADBAND.pdf (visited June 17, 2015) (“Dutz et al. Study”). Federal Communications Commission FCC 15-94 213 Rappoport et al. (2003)139 2000 -1.491 Not available Cable Modified Simulation 2013-14 -3.67 -0.0814 Dutz et al. (2009) 2005 -5.12 -0.1212 2006 -5.48 -0.1287 2007 -5.59 -0.1322 2008 -5.21 -0.1312 Rappoport et al. (2003) 2000 -0.587 -0.0145 AT&T and Telcos Modified Simulation 2013-14 -4.66 -0.1033 DSL Dutz et al. (2009) 2005 -5.71 -0.1490 2006 -4.78 -0.1365 2007 -3.98 -0.1299 2008 -4.04 -0.1356 ADSL Rappoport et al. (2003) 2000 -1.462 Not available Fiber Dutz et al. (2009) 2007 -8.70 -0.2198 2008 -8.11 -0.2160 116. Because the Modified Simulation uses more recent data than do the other studies that are cited in Table 15, it is appropriate to compare the figures from the Modified Simulation to extrapolated versions of the ones from the other studies. We first compare the Modified Simulation to the Dutz et al. Study and argue that the elasticities from the Modified Simulation are broadly consistent with a few notable patterns that the Dutz et al. Study illustrates.140 117. First, the Dutz et al. Study notes that the own-price elasticity of broadband Internet service is increasing (i.e., decreasing in magnitude), which they argue reflects an increasing tendency among households to view broadband Internet service as a necessity rather than as a luxury. The elasticity and semielasticity estimates are consistent with this trend. 118. Second, while the estimates of the Dutz et al. Study do not illustrate a clear trend with respect to the own-price elasticity and semielasticity of cable broadband, the values derived from the Modified Simulation are not drastically different from the estimates of the Dutz et al. Study, and the fact that the estimates from the Modified Simulation are lower in magnitude could plausibly reflect the fact (Continued from previous page) 138 Austan Goolsbee, The Value of Broadband and the Deadweight Loss of Taxing New Technology, NBER Working Paper (2006), available at http://www.nber.org/papers/w11994.pdf (visited June 16, 2015) (“Goolsbee Study”). 139 Paul Rappoport, Donald J. Kridel, Lester D. Taylor & James Alleman, Residential Demand for Access to the Internet, (2003), available at http://www.colorado.edu/engineering/alleman/print_files/Forecasting_the_Demand_for_Internet_Services.PDF (visited June 16, 2015) (“Rappoport et al. Study”). 140 In particular, simple inspection of the estimates in Table 15 do not raise serious concerns about the view that the Modified Simulation estimates and the Dutz et al. Study estimates are based on draws from a common time- dependent data generating process, particularly in light of the technological trends that we discuss here. Federal Communications Commission FCC 15-94 214 that, given the increased demand for streaming video content (especially in the years since the Dutz et al. Study), DSL broadband service is increasingly being viewed as an inferior product relative to cable broadband service.141 119. Finally, the estimated own-price elasticity for broadband service provided by AT&T and other Telcos reflects the underlying demands for both DSL service and fiber service. As in the case of cable service, the estimates by the Dutz et al. Study for those technologies seem roughly consistent with their analogues from the Modified Simulation, especially in light of the fact that fiber broadband service was a relatively new product during the two years in which the Dutz et al. Study includes it in the analysis. Due to consumers’ unfamiliarity with this product, and also given its fierce competition with cable broadband service – a much more well-established and familiar product – it is entirely plausible that fiber broadband service would have initially exhibited relatively elastic demand that has tapered off as consumers have become more familiar with it. 120. The results of the Goolsbee Study also appear to be roughly consistent with those of the Dutz et al. Study, especially keeping in mind that Goolsbee’s estimates reflect data from 1998. Interestingly, though, the estimates of the Rappoport et al. Study suggest that, even in 2000, the demand for broadband service was significantly less elastic than the estimates from the other studies mentioned above seem to indicate. B. Studies of the U.S. MVPD Market 121. Two studies of the U.S. MVPD market that allow for meaningful comparisons with the Modified Simulation are due to Goolsbee and Petrin in 2004142 and Crawford and Yurukoglu in 2012.143 Table 16 provides their estimates of own-price demand elasticities (and implied estimates of own-price demand semielasticities) of U.S. MVPD products. 141 Of course, the introduction of fiber broadband service is a countervailing factor that, all else equal, should lead to increasingly elastic demand for cable broadband service. However, given the limited availability of fiber broadband service, its increasing strength as a disciplining force on cable broadband is likely less important than the dwindling strength of DSL broadband service – which is much more widely available – as a disciplining force. 142 Austan Goolsbee & Amil Petrin, The Consumer Gains from Direct Broadcast Satellites and the Competition with Cable TV, 72 ECONOMETRICA 351, 351-381 (2004) (“Goolsbee-Petrin Study”). 143 Crawford-Yurukoglu Study at 643-685. Federal Communications Commission FCC 15-94 215 Table 16: Comparison of Own-Price Demand Elasticity and Semielasticity Estimates for MVPD Products from Different Studies Technology Service Type Study Years Elasticity Semielasticity Cable All Modified Simulation 2013-14 -8.07 -0.0792 Basic Goolsbee-Petrin (2004) 2001 -1.538 -0.0567 Crawford- Yurukoglu (2012) 1997-2007 -4.12 -0.1935 Premium Goolsbee-Petrin (2004) 2001 -3.175 -0.0823 Expanded Basic Crawford- Yurukoglu (2012) 1997-2007 -6.34 -0.2250 Digital Basic -13.11 -0.2942 Satellite All Modified Simulation 2013-14 -7.82 -0.0990 Goolsbee-Petrin (2004) 2001 -2.448 -0.0612 Crawford- Yurukoglu (2012) 1997-2007 -5.35 Not available 122. Again, we may view the estimated elasticities from the Modified Simulation as newer versions of their analogues from the other studies referenced in Table 16. Leaving aside the usual caveats regarding estimation errors and differences in methodologies across studies, the figures suggest that demand for cable MVPD service has become more elastic in recent years. This trend is certainly plausible given a number of factors, including the entry of fiber-based competitors (especially Telcos since 2006), the launch of a new satellite by DISH in 2006, which expanded DISH’s channel capacity significantly, and, probably to a lesser extent, the recent advent of online video distributors (“OVDs”) and the resulting increase in so-called “cord-cutting” (i.e., dropping traditional video service and relying solely on access to OVDs through a broadband connection for video service). The first and third factors listed above also may account for an increase (which the above figures suggest) in the elasticity of demand for satellite-based MVPD service. 123. The Modified Simulation semielasticities, compared with the ones based on the estimates of the Goolsbee-Petrin Study, are consistent with the above story as well. Interestingly, though, the Modified Simulation semielasticity for cable MVPD service seems to be much lower in magnitude than the ones derived from Crawford and Yurukoglu’s analysis. This discrepancy appears to be driven largely by price increases in recent years, as the elasticities themselves are not very different. VII. IMPLICATIONS 124. In considering the implications of the BH Simulation, it is important to keep in mind the limited scope of the exercise. The merger simulation is designed to provide a data-driven, quantitative answer to the following question: “Assuming that all industry participants’ product offerings remain the Federal Communications Commission FCC 15-94 216 same, what price changes arise from the changed pricing incentives created by the proposed transaction?” That is, the simulation speaks to the price effects of the transaction, holding constant the industry product mix. In particular, the simulation ignores any effect that the post-merger integration of the (currently synthetic) bundle that combines DIRECTV’s video programming service with AT&T’s broadband Internet access service would have on consumer welfare other than through a potential change in its price. 125. To answer this question, the BH Simulation simplifies many features of the industry. Notably, it abstracts away from the price discrimination (tiering) that industry participants engage in, as well as from any dynamic considerations (such as the incentives for investment, incentives to change the product mix, and frictions due to consumer switching costs).144 Finally, to make it possible to find results, any merger simulation imposes specific forms on demand systems and nesting behavior that are unlikely to match consumer behavior exactly. 126. These simplifications are made to address data limitations and reflect the limits of current modeling technology. They are common in economic analyses of the industry. The model used in the simulation was judged to be an appropriate representation of the state-of-the-art in merger simulation methods. Given the available data and the industry setting, the approach adopted by the parties represents best practice in building a merger simulation. Where we adopted slightly different models, it was due either to minor issues found in code submitted by the Applicants or to the ability of the Commission to access data unavailable to the Applicants. None of these adjustments changed our receptiveness to the general approach. 127. That said, results of the merger simulation should be viewed as only contributing to an understanding of the competitive impact of the proposed transaction. With this caveat in mind, the rest of this section highlights the implications of the Modified Simulation for understanding the likely price effects of the transaction, holding constant the industry product mix. 128. In discussing the quantitative implications, we begin by considering the Modified Simulation under Low PPR, which we believe represents a plausible outcome while also providing a conservative estimate for the consumer welfare gain, as measured by the change in consumer surplus. The Corrected Simulation predicts this number to be $0.30 per month (Table 1), while the Modified Simulation predicts this number to be $0.51 per month (Table 2). The robustness checks that we did, together with information from the Applicants’ submissions,145 suggest that these estimates have a non- trivial margin of error around them. While it is difficult to systematically quantify the effect of this margin of error, the robustness checks reported above suggest that the consumer surplus impact – the net effect of the horizontal harm, bundling benefit, and PPR – is likely to be positive, though perhaps modest. In particular, it is unlikely that consumers will be harmed at an aggregate level. Indeed, our best measure of the net effect of the horizontal harm and the bundling benefit under No PPR is that they essentially cancel each other out at the aggregate level. To the extent that we believe that the Low PPR scenario is likely to occur, we also believe that, if anything, the merger is likely to yield modest benefits to consumers in the form of lower prices. 129. The simulation is also capable of giving predictions of consumer impact by location, finding difference in the merger’s effect across DMAs. Here, the most harm done to any one DMA is a loss of $0.04 per household per month in consumer surplus when High PPRs are assumed. For many DMAs, however, the benefit is of a significantly higher magnitude. 130. It is useful to consider a bound on the harm that this transaction may generate. To do this, we consider the Horizontal Effects Simulation, in which AT&T’s video service starts as a standalone 144 See Berry-Haile Analysis at 13. 145 See Berry-Haile Additional Discussion at 41-44. Federal Communications Commission FCC 15-94 217 business that is then merged with DIRECTV. To get a worst-case estimate of the horizontal harm that the transaction may generate, we focus here on the No PPR scenario. In this case, aggregate consumer surplus falls by $0.29 per household per month. Evaluated DMA by DMA, the greatest potential for harm is in the Houston area, where the purely horizontal effect is estimated to be -$0.70 per month. The numbers serve as an indication of the downside risk associated with the merger. Our judgment is that, while these numbers indicate a potential for harm, the size of this harm, when isolated, is not so large, in and of itself, as to destroy the credibility of a claim that the improved pricing incentives involved in AT&T and DIRECTV moving from a synthetic bundle to an integrated bundle (in addition to the downward pressure this places on cable bundle prices) may provide an offsetting force. Furthermore, these figures are based on the No PPR scenario. Under the more plausible Low PPR scenario, the Horizontal Effects Simulation actually predicts a consumer surplus gain, underscoring the importance of potential downward pricing pressure as a result of PPRs. 131. In a setting in which PPRs are considered, two questions arise for the economic analysis. The first is what the appropriate level of reduction might be to input into the model, and the second is how to interpret the results that the model generates. 132. An examination of the programming data, discussed in Section IV.C, suggests that a [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] reduction in programming payments (i.e., Low PPR) is plausible (though arguably conservative). This is in contrast to the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] reduction (i.e., High PPR) used in the BH Simulation. 133. Before inputting any assumption regarding PPRs into the simulation, it is important to consider the economic structure imposed by the simulation. A reduction in costs will, regardless of the simulation, be passed through to consumers in some degree in the form of lower prices. All else being equal, this is a benefit to consumers.146 134. When applied to the simulation, the assumption of Low PPR is sufficient to make the aggregate effect on consumer surplus positive and sufficiently large to be economically distinguishable from zero. Hence, the evidence presented by the merger simulation suggests an overall effect that is positive for consumers, not only in the aggregate, but also at the DMA level for all but three DMAs. 146 As noted before, the BH Simulation abstracts away from dynamic incentives and ignores programmers (and thus programming inputs) completely. As a result, the analysis presented here does not directly address a theory of harm based on the premise that PPRs will tend to reduce programmers’ incentives to invest in high-quality programming, which will eventually lead to quality reductions and, ultimately, adverse effects on consumer welfare. Using the programming payment data presented in Section IV.C, we obtained a worst-case estimate for the total fall in affiliate fees paid by MVPDs to programmers as a result of the merger. This analysis used a run of the Modified Simulation under the assumption of a reduction of [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] per subscriber per month in AT&T’s affiliate fees. In this scenario, programmers were estimated to lose about [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent of their affiliate fees. Based on SNL Kagan, TV Networks: Economics Profile & Peer Comparison, rel. March 10, 2015, we estimate that, all else being equal, this drop in affiliate fees would constitute a loss of [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent of total programmer revenues and a loss of [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] percent of total programmer profits. The evidence in the record does not support the assertion that these reductions represent anything beyond a redistribution of surplus between programmers and MVPDs. In the absence of evidence that indicates a loss of efficiency or a harm to consumers, the Commission remains agnostic on normative aspects of the division of this surplus. Federal Communications Commission FCC 15-94 218 APPENDIX D Analysis of AT&T’s FWLL Coverage and Performance Claims and Claimed Rural Benefits TABLE OF CONTENTS Heading Paragraph # I.  INTRODUCTION .................................................................................................................................. 1  A.  Overview .......................................................................................................................................... 1  B.  Terminology ..................................................................................................................................... 2  II.  COVERAGE AND PERFORMANCE CLAIMS ................................................................................ 17  A.  Background .................................................................................................................................... 17  B.  Analysis ......................................................................................................................................... 27  1.  Coverage .................................................................................................................................. 28  2.  Performance ............................................................................................................................ 37  III. AT&T’S CLAIM THAT FWLL WOULD BENEFIT 13 MILLION RURAL CUSTOMERS ........... 42  A.  Background .................................................................................................................................... 42  B.  Analysis ......................................................................................................................................... 44  I. INTRODUCTION A. Overview 1. This Appendix contains the Commission staff’s analysis of certain technical claims made by the Applicants in connection with their proposed transaction. Section II covers claims the Applicants have made with respect to the performance and coverage of their proposed Fixed Wireless Local Loop service (“FWLL”) network. Section III analyzes the claims the Applicants put forth in their Application that an additional benefit of the transaction is the deployment of the FWLL to 13 million largely rural households.1 B. Terminology 2. This section gives a brief introduction to the terms used in this Appendix. 3. ArcGIS is a geographic information system that can be used for the management, analysis, and display of geographic data.2 4. Capacity of a cell site is typically measured by the number of simultaneous voice calls 1 AT&T initially described the potential reach of FWLL as 13 million “customer locations,” which AT&T never defined. See Application, Description of Transaction, Public Interest Showing, and Related Demonstrations, transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, at 5, 39 n.121 (filed June 11, 2014) (“Application”). However, AT&T later determined that the correct term should be “households.” See, e.g., ATT-FCC-00408122, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ; ATT-FCC-02115462, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . AT&T nonetheless continued to use the terms interchangeably. See ATT-FCC-00408122, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ; AT&T Inc. Response to Sept. 9, 2014, Information and Discovery Requests, transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Vanessa Lemmé, Media Bureau, FCC, MB Docket No. 14-90, at 204 (Oct. 7, 2014) (“AT&T Response to Sept. 9, 2014, Information Request”) (using households as the unit of calculation of a percentage). For the purposes of our analysis, we use the term “households.” 2 See ESRI, ArcGIS 9 – What is ArcGIS, http://downloads.esri.com/support/documentation/ao_/698What_is_ArcGis.pdf (visited June 18, 2015). Federal Communications Commission FCC 15-94 219 that can be made on the cell or the total volume of data throughput provided by the cell.3 This capacity is a function of both the amount of spectrum available in the cell and the radio technology used.4 The total capacity of the radio network is determined by the capacity of the individual cells. The network traffic is often distributed unevenly over the cells, with the result that the busiest cells drive the need to increase network capacity. Common ways to increase network capacity are adding cells, adding spectrum, optimizing cells to redistribute traffic more evenly, and increasing spectral efficiency through network technology or handset upgrades. Network capacity can be added by building more cells, so theoretically no more spectrum is needed. However, spectrum acquisition is often more economically attractive than cell splitting, in part because acquiring additional spectrum in a geographic area increases the potential capacity of all cells in that area. 5. A Cell Site is a transmitter/receiver location through which radio links are established between a wireless communications system and a wireless unit.5 Cell sites typically include a support structure (i.e., a tower, building or other structure that provides a desired height above the ground), antennas, cables, radios, processors, etc. One site contains one or more sectors,6 with most sites having three sectors.7 A sector corresponds to a geographic cell of radio coverage that uses a portion of the spectrum to communicate with a number of subscriber devices.8 6. Cell Splitting refers to building new sites to increase capacity. A new site is placed so that at least one of its sectors overlaps with a congested sector on a pre-existing site, taking over some of its coverage area and some of its traffic. This effectively splits a congested cell of coverage into two or more cells that can share the traffic load.9 3 SIEGMUND M. REDL, MATTHIAS K. WEBER & MALCOLM W. OLIPHANT, AN INTRODUCTION TO GSM 6 (1995). 4 See generally SAMI TABBANE, HANDBOOK OF MOBILE RADIO NETWORKS 288-300 (2000) (“HANDBOOK OF MOBILE RADIO NETWORKS”). 5 See HARRY NEWTON, NEWTON’S TELECOM DICTIONARY 770 (20th ed. 2004) (“NEWTON’S TELECOM DICTIONARY”) (defining cell site as “a transmitter/receiver location, operated by the WSP (Wireless Service Provider), through which radio links are established between the wireless system and the wireless unit …”); see also HANDBOOK OF MOBILE RADIO NETWORKS at 206-207 (“The cellular architecture was originally designed as a means of providing a region of substantial geographic size … with a communications network using a limited frequency allocation and servicing an increasing traffic demand …. The mechanism is based on the path loss property of radio waves, which means that a frequency used on one site can be reused on another site provided that the two sites are sufficiently far from each other. Each site covers an area called a cell, the size of which usually depends on user density.). 6 Sectorization is defined in the Commission’s rules as follows: The use of an antenna system at any broadband station, booster station and/or response station hub that is capable of simultaneously transmitting multiple signals over the same frequencies to different portions of the service area and/or simultaneously receiving multiple signals over the same frequencies from different portions of the service area. 47 C.F.R. § 27.4. See also HANDBOOK OF MOBILE RADIO NETWORKS at 220 (providing graphical examples of sites having between one to three sectors). 7 See HANDBOOK OF MOBILE RADIO NETWORKS at 295. 8 See id. (Each sector can be considered a new cell as it uses a different set of channels and a directional antenna.). 9 Id. at 293 (“The cell splitting technique consists of reducing cell sizes with an immediate consequence of increasing network capacity. Each cell is split up into a number of cells of a smaller size.”). Federal Communications Commission FCC 15-94 220 7. Data Speed refers to the speed of the data transmission of a network, typically measured in million bits per second (“Mbps”).10 8. GeoLytics, Inc. provides projections of the most recent census data available. The analysis in this Appendix uses GeoLytics’ 2013 census block-level projections.11 9. High-Gain Antenna (“HGA”) refers to an antenna with a narrow radio beam that is used to increase signal strength. High-gain antennas provide a more precise way of targeting radio signals and are therefore very essential to long-range wireless networks. They even amplify weak signals used in satellite communication.12 10. LTE is an acronym for Long Term Evolution, which is the latest mobile network technology standard set by the 3rd Generation Partnership Project (“3GPP”).13 LTE is typically used in so-called “4G” networks. 11. MIMO, or multiple input multiple output, is an advanced antenna technology for wireless communications in which multiple antennas are used at both the source (transmitter) and the destination (receiver). The antennas at each end of the communications circuit are combined to minimize errors and optimize data speed.14 12. Radio Channel is a frequency, or band of frequencies, assigned to a transmitter/receiver station in the cell site and used in the cell for transmissions between the station and wireless units.15 13. Radio Frequency carrier (or “RF carrier”) refers to both the radio equipment for a radio channel and the signals broadcast over the air on that radio channel.16 For example, both Universal Mobile Telecommunications System (“UMTS”) and LTE can operate with 5+5 megahertz RF carriers, where 5 megahertz is used for uplink transmissions from subscriber devices to the network and another 5 10 See Transmission Speed Definition from PC Magazine Encyclopedia, PC MAGAZINE, http://www.pcmag.com/encyclopedia/term/53109/transmission-speed (visited June 18, 2015). 11 See Block Level Estimates, GEOLYTICS, http://www.geolytics.com/USCensus,Block- Estimates,Data,Features,Products.asp (visited June 18, 2015). 12 See High-Gain Antenna (HGA), TECHOPEDIA, http://www.techopedia.com/definition/26056/high-gain-antenna- hga (visited June 25, 2015). 13 See LTE Encyclopedia, https://sites.google.com/site/lteencyclopedia/home (visited June 18, 2015). See also Magdalena Nohrborg, 3GPP, LTE, http://www.3gpp.org/technologies/keywords-acronyms/98-lte (visited June 18, 2015). 14 See Margaret Rouse, What is MIMO (multiple input, multiple output), TECHTARGET, http://searchmobilecomputing.techtarget.com/definition/MIMO (visited June 18, 2015). 15 See Radio Channel Definition and Meaning, DICTIONARY OF ENGINEERING, http://www.dictionaryofengineering.com/definition/radio-channel.html (visited June 18, 2015). 16 CARL J. WEISMAN, THE ESSENTIAL GUIDE TO RF AND WIRELESS 9 (1st ed. 2000) (“ESSENTIAL GUIDE TO RF AND WIRELESS”) (“Frequency is what separates one [Radio Frequency or RF] signal from another and it is what distinguishes one wireless application from another.”); id. at 11-12 (“Only analog signals (sine waves) are used to carry information ‘on their backs’ as they travel through the air. These analog ‘carrier’ signals can carry either analog or digital ‘information’ signals. The process of combining information signals on top of carrier signals is called modulation …. When an information signal is combined with a carrier signal the result is known as wireless communications, and the analog signal doing the carrying is called RF or the carrier ….”); GEORGE CALHOUN, DIGITAL CELLULAR RADIO 202-203 (1988) (“DIGITAL CELLULAR RADIO”) (“Most radio transmission utilizes a continuous wave of a fixed frequency, called the carrier …. The modulated carrier – i.e., the carrier with the information … actually occupies a narrow region of the spectrum …. [T]he width of this region – the occupied bandwidth – is also measured in KHz or MHz. This is what is commonly referred to as a radio channel.”). Federal Communications Commission FCC 15-94 221 megahertz is used for downlink transmissions from the network to subscriber devices. One RF carrier can support many devices at once. 14. Spectrum is the set of radio wave frequencies used by an operator to provide communications services to its subscribers.17 It is measured in Hertz (“Hz”) which represents the number of wave cycles that will pass a point in one second.18 Because radio waves travel at the speed of light, the wave length is easily calculated from the wave’s frequency. When referring to radio spectrum used for mobile broadband services, frequency is typically measured in kilohertz (1000 Hz), megahertz (“MHz”) (1 million Hz), or gigahertz (“GHz”) (1 billion Hz). These units can refer to either the frequency of a radio wave, or the bandwidth between two frequencies. For example, there is five megahertz of bandwidth between the radio frequency 1930 MHz (1.93 GHz) and 1935 MHz (1.935 GHz). Spectrum used by mobile providers is typically licensed; however, technologies that use unlicensed spectrum, such as Wi-Fi, can also be used to relieve congestion on networks that also use licensed frequencies. 15. Spectral Efficiency refers to the amount of traffic a given amount of spectrum in a cell can support.19 Newer technologies often increase spectral efficiency compared to older technologies, for example, LTE has a greater spectral efficiency than UMTS and therefore provides more capacity per RF carrier of equal size. 16. Stata is a statistical analysis package created and maintained by StataCorp LP. Its capabilities include statistical analysis, plus data management, graphics, simulations, and programming.20 II. COVERAGE AND PERFORMANCE CLAIMS A. Background 17. The Applicants propose to deploy a FWLL network that would offer specific coverage and speeds in areas outside AT&T’s wireline footprint and areas within that footprint that currently do not receive the U-verse broadband and video bundle.21 18. The Applicants would deploy the FWLL on [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .22 19. The FWLL would employ LTE technology in [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] of dedicated spectrum in the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] currently licensed to AT&T.23 Each cell site would use “the same basic equipment, spectrum and technological configuration.”24 In particular, the deployment would require 17 See NEWTON’S TELECOM DICTIONARY at 770 (defining spectrum as “[a] continuous range of frequencies, usually wide in extent within which waves have some specific common characteristics.”) 18 See ESSENTIAL GUIDE TO RF AND WIRELESS at 8 (“The number of times a signal goes through a complete up and down cycle (from point A to point E) in one second is the signal’s frequency (measured in Hertz and abbreviated Hz).”) 19 Spectral efficiency is a measure of modulation efficiency and can be defined as the number of “bits per Hertz” or the number of bits that are transmitted in a given period of time, usually one second, over a radio channel with a defined bandwidth. See DIGITAL CELLULAR RADIO at 304-305, 394. 20 See StataCorp LP, Why Use Stata?, http://www.stata.com/why-use-stata/ (visited June 18, 2015). 21 Application at 5. 22 See AT&T Response to Sept. 9, 2014, Information Request at 203; see also infra Section XI.G.3. 23 See AT&T Response to Sept. 9, 2014, Information Request at 193-194, 200. [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] 24 AT&T Response to Sept. 9, 2014, Information Request at 194. Federal Communications Commission FCC 15-94 222 AT&T to “install additional antennas and other equipment at each cell site in areas it seeks to serve.”25 The Applicants do not specify what “other” equipment would be required, but it would be typical for a deployment like FWLL to include radio sets and potentially additional power and backhaul resources as well as professionally installed HGAs at the customer premise.26 20. The Applicants state that the HGAs would be [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .27 The Applicants also state that FWLL “will incorporate advanced technologies, including professionally installed customer premises equipment that will significantly enhance spectral efficiency and signal quality.”28 21. The Applicants expect that by using HGAs to compensate for differences in spectrum propagation characteristics, FWLL coverage at the deployment frequencies would be similar to their [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] coverage.29 AT&T notes in particular that the HGAs would have a much higher gain than the antennas of [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .30 22. Depending on the number of subscribers and the distance from the cell site, the FWLL service would have maximum download speeds of up to [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .31 The Applicants expect that [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .32 25 Id. at 235. 26 See id. at 207 (“Deploying fixed WLL requires incremental equipment at existing LTE cell sites, including new antennas, radios, and base band units.”); Implementation of Section 6002(B) of the Omnibus Budget Reconciliation Act of 1993, Annual Report and Analysis of Competitive Market Conditions With Respect to Mobile Wireless Including Commercial Mobile Services, WT Docket No. 13-135, Seventeenth Report, 29 FCC Rcd 15311, 15371, ¶ 116 (WTB 2014) (observing that a lack of fiber backhaul can delay a provider’s LTE rollout); Neal Gompa, T- Mobile Announces LTE, Prepares to take US Wireless Market by Storm, EXTREMETECH, Feb. 24, 2012, available at http://www.extremetech.com/electronics/119703-t-mobile-announces-lte-prepares-to-take-the-us-wireless-market- by-storm (discussing how LTE deployment would require new backhaul) (visited June 18, 2015). 27 See ATT-FCC-02122591, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See also ATT-FCC-02122591, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ; Application, Declaration of John T. Stankey, Group President and Chief Strategy Officer, AT&T, transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, ¶ 48 (filed June 11, 2014) (“Stankey Decl.”) (stating that FWLL “will incorporate advanced technologies, including professionally installed customer premises equipment, that significantly enhance spectral efficiency and signal quality.”). See also Joint Opposition of AT&T and DIRECTV to Petitions to Deny and Condition and Reply to Comments, Reply Declaration of Michael L. Katz, transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, ¶ 42 (filed Oct. 16, 2014) (“Katz Reply Decl.”) (stating that FWLL “will utilize an outdoor antenna which will be much more powerful than the indoor device that is currently offered” and will “achieve faster speeds than current wireless broadband products”). 28 Stankey Decl. ¶ 48. 29 [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] 30 See AT&T Response to Sept. 9, 2014, Information Request at 197, 202; AT&T Inc. Response to Sept. 9, 2014, and Dec. 15, 2014, Information and Discovery Requests, transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90, Exhibit 4.1 at 26 (Dec. 19, 2014) (“AT&T Response to Dec. 15, 2014, Information Request”). 31 AT&T Response to Sept. 9, 2014, Information Request at 200. 32 Id. at 201. Federal Communications Commission FCC 15-94 223 23. With these downlink data speeds, the Applicants believe the FWLL service would perform “as well as wireline broadband services advertised today at 15-20 Mbps,” which, the Applicants note, would be faster than typical indoor wireless home Internet products like AT&T’s Wireless Home Phone and Internet.33 The Applicants assert that “even customers at the cell edge will experience speeds greater than 10 Mbps more than 90 percent of the time.”34 In addition, the Applicants claim that the data speeds would be significantly better in off-peak periods, and that customers located closer to the cell tower would experience even better speeds.35 24. AT&T would limit the FWLL capacity to no more than “ [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .”36 However, AT&T expects generally to deploy [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .37 25. AT&T notes that “ [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .”38 26. The Applicants have chosen not to submit the results of a traditional network planning analysis to support their claims. Instead of providing detailed information including proposed site locations, comprehensive link budgets, and other facts about the proposed network, the Applicants rely on assumptions that deployment parameters and operating conditions applicable to a different technology in a different spectrum band would apply to FWLL. To support their assumptions, the Applicants have provided high-level information on slide decks, summaries and test results describing various aspects of the proposed FWLL deployment, including cell site locations, coverage simulation results, link budgets, network configuration plans, and summaries of field trials. Some of the Applicants’ network engineering materials did not support their network performance claims. Similarly, certain of the Applicants’ claims in their “business case” materials related to marketing FWLL services are inconsistent with certain of their engineering materials.39 B. Analysis 27. It is axiomatic that the performance of any wireless broadband network depends on a number of factors, such as technology and equipment, spectrum, the usage patterns of the subscribers, the traffic load, and the deployment environment, including possible sources of interference that could limit the coverage or data speeds of the network. Commission staff has reviewed the materials that the Applicants have submitted. This review included carefully analyzing the materials for accuracy, 33 Stankey Decl. ¶ 49; AT&T Response to Sept. 9, 2014, Information Request at 194. See generally ATT-FCC- 02122591, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . See also Katz Reply Decl. ¶ 42 (“[F]ixed WLL will be able to achieve faster speeds than current wireless broadband products.”). 34 Stankey Decl. ¶ 49; AT&T Response to Sept. 9, 2014, Information Request at 194. 35 Stankey Decl. ¶ 49; AT&T Response to Sept. 9, 2014, Information Request at 194. AT&T states that it “has not projected minimum download speeds for each geographic area.” AT&T Response to Sept. 9, 2014, Information Request at 194. 36 AT&T Response to Sept. 9, 2014, Information Request at 200. 37 Id. 38 Id. 39 The business plan assumes the offered cell-edge data speeds from [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] Mbps during the peak 15 minutes, depending on the MIMO configuration, but the [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] study suggests only [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] Mbps peak 15-minutes cell-edge data rates. Compare AT&T Response to Sept. 9, 2014, Information Request, Exhibit 58.g.1 at 2, 11 with AT&T Response to Dec. 15, 2014, Information Request, Exhibit 6.1 at 11. Federal Communications Commission FCC 15-94 224 consistency, and feasibility. The technical information provided on the proposed FWLL deployment is presented at a higher level than is customary for a typical network engineering design and analysis. The technical data initially provided by the Applicants included the results of a network performance simulation program that differ materially from the map showing predicted coverage that was also provided. However, as set forth in greater detail below, the Applicants ultimately submitted sufficient data to enable Commission staff to understand the discrepancy and conclude that the Applicants’ coverage and performance claims are feasible. 1. Coverage 28. Coverage characteristics can differ between technologies and spectrum. Therefore, a network deployment would normally be closely engineered to the propagation characteristics of the spectrum being used and related data would be available for analysis. The Applicants have provided a FWLL coverage map based on their [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .40 The Applicants, however, plan to deploy the FWLL network on [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . 29. The spectrum used in the Applicants’ coverage maps has coverage characteristics that are different, and typically better, than the spectrum where the FWLL would be deployed. The principal compensation for potential propagation differences the Applicants describe in the record is the use of the professionally installed outdoor HGAs. 30. These coverage maps were generated using the “best server” analysis data from Forsk’s Atoll, an RF planning tool AT&T normally uses in the ordinary course of business to predict signals for use in network planning.41 The Atoll tool uses a variety of site-specific parameters including actual cell site location, cell tower height, and local topography to predict signal availability in requested areas. The “best server” approach shows only the areas where [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .42 31. Separately, the Applicants submitted the results of a network performance simulation that purport to show the predicted network download data speeds at the cell edge. Unlike the Atoll tool, the performance simulation assumes the same parameters for each cell site which are not site-specific. The results were inconsistent with the site-specific Atoll coverage predictions. 32. The performance simulation data and assumptions that the Applicants initially submitted, in fact, suggest that the coverage area of the FWLL depicted on the maps generated with the Atoll “best server” data would not support the target cell-edge downlink data speeds. In particular, as depicted in Figure 1 below, Applicants’ performance-simulation data show that target data speeds were achieved where the average FWLL cell radius was assumed to be [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .43 However, as Figure 1 also shows, the radii of some of the cells predicted using the Atoll tool are much larger, sometimes [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .44 40 AT&T Response to Sept. 9, 2014, Information Request at 197. 41 Id. Forsk’s Atoll propagation tool is used by AT&T to create coverage maps. 42 Id. 43 See id. at Exhibit 58.i.1 at 5, Exhibit 58.g.2 at 3 (reporting an inter-site distance of [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] ). 44 See id. at Exhibit 58.b, Exhibit 81.j. Federal Communications Commission FCC 15-94 225 Figure 1: Houston, TX [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] 33. In response to Commission staff inquiries about this discrepancy, the Applicants submitted additional results from four geographically diverse FWLL field trials using a frequency band near the band where they plan to deploy FWLL. In addition to the field-test data, the Applicants responded to the Commission staff inquiries with new, separate calculations of cell coverage radii based on their FWLL link budgets and a simple slope-intercept propagation model, ranging from [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] (“supplemental cell radii calculations”).45 34. Although the FWLL field trials were conducted at [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] frequencies, the performance results show that with HGAs, the measured cell-edge downlink data speeds could be greater than 10 Mbps at similar distances compared to the same cell Atoll coverage predictions in three of the four geographic areas – [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . However, for the fourth geographic area, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] , the field trial measured average cell-edge radius seems somewhat less than the same cell Atoll coverage predictions.46 35. Applying the measurements obtained by the field trials, Commission staff was able to determine that the discrepancy between the coverage maps generated with the Atoll tool and the throughput simulation data likely results from the difference in the way the Atoll tool and the simulation tool predict coverage, as well as differences in the cell site configurations. The Atoll prediction of the FWLL cell coverage uses varying base station antenna heights and actual terrain data, in contrast to the lab simulation coverage model, which uses simplified propagation without terrain variability and uniform antenna heights that appear to be generally lower than the Atoll antenna heights. Similarly, Commission staff determined that, unlike the Atoll predictions, the set of supplemental cell radii calculations submitted alongside the field test data, as described in paragraph 34 above, relied on average estimates which also do not account for varying terrain and base station antenna heights. 36. Commission staff determined that if the original, inconsistent performance and coverage simulation information were interpreted taking into consideration the actual field performance measurements, it is reasonable to conclude that the Applicants’ claim that when compensated by other factors, such as higher HGAs and likely better propagations than predicted,47 the actual FWLL network 45 See AT&T Response to Dec. 15, 2014, Information Request, Exhibit 5.1 at [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] Tab. These are simple propagation model calculations without regard to the underlying physical terrain or clutter. 46 See id., Exhibit 4.2 at 4, 9; Exhibit 4.4 at 23-24, 47-48; and Supplemental Exhibit 58.b.1 FWLL coverage maps. See also AT&T Response to Sept. 9, 2014, Information Request at 197 (describing lab modeling); AT&T Response to Sept. 9, 2014, Information Request, Exhibit 58.g.2 at 3-4; AT&T Response to Dec. 15, 2014, Information Request at 5 (stating that AT&T has conducted “further lab modeling” since its first predictions). 47 See AT&T Response to Sept. 9, 2014, Information Request, Exhibit 58.i.1 at 4; AT&T Response to Dec. 15, 2014, Information Request at 6. The fixed outdoor antenna at customer premises equipment (“CPE”) has two main advantages compared to antennas [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] : (1) roof- (continued….) Federal Communications Commission FCC 15-94 226 coverage at the target data speed would be comparable to the Atoll coverage predictions for the field trial locations. 2. Performance 37. The Applicants have proposed a FWLL network design based on several assumptions that may not be realized in the actual deployed network. In particular, the Applicants have assumed that: (1) subscriber usage patterns would be consistent with a lower speed tier than the Applicants propose to market for the FWLL deployment; (2) six-sector sites would provide [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .48 38. More specifically, the Applicants assume a conservative traffic speed tier of [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] in the lab-based simulations they have submitted to support their FWLL performance and capacity claims.49 By contrast, the Applicants elsewhere claim they would offer consumers much faster data speeds (15-20 Mbps) and a higher data cap (up to [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] per month).50 39. The record shows that when subscribers are offered higher speeds, their usage increases.51 This, in turn, would affect the network’s capacity and performance. Traffic typically associated with the higher speed and data tiers associated with the target market segment (i.e., 15-20 Mbps and up to [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] per month) places greater demands on a network than the lower speed tiers that the Applicants relied upon for their FWLL input traffic model [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .52 If improperly engineered, greater traffic loads could have an adverse effect on either maintaining the FWLL’s target data speeds or the target number of subscribers that could use the network per sector, or both. Finally, Applicants use simulations of [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .53 However, the Applicants’ own analysis indicates that [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .54 Furthermore, the simulated sector capacity may not be realized for all sites, because [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .55 40. The Applicants have provided no evidence to explain how they would address the practical issues affecting FWLL performance that could arise if any of these assumptions is not accurate. Poor network performance could, in turn, affect long-term competitiveness of FWLL, particularly in areas where there is an existing terrestrial broadband competitor. (Continued from previous page) top heights and (2) antenna gain of [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . There are also differences in foliage or clutter signal losses at higher frequencies, but those differences are counterbalanced to some degree by the higher FWLL antenna gains and heights. Foliage signal reduction is generally higher at higher frequencies. 48 See AT&T Response to Sept. 9, 2014, Information Request at 200, Exhibit 58.g.2 at 8. 49 See id. at Exhibit 58.i.1 at 3; AT&T Response to Dec. 15, 2014, Information Request, Exhibit 6.1 at 4. 50 See AT&T Response to Sept. 9, 2014, Information Request at 199. 51 See id. at Exhibit 58.f.6 at 4. 52 See supra ¶ 38. 53 See AT&T Response to Sept. 9, 2014, Information Request at 200, Exhibit 58.g.2 at 8. 54 See id. at Exhibit 58.i.1 at 13. 55 See ATT-FCC-02208834, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . Federal Communications Commission FCC 15-94 227 41. To be sure, network performance challenges could be mitigated by adding spectrum or cell sites to maintain the target quality of service. Any required expansion, either of cell count, or use of additional spectrum, would increase both capital and operating expenses.56 However, there is no evidence in the record that explains how AT&T might expand the FWLL network capacity or whether its business model would support the required additional cost. In the absence of quantifiable financial data that support a different conclusion, we believe these expenses could have a negative impact on the financial viability of the FWLL business model.57 III. AT&T’S CLAIM THAT FWLL WOULD BENEFIT 13 MILLION RURAL CUSTOMERS A. Background 42. The Applicants claim that post-transaction, FWLL would reach 13 million mostly rural households.58 Further, AT&T maintains that almost 20 percent of the 13 million households (i.e., 2.6 million households) have no access to terrestrial broadband, and 27 percent of the 13 million households (i.e., 3.5 million households) have only one terrestrial option.59 Conversely, in certain documents, AT&T claims that its FWLL network would provide broadband to [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] FWLL-eligible rural households that do not have terrestrial broadband service available.60 The record does not reconcile the claims. In any case, Applicants claim that in most households with only one terrestrial option, that option is either DSL or a relatively slow cable modem service.61 43. According to the Applicants, FWLL coverage – and, therefore, the number of people it would serve in rural areas – is determined by a complex four-part formula. First, the sites must be in a “rural” geographic area. AT&T defines geographic areas “by superimposing a grid of one mile squares on a map of the continental United States.”62 AT&T “considers a cell site ‘rural’ if it is located in a square mile with a population of less than 250 persons.”63 However, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .64 In addition, the cell site must be one where [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . Finally, the cell site must be one where [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .65 56 There is no evidence in the record that explains how AT&T might expand the FWLL network capacity. See generally AT&T Response to Sept. 9, 2014, Information Request, Exhibit 59.a.1 (showing net present values for various numerical ranges of subscribers) and Exhibit 59.l.1 (showing LTV calculations for FWLL). 57 See supra ¶ 26 & n.39. 58 See Stankey Decl. ¶ 36. See also supra n.1 (discussion and definition of “household”). 59 Application at 44. AT&T describes its methodology as follows: “ [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] .” AT&T Response to Sept. 9, 2014, Information Request at 205. 60 See ATT-FCC-02210352, [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] . Specifically, AT&T estimates that there are [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] households in no-broadband or “IP red” territories, of which [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] are rural and [BEGIN HIGHLY CONF. INFO.] [END HIGHLY CONF. INFO.] are rural and included in its FWLL deployment plans. Id. 61 Application at 44; Stankey Decl. ¶ 55. 62 AT&T Response to Sept. 9, 2014, Information Request at 195. 63 Id. 64 See id. at 203. 65 See id. Federal Communications Commission FCC 15-94 228 B. Analysis 44. AT&T has submitted a map that depicts its projected FWLL deployment area. There is insufficient information in the record to replicate this coverage map. Therefore, we assume this map was derived using the four criteria discussed above, and use this map to represent the projected FWLL deployment area in our analysis. We evaluate the Applicants’ claims regarding population and existing broadband coverage within the projected FWLL coverage area using two separate methods. 45. The first method uses GeoLytics census data and State Broadband Initiative (“SBI”) data (“Census Block Methodology”). SBI data contain a comprehensive list of terrestrial providers at the census-block level, including provider name, maximum advertised speed, and technology, among other variables.66 GeoLytics data contain 2013 projections of 2010 census data at the census-block level, including variables such as population, housing units, and block area. Using ArcGIS, Commission staff selected every census block for which the centroid of the block overlapped the projected FWLL deployment area. The resulting list of census blocks was combined with the SBI and GeoLytics data.67 The resulting dataset was used to create census-block-level estimates of broadband availability and population density.68 46. In addition, we also evaluated the Applicants’ claims using only the data that they submitted in the record (“Grid Methodology”). This second method enables an examination of population and population density; however there was not sufficient Applicant-submitted data to evaluate existing broadband coverage in the projected FWLL deployment area. Using ArcGIS, Commission staff overlaid the Applicant-submitted one mile grid with the map of the projected FWLL deployment area. The grid squares that overlapped the projected deployment area were selected, and the selected 66 In 2009, in the Recovery Act, Congress directed NTIA through the SBI program to collect more robust data about broadband deployment and to create a National Broadband Map, “a comprehensive, interactive, and searchable nationwide inventory map of existing broadband service capability and availability.” Inquiry Concerning the Deployment of Advanced Telecommunications Capability to All Americans in a Reasonable and Timely Fashion, GN Docket No. 10-159, Seventh Broadband Progress Report and Order on Reconsideration, 26 FCC Rcd 8008, 8079, Appendix F, ¶ 4 (2011) (quoting Department of Commerce, NTIA, State Broadband Data and Development Grant Program, Docket No. 0660-ZA29, Notice of Funds Availability, 74 Fed. Reg. 32545, 32546 (July 8, 2009), available at http://www.ntia.doc.gov/frnotices/2009/FR_BroadbandMappingNOFA_090708.pdf); see also Inquiry Concerning the Deployment of Advanced Telecommunications Capability to All Americans in a Reasonable and Timely Fashion, GN Docket No. 11-121, Eighth Broadband Progress Report, 27 FCC Rcd 10342, 10344, ¶ 1 (2012). These data, while useful for measuring developments in terrestrial broadband deployment, have certain limitations that likely overstate the extent of terrestrial broadband deployment. 67 Our analysis does not include broadband connections in which the customer is classified as “Business” or “Government,” nor those in which the provider is classified as a “Reseller.” Every observation in the SBI dataset provides information on an individual provider in an individual census block. Therefore, there are no SBI data for any census block without existing terrestrial broadband providers. However, census blocks that are within the projected FWLL deployment area, but do not have existing terrestrial broadband providers are included in our analysis in order to reflect the number of households with no broadband providers. 68 The SBI data is divided into two categories – large census blocks and small census blocks. Large blocks are defined as any census block with an area greater than two square miles. Small blocks are defined as any census block with an area up to two square miles. The total number of broadband providers in a census block represents deployment, or homes passed, which does not necessarily reflect the number of choices available to a particular individual or household. If there are no households in a census block, it is assumed that there is no terrestrial broadband coverage in that block. Because our analysis reflects the overall number of households with broadband coverage, this does not affect our result. Federal Communications Commission FCC 15-94 229 observations were exported and combined with demographic data to calculate population and population density for these one square mile areas.69 47. Since its 2004 Report and Order concerning deployment of wireless services in rural areas, the Commission has used a guideline definition of “rural” as an area with a population density of 100 persons or fewer per square mile.70 In our analysis, the relevant area is either a census block or a mile grid square, depending on the methodology used. We see no reason to revisit this guideline here. The Commission’s definition of a rural area as one with 100 persons or fewer per square mile conflicts with the Applicant’s definition of a rural area as one with less than 250 persons per square mile. 48. Using the Commission’s definition of “rural” at the census block level, and the Census Block Methodology described above, we estimate that approximately 4.5 million rural households would be within the projected FWLL deployment area. Using the same methodology, and the Applicants’ definition of rural, we estimate that approximately 6.9 million rural households would be within the projected FWLL deployment area. As a control, we replicated this calculation using the Applicants’ data using the Grid Methodology and the Commission’s definition of “rural.” 71 Using this method, we find there are 6.0 million FWLL eligible rural households. These estimates are substantially fewer than half of AT&T’s claim of 13 million “largely rural” households. 49. We also note that, based on the Census Block Methodology described above, there are many fewer rural households with zero or one terrestrial broadband provider than AT&T claims. Table 1 compares the Applicants’ customer-coverage claims to Commission staff estimates. 69 AT&T submitted three separate population numbers for each grid square: daytime population, nighttime population, and maximum population, which is the larger of daytime and nighttime population. See AT&T Response to Sept. 9, 2014, Information Request at 195, 198 (describing AT&T’s process for identifying “rural” areas in a grid of square miles) and Exhibit 58.c.2 (setting forth daytime, nighttime, and maximum population values via “National Morphology Grids”). The analysis in the body of this Appendix uses the maximum population. This is the value that we assume AT&T used in their analysis. 70 Facilitating the Provision of Spectrum-Based Services to Rural Areas and Promoting Opportunities for Rural Telephone Companies To Provide Spectrum-Based Services, WT Docket No. 02-381, Report and Order, 19 FCC Rcd 19078, 19087-88, ¶ 12 (2004) (“We recognize, however, that the application of a single, comprehensive definition for ‘rural area’ may not be appropriate for all purposes …. Rather than establish the 100 persons per square mile or less designation as a uniform definition to be applied in all cases, we instead believe that it is more appropriate to treat this definition as a presumption that will apply for current or future Commission wireless radio service rules, policies and analyses for which the term ‘rural area’ has not been expressly defined. By doing so, we maintain continuity with respect to existing definitions of ‘rural’ that have been tailored to apply to specific policies, while also providing a practical guideline.”). 71 The Applicant data included the population per grid square; however it did not include the number of households per grid square. We converted the resulting population numbers into households by dividing the population numbers by the census definition of average household size of 2.54 persons per household. Federal Communications Commission FCC 15-94 230 Table 1. Applicants’ Claim (number of households) Commission Staff Estimate (number of rural and non-rural housing units)72 Commission Staff Estimate (number of rural housing units with less than 250 pops per square mile) Commission Staff Estimate (number of rural housing units with less than 100 pops per square mile) Customers located in projected FWLL deployment area 13 million “largely rural” households73 15.2 million 6.9 million 4.5 million Customers located in projected FWLL deployment areas with no existing terrestrial broadband provider 2.6 million74 1.5 million 1.2 million 1.0 million Customers located in projected FWLL deployment areas with one existing terrestrial broadband provider 3.5 million75 4.5 million 2.8 million 2.0 million 50. Table 1 shows that, according to AT&T, 2.6 million households within the projected FWLL deployment have no existing terrestrial broadband service. In contrast, Commission staff estimates that within the entire projected FWLL deployment area (i.e., rural and non-rural areas), approximately 1.5 million households would have no terrestrial broadband option. In addition, based on the Applicant’s definition of “rural,” Commission staff estimates that approximately 1.2 million rural households would have no terrestrial broadband option. Further, based on the Commission’s guideline definition of “rural,” Commission staff estimates that approximately 1.0 million rural households would have no terrestrial broadband option. With regard to AT&T’s claim that 3.5 million households have only one terrestrial broadband provider,76 we observe that, even assuming Applicants’ definition of “rural,” approximately 2.8 million rural households would have one terrestrial broadband option. And based on the Commission’s guideline definition of “rural,” Commission staff estimates that approximately 2.0 million rural households would have only one terrestrial broadband option. Thus, Commission staff’s estimates are significantly lower than AT&T’s estimates, even when using AT&T’s definition of “rural,” but even more so when using the Commission’s established guideline for determining which communities are “rural.” 72 Commission staff estimates were derived using AT&T Response to Dec. 15, 2014, Information Request, Exhibit 58.b.1; SBI data; and GeoLytics census data. For more details regarding data and calculation techniques, see supra ¶¶ 44-49. 73 See Stankey Decl. ¶ 36. 74 AT&T maintains that almost 20 percent of the 13 million households (i.e., 2.6 million households) have no access to terrestrial broadband. See supra ¶ 42. 75 See supra ¶ 42. 76 AT&T claimed that 27 percent of 13 million households have only one terrestrial broadband provider, which equates to 3.5 million households. See supra ¶ 42. Federal Communications Commission FCC 15-94 231 STATEMENT OF CHAIRMAN TOM WHEELER Re: Applications of AT&T Inc. and DIRECTV for Consent to Assign or Transfer Control of Licenses and Authorizations, MB Docket No. 14-90. Today I vote in favor of an order with conditions that approves the acquisition of DIRECTV by AT&T. This transaction offers the opportunity for more competition, directly benefitting consumers, thus advancing the public interest. My vote comes after the Commission staff engaged in a state-of-the-art economic analysis and a rigorous examination of the facts. Their work supports my conclusion that this merger is pro-competitive, which is consistent with the views publicly expressed by the Department of Justice. Until now, neither company has been fully equipped to compete with cable’s bundled services, including both high speed broadband and pay TV. AT&T has had a disproportionately slow broadband network and higher programming costs. DIRECTV has no means of supplying competitive broadband access. There also is too little competition today in the availability of fixed broadband, especially at higher speeds. As I noted last fall, roughly 74% of American homes have a choice of zero or only one fixed broadband providers at 25/3 mbps. A combined AT&T-DIRECTV offers consumers a viable competitor to cable by providing a bundle of broadband and video. That new competition will challenge some of the nation’s biggest cable companies, including Comcast, Time Warner Cable, Charter, and Cox, as well as Google Fiber. Competition means lower prices for both customers of the new company and customers of cable. At the same time, the transaction carries with it two notable competitive risks. First, roughly one quarter of American households will lose a pay TV competitor – a traditional horizontal harm. Second, after the merger, AT&T, with the ability to provide Pay TV programming nationally, will have a greater incentive to harm emerging services that use broadband connections to offer consumers new choices in the selection of Pay TV. It is to directly address those risks that the Order includes targeted and merger- specific conditions. The conditions will lock in the benefits of the acquisition and address the risks of competitive harm. This is our statutory obligation. The detail set forth in the Order need not be repeated here, but the core purposes of the conditions should be emphasized. As the Order explains, with DIRECTV in hand, AT&T will now have a lesser incentive to deploy fiber to consumers. Given the lack of competition that already exists, that disincentive would be a bad outcome, but it is remedied by the requirement that AT&T make good on its promises and bring competitive high-speed broadband to 12.5 million customer locations. This additional build-out is about ten times the size of AT&T’s current fiber-to-the-premise (FTTP) deployment, increases the entire nation’s residential fiber build by more than 40 percent, and more than triples the number of metropolitan areas AT&T had previously announced plans to serve. A natural consequence of the FTTP deployment is expansion of the AT&T network to institutions and enterprises—and that offers an additional opportunity for competition which is why the Commission is requiring AT&T-DIRECTV to offer gigabit service to any E-rate eligible school or library within the areas where AT&T-DIRECTV deploys FTTP service. Most importantly, this requirement will result in the permanent expansion of fiber—a legacy that will endure long after the conditions of the decision have expired. AT&T has four years to complete the required FTTP build-out, but if it is not completed in that time period, all of the conditions we impose today will remain in effect. At various times, AT&T has announced an expansion of their fiber footprint. Then, after the Open Internet decision, AT&T announced they would not be undertaking the expansion and then quickly told the merger review team that it would, in fact, complete its prior expansion plans. Federal Communications Commission FCC 15-94 232 What is plain is this: investment in new high speed broadband is profitable now because it offers companies the opportunity to reach new customers with new products, and nothing in the implementation of the Open Internet Order has changed that. In addition, while our analysis confirmed that a potential benefit of the merger is the availability of better and lower priced bundles of video and broadband service, the Commission also concludes that the public interest requires us to ensure that a bundle of video and broadband services is not the only competitive choice for low-income subscribers who may not be able to afford bundled services. The Commission accordingly requires as a condition of the merger that AT&T-DIRECTV make available an affordable, low-price standalone broadband service to low-income consumers in its broadband service area. We also impose conditions that build on the Open Internet Order already in effect, addressing two merger-specific issues. First, in order to prevent discrimination against online video competition, AT&T will not be permitted to exclude affiliated video programming services and content from data caps or impose other discriminatory retail terms and conditions on its fixed broadband connections. Second, in order to bring greater transparency to interconnection practices, the company will be required to submit all completed interconnection agreements to the Commission, along with regular reports on network performance. These commitments are important to confirm the benefits of the transaction, and there has been concern expressed as to whether previous merger terms have actually been fulfilled. We have heard those concerns and require the company to retain an outside independent compliance officer who will monitor compliance and report to the Commission throughout the life of the conditions. In sum, this Order will serve the public interest. Transaction conditions should directly address the threat of competitive harm, and these do. Binding legal requirements must be enforceable, which is why the creation of an independent compliance officer is very important. There is too little competition in high speed broadband and in the market for bundled services and this transaction, with its targeted conditions, is a big step in the right direction. Broadband deployment should be inclusive, and this transaction will serve both low-income Americans and schools and libraries, addressing the Digital Divide. I am proud of our work here and so should be the American people. Federal Communications Commission FCC 15-94 233 STATEMENT OF COMMISSIONER MIGNON L. CLYBURN Re: Applications of AT&T Inc. and DIRECTV for Consent to Assign or Transfer Control of Licenses and Authorizations, MB Docket No. 14-90. I firmly believe that broadband is the greatest equalizer of our time. The ability to participate in e-commerce, seek employment online, benefit from telemedicine or become an entrepreneur is key to addressing many of the challenges in our Nation’s persistent poverty areas. But for our goal of universal broadband access to be realized, we need both deployment of networks and access to affordable services. This merger makes strides in achieving these two goals. First, approval of this transaction is conditioned on the deployment of fiber-to-the-home to twelve million additional households. This is a laudable result and I look forward to its implementation. Second, as the Order notes, cost remains a barrier for too many low-income consumers who wish to adopt broadband at home. I am particularly pleased that the commitments in this merger provide an affordable standalone broadband offering based on Supplemental Nutrition Assistance Program (SNAP) eligibility, which means that parents with school-aged children as well as adults, veterans, persons with disabilities and seniors without children, could benefit. This broadband option has the potential to empower consumers by making the Internet available at home for $10 per month for 10 Mbps, without any hidden or additional costs, fees for installation or equipment, or needless limits on consumers who have been challenged in obtaining or maintaining services before applying. Consumers will have the ability to participate for four years and AT&T will provide the Commission with reports on the success of their efforts every six months. While I believe that public interest benefits are significant, I do have concerns with the potential impact on smaller MVPDs and independent programming. This merged entity will have over 25 million video subscribers. A larger subscriber base and corresponding ability to leverage efficiencies in order to save programming costs are cited as benefits of this transaction, but a number of parties have expressed concerns regarding potential harms to smaller cable operators (or MVPDs) in accessing certain programming, or gaining access to content at affordable costs. On the other side, we have heard from a number of parties about the impact on independent programmers, particularly those that supply programming for diverse and niche audiences via MVPD or online. While the analysis underlying this transaction does not find that these challenges are made demonstrably worse by this merger, I believe it is now time to reevaluate our program access rules, and to examine barriers to program carriage and distribution, in order to address significant concerns about the impact on small businesses and diverse or independent programmers. Indeed, the transaction itself highlights the need for a reexamination of our rules when a provider as large as AT&T merges with a DIRECTV in part to reduce programming costs. I therefore call on the Chairman to initiate a proceeding, such as a Notice of Inquiry, in order to take a fresh look at our carriage rules and examine the challenges and barriers to independent and diverse programming. This could include the ability of small and independent programmers to achieve program carriage, as well as the ability of distributors to offer niche content in innovative ways. I also call on the Chairman to initiate (or complete, to the extent necessary) a proceeding regarding our program access rules, to evaluate their effectiveness and identify whether there are ways to reform our rules to provide a level playing field for smaller operators to remain competitive. In addition, the other conditions included in this Order to address identified harms, including non- discrimination in the application of data caps, the appointment of a Compliance Officer to track and report AT&T’s progress in carrying out the commitments enumerated in this Order, and enabling the Federal Communications Commission FCC 15-94 234 Commission to evaluate interconnection practices and performance are necessary to ensure that the transaction is in public interest. I therefore vote to approve this item, and look forward to working with the Chairman to take a hard look at program carriage, program distribution and the program access rules to ensure we address any industry-wide concerns that our rules may not be working as intended. Federal Communications Commission FCC 15-94 235 STATEMENT OF COMMISSIONER JESSICA ROSENWORCEL Re: Applications of AT&T Inc. and DIRECTV for Consent to Assign or Transfer Control of Licenses and Authorizations, MB Docket No. 14-90. In this Order we approve the merger of AT&T and DIRECTV. This decision is based on extended review and careful consideration of the consequences of combining a mix of interests in broadband, wireless, and video services. On balance, we find that this transaction, as conditioned, serves the public interest, as required under law. Communications markets are changing fast—and the future of watching is bound to look different from the past. Channel packages with limited choices are giving way to a new world where consumers watch what they want, when they want it, on any screen handy. As a result, broadband is becoming an essential tool for the distribution of video content. Consequently, this transaction is conditioned on the deployment of fiber to the premises to 12.5 million new customer locations. This deployment will result in more modern infrastructure that supports a range of activities—including online video. This development is not just good for consumers, it also addresses disincentives for deployment that could arise from this transaction. Even better, these facilities will be made available to schools and libraries in the deployment footprint. In addition, this transaction is conditioned on non-discriminatory usage-based practices that will help ensure that the online marketplace for video has fair opportunity to grow. Finally, though not the subject of any conditions in this transaction, the issue of independent programming and the complexities of securing access on traditional video distribution platforms has come up repeatedly in the record in this proceeding. I think this issue is ripe for examination, and hope that the Commission can find another forum for discussion of this important topic. Federal Communications Commission FCC 15-94 236 STATEMENT OF COMMISSIONER AJIT PAI APPROVING IN PART AND DISSENTING IN PART Re: Applications of AT&T Inc. and DIRECTV for Consent to Assign or Transfer Control of Licenses and Authorizations, MB Docket No. 14-90. The Commission’s comprehensive record conclusively demonstrates that AT&T’s acquisition of DirecTV is in the public interest. The combined entity will compete more effectively in our nation’s video and broadband markets. In particular, the transaction will allow the combined entity to offer integrated bundles of video and broadband to far more Americans than AT&T could alone. The end result will be lower-priced bundles for consumers and greater high-speed broadband deployment. I therefore support the Commission’s decision to approve this transaction. However, I cannot support the Commission’s decision to place 17 pages of conditions on that approval.1 The transaction’s benefits clearly outweigh any harms. As a result, there is no need to impose conditions upon it. Indeed, earlier this week, the U.S. Department of Justice “concluded that the combination of AT&T’s land-based internet and video business with DirecTV’s satellite-based video business does not pose a significant risk to competition.”2 It therefore concluded its “extensive investigation” into the transaction without asking for any conditions.3 The Commission should have done the same. But the FCC goes much further, demanding that AT&T satisfy a regulatory wish-list that has nothing to do with the transaction at hand. These conditions are the forced tribute that the company must offer to mollify the Capitol.4 In this regard, I dissent. Some conditions are nothing more than policymaking through the merger review process. Consider, for example, the decision (however nobly intended) to require AT&T to offer discounted broadband service. The Commission concedes that its economic model “predicts very little change in the price of AT&T’s standalone broadband post-transaction.”5 Depending upon the assumptions used, the price adjustment will range from a 0.73% decline to a 0.46% increase. Moreover, the Commission concedes that the transaction will lead to “little change (positive or negative)” in the stand-alone broadband prices offered by AT&T’s competitors.6 Notwithstanding these findings, the Commission effectively decides to get into the discount broadband business, using AT&T as its agent. For example, where technically available, the company is 1 Order at Appendix B. 2 Justice Department Will Not Challenge AT&T’s Acquisition of DirecTV, Press Release (July 21, 2015), http://go.usa.gov/37rkh. 3 See id. 4 Cf. President Snow, The Hunger Games (Lionsgate 2012) (“And so it was decreed that, each year, the various districts of Panem would offer up, in tribute, one young man and woman to fight to the death in a pageant of honor, courage and sacrifice.”). 5 Order at para. 142. 6 Order at para. 143. Federal Communications Commission FCC 15-94 237 required to offer “qualifying households”7 wireline broadband service with download speeds of at least 10 Mbps (which, according to the majority on a different day, isn’t broadband8) for no more than $10 a month.9 When the Commission instructs a regulated entity that it must offer a particular service for no more than a particular price, there is a name for that. It is called rate regulation. So notwithstanding the repeated claims by some over the past few months that the FCC has no interest in regulating retail broadband rates, the reality is far different. When given the opportunity, the Commission did not hesitate to impose rate regulation upon a broadband provider. This is merely a preview of coming attractions. Moreover, the rate regulation imposed by the Commission is not even designed to prevent price increases from occurring in the wake of the transaction. It is merely intended to cut pre-transaction prices. To give just one example, the price of AT&T’s stand-alone 6 Mbps broadband service in Austin, Texas is currently $34.95 a month.10 And the 6 Mbps broadband service portion of an Internet/phone bundle is $14.95 a month.11 Given these figures, how could requiring AT&T to offer a 10 Mbps stand- alone broadband service for no more than $10 a month possibly be necessary to remedy a harm caused by the transaction? The Commission doesn’t even make a cursory attempt to explain how it arrived at this $10 price point. Shifting gears, the Commission imposes conditions designed to remedy what it contends would be the combined entity’s increased incentive to discriminate against unaffiliated over-the-top video providers. Specifically, the Commission claims that, after the merger, AT&T would seek to hinder those over-the-top providers in order to protect DirecTV’s video product or AT&T’s own online video products. But the only detailed economic analysis and econometric modeling in the record (not to mention the Justice Department’s approval, which implies rejection of the FCC’s argument) point to the opposite conclusion. They demonstrate that any attempt by AT&T to hamper its broadband customers’ access to over-the-top video providers would only end up hurting the company.12 It would encourage customers to switch broadband providers, and many customers purchasing bundles from AT&T would likely end up taking their video business elsewhere as well. Indeed, AT&T benefits when its customers use over-the- top video providers, such as Netflix and Amazon, since “[o]nline video is a major driver of broadband demand, and desire to consumer online video leads consumers to purchase more broadband service, 7 Qualifying households are those where at least one individual participates in the Supplemental Nutrition Assistance Program and that do not have certain outstanding debts to AT&T. See Order at Appendix B § VI(2)(c). 8 See Inquiry Concerning the Deployment of Advanced Telecommunications Capability to All Americans in a Reasonable and Timely Fashion, and Possible Steps to Accelerate Such Deployment Pursuant to Section 706 of the Telecommunications Act of 1996, as Amended by the Broadband Data Improvement Act, GN Docket No. 14-126, 2015 Broadband Progress Report and Notice of Inquiry on Immediate Action to Accelerate Deployment, 30 FCC Rcd 1375, 1485 (Dissenting Statement of Commissioner Ajit Pai) (2015), http://go.usa.gov/372yw. 9 Order at Appendix B § VI(2)(a). 10 See AT&T, Check for U-Verse Availability, https://www.att.com/shop/u-verse/offers.html (entering address for Austin, Texas and selecting Internet-phone bundle). 11 See id. (entering Austin, Texas address and selecting stand-alone broadband service). 12 Joint Opposition, “An Economic Assessment of AT&T’s Proposed Acquisition of DIRECTV,” Reply Declaration of Michael L. Katz, transmitted by letter from Maureen R. Jeffreys, Counsel for AT&T, to Marlene H. Dortch, Secretary, FCC, MB Docket No. 14-90 (Oct. 16, 2014). Federal Communications Commission FCC 15-94 238 including higher speed tiers.”13 Thus, it should come as no surprise that AT&T offers a promotional package in conjunction with Amazon Prime in four cities. The Commission’s theory is also internally contradictory. For example, the Commission contends on one hand that AT&T would seek to harm unaffiliated over-the-top video providers in order to encourage customers to subscribe to DirecTV’s video package. Yet it concludes in another section that over-the-top video providers are in a different product market than multichannel video programming distributors such as DIRECTV because “for most consumers today, OVD [online video distribution] services are not substitutes for MVPD services.”14 Well, which is it? Do over-the-top video providers compete with DirecTV or not? The FCC’s creative heads-we-win-tails-AT&T-loses view of the video marketplace embraces Wilde’s dictum that “consistency is the hallmark of the unimaginative.”15 On top of this, the conditions pertaining to over-the-top video providers are highly intrusive. For example, AT&T must fork over to the FCC each and every interconnection agreement for the next four years. It also must retain an Independent Measurement Expert, approved by the FCC’s Office of General Counsel, to develop a methodology for measuring specified performance metrics for traffic exchanged at its interconnection points. And it must report regularly to the Commission on those metrics. This government-mandated surveillance is entirely unnecessary and is just another step towards putting the FCC at the core of the Internet. It has been said that the Commission’s Title II order makes the FCC the referee on the field, ready to throw the flag whenever a broadband service provider does something that it doesn’t like. This condition goes beyond that and injects the FCC into the huddle, monitoring a team’s play calling. Speaking of surveillance, the conditions imposed by the Commission in this item also place a FCC-designated monitor inside of AT&T. Specifically, the company and the Commission’s Office of General Counsel must appoint an Independent Compliance Officer to monitor AT&T’s compliance with the Commission’s demands and provide regular reports to the Commission. And if the company and Commission’s Office of General Counsel are unable to agree on an Independent Compliance Officer, the Office of General Counsel will pick one.16 This Independent Compliance Officer will have wide-ranging powers. Among other things, he or she will have the authority to interview any company personnel, to inspect and copy any document, email, or contract, and to require the company to provide any data or submit any reports for any purpose that he or she believes to be reasonably related to his or her duties.17 And he or she may hire a staff to help do all of these things.18 There is no justification for the Commission to adopt this extraordinary condition. The Commission does not point to any credible evidence that the company has failed to comply with the 13 Id. 14 Order at para. 68. 15 Moreover, to the extent that the Commission believes that the company will attempt to hurt unaffiliated, over-the- top video providers in order to assist AT&T’s over-the-top video products—a theory that is not supported by the record—it is difficult to see how such a harm would be specific to this transaction because such an incentive would already exist. 16 Order at Appendix B § VII(3)(a). 17 Order at Appendix B § VII(3)(i). 18 Order at Appendix B § VII(3)(k). Federal Communications Commission FCC 15-94 239 conditions imposed upon it by the FCC in prior transactions. And there is no reason to presume that AT&T will fail to abide by the conditions contained here. This also establishes a dangerous precedent. I have little doubt that when we consider future transactions, there will be calls for future applicants to accept Independent Compliance Officers as a condition of approval. Virtually any transaction involving companies we regulate could result in the injection of a Commission-selected solon with vast powers. Government-approved monitors placed throughout the communications industry would represent a pernicious intrusion into the affairs of private businesses and a dramatic expansion of the Commission’s authority. * * * Substance aside, I also have concerns about process. The Commission has established a 180-day shot-clock for transactional review. But it has been 408 days since AT&T and DirecTV filed their application with the Commission! In that amount of time, Jules Verne’s Phileas Fogg from Around the World in 80 Days could have circumnavigated the globe five times in the 1870s. To state the obvious, this matter has taken far too long to resolve and has made a joke of the Commission’s 180-day shot-clock. What took so long? Part of the time was wasted, over my objection, on a pointless and misguided quest to permit third-parties to review programming contracts. This dispute ended with the Commission suffering a resounding defeat in the U.S. Court of Appeals for the D.C. Circuit, which held unanimously that the Commission “offer[ed] an exceedingly thin rationale” for a “substantive and important departure from prior Commission policy.”19 And even though the D.C. Circuit gave the FCC the option on remand of trying again to give third-parties access to programming contracts, the agency approves this transaction without taking that step, belying any assertion that third-party inspection of these materials was necessary to the Commission’s consideration of this transaction.20 But this litigation was not responsible for most of the delay here. Instead, it appears that the Commission adopted a “four corners” strategy for handling this transaction that would have made the late, great University of North Carolina basketball coach Dean Smith proud. This is unacceptable. I understand that some savor the leverage gained by keeping a regulated entity under the Commission’s thumb for as long as possible. But the speed of today’s digital economy makes it critical for the FCC to move quickly. When companies remain stuck in purgatory for over a year waiting for an answer from the Commission, their business plans are placed on hold while their rivals move full speed ahead. That isn’t good for competition, it isn’t good for consumers, and it isn’t good government. For these reasons, I hope that in the future, the Commission abides by its self-imposed deadline and completes its consideration of proposed transactions much more quickly than it did here. 19 CBS Corp. v. FCC, 785 F.3d 699, 708–09 (D.C. Cir. 2015). 20 Cf. id. at 707 (“Nowhere does either the [Media] Bureau or the Commission make the jump from useful or relevant or central to necessary.”). Federal Communications Commission FCC 15-94 240 STATEMENT OF COMMISSIONER MICHAEL O’RIELLY APPROVING IN PART AND CONCURRING IN PART Re: Applications of AT&T Inc. and DIRECTV for Consent to Assign or Transfer Control of Licenses and Authorizations, MB Docket No. 14-90. More than a year after the applicants filed the transfer applications, with our 180-day “shot clock” left in pieces on the floor like a particularly raucous alarm clock on a weekend morning, the Commission now concludes its review. Although I do not subscribe to a number of premises presented in the item (is it really the Commission’s job to make a “bet” on competition as part of its overall public interest analysis? Are bundles really the future of communications offerings in the age of the cord cutter?), on balance the combination of the two companies will produce certain benefits, and the case for a lack of competitive harm is sufficiently made. At the same time, I find the conditions being imposed, albeit less onerous than some of those extracted in past merger approvals, are unrelated to the transaction at hand, outside the scope of our proper role, and harmful to consumers. Therefore I approve in part and concur in part. Some of the conditions imposed will sound strangely familiar to those who have followed the Commission closely over the last many years. Our duty to determine whether a proposed transaction will serve the public interest, however, should not be read as an opportunity to divert private resources toward favorite causes and theories. The economic analysis finds that the net effect of the transaction is beneficial to consumers,1 that the transaction has a positive effect on the price of bundled products,2 and that the loss of a video provider within AT&T’s current video footprint creates such a limited potential for competitive harm that, when balanced against the benefits of the transaction, no conditions are required.3 But the Commission just can’t pass up an opportunity to push its own objectives, even if it is unrelated to the matter at hand. For instance, the analysis suggests there will be “very little change in the price of AT&T’s standalone broadband post-transaction” and thus the applicants’ standalone broadband offering and pricing commitment was rejected as unnecessary.4 But in the very next paragraph, the importance of a standalone option to the public interest is reiterated and then used as a flawed justification to impose a mandate to offer a discounted standalone broadband program to low-income consumers.5 The imposition of this condition – which is clearly not merger-specific – is likely to result in price increases as the majority of AT&T customers will have to subsidize those AT&T customers that receive this offering. As recognized in the Order, the transaction will allow the applicants to achieve certain efficiencies, freeing up resources that the applicants proposed to invest in a mix of fiber and fixed wireless broadband deployment. The Order goes to great pains to discount any potential benefits of a fixed wireless buildout to 13 million homes in largely rural areas, including many that currently have no access to any type of terrestrial broadband, while on the other hand forcing additional fiber buildout to metropolitan areas that already have a competitive broadband market, in the hope of further increasing 1 Supra para. 105 2 Supra para. 111 3 Supra para. 127 4 Supra paras. 142, 143 5 Supra paras. 144, 145 Federal Communications Commission FCC 15-94 241 competition there. I am skeptical that this reprioritization of metropolitan over rural areas really is in the public interest when so many people continue to go completely unserved. Generally, we should be supportive of innovative, voluntary private sector solutions to connect unserved areas, but in this item we have rejected the opportunity. Consequently, millions of Americans will still be waiting indefinitely for a viable broadband offering while the applicants sink billions of dollars into large, competitive markets at the demand of the government, not market forces. Additionally, the interconnection disclosure condition inches us that much closer to rate regulation, which I have said repeatedly will be the inevitable result of the Commission’s recent net neutrality order, despite all protests to the contrary. Why collect all of this sensitive information, and threaten to release it publicly, if not to eventually intervene in these agreements? While the item states that the record does not contain “any evidence that would support blanket restrictions on all interconnection agreements”6 between the applicants and online video providers, it is clear that the table is being set for exactly that. And then there are the inane limitations on usage based billing practices and the mandate of an independent compliance officer with questionable duties despite the lack of a Commission finding of past digressions with other mergers. While I do not support the specific conditions being imposed, I am willing to concur because the applicants have indicated that they are willing to accede to them and they don’t appear to cause direct harm to other market participants. I look forward to seeing the merged company’s efforts to meet consumer demands for services in the communications landscape. 6 Supra para. 219