*Pages 1--90 from Microsoft Word - 48615.doc* Federal Communications Commission FCC 05- 96 1 Before the Federal Communications Commission Washington, D. C. 20554 In the Matter of The Commission’s Cable Horizontal and Vertical Ownership Limits ) ) ) ) MM Docket No. 92- 264 SECOND FURTHER NOTICE OF PROPOSED RULEMAKING Adopted: May 13, 2005 Released: May 17, 2005 Comment Date: 30 days after publication in the Federal Register Reply Comment Date: 45 days after publication in the Federal Register By the Commission: Commissioners Copps and Adelstein issuing a joint statement. TABLE OF CONTENTS Paragraph I. INTRODUCTION.................................................................................................................................. 1 II. SECOND FURTHER NOTICE OF PROPOSED RULEMAKING.................................................. 17 A. Legal Framework............................................................................................................................. 18 1. Statutory Objectives............................................................................................................ 18 2. Judicial Review and Previous Implementation Efforts...................................................... 21 3. Elements of the Horizontal and Vertical Limits................................................................. 27 4. Approaches to Establishing Horizontal and Vertical Limits ............................................. 38 B. Industry Developments.................................................................................................................... 49 C. Economic Basis for Horizontal Limit.............................................................................................. 61 1. Defining the Market............................................................................................................ 64 a) Programming Market ............................................................................................ 65 b) Programming Distribution Market ....................................................................... 67 c) Relevant Geographic Markets .............................................................................. 69 2. Potential Harms of Horizontal Concentration.................................................................... 71 a) Analytical Frameworks for Economic Analysis of Harms .................................. 71 (1) Open Field Approach ............................................................................ 72 (2) Monopsony Framework......................................................................... 85 (3) Bargaining Power as a Source of Unilateral Anticompetitive Action.. 90 (a) The Use of Bargaining Theory to Establish New Limits ........ 97 1 Federal Communications Commission FCC 05- 96 2 (b) Experimental Economics Study............................................. 101 b) Additional Factors in the Analysis ..................................................................... 105 (1) The Impact of Competition at the Distribution Level ......................... 106 (a) Threshold Approach ............................................................... 114 (2) The Potential for Joint Action.............................................................. 117 (3) The Impact of Independent Actions by Multiple Cable Operators..... 126 (4) The Impact of Vertical Integration ...................................................... 128 (a) Empirical Studies of Foreclosure ........................................... 130 3. Potential Benefits of Horizontal Concentration ............................................................... 137 D. Vertical Limit................................................................................................................................. 143 1. Defining the Market.......................................................................................................... 148 2. Potential Harms of Vertical Integration ........................................................................... 150 3. Potential Benefits of Vertical Integration......................................................................... 154 E. Diversity of Information Sources .................................................................................................. 163 III. PROCEDURAL MATTERS ............................................................................................................. 164 A. Comment Information ................................................................................................................... 164 B. Regulatory Flexibility Act ............................................................................................................. 165 C. Paperwork Reduction Act.............................................................................................................. 166 D. Ex Parte Information ..................................................................................................................... 167 IV. ORDERING CLAUSES .................................................................................................................... 169 APPENDICES A. LIST OF COMMENTERS B. INITIAL REGULATORY FLEXIBILITY ANALYSIS 2 Federal Communications Commission FCC 05- 96 3 I. INTRODUCTION 1. The Cable Television Consumer Protection and Competition Act of 1992 amended the Communications Act of 1934 to provide increased consumer protection and to promote increased competition in the cable television and related markets. 1 Among other things, the 1992 Act added behavioral rules for cable carriage of broadcast signals and retransmission consent; 2 rate regulation; 3 program access obligations with respect to satellite- delivered cable programming; 4 and structural rules intended to address the consequences of increased horizontal concentration and vertical integration in the cable industry. 5 Section 613( f) directed the Commission to conduct proceedings to establish reasonable limits on the number of subscribers a cable operator may serve (horizontal limit), and the number of channels a cable operator may devote to its affiliated programming networks (vertical, or channel occupancy, limit). 6 A principal goal of this comprehensive program was to foster a diverse, robust, and competitive market in the acquisition and delivery of multichannel video programming. 7 Congress intended the structural ownership limits mandated by Section 613( f) to ensure that cable operators did not use their dominant position in the multichannel video distribution (MVPD) 8 market, acting unilaterally or jointly, to unfairly impede the flow of video programming to consumers. 9 At the same time, Congress recognized that multiple system ownership could provide benefits to consumers by allowing efficiencies in the administration, distribution and procurement of programming, and by providing capital and a ready subscriber base to promote the introduction of new programming services. 10 The matters before the Commission in this proceeding stem directly from efforts begun in 1992 to implement Congress’ mandate to balance these competing interests by adopting reasonable cable ownership limits and attribution benchmarks. 11 1 Cable Television Consumer Protection and Competition Act of 1992, Pub. L. No. 102- 385, 106 Stat. 1460 (1992 Act); H. Rep. No. 628, 102d Cong., 2d Sess. 1 (1992) (House Report); Communications Act of 1934, 47 U. S. C. §§ 151, et seq. (Communications Act). 2 Communications Act §§ 614 and 615, 47 U. S. C. §§ 534 and 535. 3 Id. § 623, 47 U. S. C. § 543. 4 Id. § 628, 47 U. S. C. § 548. 5 Id. § 613( f), 47 U. S. C. § 533( f). 6 Id. 7 See S. Rep. No. 92, 102d Cong., 1 st Sess. 1, 18 (1991) (Senate Report); House Report at 27; see also 1992 Act § 2( a)( 4), (b)( 1)-( 5); 47 U. S. C. § 521 (a)( 4), (b)( 1)-( 5). 8 MVPDs include, but are not limited to, providers of cable, multichannel multipoint distribution, direct broadcast satellite, and television receive- only program distribution services that make “available for purchase by subscribers or customers, multiple channels of video programming.” 47 U. S. C. § 522( 13). 9 47 U. S. C. § 533( f)( 2)( A). 10 House Report at 41, 43; see also Senate Report at 27, 33. 11 See, e. g., Implementation of Section 11( c) of the Cable Television Consumer Protection and Competition Act of 1992 Horizontal Ownership Limits, 14 FCC Rcd 19098 (1999) (1999 Cable Ownership Order); Implementation of the Cable Television Consumer Protection and Competition Act of 1992, 14 FCC Rcd 19014 (1999) (1999 Cable (continued….) 3 Federal Communications Commission FCC 05- 96 4 2. The Commission first established a 30% cable subscriber (horizontal) ownership limit and 40% channel occupancy (vertical) rule in 1993. 12 It found that a 30% horizontal ownership limit on cable households passed “is generally appropriate to prevent the nation’s largest MSOs from gaining enhanced leverage from increased horizontal concentration,” while at the same time, “ensur[ ing] that a majority of MSOs continue to expand and benefit from the economies of scale necessary to encourage investment in new video programming services and the deployment of advanced cable technologies.” 13 With respect to the vertical limit, the Commission found that a 40% limit on the number of activated channels that can be occupied by the operator’s affiliated video programming services 14 “is appropriate to balance the goals of increasing diversity and reducing the incentive and ability of vertically integrated cable operators to favor their affiliated programming, with the benefits and efficiencies associated with vertical integration.” 15 The limit applies only to channel capacity up to 75 channels. 16 The 75- channel maximum reflected the Commission’s recognition that expanded channel capacity would reduce the need for channel occupancy limits as a means of encouraging cable operators to carry unaffiliated programming, and that the dynamic state of cable technology required that periodic review of the channel (Continued from previous page) Attribution Order); Implementation of 11( c) of the Cable Television Consumer Protection and Competition Act of 1992 Horizontal Ownership Limits, 13 FCC Rcd 14462 (1998) (1998 Horizontal Reconsideration Order); Implementation of the Cable Television Consumer Protection and Competition Act of 1992, Review of the Cable Attribution Rules, 13 FCC Rcd 12990 (1998); Regulations Governing Attribution of Broadcast and Cable/ MDS Interests, Regulations and Policies Affecting Investment in the Broadcast Industry and Reexamination of the Commission’s Cross- Interest Policy, 11 FCC Rcd 19895 (1996); Implementation of Sections 11 and 13 of the Cable Television Consumer Protection and Competition Act of 1992, 10 FCC Rcd 7364 (1995 Vertical Reconsideration Order); Implementation of Sections 11 and 13 of the Cable Television Consumer Protection and Competition Act of 1992, Horizontal and Vertical Ownership Limits, Cross- Ownership Limitations, and Anti-trafficking Provisions, 8 FCC Rcd 8565 (1993) (1993 Second Report and Order); Implementation of Sections 11 and 13 of the Cable Television Consumer Protection and Competition Act of 1992, Horizontal and Vertical Ownership Limits, 8 FCC Rcd 210 (1992). 12 1993 Second Report and Order, 8 FCC Rcd at 8567 ¶¶ 3- 4. 13 Id. at 8577 ¶ 25. 14 Id. at 8601 ¶¶ 83- 84. The “up to” 75 video channels limit was based on the technological capacity of the average cable system in 1993, which generally limited the number of channels available for distribution of video programming, absent advancements such as signal compression or “fiber to the block,” to approximately 75 channels. Id. at 8601 ¶ 84 & n. 106. The Commission further recognized that the need for a vertical limit would likely decrease as channel capacity increased, so it capped the limit for larger systems of greater than 75 channels. 15 Id. at 8594 ¶ 68. 16 For a system with 75 or fewer channels, the limit is 40% of actual activated channel capacity; 60% of activated channel capacity must be reserved for unaffiliated programming, i. e., 45 channels for a 75 channel system. For systems with 75 or more channels, the limit is applied only to 75 channels, meaning, in effect, that 45 channels on such systems must be reserved for unaffiliated programming (60% of 75). As a result, the limit for larger systems is effectively higher when expressed as a percentage of system capacity, than the limit for systems with 75 channels or fewer. 4 Federal Communications Commission FCC 05- 96 5 occupancy limit be undertaken. 17 In the 1995 Vertical Reconsideration Order, the Commission denied two petitions for reconsideration and reaffirmed its decision regarding the 40% channel occupancy limit. 18 3. To better reflect changed market conditions and allow for organic growth in subscribership, in the 1999 Cable Ownership Order the Commission revised the 30% horizontal limit to permit a cable operator to reach 30% of all MVPD subscribers, rather than solely cable subscribers. 19 As the Commission observed, this was equivalent to establishing a 36.7% cable subscriber limit. 20 This limit was based on the Commission’s determination that cable operators at certain concentration levels, “either by unilateral, independent decisions or by tacit collusion,” could effectively prevent programming networks from entering or surviving the marketplace simply by deciding not to carry a particular network, thereby impeding the flow of programming to the consumer. 21 The Commission estimated that a new cable programming network would need access to 40% of the MVPD subscribers nationwide to be viable. 22 A 30% limit, the Commission reasoned, would allow new programming networks access to a 40% “open field” by ensuring the presence of at least four cable operators in the market, and by preventing the two largest cable operators from garnering more than 60% of the market. 23 4. In proceedings implementing the 1992 Act’s broad structural rules, the Commission determined that use of its broadcast attribution standard was appropriate for defining what constitutes a cognizable interest. 24 Specifically, use of the broadcast attribution benchmarks for horizontal cable ownership and vertical channel occupancy limits was considered appropriate, because, like the broadcast ownership rules, the 1992 Act’s rules governing cable industry structure were designed to ensure competition and diversity in the video marketplace. 25 The 1993 cable horizontal and vertical ownership 17 Id. at n. 86 (measurement of the channel occupancy rule to be done on a per channel basis using the traditional 6 MHz channel definition; periodic review necessary in light of fact that it may soon be common for cable operators to provide several channels using a single 6 MHz bandwidth segment). 18 1995 Vertical Reconsideration Order, 10 FCC Rcd at 7365 ¶ 3. 19 1999 Cable Ownership Order, 14 FCC Rcd at 19101 ¶ 5. 20 Id. at 19101 ¶ 6. 21 Id. at 19114- 16 ¶¶ 38- 44. 22 The 40% “open field” was based on the Commission’s findings that in order to be viable, a new programming network needs to access approximately 15- 20 million subscribers (20% of the market), and that, even with such access, it has only a 50% chance of actually reaching subscribers given tier packaging and consumer preferences. See 1999 Cable Ownership Order, 14 FCC Rcd at 19115- 18 ¶¶ 40- 51. 23 Id. 24 1993 Second Report and Order, 8 FCC Rcd at 8579- 81, ¶¶ 30- 35 (horizontal attribution standard), 8590- 92 ¶¶ 56- 63 (vertical attribution standard). See also, Implementation of the Cable Television Consumer Protection and Competition Act of 1992, Review of the Commission’s Cable Attribution Rules, Notice of Proposed Rulemaking, 13 FCC Rcd 12990, 991- 993 ¶ 2, 4 (1998) (1998 Cable Attribution NPRM). 25 The Commission also observed that the legislative history of the 1992 Act expressly suggested use of the broadcast attribution standard. 1998 Cable Attribution NPRM, 13 FCC Rcd at 12993 ¶ 4. The “general” cable attribution rules apply to such broad structural limitations as the horizontal ownership limits, 47 C. F. R. § 76.503; channel occupancy limits, 47 C. F. R. § 76.504; cable/ SMATV cross- ownership, 47 C. F. R. § 76.501( d); and cable-( continued….) 5 Federal Communications Commission FCC 05- 96 6 attribution standards mirrored the broadcast attribution rules, and inter alia, attributed all corporate voting stock interests of 5% or more and contained an exemption to the voting stock threshold under which a minority corporate shareholder’s voting interests were not attributed in cases where a single shareholder owns more than 50% of the outstanding voting stock of the corporation. 26 Both broadcast and cable standards attributed partnership interests, except properly “insulated” limited partnership interests. 27 In 1998, the Commission launched a comprehensive review of the cable attribution rules in light of recent developments in the cable industry together with the Commission’s review, in a separate proceeding, of the broadcast attribution rules on which many of the cable attribution rules were based. 28 5. In the 1999 Cable Attribution Order, the Commission revised several aspects of its cable attribution rules to track certain changes made to the broadcast attribution rules. In addition, in a departure from the broadcast attribution rules, the 1999 Cable Attribution Order eliminated the single-majority shareholder exemption to its general cable attribution standard and relaxed one of the limited partner insulation criteria, which, if satisfied, keep a limited partnership interest from being attributed to a limited partner, to permit a broader range of activities performed on behalf of the partnership by a limited partner while still remaining insulated. 29 In general, limited partners cannot be relieved from attribution unless they are not materially involved in the management or operations of the media entity concerned (the “no material involvement” standard). In setting specific guidance as to what kind of insulation is sufficient to exempt a limited partnership interest from attribution, the Commission originally established seven criteria, collectively referred to herein as the “ILP exception,” which, if met would make it safe to presume that a limited partner will not be materially involved in the media management and operations of the partnership. 30 The sixth insulation criteria applicable to cable ownership generally barred a limited partner from performing “any services to the partnership relating to its media activities.” The Commission narrowed this prohibition to exclude only services performed by the limited partner for the partnership that are materially related to the partnership’s video programming activities, thus broadening (Continued from previous page) telco buyout prohibition, 47 C. F. R. § 76.505. In contrast, for those rules implemented under the 1992 Act to deter specific improper practices and also to promote competition and diversity, such as commercial leased access and program access, the Commission adopted additional, stricter cable attribution standards. See 1998 Cable Attribution NPRM, 13 FCC Rcd at 12993 ¶ 5; 1999 Cable Attribution Order, 14 FCC Rcd at 19054 ¶ 104. These stricter attribution standards are also referred to as “program access” attribution standards. 1999 Cable Attribution Order, 14 FCC Rcd at 19051 ¶ 93. 26 1993 Second Report and Order, 8 FCC Rcd at 8580- 81 ¶ 34. See former 47 C. F. R. § 73.3555 Note 2( b); former 47 C. F. R. § 76.501Note 2( b). For passive institutional investors, voting stock interests of 10% or more were attributable. See former 47 C. F. R. § 76.501 Note 2( c). 27 47 C. F. R. § 73.3555 Note 2( a) and (f); 47 C. F. R. § 76.501Note 2( a) and (f). 28 1998 Cable Attribution NPRM, 13 FCC Rcd at 12990 ¶ 1; citing Review of The Commission’s Regulations Governing Attribution of Broadcast Interests, Notice of Proposed Rulemaking, 10 FCC Rcd 3606 (1995); Review of The Commission’s Regulations Governing Attribution of Broadcast and Cable/ MDS Interests, Further Notice of Proposed Rulemaking, 11 FCC Rcd 19895 (1996). 29 See 1999 Cable Attribution Order, 14 FCC Rcd at 19039- 41, 19046 ¶¶ 61- 64, 81. Under the original ILP exemption, a limited partner could not be materially involved in the “media activities” of the partnership and retain insulation. See former 47 C. F. R. § 76.501 Note 2( f). 30 1999 Cable Attribution Order, 14 FCC Rcd at 19039- 40 ¶ 61. 6 Federal Communications Commission FCC 05- 96 7 the range of activities that could performed without loss of insulation for the limited partner. 31 The Commission also indicated that a limited partner’s insulation would be lost if an agreement for the sale of programming was entered into between the limited partner and the partnership. 32 6. The United States Court of Appeals for the District of Columbia Circuit in Time Warner Entertainment Co. v. FCC (Time Warner II) reversed and remanded the Commission’s 30% horizontal ownership limit and its 40% channel occupancy limit. 33 Additionally, the court vacated the Commission’s decision to eliminate the single majority shareholder exemption to its general cable attribution rules and the “no sale” aspect of the limited partnership insulation criteria. 34 The court found that the horizontal and vertical ownership limits unduly burdened cable operators’ First Amendment rights, that the Commission’s evidentiary basis for imposing the ownership limits and its rationales supporting the vacated attribution rules did not meet the applicable standards of review, and that the Commission had failed to consider sufficiently changes that have occurred in the MVPD market since passage of the 1992 Act. 35 In response, the Commission issued a Further Notice of Proposed Rulemaking (2001 Further Notice). 36 7. In the 2001 Further Notice, the Commission solicited comment on the nature of the MVPD industry, industry changes since the 1992 Cable Act, how these changes affected the implementation of horizontal and vertical limits, and various proposals for a new horizontal limit. The Commission sought to develop an evidentiary basis for setting limits, sought to establish the need for vertical limits and their optimal level, and sought comment and evidence on the attribution rules 31 See 1999 Cable Attribution Order, 14 FCC Rcd at 19039- 41 ¶¶ 61- 64; Applications for Consent to the Transfer of Control of Licenses and Section 214 Authorizations from MediaOne Group, Inc., Transferor, to AT& T Corp., Transferee, (AT& T- MediaOne Order) 15 FCC Rcd 9816, 9839 ¶ 45 (2000). 32 See 1999 Cable Attribution Order, 14 FCC Rcd at 19055 ¶ 106. 33 240 F. 3d 1126 (D. C. Cir. 2001) (Time Warner II). The D. C. Circuit upheld the underlying statute in Time Warner Entertainment Co. v. United States, 211 F. 3d 1313 (D. C. Cir. 2000) (Time Warner I). 34 Time Warner II, 240 F. 3d at 1142- 43. 35 Id. at 1130- 40. The cable ownership rules were not vacated by the court in Time Warner II. In addition, as the court noted, the Commission’s voluntary stay of enforcement of the horizontal limit “ended automatically” upon the reversal of the District Court’s decision in Daniels Cablevision, Inc. v. United States, 835 F. Supp. 1 (D. D. C. 1993) (Daniels). Time Warner II, 240 F. 3d at 1128. The cable horizontal ownership cap has been reversed and remanded, and we have not yet determined what rules will best effectuate Congress’ intent in enacting section 613( f) of the Communications Act. If presented with a proposed merger or other transaction involving a cable operator that called into question compliance with our rules during the pendency of this rulemaking, we remain obligated to ensure that the resulting firm’s national subscriber reach would not result in the harms to competition and consumers that the horizontal cap is intended to prevent (i. e., ensuring that no cable operator can unfairly impede the flow of video programming from the programmer to the consumer). 36 Implementation of Section 11 of the Cable Television Consumer Protection and Competition Act of 1992, 16 FCC Rcd 17312 (2001) (2001 Further Notice). After the 2001 Further Notice, the Commission suspended the elimination of the broadcast single majority shareholder exemption pending the outcome of this proceeding. See Review of The Commission’s Regulations Governing Attribution of Broadcast and Cable/ MDS Interests, Order, 16 FCC Rcd 22310 (2001). 7 Federal Communications Commission FCC 05- 96 8 addressed by the court. 37 The Commission also sought information concerning the contractual relationships between programmers and cable operators in order to establish the extent of cable operators’ market power and the effects of market power on the quantity and quality of programming, as well as the effects of market power on the programming costs of smaller MVPDs. 38 Further, the Commission sought comment on consumers’ access to alternative MVPDs, particularly Direct Broadcast Satellite (DBS), and their effect on competition in the MVPD market. 39 8. The 2001 Further Notice also asked commenters to address, with empirical and/ or theoretical evidence, the single majority shareholder exemption and application of the limited partnership insulation criteria to bar programming sales, and, in light of the Time Warner II decision, also sought evidence on whether to reinstate the single majority shareholder exemption for purposes of the broadcast and cable/ multipoint distribution service (MDS) attribution rules. 40 The Commission is currently reviewing these issues and expects to address expeditiously the broadcast and cable single majority shareholder exemption and the cable limited partnership insulation criteria. This Second Further Notice of Proposed Rulemaking (Second Further Notice) seeks updated and more specific comment on the Commission’s remanded cable horizontal and vertical ownership limits. 9. Commenters to the 2001 Further Notice 41 offered a range of viewpoints on the ownership questions, arguing at one end of the spectrum that the horizontal cap should remain at 30% or be lowered, 42 and proposing at the other end that the cap be eliminated. 43 Other commenters advocated using a case- by- case approach 44 or a local market- by- market approach. 45 However, none of the comments yielded a sound evidentiary basis for setting horizontal or vertical limits as demanded by the D. C. Circuit. While many commenters presented theoretical, legal or economic arguments and anecdotal evidence, no party provided a compelling approach that supported a particular horizontal or vertical limit. As discussed in detail below, the economic analyses submitted are informative, but not dispositive – we find that they are either unconvincing in light of current marketplace conditions or are simply generalized economic theories that do not provide a sound evidentiary basis for adopting a particular limit. The 37 2001 Further Notice, 16 FCC Rcd at 17320- 21 ¶ 7. 38 Id. at 17316- 34 ¶¶ 2- 45; 17338- 47 ¶¶ 50- 73; 17349- 52 ¶¶ 76- 84. 39 Id. at 17325- 28 ¶¶ 18- 26; see also Time Warner II, 240 F. 3d at 1133- 34. 40 Id. at 17355 ¶ 87. In 2004, MDS/ MMDS was renamed the Broadband Radio Service (BRS) by the Commission. See Amendment of Part 1, 21, 73, 74 and 101 of the Commission’s Rules to Facilitate the Provision of Fixed and Mobile Broadband Access, Educational and Other Advanced Services in the 2150- 2162 and 2500- 2690 MHz Bands, et al., 19 FCC Rcd 14165, 14227- 32 ¶¶ 165- 76 (2004). 41 Appendix A provides a list of commenters and the abbreviations by which they are identified herein. 42 CFA Comments at 25. 43 Time Warner Comments at 9. 44 See, e. g., Letter from Michael H. Hammer, Willkie, Farr & Gallagher, on behalf of Comcast, to Marlene H. Dortch, Secretary, FCC (Mar. 13, 2003). 45 RCN Comments at 18. 8 Federal Communications Commission FCC 05- 96 9 passage of time since this record closed gives us additional reason to provide interested parties the opportunity to both augment and refresh the evidentiary record. 10. In addition, the Commission subsequently sought to augment the record by means of a programming network survey and an experimental economics analysis. The Commission’s efforts to obtain empirical data and information through the programming network survey yielded little useful information. 46 The Commission then conducted, and released for comment, an experimental economics analysis designed to determine whether changes in concentration may impede the flow of programming to consumers (BKS Study), 47 and developed theoretical analyses designed to determine the relationship between bargaining power and buyer size in a bilateral bargaining environment. 48 The BKS Study created an experimental market that included many of the features of the actual market in which MVPDs and cable programming networks negotiate affiliate fees (e. g., trades involving differentiated products, differences in the level of non- avoidable sunk costs incurred by buyers and sellers, and the use of a sequential bilateral bargaining process to negotiate fees). The study found that increasing horizontal concentration could impede the flow of programming and, by at least one measure, indicated that impairment would be likely to occur at a level of concentration somewhere between a 44% and a 51% market share. 49 The impairment could cause networks to cease operation or reduce the quality of programming delivered to consumers. However, the BKS Study did not model some potentially important aspects of the industry (i. e., vertical integration, retail competition from DBS, entry into and exit from the cable network programming industry, differences in Most Favored Nation (MFN) agreements across different- sized buyers). Similarly, the theoretical work of Adilov and Alexander suggests that, under certain conditions, increased firm size can produce an improved bargaining position and adversely affect the flow of programming. 50 While these analyses of bargaining power show that increasing horizontal size imparts increased bargaining power to the largest buyer of programming, they are imprecise in determining the point at which such increased bargaining power impedes the flow of programming. 11. In addition to the deficiencies in the record, a number of significant events have occurred since the release of the 2001 Further Notice that must be taken into account in fashioning cable ownership limits. First, the 2002 Comcast- AT& T cable transaction resulted in one entity having a share 46 See Letter from W. Kenneth Ferree, Chief, Cable Services Bureau, FCC, to Programming Network Owners (Feb. 15, 2002). The letter sought information from programming network owners for each network in which they had an interest, including the number of subscribers at the time the network became profitable, the number of subscribers at the end of calendar years 1997- 2001, and information on the vertical integration status and genre of each network. 47 Mark Bykowsky, Anthony Kwasnica, & William Sharkey, Horizontal Concentration in the Cable Television Industry: An Experimental Analysis, FCC Office of Plans and Policy, Working Paper No. 35 (June 2002 & rev. July 2002) (BKS Study). The BKS Study was released for public comment and generated a substantial record in response. 48 Nodir Adilov & Peter J. Alexander, Asymmetric Bargaining Power and Pivotal Buyers, FCC Media Bureau Working Paper No. 13 (Sept. 2002) (Asymmetric Bargaining Power); Nodir Adilov & Peter J. Alexander, Most-Favored Customers in the Cable Industry, FCC Media Bureau Working Paper No. 14 (Sept. 2002). 49 See ¶¶ 101- 102, infra. 50 Asymmetric Bargaining Power, supra n. 48. 9 Federal Communications Commission FCC 05- 96 10 of MVPD subscribers very close to our remanded 30% ownership limit. 51 Second, the 2003 News Corp. -Hughes transaction created the first vertically integrated DBS operator, involving a number of video programming assets. 52 Third, courts have remanded media ownership rules in three decisions, requiring that the Commission more firmly base its rules on empirical data and record evidence. 12. In 2002, two of the Commission’s broadcast ownership rules were reviewed and remanded by the D. C. Circuit in Fox Television Stations, Inc. v. FCC and Sinclair Broadcasting Group, Inc. v. FCC. 53 While the court in Fox agreed that, “[ i] n the context of the regulation of broadcasting, ‘the public interest’ has historically embraced diversity (as well as localism),” it found that the Commission had “not provide[ d] an adequate basis for believing the Rule would in fact further” those interests. 54 Similarly, while the court in Sinclair found that the Commission had “adequately explained how the local ownership rule furthers diversity at the local level and is necessary in the ‘public interest’ under §202( h) of the 1996 Act,” it remanded the rule, finding that the Commission had “not provided any justification for counting fewer types of ‘voices’ in the local [television] ownership rules than it counted in its rule on cross- ownership of radio and television stations.” 55 13. In June 2003, the Commission adopted substantial revisions to its broadcast ownership rules in the Biennial Review Order. 56 We replaced the newspaper/ broadcast and radio/ television cross-ownership rules with a set of cross- media limits; modified the local television multiple ownership rule; modified the local radio ownership rule and its market definition; modified the national television ownership rule; and retained the dual network rule. 57 In 2004, in Prometheus Radio Project, et al. v. FCC 51 See Applications for Consent to the Transfer of Control of Licenses, Comcast Corporation and AT& T Corp., Transferors, to AT& T Comcast Corporation, Transferee (Comcast- AT& T Order), 17 FCC Rcd 23246 (2002). 52 See General Motors Corporation and Hughes Electronics Corporation, Transferors and The News Corporation Limited, Transferee, For Authority to Transfer Control (News- Hughes Order), 19 FCC Rcd 473 (2003). The programming assets involved in the transaction included 35 owned and operated (O& O) full- power television broadcast stations, a national television broadcast network, ten national cable programming networks, and 22 regional cable programming networks. 53 See Fox Television Stations, Inc. v. FCC, 280 F. 3d 1027, modified on rehearing, 293 F. 3d 537 (D. C. Cir. 2002) (Fox) and Sinclair Broadcasting Group, Inc. v. FCC, 284 F. 3d 148 (D. C. Cir. 2002) (Sinclair). The court in Fox remanded the Commission’s retention of the then congressionally- established 35% national television ownership rule. See 1998 Biennial Review – Review of the Commission’s Broadcast Ownership Rules and Other Rules Adopted Pursuant to Section 202 of the Telecommunications Act of 1996, 15 FCC Rcd 11058 (2000). The court in Sinclair remanded the Commission’s 1999 revision of its local television multiple ownership rule. See Review of the Commission’s Regulations Governing Television Broadcasting, 14 FCC Rcd 12903 (1999). 54 Fox, 280 F. 3d at 1042, 1043. 55 Sinclair, 284 F. 3d at 160, 162. 56 2002 Biennial Regulatory Review – Review of the Commission’s Broadcast Ownership Rules and Other Rules Adopted Pursuant to Section 202 of the Telecommunications Act of 1996, 18 FCC Rcd 13620 (2003) (2002 Biennial Review Order). See also Telecommunications Act of 1996, Pub. L. No. 104- 104, 110 Stat. 56 (1996). 57 Id. Congress subsequently amended Section 202( c) of the Telecommunications Act of 1996, directing the Commission to modify the national television ownership limit, contained in 47 C. F. R. § 73.3555, to 39%. See Consolidated Appropriations Act, 2004, Pub. L. No. 108- 199, §. 629, 118 Stat. 3 (2004). 10 Federal Communications Commission FCC 05- 96 11 (Prometheus), the Third Circuit remanded the cross- media limits, the local television multiple ownership rule, and the local radio ownership rule, finding them inadequately justified. 58 The court held that “[ d] eference to the Commission’s judgment is highest when assessing the rationality of the agency’s line-drawing endeavors,” 59 but then directed the Commission to better explain its conclusions. 60 14. These court decisions are instructive as we attempt to fashion cable horizontal and vertical ownership limits. The Fox, Sinclair, Prometheus, and Time Warner II courts all remanded ownership limits under consideration with instructions for the Commission to better justify its decisions on the basis of the record evidence. 61 Each proceeding involved line- drawing determinations to establish ownership limits, and in each the Commission attempted to create rules that would promote policy goals that inherently are not easily measured or quantified. The broadcast ownership proceedings involved an assessment of the continued public interest need for national and local broadcast ownership limits, including local cross- media ownership restrictions applicable to local broadcast television, radio, and newspaper outlets. In setting these limits, the Commission sought to demonstrate how its decisions would promote diversity and localism, as well as competition, based on a wide array of empirical and theoretical evidence. 15. In the cable ownership realm, the Commission is directed by statute to promote effective competition and ensure diversity. 62 Specifically, we must determine at what point cable horizontal reach will unfairly impede the flow of programming, a somewhat fluid concept susceptible to a variety of interpretations, and our vertical limit must be designed to achieve the statutory goals by means of a channel occupancy limit – the mechanism specified by Congress for this purpose. 63 Although courts and agencies routinely attempt to measure and quantify competition, our task in this proceeding is complicated by the possibility that the harms our rules are designed to prevent may arise at concentration levels higher than those that exist in today’s markets. As we explain in more detail below, in examining a variety of economic theories of harm relevant to cable ownership limits, it has been difficult to ascertain how hypothetical market conditions might affect competition and diversity. In the face of these difficulties, Fox, Sinclair, Prometheus, and Time Warner II instruct us to draw a reasoned and specific connection between the record evidence and each element of our horizontal and vertical ownership limits, a task we cannot adequately accomplish on the basis of the record compiled in response to the 2001 Further Notice. 58 See Prometheus Radio Project, et al. v. FCC, 373 F. 3d 372 (3rd Cir. 2004) (Prometheus). The court found that “[ b] ecause the Commission is under a statutory directive to modify the national television ownership cap to 39%, challenges to the Commission’s decision to raise the cap to 45% cap are moot.” Id. at 396. 59 Id. at 410- 11 (citations omitted). 60 Id. at 435. The court “identified several provisions in which the Commission falls short of its obligation to justify its decisions to retain, repeal, or modify its media ownership regulations with reasoned analysis.” Id. 61 Fox, 280 F. 3d at 1044; Sinclair, 284 F. 3d at 169; Prometheus, 373 F. 3d at 390; Time Warner II, 240 F. 3d at 1128. 62 See 47 U. S. C. § 533( f)( 1). 63 See 47 U. S. C. § 533( f)( 1)( A) and (B). 11 Federal Communications Commission FCC 05- 96 12 16. We therefore conclude that a Second Further Notice is necessary to update the record and provide additional input on horizontal and vertical ownership limits so that we may comply with our statutory mandate and the court’s directives in Time Warner II. We seek comment on the proposals in the record, recent developments in the industry, and our tentative conclusions described below. We ask commenters to supplement the record where possible by providing new evidence and information to support the formulation of horizontal and vertical limits, and we invite parties to undertake their own studies in order to further inform the record. We also invite comment on Media Bureau Staff Research Paper No. 2004- 1 which examines the effect of subscribership on a network’s ability to survive in the marketplace. 64 Once the record in this Second Further Notice is complete, we intend to expeditiously address the issues contained therein, and enact sustainable cable horizontal and vertical ownership limits. II. SECOND FURTHER NOTICE OF PROPOSED RULEMAKING 17. As stated above, in 1992, Congress enacted Section 613( f) of the Communications Act to address its concern that the trend towards horizontal and vertical concentration in the cable industry could affect and potentially impede the flow of programming to consumers due to cable operators’ size and market power. One goal of the 2001 Further Notice was to solicit public comment and develop an evidentiary basis for setting horizontal and vertical limits in the current dynamic and evolving communications marketplace. Unfortunately, as previously noted, the record developed thus far does not contain sufficient evidence that would allow us to set reasonable and sustainable horizontal and vertical ownership limits. This Second Further Notice is therefore necessary to update the record and provide additional input on ownership limits so that we may comply with our statutory mandate and the court’s directives in Time Warner II. We retain the original record in this proceeding, and commenters should therefore avoid merely repeating their previously filed comments. Instead, commenters should address how recent developments in the industry may affect our analysis, and provide, where available, new evidence and information to support the formulation of horizontal and vertical limits. Additionally, to develop a more focused and useful record, in this Second Further Notice, we address the viability of proposals for setting limits suggested in the record. A. Legal Framework 1. Statutory Objectives 18. In 1992, Congress recognized that cable operators’ increasing horizontal concentration and vertical integration could frustrate competition in the production and delivery of multichannel video programming. 65 Specifically, the Senate Report concluded that increased horizontal concentration could “give cable operators the power to demand that programmers . . . [provide] cable operators an exclusive 64 Keith S. Brown, A Survival Analysis of Cable Networks, Media Bureau Staff Research Paper No. 2004- 1 (rel. Dec. 7, 2004) (Survival Analysis). The Survival Analysis uses the statistical tools of survival and duration analysis to estimate how different variables affect a cable network’s probability of survival and expected length of life. Using these results, the study estimates the number of subscribers a cable network needs for any given probability of survival over a given length of time. The Survival Analysis concludes, for example, that a network growing at an average rate requires approximately 42 million subscribers to have a 70% probability of survival over its first 10 years. The study is being placed in the record of this proceeding concurrently with the release of the Second Further Notice. 65 Senate Report at 24. 12 Federal Communications Commission FCC 05- 96 13 right to carry the programming, a financial interest or some other added consideration as a condition of carriage.” 66 More generally, the Senate Report stated, “a market that is dominated by one buyer of a product, a monopsonist, does not give the seller any of the benefits of competition.” 67 Congress was also concerned that an increase in vertical integration between cable operators and programmers may provide incentives and opportunities for cable operators to favor affiliated over non- affiliated programmers. 68 19. A principal objective of the 1992 Act was to foster competition in the acquisition and delivery of multi- channel video programming by encouraging the development of alternative and new technologies, including cable and non- cable systems. 69 Congress evidenced a preference for competition over regulation in order to achieve this objective, believing that the presence of alternative cable and non-cable MVPDs would constrain cable operators’ market power in the acquisition and distribution of multi-channel video programming, 70 as well as improve their service and programming quality and curb their subscription rate increases. 71 As detailed below, however, Congress found that the cable industry, the nation’s dominant and increasingly horizontally concentrated medium for the delivery of multi- channel programming, faced virtually no competition at the local level, and only limited competition at the regional and national level. 72 Additionally, Congress found that the increase in vertical integration between cable operators and programmers provided incentives and opportunities for cable operators to favor affiliated over non- affiliated programmers and, likewise, for programmers to favor affiliated over 66 Id. 67 Id. at 33 (“ Witnesses . . . testified that, with the increased concentration in the cable industry, the large MSOs have the market power to determine what programming services can ‘make it’ on cable.”). 68 Id. at 24; see also 1992 Act §§ 2( a)( 5)-( 6); House Report at 41. 69 See generally Senate Report, House Report; see also 1992 Act §§ 2( b)( 1)-( 5). 70 See Senate Report at 12, 18, 20- 24; House Report at 30, 44; see also Annual Assessment of the Status of Competition in the Market for the Delivery of Video Programming, CS Docket No. 00- 132, 7 th Annual Report, 16 FCC Rcd 6005, 6007 n. 4 (2001) (the 1992 Act “imposed a regulatory scheme on the cable industry designed to serve as a transitional mechanism until competition develops and consumers have adequate multi- channel video programming alternatives”). In fact, experience has shown that competition does result in lower rates, improved service, and increased programming fare. Id. at 6092- 98 ¶¶ 213- 34. 71 Various provisions of the 1992 Act reflect congressional concern “about concentration of the media in the hands of a few who may control the dissemination of information” at the local, regional and national levels. Senate Report at 32. See, e. g., 47 U. S. C. § 543( b)( 1) (requiring the Commission to issue rules to protect subscribers of “any cable system that is not subject to effective competition” from excessive rates); 47 U. S. C. § 541( a)( 1) (prohibiting local authorities from granting exclusive franchises or unreasonably refusing to award additional franchises); 47 U. S. C. § 533( a)( 2) (limiting cable operators from owning MMDS or SMATV systems within their franchise areas); 47 U. S. C. § 533( d) (allowing local authorities to deny transfers of franchises that would reduce or eliminate competition in the delivery of cable services); 47 U. S. C. § 544( b)( 2)( C) (requiring the Commission to issue rules that promote the commercial availability of cable consumer equipment); 47 U. S. C. § 547( b) (prohibiting cable operators from engaging in unfair practices vis- à- vis video programmers and other MVPDs). 72 See 1992 Act §§ 2( a)( 2)-( 4), (6); see also Senate Report at 12, 13- 18, 20, 32- 34; House Report at 27, 43- 47. 13 Federal Communications Commission FCC 05- 96 14 non- affiliated operators in the distribution of video programming. 73 Thus, given the absence of competition at the time, Congress believed that certain structural limits were necessary. 74 20. To combat these harms, Section 613( f)( 1) of the Communications Act directs the Commission, “in order to enhance effective competition,” to conduct proceedings and to set a reasonable horizontal limit “on the number of cable subscribers a person is authorized to reach through cable systems owned by such person, or in which such person has an attributable interest,” and a reasonable vertical limit “on the number of channels on a cable system that can be occupied by a video programmer in which the cable operator has an attributable interest.” 75 Section 613( f)( 2) directs that, in setting these limits, “the Commission shall, among other public interest objectives:” (A) ensure that no cable operator or group of cable operators can unfairly impede, either because of the size of any individual operator or because of joint actions by a group of operators of sufficient size, the flow of video programming from the video programmer to the consumer; (B) ensure that cable operators affiliated with video programmers do not favor such programmers in determining carriage on their cable systems or do not unreasonably restrict the flow of the video programming of such programmers to other video distributors; (C) take particular account of the market structure, ownership patterns, and other relationships of the cable television industry, including the nature and market power of the local franchise, the joint ownership of cable systems and video programmers, and the various types of non- equity controlling interests; (D) account for any efficiencies and other benefits that might be gained through increased ownership or control; (E) make such rules and regulations reflect the dynamic nature of the communications marketplace; (F) not impose limitations which would bar cable operators from serving previously unserved rural areas; and (G) not impose limitations which would impair the development of diverse and high quality video programming. 76 73 See Senate Report at 24 (“ when cable systems are not subject to effective competition . . . [p] rogrammers either deal with operators of such systems on their terms or face the threat of not being carried in that market. The Committee believes this disrupts the crucial relationship between the content provider and the consumer . . . . Moreover, these concerns are exacerbated by the increased vertical integration in the cable industry.”); see also 1992 Act §§ 2( a)( 5)-( 6); House Report at 41. 74 See Senate Report at 18, 25- 26, 33; House Report at 26, 30, 40- 44. 75 47 U. S. C. § 533( f)( 1)( A)&( B). 76 47 U. S. C. § 533( f)( 2)( A)-( G). 14 Federal Communications Commission FCC 05- 96 15 2. Judicial Review and Previous Implementation Efforts 21. Facial Challenge. Section 613( f) ultimately survived the cable industry’s challenge that horizontal and vertical limits violate cable operators’ First Amendment rights by excessively limiting their speech. 77 In 2000, the Time Warner I court concluded that Section 613( f) resulted from a fear that an increase in concentration and vertical integration could result in anticompetitive behavior by cable operators toward programming suppliers, as well as toward potential new MVPD entrants. The court accepted these concerns as “well grounded in the evidence and a bit of economic common sense” and found them to be important government interests justifying an infringement of the cable operators’ right to free expression. 78 22. Horizontal Limit. In the 1993 Second Report and Order, the Commission found that a limit of 30% of households passed by all cable operators represented a careful balance between: (1) limiting the possible exertion by a cable operator of excessive market power in the purchase of video programming; and (2) ensuring that cable operators are able to expand and benefit from the economies of size necessary to encourage investment in new video programming technology and the deployment of other advanced technologies. 79 In the 1998 Horizontal Reconsideration Order, the Commission sought 77 Time Warner I, 211 F. 3d 1313. This facial challenge was launched in 1993, resulting in a judgment that same year that the horizontal “subscriber limits provision unconstitutionally abridged the First Amendment rights of cable operators,” while the vertical channel occupancy provision did not. See Daniels, 835 F. Supp. 1. The Daniels court also decided that, because “there is substantial ground for difference of opinion” as to the constitutionality of the underlying statute, it would stay its proceedings and the issuance of any relief to the plaintiff’s pending appeal. The government appealed the former ruling, while Time Warner appealed the latter. Time Warner I, 211 F. 3d at 1315. Thereafter, the Commission issued the 1993 Second Report implementing the challenged provisions, but voluntarily stayed the effective date of its rules pending the appeal in Daniels. Time Warner then challenged the cable ownership rules in Time Warner Entertainment Co., L. P. v. FCC, No. 94- 1035 (D. C. Cir. 1994). The D. C. Circuit consolidated the Daniels and Time Warner appeals in Time Warner Entertainment Co., L. P. v. FCC, 93 F. 3d 957 (D. C. Cir. 1996), and held the consolidated appeals in abeyance pending the Commission’s decision on petitions for reconsideration. Id. at 979- 80. In 1998, the Commission issued a Second Order on Reconsideration and Further Notice of Proposed Rulemaking, 13 FCC Rcd 14462 (1998), maintaining the 30% horizontal limit and denying a motion to lift the stay on enforcement of the horizontal limit. Once the Commission issued the Second Order on Reconsideration, the D. C. Circuit lifted its stay on its consideration of the consolidated Daniels and Time Warner proceedings, issuing a decision reversing the Daniels decision two years later in Time Warner I. In the Second Order on Reconsideration, the Commission also continued its stay of the effective date of the horizontal ownership rules pending a decision by the D. C. Circuit on challenges to the ownership rules and Section 613( f). Id. at 14492 ¶ 75. Once the Time Warner I court upheld the constitutionality of the underlying statute, the Commission’s voluntary stay of the effectiveness of its rules ended automatically, and the rules went into effect. See Time Warner II, 240 F. 3d at 1128. The sequential series of decisions, revisions, and appeals resulted in a seven year delay of appellate resolution of the facial challenge to the constitutionality of the underlying statute and in appellate review of the appropriateness of the Commission’s implementation of the statute not being resolved until eight years after release of the 1993 Second Report and Order, which established the basic horizontal and vertical implementation framework at issue in this proceeding. 78 Time Warner I, 211 F. 3d at 1322. 79 1993 Second Report and Order, 8 FCC Rcd at 8569, 8582- 84 ¶¶ 8, 37- 42. The Commission also stated that it intended to review the horizontal limits every five years in order to determine whether they were still reasonable under new market conditions and continued to meet their policy objectives. Id. at 8583 ¶ 40. 15 Federal Communications Commission FCC 05- 96 16 comment on possible revisions of the horizontal ownership rules and the method by which horizontal ownership is calculated. 80 In the 1999 Cable Ownership Order, the Commission revised the 30% horizontal limit on households passed by all cable operators to include all cable and non- cable MVPD subscribers in the calculation of the appropriate horizontal market, a change it stated was needed to reflect the growing impact emerging non- cable MVPDs were having on the programming marketplace. 81 Under the revised rule, a horizontal subscriber limit of no more than 30% of MVPD subscribers would be equivalent to a 36.7% limit based on cable subscribership alone. The Commission characterized its action as a “significant relaxation of the rule,” that retained the “theoretical underpinnings” of its original 30% limit while taking account of marketplace changes through revision of the definition of the relevant market as all MVPD subscribers. 82 The 30% limit continued to be based on the Commission’s underlying theory that cable operators at certain concentration levels, “either by unilateral, independent decisions or by tacit collusion,” could effectively prevent programming networks from entering or surviving in the marketplace simply by deciding not to carry a particular network. 83 Analyzing industry data, the Commission estimated that a new cable programming network would need access to 40% of the MVPD subscribers nationwide to be viable. 84 A 30% limit, the Commission reasoned, would allow new programming networks access to a 40% “open field” by ensuring the presence of at least four cable operators in the market, and by preventing the two largest cable operators from garnering more than 60% of the market. 85 In this regard, the Commission explained, “even if two operators, covering 60% of the market, individually or collusively deny carriage to a programming network, the network would still have access to 40% of the market, giving it a reasonable chance of financial viability.” 86 23. The Time Warner II court rejected the Commission’s approach to calculating the horizontal limit. The court found that the Commission lacked any evidence that cable operators would collude and that it could not simply assume that cable operators would coordinate their behavior in this fashion. The court held that Section 613( f)( 1) authorizes the Commission to set a limit that would prevent a “single company” from foreclosing entry of a programming network, 87 but does not authorize the agency to regulate the “legitimate, independent editorial choices of multiple MSOs.” 88 Having found that the Commission did not provide an adequate evidentiary basis to assume two operators might engage in joint anticompetitive conduct, the court concluded that the record would support a limit no lower than 80 13 FCC Rcd at 14464- 65 ¶ 4. 81 1999 Cable Ownership Order, 16 FCC Rcd at 19031 ¶ 37. 82 Id. 83 Id. at 19116 ¶ 43; see also ¶ 3, supra. 84 The 40% “open field” was based on the Commission’s findings that in order to be viable, a new programming network needs access to approximately 15- 20 million subscribers (20% of the market), and that, even with such access, it has only a 50% chance of actually reaching subscribers given tier packaging and consumer preferences. See 1999 Cable Ownership Order, 14 FCC Rcd at 19115- 18 ¶¶ 40- 50. 85 Id. 86 Id. at 19119 ¶ 53. 87 Time Warner II, 240 F. 3d at 1131. 88 Id. at 1130- 35. 16 Federal Communications Commission FCC 05- 96 17 60% using the 40% open field premise. 89 The court also required that in fashioning another limit, we recognize that market power depends not only on market share but on the “availability of competition.” 90 24. Vertical Limit. In the 1993 Second Report and Order, the Commission adopted a channel occupancy limit that prohibited a cable operator from carrying video programming services it owns or in which it has an attributable interest on more than 40% of its activated channels. 91 In setting the vertical ownership limit at 40%, the Commission sought to “maximize the number of voices available to cable viewers without impairing the ability or incentive of cable operators to invest in new and existing video programming services.” 92 The Commission recognized that, although Section 613( f) contemplated the establishment of some limits on cable vertical integration, “MSO investment was responsible for the development and survival of several of the most popular video programming services,” and that “vertical integration among the largest MSOs had contributed to program diversity by providing new programming services with an extensive subscriber base and information regarding viewer tastes and desires.” 93 The Commission also recognized that vertical integration can produce efficiencies with respect to video programming acquisition, distribution and marketing, which might contribute to innovative programming fare and lower subscription charges. 94 The Commission believed that the 40% limit was “high enough to preserve the benefits of vertical integration,” 95 and further relied upon the fact that most cable operators who filed comments in the rulemaking proceeding supported the 40% limit. 96 25. The Commission recognized that the need for a vertical limit would likely decrease as channel capacity increased. Thus, the Commission’s rule applies to channel capacity only “up to 75 channels.” 97 As a result, for higher capacity systems, the percentage limit is effectively much higher than 89 Id. at 1132- 33 (accepting, but not addressing the validity of, the Commission’s 40% open field premise). The court also found it unnecessary to reach the issue of whether the record supported the Commission’s conclusion that new programmers would need access to an “open field” of 40% of U. S. subscribers. Id. 90 Id. at 1134 (emphasis in original). 91 47 C. F. R. § 76.504. In calculating a system’s capacity, “activated channels” includes all commercial and non-commercial broadcast, public, educational, governmental, and leased access channels carried. See 1993 Second Report and Order, 8 FCC Rcd at 8588- 89 ¶ 54. The Commission has also defined the term “activated channel” in the digital context. See Carriage of Digital Television Broadcast Signals, Amendments to Part 76 of the Commission Rules, Implementation of the Satellite Home Viewer Improvement Act of 1999, Local Broadcast Signal Carriage Issues, Application of Network Non- Duplication, Syndicated Exclusivity and Sports Blackout Rules to Satellite Retransmission of Broadcast Signals, 16 FCC Rcd 2598, 2614- 16 ¶¶ 39- 41 (2001) (2001 Digital Must Carry Order), Second Report and Order and First Order on Reconsideration, 20 FCC Rcd 4516 (2005). 92 See 1993 Second Report and Order, 8 FCC Rcd at 8592 ¶ 64. 93 Id. at 8584- 85 ¶ 44. 94 Id. at 8593- 95 ¶ 68. 95 Id. at 8592- 96 ¶¶ 65- 71. 96 Id. 97 47 C. F. R. § 76.504( b). The 75 channel threshold thus reserves at most 45 channels for unaffiliated programming services (75 x .60 = 45). 17 Federal Communications Commission FCC 05- 96 18 40%. 98 Moreover, because future expansion of channel capacity through the use of advanced technologies or the presence of effective competition might reduce the need for the limit or render it unnecessary, the Commission stated that it would revisit the restriction at a later date. 99 In this regard, the Commission observed that “Congress has . . . indicated that a primary objective of the Act was to ‘rely on the marketplace, to the maximum extent feasible, to promote the availability to the public of a diversity of views and information’ and that the legislation was intended to protect consumer interests in the receipt of cable service ‘where cable television systems are not subject to effective competition. ’ Thus . . . further analysis as to whether the restrictions might be phased out where effective competition develops will be appropriate.” 100 In the 1995 Vertical Reconsideration Order, the Commission reaffirmed the vertical limit, barring cable operators with 75 or fewer channels from devoting more than 40% of channel capacity to affiliated programming. 101 It again found that the 40% limit “is appropriate to balance the goals of increasing diversity and reducing the incentive and ability of vertically integrated cable operators to favor their affiliated programming, with the benefits and efficiencies associated with vertical integration.” 102 The Commission found that until cable operators deployed emerging technologies such as fiber optic cable or digital signal compression, which would greatly expand channel capacities and thus obviate the need for channel occupancy limits as a means of encouraging cable operators to carry unaffiliated programming, the 75 channel maximum continued to make sense. 103 26. The Time Warner II court concluded that the Commission had not attempted to link the 40% limit with the benefits and harms resulting from common control of both programming supply and distribution sources or with current MVPD market conditions. 104 The court dismissed the Commission’s argument that “no MSO has yet complained that the 40% vertical limit has required it to alter programming,” 105 stating this “says nothing about the plans that the rule may have scuttled.” 106 Concluding that the Commission neither justified the vertical limit with record support, nor established that the limit did not burden speech more than necessary, the court reversed and remanded the limit. The court cautioned the Commission, on remand, to consider the constraining impact of competition on cable operators’ ability to favor affiliated programming at the expense of unaffiliated programming, opining that competition “precludes cable operators from exercising the market power which originally justified channel occupancy limits.” 107 98 For example, for a 200- channel system, 45 channels must be reserved for unaffiliated programming, and 155 channels, i. e., 85%, could be occupied by operator- affiliated programming. 99 1993 Second Report and Order, 8 FCC Rcd at 8601- 02 ¶¶ 83- 84. 100 Id. at 8603- 04 ¶ 84. 101 1995 Vertical Reconsideration Order, 10 FCC Rcd at 7365 ¶ 3. 102 Id. at 7367 ¶ 8. 103 Id. 104 Time Warner II, 240 F. 3d at 1137- 38. 105 Id. at 1137. 106 Id. at 1137- 38. 107 Id. at 1138. 18 Federal Communications Commission FCC 05- 96 19 3. Elements of the Horizontal and Vertical Limits 27. We next examine the stated objectives of Section 613( f) in light of the Time Warner I and Time Warner II decisions and the comments received in response to the 2001 Further Notice on the elements of the statute. Section 613( f)( 1) of the Communications Act directs the Commission to set horizontal and vertical limits in order to “enhance effective competition.” 108 Section 613( f)( 2) sets forth seven specific criteria and public interest objectives to be taken into account in setting horizontal and vertical limits. 109 These include consideration of offsetting efficiencies gained through ownership and control and establishment of limits that reflect “the dynamic nature of the communications marketplace.” 110 Effectuating Congress’ intent under this statute therefore involves a careful weighing of statutory objectives and factors in light of an MVPD marketplace that is rapidly evolving in terms of both distribution platforms and vastly expanded programming choices. 28. Enhance Effective Competition. Prefacing the statutory directive to establish both horizontal and vertical cable ownership limits is the single phrase “[ i] n order to enhance effective competition.” 111 The 2001 Further Notice discussed changes in the MVPD marketplace and assumed that non- cable MVPDs and overbuilders should be considered “competition” for this purpose since they provide outlets for programmers and alternatives for consumers. 112 The 2001 Further Notice observed that perhaps “the most important difference between the industry in 1992 and today is that in 1992 there was no clear nationwide substitute for cable,” whereas today, “DBS has a national footprint and, although there are questions concerning DBS’ ability to constrain cable prices, it appears that DBS currently offers an effective alternative path through which program networks can reach subscribers.” 113 At the same time, the Commission recognized that this does not suggest that consumers necessarily enjoy the effects of strong competition in the MVPD market, but rather, simply that there are alternatives to cable available to consumers and programmers which were not available in 1992. 114 In addition, the 2001 Further Notice observed that in the context of Section 613( f), effective competition “seems to mean competition sufficient to provide alternative means for programmers viably to reach consumers, thus protecting consumer choice and welfare.” 115 Comment was sought on the impact of the competitive presence of DBS on cable operators’ market power generally and on their ability to select programming for reasons 108 47 U. S. C. § 533( f)( 1). 109 See 47 U. S. C. § 533( f)( 2)( A)-( G). 110 See 47 U. S. C. § 533( f)( 2)( D), (E). 111 47 U. S. C. § 533( f)( 1). 112 2001 Further Notice, 16 FCC Rcd at 17327 ¶ 23. 113 Id. at 17326- 27 ¶ 22, citing, 2000 Price Survey. See also EchoStar Communications Corporation, General Motors Corporation, Hughes Electronics Corporation (Transferors) and EchoStar Communications Corporation (Transferees), 17 FCC Rcd 20559, 20605- 09 ¶¶ 106- 16 (EchoStar- DirecTV HDO) (tentative conclusion that for purposes of initial analyses of EchoStar/ DirecTV merger application, relevant product market was MVPD market; issue of whether DBS is in fact a closer substitute for DBS than cable designated for hearing). 114 Id. at 17327 ¶ 24. 115 Id. 19 Federal Communications Commission FCC 05- 96 20 other than quality and/ or viewer interest, and on the extent to which advertisers view DBS as an effective substitute for cable in reaching viewers. 116 29. Comcast contends that cable faces competition in every market because DBS is a clear substitute for cable, noting that 60% of all DBS subscribers are in areas served by cable, and nearly half of all current DBS subscribers are former cable subscribers. 117 Writer’s Guild argues that DBS providers are not yet significant enough to compete in the market for the purchase of programming and limit the market power of cable operators, and that DBS will have even less ability to compete if the 30% horizontal ownership cap is lifted. 118 Writer’s Guild further contends that the limited reach of DBS and other competitive MVPDs restricts their ability to make a program service viable, and that a new network still cannot be viable without cable carriage. 119 Similarly, CFA argues that satellite remains primarily a niche market player serving either rural communities in which cable is inferior or unavailable or serving high- volume specialty programming markets. 120 Further, CFA cites a Consumers Union Survey that appears to indicate that DBS and cable are viewed differently by consumers and function more as “complements” rather than “substitutes” in the market. 121 CFA also argues that “enhance” requires the Commission to do more than merely protect competition. 122 30. The record compiled in response to the 2001 Further Notice is now four years old. Total DBS subscribership has increased during this time from about 16.1 million households to approximately 23.2 million households, a factor that must be taken into account in fashioning rules intended to enhance effective competition. 123 In addition, News Corp. ’s acquisition of DirecTV created, for the first time, a DBS operator that is vertically integrated with programming networks. We therefore seek updated evidence and analysis of the role of DBS competition in providing alternative means for programmers to viably reach consumers, thus protecting consumer choice and welfare, and comment on the weight to be given such competition in establishing the ownership limits. We seek comment on how a vertical limit can enhance effective competition if programming rejected by an incumbent cable operator can be carried on an alternative MVPD, or via other means of electronic delivery to the consumer. Additionally, the Commission must consider the extent to which horizontal and vertical limits are intended to promote 116 Id. ¶ 23. 117 Comcast Reply Comments at 11- 13. 118 Writer’s Guild Comments at 9. 119 Id. 120 CFA Comments at 151- 53. 121 Id. at 159- 63. 122 Id. at 16. CFA further argues that the antitrust law alone supports a 30% horizontal ownership limit, and that because the antitrust law is intended only to protect, not enhance, competition, the Commission cannot adopt a limit higher than 30%. Id. at 25- 29, citing U. S. v. Philadelphia National Bank, 374 U. S. 270 (1966) and FTC v. H. J. Heinz Company, 246 F. 3d 708 (DC Cir. 2001). 123 See Annual Assessment of the Status of Competition in the Market for the Delivery of Video Programming, 20 FCC Rcd 2755, 2792- 93 ¶ 54 (2005) (11 th Annual Report). See also further discussion on the significant recent growth in DBS subscribership in ¶ 52, infra. 20 Federal Communications Commission FCC 05- 96 21 competition in the programming market. We tentatively conclude that “enhance effective competition” applies to MVPD competition as well as competition in the supply of programming and seek comment on this tentative conclusion. 31. Not Unfairly Impede the Flow of Programming. Section 613( f)( 2)( A) requires that the Commission “ensure that no cable operator or group of cable operators can unfairly impede, either because of the size of any individual operator or because of joint actions by a group of operators of sufficient size, the flow of video programming from the video programmer to the consumer.” 124 The court in Time Warner II held that a broad interpretation of “unfair” “is plausible only for actions that impinge on the interest in competition that lay at the heart of Congress’s concern.” 125 How should “unfair” be defined after the Time Warner II decision? Does “unfairly” suggest that our analysis focus on “efficiency” as used in a purely economic model? Or, should our analysis rely much less on economic concepts, as suggested by AT& T when it argues that if there are at least two outlets and no collusion, a programmer’s failure to reach homes is the result of “legitimate, independent editorial choices” and cannot be deemed unfair? 126 We also seek comment on the ability and incentive of an individual cable operator, or group of cable operators, to restrict the flow of programming to the consumer. Is it possible for the “flow of video programming” to be quantified, and if so, what amount must be unfairly impeded before being constrained by an ownership limit? The BKS Study attempted to quantify the level of cable operator concentration where the flow of programming becomes restricted. We seek comment on whether experimental economics studies, other types of studies, economic theory, or experience in other industries would be useful in identifying the point at which horizontal concentration among cable operators is likely to unfairly impede the flow of programming. 32. Neither Favor Nor Unfairly Restrict Affiliated Programming. Section 613( f)( 2)( B) of the Act directs the Commission to ensure that cable operators affiliated with video programmers “do not favor such programmers in determining carriage” nor “unreasonably restrict” the flow of programming from such programmers to other MVPDs. 127 RCN proposes that the Commission adopt regulations, based on authority in Section 613( f) and its ancillary jurisdiction, that would, on a market- by- market basis, measure market power through an analysis of an entity’s ability to control access to “sought- after programming.” 128 According to RCN, if MVPD entrants cannot gain access to this programming due to the incumbent’s ability to control that programming, then a presumptive finding of market power would be made, which would compel the owner of the programming to make it available on “industry- standard terms.” RCN recommends limiting this rule to programming that is “unique and otherwise unobtainable,” thus excluding programming that a competitor could produce itself. 129 In contrast, Cablevision argues that the proliferation of unaffiliated networks, the emergence of the broadcast networks as a significant competitive force in the cable programming market, and the strength and durability of competition from 124 47 U. S. C. § 533( f)( 2)( A). 125 Time Warner II, 240 F. 3d at 1135. 126 AT& T Comments at 13- 14, citing Time Warner II, 240 F. 3d at 1135. 127 47 U. S. C. § 533( f)( 2)( B). 128 RCN Comments at 15- 18. 129 Id. at 16- 17. 21 Federal Communications Commission FCC 05- 96 22 alternative MVPDs precludes cable operators from using vertical integration to thwart competition from rival MVPDs and programmers. 130 Cablevision further argues that rival MVPDs have access to a broad range of non- vertically integrated programming, as well as the ability to enter into programming investments themselves. 131 We seek further comment on how ownership limits may further the statutory objective that cable operators not favor affiliated programmers in determining carriage nor “unreasonably restrict” the flow of programming from such programmers to other MVPDs in light of current marketplace conditions. In addition, we seek comment on the types of activity that would constitute an unreasonable restriction on the flow of programming from affiliated programmers to other MVPDs, and how ownership limits could address such activity. 33. Market Structures; Industry Relationships; Joint Ownership; Nature and Market Power of Local Franchise. Section 613( f)( 2)( C) of the Act directs the Commission to “take particular account of the market structure, ownership patterns, and other relationships of the cable television industry, including the nature and market power of the local franchise, the joint ownership of cable systems and video programmers, and the various types of non- equity controlling interests.” 132 In the 2001 Further Notice, we asked for comment on the existing market structure and ownership patterns in the cable industry. 133 We received no comments on precisely how we should interpret the terms in Section 613( f)( 2)( C). Cablevision, however, comments that the Commission should follow the congressional directive to rely on the marketplace “to the maximum extent feasible” when establishing rules under the 1992 Act. 134 We seek comment on the meaning of these statutory terms and their effect with regard to setting effective ownership limits. In particular, do “ownership patterns” and “the nature and market power of the local franchise” refer to clustering, or some other phenomenon? Does “joint ownership of cable systems and video programmers, and the various types of non- equity controlling interests” entail only issues of vertical integration and our attribution rules, or was Congress referring to something more? 34. Offsetting Benefits and Need for Rules to Reflect Dynamic Marketplace. In addition to accounting for potential harms that may occur as a result of horizontal concentration and vertical integration, Section 613( f)( 2) requires that the Commission “account for any . . . benefits that might be gained through increased ownership or control; 135 make such rules and regulations reflect the dynamic nature of the communications marketplace; 136 and not impose limitations which would bar cable operators from serving previously unserved rural areas. 137 AT& T states that there are clear public interest benefits to increased cable concentration, mostly as the result of cable operators’ economies of scale, and that the 130 Cablevision Comments at 9. 131 Id. at 9- 10, citing, for example, a transaction between EchoStar, Vivendi, and the USA Network. 132 47 U. S. C. § 533( f)( 2)( C). 133 2001 Further Notice, 16 FCC Rcd at 17334 ¶ 44. 134 Cablevision Comments at 10- 11, citing the 1992 Act, § 2( b)( 2). 135 47 U. S. C. § 533( f)( 2)( D). 136 47 U. S. C. § 533( f)( 2)( E). 137 47 U. S. C. § 533( f)( 2)( F). 22 Federal Communications Commission FCC 05- 96 23 Commission is required to take such benefits into account. 138 Citing Section 613( f)( 2)( E), that the ownership rules must reflect the dynamic nature of the communications marketplace, Progress and Freedom Foundation (PFF) explains that the cable industry’s recent evolution from strictly analog video programming to digital video and non- video offerings has made the industry much more dynamic. 139 We seek comment on the effect of these considerations of potential benefits with regard to setting ownership limits. 140 35. Rules Not to Impair Development of Diverse and High Quality Programming. Section 613( f)( 2)( G) requires the Commission to ensure that any limits imposed do not “impair the development of diverse and high quality video programming.” Time Warner I upheld the constitutionality of Section 613( f), finding that Congress reasonably concluded that dramatic concentration in the cable industry “threatened the diversity of information available to the public and could form a barrier to the entry of new cable programmers.” 141 In Time Warner II, however, the court concluded that “Congress has not given the Commission authority to impose, solely on the basis of the ‘diversity’ precept, a limit that does more than guarantee a programmer two possible outlets (each of them a market adequate for viability).” 142 Nevertheless, the court suggested that diversity, while not the primary concern of the statute, is a factor entitled to consideration. 143 To fully implement the provisions of Section 613( f), as well as abide by the court’s directives in Time Warner II, in the 2001 Further Notice, we asked for comment on the scope of our diversity goal in light of the court’s ruling. 144 36. In response to the 2001 Further Notice, AT& T argues that Congress’ primary concern in authorizing ownership limits is fair competition, while diversity is a “byproduct” of requirements that ensure there are at least two outlets for video programming. 145 Comcast argues that the availability of two conduits through which a programmer could reach the number of viewers needed for viability is the absolute limit on diversity as a justification for horizontal ownership restrictions. 146 CFA argues, 138 AT& T Comments at 69, citing AT& T Comments, Ordover Decl. at ¶129. 139 PFF Comments at 9, 12- 16. 140 We further discuss each of these goals in the context of determining horizontal and vertical ownership limits as appropriate in Sections II. C. and D., infra. 141 Time Warner I, 211 F. 3d at 1320. The court further held that “[ i] t is enough that, having determined that ‘[ c] oncentration has grown dramatically in the cable industry, ’ S. Rep. at 32, 1992 U. S. C. C. A. N. at 1165, the Congress reasonably concluded that this concentration threatened the diversity of information available to the public and could form a barrier to the entry of new cable programmers. That is hardly an unreasonable inference.” Id. 142 Time Warner II, 240 F. 3d at 1135. 143 Id. at 1136, noting “the duality of interests [competition and diversity] at work in this section.” 144 Pursuant to Section 613, our cable horizontal and vertical ownership limits were designed to promote a diversity of programming choices for consumers. 145 AT& T Comments at 11- 12, citing Time Warner II, 240 F. 3d at 1136. 146 Comcast Comments at 16. 23 Federal Communications Commission FCC 05- 96 24 however, that while the court found that the Commission cannot rely solely on diversity as a rationale for its horizontal ownership limit, diversity remains one of the two primary concerns animating the 1992 Act, and the Commission must act on a prophylactic basis to accomplish the dual goals of competition and diversity. 147 CFA also asserts that the Time Warner II decision requires the Commission to: (1) clearly articulate economic and legal rationales supporting its decision; (2) prove that its enhancement of diversity under the rule is substantial rather than de minimis; 148 and (3) recognize that Congress’ overarching purpose was to enhance effective competition, which, in turn, protects diversity. 149 RCN argues that despite the court’s holding in Time Warner II, the Commission has ample authority under Section 613, as well as other sections of the Act, to promulgate rules that promote diversity. 150 We seek comment on the role and weight diversity concerns should play in setting our ownership limits. If our limits ensure adequate competition in the MVPD marketplace such that the flow of programming is not unfairly impeded, would diversity likewise be ensured? 37. We are interested in whether the widespread availability of DBS service, along with its continued strong growth in subscribership since the 2001 Further Notice, provide an adequate outlet for programming such that diversity is ensured. 151 Writer’s Guild argues that DBS operators are not a source of programming diversity, because their limited subscribership restricts their ability to make programming services viable. 152 However, as we noted previously, since the record closed on the 2001 Further Notice, total DBS subscribership has grown from approximately 16.1 million in June 2001 to 23.2 million in June 2004, an increase of almost 45%. 153 At the same time, it appears that a number of new networks have launched successfully using business models that do not require the same subscriber reach that more established, general interest networks enjoy. 154 In assessing how a limit could promote 147 CFA Comments at 13- 16. 148 CFA cites to the Commission’s Opposition to Certiorari in Time Warner II, where we stated: “The court of appeals found the promotion of diversity to be an insufficient justification for the rule because ‘at some point . . . the marginal value of such an increment in diversity would not qualify as an important governmental interest’ . . . . That concern about de minimis enhancements in diversity, however, has no relevance here . . . . The court’s ability to imagine hypothetical situations where the incremental increase in diversity might not justify a regulation thus provides no basis for invalidating a regulation whose actual and foreseeable operation substantially enhances the Congressional goal of diversity.” Id., citing FCC Opposition to Cert. at 10- 11. 149 Id. 150 RCN argues that the following sections of the Act also apply: Sections 601; 612( a), (e)( 2)-( 3), (g); and 628( a), (c)( 1), (c)( 4)( D). RCN Comments at 9. 151 As of June 2004, DirecTV is the second largest MVPD and EchoStar (DISH Network) is the fourth. See 11 th Annual Report, 20 FCC Rcd at 2793 ¶ 55. 152 Writer’s Guild Comments at 9. 153 See n. 123, supra. 154 For example, College Sports Television, The Tennis Channel, and Reality Central have focused on seeking carriage on cable operators’ digital tiers, which generally reach a smaller segment of a cable operator’s subscribers. See discussion in Section II. B., infra. However, Oxygen Network states that carriage on a cable operator’s analog tier – providing assurance of widespread distribution – is essential to obtain the financing (continued….) 24 Federal Communications Commission FCC 05- 96 25 diversity, should we be concerned with maximizing the range of different program types, or maximizing competing sources of each type of programming, or both? 4. Approaches to Establishing Horizontal and Vertical Limits 38. Scope of Legal Authority. The Time Warner II court suggested several ways cable operators could unfairly impede the flow of programming that might form the basis of a horizontal limit. 155 The court explained that the Commission might justify a limit if it could establish that a single large cable operator acting alone would be able to act anticompetitively by “extort[ ing] equity from programmers or forc[ ing] exclusive contracts . . . while serving somewhat less than [the market share] . . . that would allow it unilaterally to lock out a new cable programmer . . . .” 156 It found, however, that the Commission failed to offer any evidence or theory of anticompetitive harm arising from the actions of a single cable operator. 157 In addition, Time Warner II acknowledged that a cable operator might be able to “exploi[ t its] monopoly position in a specific cable market to extract rents that would otherwise flow to programmers,” but questioned whether such action would give rise to an important government interest justifying a burden on speech. 158 The court further noted that we must show how cable operators’ hard bargaining with programmers to lower programming costs is unfair. 159 Finally, the court stated that the Commission would be justified in ensuring “at least two conduits through which” programmers may reach an adequate number of consumers. 160 The court found that a cable operator’s size would constitute an unfair impediment to the flow of programming if that operator were the only viable conduit for programming “independent of concerns over anticompetitive conduct.” 161 In the 2001 Further Notice, in accordance with our statutory mandate, First Amendment principles, and Time Warner II, we sought (Continued from previous page) necessary to develop and launch an independent programming service. See Oxygen Comments filed in MB Docket No. 04- 207 (A La Carte Proceeding) at 3 (Jul. 15, 2004). 155 Time Warner II, 240 F. 3d at 1133. 156 Id. We note that, in 1992, Congress instructed the Commission to adopt rules prohibiting cable operators from demanding equity in exchange for carriage. See 47 U. S. C § 536; 47 C. F. R. § 76.1301. Despite these protections, Time Warner II recognizes that “a single MSO, acting alone rather than ‘jointly, ’ might perhaps be able to do so while serving somewhat less than the 60% of the market (i. e., less than the fraction that would allow it unilaterally to lock out a new cable programmer) despite the existence of antitrust laws and specific behavioral prohibitions enacted as part of the 1992 Cable Act, see 47 U. S. C. § 536, and the risk might justify a prophylactic limit [horizontal cap] under the statute.” Time Warner II, 240 F. 3d at 1133. 157 Id. at 1132- 34. 158 Id. at 1133. The term “rents” usually refers to “all payments to inputs that are above the minimum required to make these inputs available to the industry.” Edwin Mansfield and Gary Yohe, MICROECONOMICS, 462 (10th ed., 2000). Here, large cable operators can use their size and concomitant bargaining power to claim more of the “return” or “surplus” of their deals with programmers. Continued lowered returns on programmers’ investment can create an incentive for underinvestment in programming, which can result in inefficiency. 159 Time Warner II, 240 F. 3d at 1136 n. 6. 160 Id. at 1131- 32, 1135. 161 Id. at 1134- 35. 25 Federal Communications Commission FCC 05- 96 26 comment on the state of competition in the MVPD market to ensure that our rules are reasonable and serve the public interest. 162 39. Horizontal Limits. In determining how to set a horizontal limit, we are again guided by the language in the statute and the court’s consideration and rejection of our prior limits. In ruling that the Commission had failed to meet the required evidentiary standard, the court in Time Warner II stated that we must base our limits on a “non- conjectural risk” of economic harm. 163 Discussing joint action, the court faulted the Commission’s failure to point to examples of collusion and stated that “[ s] ubstantial evidence does not require a complete factual record – we must give appropriate deference to predictive judgments that necessarily involve the expertise and experience of the agency.” 164 The court further stated that the standard requires the Commission to put forth some evidence that indicates prospects for collusion. 165 In discussing other forms of anticompetitive action, the court suggested that sound theory could provide an evidentiary basis for limits, stating “there are theories of anticompetitive behavior other than collusion that may be relevant to the horizontal limit and on which the FCC may be able to rely on remand.” 166 40. In response, cable operators generally oppose the imposition of any ownership limits. 167 Time Warner argues that because foreclosure 168 of entry by video programming services does not constitute a non- conjectural problem, a subscriber limit is neither necessary nor appropriate. 169 Other commenters, pointing to statutory provisions that require us to take account of changing market conditions, as well as the court’s instruction to consider the role of DBS in the MVPD marketplace, argue that conditions have changed so markedly since 1992 that the horizontal limits envisioned by Section 613( f) are no longer necessary. 170 162 2001 Further Notice, 16 FCC Rcd at 17320- 21 ¶ 7. 163 Time Warner II, 240 F. 3d at 1132. 164 Id. at 1133. 165 Id. 166 Id. 167 See, e. g., Time Warner Comments at 6- 19; AT& T Comments at 4- 5; Comcast Comments at 17- 18. To support their position, cable operators rely heavily on the arguments of economist Alexander Raskovich, who submitted comments containing his article on pivotal buyers. See Raskovich Comments, later revised and published as Alexander Raskovich, Pivotal Buyers and Bargaining Position, 51 J. OF INDUS. ECON. 4, 405- 26 (Dec. 2003) (Raskovich, Pivotal Buyers and Bargaining Position). Time Warner’s experts, Paul L. Joskow and Linda McLaughlin, and AT& T’s expert, Janusz Ordover, rely on Raskovich’s arguments. See Jaskow & McLauglin, “An Economic Analysis of Subscriber Limits” (attached to Time Warner Comments and hereinafter referred to as “Joskow and McLaughlin”) at 15 n. 30; AT& T Comments, Ordover Decl. at 41. 168 “Foreclosure” means that a large vertically integrated cable operator’s decision not to grant a programmer carriage could induce the programmer to exit the market or could deter the programmer from entering the market. 169 Time Warner Comments at 9- 14. 170 See, e. g., PFF Comments at 7. 26 Federal Communications Commission FCC 05- 96 27 41. In ex parte filings since the 2001 Further Notice, Comcast has proposed a burden- shifting approach to setting cable ownership limits. 171 Comcast argues that marketplace facts demonstrate that the quantity and quality of video programming available to consumers, as well as the source diversity and content diversity, has never been greater. Comcast further argues that there is no evidence of current impediments to the flow of video programming to consumers, or that the number of one cable company’s subscribers would create an impediment. In light of the record, as well as the court’s decision in Time Warner II, Comcast believes that a sustainable hard limit is unattainable. Comcast therefore proposes that the Commission adopt an approach that uses a procedural trigger rather than a hard cap. 172 42. Under such an approach, Comcast urges, all proposed mergers would be reviewed and would be subject to a public interest analysis, but only those above a specified “soft cap” would require a more detailed information submission and market analysis. 173 Proponents of mergers below the soft cap would still bear the burden of proving, by a preponderance of the evidence, that the proposed transaction serves the public interest. However, they would be able to establish a prima facie case that the merger is in the public interest by certifying that the combined entity’s size does not exceed the soft cap. The burden of proof would then shift to the opponents of the merger, who would need to show the existence of extraordinary circumstances that pose competitive concerns. 174 In essence, Comcast appears to be proposing a form of case- by- case review premised on a procedural trigger, rather than a set numerical limit to the number of subscribers an operator may reach. 43. Some commenters argue that the Commission may satisfy its statutory obligations under Section 613( f) without establishing some form of horizontal limit. 175 As noted above, Time Warner argues that because foreclosure of entry by new video programming services does not constitute a non-conjectural problem, a horizontal limit is neither necessary nor appropriate. 176 Time Warner also argues that Section 613( f) grants the Commission discretion to not impose any limit. 177 AT& T argues that the word “reasonable” in Section 613( f) must be read to permit the Commission to decline to adopt a horizontal limit, particularly where, as here, a regulation restricts speech. 178 Comcast asserts that Congress authorized horizontal limits only to the extent needed to prevent unfair impediments to the flow 171 See Letter from Michael H. Hammer, Willkie, Farr & Gallagher, on behalf of Comcast, to Marlene H. Dortch, Secretary, FCC (Mar. 13, 2003); Letter from James L. Casserly, Willkie, Farr & Gallagher, on behalf of Comcast, to Marlene H. Dortch, Secretary, FCC (Jan. 31, 2003). 172 Id. 173 Id. 174 Id. 175 See Time Warner Comments at 9; AT& T Reply Comments at 18- 19; Comcast Reply Comments at 4- 5. 176 Time Warner Comments at 9. 177 Id. 178 AT& T Reply Comments at 18- 19. 27 Federal Communications Commission FCC 05- 96 28 of programming to consumers, and that the record does not show that cable operators can impede the flow of programming to consumers, “now or in the future.” 179 44. We do not agree with commenters who argue that we have the discretion to forgo imposing a horizontal limit. Section 613( f) clearly states “the Commission shall . . . conduct a proceeding . . . to prescribe rules and regulations establishing reasonable limits on the number of cable subscribers a person is authorized to reach . . . .” 180 Further, Time Warner I held that “[ t] he ‘subscriber limits’ provision directs the Federal Communications Commission to limit the number of subscribers a cable operator may reach.” 181 We therefore tentatively conclude that the language of Section 613( f) requires us to set some limit on the number of MVPD subscribers one entity may reach. 182 While the current record contains no evidence of the exact percentage of MVPD subscribers attributed to one entity at which the flow of programming may be impeded, it does provide evidence that if an entity is unencumbered in its subscriber reach, harms are likely to occur. 183 We also tentatively conclude that Congress gave the Commission significant discretion in determining the ownership limits, both in their absolute level as well as in their form and structure. In particular, neither the statute nor the legislative history states a clear preference either for hard limits or against other types of limits. We seek comment on our tentative conclusions that the statutory language requires us to set a reasonable limit on the number of subscribers that a cable operator may reach, and on the scope of our discretion to fashion the form and structure of such limits under Section 613( f). 45. Vertical Limits. In the 2001 Further Notice, we asked for comment on how the changes in the MVPD market and in the level of vertical integration for cable MVPDs may have affected cable operators’ ability to favor affiliated over unaffiliated programming. 184 We also sought comment on how application of stringent vertical restrictions might impact economic efficiencies and affect cable operators’ investment in, and production of, diverse and high quality programming. 185 Finally, we asked 179 Comcast Reply Comments at 4- 5. 180 47 U. S. C. § 533( f)( 1)( A). In addressing Section 613( f)( 1), the legislative history states: “The FCC is given discretion in establishing the reasonable limits on horizontal and vertical integration; however, the legislation is clear that the FCC must adopt some limitations. The Committee believes that it has given the FCC enough discretion in the legislation to strike the proper balance. The Committee, therefore, will permit the FCC to establish limits that best serve the public interest. The Committee will then review this decision. Because these markets are dynamic, the FCC should revisit these limitations at appropriate times to ensure that they accurately reflect the policies of the legislation.” Senate Report at 80 (emphasis added). Compare House Report at 43 (the House bill directed the Commission to impose limits on horizontal integration, but the House Bill’s vertical integration provision, which was not enacted, required the Commission only to conduct a study “to consider the necessity and appropriateness of imposing limitations on vertical integration”). 181 Time Warner I, 211 F. 3d at 1315. 182 47 U. S. C. § 533( f)( 1). 183 Among the findings of the BKS Study was that increases in the size of the largest cable operator can lead to situations in which economic efficiency is reduced, some programmers fail to recover their costs, and smaller cable operators pay more for programming. BKS Study at 3- 5. 184 2001 Further Notice, 16 FCC Rcd at 17350 ¶ 80. 185 Id. 28 Federal Communications Commission FCC 05- 96 29 commenters to address the economic basis underlying the concern with vertical integration and market foreclosure. 186 46. In response to the 2001 Further Notice, CFA argues that although horizontal market power is the primary focus of this proceeding, vertical market power is the driving force behind the horizontal ownership cap. 187 It argues that vertical market power results in anticompetitive conduct, and that when dominant firms become integrated across markets for critical inputs, there are potential problems. CFA also argues that vertical integration can create barriers to entry. 188 However, CFA fails to offer any argument or evidence on how a channel occupancy limit can prevent the harms it alleges. 47. In its comments, Cablevision argues that given technological advancements and today’s “vigorously competitive” MVPD marketplace, no channel occupancy limit will survive constitutional scrutiny. 189 Cablevision argues that competition from DBS significantly affects a cable operator’s incentive and ability to favor affiliated programming, since the use of any program carriage criteria other than viewer preference risks driving subscribers into the arms of competitors. 190 Cablevision further argues that Section 613( f) does not require the establishment of channel occupancy limits if the Commission determines that marketplace conditions obviate the need for such rules. 191 NCTA argues that Congress’ concerns in 1992 clearly were premised on very different market conditions than those in existence today. 192 NCTA argues that limiting the number of channels that may be occupied by vertically integrated programmers is no longer necessary or useful to advance the government’s interest in ensuring that cable operators do not discriminate against unaffiliated programming. 193 Time Warner argues that discrimination on the basis of affiliation is already targeted by other rules. 194 Time Warner further argues that Section 613 only requires the Commission to “conduct a proceeding … ,” and that if, after such a proceeding, the Commission finds that no limit is justified, then “reasonable limits” are no limits at all. 195 48. The express language of Section 613( f)( 1)( B) directs the Commission to conduct a proceeding “to prescribe rules and regulations establishing reasonable limits on the number of channels on a cable system that can be occupied by a video programmer in which a cable operator has an 186 Id. at 17350- 51 ¶ 81. 187 CFA Comments at 93- 104. 188 Id. 189 Cablevision Comments at 5. 190 Id. at 7. 191 Id. at 10. 192 NCTA Comments at 22. 193 Id. at 22- 23. 194 Time Warner Comments at 36, citing 47 USC § 536( a)( 3) and 47 C. F. R. §76.1301( c). 195 Id. at 37. 29 Federal Communications Commission FCC 05- 96 30 attributable interest.” 196 We are not persuaded that the qualifier “reasonable” can reasonably be construed to mean “no” limits, a reading which would effectively grant the Commission discretion to forgo altogether – that is, forbear from – establishment of a vertical limit. 197 Consistent with our conclusions above, 198 we tentatively conclude that Section 613( f) requires us to set both cable horizontal ownership and vertical channel occupancy limits at some number. As discussed in greater detail below, we seek comment on how we can set both horizontal ownership and channel occupancy limits that will survive constitutional scrutiny in light of present circumstances. 199 B. Industry Developments 49. There have been significant changes in the MVPD industry in the last several years that bear upon the question of establishing reasonable cable horizontal and vertical ownership limits. The current MVPD market differs dramatically from that which existed when Congress enacted the subscriber and channel occupancy provisions of the 1992 Act. First, in 1992, there was minimal competition to cable; competition, particularly from DBS providers, has significantly increased since then. Second, cable horizontal concentration and regional clustering have increased since 1992. Third, since 1992, cable plant upgrades have resulted in new, advanced digital services and significantly increased channel capacity. Fourth, the number of national programming networks, and their diversity in terms of sources and content, has increased. Fifth, vertical integration between cable operators and cable programming networks has decreased in percentage terms. 200 50. Cable operators, as well as other MVPDs, have been increasing their plant capacity, and have upgraded and enhanced system capabilities. As a result, MVPDs are offering substantially more programming networks and are rolling out new, advanced services to their customers. In 1992, most cable systems had a channel capacity of between 30 and 53 analog channels. 201 Today, cable operators are choosing to provide, on average, 70 analog video channels and approximately 150 digital video channels, with enough additional bandwidth to provide high- definition television, video- on- demand, Internet access services, and both circuit- switched and IP- based voice services. 202 196 47 U. S. C. § 533( f)( 1)( B). 197 Title VI of the Communications Act contains no provision granting the Commission authority to forbear from applying its rules. Compare Section 10 of the Communications Act, added by the Telecommunications Act of 1996, 47 U. S. C. § 160 (upon appropriate findings, the Commission may apply forbearance authority to a telecommunications carrier or service in some or all markets); 47 U. S. C. § 332( c)( 1) (authorizes the Commission to specify that certain provision of Title II shall not apply to commercial mobile radio service providers). 198 See ¶ 44, supra. 199 See Section II. C., infra. 200 See Annual Assessment of the Status of Competition in the Market for the Delivery of Video Programming (10 th Annual Report), 19 FCC Rcd 1606, 1690- 91 ¶¶ 141- 42; 11 th Annual Report, 20 FCC Rcd at 2832 ¶ 145. 201 See House Report at 31. 202 See 11 th Annual Report, 20 FCC Rcd at 2772 Table 3. 30 Federal Communications Commission FCC 05- 96 31 51. In addition to, and possibly as a result of the increased plant capacity of cable operators, the number of national programming networks has increased dramatically in recent years. In 1994, 106 satellite- delivered national programming networks were in operation. By 2001, there were 294. Just two years after the 2001 Further Notice, the number of networks had increased by 45 channels, to 339, and we now report 388 national programming networks. 203 Similarly, competition among programming networks and their diversity of source and content has increased. We recognize, however, that while the total number of available programming networks has increased, individual viewers tend to concentrate their viewing among a small number of networks. 204 In 1992, there was only one non- broadcast national news network, CNN. Today, CNN competes with MSNBC, Fox News Channel, CNBC, and Bloomberg, among others, for viewers. In children’s programming, consumers can now choose from Nickelodeon, several Disney networks, Cartoon Network, and Noggin. With respect to basic service movie channels, before 1992, there was only AMC; now there are TCM, Fox Movie Channel, Sundance, Independent Film Channel and the Lifetime Movie Network. Today, there is also a great variety of more specialized niche programming, such as Food Network, Sci- Fi, Golf, HGTV, Outdoor Life, and the Speed Channel. Even in niches in which an existing network enjoys a strong brand name, new networks are entering, as National Geographic has entered to challenge Discovery. 205 52. MVPDs are the primary purchasers of multichannel video programming targeted to a national audience. Although non- incumbent MVPDs continue to increase their share of the MVPD market, cable operators serve approximately 72% of total MVPD households. 206 In addition, as of December 2003, the top ten cable operators accounted for approximately 80% of total cable subscribers. 207 Further, as stated previously, since the 2001 Further Notice, the 2002 Comcast- AT& T cable transaction resulted in Comcast having the largest share of U. S. MVPD subscribers, which is very close to our remanded 30% ownership limit. 208 The 2003 News Corporation- Hughes transaction resulted in DirecTV, already one of the top three MVPDs, becoming vertically integrated with a substantial amount of cable and broadcast programming assets. 209 Also, significant growth in the number of DBS subscribers continues. DirecTV is the second largest MVPD, with approximately 13 million subscribers 203 10 th Annual Report, 19 FCC Rcd at 1690- 91 ¶ 142; 11 th Annual Report, 20 FCC Rcd at 2832 ¶ 145. 204 For example, Nielsen Media Research estimates that the average cable household watches approximately 17 of the 100 plus channels available. See Nielsen Media Research, Television Audience 2004, Feb. 2005, at 13. 205 See 11 th Annual Report, 20 FCC Rcd at 2885- 94 Appendix C, Table C- 3. See also Annys Shin, Channeling a New Wave of Viewers, National Geographic Pursues a Market with Distinction, WASHINGTON POST, July 12, 2004, at E- 1. 206 Of 92,295,766 MVPD households, cable subscribers accounted for 66,100,000. See 11 th Annual Report, 20 FCC Rcd at 2869 Appendix B, Table B- 1. 207 Cable Developments 2004, NCTA, at 7, 28. 208 See Comcast- AT& T Order, 17 FCC Rcd at 23263 ¶ 48 (2002). As of March 22, 2005, Comcast had a total of approximately 26. 3 million attributable cable subscribers, or approximately 28. 5% of all U. S. MVPD subscribers. See Letter from Peter H. Feinberg, Comcast Cable Communications, LLC, to Marlene H. Dortch, Secretary, FCC, MM Docket No. 92- 264 at 2 (Mar. 22, 2005), citing Kagan Media Money, Mar. 2, 2005, at 7 (noting that there are approximately 92. 2 million MVPD subscribers nationwide). 209 See News- Hughes Order, 19 FCC Rcd 473 (2004). See also n. 52, supra. 31 Federal Communications Commission FCC 05- 96 32 as of June 30, 2004, an increase of 30% from the approximately 10 million subscribers as of June 2001. 210 EchoStar is the fourth largest MVPD, with approximately10 million subscribers as of June 30, 2004, an increase of 67% over its approximately 6 million subscribers as of June 2001. 211 Basic cable subscriptions since the 2001 Further Notice have actually declined. 212 And the share of national programming services that are vertically integrated with cable operators has also declined, decreasing from 35% in June 2001 to 23% as of June 2004. 213 Vertically integrated programming networks also continue to be the among the most widely available and most popular cable programming networks. In 2004, seven of the top 20 networks ranked by subscribership and three of the top 15 networks based on prime time ratings were vertically integrated; 214 similarly, in 2001, vertically integrated networks represented nine of the top 20 networks ranked by subscribership and six of the top 15 networks based on prime time ratings. 215 53. Cable operators have also been increasingly upgrading their systems and rolling out advanced digital services such as high- definition television (HDTV), video- on- demand (VOD), high-speed Internet access, and cable telephony (including voice over internet protocol (VoIP)). With digital deployment, depending on the allocation of channels between digital and analog use and the compression ratio employed, cable systems serving the vast majority of cable subscribers now are capable of offering those subscribers well over 200 channels of programming and advanced services. 216 54. With the growth of system capacity, there has been a rise in the number of cable networks that are seeking to be positioned primarily on cable operators’ digital tiers. These networks are generally focused on specialized content, such as movies or sports. 217 College Sports Television has achieved carriage on systems giving it seven million subscribers, mostly on digital sports tiers. 218 In January 2004, Crown Media launched a digital 24- hour channel, Hallmark Movie Channel for placement on digital cable systems. 219 More often than not, these channels are either packaged in wide ranging 210 See 11 th Annual Report, 20 FCC Rcd at 2792- 93 ¶¶ 54- 55. 211 Id. 212 In 2001, there were 66.9 million basic cable subscribers. In June 2004, the number had fallen to 66.1 million. See 11 th Annual Report, 20 FCC Rcd at 2869 Appendix B, Table B- 1. 213 10 th Annual Report, 19 FCC Rcd at 1690- 91 ¶ 142; 11 th Annual Report, 20 FCC Rcd at 2832 ¶ 145. 214 11 th Annual Report, 20 FCC Rcd at 2834- 35, 2901- 02 ¶¶ 150- 51, Appendix C, Tables C- 6, C- 7. 215 See Annual Assessment of the Status of Competition in the Market for the Delivery of Video Programming, (8 th Annual Report), 17 FCC Rcd 1244, 1363- 64, Appendix D, Tables D- 6, D- 7 (2002). 216 Id. For example, Comcast reports that a typical upgraded 750 MHz plant can provide 84 analog video channels, 216 digital video channels, eight HDTV channels, VOD service for 400 digital video customers at any one time, high speed data service for 400 subscribers, and telephone service for 300 customers. Id. at 1625 ¶ 25 n. 59. 217 See R. Thomas Umstead, Diginets Hit the Screen, MULTICHANNEL NEWS, Dec. 8, 2003. 218 R. Thomas Umstead, CSTV Continues to Fight for Acceptance, MULTICHANNEL NEWS, Apr. 12, 2004. 219 Crown Media Holdings, Inc., Crown Media Announces Hallmark Movie Channel (press release), Nov. 11, (continued….) 32 Federal Communications Commission FCC 05- 96 33 programming genre packages or in niche packages, such as sports or family only programming. 220 It remains to be seen how specialized digital programming tiers will become. It is possible that the more narrow digital tiers become, a new network in an already existing genre or niche may face difficulty gaining carriage if it is perceived as being duplicative of another network’s programming. We seek comment on the impact of digital tiers on carriage for independent networks. 55. Video- On- Demand/ Subscription Video- On- Demand. VOD permits subscribers to instantly access video programming content on a program by program basis. Similarly, subscription VOD (SVOD) allows a programmer to create a library of content that can be accessed at any time and as often as desired for a monthly subscription fee. 221 Several new networks as well as networks that are seeking cable carriage have announced that they will create content solely for video- on- demand placement. For example, Reality Central, a 24- hour reality programming network, is offering cable companies versions of its exclusive and original programming for broadband and VOD/ SVOD. 222 Although VOD presents programmers with a new venue through which to present their content, this business model may constrain the growth of new networks as consumers balk at accepting additional subscription fees to access new and independent programming. We seek comment on the effect that VOD/ SVOD may have on the opportunity for independent programmers to gain distribution of their programming. 56. Internet Distribution. Increasingly, programmers are directing their content over the Internet for programming distribution. In June 2004, Real Networks, Inc., a streaming media company, announced a partnership with Starz, a provider of premium movie services, to provide cable modem, digital subscriber line (DSL) and other high- speed data users a subscription- based service allowing them to download up to 100 first- run and library- based movies. 223 Movielink, MovieFlix and CinemaNow also offer the ability to access programming via high speed data access, bypassing the traditional video services offered by cable and DBS operators. 224 In addition, Comcast has begun to provide video (Continued from previous page) 2003. 220 For example, Cable One, which operates cable systems in the Northwestern and Midwestern United States, offers a “Digital Value Pak” that includes a wide range of channels such as Outdoor Channel, Golf Channel, Fuel, National Geographic Channel, Court TV, Discovery Kids, G4TechTV. Cable One also offers a “Digital Faith and Family Pak” offering family and faith- based network programming. See Cable One, Inc., at http:// www. cableone. net/ (visited May 12, 2005). 221 See Simon Bernholt and Pascal Volle, SVOD: The Optimum Business Model Remains Unclear, Mercer Management Consulting, Media Context, Oct. 2002. 222 Reality Central has signed carriage agreements with Insight, which, in addition to carriage on Insight’s cable system, includes a cable modem feed, interactive programming feed and daily VOD content. Launch Pad, CableFax, Feb. 19, 2004. See also Reality Central, Inc., Mediacom To Carry Reality Central (press release), May 26, 2004. 223 Alan Breznik, Starz, RealNetworks Start Movie SVOD Service on Web, CABLE DATACOM NEWS, July 1, 2004 (Breznik Article). The service costs $12.95 per month. In addition to a library of movies, which will rotate 25% on a monthly basis, subscribers will also have access to a streamed version of the Starz Channel programming that can be subscribed to through a cable or DBS provider. See also RealNetworks, Inc., Starz and RealNetworks Launch First Subscription Premium Movie Service for Broadband (press release), June 14, 2004. 224 Breznik Article. 33 Federal Communications Commission FCC 05- 96 34 programming to its Internet customers over its high- speed data lines. On July 21, 2004, Comcast, ABC News, and Walt Disney Internet Group announced a broadband content distribution agreement 225 in which Comcast will provide its customers with ABC News live and ABC News on demand video, 226 and will also offer an online Kids Channel with interactive games, activities, and videos from Disney. These developments would seem to indicate that programmers have alternative distribution platforms for some types of content. We seek comment on the impact, if any, of Internet- based delivery on the ability of programming producers to reach consumers. 57. Even with the introduction of these additional channels and services, however, cable operators may once again face capacity constraints for the distribution of some types of content. 227 In an effort to address potential constraints proactively, the cable industry is investigating various new technologies to allow operators to attain more capacity over their upgraded plant. Three cable providers, Comcast, Cox, and Time Warner, are reported to be pursuing a project entitled “Next Generation Network Architecture” (NGNA), which is attempting to squeeze more carriage capacity over their upgraded plant through means of various compression technologies and customer premises equipment (CPE). The NGNA project seeks to define the features of a next- generation, all- digital cable network, which could have broad implications for functionality and cost. 228 The effort involves rethinking cable’s basic technologies, including everything from encryption strategies to set- top boxes that can be dramatically upgraded via software uploads, to create more carriage capacity by completely migrating cable service from analog to digital transmission so that all services could be provided utilizing Internet Protocol. 229 In addition, instead of offering hundreds of channels at once, cable operators might offer “switched video,” treating every channel the way current systems treat VOD: sending channels only when requested by a customer via remote control. 230 58. We are interested in obtaining information on existing and planned channel capacity and usage, both analog and digital, particularly with regard to the relationship between horizontal ownership and independent cable network distribution. We seek comment on the opportunities, if any, that the increased channel capacity of cable systems provides to independent programmers seeking to launch new channels. To what extent are new programming services launching on digital tiers or VOD? We also seek information on how cable operators apportion channel capacity among cable networks they own or have an ownership interest in, and what relationship, if any, that has to the ability of independent cable networks to obtain carriage. We also request information on how channel capacity is being used, and 225 See Comcast, ABC News, and Walt Disney Internet Group Sign Landmark Broadband Distribution Deal (press release), Jul. 21, 2004, available at http:// www. cmcsk. com/ phoenix. zhtml? c= 118591& p= irol- newsArticle& ID= 594276& highlight= (visited May 12, 2005). 226 Comcast high- speed Internet customers will be able to access reports from Nightline, World News Tonight, Good Morning America, This Week, 20/ 20, and Primetime Live. Id. 227 See Communications Daily, June 21, 2004, at 4. 228 Jeff Baumgartner, NGNA: A Sneak- Peek at Cable’s Battle Plan for the Future, CED MAGAZINE (Sneak- Peek at NGNA) (May 2004), at http:// testced. cahners1. com/ ced/ 2004/ 0504/ 05a. htm (visited May 12, 2005). 229 Id. See also John H. Higgins and Ken Kerschbaumer, Cable Operators: Still No Space 85 Billion Dollars Later, BROADCASTING & CABLE, June 14, 2004 at 38 (Still No Space). 230 Id. at 38. 34 Federal Communications Commission FCC 05- 96 35 whether excess capacity on systems exists. Further, we are interested in information on plans for increasing channel capacity through projects such as NGNA, and comment on the implications of such efforts. 59. Unaffiliated Programming Networks. In an effort to fine tune our inquiry in light of industry developments, we find it useful to examine some of the factors that have been integral to the success of new programming networks that are not affiliated with any cable operator. Generally, successful independent networks that have launched in the past ten years have possessed one or more of the following: a strong, experienced executive suite; programming niches with a track record of viewer support; investor interest from existing content companies, cable distributors or venture capital firms; and a degree of flexibility in negotiating with cable operators for carriage. 231 Are any of these factors critical to obtaining carriage, or are they assumed within the industry as must- have attributes necessary to initiate distribution discussions with cable operators? What specific factors do independent networks lack that retard their ability to obtain carriage? For example, retransmission consent leverage has been used successfully by Disney/ ABC and Viacom to gain cable carriage for their new cable networks. 232 In the News- Hughes Order, we found that the transaction would increase News Corp. ’s incentive and ability to 231 Independent networks such as The Tennis Channel, College Sports Television, The Game Show Network, and Oxygen appear to have leveraged one or more of these attributes to build a distribution base. For example, The Tennis Channel was founded by former executives of Viacom, has secured distribution rights for major U. S. and international tennis matches, has received venture capital financing to support its commercial launch and the creation of original programming, and has focused primarily on obtaining carriage on cable operators’ digital tiers. The Tennis Channel launched in May 2003 with three million subscribers. By May 2005, the network was available in approximately 56 million cable homes. See Bob Keisser, Dirtbag Commentator Delivers Sales Pitch, PRESS- TELEGRAM (Long Beach, CA), May 12, 2005. See also, e. g., Jim McConville, Biondi- Led Group Lobbing Tennis Channel at Cable, HOLLYWOOD REPORTER, Aug. 29, 2001; Larry Stewart, Fledgling Tennis Channel Gains Momentum, LOS ANGELES TIMES, June 17, 2002. Similarly, College Sports Television’s founders have prior cable network experience, attracted approximately $125 million in pre- launch financing, and have secured the rights to a large number of college sporting events. It too has focused on obtaining carriage on cable operators digital tiers and credits its willingness to be placed on a digital sports tier for its early success in obtaining distribution. See R. Thomas Umstead, CSTV Continues Fight for Acceptance, MULTICHANNEL NEWS, Apr. 12, 2004. The founders have stated, however, that long- term they will need wider distribution than sports tiers to remain viable. Id. 232 In April 1999, Disney/ ABC created SoapNet, a 24- hour soap opera channel, in part in response to Sony Corporation’s SoapCity, which was created in 1997. Disney used its retransmission consent negotiations involving carriage of its owned and operated local ABC broadcast stations to secure an initial distribution on smaller cable systems covering between three and five million subscribers for its January 2000 launch date. Sony, which did not have the benefit of retransmission consent leverage, failed to secure cable carriage for SoapCity, ultimately diverting the network to an online- only business model. See, e. g., Linda Moss, Disney’s Retransmission Clout Comes to SoapNet’s Aid, MULTICHANNEL NEWS, Nov. 1, 1999; Jim McConville, SoapNet All Set; Soap City Stalled, ELECTRONIC MEDIA, Jan. 17, 2000. With respect to Viacom, it announced a multi- year distribution agreement with Comcast covering its broadcast and cable networks. Viacom’s press release describing the agreement states that “retransmission will make it possible for Comcast subscribers in CBS owned-and- operated markets to receive CBS’s industry leading lineup of high definition television programming” and that “( u) nder the terms of the affiliation agreements, Viacom’s MTV Networks (including Spike TV, Comedy Central, CMT and the MTVN Digital Suite) and the BET Networks will continue to be available on Comcast systems nationwide. Additionally, Comcast will augment its digital suite of services by launching Nicktoons and MTV Hits and increasing the distribution of MTV2, Nickelodeon GAS, VH1 Classic and VH1 Country.” See Viacom, Inc., Viacom and Comcast Sign Multi- Year Affiliation Agreement (press release), Dec. 19, 2003. 35 Federal Communications Commission FCC 05- 96 36 temporarily withhold from other MVPDs access to the signals of its television broadcast stations as a negotiating tactic, and we designed a remedy to this potential harm in our merger conditions. 233 Are independent networks being squeezed out of distribution due to available slots being assigned to larger, vertically integrated programmers? Similarly, cable operator investment has been attributed as a critical factor in the early success of most networks to gain carriage. 234 One analyst found that the major factor determining a successful cable network launch was whether the network was backed either by a major programmer or by a cable operator with large scale distribution. 235 60. We seek comment on whether there is a relationship between ownership limits, either at present levels or some alternative limit, and the ability of independent programmers to gain carriage from cable operators, and remain independent, viable entities. C. Economic Basis for Horizontal Limit 61. In this section, we discuss potential harms and benefits of horizontal concentration and proposed economic foundations for establishing a horizontal limit on cable operator size. In response to the 2001 Further Notice, commenters’ positions range from arguing that the horizontal cap should remain at 30% or be lowered, 236 to elimination of the cap, 237 to the adoption of a nationwide case- by- case approach, 238 to using a local market- by- market approach. 239 PFF does not advocate specific horizontal or vertical ownership limits. Instead, PFF urges the Commission to allow the markers the 2001 Further 233 See News- Hughes Order, 19 FCC Rcd at 568, 572- 73 ¶¶ 209, 220- 21. 234 For example, Oxygen Network received a $100 million investment from Paul Allen’s Vulcan Ventures. Paul Allen is the chairman of Charter Communications and is its largest stockholder. See Jim McConville, Allen Buys Again: Invests $100M in Oxygen Media, ELECTRONIC MEDIA, June 7, 1999. Vulcan Ventures received 7% equity ownership in Oxygen Media and a seat on the company’s board of directors. In addition, Oxygen received analog carriage on Charter Communications systems, which served approximately 2. 3 million subscribers. According to Oxygen’s website, it is currently available in approximately 54 million households. See http:// www. oxygen. com/ basics/ about/? slot= footer (visited May 12, 2005). 235 Kagan World Media, Upstart Networks: It’s All About Who You Know, Cable Program Investor, Sept 12, 2003, at 4. Kagan reviewed 116 independent, cable operator- owned, and major programmer- owned networks, finding that in 83 of 116 launches, or approximately 72%, involved networks that had investment participation by either a content programmer or a distributor. 236 CFA Comments at 25. Some cable networks state that 50 million subscribers is the approximate threshold for achieving meaningful national advertising revenues in order to ensure viability. See GSN Comments in MB Docket No. 04- 207 at 3- 4 (Jul. 15, 2004); Crown Media Comments in MB Docket No. 04- 207 at 6 (Jul. 15, 2004); Viacom Comments in MB Docket No. 04- 207 at 17 (Jul. 15, 2004). As noted in Time Warner II, in setting the 30% horizontal ownership limit, the Commission found that “the average cable network needs to reach 15 million subscribers to be economically viable.” Time Warner II, 240 F. 3d at 1131, citing 1999 Cable Ownership Order, 14 FCC Rcd at 19114- 16 ¶¶ 40- 42. 237 Time Warner Comments at 9. 238 See Letter from Michael H. Hammer, Willkie, Farr & Gallagher, on behalf of Comcast, to Marlene H. Dortch, Secretary, FCC (Mar. 13, 2003). 239 RCN Comments at 18. 36 Federal Communications Commission FCC 05- 96 37 Notice identified – the statutory mandate, the Time Warner II decision, First Amendment principles, and MVPD market conditions – to guide its action on remand. PFF argues that if the Commission heeds these guideposts, it will adopt minimally restrictive ownership limits. 240 AT& T and Time Warner generally oppose any type of horizontal limit, arguing that in today’s competitive environment, a subscriber limit is neither necessary nor appropriate. 241 Comcast likewise argues that, based on Time Warner II and the record, no sustainable ownership limit can reasonably be set. 242 However, Comcast states that if the Commission concludes that some limit is required, the Commission should adopt a burden- shifting approach. 243 62. Comments filed by competitive MVPDs (i. e., overbuilders) focus mainly on the clustering taking place in some markets by large cable operators. They argue that programmers often accede to the dominance of incumbent cable operators and refuse to sell their programming to overbuilders, who lack a critical mass of subscribers. 244 The Broadband Service Providers Association (BSPA) argues that incumbent cable operators already have the incentive and ability to use their control over sports and other regional programming to foreclose entry by overbuilders. 245 BSPA urges the Commission to address access to terrestrially- delivered programming in this proceeding pursuant to Section 613( f) or Section 628 (i. e., program access rules). 246 RCN argues that the ultimate significance of a national cap is what it means for the local distribution of programming because competition occurs at the local level. 247 RCN is not principally concerned about the total number of MVPD subscribers served by any particular cable operator, but with whether it is frozen out of a target market by anticompetitive tactics. 248 RCN therefore urges the Commission to develop a local market- by- market approach to ownership limits. 249 63. None of the comments filed in response to the 2001 Further Notice yields a sound evidentiary basis for setting horizontal or vertical limits. While many commenters presented theoretical, legal or economic arguments and anecdotal evidence, no party provided a compelling approach that supported a particular horizontal or vertical limit. In this section, we discuss and seek comment on an 240 PFF Comments at 7. 241 AT& T Comments at 4- 5; Time Warner Comments at 6- 19. 242 Comcast Comments at 17- 18. 243 See Letter from Michael H. Hammer, Willkie, Farr & Gallagher, on behalf of Comcast, to Marlene H. Dortch, Secretary, FCC (Mar. 13, 2003); Letter from James L. Casserly, Willkie, Farr & Gallagher, on behalf of Comcast, to Marlene H. Dortch, Secretary, FCC (Jan. 31, 2003). 244 CMVPDs Comments at 6- 7. 245 BSPA Comments at 4- 5. 246 Id. 247 RCN Comments at 6. 248 Id. 249 Id. at 18. 37 Federal Communications Commission FCC 05- 96 38 economic foundation for a horizontal limit. 250 We start by discussing our proposed definitions for the various markets and solicit additional information on some remaining questions. Next, we analyze the potential harms of horizontal concentration through various frameworks, discuss the comments received in response to the 2001 Further Notice, and discuss the potential benefits of horizontal concentration. Finally, we request further information on how a horizontal limit can prevent potential harms while protecting potential benefits. 1. Defining the Market 64. The first step in our analysis of whether increasing cable operator size and concentration is likely to reduce competition and impede the flow of programming is to define the markets involved. In the 2001 Further Notice, we proposed a definition of markets, in which we distinguished between three separate but interrelated markets: the production of programming; the packaging of programming in networks; and the distribution of programming to consumers. 251 While we have received comments on these proposed market definitions, we find that some key questions remain unresolved. We therefore seek comment on certain questions discussed below, and seek further analysis and evidence to help resolve the issues raised. a) Programming Market 65. In the 2001 Further Notice, we distinguished between the producers of programming and the networks that packaged this programming and distributed it to subscribers using MVPD facilities. The focus of our analysis was on the ability of networks to gain carriage on cable operators’ systems. 252 AT& T argues that we should not be concerned with networks’ ability to enter the market, but instead should focus on program producers’ ability to find outlets to distribute their programming to the public. 253 Thus, if each network is viewed as simply a conduit for distributing programming to consumers, the problem would then become whether there are sufficient conduits available to ensure that there is a competitive marketplace for programming, which will allow programming to flow unimpeded to consumers. 254 Under this theory, the ability of networks to enter the MVPD marketplace would not be important if there are sufficient conduits for programming to reach consumers. Indeed, we observe that some programs that were rejected or dropped by one network have been picked up by other networks. 255 66. If, on the other hand, networks play a significant role in developing and producing original and high quality programming, then the entry of new networks will encourage the production 250 Vertical limits are discussed below in Section II, D. 251 2001 Further Notice, 16 FCC Rcd at 17321- 28 ¶¶ 8- 26. 252 Id. at 17321- 25 ¶¶ 9- 17. 253 AT& T Comments, Ordover Decl. at 59. 254 Id. This analysis might have to be performed for each niche/ genre, to the extent that the market is segmented by niche, as discussed below. Thus a sports channel might not provide a suitable alternative conduit for news, movies, or science fiction programming. 255 See, e. g., Bill Carter, ABC Under Disney: Kingdom, Yes. Magic, No., The NEW YORK TIMES, Mar. 8, 2004. ABC rejected “Survivor” and “The Apprentice,” which found homes on CBS and NBC, respectively. 38 Federal Communications Commission FCC 05- 96 39 and distribution of new programming to consumers. In support of this theory, we note that many networks contract for programming appropriate to their genre, suggesting that these networks may play a critical role in the development and production of programming. 256 We seek comment generally on the role that networks play in the production and distribution of programming, and on the role of niche networks in the development of genre- specific programs that may target audiences that are too small and specific to make them attractive to general entertainment networks or networks serving other genres. b) Programming Distribution Market 67. We previously determined that the programming distribution market should be measured by the number of subscribers rather than the number of homes passed, and that DBS subscribers should be included in the count of total subscribers to which the limit is applied; that is, that the limit should be formulated as a percentage of all MVPD subscribers, rather than as a percentage of cable homes passed. 257 CFA argues that we should not include DBS subscribers in the calculation of total subscribers. CFA claims that DBS services appeal only to customers seeking a higher quality and higher priced product, and are not a substitute for cable services for the typical “lunch bucket” cable subscriber. 258 Some commenters have also discussed the importance of alternatives to MVPDs, such as the sale and rental of DVDs and videocassettes, to distribute programs. 259 We again seek comment on the appropriate definition of the programming distribution market. We specifically seek comment on our decision to include DBS subscribers in the formulation and application of a limit. We observe that DirecTV and EchoStar rank among the top five MVPDs today, 260 and that DBS equipment prices have fallen significantly such that DBS has become more comparable to cable service. 261 68. We seriously question, however, whether other physical conduits, such as theatrical showings in movie theaters and sales and rentals of VHS tapes and DVDs, should be included in our analysis of the distribution market. The economics literature indicates that in many cases these conduits merely represent separate exhibition windows and not alternative means of entry. 262 We tentatively 256 For example, Oxygen Network launched in February 2000 with 55 hours of original programming. Ron Grover, Does Oxygen Have Enough Money to Burn, Business Week Online, Dec. 10, 1999. CSTV has developed 100 original 30- minute shows. See College Sports Television, CSTV Goes Back to School with Launch of CSTV U. Programming Initiative Featuring “Curriculum” of 100 30- Minute Instructional and Educational Shows Aimed at Aspiring Athletes (press release), Apr. 12, 2004. 257 1999 Cable Ownership Order, 14 FCC Rcd at 19101 ¶ 5. 258 CFA Comments at 45, 151- 71. CFA refers to those consumers that only purchase the basic and expanded basic tiers of cable service as the “lunch bucket crowd” and estimates that 42 million subscribers fall in this category. CFA Comments at 159. 259 See AT& T Comments, Ordover Decl. at 56- 57. 260 See ¶ 52, supra. 261 See 10 th Annual Report, 19 FCC Rcd at 1609, 1652 ¶¶ 5, 68; 11 th Annual Report, 20 FCC Rcd at 2794- 95 ¶¶ 56- 57. See also U. S Government Accountability Office, Direct Broadcast Satellite Subscribership Has Grown Rapidly, but Varies Across Different Types of Markets, GAO- 05- 257 (Apr. 2005) (2005 GAO Report). 262 Owen & Wildman, VIDEO ECONOMICS 26- 38 (1992); Barry Litman, The Motion Picture Entertainment Industry, THE STRUCTURE OF AM. INDUS. 199- 200 (Walter Adams, ed. 8 th ed. 1990). 39 Federal Communications Commission FCC 05- 96 40 conclude these other conduits should not be considered part of the same market of programming network distribution because they are not a means for network programmers to distribute their programming. We seek comment on this tentative conclusion. c) Relevant Geographic Markets 69. In the 2001 Further Notice, we recognized that “[ t] he geographic market for certain types of niche programming may . . . be national or international in scope” and sought comment on this conclusion. 263 Some commenters allege that the market for programming is international. 264 Other commenters say we should also consider regional markets. 265 70. We continue to find it reasonable to concentrate our inquiry on the effects of cable concentration in the United States, and ask for comment on this tentative conclusion. We have concluded in the past that the programming market is at least national. 266 No commenter has presented economic data that define the contours of the programming market. Instead, commenters make the uncontroversial point that domestic programmers sell some programming to international buyers and also rely on distribution outlets other than cable or DBS. We tentatively conclude that the relevant geographic market is, for purposes of the Section 613( f) analysis, no greater than the United States. We also believe that regional markets may be relevant when considering programming, such as regional sports and news networks, that is only of interest to, or available in, a particular region. As discussed further below, 267 we seek comment on whether and how the existence of regional markets should affect our development of horizontal and vertical limits. Would a regional limit on concentration better effectuate any of the statutory purposes set forth in Section 613( f)( 2), and if so, under what circumstances, and what would be the measure? 2. Potential Harms of Horizontal Concentration a) Analytical Frameworks for Economic Analysis of Harms 71. In this section, we seek further comment on the appropriate economic framework for determining whether, and at what level, a cable operator’s size is likely to impede the flow of programming to consumers or diminish effective competition. As described above, we have not found sufficient economic evidence in the record to make such a determination. We discuss in this section the strengths and weaknesses of a number of analytical frameworks and economic theories that have been proposed. We also discuss arguments and evidence that have been put in the record concerning the viability and usefulness of these analytical frameworks. Within the context of each analytical framework, we further consider the strength of the evidence concerning whether a horizontal limit is necessary to ensure the flow of programming. 263 2001 Further Notice, 16 FCC Rcd at 18532 ¶ 9. 264 See, e. g., AT& T Comments at 20- 24; Cablevision Comments at 21. 265 See, e. g., CFA Comments at 112- 17. 266 AT& T- Comcast, 17 FCC Rcd at 23269 ¶ 43. 267 See ¶ 148, infra. 40 Federal Communications Commission FCC 05- 96 41 (1) Open Field Approach 72. In the 1999 Cable Ownership Order, the Commission adopted horizontal limits based on a theory that cable operators at certain concentration levels could effectively prevent programming networks from entering or surviving in the marketplace simply by deciding not to carry them. 268 The Commission found that a new programming network needs to access 15 to 20 million subscribers and that the typical programming network had only a 50% chance of actually reaching all available MVPD subscribers. 269 The Commission concluded that a programmer needed to have an “open field” of 40% of MVPD subscribers nationwide and that a 30% MVPD subscriber limit would assure that a 40% open field remained even if the two largest cable operators decided not to carry it. 270 The Commission determined that calculations of the horizontal limit should include all MVPD subscribers, including non-cable MVPD subscribers, to take into account the increased market share of non- cable MVPDs. 271 73. The Time Warner II court rejected certain aspects of this approach, finding that the Commission lacked any evidence that cable operators would collude and that it could not simply assume that cable operators would coordinate their behavior. 272 Further, the court held that Section 613( f)( 1) does not authorize the agency to regulate the “legitimate, independent editorial choices of multiple MSOs.” 273 Thus, the court found that the record supported only a 60% limit under the Commission’s 40% open field premise. 274 However, the court did not reach the question of whether the 40% open field assumption was reasonable. In discussing the open field approach, the court admonished the Commission that market share does not necessarily equate with market power. 275 The court stated that on remand the Commission should take into account relevant measures of market power, and elasticities of supply and demand vis- à- vis other MVPD offerings, mainly DBS. 276 The court was specifically referring to the effects of retail competition from DBS and other MVPDs and consumer demand for programming on a cable operator’s incentive to foreclose an unaffiliated programming rival from carriage in an effort to favor a competing programming network that is affiliated with the cable operator. 277 268 1999 Cable Ownership Order, 14 FCC Rcd at 19116 ¶ 43. 269 Id. at 19114- 18 ¶¶ 40- 50. 270 Id. at 19118- 21 ¶¶ 51- 57. 271 Id. at 19121 ¶ 57. 272 Time Warner II, 240 F. 3d at 1133- 34. 273 Id. at 1130- 35. 274 Id. at 1132- 33 (accepting, but not addressing the validity of, the Commission’s 40% open field premise). 275 Id. at 1134. 276 Id. 277 Id. 41 Federal Communications Commission FCC 05- 96 42 74. Issues in Utilizing an Open Field Approach. In the 2001 Further Notice, the Commission asked for comment on the open field approach. 278 Pursuant to the court’s directive in Time Warner II, the 2001 Further Notice sought comment on what horizontal limit would be necessary to ensure that programmers have access to at least two viable outlets, “independent of concerns over anticompetitive conduct.” 279 In response, several commenters claim that an open field approach cannot justify a horizontal limit. 280 For example, commenters point out that many successful programming networks reach fewer than 15 million providers. 281 AT& T lists nearly 50 such networks, and points out that some successful networks are more than five years old and have fewer than three million subscribers. 282 AT& T argues that because advertising supports programming, networks can be viable even if they reach fewer than 15 million MVPD subscribers. 283 More generally, AT& T argues that the open field approach assumes that all services need the same size open field to achieve viability, when in reality the open field requirement is highly individualized and depends on the unique characteristics of each programming package. 284 Commenters also dispute the methods the Commission used to move from the 20% of the industry necessary for network survival to the 30% limit, such as the 50% success rate assumption, 285 and theories of collusion. 286 75. An examination of the subscriber numbers AT& T cites indicates that many of the programming networks it discusses have subsequently achieved substantial subscriber growth. For instance, AT& T lists Oxygen with 14.7 million subscribers in 2001, but by February 2004, Oxygen had grown to 49 million subscribers. 287 Similarly, AT& T states that the National Geographic Channel had 14.1 million subscribers in 2001, whereas by January 2004, the channel had 47 million subscribers. 288 AT& T also lists the Style Channel in 2001 with 11 million subscribers, but by January 2004 it had 34 million subscribers. 289 These statistics may undermine AT& T’s point that networks can survive without 278 2001 Further Notice, 16 FCC Rcd at 17338- 41 ¶¶ 52- 59. 279 Id. at 17339- 41 ¶¶ 55- 58, citing Time Warner II, 240 F. 3d at 1134- 35. 280 See AT& T Comments at 61- 66; AT& T Comments, Besen Decl. ¶¶ 3, 11, 14; AT& T Comments, Ordover Decl. at ¶ 145; Time Warner Comments at 19- 28; Time Warner Reply Comments at 14- 18. 281 AT& T Comments at 60- 65; AT& T Comments, Besen Decl. at 3- 6; Time Warner Comments at 24- 26; Time Warner Reply Comments at 17. 282 AT& T Comments at 61- 65. 283 AT& T Comments, Besen Decl. ¶3. 284 AT& T Comments at 62- 5; AT& T Comments, Besen Decl. ¶¶ 3, 11, 14. 285 AT& T Comments at 65- 66; Time Warner Comments at 27- 28; Time Warner Reply Comments at 18. 286 AT& T Comments at 66- 68; NCTA Comments at 18- 20; Time Warner Comments at 20- 23; Time Warner Reply Comments at 15- 16. 287 AT& T Comments at 60- 61; NCTA Cable Developments 2004 at 143. 288 AT& T Comments at 60- 61; NCTA Cable Developments 2004 at 134. 289 AT& T Comments at 60- 61; NCTA Cable Developments 2004 at 172. 42 Federal Communications Commission FCC 05- 96 43 more than 15 million subscribers, and they also point to a larger factor that must play a part in our analysis. Programmers may need not just a certain number of subscribers at any point in time, but must also maintain continued growth after that time to have a probability of survival. AT& T’s numbers were accurate at the time submitted, but those subscriber numbers have increased over the intervening years. Presumably, the possibility of this kind of growth was a factor in the programmers’ decisions to enter or remain in the market. 76. It does appear that some programming networks can survive with access to few subscribers, perhaps because they have unusually high advertising revenues, obtain high affiliate fees from MVPDs, or have lower- cost programming. 290 Similarly, we expect that there are other programming networks that require access to higher levels of subscribers. 291 The statute does not refer to particular types of programming networks, but rather to programming generally. The simple fact that some networks may be able to survive with fewer subscribers than others does not invalidate the use of averaged data to fashion a limit; rather, it suggests that if we use averaged data, we must recognize that it may underestimate the viability requirements of high- cost networks. Clearly different types of networks need access to different numbers of subscribers. We seek comment on whether we should focus our analysis on the minimum number of subscribers needed by an average network, or instead examine separately the requirements of networks with high- cost and with low- cost programming. 77. AT& T also lists a large number of spin- off networks, such as CNN International, and a variety of Discovery networks, as surviving with very few subscribers. 292 Presumably, programmers with the financial backing of large corporations, and lower costs from economies of scope, can survive longer with fewer subscribers than networks without such backing. 293 The flow of programming from sources other than large corporations could be impeded unfairly even while programmers with the backing of large corporations could survive. We believe that preserving access only for programmers with this kind of financial backing would not serve the goals of Section 613( f). 78. Time Warner argues that there was plentiful entry in the period 1992- 2001, including entry by independent networks. 294 It points to a number of factors, including the increase in cable channel capacity and the rise of DBS competition. It also presents evidence that large cable operators 290 For example, regional sports networks such as those in the Comcast SportsNet and Fox Sports Net networks typically have fewer than five million subscribers. NCTA Cable Developments 2003 at 178 and 180- 189. GoodLife TV Network, a general entertainment network which began service in 1985, survives today with 10 million subscribers. NCTA Cable Developments 2004 at 102. 291 See, e. g., the Survival Analysis, which found that a national, non- premium network growing at the average rate requires over 19 million subscribers at the end of five years to have a 70% probability of survival over its first five years, and over 41. 5 million subscribers to have a 70% probability of survival over its first ten years. Id. at 29. 292 AT& T Comments at 60- 61. 293 AT& T also lists TV Japan as surviving with very few subscribers. AT& T Comments at 60- 61. TV Japan, however, is owned by NHK, Japan’s largest broadcaster. 294 Time Warner Comments, Joskow and McLaughlin Decl. at 2- 7. 43 Federal Communications Commission FCC 05- 96 44 have tended to carry more programming over time, both affiliated and unaffiliated, indicating that the increase in size of operators will help, rather than hinder, entry by new networks. 295 79. CFA, on the other hand, argues that the Commission’s 30% limit was too high, and it would instead support a lower limit in the range of 20% to 25%. 296 CFA contends that anticompetitive behavior can occur at levels short of complete foreclosure and in markets with more than two dominant firms. CFA claims that a far greater open field therefore may be necessary for competitive entry by a new programmer, as much as 30 to 40 million subscribers instead of the 15 million figure previously relied on by the Commission, resulting in a limit in the range of 15% to 33%. 297 CFA adds that increased programming costs further underscore the need for an open field of 20 to 25 million subscribers. 298 Writer’s Guild argues that DBS operators are not a source of programming diversity, because the limited reach of DBS and other competitive MVPDs restricts their ability to a make programming service viable. 299 80. We seek additional comment on whether we should continue to use an open field approach, and whether this best meets Congress’s goal of ensuring the flow of programming to consumers. Commenters should focus on a programmer’s ability to survive in the marketplace without carriage by the largest operator. In other words, we seek to ensure that a programmer denied carriage by the largest operator could nevertheless survive in the marketplace if it gained carriage on all remaining MVPDs. In effect, the programmer viability analysis seeks to identify the subscriber reach necessary for a cable operator to become a pivotal buyer, such that a programming network must gain access to at least some of this operator’s subscribers to enter or survive in the market. Commenters advocating the use of an open field approach should also address how we should determine the size of the open field, recognizing that different types of networks may require different subscriber reaches to be viable, depending on the cost of the programming, the target audience, and projected advertising revenue. 81. Calculating a Limit under the Open Field Approach. Developing a defensible horizontal limit under the open field approach requires an analysis of the number of subscribers a programmer requires in order to remain viable. Because the court did not specifically address the 40% open field assumption, 300 and because of the passage of time since the Commission first developed the estimates on which it rests, the Commission attempted on remand to gather fresh data on programmer viability. Selected programmers were asked to provide data that would allow us to correlate programmer characteristics with profitability and thereby determine what subscriber reach is necessary for survival in the market. 301 The survey asked cable programming networks to report five years of subscribership 295 Id. at 5- 8. 296 CFA Comments at 195- 201. 297 Id. at 196- 98. 298 Id. at 198- 200. 299 Writer’s Guild Comments at 9. 300 Time Warner II, 240 F. 3d at 1132. 301 A copy of the letter sent to programmers can be found in the record of this proceeding. See Letter from W. Kenneth Ferree, Chief, Cable Services Bureau, FCC, to Programming Network Owners (Feb. 15, 2002). 44 Federal Communications Commission FCC 05- 96 45 information, the year they became profitable, their subscribership at time of profitability, whether they are vertically integrated, and their geographic reach. Unfortunately, the vast majority of responses were incomplete. Most responses came back with little more than address and subscribership information. By itself, this information is not sufficient to allow us to draw any conclusions about viability. 82. Although the primary purpose of the cable horizontal ownership rules is to ensure that the flow of video programming to consumers not be unfairly impeded by cable operators, the record in this proceeding generated almost no comments from independent cable programming networks on the number of subscribers required to remain viable. 302 A more recent inquiry generated a substantial volume of comments opposing mandatory a la carte and “themed tier” service offerings, in which independent cable programming networks estimated the number of subscribers they require to remain viable. On May 15, 2004, the Media Bureau issued a Public Notice (A La Carte PN) on factual questions regarding the provision of a la carte and “themed- tier” services on cable television and direct broadcast satellite systems. 303 The Bureau sought this information in response to specific requests from members of Congress for a report on these issues. 304 On November 19, 2004, the Media Bureau released a report (A La Carte Report) on the efficacy of a la carte pricing in the pay- television industry. 305 Several video programmers responded to the A La Carte PN alleging adverse impacts of mandated a la carte or themed tier offerings. 306 In support of their positions, the programmers identified certain subscriber targets they claimed were necessary to ensure network survival, which they generally claimed would not be possible to accomplish under an a la carte or themed tier regime. Programmer comments on network survival reflect the following: (1) the Commission’s suggestion that programming services may survive with a subscriber base of 15 to 20 million subscribers is “long out of date;” 307 (2) on average, niche networks’ revenues are split roughly 50- 50 between advertising and license fees, both of which are essential to the survival of a niche network because neither is sufficient standing alone, with both tied directly to the network’s distribution level – the total number of subscribers who can view the network; 308 (3) a typical new network does not launch until it can gain commitments from MVPDs for 302 The America Channel reports that investors desire to reach 50 million subscribers within 5 to 7 years of the launch of a cable programming network. America Channel ex parte Comments (Dec. 9, 2004, and Jan. 3, 2005). 303 See Comment Requested on A La Carte and Themed Tier Programming and Pricing Options for Programming Distribution on Cable Television and Direct Broadcast Satellite Systems, 19 FCC Rcd 9291 (2004). The Bureau also conducted a symposium to explore the advantages and disadvantages of an a la carte marketing scheme, including possible effects on retail prices. 304 See May 18, 2004 Letter from Congressmen Barton, Dingell, Upton, Markey and Deal of the House Committee on Energy and Commerce to FCC Chairman Powell, and May 19, 2004 Letter from Senator John McCain, Chairman, Senate Committee on Commerce, Science and Transportation to FCC Chairman Powell. 305 See Report on the Packaging and Sale of Video Programming Services to the Public (A La Carte Report) (MB Nov. 18, 2004). 306 See, e. g., Comments filed in MB Docket No. 04- 207 (A La Carte Proceeding), Oxygen Comments at 2- 8; A& E Comments at 15- 25; Crown Media Comments at 7- 12; TV One Comments at 1- 3, Decl. of Larry D. Gerbrandt at 4- 11. 307 See Crown Media Comments at 6; GSN Comments at 3- 4. 308 A& E Comments at 19. 45 Federal Communications Commission FCC 05- 96 46 distribution to at least 10 million homes within the first two years, and a typical start- up business plan is to reach a minimum of 30 million households within the first three to five years of launch in order to attract sufficient advertising fees to make up for the fact that during the early years a new network receives minimal (if any) affiliate license fees; 309 (4) because new networks receive minimal or no affiliate fees, the primary source of revenue for a start- up is advertising revenue, and advertising revenue only becomes viable once a network has 20- 25 million subscribers, but even at these subscriber levels, it is impossible to sell meaningful national advertising; 310 (5) although Nielsen will rate a network with 20 to 25 million subscribers, ratings data at these levels are unstable and of little value until the network reaches current survivability targets, somewhere between 40 to 60 million subscribers; 311 and (6) because advertisers are primarily interested in subscriber growth, even at the 50 million or more subscriber level, a network must be able to demonstrate that its distribution is growing, or risk advertiser abandonment. 312 We find this data relevant to our analysis of reasonable horizontal ownership limits and intend to incorporate by reference the data filed in MB Docket No. 04- 207 into the record of this proceeding. We seek comment generally on how this data should be applied and specifically on the impact of these changing subscriber targets on the calculation of the number of subscribers a programmer requires for launching and remaining viable in today’s marketplace. 83. Regardless of the horizontal ownership limit we adopt, reliable MVPD subscribership data is needed to determine whether a violation occurs. The 1999 Cable Ownership Order endorsed the use of published, current and widely- cited industry data to establish the number of MVPD subscribers nationwide, for purposes of determining a cable operator’s share of the market. 313 Subscriber data are currently available from a variety of published sources; however, at times, these publicly available 309 See TV One Comments, Decl. of Larry D. Gerbrandt at 4. 310 Id. at 6; GSN Comments at 3- 4; Oxygen Comments at 4. 311 A& E Comments at 19 (a national niche network needs to achieve a threshold level of at least 30 to 40 million subscribers); Bloomberg Comments at 5 (once the service reaches 40 million subscribers, it will be able to generate higher affiliate and advertising fees to sustain the service over the long- term); Oxygen Comments at 4 (ratings data useful only once network reaches 45 to 50 million subscribers); GSN Comments at 3- 4 (50 million is the approximate threshold for achieving meaningful national advertising revenues); TV One Comments, Decl. of Larry D. Gerbrandt at 4 (combination of advertising and affiliate fees exceeds operating, marketing and programming expenses when network reaches 40 million or more households); Viacom Comments at 19 (50 million households are necessary in order to reach a meaningful number of viewers); Crown Media Comments at 6 (more realistic plateau for meaningful advertising revenues is now approaching 50 to 60 million subscribers; “with nearly 26 million full and part- time subscribers, the performance of the Hallmark Channel’s predecessor was stagnant and its financial prospects were dim”). 312 GSN Comments at 4. See also Morgan Stanley Equity Research, News Corp., Highlights from the Media Conference, September 9, 2004 at 1 (“ Currently, the peak potential distribution for smaller cable networks is estimated at 40- 50 million subscribers, but this number is likely to increase over time, creating further opportunities for growth at Fox’s youngest networks.”). 313 See 1999 Cable Ownership Order, 14 FCC Rcd at 19112 at ¶ 35. In the 1999 Cable Ownership Order, we recognized that “not all of the data used by the industry is identical and that some degree of estimation and double counting may be involved.” Id. At that time, the Commission found it unnecessary for the Commission or firms subject to the ownership limit to refine generally accepted industry estimates because the rule was based on estimates. Id. 46 Federal Communications Commission FCC 05- 96 47 sources are inconsistent with one another. In addition, firm- specific subscriber figures submitted to the Commission may differ from figures reported in other contexts (e. g., in SEC filings or investment reports). We seek comment on whether we should take steps to address the reliability of any subscriber data we may use in applying the horizontal limit, and whether the Commission should adopt its own data collection procedures to obtain industry- wide subscriber data. 84. As noted above, the Media Bureau recently released Media Bureau Staff Research Paper No. 2004- 1 (Survival Analysis), which focuses on the actual failure and success rates of networks and the relationship of those rates to subscriber reach. 314 We seek comment on the value of this method in developing a horizontal limit under the open field approach. We also seek comment on the method as applied in the Survival Analysis and alternatives and refinements to the methods employed. (2) Monopsony Framework 85. Time Warner II faulted the Commission’s open field analysis for failing to identify a non- conjectural harm, and for providing no analysis of whether cable operators have the ability to exercise market power. 315 It stated that the statute requires the Commission “to assess the determinants of market power in the cable industry and to draw a connection between market power and the limit set.” 316 In the 2001 Further Notice we sought comment on the harms that might result from high concentration, including those that result from anti- competitive behavior. 317 We asked at what level of concentration a large cable operator gains sufficient market power to be able to refuse carriage of programming for reasons other than consumer demand. 318 We further asked questions that would help answer whether cable operators have market power, and whether they have an incentive to exercise it. Our questions concerning the concentration of the market, and whether a 40% open field was sufficient for entry by the typical programming network, relate to two key determinants of market power, which are the number of competitors in the market and the ability of firms to enter the market. 319 In this and the next sections we discuss the comments we received, and the further questions that we find need to be answered, concerning several theories of harm based on market power, and how a horizontal limit can eliminate those harms. 314 See ¶ 16 supra. 315 “Having failed to identify a non- conjectural harm, the Commission could not possibly have addressed the connection between the harm and market power. But the assessment of a real risk of anti- competitive behavior- - collusive or not-- is itself dependent on an understanding of market power … .” Time Warner II, 240 F. 3d at 1133- 34. 316 Id. 317 2001 Further Notice, 16 FCC Rcd at 17340 ¶ 57. 318 Id. at 17328, 17340- 41 ¶¶ 28, 58. 319 Id. at 17340- 41 ¶ 58. 47 Federal Communications Commission FCC 05- 96 48 86. In response to the 2001 Further Notice, some commenters argue that the market for programming does not meet the key conditions necessary for the monopsony 320 model to be applicable. For example, they argue that the supply curve is not upward sloping, and that buyers cannot force the price down by reducing their purchases, because it is costless to supply programming to one more subscriber, if the service is already being provided to other subscribers. 321 Other commenters argue that the talent used to provide programming have plenty of other employment opportunities, such as theatrical motion pictures and broadcast network programming, and therefore the supply of such services is likely to be flat. 322 87. To support using the theory of monopsony to demonstrate how a large purchaser of programming could cause harm to the market, CFA points to numerous, widely accepted economic theories that state a monopsonist would have the power to decrease programmers’ output and the prices they receive. 323 It claims that these theories apply to cable operators’ relationship to programmers. 324 We seek comment on the appropriateness of applying standard monopsony arguments to our analysis of the specific nature of the programming market. 88. Commenters proposing monopsony as an analytical framework should address how monopsony power can be measured. Can Pigou’s Index (called by Pigou the “rate of exploitation”), which is the monopsony version of Lerner’s Index, be used to measure monopsony power here? 325 Are there other measures that indicate that monopsony power is being exercised? For example, is the failure of some networks to gain carriage an indication that monopsony power is being exercised, or is it due to the low quality of those networks? Are launch fees 326 a means of extracting monopsony rents, or are they serving a filtering function to help weed out low- quality networks? 327 Some cable operators have 320 A firm is called a “monopsony” if it is the only buyer in a market, and a firm that is the only seller in a market is called a “monopoly.” Dennis W. Carlton and Jeffrey Perloff, MODERN INDUSTRIAL ORGANIZATION 87, 105 (3rd Ed., 2000) (Carlton and Perloff). 321 Time Warner Comments, Joskow and McLaughlin Decl. at 8- 9; AT& T Comments, Ordover Decl. at 12- 13. 322 Time Warner Comments, Joskow and McLaughlin Decl. at 9. 323 CFA Comments at 91- 92. 324 Id. at 26, 28 (citing FTC v. H. J. Heinz Co. ¸ 246 F. 3d 708, 715 (D. C. Cir. 2001); United States v. Philadelphia National Bank, 374 U. S. 270 (1966)). 325 Pigou’s Index is the measure of the difference between the input price and the marginal revenue product. Pigou’s Index = E = (MRP- w)/ w = 1/ e s , where MRP = marginal revenue product = revenue produced by purchasing one more unit of the input, w = the cost of the input, and e s = the price elasticity of supply. William Boal and Michael Ransom, Monopsony in the Labor Market, 35 J. OF ECON. LITERATURE 1, 86- 112 (Mar. 1997). 326 In the case of a new programming network, an MVPD may demand that the programmer pay it for the right to access its subscribers (a practice sometimes referred to as a “launch fee”). 327 Launch fees in the cable industry have certain similarities to “slotting allowances” in the grocery industry. Slotting allowances are payments by manufacturers to grocers for stocking new products. The Staff of the Federal Trade Commission has identified a number of possible benefits and harms of the practice, including its use as a signal of quality and as a by- product of market power at the retail level. Federal Trade Commission Staff, Report (continued….) 48 Federal Communications Commission FCC 05- 96 49 sought equity from programming networks as part of their carriage negotiation stance. 328 Is this an exercise of monopsony power, or is it a more efficient mechanism of risk- and profit- sharing than a simple fixed price for carriage? Does the alleged need for a new network to have bargaining leverage, usually in the form of an affiliated broadcast station or popular network, 329 indicate that monopsony power is being exercised against independent networks lacking popular affiliates, or is this bargaining just an efficient means of achieving the transaction at lower cost? 330 89. The most significant challenge to the use of the monopsony model would appear to come from the need for prices for individual transactions to be publicly known and to vary with the market- clearing price. The market for programming appears to be characterized by private bilateral negotiations that yield complex prices that are not made public. If this is the case, is there a market price that could be affected by the monopsonist’s purchasing decisions? If the existence of private negotiations with nonpublic terms of agreement implies that there is no market price, then we ask whether a bilateral bargaining model would be more useful for analyzing the programming market than the monopsony model. We discuss the use of bilateral bargaining models in the next section. (3) Bargaining Power as a Source of Unilateral Anticompetitive Action 90. Bargaining theory is an alternative framework to the theory of monopsony for analyzing how a large purchaser of programming services could exercise market power and cause harm to the market. In the 2001 Further Notice we suggested that at much higher levels of concentration cable operators could use their bargaining power to force down the prices they pay for programming, which could harm the flow of programming. 331 We explore here bargaining power as a source of unilateral anticompetitive action. Bargaining theory may better describe and model the private negotiations and non- public terms of agreements typically employed in the purchase of programming by cable operators, as compared to the theory of monopsony. Bargaining theory is often used to model bilateral (Continued from previous page) on the Federal Trade Commission Workshop on Slotting Allowances and Other Marketing Practices in the Grocery Industry, February 2001. 328 See The Bridge July 2004 (“. . . many big providers will want a piece of the new channel company”); The Cable Center, Peter Barton: An Oral History (“ The theory was, why not own - because we had the leverage to own - why not insist on owning a piece of programming service.”) available at http:// www. cablecenter. org/ library/ collections/ oral_ histories/ subjects. cfm. 329 Some of these may involve retransmission consent agreements of local broadcast stations, while others may involve the carriage of new networks’ negotiations for carriage of a particularly popular network. 330 The tying of carriage of a new affiliated network to an agreement for carriage of a popular network or station can be mutually beneficial to the cable operator and network, since the cable operator may provide something that is lower cost (the channel capacity), compared to paying the full cash value of the popular network, while the network gains carriage of a potentially- profitable affiliate. In other words, it may be of lower cost to each side than would strictly cash deals of the cable operator paying a popular network for carriage, or for a program producer paying to get carriage for its new, unproven network. 331 2001 Further Notice, 16 FCC Rcd at 17327- 28, 17333- 34 ¶¶ 26, 43. 49 Federal Communications Commission FCC 05- 96 50 negotiations, 332 and is usually better able to handle complex market structures, and to take into account transaction- specific factors. 91. Several specific institutional features shape the economics of negotiations between programming networks and MVPDs. For example, prior to entering negotiations, MVPDs and programming networks make substantial investments. For MVPDs, these investments include the construction of their video distribution systems. These costs, however, are systemic rather than specific to the contractual relationship with any single programming network. For programming networks the investments include the acquisition or production of programs. Importantly, the programming network must continue to incur these costs on an ongoing basis to develop new content. Moreover, a significant portion of these investments are considered specialized, in that a substantial portion of the investments would be lost if the programming network failed to obtain carriage on cable systems or other MVPDs. 92. After making these investments, programmers and MVPDs enter into contractual negotiations. During this process, programming networks attempt to maximize their affiliate fees and advertising revenue. In contrast with programmers, which compete fiercely for carriage, MVPDs generally do not compete with each other to acquire programming. 333 On the other hand, MVPDs likely attempt to maximize their subscription and advertising revenue while attempting to minimize their affiliate payments. In economic terms, each MVPD and each programmer are engaged in a bilateral bargaining problem. 93. The bargaining and contract theory literature has established that when at least one side of the negotiation has sufficient bargaining power, inefficiencies can arise. 334 One source of inefficiency is directly related to the relative bargaining power of the parties. Many programming networks compete among themselves for the right to sell programming to an MVPD and thereby acquire access to its subscribers. Thus, the cost the MVPD incurs from not reaching an agreement with any particular programming network is low because of the willingness of competing programming networks to sell to it. However, the cost a programming network incurs from not reaching an agreement with a large cable operator may be high if access to that operator’s subscribers is needed for it to remain viable and earn a profit. 335 According to this reasoning, because of the existence of one or more close substitutes, some programming networks may have very little bargaining power in negotiations with MVPDs. However, other programming networks may have few close substitutes and, if popular, may have substantial 332 Drew Fudenberg and Jean Tirole, GAME THEORY, ch. 10 (1993). 333 Absent an exclusivity provision, one MVPD’s acquisition of program carriage rights does not diminish the supply of programming available to other MVPDs. 334 We use the term economic efficiency in its generally accepted sense – the maximization of society’s scarce resources. However, in the discussion of the BKS Study, the term efficiency is used in a narrower sense of “trading efficiency,” which is one of the three sources of (in) efficiency discussed in this section. 335 For example, Comcast is generally viewed as an important distributor for video programmers. See George Anders, Want to Start a TV Channel?; Amy Banse, WALL ST. J., Jan. 19, 2004 (“ If you’ve got Comcast behind you, you’re practically guaranteed of being a success.”); Andrew Grossman, NBA TV scores Comcast Deal, HOLLYWOOD REPORTER, Mar. 10, 2004, at 4, (referring to Comcast as “the cable gorilla that every programmer needs”). 50 Federal Communications Commission FCC 05- 96 51 bargaining power over MVPDs. 336 Moreover, the situations in which a programming network can be expected to have the least amount of bargaining power relative to a cable operator are those in which the investment costs of the programmer are high, and the cost incurred by the cable operator from declining to carry it is low. Thus, it is plausible that programming networks with low relative bargaining power may be unable to recover their fixed programming costs. In this instance, the bargaining power of a cable operator may induce a programming network to exit the market, or to reduce its costs by lowering the quality of its programming. We ask for comment on whether a cable operator of sufficient size would have the bargaining power to force prices down, and whether this would reduce the quality and flow of programming, and create economic inefficiency. Furthermore, we ask how we can determine at what level this would occur. 94. A second and closely related form of inefficiency is known as the “hold- up problem” or the “underinvestment effect.” 337 Here, the party undertaking a relationship- specific investment, in this case the programming network, realizes that once its investments are sunk they cannot be recouped if bargaining breaks down. 338 Therefore, the more a firm invests upfront in relationship- specific assets, the weaker its bargaining position, and the lower its expected surplus from the negotiation. Anticipating this, a firm in this circumstance will under- invest, relative to the economic optimum, in relationship-specific assets. 339 Viewed broadly, the fear that parties may be held- up by other parties may lead to too little investment in specialized assets, i. e., programming, compared with the level of investment that maximizes economic efficiency. Thus, even if programming networks are able to negotiate successfully with a sufficient number of MVPDs under favorable terms, economic efficiency may not be maximized because of the relationship- specific nature of the programming network’s investments. We seek comment on whether contracts can be written that overcome this problem such that the possibility of hold- up can be reduced or eliminated. 95. A third source of inefficiency occurs when mutually beneficial trades fail to occur because the two parties in the bargaining process are uncertain about the size of the surplus available from a completed deal. In this case, each party may demand a larger amount than is available and a complete bargaining breakdown or delay might occur, which in the present context would result in the withholding of otherwise valued programs from the market. 340 Any such breakdown or delay would 336 See News- Hughes Order, 19 FCC Rcd at 543- 48 ¶¶ 147- 62. 337 See Oliver Williamson, MARKETS AND HIERARCHIES: ANALYSIS AND ANTITRUST IMPLICATIONS (1975). Williamson’s insight has been widely adopted in the study of contracts as well as vertical relationships in antitrust and regulation. See, e. g., J. J. Laffont and J. Tirole, A THEORY OF INCENTIVES IN PROCUREMENT AND REGULATION, 99- 100 (1993). 338 The seller can reduce the likelihood of having the bargaining breakdown by increasing its willingness to accept a low price from the buyer. 339 Programming networks and MVPDs have a unique relationship in that their investments are “co- specialized”; the economic value of their respective investments depends upon the behavior of the other party. Specifically, the value of the investments made by programming networks depends, in part, on the investment and other decisions made by MVPDs. Likewise, the value of the investments made by MVPDs depends, in part, on the investment and other decisions made by the programming networks. The co- specialized nature of their respective investments means that the interests of the programming networks and MVPDs are imperfectly aligned. 340 When the number of programming networks exceeds the number of channels the MVPD has allocated to carry those channels, an economic inefficiency may also be caused by the MVPDs’ uncertainty regarding the size of the (continued….) 51 Federal Communications Commission FCC 05- 96 52 impede the flow of programming. Economic theory has shown, however, that as the number of competitors on each side of the market increases, the likelihood of breakdown or delay is diminished, so that markets that have many competitors are approximately efficient. 341 Thus, the level of the inefficiency is directly and inversely related to the number of competing buyers and sellers. We ask whether an increase in cable concentration will likely lead to such breakdowns occurring, and thus increase the level of inefficiency. 96. If bargaining power is the frame of reference, then the economic question before the Commission is whether an increasing level of concentration among cable operators is likely to reduce the bargaining power of programmers to such an extent that (1) programmers cannot recover their costs, (2) the hold- up problem is amplified, or (3) the likelihood of bargaining breakdown increases. We seek comment on which of these economic inefficiencies may rise to the level of reducing the flow of programming to consumers. (a) The Use of Bargaining Theory to Establish New Limits 97. Cable industry commenters draw on the work of Alexander Raskovich to argue that large firm size could, in fact, weaken a cable operator’s bargaining position. For example, AT& T suggests that increased firm size reduces the likelihood of hold- up, because a larger cable operator can less credibly threaten to free- ride than a smaller cable operator, since the larger the operator, the more it will lose from failure to carry programming consumers value. 342 Moreover, if a buyer becomes so large that it becomes “pivotal” to a supplier’s production decision, the buyer cannot credibly abdicate responsibility for ensuring that the supplier’s costs are covered. 343 Time Warner, relying on Raskovich as well as Chipty and Snyder (1999), 344 claims that the larger cable operators’ decreased bargaining position results in larger operators “sharing in efficiencies that they have helped to create rather than exerting greater buyer market power.” 345 (Continued from previous page) rents available from carrying each channel. 341 The original result on the inefficiency of bilateral bargaining is know as the Myerson- Satterthwaite Theorem. See Roger B. Myerson and Mark A. Satterthwaite, Efficient Mechanisms for Bilateral Trading, 29 J. ECON. THEORY 2, 265- 81 (Apr. 1983). The proof that as individual bargaining power goes away the market becomes efficient is in Thomas Gresik and Mark Satterthwaite, The Rate at Which a Simple Market Converges to Efficiency as the Number of Traders Increases: An Asymptotic Result for Optimal Trading Mechanisms, 48 J. ECON. THEORY 1, 304- 32 (Jun. 1989). For an exposition of these and related results, see Drew Fudenberg and Jean Tirole, GAME THEORY (1993). 342 AT& T Comments, Ordover Decl. at 47- 44; AT& T Comments at 47. Assuming programmers recoup their programming investment by selling to a cable operator, they can sell their product to other operators at a substantially reduced price. Thus, competing cable operators “free- ride” on the operator who paid the up- front fixed costs of the seller. However, the ubiquity of so- called most- favored- nation clauses in programming contracts resolves this free- rider problem and protects the cable operator who initially purchases the programming from opportunism on the part of the programmer and other operators. 343 AT& T Comments, Ordover Decl. ¶¶ 47- 48. 344 Tasneem Chipty & Christopher Snyder, The Role of Firm Size in Bilateral Bargaining: A Study of the Cable Television Industry, 81 REV. ECON. & STAT. 2, 326- 40 (1999). 345 Time Warner Comments, Joskow and McLaughlin Decl. at 15. 52 Federal Communications Commission FCC 05- 96 53 98. Raskovich’s model is a generalization of the work of Chipty and Snyder, who construct a bargaining framework in which a program seller engages in simultaneous bilateral bargaining with multiple program buyers. Raskovich amends the model of Chipty and Snyder to include pivotal buyers, that is, buyers without whom sellers would produce zero output. 346 Assuming that there is an even split between buyers and seller (i. e., 50%- 50% of a trade’s surplus), Raskovich demonstrates conditions under which the pivotal buyer finds its bargaining position worsened. 347 Raskovich posits a situation in which a buyer becomes so large through merger that only the buyer can cover the seller’s cost of producing programming. In this context, the programmer’s surplus from bargaining with the single large cable operator would be greater than the sum of the surpluses the programmer would receive from the two buyers prior to the merger. This implies that once a cable operator reaches a sufficient size, its payments to programmers will increase. 348 99. Neither the Chipty and Snyder model nor the Raskovich model persuades us that limits on cable operator size are unnecessary. Adilov and Alexander show that if there are asymmetries in bargaining power, i. e., the surplus split varies from 50%- 50%, the results of Chipty and Snyder and Raskovich may not hold. 349 Rather, they demonstrate that where bargaining power is not symmetric, mergers could improve a cable operator’s bargaining position and decrease payments to programmers, even when the merged firm becomes pivotal. 350 100. We find it unlikely that bargaining power is symmetric across all buyers regardless of size. No commenters offer any reasons or evidence to support the assumption that it is. Adilov, using basic data from the BKS Study, estimates bargaining power directly. 351 Adilov’s results reveal statistically significant differences in individual buyers’ bargaining power, a result that is not consistent with an assumption of constant bargaining power across firm size. The data generated from the BKS Study also show that buyers and sellers did not split the economic surplus evenly under all conditions. 352 We seek comment on the usefulness of the technical analyses contained in bilateral bargaining theory in light of the wide range of results it appears to generate. (b) Experimental Economics Study 346 Raskovich Comments at 3- 4; Raskovich, Pivotal Buyers and Bargaining Position. 347 Id. at 11- 22. 348 Id. at 22. 349 Adilov and Alexander generalize the models of Raskovich and Chipty and Snyder to allow for asymmetric bargaining power (i. e., the split does not have to be 50%- 50%). See Adilov & Alexander, Asymmetric Bargaining Power, supra. 350 Id. at 8. 351 Adilov ex parte statement (Jan. 9, 2003) (submitting Nodir Adilov, Firm Size and Bargaining Power: A Non-Linear Least Squares Estimate from the Cable Industry, Working Paper, Department of Economics, Cornell University (Nov. 2002)). 352 Id. at 11. 53 Federal Communications Commission FCC 05- 96 54 101. In 2002, the Commission launched the BKS Study, regarding the extent to which different levels of horizontal concentration among MVPDs might affect the flow of video programming to consumers. The study utilized the methodology of experimental economics, which examines economic interactions among market participants in controlled laboratory settings. The study was designed to aid the Commission in its evaluation of a horizontal limit on remand from the court in Time Warner II. The Commission placed the BKS Study in the record of this proceeding and sought comment on it. 353 To assist the public in its analysis of the study, the Commission also made available the raw data upon which the study’s conclusions are based. 354 102. The BKS Study created an experimental market that parallels, in significant ways, the market in which programming networks and MVPDs negotiate affiliate fees. 355 Economic experiments were run under different levels of horizontal concentration among cable buyers. 356 The concentration levels were chosen so as to generate a wide size range for buyers while, at the same time, depicting concentration levels that may occur in the future absent government intervention. The study assessed the effects of horizontal concentration using four measures: economic efficiency, buyer’s bargaining power, buyer surplus, and seller profits and losses. 357 The results for each measure varied, but by at least one measure – seller profits and losses – the study found that all except the most popular programming networks fared significantly worse in the market dominated by a single 51% buyer than in the market in which the two largest buyers served 44% and 39% of subscribers. 358 The adverse effects on seller profits in these hypothetical markets could induce sellers to either exit the market or lower the quality of their programming, particularly if alternative investments offered better comparative returns. 103. Commenters raise several objections to reliance on the BKS Study in setting a horizontal ownership limit. Perhaps the most common criticism of the study concerns “parallelism.” Critics claim that because experiments cannot mirror the “real world” perfectly, the study cannot provide useful evidence. 359 For example, commenters noted the experiment failed to model the DBS operator as a direct competitor to the cable operators. Commenters also raise a number of arguments concerning the study’s methodology. For instance, AT& T argues the study’s alleged poor design induced subjects to engage in 353 Public Notice, Media Bureau Seeks Comment on Experimental Economics Study Examining Horizontal Concentration in the Cable Industry, 17 FCC Rcd 10544 (2002). 354 See www. fcc. gov/ osp/ workingp. html. 355 BKS Study at 9- 21. 356 The experiment measured effects under three different hypothetical market configurations: (1) a market in which the largest buyer served 27% of all subscribers; (2) a market with two large buyers serving 44% and 39% of the market; and (3) a market in which one large buyer served 51% of the market and the next largest buyer served only 17%. Id. at 15. 357 Id. at 22- 26. 358 Id. at 45- 48. In this environment, the least popular networks suffered greater losses than in the 44%/ 39% market, and the moderately popular network enjoyed lower profits than in the 44%/ 39% market. Id. 359 AT& T Supplemental Comments at 8; Comcast Supplemental Comments at 6, 8; Time Warner Supplemental Comments at 5; SBC Supplemental Comments at 1. 54 Federal Communications Commission FCC 05- 96 55 loss avoidance rather than profit- maximizing behavior. 360 Time Warner, on the other hand, claims that the study induced subjects to act erratically. 361 104. We recognize that the BKS study has limitations in that DBS was not modeled as a competitor, and the study did not include vertically integrated players. However, we believe that experimental economics can be a useful tool for evaluating the effects of increasing concentration. We seek comment on whether we should continue to consider experimental economics, as well as additional analytical methods that may help us devise a limit. b) Additional Factors in the Analysis 105. In the previous section we presented three frameworks for analyzing the potential harms associated with horizontal concentration among buyers in the programming market. In this section we discuss four factors that should be considered when designing, evaluating, and applying an analytical framework. For each of these factors we seek comment on its weight and importance in each of the analytical frameworks we have examined, as well as suggestions on how to incorporate the factor into the analytical frameworks. (1) The Impact of Competition at the Distribution Level 106. The Time Warner II court criticized the Commission for failing to examine whether cable operators had market power in the distribution market, and, in particular, for failing to take into account the growth of competition from DBS providers. 362 It also expressed concern that the Commission’s analysis was focused too narrowly on cable operators’ current market share, and that a proper analysis of market power should include consideration of “the availability of competition,” and its impact on the elasticities of supply and demand. 363 It pointed out that MVPDs that attempted to exercise market power by refusing to carry new programming might find their customers switching to other MVPDs. 364 360 AT& T Supplemental Comments at 14. 361 Time Warner Supplemental Comments at 10. In addition to the above objections, there are a standard set of criticisms leveled against economic experiments. See Vernon Smith, Method in Experiment: Rhetoric and Reality, 5 EXPERIMENTAL ECONOMICS 2, 91- 100 (Oct. 2002). For example, it is sometimes asserted that the subjects participating in the experiments do not have the requisite level of knowledge and experience, and that parties to actual negotiations use consultants to assist in the decision- making process. Another criticism is that the instructions that describe the economic environment in which the subjects participate may have been unclear or inadequate, thereby leading to anomalous subject behavior. A closely related criticism is that, perhaps because of the complexity of the economic environment, subjects were not given sufficient opportunity to learn how best to behave in the economic experiment. 362 Time Warner II, 240 F. 3d at 1134. 363 Id. (emphasis in original). 364 Id. 55 Federal Communications Commission FCC 05- 96 56 107. In the 2001 Further Notice we noted the growth of DBS’ share of the MVPD market. 365 We sought comment on the impact of DBS’ growth and presence on cable operators’ market power and on their incentive to choose programming for reasons other than quality. 366 We also sought comment on what level of competition in the MVPD market would be “sufficient to provide alternative means for programmers viably to reach consumers,” and on the appropriate measure for determining when this level of competition is reached. 367 108. In response, cable commenters argue that the Commission must conduct a “dynamic” examination of market power. They suggest that the ability to exercise market power depends not only on market share, but on the elasticities of supply and demand. Thus, to determine cable operators’ market power, we should consider barriers to entry and emerging competition, as well as their long- run effects. These commenters maintain that a dynamic analysis of the MVPD market indicates that the Commission need not impose any limits, since programmers have so many different outlets for their product that cable operators hold no deleterious market power. 368 They point out that when Section 613( f) was enacted, cable and broadcast television were the primary outlets for distributing video programming, while in the interim, other forms of video distribution, primarily DBS, have become much more widespread, such that cable’s share of all MVPD subscribers has been reduced by almost 25%. 369 Furthermore, many commenters contend that DBS is growing at a fast rate – a rate that exceeds cable’s growth rate – and that it is offering digital technology that has vastly expanded channel capacity. In addition, Comcast points out that while DBS was originally predicted to thrive only in areas not served by cable, today, at least half of all new DBS subscribers are switching from cable. 370 These commenters maintain that because any dissatisfied cable customer can switch to DBS, cable operators have no incentive to lower the quality or quantity of programming. Therefore, in their view, limits are not necessary. 371 109. CFA argues that DBS is not a substitute for cable, because of its higher price and quality. 372 It argues that DBS provides a high end product that is not attractive to the typical “lunch bucket” consumer of cable services. 373 CFA claims that DBS providers prefer to compete in terms of 365 2001 Further Notice, 16 FCC Rcd at 17326 ¶ 21. 366 Id. at 17326- 27 ¶ 22. 367 Id. at 17327 ¶¶ 23- 24. 368 Comcast Comments at 29- 31; Time Warner Comments, Joskow and McLaughlin Decl. at 2- 6, 11, 21; NCTA Comments, Shelanski Decl. at 8; AT& T Comments, Ordover Decl. at 62; Time Warner Comments at 13- 14; AT& T Reply Comments at 6. 369 Comcast Comments at 17- 29; NCTA Comments at 11- 14. 370 Comcast Comments at 23. 371 Comcast Comments at 17- 21; Time Warner Comments, Joskow and McLaughlin Decl. at 6- 7; AT& T Comments, Ordover Decl. at 10, 23- 26; PFF Comments at 18; AT& T Reply Comments at 7. 372 CFA Comments at 163- 70. See also Writers Guild Comments at 9- 10 (rejecting the notion that the existence of DBS could render horizontal limits unnecessary). 373 CFA Comments 170- 71. 56 Federal Communications Commission FCC 05- 96 57 programming and not price, and that the rise of DBS competition has failed to limit cable rate increases. 374 CFA points to survey data that show that rural areas often lack cable service, and that a large proportion of satellite customers live in rural areas. 375 CFA claims that the survey data demonstrate that for most satellite customers cable is not a substitute, either because it is not available, or because consumers view it as a complement. 376 As supporting evidence CFA provides an analysis of consumer survey data and consumer monthly bills that shows that DBS services are considered higher quality, and cost more than basic cable. 377 110. We seek comment on CFA’s arguments and evidence, especially in light of the rapid growth of DBS subscribership and recent changes in the prices and programming DBS operators offer. We find no evidence that cable subscribers are substantially less affluent than satellite subscribers. In 2003, average household income of cable subscribers was $48,700, while that of satellite subscribers was $51,600. 378 In addition, we note that recent reports suggest that cable subscriber growth has stalled, while DBS subscribership continues to grow at a rapid rate. 379 111. We also seek comment on whether a dynamic analysis of the type envisioned by cable commenters is necessary, and comment on how we could perform such an analysis. A number of factors suggest that a dynamic analysis is not necessary. First, barriers to entry in the MVPD market remain high for new entrants. 380 Cable overbuilders in particular have faced many obstacles in their attempts to enter and survive in the marketplace, and many overbuilders have scaled back construction plans or failed. 381 Because the costs of building competing cable systems are high, overbuilders today generally 374 Id. at 155- 57. 375 Id. at 159- 60. 376 According to CFA: “Thus, in this survey, just under 60% of respondents either cannot get cable or appear to view it as a complement, rather than a substitute.” Id. at 163. 377 Id. at 163- 70. See also 2005 GAO Report at 7- 8. 378 Calculated by FCC staff using survey data from TNS Telecoms ReQuest Market Monitor™, Bill Harvesting®. 379 See 11 th Annual Report, 20 FCC Rcd at 2758, 59 ¶¶ 5, 7; Peter Grant, Cable Trouble: Subscriber Growth Stalls As Satellite TV Soars, WALL ST. J., Aug. 4, 2004, at B1. 380 Because of the high sunk costs and specialized assets needed to enter this market, one possible “dynamic” theory, that of contestable market theory, may not apply to this market. See William J. Baumol, John C. Panzar & Robert D. Willig, Contestable Markets and the Theory of Industry Structure (1988). 381 See, e. g., 8 th Annual Report, 17 FCC Rcd at 1294- 97 ¶¶ 107- 15; Annual Assessment of the Status of Competition in the Market for the Delivery of Video Programming, 17 FCC Rcd 26901, 26948- 52 ¶¶ 102- 11 (2002) (9 th Annual Report). See also RCN Comments at 3- 4. There have been recent announcements regarding some Bell Operating Companies’ plans to roll out fiber optic cable to the home (FTTH) to provide advanced digital services to their customers, including multichannel video programming. See, e. g., Verizon, Verizon Poised to Deliver First Set of Services to Customers Over Its Fiber- to- the- Premises Network (press release), July 19, 2004, available at http:// newscenter. verizon. com/. The announcements generally involve limited “test market” areas. Widespread deployment, if determined to be feasible, is still several years away. See also 11 th Annual Report, 20 FCC Rcd at 2823- 25 ¶¶ 127- 28. 57 Federal Communications Commission FCC 05- 96 58 are concentrated in high- density portions of urban and suburban markets. 382 New satellite- based competitors, on the other hand, must contend with physical spectrum constraints and limited orbital capacity. 383 112. Second, the number of existing MVPD competitors in most geographic areas remains small for the distribution market, three in most cases, and two in others. 384 Because this market is still highly concentrated, cable operators are likely to retain substantial market power in those areas. We note that commenters make no serious attempt to calculate the effect of DBS on cable operators’ bargaining power. We seek comment on the degree to which the presence of DBS distribution alternatives acts to curb cable operators’ bargaining power in the total programming market. 385 113. We also find that commenters have failed to provide a method of analyzing the effects of competition in the MVPD market that would allow us to establish a specific limit. Despite their criticisms of the static model, no commenter described, demonstrated, or utilized a theoretical framework that could incorporate the competitive effects that were alleged to be important. 386 We seek comment on whether we can modify the model to incorporate these important competitive effects, or develop a new framework, taking all relevant effects into account that would enable us to carry out our statutory responsibility under Section 613( f). (a) Threshold Approach 114. In the 2001 Further Notice the Commission asked for comment on whether to assess a cable operator’s market power in the MVPD market with a measure other than its market share in the national market, and to use this alternative measure in a so- called threshold approach. Under this approach, the Commission would determine the level of competition from DBS and other MVPDs necessary to prevent the harms identified by Congress in Section 613( f). 387 As long as competition exceeded this threshold, no horizontal limit would be necessary. The threshold would denote a level of competition at which the market afforded sufficient alternative means, in addition to cable, for video 382 See 8th Annual Report, 17 FCC Rcd at 1294 ¶ 107. 383 See EchoStar- DirecTV HDO, 17 FCC Rcd at 20616- 19 ¶¶ 140- 50. See also Morgan Stanley, “Cablevision Plans To Spin Off Satellite and Theater Assets,” June 3, 2003 (“ Even with all 13 frequencies, there is not enough spectrum to effectively offer local into local.”) 384 See EchoStar- DirecTV HDO, 17 FCC Rcd at 20612- 14 ¶¶ 126- 32. 385 We note that subscribership totals for existing non- cable MVPDs, especially DBS, are included in our calculations of market shares of the programming market for cable operators. Thus the importance of DBS providers in providing a competitive alternative to cable providers could be considered to be reflected in DBS’ shares of the total MVPD subscribers. 386 It is important to recognize that although commenters have called for a dynamic market analysis, they have failed to provide a mathematically rigorous dynamic model. The Raskovich model, which commenters rely upon extensively, is a static model. We are not aware of a dynamic model of the MVPD industry. 387 2001 Further Notice, 16 FCC Rcd at 17343 ¶ 64. 58 Federal Communications Commission FCC 05- 96 59 programmers to reach consumers. The 2001 Further Notice proposed several measures that could be used in a threshold test discussed below, and asked for comment on these. 388 115. PFF advocates a threshold approach, and argues that the Commission should find that the existence of a single MVPD competitor to incumbent cable operators is sufficient to curb the harms envisioned by Section 613( f). 389 RCN supports a similar approach for measuring and addressing market power to control sought- after programming in individual markets and stresses that the threshold approach should be applied on a market- by- market basis and not simply applied nationally. 390 CFA opposes the proposed threshold approach, arguing that it does not meet the statutory requirement that the Commission “shall” set a horizontal ownership limit. 391 CFA further contends that the approach is unworkable because the Commission could not effectively enforce it. 392 116. We seek additional comment on the use of the threshold approach in establishing a horizontal ownership cap. In the 2001 Further Notice, the Commission requested comment on whether the Implicit Lerner Index, the “q” ratio, or the Herfindahl- Hirschman Index (HHI) could be used in the threshold approach. 393 We seek additional comment on the use of these measures, as well as alternative measures of market performance. How well does the economic theory underlying these measures comport with the characteristics of the video programming and MVPD markets? How do the numeric values of these measures relate to the degree to which the flow of video programming from the programmer to the consumer may be unfairly impeded? If we adopted any of these measures, how would they be calculated and applied to determine the appropriate horizontal limit? 388 Id. at 17345- 46 ¶¶ 69- 70. In the 2001 Further Notice, we also sought comment on a restriction on cable- DBS cross- ownership as it would relate to the adoption of a threshold approach. Id. at 17345 ¶ 68. We received comments both supporting (Writers Guild Comments at 10) and opposing (Cablevision Comments at 5, 15- 16) adoption of such a restriction. Since we are not adopting the threshold approach, we do not reach the question of whether a cable- DBS cross- ownership restriction is necessary to promote the goals of Section 613( f). 389 PFF Comments at 17- 18. At the same time, PFF recommends that the Commission be permitted to continue to gather evidence of the existence of market power and resulting harms to consumers in the MVPD marketplace and fashion remedies where appropriate. Id. 390 RCN Comments at 17- 18. 391 CFA Comments at 21- 25. 392 Id. at 24- 25. 393 2001 Further Notice, 16 FCC Rcd at 17342- 43 ¶¶ 62- 64. The Commission has previously examined the q-ratio (also known as “Tobin’s Q”). For instance, commenters to the 1990 Report calculated q- ratios for the cable industry of between 3.3 and 4.3. We noted in the report that although the high q- ratios indicated some, or even substantial, market power in 1990, application in this context must be made carefully because the q- ratio is sensitive to the assumptions made in its calculation and to specific industry characteristics. See Competition, Rate Deregulation and the Commission’s Policies Relating To the Provision of Cable Television Service, 5 FCC Rcd 4962, 4997- 5003 ¶¶ 54- 70 (1990) (“ 1990 Report”). Similar conclusions were made in the First Annual Video Competition Report. See Annual Assessment of the Status of Competition in the Market for the Delivery of Video Programming, 9 FCC Rcd 7442, 7542- 45 ¶¶ 204- 12 (1994). 59 Federal Communications Commission FCC 05- 96 60 (2) The Potential for Joint Action 117. Section 613( f)( 2)( A) of the Communications Act requires that the Commission ensure that cable operators, either singly or as a group, because of their size or because of their joint actions, not be able to unfairly impede the flow of video programming. 394 In the 1999 Cable Ownership Order, the Commission assumed that multiple avenues of entry were necessary to ensure the unimpeded flow of programming. The Commission utilized an open field approach to set a 30% limit, which guaranteed that even if there were collusion between the two largest players to attempt to prevent entry by a programmer, that programmer would still be able to gain enough subscribers through carriage on other systems. The 1999 order also hypothesized that the two largest operators might effectively preclude entry of a new programmer by tacitly reaching the same carriage decision. The Time Warner II court rejected the Commission’s analysis, arguing that the risk of collusion had been inadequately substantiated, and that the Commission had failed to demonstrate that the legitimate, independent editorial choices of multiple cable operators could “unfairly” impede the flow of video programming. 395 It found the Commission had not presented evidence that collusion was likely and therefore had not adequately supported its limits. 118. We ask whether Section 613( f)( 2)( A) of the Communications Act requires the Commission to examine the possibility of joint action, in which firms act to maximize their joint benefits by reducing competition, either through overt collusion, which is generally prohibited by the antitrust laws, 396 or tacit collusion, without direct communication between the firms. 397 We also ask whether such an analysis would be consistent with the court’s findings in Time Warner II. Because the language of the Act refers to cable operators’ “joint actions,” and because the economics and legal literatures (including the Horizontal Merger Guidelines) 398 acknowledge the possibility of tacit collusion in certain 394 Section 613( f)( 2): “In prescribing rules and regulations under paragraph (1), the Commission shall, among other public interest objectives – (A) ensure that no cable operator or group of cable operators can unfairly impede, either because of the size of any individual operator or because of joint actions by a group of operators of sufficient size, the flow of video programming from the video programmer to the consumer.” 395 Time Warner II, 240 F. 3d at 1134- 36. 396 In most cases, for example, communications between or among firms in order to fix prices is per se illegal. 397 “Tacit collusion” is the standard term used in the economics literature used to refer to competing firms acting to maximize their joint benefits by reducing competition between them without directly communicating with each other. Other terms used are “conscious parallelism” and “tacit coordination.” See, e. g., Carlton and Perloff at 134; Marc Ivaldi, Bruno Jullien, Patrick Rey, Paul Seabright, and Jean Tirole, The Economics of Tacit Collusion, n. 2, Final Report for DG Competition, European Commission, Institut D’Economie Industrielle (Toulouse) (Mar. 2003); F. M. Scherer and David Ross, INDUSTRIAL MARKET STRUCTURE AND ECONOMIC PERFORMANCE, 265, 339 (3rd ed., 1990). The 1997 Horizontal Merger Guidelines use the broader term “coordinated interaction” in section 2.1 to refer to “actions by a group of firms that are profitable for each of them only as a result of the accommodating reactions of the others.” It is a broader term because it refers to both tacit and express collusion. 1997 Horizontal Merger Guidelines, section 2.1. 398 1997 Horizontal Merger Guidelines, section 2.1. See, e. g., Carlton and Perloff at 134; Ivaldi, Jullien, Rey, Seabright, and Tirole; Scherer and Ross at 205- 06, ch. 7, 9; Herbert Hovenkamp, Antitrust Law: An Analysis Of Antitrust Principles And Their Application, XII ASPEN LAW & BUSINESS, 21 (1999). 60 Federal Communications Commission FCC 05- 96 61 circumstances, we tentatively conclude that we should determine whether joint action 399 by cable operators is likely, and if we determine that it is likely, we should factor this into the analysis. 119. We note that an explicit agreement among firms in a given market may not be necessary for that market to be characterized by joint action. Such collusive behavior may arise as a result of “conscious parallelism” in the behavior of firms. Conscious parallelism can arise without any explicit agreement among firms, but simply as the result of a rational calculation by each firm of the consequences of its actions for competing firms, particularly taking into account the most likely reactions of those firms. 400 This kind of coordinated action is difficult to detect or control. As one court observed: “Tacit coordination is feared by antitrust policy even more than express collusion, for tacit collusion, even when observed, cannot easily be controlled directly by the antitrust laws. It is a central object of merger policy to obstruct the creation or reinforcement by merger of such oligopolistic market structures in which tacit coordination can occur.” 401 120. We sought comment and economic evidence in the 2001 Further Notice on whether cable operators have the incentives to engage in collusive behavior, and on what kinds of coordinated or collusive conduct would be relevant to the establishment of a limit. 402 We also sought information and evidence on whether cable operators’ existing activities constitute collusion. 403 Cable commenters argue that cable operators lack an incentive to collude, and that this is evidenced by their past behavior, which shows that cable operators have not disfavored unaffiliated programming nor hindered the flow of programming. 404 They also argue that collusion to block entry by a rival programmer is a violation of the antitrust laws. 405 Time Warner argues that cable operators that are not vertically integrated do not have an incentive to collude, because they do not compete with each other. 406 Cable commenters also argue that reducing purchases of programming will not yield lower prices and that therefore cable operators have nothing to gain from colluding in order to jointly exercise monopsony power. 407 AT& T 399 We focus our analysis on the likelihood of tacit collusion, since overt collusion is per se illegal under the antitrust laws. Scherer and Ross at ch. 9. 400 See, e. g., D. F. Turner, The Definition of Agreement Under the Sherman Act: Conscious Parallelism and Refusal to Deal, 75 HARV. L. REV. 655 (Feb. 1962) at 661; Scherer and Ross at 339- 46. 401 Federal Trade Comm’n v. H. J. Heinz Co., 246 F. 3d 708, 725 (D. C. Cir. 2001) (citing Phillip E. Areeda, Herbert Hovenkamp & John L. Solow, Antitrust Law § 901b2 at 9 (rev. ed. 1998)). 402 2001 Further Notice, 16 FCC Rcd at 17340 ¶ 56. 403 Id. 404 AT& T Comments at 66- 67; AT& T Comments, Ordover Decl. at 77- 82; AT& T Comments, Besen Decl. at 6- 14; Time Warner Comments, Joskow and McLaughlin Decl. at 19- 20; NCTA Comments at 17- 20; Time Warner Comments at 20- 22. 405 Time Warner Comments, Joskow and McLaughlin Decl. at 20; Time Warner Comments at 21; AT& T Comments, Ordover Decl. at 77. 406 Time Warner Comments, Joskow and McLaughlin Decl. at 19- 20. 407 Id. at 20; AT& T Comments, Ordover Decl. at 78; NCTA Comments at 19- 20. 61 Federal Communications Commission FCC 05- 96 62 argues that the possibility that cable operators would jointly engage in vertical foreclosure is “particularly far- fetched,” because joint action increases the costs and reduces the benefits of such an action. 408 Time Warner argues that the characteristics of the industry make collusion unlikely. It argues that the wide variation in the value of carrying particular programming to cable operators, and the complex non- public nature of program carriage contracts, “make reaching a formal or tacit agreement, policing it, and punishing cheating extremely difficult.” 409 121. We are not persuaded by the comments that joint action could not occur under certain circumstances. Much of the empirical evidence cited by the cable industry’s economists is based on past performance of the market, when cable operators were much smaller than we are contemplating today, although even then there were reports of various forms of joint action. 410 We also find unpersuasive arguments that cable operators lack an incentive to act jointly to gain advantage. If we determine that a cable operator of a sufficient size would find it profitable to engage in conduct of the types discussed above, 411 then the possibility exists that two or more smaller cable operators, whose combined size is also sufficiently large, would seek to maximize their profits by jointly engaging in this conduct. Because we remain concerned about the possibility of joint action, we seek further comment on whether cable operators have the incentive and ability to engage in joint action. If joint action is likely, we ask how many cable operators are likely to engage in joint action on any programming decision, 412 and how we should use the findings on these issues to devise a horizontal limit. 122. We first ask whether cable operators have an incentive to engage in joint action with respect to the acquisition of programming. If a single firm of sufficient size has an incentive to engage 408 AT& T Comments, Ordover Decl. at 79. In contrast, Time Warner believes that only vertically- integrated cable operators could have an incentive to collude. It argues that the legal restrictions on collusion, the wide variation in benefits for cable operators’ carrying a program, and the complexity of program carriage contracts, make it difficult to collude. Time Warner Comments, Joskow and McLaughlin Decl. at 20. 409 Time Warner Comments, Joskow and McLaughlin Decl. at 20; see also AT& T Comments, Ordover Decl. at 78. 410 See, e. g., Leo Hindery and Leslie Cauley, The Biggest Game of All: The Inside Strategies, Tactics, And Temperaments That Make Great Dealmakers Great, New York: The Free Press, 2003, at 185. (“ Some years ago, Group W, the cable arm of Westinghouse, tried to launch a news service to compete with CNN. Cable operators locked arms – and turned Group W down flat. After months of trying to get carriage for the service and getting nowhere, Group W finally shut the whole thing down.”) 411 For example, the possibility of a cable operator using its bargaining power to force down the price for programming to below competitive levels, and engaging in vertical foreclosure to reduce entry and competition in the programming or distribution markets. See Sections II. C., supra (analysis of monopsony power, bargaining power, and vertical foreclosure). 412 Clearly, the greater the number of cable operators that are likely to act jointly, the smaller the individual threshold size we should be concerned with. Thus if, for example, the threshold size for anticompetitive conduct were found to be 60% for one firm, the threshold assuming joint action by two firms would be 30%, by three firms 20%, and by four firms 15%, to achieve the same benefits (assuming that the threshold size remains constant as the number of firms acting jointly increases). The economics literature suggests that the ability and desire to act jointly decreases as the number of firms participating increases. Carlton and Perloff at 132- 34; Scherer and Ross at 277- 79. 62 Federal Communications Commission FCC 05- 96 63 in anticompetitive behavior, does it necessarily follow that a group of smaller firms would have an incentive to collectively engage in similar behavior? 413 Assuming that a single firm did have an incentive to engage in anticompetitive behavior of some sort, 414 a key question that follows is whether cable operators are likely to have similar or divergent interests in their purchase of programming. The MVPD market appears to exhibit a number of characteristics that could provide an incentive for cable operators to coordinate their actions in the purchase of programming. Cable operators are dependent on quality programming to attract and keep subscribers; programming networks depend on MVPDs for distribution of their programming to consumers, the most important of which in most geographic markets are the cable operators; and cable operators compete against DBS and other entrants, but generally not against each other. These facts align cable operators’ interests in such a way that makes joint action potentially desirable for them. 415 In certain aspects cable operators may have similar interests, while in others their interests may diverge. For example, it is likely that all cable operators want to minimize their payments for the programming networks they carry, or at least to extract the best value possible for the lowest price. Yet they may differ in the cost- quality tradeoffs they might accept, with some cable operators preferring lower- quality, lower- cost programming. They may also have a divergent interest in desiring to shift the burden of paying the fixed costs of programming onto other cable operators. Cable operators may also diverge in their interests in discriminating against rival networks. Some cable operators may not have the same interest in foreclosing an independent network as an operator who owns a rival affiliated network, and may in fact prefer to maintain competition in network programming. Certain practices may increase cable operators’ incentives to act jointly. For example, joint ventures by cable operators in providing network programming potentially give the co-owners a shared incentive to discriminate against rival networks. 416 We seek comment on this analysis. 413 If so, it could be argued that any time we find that a single cable operator of a particular threshold size is likely to successfully engage in anticompetitive behavior, we should then consider whether multiple operators whose combined size achieves at least that threshold size, would likely engage in collusive behavior to achieve the same results. 414 The question of whether a single firm of sufficient size would have an incentive to engage in anticompetitive behavior is discussed above in the sections on monopsony power, bargaining power, and vertical foreclosure. See Sections II. C., supra. See also Time Warner II, 240 F. 3d at 1130 (“ The Commission is on solid ground in asserting authority to be sure that no single company could be in a position singlehandedly to deal a programmer a death blow.”); Time Warner I, 211 F. 3d at 1320 (“ Congress reasonably concluded that this concentration [in the cable industry] threatened the diversity of information available to the public and could form a barrier to entry of new cable programmers.”). Aside from the question of a single firm’s behavior, we ask whether it is possible that two or more firms could have an incentive to collude and jointly engage in anticompetitive behavior. 415 In certain key aspects, much of the discussion of the theory of collusion in the economics literature does not apply to this market. While in the usual discussion of a cartel selling a good, each member’s sales will potentially reduce the sales of the other members, here, because the product is characterized by non- rivalrous consumption, each purchase does not reduce the quantity available, or the purchases by others, of the product. In addition, cable operators do not compete with each other, so they cannot gain market share by cheating on the cartel. 416 Many cable operators participate in joint ventures to provide network programming. For example, Cox Communications and Advance/ Newhouse Communications each own 25% of Discovery Communications, which owns cable networks such as the Discovery Channel, Animal Planet, and TLC. See 11 th Annual Report, 20 FCC Rcd at 2874- 76 Appendix C, Table C- 1. See also Annys Shin, Big Discovery Shareholders Refuse to Join Liberty Spinoff, WASHINGTON POST, May 13, 2005, at E1. 63 Federal Communications Commission FCC 05- 96 64 123. We also seek comment on whether cable operators have the ability to engage in joint action, and we seek economic analysis and evidence indicating the ease of engaging in joint action in this market. Some general characteristics of the industry may facilitate joint action. The MVPD market is highly concentrated with high barriers to entry by new competitors and an absence of close substitutes for the services both bought and sold by the industry. These characteristics are similar to the general characteristics described in the economic literature as leading to either overt collusion or conscious parallelism in behavior. 417 124. Other general characteristics of the industry may make joint action more difficult. The product purchased, network programming, is heterogeneous between networks. Therefore establishing a uniform price schedule for the purchase of all programming 418 would be very difficult. 419 However, each programmer’s product (i. e., its programming network), which is offered to all MVPDs, is homogenous, providing cable operators the opportunity to engage in joint action with respect to the price paid for that network. The use of private negotiations, with non- public terms of agreement, would appear to make it very difficult for cable operators to tacitly engage in joint action, at least in terms of prices. The complexity of these agreements, and the tendency to specify lower per- subscriber prices for larger purchases of programming in these agreements, also mitigate against joint action. 420 Joint action for the purpose of vertical foreclosure of rival networks, however, will not be hindered by these practices, since network carriage is easy to observe. 125. We seek comment on the harms (or benefits) that could be caused by joint action. We do not wish to promulgate regulations that prevent beneficial joint action, but we are concerned nonetheless about the possibility for harmful joint action. Some joint actions may harm consumers by making the potential harms arising from horizontal concentration possible at lower levels. Joint action can be particularly harmful because it creates the inefficiency attendant to a monopoly, but denies consumers the efficiencies that might result from a merger. 421 In addition, joint action that seeks to shift costs onto rival MVPDs or to favor affiliated programming over unaffiliated programming distorts the market and denies consumers the benefits of fair competition. On the other hand, some forms of joint action among cable operators benefit consumers and are desirable. For instance, joint action to save struggling networks or joint ventures to launch new networks may preserve or increase the diversity and quality of programming available to consumers. 422 The effects of other actions are indeterminate in that 417 See Pepell, Richards and Norman, Industrial Organization at 383. These general characteristics are as follows: (1) there are very few substitutes available to consumers for the products sold (or bought) by the firms in the market (i. e., the demand curves for the products are relatively inelastic); (2) there is little or no prospect of competitive entry into the market; (3) the cost of reaching a cooperative agreement among firms in the market is small due to the high level of industry concentration, the small number of firms in the market, the similar cost conditions among the firms and/ or the lack of significant product differentiation among the firms; (4) the cost of maintaining a cooperative agreement among the firms is also small due to frequent interaction among the firms; and (5) market conditions tend to be relatively stable. 418 Price- fixing is a traditional goal of cartels. 419 Time Warner Comments, Joskow and McLaughlin Decl. at 20. 420 Id. 421 See John E. Kwoka, Jr. and Lawrence J. White, THE ANTITRUST REVOLUTION, 167 (3 rd ed., 1999). 422 “For example, on several occasions, MSO investment has enabled a programming service to remain in (continued….) 64 Federal Communications Commission FCC 05- 96 65 they may benefit consumers, but may also harm them. For instance, joint bargaining for lower programming costs may lower cable operator costs and allow them to charge lower prices, but it may also harm programmers or MVPD rivals by reducing the amount and quality of programming available, and thus deny consumers quality programming or the benefits of competition. We seek comment on the likelihood that joint action will impede the flow of programming, either by forcing down the price of programming paid by cable operators to a level that reduces the quantity and/ or quality of programming, or by foreclosing entry by either rival unaffiliated network programmers or by competing MVPDs. (3) The Impact of Independent Actions by Multiple Cable Operators 126. We ask whether there are theories addressing how multiple cable operators that are acting independently could unfairly impede the flow of programming, as discussed in Time Warner II. The open field approach the Commission used in the 1999 Cable Ownership Order assumed that multiple avenues of entry were necessary to ensure the unimpeded flow of programming. The Commission pointed out that a 30% limit would ensure that there were at least four cable operators. The Time Warner II court held that the Commission had failed to demonstrate that the legitimate, independent editorial choices of multiple cable operators could “unfairly” impede the flow of video programming. In the 2001 Further Notice we sought comment on the possible harms that could arise from high levels of horizontal concentration. 423 We also sought comment on the possible effects of the level of concentration on the amount and diversity of programming. 424 AT& T argued that if there are at least two outlets and no collusion, a programmer’s failure to reach homes is the result of “legitimate, independent editorial choices” and cannot be deemed unfair. 425 Comcast argues that cable operators are unable to exercise editorial oversight and impair the ability of a program producer to access viewers through broadcast stations because the stations can secure carriage through the exercise of their must-carry rights. 426 The Writers Guild of America argues that consolidation in program production and distribution has already eroded quality and creativity and reduced diversity. 427 We seek comment on whether there are analytical approaches that would establish whether multiple cable operators, acting independently and with no attempt to overtly or tacitly coordinate their actions with other cable operators, could harm the market or the ability of programmers to gain carriage. We further ask whether such approaches would be consistent with the court’s holding in Time Warner II that promoting diversity alone is not a sufficient basis for crafting a limit designed to address multiple cable operations’ independent editorial choices. 428 (Continued from previous page) operation when it otherwise would have been forced to discontinue its programming. MSO commenters emphasize that the cable industry provided critical financial support to sustain both Turner Broadcasting (owner of WTBS and CNN) and C- SPAN. In addition, NCTA quotes Discovery Channel Chairman John S. Hendricks’ statement that cable operators’ investment ‘rescue[ d] ’ his programming service.” 1990 Report, 5 FCC Rcd at 5009 ¶ 83 (citations omitted). 423 2001 Further Notice, 16 FCC Rcd at 17340 ¶ 57. 424 Id. at 17330- 31 ¶ 35. 425 AT& T Comments at 13- 14. 426 Comcast Comments at 25- 26. See Communications Act §§ 614- 615, 47 U. S. C. §§ 534- 535. 427 Writers Guild of America Comments at 5. 428 See Time Warner II, 240 F. 3d at 1135. 65 Federal Communications Commission FCC 05- 96 66 127. We seek comment on whether and to what extent the independent decisions of cable operators regarding carriage of new networks should be considered, and how the actions of independent cable operators, not acting in overt or tacit collusion, could cause harm to the market and to independent programmers. We seek comment on the ability of cable operators to identify networks that will be successful, and the cost to programmers and to consumers of cable operator errors in predicting the value of new networks. We also request information on whether the existence of two powerful, incumbent DBS operators affects these relationships. 429 (4) The Impact of Vertical Integration 128. In the 2001 Further Notice we asked whether large cable operators with programming interests would have an incentive to unfairly favor affiliated programming over unaffiliated programming, and whether they could withhold their affiliated programming from competitors in order to disadvantage or prevent entry by competing MVPDs, such as cable overbuilders. 430 We also asked if they could use their size to gain large programming license fee discounts and exclusive contracts with nonaffiliated programming, and whether this would harm rival MVPDs, lessen competition, and reduce the flow of programming to consumers. 431 We sought comment and empirical evidence on whether these problems have occurred in the past or are likely to occur if cable operators are not constrained by an ownership limit. 432 As discussed below, we find the studies and analysis submitted in the record on the issue of vertical foreclosure to be insufficient evidence to support a particular horizontal limit on subscribership, and we seek further comment and empirical evidence on the likelihood of vertical foreclosure and the ability of a horizontal limit to reduce that likelihood. 129. We seek comment and evidence on whether a large cable operator that reaches a threshold size will have the incentive and ability to engage in consumer foreclosure. We further ask whether an open field approach, such as that employed by the Commission in the 1999 Cable Ownership Order, in conjunction with our program access rules, 433 would be sufficient to ensure that a large cable operator would not be able to successfully engage in vertical foreclosure. (a) Empirical Studies of Foreclosure 130. In response to the 2001 Further Notice, empirical studies were submitted to the Commission that examined whether vertically integrated cable operators have favored their affiliated programming services and are likely to do so in the future. 434 CFA alleges that large cable operators 429 Several independent networks, such as CSTV, NFL Network and Reality TV, secured their first distribution deals on DBS systems before securing distribution on cable systems. 430 2001 Further Notice, 16 FCC Rcd at 17328- 29 ¶ 29. 431 Id. at 17329 ¶ 30. 432 Id. 433 See 47 C. F. R. §§ 76.1000- 1003. 434 AT& T Comments, Besen Decl. at 10- 14; Time Warner Comments, Joskow and McLaughlin Decl. at 2- 7. 66 Federal Communications Commission FCC 05- 96 67 already engage in foreclosure. CFA cites several examples of alleged abuses. 435 The specific allegations of abuses that CFA raises, however, are either anecdotal or unsubstantiated. Most of the anecdotal examples occurred several years or even decades ago. CFA also points to two studies which examined the effects of vertical integration on the carriage of cable programming, and found that vertically integrated cable operators may have favored their affiliated programming services in the past. For example, using an econometric model of the cable TV industry, Chipty found that vertically integrated cable operators tend to exclude programming services provided by their rivals. 436 Similarly, Waterman and Weiss empirically examined the effects of vertical integration in the cable industry. They found that vertically integrated cable operators tend to favor the programming providers with which they have an ownership affiliation. 437 They also found that increasing channel capacity reduces, but does not eliminate, this tendency. 438 131. Although these academic papers indicate that some foreclosure may have occurred in the past, they use data from a time when channel capacity was more constrained. For instance, Waterman and Weiss chose to examine cable systems with 54 or more channels separately, because this represented the state- of- the art technology at the time, and found that systems with more channels carried more networks, including rivals to affiliated programming. 439 Today, most cable subscribers have access to more than 54 channels, and consumers purchasing digital tiers often have access to over 100 channels of programming. Given the Waterman and Weiss finding that the tendency to favor affiliated programming diminished with increased channel capacity, combined with the increase of channel capacity since their study was performed, we ask whether it is possible to conclude that the behavior they observed is likely to continue. The data used by Chipty are also quite old, covering 1991, and examine home shopping networks, which may present cable operators with different financial incentives than other types of networks. 440 In addition, the significant increase in retail competition from DBS could raise the cost to cable operators of favoring affiliated networks, and thus act as a deterrent to a policy of foreclosure. 441 Since the industry has undergone tremendous change since these studies were performed, we tentatively conclude that these studies are of little probative value in our analysis. 132. Cable operators have submitted studies that purport to show that they have no theoretical incentive to favor affiliated programming networks and not carry attractive unaffiliated programming 435 CFA reports examples where MVPDs were denied access to New England Cable News and TVLand by AT& T Broadband’s Headend in the Sky. CFA Comments at 128- 29. 436 Tasneem Chipty, Vertical Integration, Market Foreclosure, and Consumer Welfare in the Cable Television Industry, 91 THE AMERICAN ECONOMIC REVIEW 3, 428- 53 (Jun. 2001) (Tasneem Chipty, Vertical Integration). 437 Waterman and Weiss at 101. Waterman and Weiss examine both premium networks and basic cable networks. 438 Id. 439 Waterman and Weiss found that as capacity expands, vertically integrated systems tend to increase the carriage of all networks, including those of rival, unaffiliated networks. Waterman and Weiss at 93, 100- 01. 440 See Chipty at 429, 432- 33, 436- 39. 441 See AT& T Comments at 51; NCTA Comments at 13- 14. We discuss CFA’s argument that DBS does not act as a constraint on cable operators’ behavior in ¶ 109 and 110, supra. 67 Federal Communications Commission FCC 05- 96 68 networks; 442 that programmers could use alternative distribution channels (such as broadcast TV, foreign MVPDs, and DVD sales) if a cable operator attempted to foreclose rival networks; 443 that larger cable operators have tended to have more channel capacity and carry more channels; 444 that cable operators have not engaged in foreclosure in the past, and there has been plentiful entry; 445 and that a cable operator’s incentive to foreclose shrinks as its size increases. 446 They argue that this evidence demonstrates that an increase in cable concentration will not increase the likelihood of foreclosure and reduce the flow of programming. 447 133. For a number of reasons, we tentatively conclude that these studies fail to prove that future increases in cable concentration will not increase the incentives and ability of vertically integrated cable operators to engage in vertical foreclosure. First, much of the evidence presented on past entry looks solely at aggregate data. If, indeed, the programming network market is segmented according to genre and type of programming, then a policy of vertical foreclosure might only be profitable in particular submarkets. Second, evidence that cable operators have not engaged in foreclosure in the past does not prove that they will not do so in the future, especially if they are still growing in size. The Commission has previously found, 448 and the cable operators’ evidence does not refute, that only if the cable operators exceed a particular threshold, will a policy of foreclosure likely be successful. 449 134. Third, the argument that the cable operator’s incentive to foreclose shrinks as the cable operator grows in size, which is integral to AT& T’s analysis, 450 fails to take into account the key point that the cable operator’s ability to successfully foreclose rival programming networks grows with each increase in subscriber reach. If the likelihood of engaging in foreclosure depends not just on the benefit if successful, but also on the likelihood of success, then an increase in size may make a policy of foreclosure viable where such a policy was previously unprofitable. 442 AT& T Comments, Besen Decl. at 6- 8; AT& T Comments, Ordover Decl. at 48- 52. 443 AT& T Comments, Ordover Decl. at 52- 65. Ordover focuses his analysis on program developers’ ability to find outlets to distribute their programming, and not on the ability of a new programming network to enter the market. 444 Time Warner Comments, Joskow and McLaughlin Decl. at 5- 6. 445 Id. at 2- 4; AT& T Comments, Besen Decl. at 10- 14. 446 AT& T Comments, Besen Decl. at 14- 20; AT& T Comments, Ordover Decl. at 48- 52. 447 AT& T Comments, Besen Decl. at 14- 21; AT& T Comments, Ordover Decl. at 48- 68. 448 1999 Cable Ownership Order, 14 FCC Rcd at 19119 ¶ 55. 449 Since there is no data on the behavior of domestic cable operators that exceed the Commission’s limits, because it has not happened yet, it appears that empirical evidence can say little about the effect of allowing operators to grow larger than their current size. 450 AT& T Comments, Besen Decl. at 14- 20; AT& T Comments, Ordover Decl. at 48- 52. 68 Federal Communications Commission FCC 05- 96 69 135. Fourth, while alternative distribution channels do exist, it is not certain that these channels are available to a new programmer that is entering the market, 451 nor that they generate the kinds of revenue necessary to support high quality original programming. 452 Some of these distribution channels may not be appropriate for serial programming such as a TV series, or programming designed for a particular genre or niche. We ask for more evidence that these alternative distribution channels are available to the kinds of new programming found on cable TV, and will provide sufficient revenues to provide a means of entering the market. We also ask whether a programming network could make use of these alternative distribution channels for distributing its regular programming, as opposed to a program producer attempting to distribute a single piece of programming, such as a movie. 136. We find that cable operators potentially have an incentive to engage in vertical foreclosure, and that the evidence presented about their past behavior does not rule out the possibility that a cable operator of larger size could, in the future, have the incentive and ability to discriminate against or foreclose an unaffiliated network. We seek comment on independent analyses that have been performed on this issue since the close of the comment period in the 2001 Further Notice. 453 Studies submitted by commenters should be based on current technological and market conditions. Studies should predict the likelihood of vertical foreclosure if there were growth in industry concentration. The changes in both cable operators’ incentive and ability to engage in vertical foreclosure should be taken into account in any studies. 3. Potential Benefits of Horizontal Concentration 137. In the 2001 Further Notice, we asked about the benefits of horizontal concentration, such as economies of scale, development of new programming, digital deployment, and investment in non- video services. 454 Some commenters have claimed that concentration would bring such benefits. 455 They have not attempted to quantify these benefits or otherwise substantiate their claims in any meaningful fashion. We have no evidence on the record that would help us identify these benefits or evaluate them at concentrations higher than those that exist today. Further, many of the purported benefits are emerging at current levels of concentration. 456 Therefore, although we discuss some theoretical benefits of concentration below, at this point we have received no conclusive evidence that additional concentration is necessary to produce these benefits. 451 Some channels, such as DVD sales and overseas markets, may be open mostly to programming that has proven itself through an established early window channel. Distribution on a programming network is important not just for generating immediate revenues, but also for advertising the programming, by creating a reputation for the program. This generates further revenues in DVD sales and overseas showings. 452 See ¶ 82, supra. 453 See, e. g., Michael E. Clements and Amy D. Abramowitz, Ownership Affiliation and the Programming Decisions of Cable Operators, available at http:// web. si. umich. edu/ tprc/ papers/ 2004/ 289/ TPRC2004. pdf and General Accounting Office, Issues Related to Competition and Subscriber Rates in the Cable Television Industry (Oct. 2003). 454 2001 Further Notice, 16 FCC Rcd at 17331- 32, ¶¶ 36- 40. 455 See, e. g., AT& T Comments at 69- 70. 456 See, e. g., 11 th Annual Report, 20 FCC Rcd at 2779- 92 ¶¶ 36- 52; AT& T Comments at 24. 69 Federal Communications Commission FCC 05- 96 70 138. Commenters argue that the largest operator in a concentrated market may enjoy efficiencies as a result of economies of size and scale. 457 The fixed costs of providing service can be spread over a larger customer base. One study referenced by commenters suggests that cable operator growth is due to increased efficiencies, and that bargaining power does not increase with size. 458 Even if a cable operator’s bargaining power does increase with size, the operator may pass some of its savings on to consumers in the form of lower rates (or smaller rate increases). Another study suggests that large operators do pass a small percentage of their programming cost savings onto consumers. 459 The 2001 Price Survey, however, found the opposite, that cable rates increase with cable operator size. 460 Other studies have reached the same conclusion. 461 139. AT& T suggests that cable ownership rules could impede cable operators from gaining the scale necessary to offer high- speed Internet, digital cable, and telephony services, potential benefits to consumers. 462 High- speed Internet and digital cable services, however, have been deployed rapidly throughout the country, by large and small cable operators alike, 463 and AT& T offers no evidence that speed of deployment would increase with increased industry concentration. NCTA reports that 91% of households passed by cable now have access to cable advanced services. 464 We tentatively conclude that further concentration is not necessary to speed development and delivery of these services. We seek comment on this tentative conclusion. We also seek comment on the relevance of the deployment of high- speed Internet and telephony services to this proceeding, since they generally do not involve the goals specified in Section 613( f)( 2), in particular that cable operators do not unfairly impede the flow of programming to consumers, and do not favor affiliated programming or unreasonably restrict the flow of affiliated programming to other video distributors. 140. Commenters argue that high levels of concentration may provide direct benefits to programmers, in particular by better enabling programmers to recover their costs. 465 Programming involves high fixed costs to produce, and low marginal costs for distribution. Uncertainty about whether the programmer will gain sufficient carriage to recover its fixed costs can act as a barrier to entry. Time 457 See Time Warner Comments, Joskow and McLaughlin Decl. at 14- 16. 458 Chipty and Snyder at 326; Time Warner Comments, Joskow and McLaughlin Decl. at 15. 459 George S. Ford and John D. Jackson, Horizontal Concentration and Vertical Integration in the Cable Television Industry, REVIEW OF INDUSTRIAL ORGANIZATION, Vol. 12 (1997). 460 See 2001 Price Survey, 17 FCC Rcd at 6318 ¶ 45. 461 See General Accounting Office, The Effects of Competition from Satellite Providers on Cable Rates (Jul. 2000); W. M. Emmons and R. A. Prager, The Effects of Market Structure and Ownership on Prices and Service Offerings in the U. S. Cable Television Industry, RAND JOURNAL 732- 50 (Winter 1997). 462 AT& T Comments, Ordover Decl. at 17- 18, 70- 73. 463 10 th Annual Report, 19 FCC Rcd at 1636- 48 ¶¶ 39- 60. 464 National Cable & Telecommunications Association 2004 Year- End Industry Overview at 5. 465 Time Warner Comments, Joskow and McLaughlin Decl. at 14- 15. 70 Federal Communications Commission FCC 05- 96 71 Warner points out that a carriage commitment from a large cable operator can reduce this uncertainty, and make entry easier. 466 141. Commenters also argue that increasing concentration can help solve the potential problem of multiple small cable operators attempting to free ride on the payments made by the other cable operators, in which each cable operator forces down the price it pays to a level that fails to cover an adequate share of the fixed costs. 467 They state that each cable operator would prefer to pay just the marginal cost of providing the programming, and let the other buyers pay for the fixed costs of producing the programming. If, instead, there were a single buyer that was large enough that its purchasing decision would affect the viability of a programmer, then it would have to consider the effects of the price it demands on the financial viability of the programmer, and hence the likelihood the programmer will stay in the market. According to this view, a large “pivotal” buyer will be less likely to demand discounts that threaten the viability of the programmer. 468 142. The realization of this potential benefit, however, depends upon several factors that are not likely to occur in practice. In a highly concentrated industry, operators may demand large discounts from programming networks because of the market power they enjoy. Because MVPDs depend upon programmers for content, even a monopsonist MVPD would not generally want to demand prices so low that they harm programming networks’ ability to provide programming. In order to ensure that programming networks receive sufficient payment to cover their fixed costs, however, the operator would have to have an intimate knowledge of the cost structure of particular networks, which is unlikely in practice. As a result, even a pivotal buyer might unwittingly force video programming networks to accept compensation that does not cover all of their relevant costs, thus reducing their ability to provide high quality programming or, possibly, forcing some out of business. 469 The argument also assumes that buyers want to ensure the financial viability of their suppliers. Large buyers, however, may decide that pursuing a policy of forcing prices down is more profitable, because the resulting reduction in purchasing costs outweighs the loss of some higher- cost programmers that may be forced to exit the business. Being a pivotal buyer may also give the cable operator an incentive to vertically integrate and favor its affiliated programming networks, since it can ensure that no competing programming networks can enter the market. For these reasons, we tentatively conclude that commenters have not demonstrated that allowing a cable operator to become large enough to become a pivotal buyer will improve the flow of programming, and should therefore be counted as a benefit of increased horizontal concentration. D. Vertical Limit 143. Section 613( f) of the Communications Act directs the Commission to “prescribe rules and regulations establishing reasonable limits on the number of channels on a cable system that can be 466 Id. 467 Raskovich Comments; Raskovich, Pivotal Buyers and Bargaining Position; Time Warner Comments, Joskow and McLaughlin Decl. at 15- 16; AT& T Comments, Ordover Decl. at 39- 45. 468 Id. 469 Offering higher payments to keep just the high cost networks in business is not likely to solve the problem here, because cable operators may not know which networks are high cost. If they offered higher payments to all networks that are high cost, all networks would claim to be high cost. 71 Federal Communications Commission FCC 05- 96 72 occupied by a video programmer in which a cable operator has an attributable interest.” 470 Among other things, in setting the limit, the Commission is directed to “ensure that cable operators affiliated with video programmers do not favor such programmers in determining carriage . . .” 471 and to refrain from “impos[ ing] limitations which would impair the development of diverse and high quality video programming.” 472 In 1993 the Commission found that a 40% limit on the number of activated channels that can be occupied by affiliated video programming services struck an appropriate balance between the goals of reducing the incentive and ability of vertically integrated cable operators to favor their affiliated programming, increasing diversity, and permitting cable operators to realize the benefits and efficiencies associated with vertical integration. 473 144. The Time Warner II decision reversed and remanded the 40% channel occupancy limit, finding that the Commission had failed to justify its vertical limit with record evidence, and had failed to adequately consider the benefits and harms of vertical integration or current MVPD market conditions in its analysis. 474 In the 2001 Further Notice, the Commission sought comment on how it could fashion meaningful and relevant channel occupancy limits given the changes that have occurred in the MVPD industry. 475 The 2001 Further Notice requested comment on the economic underpinnings of the statutory requirement, and asked commenters to address the economic basis underlying the concern with vertical integration and market foreclosure. 476 145. In response to the 2001 Further Notice, several commenters assert that the Commission should not adopt any channel occupancy rules and should not limit the carriage of affiliated programming. 477 Cablevision argues that given the technological advancements and today’s “vigorously competitive” MVPD marketplace, no channel occupancy limit will survive constitutional scrutiny. 478 NCTA contends that competition in the sale of video programming has effectively eliminated incentives to discriminate, and that if a cable operator refuses to carry attractive programming services, it will not only fail to attract subscribers and fail to maximize revenue from existing subscribers, it may lose subscribers. 479 Other commenters, however, assert that horizontal concentration and vertical integration 470 See 47 U. S. C. § 533( f)( 1)( A)-( B). 471 47 U. S. C. § 533( f)( 2)( B). 472 47 U. S. C. § 533( f)( 2)( G). The Commission is also directed to consider the other public interest objectives listed in Section 613( f)( 2). See 47 U. S. C. § 533( f)( 2)( A), (C)-( F). 473 1993 Second Report and Order, 8 FCC Rcd at 8593- 95 ¶ 68. 474 Time Warner II, 240 F. 3d at 1137- 39. 475 2001 Further Notice, 16 FCC Rcd at 17350- 51 ¶ 81. 476 Id. 477 See Cablevision Comments at 5- 11; Comcast Comments at 29- 33; NCTA Comments at 20- 23; Time Warner Comments at 35- 37. 478 Cablevision Comments at 5. 479 NCTA Comments at 11, 14. 72 Federal Communications Commission FCC 05- 96 73 in the MVPD industry require that the Commission enact and enforce a strict channel occupancy limit. 480 CFA argues that vertical integration of cable firms facilitates the imposition of higher costs on programming rivals or a degradation in their quality of service (by withholding desired programming) to gain an advantage. 481 Writer’s Guild contends that the Commission should not only retain the existing 40% channel occupancy limit, but should strengthen it through ownership limits on both cable and broadcast networks, regardless of whether the owner is a cable operator. 482 146. Both Congress and the Commission have long recognized that vertical integration produces efficiencies in the production, distribution, and marketing of video programming, enabling cable operators to make additional investments in both distribution plant and programming. 483 Congress and the Commission, on the other hand, also have been concerned that such integration may provide an incentive for cable operators to engage in strategic, anticompetitive behavior. 484 The economics literature provides support for both propositions: vertical integration between programmers and MVPDs can result in both efficiency gains (which can lower prices) and market foreclosure (which can lead to higher prices). 485 While the public interest objectives enumerated in Section 613( f)( 2)( A)-( G) direct that we take account of the risks and benefits of vertical integration in the cable industry together with prevailing market conditions in choosing what limit is “reasonable,” the record before us provides no new evidentiary support or metrics with which to better calculate a limit that is “reasonable” in today’s marketplace. None of the comments filed in response to the 2001 Further Notice yielded a sound evidentiary basis for either retaining the current 40% vertical limit or for setting a different limit. Nonetheless, we disagree with commenters who assert that the Commission should not adopt any channel occupancy rules and should not limit carriage of affiliated programming. 486 147. The statute expressly directs the Commission to conduct a proceeding and “to prescribe rules and regulations establishing reasonable limits on the number of channels on a cable system that can 480 See CFA Comments at 93- 105; Writer’s Guild Comments at 15. 481 CFA Comments at 96- 97. 482 Writer’s Guild Comments at 15. 483 See Senate Report at 26- 27, 81; House Report at 41; 1995 Vertical Reconsideration Order, 10 FCC Rcd at 7365- 66 ¶¶ 5- 6; 1993 Second Report and Order, 8 FCC Rcd at 8584- 85 ¶¶ 43- 44; Initial Notice, 8 FCC Rcd at 218- 19 ¶¶ 44- 45. 484 See Senate Report at 25- 27, 81; House Report at 41; 1995 Vertical Reconsideration Order, 10 FCC Rcd at 7365 ¶ 4; 1993 Second Report and Order, 8 FCC Rcd at 8583- 84 ¶¶ 41- 42; Initial Notice, 8 FCC Rcd at 218 ¶¶ 42- 43. Cf. generally Program Access Order, 17 FCC Rcd at 12135- 50 ¶¶ 24- 55 (discussing ability and incentive of vertically integrated programming networks to favor affiliated cable operators). 485 See Tasneem Chipty, Vertical Integration, supra. Using an econometric model of the cable industry, Chipty found that the harmful effects of integration due to foreclosure are offset by the efficiency- enhancing effects of integration. See also Michael E. Clements and Amy D. Abramowitz, Ownership Affiliation and the Programming Decisions of Cable Operators, available at http:// web. si. umich. edu/ tprc/ papers/ 2004/ 289/ TPRC2004. pdf. 486 See Cablevision Comments at 5- 11; Comcast Comments at 29- 33; NCTA Comments at 20- 23; Time Warner Comments at 35- 37. 73 Federal Communications Commission FCC 05- 96 74 be occupied by a video programmer in which a cable operator has an attributable interest.” 487 Further, in the examination of the scope of our legal authority under Section 613( f), we found that the Commission lacks express authority under Title VI to forbear from the implementation and enforcement of its provisions. 488 Thus, we are bound to follow Congress’ statutory directive that a vertical limit be set, and the challenge in implementing Section 613( f)( 1)( B) in light of Time Warner II remains one of finding and adequately justifying a reasonable numerical limit that permits cable operators to enjoy the benefits of vertical integration, protects against any potential harms of discrimination against rival programming that may exist, and takes account of the vastly changed technological and competitive landscape that characterizes today’s MVPD marketplace, while not burdening substantially more speech than necessary. 489 We again request comment and empirical and theoretical evidence to assist in the development of reasonable limits and in the articulation of how such limits address the statutory goals. 1. Defining the Market 148. We seek comment on how to define the programming and distribution markets for the purposes of determining an appropriate limit on channel occupancy by vertically integrated cable operators. In the 2001 Further Notice we proposed that programming could be classified into two broad categories, general entertainment and niche programming. 490 We also suggested that programming networks vary according to whether they focus on a particular subject or are more general purpose, whether they gain a large nationwide audience, how narrowly focused they are in a particular subject, and whether they are national or regional in scope. 491 We received little comment on whether these differences in the types of programming networks affect a cable operator’s incentive and ability to engage in vertical foreclosure. As we discuss above, we ask whether the market for programming should be segmented according to the type of programming network involved. Could the incentive and ability of cable operators to engage in vertical foreclosure vary according to the type of programming network? We note that a channel occupancy limit only ensures that cable operators carry some unaffiliated programming networks. It does not prevent a cable operator from discriminating against any specific programming network. If we were to determine that the incentive and ability for a cable operator to discriminate varies according to the submarket involved, how could a channel occupancy limit prevent discrimination against rival programming networks? 149. We also seek comment on whether placement of networks on different tiers affects how vertical foreclosure might be implemented by a cable operator, and whether our rules should be applied on a tier- specific basis. Networks are often placed on different tiers, or in different packages of programming made available to consumers. Also, cable operators typically have a much greater channel capacity on their digital tiers, but fewer customers have access to this tier, compared to the analog portions of their network. We ask whether our analysis should take into consideration the existence of 487 See ¶ 48, supra. 488 Id. 489 See Time Warner II, 240 F. 3d at 1137. 490 2001 Further Notice, 16 FCC Rcd at 17321 ¶ 9. 491 Id. at 17322- 23 ¶¶ 12- 13. 74 Federal Communications Commission FCC 05- 96 75 tiers and packages, with their differences in technical characteristics, numbers of channels, and pricing. We also ask whether a vertical limit should be applied on a tier- specific or package- specific basis. 2. Potential Harms of Vertical Integration 150. In the 2001 Further Notice we asked commenters to “address the economic basis underlying the concern with vertical integration and market foreclosure.” We asked whether the necessary conditions existed in the MVPD industry for cable operators to profitably engage in vertical foreclosure, and for this foreclosure to be harmful to the flow of programming. 492 We also sought comment on whether current and likely future developments in the MVPD market will mitigate past concerns regarding the ability of cable operators to discriminate against unaffiliated programming networks. 493 151. We discussed above how vertical integration can create an incentive for a large vertically integrated cable operator to engage in foreclosure, by not carrying a rival programming service that competes with its affiliated programming service. 494 We also discussed the types and causes of vertical foreclosure, and the harms that this can cause. 495 152. In their responses to the 2001 Further Notice, cable operators point to market factors that make vertical foreclosure unlikely. First, they state that a programmer can obtain carriage despite a cable operator’s preference not to carry the programmer’s service under several scenarios: 496 (1) where the programmer is seeking carriage of a broadcast network entitled to “must carry” status under the Commission’s rules; 497 (2) where the programmer is seeking carriage of a “must have” programming network that consumers demand; and (3) where the programmer is seeking carriage of a service pursuant to the Commission’s leased access rules. 498 Second, they assert that discrimination on the basis of affiliation is already targeted by the program access rules. 499 Third, they argue that competition from alternative MVPDs such as DBS makes it unprofitable for a cable operator to engage in foreclosure, since failure to carry unaffiliated popular networks will drive customers to other MVPDs. 500 Lastly, they argue that market conditions have changed to make foreclosure unlikely, citing in particular the increase in channel capacity of cable systems. 501 492 2001 Further Notice, 16 FCC Rcd at 17350- 51 ¶ 81. 493 Id. at 17351- 52 ¶ 83. 494 See Section II. C. 2. c. (4), supra. 495 Id. 496 AT& T Comments at 50- 51; Comcast Comments at 25- 28. 497 47 C. F. R. § 76.56. 498 47 C. F. R. § 76.701. 499 Time Warner Comments at 35- 37 (citing 47 U. S. C. § 536( a)( 3); 47 C. F. R. §76.1301( c)). 500 Cablevision Comments at 7- 10; NCTA Comments at 21. 501 Cablevision Comments at 7- 9. 75 Federal Communications Commission FCC 05- 96 76 153. Nonetheless, the terms of Section 613( f)( 1)( B) require that we establish reasonable limits on channel occupancy, and we therefore again seek empirical, theoretical and anecdotal evidence to support our effort to carry out this statutory mandate. 3. Potential Benefits of Vertical Integration 154. In the 2001 Further Notice we asked commenters to discuss the benefits of vertical integration, and the extent to which these benefits mitigate or outweigh the harms caused by cable operators favoring affiliated programming. 502 We asked how these benefits should affect the fashioning of vertical limits and restraints. We sought comment on what impact relaxing or modifying the current limit of 40% might have on producing economic efficiencies, fostering innovation in services, and encouraging greater investment in and development of diverse and responsive programming. 503 We also asked whether the existence of these benefits means that we should employ alternative regulatory restrictions, other than imposing limits on cable operators’ carriage of affiliated programming, to prevent foreclosure. 504 155. In response, cable commenters argued that vertical integration provides efficiencies, by increasing the likelihood of financing for new networks and reducing the likelihood of “hold- up.” 505 They also argue that it eliminates the problem of double marginalization, which occurs when both upstream and downstream firms attempt to exercise market power by charging above- cost prices. 506 Commenters failed, however, to demonstrate that the benefits of vertical integration will always exceed the potential harms from vertical foreclosure. They also failed to identify those circumstances in which the benefits from a particular vertical investment or merger, for example a cable operator investment intended to create a new programming network in an underserved market niche, are large enough to warrant exemption from the vertical limits. We thus seek further comment on whether and when the benefits of vertical integration mitigate the potential harms that might result, either generally or for particular vertical combinations. 156. The literature indicates that historically content providers have received benefits from vertical integration with distributors. 507 In the multichannel video programming industry, three kinds of benefits can result from vertical integration: transaction efficiencies, enhanced availability of capital and creative resources, and risk reduction through signaling commitment. 508 We examine each below. 502 2001 Further Notice, 16 FCC Rcd at 17351 ¶ 82. 503 Id. at 17352 ¶ 84. 504 Id. at 17351 ¶ 82. 505 Time Warner Comments, Joskow and McLaughlin Decl. at 22. (For a discussion of the problem of “hold- up,” see ¶ 94, supra.) 506 Id. at 23. 507 See Waterman and Weiss at 45- 54. 508 Id. at 47- 49. 76 Federal Communications Commission FCC 05- 96 77 157. Transaction Efficiency. Vertical integration may increase transaction efficiency by allowing more efficient contracting between entities. An affiliation agreement between cable operators and programming networks may reduce the incentive of each to engage in post- contractual opportunistic behavior. 509 Such opportunistic behavior is especially likely to occur if market conditions are likely to change. It can also reduce the inefficiency of both parties attempting to exercise market power by charging prices above cost, in what is often called “double marginalization.” 510 Each of these inefficiencies will result in higher consumer prices, and eliminating the inefficiencies therefore can lower prices. 158. Resources. Vertical integration makes available to programming networks the capital resources of large cable operators. Since only the large cable operators tend to be vertically integrated, this can offer programmers significant financial resources, potentially facilitating the development of a greater quantity of programming and higher quality programming. This is particularly important to new networks, which usually incur losses for the first several years of operation. 511 In addition, vertical relationships allow programming networks access to creative and management resources at the cable operators, which may increase the efficiency of operation at the networks. Finally, cable operators have a direct relationship with subscribers, and access to this relationship through vertical integration may give programmers better knowledge about consumer demand. 159. Signaling Commitment. Signaling commitment is the commitment to carry a network, perhaps even before it launches. Vertical integration creates at least the appearance of signaling commitment, which may allow new programmers access to capital from sources other than the affiliated MVPD and the ability to acquire talent and content. Absence of signaling commitment may cause networks to exit the market, or never to enter the market in the first place. Thus, vertical integration may reduce the risk of failure for new networks, thereby increasing program diversity. 509 An example of opportunistic behavior on the part of the cable operator would be failure to promote an unaffiliated programming network. Opportunistic behavior by a programmer might involve reducing quality or providing content of a different sort than what was promised (for example, promising to provide programming for a niche that was previously unserved, and then switching to a genre with a larger audience). In addition, one of the parties could in the future attempt to take advantage of the other party’s committed sunk costs by demanding a more advantageous price, which would reduce the incentive of the other party to commit to the transaction (i. e., the “hold- up problem”). Oliver Williamson, MARKETS AND HIERARCHIES: ANALYSIS AND ANTITRUST IMPLICATIONS (1975). See ¶ 94, supra. 510 Double marginalization occurs when both the license fee that a programmer charges a cable operator, and the retail rate charged by the cable operator to subscribers, are set above cost. This will cause the operator to set its subscription rate higher than the most efficient level, thus reducing subscription levels and reducing total revenue to both the operator and the programmer. Joint ownership is one means of eliminating this problem. See Waterman and Weiss at 48- 49. 511 One example of this is the Radio One- Comcast partnership to launch a network to compete with Black Entertainment Television. Krissah Williams, Radio One, Comcast in Cable Deal, WASHINGTON POST, Jan. 13, 2003, at A1.; see also Krissah Williams, Comcast Alliance May Be Key to Cable Channel’s Success, WASHINGTON POST, Jan. 15, 2003, at E1. 77 Federal Communications Commission FCC 05- 96 78 160. Vertical integration can be particularly beneficial in the development and launch of local and regional programming. 512 Incumbent cable operators, with their direct connection to local communities, experience in ownership of programming, and large financial resources, may be uniquely positioned to develop and distribute local and regional programming. For instance, in response to the Commission’s 2002 Competition Report Notice of Inquiry, AT& T claimed that with increased clustering, it was able to develop and offer more local and regional programming to consumers. 513 Additionally, upgrades to cable systems are allowing operators to expand channel capacity, which facilitates inclusion of local and regional programming in cable system offerings. 161. We lack record evidence concerning the actual benefits of vertical integration, and we seek comment on whether there are benefits from vertical integration, and in particular whether and how much vertical integration has increased the flow of programming. We ask whether there are means of directly measuring these benefits, such as the added resources gained by programmers from vertical integration, and the reduction in double- marginalization. We also ask what metrics to use to measure the resulting benefits to consumers, such as the increase in programming. We ask what information and data would be needed to calculate these benefits. 162. In sum, on the record before us, we lack the evidence necessary under Time Warner II to establish a firm channel occupancy limit that would both preserve the benefits of vertical integration and protect against potential harms without unduly burdening cable operators’ First Amendment rights. We seek additional evidence or suggested approaches that would support a specific limit in light of current market conditions, consistent with the statutory obligation to establish a reasonable limit on the number of channels that a cable operator may occupy with video programming in which the operator has an attributable interest. E. Diversity of Information Sources 163. Section 612( g) of the Communications Act provides that at such time as cable systems with 36 or more activated channels are available to 70% of households within the United States and are subscribed to by 70% of those households, the Commission may promulgate any additional rules necessary to promote diversity of information sources. 514 In its Eleventh Annual Report, the Commission surveyed available data sources to determine whether this threshold has been met. 515 The Commission found that cable systems with 36 or more channels are available to 79.8% (84,415,707 ÷ 105,842,000) of 512 See 9 th Annual Report, 17 FCC Rcd at 26957 ¶ 128; 2002 Program Access Order 17 FCC Rcd at 12132, 12148- 49 ¶¶ 19, 54 (most regional programming networks are vertically integrated). 513 See 9 th Annual Report, 17 FCC Rcd at 26956 ¶ 127. MVPD competitors responded in the record that such clustering was allowing anticompetitive behavior by cable incumbents, such as migrating local and regional programming to terrestrial delivery so that the incumbent could deny downstream rivals access to the programming. Id. at 26956- 57 ¶ 128. 514 47 U. S. C. § 532( g). 515 See 11 th Annual Report, 20 FCC Rcd at 2767- 68 ¶ 20. 78 Federal Communications Commission FCC 05- 96 81 168. Contact Information. The Media Bureau contact for this proceeding is Barbara Esbin or Patrick Webre at (202) 418- 7200. Press inquiries should be directed to Rebecca Fisher at (202) 418- 2359, TTY: (202) 418- 7365 or (888) 835- 5322. 81 Federal Communications Commission FCC 05- 96 82 IV. ORDERING CLAUSES 169. Accordingly, IT IS ORDERED, that pursuant to authority contained in sections 2( a), 4( i), 303, 307, 309, 310, and 613 of the Communications Act of 1934, as amended, 47 U. S. C. §§ 152( a), 154( i), 303, 307, 309, 310, and 533, this Second Further Notice of Proposed Rulemaking IS ADOPTED. 170. IT IS FURTHER ORDERED that, pursuant to the authority contained in sections 2( a), 4( i), 303, 307, 309, 310, and 613 of the Communications Act of 1934, as amended, 47 U. S. C. §§ 152( a), 154( i), 303, 307, 309, 310, and 533, NOTICE IS HEREBY GIVEN of the proposals described in this Second Further Notice of Proposed Rulemaking. 171. IT IS FURTHER ORDERED that the Commission’s Consumer and Governmental Affairs Bureau, Reference Information Center, SHALL SEND a copy of this Second Further Notice of Proposed Rulemaking, including the Initial Regulatory Flexibility Analysis, to the Chief Counsel for Advocacy of the Small Business Administration. FEDERAL COMMUNICATIONS COMMISSION Marlene H. Dortch Secretary 82 Federal Communications Commission FCC 05- 96 83 APPENDIX A LIST OF COMMENTERS INITIAL COMMENTS Alexander Raskovich, Economist Altrio Communications, Inc. et al. (“ CMVPDs”) [jointly filed on behalf of Altrio Communications, Inc., BellSouth Entertainment, LLC, the Independent Multi- Family Communications Council, Qwest Broadband Services, Inc., the Wireless Communications Association International, Inc.] AT& T Broadband (“ AT& T”) Broadband Service Providers Association (“ BSPA”) Cablevision Systems Corporation (“ Cablevision”) Comcast Corporation (“ Comcast”) Concerned Consumers* Consumer Federation of America, et al. (“ CFA”) [jointly filed by the following: Alliance for Community Media, Association for Independent Video and Filmmakers, Center for Digital Democracy, Consumer Federation of America, Consumers Union, Media Access Project, National Alliance for Media Arts and Culture, and the United Church of Christ, Inc.] National Cable & Telecommunications Association (“ NCTA”) Media General, Inc. (“ Media General”) National Association of Broadcasters (“ NAB”) Paxson Communications Corporation (“ Paxson”) Progress & Freedom Foundation (“ PFF”) RCN Telecom Services, Inc. (“ RCN”) Sherjan Broadcasting Co., Inc. (“ Sherjan”) Time Warner Cable (“ Time Warner”) United States Conference of Catholic Bishops (“ USCCB”) Viacom, Inc. (“ Viacom”) Writer’s Guild of America (“ Writer’s Guild”) * late filed REPLY COMMENTS AT& T Broadband (“ AT& T”) Comcast Corporation (“ Comcast”) Consumer Federation of America, et al. (“ CFA”) [jointly filed by the following: Alliance for Community Media, Association for Independent Video and Filmmakers, Center for Digital Democracy, Consumer Federation of America, Consumers Union, Media Access Project, National Alliance for Media Arts and Culture, National Association of Telecommunications Officers and Advisors, and the United Church of Christ, Inc.] Fox Entertainment Group et al. (“ Fox”) [jointly filed by the following: Fox Entertainment Group, Inc., National Broadcasting Company, Inc., the Walt Disney Company, and Viacom, Inc.] Media Access Project et al. (“ MAP”) [jointly filed by the following: Center for Digital Democracy, Consumer Federation of America, Consumers Union, and Media Access Project] National Association of Broadcasters (“ NAB”) National Cable & Telecommunications Association (“ NCTA”) 83 Federal Communications Commission FCC 05- 96 84 Paxson Communications Corporation (“ Paxson”) RCN Telecom Services, Inc. (“ RCN”) Time Warner Cable (“ Time Warner”) United States Department of Justice (“ DOJ”) World Satellite Network, Inc. (“ World Satellite”) Writer’s Guild of America (“ Writer’s Guild”) INITIAL COMMENTS ON EXPERIMENTAL STUDY (“ SUPPLEMENTAL COMMENTS”) AT& T Corporation (“ AT& T”) Comcast Corporation (“ Comcast”) National Cable & Telecommunications Association (“ NCTA”) RCN Telecom Services, Inc.(“ RCN”) SBC Communications, Inc. (“ SBC”) Time Warner Cable (“ Time Warner”) REPLY COMMENTS ON EXPERIMENTAL STUDY (“ SUPPLEMENTAL REPLY COMMENTS’) AT& T Corporation (“ AT& T”) Time Warner Cable (“ Time Warner”) 84 Federal Communications Commission FCC 05- 96 85 APPENDIX B INITIAL REGULATORY FLEXIBILITY ANALYSIS 1. As required by the Regulatory Flexibility Act of 1980, as amended (RFA), 1 the Commission has prepared this Initial Regulatory Flexibility Analysis (IRFA) of the possible significant economic impact on a substantial number of small entities by the policies and rules considered in the Second Further Notice of Proposed Rulemaking (Second Further Notice). Written public comments are requested on this IRFA. Comments must be identified as responses to this IRFA and must be filed by the deadlines for comments provided on the first page of this document. The Commission will send a copy of the Second Further Notice, including this IRFA, to the Chief Counsel for Advocacy of the Small Business Administration (SBA). 2 In addition, the Second Further Notice and the IRFA (or summaries thereof) will be published in the Federal Register. 3 A. Need for, and Objectives of, the Proposed Rules 2. Section 613( f) of the Communications Act is intended, in part, to foster a diverse, robust, and competitive market in the acquisition and delivery of multichannel video programming. Specifically, Section 613( f) requires the Commission to establish reasonable limits on the number of cable subscribers that may be reached through commonly owned or attributed systems (horizontal limits) and on the number of channels that can be occupied by the cable system’s owned or attributed video programming services (vertical limits). Congress intended these limits to ensure that cable operators do not use their horizontal reach in the multichannel video distribution (MVPD) market, acting unilaterally or jointly, to unfairly impede the flow of video programming to consumers. However, Congress recognized that multiple system ownership could benefit consumers by allowing efficiencies in the administration, distribution, and procurement of programming, and by providing capital and a ready subscriber base to promote the introduction of new programming services. Pursuant to its statutory mandate, and balancing these competing interests, the Commission has adopted and periodically revised cable ownership limits. 3. The Commission first established horizontal and vertical ownership limits in 1993. 4 The horizontal limit bars cable operators from serving more than 30% of all U. S. MVPD subscribers. The vertical limit bars cable operators with 75 or fewer channels from devoting more than 40% of channel capacity to affiliated programming. In Time Warner Entertainment Co. v. FCC, 240 F. 3d 1126 (D. C. Cir. 2001) (Time Warner II), the D. C. Circuit remanded the Commission’s horizontal and vertical limits, finding that the horizontal and vertical ownership limits unduly burdened cable operators’ First Amendment rights, that the Commission’s evidentiary basis for imposing the ownership limits and its 1 See 5 U. S. C. § 603. The RFA, 5 U. S. C. § 601 et. seq., has been amended by the Small Business Regulatory Enforcement Fairness Act of 1996 (“ SBREFA”), Pub. L. No. 104- 121, Title II, 110 Stat. 847 (1996). 2 See 5 U. S. C. § 603( a). 3 Id. 4 See Implementation of Sections 11 and 13 of the Cable Television Consumer Protection and Competition Act of 1992, Horizontal and Vertical Ownership Limits, Cross- Ownership Limitations, and Anti- trafficking Provisions, 8 FCC Rcd 8565, 8567 ¶¶ 3- 4 (1993) (1993 Second Report and Order). 85 Federal Communications Commission FCC 05- 96 86 rationales supporting the vacated attribution rules did not meet the applicable standards of review, and that the Commission had failed to consider sufficiently changes that have occurred in the MVPD market since passage of the 1992 Act. The Commission thereafter issued a Further Notice of Proposed Rulemaking (2001 Further Notice) soliciting comment aimed at establishing a sound record on which to base cable horizontal and vertical limits. 5 4. None of the comments to the 2001 Further Notice yielded a sound evidentiary basis for setting horizontal or vertical limits. While many commenters presented theoretical, legal, or economic arguments and anecdotal evidence, no party provided a compelling approach that supported a particular horizontal or vertical limit. The Commission subsequently sought to augment the record by means of a programming network survey and econometric analysis, with limited results. The Commission concludes that a Second Further Notice is necessary to update the record and provide additional input on horizontal and vertical ownership limits so that the Commission may comply with the statutory mandate and the court’s directives. 5. In the Second Further Notice, the Commission seeks comment on how recent developments in the industry may affect the issues before us. Additionally, to develop a more focused and useful record, the Commission addresses the viability of proposals for setting limits suggested in the record. B. Legal Basis 6. The authority for the action proposed in this rulemaking is contained in Sections 2( a), 4( i), 303, 307, 309, 310, and 613 of the Communications Act of 1934, as amended, 47 U. S. C. §§ 152( a), 154( i), 303, 307, 309, 310, 533. C. Description and Estimate of the Number of Small Entities to Which the Proposed Rules Will Apply 7. The RFA directs agencies to provide a description and, where feasible, an estimate of the number of small entities that will be affected by the rules. 6 The RFA defines the term “small entity” as having the same meaning as the terms “small business,” “small organization,” and “small governmental jurisdiction.” 7 In addition, the term “small business” has the same meaning as the term “small business concern” under the Small Business Act, unless the Commission has developed one or more definitions that are appropriate to its activities. 8 Under the Small Business Act, a “small business concern” is one 5 See Implementation of Section 11 of the Cable Television Consumer Protection and Competition Act of 1992, 16 FCC Rcd 17312 (2001) (2001 Further Notice). The Initial Regulatory Flexibility Analysis for the 2001 Further Notice is at 16 FCC Rcd at 17360- 17369. 6 5 U. S. C. § 603( b)( 3). 7 5 U. S. C. § 601( 6). 8 Id. § 601( 3) (incorporating by reference the definition of “small business concern” in the Small Business Act, 15 U. S. C. § 632). Pursuant to 5 U. S. C. § 601( 3), the statutory definition of a small business applies “unless an agency, after consultation with the Office of Advocacy of the Small Business Administration and after opportunity for public comment, establishes one or more definitions of such term which are appropriate to the activities of the agency and publishes such definition( s) in the Federal Register.” 86 Federal Communications Commission FCC 05- 96 87 which: (1) is independently owned and operated; (2) is not dominant in its field of operation; and (3) satisfies any additional criteria established by the SBA. 9 8. Cable and Other Program Distribution. This category includes cable systems operators, closed circuit television services, direct broadcast satellite services, multipoint distribution systems, satellite master antenna systems, and subscription television services. The SBA has developed a small business size standard for this census category, which includes all such companies generating $12.5 million or less in revenue annually. 10 According to the Census Bureau data for 1997, there were 1,311 firms in this category, total, that operated for the entire year. 11 Of this total, 1,180 firms had annual receipts of under $10 million and an additional 52 firms had receipts of $10 million or more, but less than $25 million. Consequently, the Commission estimates that the majority of providers in this service category are small businesses that may be affected by the rules and policies in the Second Further Notice. We note, however, that the ownership rules at issue here apply only to cable operators, and not other MVPD providers. 9. Cable System Operators (Rate Regulation Standard). The Commission has developed, with SBA’s approval, our own definition of a small cable system operator for purposes of rate regulation. Under the Commission’s rules, a “small cable company” is one serving fewer than 400,000 subscribers nationwide. 12 We last estimated that there were 1,439 cable operators that qualified as small cable companies at the end of 1995. 13 Since then, some of those companies may have grown to serve over 400,000 subscribers, and others may have been involved in transactions that caused them to be combined with other cable operators. Consequently, we estimate that there are fewer than 1,439 small cable companies that may be affected by the adopted rules. 10. Cable System Operators (Telecom Act Standard). The Communications Act of 1934, as amended, also contains a size standard for a “small cable operator,” which is “a cable operator that, directly or through an affiliate, serves in the aggregate fewer than one percent of all subscribers in the United States and is not affiliated with any entity or entities whose gross annual revenues in the aggregate exceed $250,000,000.” 14 The Commission has determined that there are 67.7 million subscribers in the 9 15 U. S. C. § 632. 10 13 C. F. R. § 121.201, NAICS Code 517510. 11 U. S. Census Bureau, 1997. Economics and Statistics Administration, Bureau of Census, U. S. Department of Commerce, 1997 Economic Census, Subject Series – Establishment and Firm Size, Information Sector 51, Table 4 at 50 (2000). The amount of $10 million was used to estimate the number of small business firms because the relevant Census categories stopped at $9,999,999 and began at $10,000,000. No category for $12.5 million existed. Thus, the number is as accurate as it is possible to calculate with the available information. 12 47 C. F. R. § 76.901( e). The Commission developed this definition based on its determinations that a small cable company is one with annual revenues of $100 million or less. See Implementation of Sections of the Cable Television Consumer Protection and Competition Act of 1992: Rate Regulation, MM Doc. Nos. 92- 266 and 93- 215, Sixth Report and Order and Eleventh Order on Reconsideration, 10 FCC Rcd 7393, 7408- 7409 ¶¶ 28- 30 (1995). 13 Paul Kagan Assocs., Inc., Cable TV Investor, Feb. 29, 1996 (based on figures for Dec. 30, 1995). 14 47 U. S. C. § 543( m)( 2). 87 Federal Communications Commission FCC 05- 96 88 United States. 15 Therefore, an operator serving fewer than 677,000 subscribers shall be deemed a small operator, if its annual revenues, when combined with the total annual revenues of all of its affiliates, do not exceed $250 million in the aggregate. 16 Based on available data, we estimate that the number of cable operators serving 677,000 subscribers or less totals approximately 1,450. 17 The Commission neither requests nor collects information on whether cable system operators are affiliated with entities whose gross annual revenues exceed $250 million, 18 and therefore is unable at this time to estimate more accurately the number of cable system operators that would qualify as small cable operators under the size standard contained in the Communications Act. 11. Private Cable Operators (PCOs) or Satellite Master Antenna Television (SMATV) Systems. PCOs, also known as SMATV systems or private communication operators, are video distribution facilities that use closed transmission paths without using any public right- of- way. PCOs acquire video programming and distribute it via terrestrial wiring in urban and suburban multiple dwelling units such as apartments and condominiums, and commercial multiple tenant units such as hotels and office buildings. The SBA definition of small entities for cable and other program distribution services includes PCOs. 19 In 2003, there were approximately 250 PCOs operating in the United States. PCOs often serve approximately 3,000- 4,000 subscribers, but the larger operations serve as many as 15,000- 55,000 subscribers. 20 As of June 2004, PCOs served 1.1 million subscribers, down 100,000 subscribers from June 2003. 21 Because these operators are not rate regulated, they are not required to file financial data with the Commission. Furthermore, we are not aware of any privately published financial information regarding these operators. Based on the estimated number of operators and the estimated number of units served by the largest ten SMATVs, we believe that a substantial number of SMATV operators qualify as small entities. D. Description of Projected Reporting, Recordkeeping and Other Compliance Requirements 12. None proposed. 15 See FCC Announces New Subscriber Count for the Definition of Small Cable Operator, Public Notice, 16 FCC Rcd 2225 (2001). 16 47 C. F. R. § 76.901( f). 17 See FCC Announces New Subscriber Count for the Definition of Small Cable Operator, Public Notice, 16 FCC Rcd 2225 (2001). 18 The Commission does receive such information on a case- by- case basis if a cable operator appeals a local franchise authority’s finding that the operator does not qualify as a small cable operator pursuant to § 76.901( f) of the Commission’s rules. See 47 U. S. C. § 573. 19 13 C. F. R. § 121.201, NAICS Code 517510. Small entities are defined as all such companies generating $12.5 million or less in annual receipts. Id. 20 See Annual Assessment of the Status of Competition in the Market for the Delivery of Video Programming, 19 FCC Rcd 1606, 1666 ¶ 90 (2004). 21 See Annual Assessment of the Status of Competition in the Market for the Delivery of Video Programming, 20 FCC Rcd 2755, 2816 ¶ 110 (2005). 88 Federal Communications Commission FCC 05- 96 89 E. Steps Taken to Minimize Significant Impact on Small Entities and Significant Alternatives Considered 13. The RFA requires an agency to describe any significant alternatives specifically affecting small entities that it has considered in proposing regulatory approaches, which may include, among others, the following four alternatives: (1) the establishment of differing compliance or reporting requirements or timetables that take into account the resources available to small entities; (2) the clarification, consolidation, or simplification of compliance or reporting requirements under the rule for small entities; (3) the use of performance, rather than design, standards; and (4) an exemption from coverage of the rule, or any part thereof, for small entities. 22 14. The cable ownership limits are intended to prevent large cable entities from unfairly impeding the flow of video programming to consumers through their horizontal reach and/ or their vertical integration. Any horizontal or vertical limits adopted by the Commission would directly impact large cable entities, and we anticipate that they will have little adverse impact on small entities. The Second Further Notice discusses several potential scenarios in which small entities may suffer harm from large entities, either through their horizontal reach, their vertical integration, or both, and seeks comment on crafting rules that prevent harms to small entities, which could, in turn, protect the flow of programming to consumers. F. Federal Rules Which Duplicate, Overlap, or Conflict with the Commission’s Proposals 15. None. 22 5 U. S. C. § 603( c). 89 Federal Communications Commission FCC 05- 96 90 JOINT STATEMENT OF COMMISSIONERS MICHAEL J. COPPS AND JONATHAN S. ADELSTEIN Re: Implementation of Section 11 of the Cable Television Consumer Protection and Competition Act of 1992, et al., Second Further Notice of Proposed Rulemaking, CS Docket No. 98- 82 et al. It is with some disappointment that we vote to approve today’s Second Further Notice of Proposed Rulemaking. After the D. C. Circuit reversed our prior rules, the Commission sought public comment, in September 2001, on how to fashion new standards, consistent with the court’s opinion. Now, almost four years later, we still do not resolve these issues and provide much- needed certainty, but instead seek another round of comments. The record we adduced before, limited though it was, has grown stale, and needs to be refreshed and updated. Once the new record is compiled, we hope the Commission will prioritize this proceeding and move to a decision. Toward that end, we’re pleased that today’s item, even if it does not establish new numerical limits, does resolve some issues. Most importantly, the item puts to rest the notion that the Commission could simply decide that horizontal and vertical limits of some kind aren’t necessary. The item reiterates the clear language of the law: the Commission “shall . . . prescribe rules and regulations establishing reasonable limits” for a cable operator’s subscriber reach, as well as the number of its own channels it can run on its system. Against this backdrop, we hope cable operators and other parties do not argue that there should be no numerical limits, but instead provide appropriate and necessary information to help us implement the clear command of the statute. Given that the mandate dates back not to the now almost ten- year- old Telecommunications Act of 1996, but even worse, the Cable Television Consumer Protection and Competition Act of 1992, we need to work efficiently and productively to establish numerical limits which satisfy the statutory purposes expressed in section 613( f)( 2) of the Act as soon as possible. 90