No. 16-3830-cv United States Court of Appeals for the Second Circuit ________________________ UNITED STATES OF AMERICA, Plaintiff-Appellant, v. BROADCAST MUSIC, INC., Defendant-Appellee. ________________________ On Appeal from the United States District Court for the Southern District of New York ________________________ BRIEF OF INDUSTRY PARTICIPANTS AS AMICI CURIAE IN SUPPORT OF THE UNITED STATES OF AMERICA (See inside cover for list of Amici Curiae) ________________________ Kenneth L. Steinthal Jeffrey S. Bucholtz Joseph R. Wetzel KING & SPALDING LLP Ethan P. Davis 1700 Pennsylvania Avenue, NW KING & SPALDING LLP Washington, DC 20006 101 2nd Street, Suite 2300 Telephone: (202) 737-0500 San Francisco, CA 94105 Facsimile: (202) 626-3737 Telephone: (415) 318-1200 Facsimile: (415) 318-1300 (See inside cover for complete counsel for Amici Curiae) DATE: May 25, 2017 Paul M. Fakler ARENT FOX LLP 1675 Broadway New York, NY 10019-5820 Telephone: (212) 457-5445 Counsel for Music Choice Bruce D. Sokler MINTZ, LEVIN, COHN, FERRIS, GLOVSKY AND POPEO, P.C. 701 Pennsylvania Ave., NW, Suite 900 Washington, DC 20004 Telephone: (202) 434-7303 Counsel for NCTA – The Internet & Television Association Angelo I. Amador RESTAURANT LAW CENTER 2055 L Street, N.W., Suite 700 Washington, D.C. 20036 Counsel For Amicus Curiae Restaurant Law Center Charles Biggio Gary Greenstein WILSON SONSINI GOODRICH & ROSATI 1301 Avenue of the Americas 40th Floor New York, NY 10019 Telephone: (212) 497-7780 1700 K Street, NW, 5th Floor Washington, DC 20006 Counsel for Spotify, Inc. and Pandora, Inc. Kenneth L. Steinthal Joseph R. Wetzel Ethan P. Davis KING & SPALDING LLP 101 2nd Street, Suite 2300 San Francisco, CA 94105 Telephone: (415) 318-1200 Jeffrey S. Bucholtz KING & SPALDING LLP 1700 Pennsylvania Avenue, NW Washington, DC 20006 Telephone: (202) 737-0500 Counsel for American Beverage Licensees, Computer & Communications Industry Association, Google Inc., iHeartMedia, Inc., Mood Media Corporation, National Association of Broadcasters, Netflix, Inc., Radio Music License Committee, Restaurant Law Center, SoundCloud Ltd., Viacom Inc., and WineAmerica. Karyn Ablin FLETCHER, HEALD & HIDRETH, P.L.C. 1300 North 17th Street, 11th Fl. Arlington, VA 22209 Telephone: (703) 812-0443 Counsel for the National Religious Broadcasters Music License Committee CORPORATE DISCLOSURE STATEMENT 1. Alphabet Inc., a publicly traded company (NASDAQ: GOOG, GOOGL), has more than 10% ownership of Google Inc. No publicly held company owns 10% or more of Alphabet Inc.’s stock. 2. iHeartMedia, Inc. states that it has no parent corporation and no publicly held corporation owns 10% or more of its stock. 3. Mood Media Corporation has 183,694,082 Company Common Shares issued and outstanding. To the knowledge of the directors and executive officers of the Company, no person or company beneficially owns, or exercises control or direction over, directly or indirectly, Company Common Shares in aggregate entitled to 10% or more of the votes other than as follows: Arbiter Partners Capital Management, LLC Fidelity 32,128,324 20,482,664 17.5% 11.2% “Fidelity” means Fidelity Management & Research Company, Pyramis Global Advisors, LLC, Pyramis Global Advisors Trust Company, Strategic Advisors Incorporated, FIL Limited, Crosby Advisors LLC, and Fidelity SelectCo, LLC. 4. Netflix, Inc. has no parent corporation nor is there any publicly held corporation that owns 10% or more of its stock. 5. The National Religious Broadcasters Music License Committee (“NRBMLC”) is a standing committee of the National Religious Broadcasters (“NRB”), an association representing more than 1,300 radio and television stations, program producers, multimedia developers, and related organizations around the world. The NRB is a non-profit corporation that has no parent companies, and no publicly held company has a 10% or greater ownership interest in the NRB. 6. Pandora Media, Inc. states that it has no parent corporation and no publicly held corporation owns 10% or more of its stock. i 7. Radio Music License Committee has no parent corporation nor is there any publicly held corporation that owns 10% or more of its stock. 8. SoundCloud Ltd. has no parent corporation and there is no publicly held company that holds more than 10% of the stock of SoundCloud Ltd. 9. Spotify USA Inc. is a wholly-owned subsidiary of Spotify AB, a company organized under the laws of Sweden. Spotify AB is a wholly-owned subsidiary of Spotify Technology S.A., a company organized under the laws of the Grand Duchy of Luxembourg. Spotify Technology S.A. does not have a parent corporation and no publicly held corporation owns 10% or more of its stock. /s/ Kenneth L. Steinthal Kenneth L. Steinthal ii TABLE OF CONTENTS CORPORATE DISCLOSURE STATEMENT............................................. i  TABLE OF CONTENTS .......................................................................... iii  TABLE OF AUTHORITIES ...................................................................... iv  INTERESTS OF AMICI CURIAE ............................................................. 1  FACTUAL BACKGROUND ....................................................................... 2  A.  BMI and ASCAP ................................................................. 2  B.  Blanket Licenses ................................................................ 4  C.  Licensing Under the Consent Decrees .............................. 6  D.  DOJ’s Review ...................................................................... 8  E.  Split Works ....................................................................... 10  F.  The District Court’s Order ............................................... 10  SUMMARY OF ARGUMENT .................................................................. 11  ARGUMENT ............................................................................................. 15  I.  II.  The District Court Misinterpreted the Decree ......................... 15  A.  The Decree’s Plain Language and Purpose Require Full-Work Licensing ......................................................... 15 B. The District Court Ignored Copyright Law..................... 21 The District Court’s Misinterpretation Of The Decree Would Have Devastating Real-World Consequences .............. 28  A.  Bars and Restaurants ...................................................... 28 B. Audiovisual Content Providers ........................................ 30 C. Radio Stations and Digital Music Services ..................... 33 CONCLUSION ......................................................................................... 36  CERTIFICATE OF COMPLIANCE  CERTIFICATE OF SERVICE  iii TABLE OF AUTHORITIES Cases  Alden-Rochelle, Inc. v. ASCAP, 80 F. Supp. 888 (S.D.N.Y. 1948) ........................................................... 31 BMI v. Columbia Broad. Sys., Inc., 441 U.S. 1 (1979) ........................................................................... passim BMI v. Pandora Media, Inc., 140 F. Supp. 3d 267 (S.D.N.Y. 2015) .............................................. 20, 27 BMI v. Pandora Media, Inc., 2013 WL 6697788 (S.D.N.Y. Dec. 19, 2013) ........................................... 3 Brownstein v. Lindsay, 742 F.3d 55 (3d Cir. 2014) ..................................................................... 21 Buffalo Broad. Co. v. ASCAP, 744 F.2d 917 (2d Cir. 1984) ..................................................................... 5 Childress v. Taylor, 945 F.2d 500 (2d Cir. 1991) ................................................................... 22 Davis v. Blige, 505 F.3d 90 (2d Cir. 2007) ............................................................... 21, 22 In re Pandora Media, Inc., 6 F. Supp. 3d 317 (S.D.N.Y. 2014) .................................................... 4, 20 In re Pandora Media, Inc., 785 F.3d 73 (2d Cir. 2015) (per curiam) ....................................... passim M. Witmark & Sons v. Jensen, 80 F. Supp. 843 (D. Minn. 1948) ........................................................... 31 Radio Music License Comm., Inc. v. Global Music Rights, LLP, No. 16-cv-06076-CDJ (E.D. Pa.).............................................................. 4 Steele v. Turner Broad. Sys., Inc., 646 F. Supp. 2d 185 (D. Mass. 2009) .................................................... 30 U.S. v. ASCAP (In re Application of Buffalo Broad. Co.), 1993 WL 60687 (S.D.N.Y. Mar. 1, 1993), aff’d in part, vacated in part, 157 F.R.D. 173 (S.D.N.Y. September 2, 1994) ..................................... 22 iv U.S. v. BMI (In re Application of Music Choice), 426 F.3d 91 (2d Cir. 2005) ....................................................................... 7 U.S. v. BMI, 275 F.3d 168 (2d Cir. 2001) ................................................................. 2, 6 Statutes  17 U.S.C. § 101.............................................................................. 12, 22, 26 17 U.S.C. § 205(e) ..................................................................................... 25 Other Authorities  1 Nimmer on Copyright § 6.10[C] ............................................................ 24 1 Nimmer on Copyright § 10.07[B] .......................................................... 25 2 Patry on Copyright § 5:7 (2015) ............................................................ 21 2014 ASCAP Annual Report, available at http://www.ascap.com/-/media/files/pdf/about/annualreports/ascap_annual_report_2014.pdf .................................................. 3 Ben Sisario, SESAC Settles Antitrust Lawsuit Over Royalty Rates, N.Y. Times, July 23, 2015, https://www.nytimes.com/2015/07/24/business/ media/sesac-settles-antitrust-lawsuit-over-royalty-rates.html............. 4 BMI, Licensing FAQ: What is BMI? http://www.bmi.com/licensing ................................................................. 3 BMI, Our Role, http://www.bmi.com/about/#ourrole..................................... 24 BMI, Quarterly Distribution Update (September 2015), available at http://www.bmi.com/distribution/letter/572233 ................ 3 Comments of Nat’l Music Publishers’ Ass’n, Sept. 22, 2015, available at https://www.justice.gov/atr/public/ascapbmi2015/ ascapbmi22.pdf ...................................................................................... 10 v Dan Kopf, How Many People Take Credit for Writing a Hit Song?, Priceonomics, Oct. 30, 2015, http://priceonomics.com/how-manypeople-take-credit-for-writing-a-hit-song/ ............................................ 10 Donald S. Passman, All You Need To Know About The Music Business (9th ed. 2015) ..... 10 Ed Christman, ASCAP and Radio Group’s 5-Year Pact Doesn’t Address the Elephant in the Room, Billboard, Jan. 3, 2017, http://www.billboard.com/articles/business/7640666/ascap-rmlc-radiolicensing-agreement-analysis ................................................................. 3 Ed Christman, Inside the New Royalty Split for ‘Uptown Funk’: Who Gets Paid What, Billboard, May 4, 2015, http://www.billboard.com /articles/business/6553861/uptown-funk-royalties-who-gets-paid ...... 34 Global Music Rights, http://www.globalmusicrights.com ......................................................... 4 Helienne Lindvall, Calculating The Credits Behind Songwriting, The Guardian, June 24, 2008, https://www.theguardian.com /music/musicblog/2008/jun/24/calculatingthecreditsbehind ................ 34 Kristi Helm, Music licensing group’s targets include local restaurants, The Seattle Times, Aug. 2, 2007, http://www.seattletimes.com /business/music-licensing-groups-targets-include-local-restaurants/ . 30 Maureen A. Carlson, Music Copyright Disputes: Sam Smith Forced to Give Song Writing Credits to Tom Petty, JUX Law Firm, Feb. 19, 2015, http://jux.law/music-copyright-disputes-sam-smith-forced-to-givesong-writing-credits-to-tom-petty/ ........................................................ 34 Ned Hunter, Colorado Springs bar owner fined $21,000 over karaoke, The Gazette, Apr. 21, 2014, http://gazette.com/colorado-springs-barowner-fined-21000-over-karaoke/article/1518541 ............................... 30 vi Legislative Material  H.R.Rep. No. 1476, 94th Cong., 2d Sess. (1976) ..................................... 22 vii INTERESTS OF AMICI CURIAE Amici curiae are American Beverage Licensees, Computer & Communications Industry Association, Google Inc., iHeartMedia, Inc., Mood Media Corporation, Music Choice, National Association of Broadcasters, the National Religious Broadcasters Music License Committee, NCTA – The Internet & Television Association, Netflix, Inc., Pandora Media, Inc., the Radio Music License Committee, Restaurant Law Center, SoundCloud Ltd., Spotify USA Inc., Viacom Inc., and WineAmerica. Each amicus is either a BMI licensee or an industry association that represents the interests of BMI licensees. Amici include bar and restaurant owners, radio broadcasters, television broadcasters, cable television networks, music and audiovisual content services, and background music services.1 Because amici would be directly and severely harmed if the district court’s decision were to stand, they wish to be heard in support of the government on why the consent decree requires that BMI license compositions on a full-work basis—that is, license the right to actually Both parties consented to the filing of this brief. No party’s counsel authored this brief in whole or in part, and no person or entity other than amici, their members, and their counsel contributed money intended to fund its preparation or submission. 1 1 perform the compositions in its repertory, rather than license only “fractional” interests in those compositions that are worthless unless the licensee obtains additional licenses for the remaining fractional interests. If the district court had followed orderly procedures and allowed briefing on this critical question instead of deciding it at a pre-motion conference, amici would have sought to be heard below to explain how any departure from the longstanding industry practice of full-work licensing would harm their businesses. Amici’s experience negotiating and relying on blanket licenses under the BMI decree could help the Court understand why the district court’s holding that the decree permits fractional licensing conflicts with the decree’s plain language and undermines its fundamental purpose of controlling BMI’s market power. FACTUAL BACKGROUND A. BMI and ASCAP BMI, like the American Society of Composers, Authors, and Publishers (ASCAP), is a performing rights organization (PRO). It grants licenses to music users, collects license fees from them, and distributes those royalties among its affiliated copyright holders. U.S. v. BMI, 275 F.3d 168, 171 (2d Cir. 2001). BMI licenses non-exclusive public- performance rights in musical compositions on behalf of its affiliates to 2 a wide variety of music users, from bar owners to television and radio stations to digital music and audiovisual content distributors and more. BMI’s affiliates alone comprise approximately 750,000 composers, songwriters, and publishers, and BMI boasts that its repertory includes nearly 12 million compositions. See BMI, Licensing FAQ: What is BMI? http://www.bmi.com/licensing; see also BMI v. Pandora Media, Inc., 2013 WL 6697788, at *1 (S.D.N.Y. Dec. 19, 2013). Together, BMI’s and ASCAP’s repertories include about 90% of the musical compositions publicly performed in the United States.2 Each takes in over $1 billion in license fees annually.3 BMI claims that “[a]pproximately one out of every two songs played on radio is BMI-licensed music.” See http://www.bmi.com/licensing/#faqs. Two smaller, for-profit PROs license performance rights without direct government oversight: SESAC, Inc., founded in the 1930s, and See Ed Christman, ASCAP and Radio Group’s 5-Year Pact Doesn’t Address the Elephant in the Room, Billboard, Jan. 3, 2017, http://www.billboard.com/articles/business/7640666/ascap-rmlc-radiolicensing-agreement-analysis. 3 See 2014 ASCAP Annual Report, available at http://www.ascap.com//media/files/pdf/about/annual-reports/ascap_annual_report_2014.pdf; BMI, Quarterly Distribution Update (September 2015), available at http://www.bmi.com/distribution/letter/572233. 2 3 Global Music Rights (GMR), founded in 2013.4 SESAC has confronted antitrust issues of its own in recent years, and GMR is the defendant in an antitrust suit brought by the Radio Music License Committee in 2016.5 B. Blanket Licenses BMI and ASCAP operate primarily through blanket licenses, which give licensees the right to perform any of the compositions in their repertories as often as they want for a set period of time. BMI v. Columbia Broad. Sys., Inc., 441 U.S. 1, 5 (1979); see In re Pandora Media, Inc., 6 F. Supp. 3d 317, 322 (S.D.N.Y. 2014), aff’d, 785 F.3d 73 (2d Cir. 2015) (per curiam). These blanket licenses allow music users immediate access to millions of songs “without having to contact each copyright holder.” Pandora Media, 6 F. Supp. 3d at 322. The PROs and the blanket license: developed together out of the practical situation in the marketplace: thousands of users, thouIn re Pandora Media, Inc., 6 F. Supp. 3d 317, 351 & n.55 (S.D.N.Y. 2014); Global Music Rights, http://www.globalmusicrights.com. 5 See Ben Sisario, SESAC Settles Antitrust Lawsuit Over Royalty Rates, N.Y. Times, July 23, 2015, https://www.nytimes.com/2015/07/24/ business/media/sesac-settles-antitrust-lawsuit-over-royalty-rates.html; Radio Music License Comm., Inc. v. Global Music Rights, LLP, No. 16cv-06076-CDJ (E.D. Pa.). 4 4 sands of copyright owners, and millions of compositions. Most users want unplanned, rapid, and indemnified access to any and all of the repertory of compositions, and . . . owners want a reliable method of collecting for the use of their copyrights. BMI, 441 U.S. at 20. Accordingly, a “middleman with a blanket license was an obvious necessity if the thousands of individual negotiations, a virtual impossibility, were to be avoided . . . . Historically, the market for public-performance rights organized itself largely around the singlefee blanket license, which gave unlimited access to the repertory and reliable protection against infringement.” Id. at 20-21. Blanket licenses, when properly implemented, reduce the transaction costs of licensing copyrighted compositions. See Buffalo Broad. Co. v. ASCAP, 744 F.2d 917, 934 (2d Cir. 1984) (Winter, J., concurring). They let songwriters monetize their works immediately upon signing with the PRO and enable music users to begin performing the works immediately upon signing without fear of infringement. At the same time, the PROs’ collective bargaining on behalf of otherwise competing rights holders affords them massive market power. And because blanket licenses offer all songs in the PRO’s repertory on an all-or-nothing basis, they prevent songs from competing with each 5 other based on price. Thus, although collective licensing can create efficiencies, it also raises profound antitrust concerns. C. Licensing Under the Consent Decrees The United States sued BMI and ASCAP “for unlawfully monopolizing the licensing of performing rights.” BMI, 275 F.3d at 172. Both suits were settled by consent decrees that protected against the anticompetitive threats posed by collective licensing while maintaining the availability of blanket licensing—the critical benefit, possible only through collective licensing, that saved the PROs from condemnation as a per se restraint of trade. BMI, 441 U.S. at 20. In particular, BMI’s decree specifies the content of BMI’s repertory and requires BMI to offer a license to perform all of it. First, the decree provides that BMI’s “repertory” includes all of “those compositions, the right of public performance of which [BMI] has or hereafter shall have the right to license or sublicense.” BMI Decree § II(C); see ASCAP Decree § II(C). Second, it provides that BMI must, after receiving “a written application from an applicant for a license for the right of public performance of any, some or all of the compositions in [BMI]’s repertory, advise the applicant in writing of the fee which it deems reasonable for 6 the license requested.” BMI Decree § XIV(A); see ASCAP Decree § IX.A. Finally, to reduce the risk that BMI will use its market power to demand a supracompetitive fee, it provides that if BMI and an applicant cannot agree, the court will determine “a reasonable fee for the license requested.” BMI Decree § XIV(A); see ASCAP Decree § IX.A; U.S. v. BMI (In re Application of Music Choice), 426 F.3d 91, 96 (2d Cir. 2005) (“[R]atesetting courts . . . exist as a result of monopolists exercising disproportionate power over the market for music rights.”).6 The Supreme Court, confronted with the issue whether BMI’s and ASCAP’s blanket licensing activities were per se unlawful price fixing, recognized that they “plainly involve concerted action.” BMI, 441 U.S. at 4, 10. Nevertheless, there “are situations in which competitors have been permitted to form joint selling agencies or other pooled activities, subject to strict limitations under the antitrust laws to guarantee against abuse of the collective power thus created.” Id. at 14. These protections do not fully neutralize the PROs’ market power. Rate-court proceedings are expensive and lengthy. Smaller music users may lack the resources to litigate when faced with supracompetitive rate demands. 6 7 In concluding that the PROs’ blanket licensing was not per se illegal, the Court relied heavily on the efficiencies that those licenses provide users—namely, “unplanned, rapid, and indemnified access” to all the works in ASCAP’s and BMI’s repertories. Id. at 20. D. DOJ’s Review In 2014, at ASCAP’s and BMI’s request, the Antitrust Division opened an inquiry into the decrees’ operation and effectiveness. BMI and ASCAP sought the Division’s assent to decree modifications to allow what they called “partial withdrawals,” which would let BMI and ASCAP affiliates keep their works in BMI’s and ASCAP’s repertories for some licensees, while preventing the PROs from licensing those works to selected licensees. Both consent decree courts, and ultimately this Court, concluded that the decrees did not permit partial withdrawals. See Pandora Media, 785 F.3d at 77–78. The Antitrust Division declined to agree to decree modifications that would have permitted partial withdrawals. JA66–67. As the Division’s review progressed, the question of how “split works”—that is, works with multiple owners not all of whom are affiliated with the same PRO—should be handled became a significant issue. 8 May BMI (and ASCAP) license split works on a “fractional” basis whereby the licensee cannot perform such works without first getting additional licenses to the remaining fractional interests? Or does a blanket license under the decrees convey the right to actually perform such works and thus provide the rapid and indemnified access emphasized by the Supreme Court? After inviting comments from interested parties, the Antitrust Division concluded that that the typical BMI license historically had granted whole-work rights and that the decrees’ plain language did not permit fractional licensing. The Division explained that because the PROs’ repertories are defined by reference to “works” or “compositions”—not fractional ownership interests in works or compositions— and because the decrees require the PROs to license the right to perform all works in their repertories upon request by a music user, the decrees “require ASCAP and BMI to offer full-work licenses.” JA66. Allowing fractional licensing would create “additional impediments to the public performance of music” and undermine the very blanket-licensing efficiencies that the Supreme Court and this Court had relied upon in permitting BMI and ASCAP to survive antitrust attack. JA78–79. 9 E. Split Works As the government’s investigation made clear, most compositions have co-owners, and often many of them. Donald S. Passman, All You Need To Know About The Music Business 317-20 (9th ed. 2015). Although BMI asserts that split works represent a minority of the songs in its repertory, JA57, by some estimates as much as 85% of all musical works are co-owned, and the number of co-writers of songs is reportedly increasing. See Dan Kopf, How Many People Take Credit for Writing a Hit Song?, Priceonomics, Oct. 30, 2015, http://priceonomics.com/howmany-people-take-credit-for-writing-a-hit-song/. For example, every song in the 2014 Billboard Hot 100 has co-owners. Comments of Nat’l Music Publishers’ Ass’n, Sept. 22, 2015, at 4, available at https://www. justice.gov/atr/public/ascapbmi2015/ascapbmi22.pdf. In total, owner- ship of those 100 compositions is divided among over 1,300 fractional interests, and the overwhelming majority of those compositions have coowners that belong to separate PROs. Id. F. The District Court’s Order After DOJ issued its closing statement, BMI asked Judge Stanton for a conference to discuss an intended motion to request the court to reject DOJ’s interpretation of the decree. The court, however, dispensed 10 with actual motion practice and ruled the day of the pre-motion conference. As a result, amici—the entities that actually negotiate blanket licenses with the PROs and are directly affected by the court’s order— had no opportunity to be heard. Judge Stanton concluded that “[t]he phrase in Art. II (C) of the Consent Decree defining BMI’s repertory as ‘those compositions, the right of public performance of which [BMI] has . . . the right to license or sublicense’ is descriptive, not prescriptive” and that the decree “neither bars fractional licensing nor requires full-work licensing.” JA12. SUMMARY OF ARGUMENT The district court’s order ignores the plain language of the BMI consent decree and, equally plainly, the antitrust balance struck by the BMI (and ASCAP) decree(s) and relied on by the Supreme Court. Nothing in the text of the decrees supports the notion that BMI’s and ASCAP’s repertories are composed of fractions of ownership interests in compositions. To the contrary, the decrees define (i) the PROs’ repertories as composed of “works” or “compositions” and (ii) the PROs’ licenses as providing the right to perform the “works” or “compositions” in the 11 repertory. A license cannot grant the right to perform a work if it covers only a fractional interest in the work. Moreover, under basic copyright principles, it makes no sense to speak of licensing the performance right in a “fractional” interest in a joint work. If a work is “prepared by two or more authors with the intention that their contributions be merged into inseparable or interdependent parts of a unitary whole,” 17 U.S.C. § 101, there is no separable fraction to perform. Compositions may be fractionally owned, but only the composition may be performed. Under copyright law, and under the terms of BMI’s agreements with its affiliates, compositions co-owned by a BMI affiliate and non-affiliate(s) are in BMI’s repertory. And under the decree, if a composition is in BMI’s repertory, BMI’s blanket license must include the right to publicly perform that composition. BMI’s own licensing practices confirm that both BMI itself and its affiliates have long understood that the decree mandates full-work licensing. BMI’s (and ASCAP’s) member agreements require affiliated copyright owners to grant the PROs the right to license works that they own in whole or in part. Likewise, BMI and ASCAP licenses grant mu- 12 sic users the right to perform compositions, not an illusory right to perform fractional ownership interests in compositions. The district court’s order also upends the decrees’ careful balancing of antitrust concerns. The Supreme Court withheld per se condemnation of collective licensing because of the unique benefits offered by ASCAP’s and BMI’s blanket licensing, including specifically the value to music users of immediate, indemnified access to the PROs’ vast repertories. This benefit depends on licensees actually being able to perform works without risk of infringement liability. If a BMI blanket license conveys no right to perform the millions of split works co-owned by a BMI affiliate and non-BMI-affiliates, then there is no immediate, indemnified access to BMI’s repertory. Instead, a BMI licensee would still need to negotiate with all other co-owners of those millions of works (or their PROs, if applicable) before they could perform them without risk of infringement liability. This disruption of the decree’s careful antitrust balance is not merely academic. It threatens widespread and devastating consequences in the marketplace. “Every day, hundreds of thousands of restaurants, radio stations, online services, television stations, performance 13 venues, and countless other establishments publicly perform musical works.” JA68. Under the decrees, those users have relied on blanket licenses to perform any or all of the works in the BMI (and ASCAP) repertories. If the district court’s decision is allowed to stand, it would be significantly more expensive, burdensome, and difficult (if not impossible) for music users to obtain the rights they need to perform split works. Users that cannot control the music they perform, such as restaurants, bars, radio and television stations and cable/internet-delivered program services (which transmit syndicated programming and other programs/movies often decades old), would be particularly vulnerable to inflated fee demands by hold-out co-owners of split works. Fractional licensing also would severely undercut the value of the per-program license, another key procompetitive protection mandated by BMI’s decree. JA20. Equally fundamentally, “fractional” licensing would give every coowner hold-up power that would artificially increase its bargaining leverage. This would further splinter the market for licensing performance rights, as rights holders would have an incentive to take advantage of 14 this hold-up opportunity by withdrawing from BMI and ASCAP and either forming new PROs not constrained by the decrees or engaging in direct licensing without a PRO. Amici have seen firsthand this precise phenomenon in the last three years as new unregulated PROs like GMR have been formed and certain publishers have withdrawn from BMI and ASCAP. In sum, the district court’s decision contravenes the plain language of the decree and fundamental copyright law principles, destroys the decree’s antitrust balance, and threatens severe anticompetitive harm to the licensing marketplace. If the unique efficiencies of immediate, indemnified access to BMI’s vast repertory vanish, so does the justification for tolerating the anticompetitive concerns inherent in collective licensing. This Court should reverse the district court’s judgment and hold that the decrees require full-work licensing. ARGUMENT I. The District Court Misinterpreted the Decree A. The Decree’s Plain Language and Purpose Require Full-Work Licensing The government’s brief compellingly explains how the plain language and purpose of the decree require BMI to license the right to per- 15 form any and all of the compositions in its repertory. Rather than repeat arguments in the government’s brief, amici add only a few points. First, Judge Stanton went astray by asking whether a provision in the decree expressly proscribed fractional licensing. See JA9–10. The decree defines BMI’s repertory in terms of “compositions” and requires BMI to grant a license to “perform” all such “compositions.” JA26–28 (BMI Decree §§ II(C), XIV(A)); accord ASCAP Decree § VI. The only reasonable reading of that language is that the decree requires BMI to license the right to perform compositions—that is, actual compositions that can be performed, not fractional ownership interests that cannot be. Because the decree makes clear that full-work licensing is required, an express prohibition on fractional licensing would have been redundant. Judge Stanton’s reasoning also cannot be squared with this Court’s decision in Pandora Media, which held that the “plain language of the consent decree unambiguously precludes ASCAP from accepting” partial withdrawals through which certain publishers sought to have ASCAP license their music to some users but not others. 785 F.3d at 77–78. The ASCAP decree does not prohibit selective withdrawals in so 16 many words, but this Court (in language no less applicable to BMI) held that the selective withdrawals contravened the decree because “[t]he decree’s definition of ‘ASCAP repertory’ and other provisions of the decree establish that ASCAP has essentially equivalent rights across all of the works licensed to it.” Id. at 77. Just as the ASCAP and BMI decrees unambiguously preclude those PROs from redefining the performance rights they are required to grant by licensing those rights to some users but not others, so, too, do those decrees unambiguously preclude those PROs from redefining their repertory to consist of fractional interests in compositions rather than the actual “compositions” and “works” that the decrees explicitly obligate them to license. The BMI decree, in short, makes crystal clear that the publicperformance right that BMI must license is the right to perform any and all of the compositions in its repertory. Because fractional licensing would not convey the right to perform many compositions in BMI’s repertory, it contravenes the decree. A provision that explicitly anticipated and rejected this effort by BMI to defeat the decree’s plain language would be superfluous, just as the decree’s failure to anticipate and ex- 17 plicitly prohibit the selective-withdrawal gambit could not obscure the clarity with which that gambit violated the decree. Second, it is no accident that the decrees require full-work licensing. Only full-work licensing provides the procompetitive benefits that justify collective licensing in the first place, including the essential benefit of “unplanned, rapid, and indemnified access” to works under a PRO blanket license. BMI, 441 U.S. at 20. Fractional licensing would eviscerate this benefit. In a fractional-licensing regime, far from providing “unplanned, rapid” access to all the compositions in its repertory, a BMI blanket license would still require a music user to obtain additional licenses to perform the millions of split works with multiple owners not all of whom are BMI affiliates to avoid the risk of infringement liability. As a practical matter, the music user’s immediate, indemnified ability to perform even BMI’s 100%-controlled works would be delayed until such time (if at all) when it could identify this subset of BMI’s total repertory. And far from conveying “indemnified access,” a fractionallylimited BMI license would subject the music user to hold-up demands by the other co-owners of split works (or their PROs). The “last co- 18 owner standing” whose license is needed would be able to charge a supracompetitive fee because the user already would have sunk extensive expense into other licenses covering the same works, but still would not be able to perform the split works without substantial infringement exposure. To make matters worse, if a split work has a co-owner who is not a member of BMI or ASCAP, the music user would be unable to ask one of the consent decree courts to set a reasonable fee if the co-owner took advantage of its hold-up power to demand an unreasonable fee. These opportunities for supracompetitive pricing would create strong incentives for rights holders to withdraw from BMI and ASCAP. Why, after all, would a major publisher want to be constrained by the decrees and the district courts when it could instead extract higher rates by fracturing the licensing marketplace? BMI and ASCAP have already sought to benefit from the hold-up power of unregulated PROs and “withdrawn” publishers by pointing to the hold-up rates the latter are able to achieve as new purported “benchmarks” in the consent decree courts. That is exactly the pirouette that ASCAP, BMI, Sony/ATV Music Publishing, and Universal Music Publishing Group tried to turn 19 through selective withdrawals: the largest publishers selectively withdrew in order to use their market power to extract higher rates in direct licenses so that ASCAP and BMI could then cite those higher rates as new benchmarks in rate court cases. Pandora Media, 6 F. Supp. 3d at 354–58; BMI v. Pandora Media, Inc., 140 F. Supp. 3d 267, 284–89 (S.D.N.Y. 2015). Construing the decrees to permit fractional licensing would allow BMI and ASCAP to benefit from the hold-up power that fractional licensing would create. For these reasons, it is no answer to the problems posed by fractional licensing to say that music users should just obtain licenses from the other PROs. That is unjustifiably coercive in the same way a tie-in is coercive: “We (BMI) will grant a (meaningful) license only on the condition that you take a license from others.” It also ignores the realworld dynamics that fractional licensing would unleash: if BMI could engage in fractional licensing, more unregulated PROs would spring up, and more publishers would withdraw from PROs, because they would be able to hold up music users who have already sunk extensive investments into licenses with some co-owners but still need licenses from other last-remaining co-owners before performing a split work without 20 infringement exposure. With these incentives toward further fracturing of the licensing marketplace, the hold-ups would never end. In sum, BMI’s and ASCAP’s collective licensing escaped antitrust condemnation only because of the unique benefits provided by blanket licensing—and possible only in a regime of full-work licensing. It makes no sense to interpret the decree to substantially alter this antitrust balance essential to the PROs’ lawful existence. B. The District Court Ignored Copyright Law The district court’s order was based on the misguided notion that it would raise a copyright-law problem to interpret the decree to require full-work licensing. But full-work licensing is not just consistent with copyright law, it is the default rule. Under copyright law, where there are joint owners of a work, each co-owner may grant a non-exclusive license to use the work—the work as a whole, not a fraction of it—without the consent of the other coowners. Davis v. Blige, 505 F.3d 90, 100 (2d Cir. 2007); Brownstein v. Lindsay, 742 F.3d 55, 68 (3d Cir. 2014); 2 Patry on Copyright § 5:7 (2015). This rule flows from the nature of joint works—that is, works “prepared by two or more authors with the intention that their contri- 21 butions be merged into inseparable or interdependent parts of a unitary whole.” 17 U.S.C. § 101. Thus, under copyright law, a joint work is not divisible into fractions that can be licensed apart from the work as a whole. Davis, 505 F.3d at 98. Because “[t]he touchstone [of a joint work] is the intention, at the time the writing is done, that the parts be absorbed or combined into an integrated unit,” joint owners hold undivided interests in a work. Childress v. Taylor, 945 F.2d 500, 505 (2d Cir. 1991) (quoting H.R.Rep. No. 1476, 94th Cong., 2d Sess. 120 (1976)). A co-owner therefore may unilaterally grant non-exclusive licenses to that work, and the licensee assumes no obligation to other co-owners; instead, the co-owner who granted the license is responsible for accounting to the other co-owners. Id.; U.S. v. ASCAP (In re Application of Buffalo Broad. Co.), 1993 WL 60687, at *79–80 (S.D.N.Y. Mar. 1, 1993) (“once a broadcaster has obtained a license from one of two joint copyright holders, he is immune from copyright liability to the other copyright holder”), aff’d in part, vacated in part, 157 F.R.D. 173 (S.D.N.Y. September 2, 1994). Consequently, if one co-owner of a composition is a BMI affiliate, that co-owner can grant BMI the right to license the right to perform 22 the composition regardless of whether the other co-owners are BMI affiliates, ASCAP members, affiliates of another PRO, or affiliates of no PRO. See U.S. Br. 10, 14. Joint owners may be able to depart from this default rule by agreeing among themselves on how they will license a jointly-owned work. But in reality, as the government’s brief explains, the licensing of the public-performance right has historically occurred via PRO full-work licenses. Id. at 10–11, 39–41.7 BMI’s own agreements leave no room for doubt about this. BMI’s affiliation agreement states in plain English that the affiliate grants BMI the right to license “[a]ll music compositions . . . composed by [the affiliate] alone or with one or more co-writers.” JA148 (emphasis added). And BMI has consistently held itself out to music users as offering full-work licensing. See, e.g., U.S. Br. 40–41; JA74–75. BMI’s most recent pronouncement, for example, that it has nearly 12 million songs in Because the decree concerns the public-performance right, it is of no moment that co-owners of works commonly license a different right— synchronization (or “synch”) rights, i.e., the right to reproduce and integrate a composition into an audiovisual work—on an individual-song basis under terms requiring all co-owners to consent before the licensee may use the work. See JA77-78. 7 23 its repertory, cannot be true unless that repertory includes split works.8 In short, BMI has been obtaining full-work licensing rights for split works from its affiliates and offering full-work licensing to its licensees. BMI’s claim that possible contracts among co-owners barring one coowner from licensing a split work without the other co-owners’ consent pose a problem for full-work licensing ignores this clear history as well as the decree’s clear mandate.9 In any event, to whatever extent BMI affiliates or ASCAP members may have entered into such contracts with their co-owners, that would be an issue among those co-owners and would not affect the blanket license conveyed by BMI or ASCAP. See 1 Nimmer on Copyright § 6.10[C] (so long as licensee did not have notice of contractual restriction among co-owners, one co-owner’s grant of license is valid for licensee even if that co-owner breached contract with co-owners); see also See BMI, Our Role, http://www.bmi.com/about/#ourrole (BMI “offer[s] blanket music licenses that permit [licensees] to play nearly 12 million musical works”). 9 As the government explains, BMI’s reliance on its history of making payments to affiliates based on their fractional interests in compositions is just an effort to muddy the waters. U.S. Br. 42–45. BMI and its affiliates are free to agree to a royalty-payment system based on fractional ownership interests, but BMI’s claim that such a payment system shows that BMI has been offering fractional licenses is nonsense. 8 24 17 U.S.C. § 205(e); 1 Nimmer on Copyright § 10.07[B] (even where copyright owner has assigned all rights in a work to another party, a nonexclusive license granted by the prior owner after the assignment is valid if licensee paid consideration and took license without notice of the assignment). Indeed, BMI itself acknowledged in a letter to DOJ that any private restriction on whole-work licensing could be enforced only against a licensee with notice of the restriction. JA93. Further, any such private agreements incorporated into the operations of a significant collective licensing body would have to pass antitrust muster. Ultimately, this Court need not decide in this appeal what the legal effect would be of a contract among co-owners barring a co-owner from licensing a work without the other co-owners’ consent if one coowner licensed the work through BMI in breach of that contract. DOJ specified that it would not enforce the decrees’ requirement of full-work licensing for a year so that BMI, ASCAP, and co-owners could use that grace period to address any issues that may need to be addressed concerning such contracts among co-owners. The critical issue in this appeal is not what happens if co-owners breach contracts among themselves relating to the licensing of joint works, but rather the nature of a 25 joint work, the plain language of the decree, and the consequences of disrupting the decree’s careful balancing of pro- and anticompetitive effects. Given that a joint work by definition is “a unitary whole” and its co-owners’ “contributions” by definition have “merged into inseparable or interdependent parts of [that] unitary whole,” 17 U.S.C. § 101, the very notion of a “fraction” of a work for purposes of performance rights is a fiction. True, co-owners may each have a percentage (and thus a “fractional”) economic interest in a work. But that does not mean, as BMI suggested below, that the work can be subdivided into fractions to be licensed and performed separately. That notion is illusory: a fractional license to a split work is, if not an oxymoron as a legal matter, useless as a practical matter. Why, one wonders, would BMI want to shift to a regime in which it offers licenses that do not allow its licensees to perform millions of jointly-owned compositions? A full-work license that allows the licensee to perform all the compositions in BMI’s repertory is much more valuable than a fractional license that does not. Sellers normally do not seek to dramatically cut the value of their products; doing so normally would 26 cause demand or price or both to fall off a cliff. BMI and ASCAP, however, need not fear such competitive consequences. Each knows that music users typically need blanket licenses from both to protect against infringement liability. BMI and ASCAP thus can safely reduce their licenses’ value, secure in the knowledge that music users will still need to buy those licenses anyway. Further confirmation of the anticompetitive nature of BMI’s desire for fractional licensing is that the fractional-licensing initiative arose out of DOJ’s review of the selective-withdrawal scheme rejected by this Court. Both initiatives seemingly hurt the PROs in the short term by leaving them with less to offer to licensees. But both are strategies to escape the decree’s constraints by using publisher licensing outside of those constraints and court oversight to generate higher prices that the PROs then will use in rate court cases to drive up their own rates. See Pandora Media, 785 F.3d at 77–78; Pandora Media, 140 F. Supp. 3d at 284–89. The only way to make sense of BMI’s desire to engage in fractional licensing is as a strategy to undermine the decree’s protections against 27 BMI’s market power. If the decree were ambiguous—and it is not—this would hardly be a good reason to resolve the ambiguity in BMI’s favor. II. The District Court’s Misinterpretation Of The Decree Would Have Devastating Real-World Consequences Examples from three types of amici demonstrate three factors that compound the harm caused by the district court’s order: lack of control, lack of information, and timing. The order ignores these market realities and the decree’s careful balancing act and would be devastating for the licensing marketplace. A. Bars and Restaurants In bars, clubs, and restaurants, the specific songs performed to entertain or as background music usually are not selected by the premises owner, but rather by a band, DJ, or radio station. Rights holders are entitled to royalties for these performances, and these establishments historically have relied on blanket licenses from a very small number of PROs to avoid infringement exposure. If BMI’s licenses became fractional, these licensees would have no reliable way to determine which songs are “split” works, which means they would not be able to determine which songs they could allow the band or DJ (or radio station, if they could somehow control it) to play. 28 Even if they obtained licenses from every extant PRO, they would face the risk that fractional owners of split works would keep their interests outside of a PRO to increase leverage. And structurally, as explained above, a switch to fractional licensing would create a hold-up opportunity and give rights holders a powerful incentive to create new PROs not subject to the decrees or to withdraw from PROs—a very real prospect given that Sony/ATV Music Publishing and Universal Music Publishing Group withdrew from BMI in 2014 when their attempt to selectively withdraw failed. See supra at 19–20; JA71–72, JA94; Pandora Media, 785 F.3d at 76. In a fractional-licensing regime, the price of obtaining licenses from the four PROs that currently exist would thus go up (even if their repertories decreased and the consent decree courts blocked BMI and ASCAP from using hold-up prices to drive up their own rates), and the creation of new, smaller PROs would compound the inefficiency and hold-up problem. Fundamentally, because these licensees would have no practical way to identify split works (or their owners), they would have no practical way to avoid crushing infringement exposure. Cf. Ned Hunter, Colorado Springs bar owner fined $21,000 over karaoke, The Gazette, Apr. 29 21, 2014, http://gazette.com/colorado-springs-bar-owner-fined-21000- over-karaoke/article/1518541; Kristi Helm, Music licensing group’s targets include local restaurants, The Seattle Times, Aug. 2, 2007, http://www.seattletimes.com/business/music-licensing-groups-targetsinclude-local-restaurants/. The result would likely be that these licensees would have to forgo many of the music offerings they currently provide—leading to reduced output and, ironically, less revenue for songwriters. B. Audiovisual Content Providers Judge Stanton’s decision would also harm licensees like Netflix, Viacom, and many other entities (including NCTA members) that transmit third-party-produced audiovisual content containing copyrighted music, like television programs and movies. To use music in an audiovisual work, the creator must obtain synchronization licenses from the owners of any compositions reproduced in timed relation with the visual content. See Steele v. Turner Broad. Sys., Inc., 646 F. Supp. 2d 185, 193 (D. Mass. 2009). With few exceptions, industry practice is for the audiovisual content producers not to secure public-performance rights at the same time as they secure syn- 30 chronization rights. The responsibility for securing performance rights typically falls instead to the broadcaster or non-theatrical exhibitor, including TV stations, cable networks and other audiovisual content services. This puts the downstream exhibitor in the disadvantageous position of having to negotiate for performance rights after the music is already “in the can” and it cannot control which compositions are synched with the programming. The PROs’ full-work blanket licenses, subject to oversight under the decrees, allow exhibitors to substantially mitigate this problem.10 If BMI engaged in fractional licensing, co-owners of split works could continue to withhold performance rights at the outset and effectively deprive the downstream exhibitor of license coverage for alreadyThe music-in-the-can issue was addressed in the context of theatrical public-performance rights in Alden-Rochelle, Inc. v. ASCAP, where the court found that ASCAP’s practice of prohibiting its members from granting performance rights along with synchronization rights to motion-picture producers violated the antitrust laws. 80 F. Supp. 888, 893–94 (S.D.N.Y. 1948); see also M. Witmark & Sons v. Jensen, 80 F. Supp. 843, 847–50 (D. Minn. 1948). After Alden-Rochelle, the ASCAP decree was amended to prevent ASCAP from licensing movie theaters for public performances of music synched with motion pictures. As a consequence, performance rights in theatrically-distributed films are licensed in a price-competitive marketplace at the time of production. ASCAP Decree § IV.E. However, Alden-Rochelle’s protections have yet to be extended beyond movie theaters. 10 31 produced content unless the exhibitor took licenses at whatever fee the rights holder might charge. The exhibitor would be forced to engage with the rights holders at a time when it has no option but to use the music already in the can.11 The major publishers recognize the inordinate market power this timing issue gives them. In the limited circumstances where exhibitors produce some of the programming they transmit and have sought to obtain performance rights and synch rights together at the point of production to avoid this “in the can” problem, some publishers have refused to license performance rights directly at the same time they grant synch licenses—thus perpetuating the problem. Requiring downstream exhibitors to identify and engage in afterthe-fact negotiations with every co-owner of split works embodied in their programming would be fundamentally anticompetitive as well as impracticable. Those licensees could well forgo offering entire television Standard distribution contracts for third-party-produced programming often prohibit television exhibitors like Netflix from altering the musical content of the programming. The same is true of many radio program syndication agreements. 11 32 series/episodes or films as a result of the hold-up power of one co-owner of one composition. Fractional licensing in these circumstances vests veto power in potentially numerous rights holders, each of whom may have a relatively small interest in the work but can unilaterally veto its use. When each co-owner can say “no,” but no co-owner is empowered to provide a meaningful “yes,” it is far easier to hold up progress than to achieve it. C. Radio Stations and Digital Music Services It is equally impossible for the radio industry to function without the rapid, indemnified access provided by PRO full-work licensing. Radio stations, like audiovisual services, transmit syndicated programming and advertisements that include musical works selected by third parties. The broadcasters are in no position to identify the works’ owners, let alone license the works, independent of securing full-work PRO blanket licenses. This problem is exacerbated for the radio and digital music industries because record labels typically release sound recordings for airplay weeks or months before the ownership of the compositions embodied in those recordings is made known (or even determined). JA77. Moreover, 33 there are often extended disputes about who controls the rights to compositions.12 For example, the hit “Uptown Funk” originally had six credited songwriters before a dispute resulted in ten getting credit.13 As a result, without the immediate, indemnified access provided by full-work PRO licensing, radio stations and digital music services like those operated by Google Inc., iHeartRadio, Mood Media Corporation, Music Choice, Pandora, SoundCloud, and Spotify would face immense exposure for the performance of compositions that inevitably will include split works. The leverage enjoyed by co-owners of split works who are not affiliates of BMI or members of ASCAP would be enormous. And the alternative—not playing new releases for fear of unknown infringement exposure—would be contrary to the public interest. See, e.g., Helienne Lindvall, Calculating The Credits Behind Songwriting, The Guardian, June 24, 2008, https://www.theguardian.com /music/musicblog/2008/jun/24/calculatingthecreditsbehind; Maureen A. Carlson, Music Copyright Disputes: Sam Smith Forced to Give Song Writing Credits to Tom Petty, JUX Law Firm, Feb. 19, 2015, http://jux.law/music-copyright-disputes-sam-smith-forced-to-give-songwriting-credits-to-tom-petty/. 13 Ed Christman, Inside the New Royalty Split for ‘Uptown Funk’: Who Gets Paid What, Billboard, May 4, 2015, http://www.billboard.com /articles/business/6553861/uptown-funk-royalties-who-gets-paid. 12 34 Yet another problem that radio stations and other licensees would face arises from fractional licensing’s harmful effect on the usefulness of the per-program license, a core competitive protection mandated by BMI’s and ASCAP’s decrees and a key focus of amicus the National Religious Broadcasters Music License Committee (NRBMLC), which represents hundreds of radio stations with limited music use that hold a per-program license with BMI. That form of license reduces BMI’s and ASCAP’s market power by, among other things, enabling users to limit their license fee obligations by paying those PROs only for programming periods that include one or more otherwise unlicensed compositions in BMI’s or ASCAP’s repertory. In the radio industry, the fees applicable to these programming periods do not vary based on whether a split work or a single-owner work is featured in a programming period. Under the longstanding full-work licensing regime, a radio perprogram licensee that broadcasts programming periods including only split works owned by both BMI- and non-BMI-affiliates can clear these periods by paying BMI the fee for that period (which is just as high as if those works were wholly owned by BMI affiliates). Under fractional licensing, however, that licensee would need to pay that same BMI fee 35 plus additional license fees due to other co-owners of the same works (or their PROs, if applicable). As a result, the per-program license would become more expensive and far less transactionally efficient, rendering it a much less viable alternative to the blanket license – thus diminishing the competitive protections offered by the decree (and also diminishing inter-PRO competition in offering attractive alternative license options to users). CONCLUSION The district court’s decision conflicts with the consent decrees, with how all the market participants—including BMI and ASCAP— have interpreted those decrees, and with the default rules of copyright. Should it stand, it would overturn the balance struck in the decrees between the efficiencies a collective licensing scheme affords and the protections needed against its anticompetitive effects. This Court should reverse and hold that the decree requires full-work licensing. 36 Respectfully submitted. /s/ Kenneth L. Steinthal Kenneth L. Steinthal Joseph R. Wetzel Ethan P. Davis KING & SPALDING LLP 101 2nd Street, Suite 2300 San Francisco, CA 94105 Telephone: (415) 318-1200 Paul M. Fakler ARENT FOX LLP 1675 Broadway New York, NY 10019-5820 Telephone: (212) 457-5445 Counsel for Music Choice Bruce D. Sokler MINTZ, LEVIN, COHN, FERRIS, GLOVSKY AND POPEO, P.C. 701 Pennsylvania Ave., NW, Suite 900 Washington, DC 20004 Telephone: (202) 434-7303 Counsel for NCTA – The Internet & Television Association Angelo I. Amador RESTAURANT LAW CENTER 2055 L Street, N.W., Suite 700 Washington, D.C. 20036 Counsel For Amicus Curiae Restaurant Law Center Charles Biggio Gary Greenstein WILSON SONSINI GOODRICH & ROSATI 1301 Avenue of the Americas New York, NY 10019 Telephone: (212) 497-7780 1700 K Street, NW, 5th Floor Washington, DC 20006 Counsel for Spotify, Inc. and Pandora, Inc. 37 Jeffrey S. Bucholtz KING & SPALDING LLP 1700 Pennsylvania Avenue, NW Washington, DC 20006 Telephone: (202) 737-0500 Counsel for American Beverage Licensees, Computer & Communications Industry Association, Google Inc., iHeartMedia, Inc., Mood Media Corporation, National Association of Broadcasters, Netflix, Inc., Radio Music License Committee, Restaurant Law Center, SoundCloud Ltd., Viacom Inc., and WineAmerica. Karyn Ablin FLETCHER, HEALD & HIDRETH, P.L.C. 1300 North 17th Street, 11th Fl. Arlington, VA 22209 Telephone: (703) 812-0443 Counsel for the National Religious Broadcasters Music License Committee CERTIFICATE OF COMPLIANCE Pursuant to Federal Rule of Appellate Procedure Rule 32(a) and Cir. Rule 32.1(a), I certify that this brief complies with the length limitations set forth in Fed. R. App. P. 32(a)(7) because it contains 6,973 words, as counted by Microsoft Word, excluding the items that may be excluded under Fed. Rule App. Proc. 32(a)(7)(B)(iii). /s/ Kenneth L. Steinthal CERTIFICATE OF SERVICE I certify that on May 25, 2017, I filed the foregoing brief via the Court’s electronic case filing (ECF) system. Pursuant to Cir. Rule 25.1(h), the resulting Notice of Docket Activity generated by the ECF system constitutes service on counsel for the parties. /s/ Kenneth L. Steinthal