COMPULSORY LICENSES IN PEER?TO?PEER FILE SHARING: A WORKABLE SOLUTION

[A pdf version of this article, though without the table and charts, is available at

A. The Compulsory License for Making and Distributing Phonorecords

The most enduring compulsory license is the original one, adopted in the Copyright Act of 1909. The elaborate scheme was Congress's response to the Supreme Court's decision in White-Smith v. Apollo[12] holding that piano rolls, and, by extension, phonorecords, were not “copies” of the musical works they recorded. That holding meant that the creators of phonorecords or other mechanical reproductions of musical works did not have to pay the owners of copyrights in the songs they reproduced.

In 1909, Congress legislatively overruled the White-Smith case by providing that the making of phonorecords or other mechanical versions of songs was subject to copyright protection. Congress created the phonorecord compulsory license to protect against the monopolization of music by the sound recording industry, and to assure that performers would have access to any songs they wanted to “cover” by making their own recordings at a reasonable price.[13] The provision has stood the test of time, increasing from 2 cents per song in 1909 to 9.1 cents per song (or 1.75 cents per minute of playing time) scheduled to go into effect in 2006.[14]

The success of this original compulsory license may have inspired Congress to adopt other compulsory licenses in the 1976 Copyright Act. But the phonorecord license arose in a context significantly different from any of the other compulsory licenses, and particularly the peer-to-peer environment. The phonorecord compulsory license does not involve the “pooling” of funds, but [72] rather the direct payment by a user/performer (or the performer's recording company) to the owner of copyright in the underlying musical work (or payments made through the Harry Fox Agency as a designated intermediary).

The phonorecord license thus is simpler to administer than the later, more complicated compulsory licensing schemes. It also tracks more closely the private contract negotiation that would have occurred in the absence of the compulsory license.[15]

At least part of the justification for interfering with the normal market in musical works was the fact that the users—the performers and record companies involved in making new versions of older works—also contributed creatively to the pool of available versions of songs. This is not the case in the typical peer-to-peer transaction, which usually involves the simple multiplication (and potential displacement) of copies of works that are already available through commercial channels. A different situation might pertain if file sharing produced a large number of derivative works, through sophisticated digital editing and manipulation. But this has not been the case to date.

The story of the first compulsory license, however, is not finished. As electronic dissemination of musical works displaces the traditional sale of phonorecords and CDs, any compulsory license pegged only to the old technology soon would be doomed to failure. In 1995, Congress updated section 115 to compensate music copyright owners for the digital delivery of works authorized under the compulsory license, as well as the sale of old-fashioned “phonorecords” (defined broadly enough to include CDs).[16]

Prior to 1995, though there was an exclusive performance right in the underlying music, there was no exclusive performance right in sound recordings as such. In 1995, however, Congress created such a right. It was limited to the digital performance or transmission of such works, with lots of exceptions that nullified much of the potential impact of the new right.[26] As part of the package, Congress created a compulsory license that applied to some non-interactive digital transmission services. Such a compulsory license might seem relatively easy to set up, since it involves a relatively finite number of webcasters, who do or could operate their websites for profit, and who presumably are in a position to absorb reasonable performance fees.

After Congress adopted the complicated new right and incorporated the compulsory license into section 114 of the Act, observers waited to see how the compulsory license would work out. Even before any fees had been collected under the license, however, it became obvious that the statutory language was unclear. Did it apply to “streaming audio”? No one knew. By 1998, as part of the Digital Millennium Copyright Act, Congress revised the language to clear up some of the ambiguities.[27] A Copyright Arbitration Royalty Panel was established to recommend the initial rates for the compulsory license;[28] it came up with a proposed rate of 0.14 cents for each song streamed on an Internet-only webcast, and 0.07 cents for each song included as part of an AM or FM radio retransmission. After much public discussion and complaint, the Librarian of [75] Congress adopted a compromise rate of 0.07 cents for each song delivered, whether by AM, FM, or Internet-only transmission.

Many people thought that the rates were outrageous, and that smaller operators could not afford them. Congress intervened by passing the Small Webcaster Settlement Act of 2002.[29] Currently, the webcasting royalty rates are divided into nine categories of digital audio services, depending upon such factors as whether the service is commercial or noncommercial. Fees range from as low as $200 for noncommercial webcasters devoted primarily to news, talk, and sports, to 10% of gross proceeds for such commercial services as XM Satellite Radio and SIRIUS Satellite Radio.

Since its rocky start, the compulsory license has begun generating at least a moderate flow of revenue, reaching as high as $35 million in 2005.[30] While the fees might seem to bode well as a model for a peer-to-peer compulsory license, the comparison is misleading. Much of the revenues generated by the new digital performance right are attributable to commercial satellite radio services such as XM and SIRIUS. Most peer-to-peer exchanges on the Internet, by contrast, will presumably be in a non-commercial setting, where revenues are not likely to be generated, and funds will not likely be available for distribution.

For almost two decades, the broadcast and cable industries fought over whether and how much cable systems should pay rights holders for cable systems' retransmission of programs broadcast by television stations. As a first step to establish a bargaining advantage, television broadcast networks and producers sued to establish that cable use of copyrighted broadcast programming was a copyright infringement. Partly out of fear of strangling the then-emerging cable industry, the Supreme Court twice flatly held that this type of use was “passive” in nature, and thus created no liability.[31]

After the Teleprompter decision, the broadcast and production interests got the message that no judicial relief was in sight, and turned their attention to the [76] decades-old Congressional fight over cable fees. The result was a compulsory license in section 111 of the 1976 Copyright Revision Act, which went into effect in 1978. This hideously complicated provision provided that cable operators could carry both local and distant broadcast television signals for a fee mandated by the Act, subject to periodic adjustments by the former Copyright Royalty Tribunal. (Later, upon the abolition of the Tribunal, Copyright Arbitration Royalty Panels were appointed by the Librarian of Congress. Most recently, in November 2004, the panels were themselves replaced by a new system of Copyright Royalty Judges, to take effect in 2005.) The fee was based upon the number of “distant signal equivalents” (“DSEs”) that a cable system imported, counting a distant independent station as one and a network-affiliated station or educational station as 1/4. The number of DSEs was multiplied by a figure initially set by Congress and later adjusted by the Tribunal, to establish the percentage of their gross revenues charged for importing distant television signals.[32] The revenues collected by the licensing system then were divided among the copyright owners, after elaborate hearings that typically held up distributions for many years. The big winners in this process generally were broadcast programming and sports rightsholders.[33]

The percentage of gross revenues for each DSE has increased over the years. Similarly, the total gross revenues of cable systems has increased steadily every year. (See Table I and Graph I, reproduced at the end of this article, showing an increase from just over $1 billion in revenues when the Copyright Act was first passed, to almost $30 billion in 2002.) But the total payment under the cable compulsory license (Graph II) actually has decreased in the last decade. After peaking near $200 million in 1989, it has gone down to only about $120 million in the last few years. (In part, this is offset by an increase in the compulsory licensing fees for satellite distribution systems under section 119, described below, which in 2002 amounted to almost $69 million.[34])

Why have the royalties under the compulsory license decreased? Quite simply, cable systems do not import as many distant signals as in the early days. Today, viewers are interested not in distant signals, but rather in satellite [77] networks—free, per-channel, or pay-per-view—for which cable operators negotiate fees in a free marketplace. Even in its infancy, the cable compulsory license system was implemented against the backdrop of FCC regulations that severely limited the number of DSEs a cable system could import.[35] While the FCC long ago repealed the limitation, the section 111 fees effectively continue the cap on distant signals, by pricing the importation of a DSE that would have been barred by the earlier FCC rules at about 4 percent of gross revenue.[36] Cable operators thus do not view distant signal importation as a useful market strategy.

Broadcasters and cable operators also have fought over the rebroadcast of local over-the-air channels on cable systems within the same viewing area. Under the FCC's rules in the 1970s, cable systems were required to carry local programming under “must carry” rules.[37] Presumably, the local station operators did not lose money by this arrangement: broadcasters kept their local viewers—by being carried on cable systems—and were able to charge advertisers for them. The cable compulsory license did not compensate for the retransmission of local stations, since the cable operators were required to carry these signals in any event, and the local broadcasters wanted it that way; the DSE figure was based totally upon the importation of signals from outside the viewing area )and not upon retransmission of local television signals.

Most cable subscribers today watch satellite-delivered non-broadcast programming, for which the copyright model is not a compulsory license, but rather a negotiated contract. The broadcasters quickly began to figure out that the real money was in non-broadcast satellite networks.

With the decrease in carriage of distant signals, payments under section 111 naturally went down. The statute explicitly requires payments only for signals carried beyond their normal licensed area—that is, distant signals.[38] Congress' theory quite reasonably seems to have been that broadcasters benefited from cable carriage of their signals; if the cable operators had any incentive not to carry local signals, broadcasters naturally would lose viewers—and hence advertising revenues—)in their home markets. There was and is no need to impose a compulsory copyright scheme on local signals. Indeed, in many cases [78] broadcasters assist local systems in receiving high-quality signals, by building direct fiberoptic or microwave connections to cable operators.

After the widespread development of satellite cable channels in the late 1980s, cable operators had a declining need to import distant signals.[39] And since systems do not pay for local signals, it was inevitable that copyright payments would fall—as discussed and as set forth in Table I and Graph II at the end of this article.

Although beyond the scope of this paper, the change in compulsory copyright's significance is a good illustration of government's inability to predict rapid changes in market forces. In the decade after section 111's enactment, market changes reduced its importance significantly. Although satellite transmission existed at the time of the 1976 Copyright Revision Act, its drafters simply did not foresee its effect upon the relevance of signal importation and hence of a compulsory copyright scheme oriented around distant signals.

At the same time that section 111 was becoming less relevant, broadcasters and cable operators were moving to a system of private negotiations. To accommodate the shift, Congress, in the Cable Television Consumer Protection and Copyright Act of 1992, provided for “retransmission consent” (“RTC”) as an alternative to must-carry and effectively a supplement to section 111 royalties.[40] (Section 111 applies to owners of copyright in the individual programs; RTC extends rights to the broadcasters themselves, based upon their broadcast signal, and without regard to the ownership of any copyrights.)

Effective in 1993, section 325(b)(3) of the Communications Act allowed broadcasters and rights holders to negotiate for permission to carry their signals. This approach carries with it a risk under section 325(b)(4); if a broadcaster is unable to reach a retransmission consent agreement with a cable operator, it gives up its right to cable carriage locally under the current version of the “must carry” rules. But broadcasters appear to have sought such arrangements quite eagerly.

Instead of competing for relatively small slices of the compulsory copyright pie, after 1993 broadcasters seem to have preferred using the RTC option to negotiate for compensation. This apparently has not resulted in any purely financial windfalls. Instead, to the extent that the results of these negotiations are visible, they seem to reflect an increased reliance upon a form of barter.

Because the RTC deals are proprietary in nature, their details are never disclosed. Aside from the contracts' private nature, cable operators naturally fear [79] that if they make a highly favorable deal with one popular local broadcaster, others will demand the same terms. Nevertheless, discussions with cable industry executives indicate some broad outlines of RTC agreements.

According to an industry trade association representative,[41] RTC deals never include outright monetary compensation. In the early days of RTC, a few broadcasters demanded cash and met instant rejection.[42] Instead, these arrangements generally involve reciprocal dealings. For example, it was not an accident that shortly after the major broadcast networks shifted to retransmission consent negotiations, most of them struck industry-wide cable agreements to create new cable networks with a network “brand”—e.g., CNBC, MSNBC, FNC. The broadcasters were anxious to expand into new video media, which resulted in new network-run cable channels. In some cases, cable operators received favorable terms under these agreements—for example, carriage rights to both a broadcast and a cable network for less than the cost of the former alone.

The key to these transactions was that the cable industry could give the networks something more valuable than small cash payments—that is, national coverage. (In some cases, these arrangements also exist between cable operators and strong non-network group-owned stations.) Cable operators claim that they do not agree to or continue to carry cable networks with little audience interest. And some networks have had little success in launching new cable networks, even with the help of RTC agreements.

The general counsel at a major cable company indicated that other types of deals also are customary.[43] Since systems generally have excess advertising time on cable satellite channels, they often give or sell it at nominal rates to local network affiliates for running promotional material for upcoming network programs. Alternatively, an RTC agreement may commit cable operators to buy promotional time from local stations, at relatively low rates. Or broadcasters and cable operators may agree to share unused production time in their studios, for nominal payments.

This combination of carrying broadcasters' cable networks, giving excess advertising time to broadcasters, and sharing production capacity may or may not have real economic value. As the cable general counsel above noted, “It's the principle rather than the economic value. No one wants to admit paying cash. There would be network carriage and advertising agreements in any event, but the existence of RTC encourages and increases it.”

[80] While the compulsory licensing system may have represented an unhappy truce in the 1970s, it has been replaced to a large extent by negotiated agreements between the broadcasters and owners of programming, and the cable as well as satellite operators who control access to most viewers. Like the jukebox compulsory license that eventually yielded to industry negotiations, perhaps the best compulsory licenses are the ones that fade away—which section 111 basically began to do after its first decade.

III. OTHER CONSIDERATIONS

In considering the treatment of new technologies within the overall framework of copyright, it is important to remember that copyright is not necessarily, or even principally, a barrier to the dissemination of creative works. As stated by the Supreme Court in Harper & Row Publishers, Inc. v. Nation Enterprises:[47] “it should not be forgotten that the Framers intended copyright itself to be the engine of free expression. By establishing a marketable right to the use of one's expression, copyright supplies the economic incentive to create and disseminate ideas . . .”

For example, ASCAP, perhaps the best existing model for a collective rights organization, was not created by a compulsory license set by Congress, but resulted from collective bargaining among the various parties, with periodic oversight by the courts through the lens of antitrust law,[48] and periodic adjustments of rights by Congress (as in the so-called “Fairness in Music Licensing Act of 1998”[49]).

An initial determination that a use is covered by copyright gives a copyright owner considerable leverage in setting the fees for distribution or performance of such works, of course, but the copyright owner makes no money if there are no distributions. And an initial determination that copyright does not extend to a particular use, such as in the case of jukeboxes, cable, or the Betamax, will shift the bargaining power in favor of the users in any later consideration of a compulsory license.

[82] On the other hand, a compulsory license is not the only means of placing limitations upon the rights of copyright owners. There are dozens of specific exceptions and limitations to the rights of copyright, including several in section 110[50] (covering certain “nonprofit” uses), and limitations resulting from basic principles of copyright, such as fair use, the idea-expression distinction, and the limitations upon copyright in works of utility. Many socially beneficial uses of copyrighted works on the Internet, even by people not owning the copyright, will be protected by these doctrines.

Although much maligned in the Internet community, the Digital Millennium Copyright Act (DMCA)[51] gives owners of works the right to control their works through copy protection systems and the use of copy management information systems. Anyone seriously considering a compulsory license will have to work through the interplay between such a license and the workings of the DMCA.

For example, would the existence of a compulsory license to disseminate works on the Internet trump the DMCA? Presumably not, unless we essentially want to dismantle the DMCA and require that copyright owners unlock their copyright protection systems. If the existence of a compulsory license lessened the economic value of copyrighted works, particularly those initially supplied in digital form, the net effect of a compulsory license might be to convince many copyright owners to adopt more technically intrusive copy protection systems—a result that would presumably undermine the whole purpose behind such a compulsory license.

One also must keep in mind the increasing international role in deciding copyright policy. Take, for example, the recently proposed “Public Domain Enhancement Act,”[52] introduced in Congress in 2003, that would impose a maintenance fee for continuing copyright beyond 50 years from first publication. Whatever the merits of such a requirement, it seems to fly directly in the face of the Berne Convention,[53] which prohibits such formalities as a limitation on copyright. It was only in 1988 that the United States finally did away with the requirement of copyright notice and registration, as a condition to joining Berne in the first place.[54]

Another recent international development of considerable relevance is the updating of the General Agreement on Tariffs and Trade to include intellectual property rights, under the new structure of the World Trade Organization. In a [83] recent decision,[55] a WTO panel held the U.S. exemption of certain restaurants and business establishments for retransmission of musical works received over the airwaves (section 110(5)) to be in violation of Berne obligations. The panel disapproved of national exceptions or limitations that “conflict with a normal exploitation of the work.” It is quite possible that too broad a compulsory license also would be in violation of Berne obligations, triggering possible retaliatory sanctions in the WTO.

Table I
Section 111 Fees Compared to Gross Basic Industry Revenues

1978-2002

Year	Cable Royalty Fees*	Basic Cable Revenue **
1978	$ 12,910,027	$ 1,147,000,000
1979	$ 15,889,793	$ 1,332,000,000
1980	$ 20,044,492	$ 1,615,000,000
1981	$ 30,886,119	$ 2,023,000,000
1982	$ 41,156,873	$ 2,515,000,000
1983	$ 72,774,961	$ 3,041,000,000
1984	$ 92,272,898	$ 3,534,000,000
1985	$ 104,777,269	$ 4,138,000,000
1986	$ 124,725,475	$ 4,887,000,000
1987	$ 163,163,192	$ 6,016,000,000
1988	$ 193,103,897	$ 7,345,000,000
1989	$ 208,126,070	$ 8,670,000,000
1990	$ 170,335,290	$ 10,174,000,000
1991	$ 180,755,077	$ 11,418,000,000
1992	$ 188,537,115	$ 12,433,000,000
1993	$ 185,359,636	$ 13,528,000,000
1994	$ 161,271,446	$ 15,164,000,000
1995	$ 165,867,789	$ 16,860,000,000
1996	$ 177,604,829	$ 18,395,000,000
1997	$ 154,389,741	$ 20,383,000,000
1998	$ 108,244,085	$ 21,830,000,000
1999	$ 108,240,071	$ 23,135,000,000
2000	$ 120,177,595	$ 24,729,000,000
2001	$ 121,845,046	$ 27,031,000,000
2002	$ 120,795,554	$ 28,492,000,000 ***

*Source US Copyright Office, July 2003
**Source U.S. Census Bureau, Statistical Abstract of the United States: 2002

***Kagan, World Media, a PRIMEDIA Company, Broadband Cable Financial Databook, 2002

Graph I

Cable Basic Gross Revenues

1978-2002

* Distinguished University Visiting Professor, Southern Illinois University Law School, 2005-2006. Professor of Law and Director, Media Center, New York Law School. B.A., 1966, Wesleyan University; J.D., 1969, Cornell University; LL.M., 1972, Columbia University; J.S.D., 1979, Columbia University.

** Intellectual Property Consultant, New York City, www.edwardsamuels.com. B.A. 1971, Yale University; J.D., 1974, Columbia University.

[1] Immunity could undermine the primary purpose of copyright law, which is to foster the creation of new works by granting authors exclusive rights in their works.

[2] Neil Weinstock Netanel, Impose a Noncommercial Use Levy to Allow Free Peer-to-Peer File Sharing, 17 Harv. J.L. & Tech. 1 (2003). In MGM v. Grokster, 545 U.S. ___ (June 27, 2005), the Supreme Court of the United States held that distributors of a file-sharing “device” can be held liable for contributory copyright infringement if their “object” is to “promote its use to infringe copyright.” Although the Court purported to be restating the reasoning of Sony Corp. of America v. Universal City Studios, Inc., 464 U.S. 417 (1984), the ground has shifted away from the amount of non-infringing use and to the intent of the distributor. Although the case was remanded for further proceedings, the Court found sufficient evidence of Grokster’s intent that it is hard to imagine the lower courts finding anything other than copyright liability. It can be expected that this finding of liability will encourage the promoters of a peer-to-peer compulsory license to work all the more diligently for such a compromise.

[3] As discussed below, for example, the cable television compulsory copyright license filled a gap by resolving disputes between copyright owners and cable operators for a little more than a decade while the multichannel industry was developing. As soon as relations between broadcasters and cable operators stabilized, however, the industries migrated to a private law system of negotiated settlements under “retransmission consent” statutory provisions.

[4] 7 U.S.C. § 1(e) (1909); 17 U.S.C. § 115 (2000).

[5] 17 U.S.C. § 116 (2000) (Former 17 U.S.C. § 116 repealed and replaced by this new § 116, December 17, 1993, 107 Stat. 2309).

[6] 17 U.S.C. §§ 1003–07 (2000).

[7] 17 U.S.C. § 114(d)–(h) (2000).

[8] See infra § II(A).

[9] 17 U.S.C. § 118 (2000).

[10] 17 U.S.C. § 119 (2000).

[11] 17 U.S.C. § 122 (2000).

[12]209 U.S. 1 (1908).

[13] See generally H.R. 2222, 60th Cong. (2d Sess. 1909).

[14] See 17 U.S.C. § 1(e) (1909); 37 C.F.R. § 255.3(m) (1998).

[15] To this extent, it thus resembles the system of retransmission consent in the cable industry. See discussion infra § II (B).

[16] 17 U.S.C. § 101 (2000); 17 U.S.C. § 115(d) (2000).

[17] 17 U.S.C. § 1(e), para. 3 (1909) (now superseded).

[18] 17 U.S.C. § 116 (1976) (now superseded).

[19] 37 C.F.R. § 254.3 (2003).

[20]17 U.S.C. § 116A (1988); 17 U.S.C. § 116 (1993).

[21] Robert A. Gorman & Jane C. Ginsberg, Copyright; Cases and Materials 608 (6th ed. 2002).

[22] The Berne Convention for the Protection of Literacy and Artistic Works, art. 11(1), Paris revision, July 24, 1971.

[23] Sony Corp. of America v. Universal City Studios, Inc., 464 U.S. 417 (1984).

[24] 17 U.S.C. §§ 1001–10 (2000).

[25] U.S. Copyright Office., The Annual Report of the Register of Copyrights, available at www.copyright.gov/reports/index.html. The Annual Reports of the Register of Copyrights for 2001, 2002, and 2003 state that AHRA fees were $3.32 million in 2000, $4.124 million in 2001, and $3.448 million in 2002.

[26] 17 U.S.C. § 114(d) (2000).

[27] Digital Millennium Copyright Act, Pub. L. No. 105–304, § 405(a)(1)–(4), 112 Stat. 2890 (1998).

[28]17 U.S.C. § 114(f)(1) (2000).

[29] Small Webcaster Settlement Act of 2002, Pub. L. No. 107–321, 116 Stat. 2780 (2002).

[30] See Ben Sisario, Old Songs Generate New Cash for Artists, N.Y. Times, Dec. 28, 2004. § E, at 1. For current information, see the website run by SoundExchange, the organization assigned the task of collecting and distributing the compulsory fees at www.soundexchange.com.

[31] Fortnightly Corp. v. United Artists, Inc., 392 U.S. 390 (1968) and Teleprompter Corp. v. Columbia Broadcasting Sys., Inc., 415 U.S. 394 (1974).

[32] The relevant gross revenues for the computation do not include payments from “on-demand” channels, but rather only on “basic” tiers with broadcast signals. Cablevision Systems Dev. Co. v. Motion Picture Ass’n of Am. Inc., 836 F. 2d 599 (D.C. Cir. 1988), cert. denied, 487 U.S. 1235. This creates a bit of a problem when a cable system includes distant signals in a higher or “enhanced” tier—which is uncommon. Since attempting to apportion a system’s revenues between broadcast and non-broadcast revenues would produce major transactional costs, however, first the Copyright Royalty Tribunal and now the Copyright Office have chosen simply to ignore these rare cases.

[33] Daniel L. Brenner, Monroe E. Price, & Michael Meyerson, Cable Television and Other Nonbroadcast Video: Law and Policy § 9.19 (Clark Boardman Callaghan) (1986).

[34] See infra § II B.

[35] Brenner, Price, & Meyerson, supra note 35, at § 9.19.

[36] 17 C.F.R. § 308.

[37] The FCC’s “must carry” rules were codified by Congress in the 1992 Cable Television Consumer Protection and Competition Act, 47 U.S.C. §§ 534–35. The statute and its implementation were upheld by the Supreme Court in Turner Broadcasting System, Inc. v. FCC, 512 U.S. 622 (1994) and Turner Broadcasting Sys., Inc. v. FCC, 520 U.S. 180 (1997).

[38] 17 U.S.C. § 111(d)(1)(B)(1) (2000). See, e.g., Brenner, Price, & Meyerson, supra note 35, at § 9:15; Ferris & Lloyd, Telecommunications Regulation: Cable, Broadcasting, Satellite, and the Internet, § 7.12(1) (LexisNexis 2004).

[39] Distant signals still are important in some circumstances, where a station in one market is particularly attractive in another–because of program content, language, or the like. For example, cable systems in Puerto Rico carry several New York City signals; because many Puerto Rican residents have friends or relatives in New York City, developments there naturally are of interest.

[40] 47 U.S.C. § 325(b) (2000).

[41] Confidential interview with senior management, cable trade association, December 14, 2004.

[42] Confidential interview with chief executive officer, cable multiple systems operator, November 18, 2004.

[43] Confidential interview with general counsel of cable multiple systems operator, December 17, 2004.

[44] See A Review of the Copyright Licensing Regimes Covering Retransmission of Broadcast Signals (August 1997), www.copyright.gov/reports/study.pdf.

The Copyright Office concludes that it would be inappropriate for Congress to grant Internet retransmitters the benefits of compulsory licensing. The primary argument against an Internet compulsory license is the vast technological and regulatory differences between Internet retransmitters and the cable systems and satellite carriers that now enjoy compulsory licensing. The instantaneous worldwide dissemination of broadcast signals via the Internet poses major issues regarding the national and international licensing of the signals that have not been fully addressed by federal and international policymakers, and it would be premature for Congress to legislate a copyright compulsory license to benefit Internet retransmitters.

Executive summary, www.copyright.gov/reports/exsum.pdf, at 13.

[45] 17 U.S.C. § 118 (2000).

[46] 17 U.S.C. § 122 (2000).

[47] 471 U.S. 539, 558 (1985).

[48] E.g., Broadcast Music, Inc. v. Columbia Broadcasting Sys. Inc., 441 U.S. 1 (1979).