The Authors Guild et al v. Google Inc.

Filing 261

MOTION to File Amicus Brief of Computer & Communications Industry Association. Document filed by Computer and Communications Industry Association. (Attachments: #1 CCIA Amicus Curiae Brief)(Schruers, Matthew)

UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK

The Authors Guild, Inc., Association of American Publishers, Inc., et al., Plaintiffs, v. Google Inc., Defendant. Case No. 05 CV 8136 (DC)

BRIEF OF AMICUS CURIAE COMPUTER & COMMUNICATIONS INDUSTRY ASSOCIATION (CCIA) ON PROPOSED SETTLEMENT

Matthew Schruers (pro hac vice) D.C. Bar No. 484839 Senior Counsel, Litigation & Legislative Affairs Counsel for Amicus Curiae Computer & Communications Industry Association (CCIA) 900 Seventeenth Street NW, Suite 1100 Washington, D.C. 20006 [phone] [e-mail]

TABLE OF CONTENTS Page TABLE OF CONTENTS............................................................................................. ii TABLE OF AUTHORITIES....................................................................................... iii ARGUMENT............................................................................................................... 1 I. About CCIA....................................................................................................... 1 II. Summary .......................................................................................................... 1 III. Competition Issues ........................................................................................... 2 A. Market Overview........................................................................................ 2 B. The Relationship Between Copyright and Monopoly .................................. 3 C. Competition Concerns in the New Economy ............................................... 5 1. Interoperability and Lock-in................................................................... 6 2. Leveraging, Tying, and Bundling........................................................... 8 D. Competition Concerns Appear to be Overstated.......................................... 9 1. The settlement aids market entry...........................................................10 2. "Orphan works" arguments..................................................................13 (a) The number of truly orphaned works ..............................................13 (b) The "market" for orphan works ......................................................14 3. Settlement Section 3.8(a) ......................................................................15 4. The Book Rights Registry.....................................................................16 IV. Other Speculative Concerns ...........................................................................16 V. Conclusion .....................................................................................................17

ii

TABLE OF AUTHORITIES CASES Page(s)

Atari Games Corp. v. Nintendo of America, Inc., 975 F.2d 832 (Fed. Cir. 1992)........................................................................... 6 Bateman v. Mnemonics, Inc., 79 F.3d 1532 (11th Cir. 1996) ....................................... 6 Broadcom Corp. v. Qualcomm, Inc., 501 F.3d 297 (3d Cir. 2007) .............................6, 7 Data Gen. Corp. v. Grumman Sys. Support Corp., 36 F.3d. 1147 (1st Cir. 1994)............................................................................ 8 Dowling v. United States, 473 U.S. 207 (1985) ........................................................... 4 DSC Communications Corp. v. DGI Techs., 898 F. Supp. 1183 (N.D. Tex. 1995), aff'd, 81 F.3d 597 (5th Cir. 1996) ........................................ 6 DSC Communications Corp. v. Pulse Communications, Inc., 976 F. Supp. 359 (E.D. Va. 1997), aff'd in part, rev'd in part, and vacated in part, 170 F.3d 1354 (Fed. Cir. 1999)......................................... 6 Eldred v. Ashcroft, 537 U.S. 186 (2003)....................................................................... 4 Harper & Row v. Nation Enters., 471 U.S. 539 (1985) ................................................ 4 Illinois Tool v. Independent Ink Co., 547 U.S. 28 (2006) .............................................. 8 New York Times v. Tasini, 533 U.S. 483 (2001) ........................................................... 2 Sony Computer Entm't, Inc. v. Connectix Corp., 203 F.3d 596 (9th Cir.), cert. denied, 531 U.S. 871 (2000)..................................................... 6 Sony Corp. of America v. Universal Studios, Inc., 464 U.S. 417 (1984).....................3, 4 Stewart v. Abend, 495 U.S. 207, 237 (1990) ................................................................. 3 Twentieth Century Music Corp. v. Aiken, 422 U.S. 151 (1975) .................................... 4 LEGISLATIVE MATERIALS 17 U.S.C. § 102(b) ....................................................................................................... 4 17 U.S.C. § 107............................................................................................................ 5 17 U.S.C. §§ 107-122................................................................................................... 5 H.R. Rep. No. 94-1476 (1976) ...................................................................................13

iii

OTHER MATERIALS III.B Philip Areeda & Herbert Hovenkamp, ANTITRUST LAW (3d ed. 2008) ................15 David Balto, The Earth is Not Flat: the Public Interest and the Google Book Search Settlement: A Reply to Grimmelmann (Jul. 22, 2009), available at <http://www.acslaw.org/node/13812>....................................13, 14 Dennis W. Carlton & Michael Waldman, The Strategic Use of Tying to Preserve and Create Market Power in Evolving Industries, 33 RAND J. OF ECON. 194 (2002) .................................................................... 9 Linda R. Cohen & Roger G. Noll, Intellectual Property, Antitrust and the New Economy, 62 U. PITT. L. REV. 453 (2001) ........................................... 8 Einer R. Elhauge, Why the Google Books Settlement is Procompetitive, Harvard John M. Olin Center for Law, Econ., & Bus. Discussion Paper No. 646 (Aug. 21, 2009), available at <http://ssrn.com/abstract=1459028>.......................................................... 10-12 Fed. R. Civ. P. 23......................................................................................................... 5 Jerry A. Hausman & J. Gregory Spivak, Google and the Proper Antitrust Scrutiny of Orphan Works, 5(3) J. COMPETITION LAW & ECON. 411 (2009) ...................................................................................... 10, 15 Herbert Hovenkamp et al., IP AND ANTITRUST (2004 Suppl.)....................................... 5 Michael L. Katz & Carl Shapiro, Antitrust in Software Markets, in COMPETITION, INNOVATION & THE MICROSOFT MONOPOLY: ANTITRUST IN THE DIGITAL MARKETPLACE (Eisenach & Lenard eds. 1998)..............................................7, 9 Mark A. Lemley, An Antitrust Assessment of the Google Book Search Settlement, (2009), available a <http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1431555>................ 10-14 Mark A. Lemley, IP Rights and Standard Setting Organizations, 90 CAL. L. REV. 1889 (2002) ........................................................................... 7 A Speech Delivered in the House of Commons, Feb. 5, 1841, reprinted in G.M. Young, ed., THOMAS B. MCCAULAY: PROSE & POETRY (1970) ........... 4 Peter S. Menell, An Analysis of the Scope of Copyright Protection for Application Programs, 41 Stan. L. Rev. 1045 (1989)........................................ 6 Report on Orphan Works: A Report of the Register of Copyrights (Jan. 2006) available at <www.copyright.gov/orphan/orphan-report.pdf> ................... 11, 15

iv

ARGUMENT Pursuant to the Court's orders dated November 14, 2008 (Order Granting Preliminary Settlement Approval) and April 28, 2009 (Order Granting Extension), the Computer & Communications Industry Association (CCIA) submits the following brief amicus curiae on the proposed Settlement Agreement in this case. I. About CCIA The Computer & Communications Industry Association is a non-profit trade association dedicated to promoting "open markets, open systems, and open networks." CCIA members participate in many sectors of the computer, information technology, and telecommunications industries and range in size from small entrepreneurial firms to the largest in the industry. CCIA members employ nearly one million people and generate annual revenues exceeding $200 billion.1 CCIA has been an advocate or litigant in the most significant disputes over information technology competition for nearly 40 years, including the telephone monopoly, mainframe computers, and PC operating systems. II. Summary This brief focuses principally on claims about competition issues surrounding the proposed settlement. The brief begins with a review of the market, considering the relationship between the copyright "monopoly" and antitrust. The brief identifies considerations about competition in "new economy," high-technology markets, and explains how aspects of the Book Search product ameliorate any competition concerns. The brief then reviews why competition A complete list of CCIA's members, which includes defendant Google, Inc. and other search engine providers, is available online at <http://www.ccianet.org/members.html>. By way of disclosure, it is stated that CCIA has no parent company of any kind, and no publicly held corporation has an ownership stake of 10% or more in CCIA. 0
1

concerns appear to be overstated, noting certain benefits of the settlement, and observing why competition concerns lack foundation. The brief concludes that alleged competition concerns provide no basis for disapproving this historic settlement.2 III. Competition Issues A. Market Overview Products such as Amazon's Kindle and the proliferation of personal communications devices have contributed to significant growth in the market for digital books.3 Yet, notwithstanding the Kindle's success, digital books still account for less than 1% of sales. With respect to out-of-print books, consumer demand is, not surprisingly, minimal. If demand were high, books would not go out of print. As such, estimates indicate that out-of-print books account for only 3% of sales. Nevertheless, the perceived success of the Kindle has prompted others ­ including Barnes & Noble, Sony, and Samsung ­ to announce competing devices in the "increasingly competitive digital book war," the combined non-public-domain libraries of which substantially exceed 1 The settlement is impressive in scope but not without precedent. On Sept. 27, 2005, the Southern District of New York approved a settlement in the litigation In re Literary Works in Electronic Databases Copyright Litigation, MDL No. 1379. Like the Book Settlement, the database settlement provided a prospective non-exclusive license to vast libraries of copyrighted works which had been digitized without advance authorization, and permitted licensees to sublicense. So sweeping was the MDL release that it also included unregistered works, the jurisdictional nature of which is now under review by the Supreme Court in Reed Elsevier v. Muchnick (08-103). No serious questions were raised, however, regarding the propriety of a class copyright settlement entailing a prospective license. Arguments that copyright should be uniquely immune from class action overlook this precedent and reflect a curious copyright exceptionalism that finds no support in federal law, and contradicts the Supreme Court's suggestion in New York Times v. Tasini that copyright infringement issues were capable of resolution on a collective basis. 533 U.S. 483, 505 (2001). 3 Although total sales remain a sliver of the overall market for literary works, sales estimates from the Association of American Publishers show a 57.8% growth rate from 2002-08. See Association of American Publishers, "2008 S1 Report, Estimated Book Publishing Industry Net Sales 2002-2008," available at <http://www.publishers.org>. 1
2

million.4 Existing competition recently prompted Sony to cut prices on best-sellers in the library affiliated with its product.5 At the same time, the sale of printed books online, enabled by industry veterans such as Barnes & Noble and Amazon.com is now being supplemented with innovative new print-on-demand services, some of which will make hundreds of thousands of never-before-accessible books available to the public.6 It is into this market that Google's product will eventually enter.7 B. The Relationship Between Copyright and Monopoly Like these competing services, Google is not a rights-holder, but a licensee. Unlike the copyright holders whose works Google has a non-exclusive license to use, Google does not have a right backed by the federal government to exclude others from accessing, displaying, reproducing, or distributing the books at issue. Thus, the only "monopoly" before this Court is the federal privilege granted to the plaintiffs. As recipients of a "monopoly privilege" over their protected work, whose "every [unauthorized] commercial use... is presumptively an unfair exploitation", Stewart v. Abend, 495 U.S. 207, 237 (1990) (quoting Sony Corp. of America v. Universal Studios, Inc., 464 U.S. 417, 451 (1984))(brackets in original), the plaintiffs are encouraged by federal law to exclude others from the market for their individual work. This is See Michelle Megna, "Sony Plans New Kindle Rival, Adopts Open Format," Internet News, Aug. 13, 2009, available at <http://www.internetnews.com/ec-news/print.php/3830966>. See also Michelle Megna, "Barnes & Noble's Reply to Kindle to Debut in 2010," Internet News, July 21, 2009, available at <http://www.internetnews.com/ec-news/print.php/3830966>; "Samsung takes on giant Amazon in e-books," Korea Herald, July 28, 2009, available at <http://www.koreaherald.co.kr/NEWKHSITE/data/html_dir/2009/07/28/200907280064.asp>. 5 See Brad Stone & Mokoto Rich, "Sony to Cut E-Book Prices and Offer New Readers," N.Y. Times, Aug. 4, 2009, available at <http://www.nytimes.com/2009/08/05/technology/personaltech/05sony.html>. 6 See, e.g., "Agreement With Amazon Will Make U-M Digital Books Widely Available," July 21, 2009, available at <http://www.lib.umich.edu/news/amazon-agreement>. 7 Google's product will be further limited by the fact that while public domain works can be downloaded, in-copyright books will be text-only, online-only. 2
4

not to say, however, that every rightsholder violates antitrust laws; the market controlled by the rightsholder of any given book is narrow and is normally disciplined by the substitutability of competing books, as well as other literary works available in electronic and physical form.8 Thus, the cost of exclusion imposed on society has long been perceived in Anglo-American law as a necessary evil ­ "a tax on readers for the purpose of giving a bounty to writers."9 The ability to exclude has been one of the tools in the U.S. innovation policy toolbox since 1790. The Supreme Court has long recognized that Congress intended for "the monopoly created by copyright" to "reward[] the individual author in order to benefit the public." Harper & Row v. Nation Enters., 471 U.S. 539, 546 (1985) (quoting Sony). For public policy reasons, the scope of this monopoly is subordinated to broad policy objectives. "The limited scope of the copyright holder's statutory monopoly... reflects a balance of competing claims upon the public interest: Creative work is to be encouraged and rewarded, but private motivation must ultimately serve the cause of promoting broad public availability of literature, music, and the other arts." Twentieth Century Music Corp. v. Aiken, 422 U.S. 151, 156 (1975). Because "the ultimate aim" of copyright is "to stimulate artistic creativity for the general public good", id., Congress can and has modified the bounds of the copyright privilege over time. At times Congress expands it, such as extending terms to the prodigious lengths that have Copyright doctrines like the idea-expression dichotomy further support this substitutability, and thus normally prevent rightsholders from abusing the statutory monopoly. Thus, while the `Harry Potter' rightsholders might exclude others from reproducing the works of the Potter franchise, their section 106 rights are "a series of carefully defined and carefully delimited interests to which the law affords correspondingly exact protections", see Dowling v. United States, 473 U.S. 207, 216 (1985), and therefore do not extend to other fictional works about boy wizards destined for greatness. As a result, the Potter rightsholders' ability to price at abovemarket levels for fictional works is undercut by the ease with which similar but non-infringing works about wizardry can enter the market. See 17 U.S.C. § 102(b); see also Eldred v. Ashcroft, 537 U.S. 186, 219 (2003) (noting idea-expression dichotomy as one of copyright's built-in First Amendment safeguards). 9 A Speech Delivered in the House of Commons, Feb. 5, 1841, reprinted in G.M. Young, ed., THOMAS B. MCCAULAY: PROSE & POETRY (1970). 3
8

contributed to the orphan works problem, and at other times it has narrowed it with exceptions, see, e.g., 17 U.S.C. §§ 107-122, including the fair use exception, codified in 1976, that underlies this litigation. Id. § 107. In sum, the monopoly before the Court is not the privately negotiated, non-exclusive license embodied in the settlement, but rather the privilege granted by Congress to the plaintiffs for policy reasons. Although long in term, that privilege is narrow in scope, and the contours of the term and scope are well within Congress's powers to modify should it see fit to do so. Given these limitations, a mere licensee ­ least of all a non-exclusive licensee ­ has little ability to abuse this federal privilege.10 C. Competition Concerns in the New Economy As a trade association that advocates for open markets in the high technology industries, CCIA has identified various phenomena that suggest the presence of threats to competition. When these phenomena are apparent in a market, transaction, or business practice, CCIA advocates aggressively in support of a competitive marketplace. Because the characteristic threats to competition in high-tech markets are not evident in this case, CCIA views the stated concerns about competition with skepticism. There is little evidence to support the proposition that some competition concern necessitates disapproval of the settlement under Fed. R. Civ. P. 23. At present, CCIA is unaware of any potential anticompetitive scenario which could evolve that would be beyond this Court's capacity to resolve through its continuing supervision of the

As leading scholars of IP and antitrust note in the context of patents (where the right at issue is broader than copyright), "[t]he granting of a non-exclusive license itself almost never harms competition, regardless of the presence or absence of any IP dispute." Herbert Hovenkamp et al., IP AND ANTITRUST, § 7.4a (2004 Suppl.).

10

4

settlement.11 The following highlights traditional competition concerns that CCIA has identified in high-technology markets. 1. Interoperability and Lock-in In complex-product markets, interoperability and open standards are crucial to promoting competition. Two technology products can interoperate only if they conform to the same set of rules, or "interface specifications." If a company could exercise proprietary control over the interface specifications implemented by its products, that company could determine which products made by other firms ­ if any ­ could interoperate with its software, which would pose risks to consumer welfare.12 Prohibiting competitors from accessing the de facto standard interface specifications could lock users into a platform, operating system, hardware solution, or network software environment. The provider of the service may thereby attempt to leverage less competitive products, and inhibit users from transferring data between different environments. Accordingly, interoperability is a key antidote to anticompetitive behavior, and efforts to prevent or enable interoperability should be a key indicator of whether anticompetitive motives are at work. Numerous federal appellate courts have refused to allow intellectual property to interfere with interoperability, so as to promote competition.13

Settlement § 17.23. See, e.g., Peter S. Menell, An Analysis of the Scope of Copyright Protection for Application Programs, 41 Stan. L. Rev. 1045, 1082, 1097 n.281 (1989). 13 See, e.g., Atari Games Corp. v. Nintendo of America, Inc., 975 F.2d 832 (Fed. Cir. 1992); Bateman v. Mnemonics, Inc., 79 F.3d 1532 (11th Cir. 1996); DSC Communications Corp. v. DGI Techs., 898 F. Supp. 1183 (N.D. Tex. 1995), aff'd, 81 F.3d 597 (5th Cir. 1996); DSC Communications Corp. v. Pulse Communications, Inc., 976 F. Supp. 359 (E.D. Va. 1997), aff'd in part, rev'd in part, and vacated in part, 170 F.3d 1354 (Fed. Cir. 1999); Sony Computer Entm't, Inc. v. Connectix Corp., 203 F.3d 596 (9th Cir.), cert. denied, 531 U.S. 871 (2000); See also Broadcom Corp. v. Qualcomm, Inc., 501 F.3d 297 (3d Cir. 2007) (patent dispute).
12

11

5

CCIA has previously taken issue with anticompetitive behaviors relating to interoperability, particularly with respect to interface specifications, licensing, and standards ­ which may include file formats. Competitive concerns involving interoperability may arise in situations such as: (a) when a company asserts IP rights on interface specifications or standards and refuses to license on a royalty-free or at least a reasonable and non-discriminatory basis; (b) when a company deceives competitors ex ante about whether a specification or standard contains IP,14 (c) when a company affirmatively discourages interoperability efforts; (d) when a company intentionally provides nominally open reference schema for a standard with incomplete information or proprietary elements therein, or; (e) when a company first openly licenses IP related to a specification or standard and then subsequently withdraws that license without any reasonable business justification once it has become a de facto standard, leaving licensors with few options but to pay to use the newly-closed system.15 Google, however, has engaged in business deals in the digital book space that employ the widely accessible, non-proprietary ePub file format, which is based upon open standards, and makes over a million public domain works downloadable in the same file format, as well as in the widely adopted PDF file format.16 A commitment to an open file format indicates that a

See Broadcom, 501 F.3d at 313 ("Misrepresentations concerning the cost of implementing a given technology may confer an unfair advantage and bias the competitive process in favor of that technology's inclusion in the standard."). 15 See Mark A. Lemley, IP Rights and Standard Setting Organizations, 90 CAL. L. REV. 1889, 1899 (2002); Michael L. Katz & Carl Shapiro, Antitrust in Software Markets, in COMPETITION, INNOVATION & THE MICROSOFT MONOPOLY: ANTITRUST IN THE DIGITAL MARKETPLACE 65 (Eisenach & Lenard eds. 1998). 16 See Chris Snyder, "Sony-Google E-book Deal A Win for ePub, Openness," Wired Magazine, Mar. 19, 2009, available at <http://www.wired.com/epicenter/2009/03/sony-googlee-b/>. In this context, Google has stated "We believe in an open platform for accessing and reading books". The court's continuing jurisdiction over this settlement should ensure that Google does not forsake this commitment. See also Brandon Badger, "Inside Google Books: Download Over a Million Public Domain Books from Google Books in the Open EPUB 6

14

product or service is less likely to prevent competitors from interoperating. Google also supports Creative Commons licensing, providing users an easy mechanism to choose from a wide variety of permissive licensing options, should they choose to distribute works under such terms. Moreover, the fact that the book service will be accessible through a general-purpose web browser means that any web-enabled device can access it. Because of the general openness of the Google Book platform, there presently appears to be minimal risk of competition problems relating to interoperability or lock-in. 2. Leveraging, Tying, and Bundling Leveraging occurs when a developer of a platform or product with significant market power attempts to exert influence in a related market by altering the dominant product to diminish functionality or interoperability, so as to make it less functional with other companies' ancillary products and more functional with the dominant firm's ancillary products. Leveraging can also be enhanced through the use of IP rights, some examples of which are noted above. In some circumstances, antitrust policy should preclude rightsholders from leveraging IP rights into other markets.17 To this end, federal courts have ruled against such licensing practices that lacked business justification.18 Tying occurs when a party refuses to sell one product unless the consumer also purchases another, usually complementary product. Bundling occurs when two products are packaged together at a price less than their combined, individual prices. While both practices are often unobjectionable, and can in many cases have positive effects on consumer Format," Aug. 26, 2009, available at <http://booksearch.blogspot.com/2009/08/download-overmillion-public-domain.html> (discussing available file formats). 17 See Linda R. Cohen & Roger G. Noll, Intellectual Property, Antitrust and the New Economy, 62 U. PITT. L. REV. 453, 469 (2001), but see Illinois Tool v. Independent Ink Co., 547 U.S. 28, 46 (2006). 18 See, e.g., Data Gen. Corp. v. Grumman Sys. Support Corp., 36 F.3d. 1147, 1183 (1st Cir. 1994) ("In general, a business justification is valid if it relates directly or indirectly to the enhancement of consumer welfare.") 7

welfare, they can also be used to foreclose competition in the ancillary market, particularly in highly innovative markets.19 These traditional concerns do not arise with respect to the settlement at issue. Google's approach to the Book Search project is characterized by considerable openness to third parties. Unlike some of its competitors, the existing book service is not specifically tied to a proprietary device, nor have users of the existing book service been forced to purchase unrelated products or services through unnecessary bundling. Additionally, Google has made publicly available a variety of application programming interfaces (APIs), which enable the product to be integrated into third party sites. These APIs provide considerable functionality to third parties, including performing full-text searches, retrieving information, previewing, and linking.20 Instead of walling users into a proprietary ecosystem, the current service is a broadly open platform. D. Competition Concerns Appear to be Overstated. Thus far, vague concerns have been voiced about a potential "monopoly" resulting from the settlement agreement. To begin with, as noted above, the monopolies at issue are the rights in the underlying works, which are narrow in scope, and which can be modified by Congress as it chooses. A studied analysis of the "monopoly" concerns raised suggests that they do not merit rejecting the settlement.

See Katz & Shapiro, supra note 15, at 68-71; see also Dennis W. Carlton & Michael Waldman, The Strategic Use of Tying to Preserve and Create Market Power in Evolving Industries, 33 RAND J. OF ECON. 194, 194-220 (2002). 20 These APIs have been utilized by parties such as Stanford, the University of California, and Buy.com. See "Google Book Search APIs," available at < http://code.google.com/apis/books/>. 8

19

1. The settlement aids market entry. Antitrust problems manifest when there are barriers to entry that prevent potential competitors from entering the market.21 Undoubtedly, the cost of scanning works is high, the rights-clearance process difficult, and the profits speculative, but an antitrust analysis of this settlement must focus on whether it create new barriers to entry ­ which must also be weighed against any reduction of barriers to entry.22 The difficulties in entering this market spring not from the settlement but rather the presently inescapable attributes of book scanning, some of which are a direct product of copyright law. Stated otherwise, the labor and technological costs that would confront any similar undertaking are inherent in book scanning. Book scanning may be hard, but the proposed settlement makes it no harder. As economists Hausman and Spivak note, estimates of the cost of entry into this market are on the order of $100 million, which although significant could be mustered by Google competitors.23 (Indeed, major technology firms regularly dole out such amounts to resolve dubious patent infringement claims.) The settlement agreement will ease entry into the market for several reasons. The settlement establishes the not-for-profit Books Rights Registry ("BRR" or "the Registry"),24 which is charged with clarifying the status of in-copyright, out-of-print works.25 The Registry will be directed to seek out rightsholders of in-copyright books, and also pays out royalties, thus

See Mark A. Lemley, An Antitrust Assessment of the Google Book Search Settlement, at 6 (2009), available at <http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1431555> (hereinafter "Lemley"). 22 Id. at 5. 23 Jerry A. Hausman & J. Gregory Spivak, Google and the Proper Antitrust Scrutiny of Orphan Works, 5(3) J. COMPETITION LAW & ECON. 411, 412 (2009) (hereinafter "Hausman & Spivak"). 24 Lemley at 2. 25 See Einer R. Elhauge, Why the Google Books Settlement is Procompetitive, Harvard John M. Olin Center for Law, Econ., & Bus. Discussion Paper No. 646, at 10 (Aug. 21, 2009), available at <http://ssrn.com/abstract=1459028> (hereinafter "Elhauge"). 9

21

creating for the first time an incentive for rightholders to step forward and claim potentially orphaned works.26 Because it may be difficult in some cases to determine who, if anyone, maintains the rights in an out-of-print work, the resolution of gridlock or uncertainty regarding a work's rights status would benefit the entire marketplace. Because the settlement is nonexclusive, potential Google competitors may take advantage of the Registry's labors and license newly-clarified books for competing services. Competitors can also obtain information on the status of orphaned works so as to assess the risk of providing those works.27 Just as the Registry will help reunite orphan works with their rightsholders, it will also help address the equally significant problem of uncertainty and rights gridlock. For example, author-rights reversion and publisher contracts may leave unclear which party retains certain rights, and yet none of the potential rightsholders have sufficient economic interest to address this problem. By resolving uncertain or gridlocked claims over works, the Registry will reduce the costs of licensing dormant yet non-orphaned works. This feat could not be achieved by orphan works legislation, and would therefore likely complement any eventual orphan works legislation. In fact, the Copyright Office has suggested that such activity would be "indispensable" to any orphan works solution.28

Elhauge at 2. See Lemley at 7. Potential competitors can draw conclusions about infringement liability exposure by analyzing how long an apparently orphaned work has remained unclaimed notwithstanding royalties available for collection. 28 Report on Orphan Works: A Report of the Register of Copyrights (Jan. 2006), at 106 available at <www.copyright.gov/orphan/orphan-report.pdf> ("It is important to stress, however, that we believe that registries are critically important, if not indispensable, to addressing the orphan works problem.... It is our view that such registries are better developed in the private sector, and organically become part of the reasonable search by users by creating incentives for authors and owners to ensure that their information is included in the relevant databases.") (emphasis supplied). See also "Google's Big Book Case," The Economist, Sept. 3, 2009, available at <http://www.economist.com/opinion/displaystory.cfm?story_id=14363287> (lauding fact that settlement will help "to resolve the legal status of many texts subject to
27

26

10

Status clarification is therefore a positive externality whose benefits the settling parties cannot fully capture. Because potential book-scanning competitors cannot be excluded from utilizing knowledge about a work's status acquired by the Registry,29 the settlement lowers the cost of entry of follow-on book scanning initiatives. By thus lowering risks and licensing costs, the settlement may encourage competitors to enter the market, or may encourage participants who exited the market to reenter.30 For instance, an orphan work that has not been claimed after a period of time will indicate to a potential competitor that the rightsholder lacks interest or ability, thus lowering the potential risks inherent in offering the work without a license.31 Google's project also makes available digitized versions of Copyright Office records. Not only will this increase public access to government records, a valuable object in its own right, it will also help decrease the universe of orphaned works by identifying the set which did not comply with formalities and have therefore entered the public domain.32 Finally, non-approval of the settlement would actually limit consumer choice by prohibiting market "entry" by out-of-print, in-copyright works. Consumers who can choose from these newly available works will have more choices than prior to the settlement.33

absurdly long copyright periods and murky ownership"). 29 See Settlement § 6.6(d). 30 See Elhauge at 11. 31 Id. This would be desirable where the intended use is likely a permitted or privileged use under a limitation and exception, but nevertheless entails litigation risks. 32 Lemley at n.6. See also "U.S. Copyright Renewal Records Available for Download," June 23, 2008, available at <http://booksearch.blogspot.com/2008/06/us-copyright-renewal-recordsavailable.html>. 33 While Google's license to make available any truly orphaned works will be unique in the marketplace, it bears noting that such works will also be available at a free Public Access Service terminal in each public library. Thus, whatever supposed premium Google might charge for books that no one appears to want will be checked by the free availability of that work at the public library. For similar reasons, claims about cartelization are also exceedingly improbable: in-copyright works will be disciplined by numerous sources of "free" works ­ Public Access 11

2. "Orphan works" arguments Some critics have expressed concerns regarding Google's license to use orphan works. This counterintuitive argument ­ that Google's license to use books that no one wanted would create a product with which no one can compete ­ fails to surmount several factual hurdles. (a) The number of truly orphaned works This criticism likely overstates the amount of orphan works that currently exist and will exist after implementation of the settlement.34 As noted above, the settlement contains mechanisms designed to decrease the number of orphan works, such that whatever control would exist will diminish over time as the Registry's labors bear fruit and copyrights expire. By one estimate, there may be as few as 1.4 million true orphan works and that Google has likely only scanned about 580,000.35 Rights-clarification may well establish that the failure to renew copyrights before 1976 caused many works to enter the public domain. The legislative history of the 1976 Copyright Act suggests that 85% of works may not have been renewed,36 meaning that the rights clarification process may dramatically expand the universe of public domain works available to the public. As one commentator noted, "the combined effect of these rights clarification efforts and financial incentives will be to clarify the copyright status of hundreds of Service terminals, public domain books, and scholars who publish for reputational gains, rather than profit. See Hausman & Spivak at 434. 34 See David Balto, The Earth is Not Flat: the Public Interest and the Google Book Search Settlement: A Reply to Grimmelmann (Jul. 22, 2009), available at <http://www.acslaw.org/node/13812> (hereinafter "Balto"). 35 Lemley at 6. Given the imprecision with which the term "orphan works" is used, it is difficult to validate the scope of any alleged monopoly over them. It bears noting, however, that a substantial portion of works which are in-copyright but out-of-print are not orphaned, and that the set of nominally orphaned works will diminish once royalties are available. At least for the purposes of this settlement, "orphan works" will only consist of works that no one has been willing to keep track of nor, after implementation of the settlement, come forward and claim rights to, in spite of available royalties. See Lemley at 6. 36 H.R. Rep. No. 94-1476, at 136 (1976). According to Lemley, a 1961 Copyright Office study found that less than 10% of works complied with renewal requirements. See Lemley at n.6. 12

thousands, if not millions of works, which will be an enormous improvement over the status quo."37 CCIA has thus far reviewed no evidence indicating that the absolute number of orphan works will not substantially diminish as a result of the Registry's efforts. (b) The "market" for orphan works Criticism of the settlement appears to implicitly assume there is a separate market for orphaned books over which exclusive control would be economically relevant, an assumption that does not survive scrutiny. Surely, consumers do not differentiate books into separate markets based on the clarity of who owns the rights. Further, the fact that a work is orphaned suggests not only that is it substitutable for other works, but that the work may have gone out of print because other works were superior. At the least, this strongly suggests that many orphaned works are comparable to works that haven't been orphaned, and thus do not constitute their own market.38 Not only is there no separate market for orphaned works, there may be no market at all. As Authors Guild President Roy Blount noted, orphaned works by definition provided little commercial appeal for print publishers.39 Allegations of "monopoly", however, imply that orphaned works represent a vast mine of untapped economic potential. Whatever criticisms might be reasonably voiced about the publishing industry, this proposition ­ that the industry so fundamentally misunderstands its own business as to leave lying fallow hundreds of thousands of profitable works ­ is the most sweeping. Yet scant evidence supports this view. Indeed, a review of the Copyright Office's 2006 orphan works report suggests that the orphan works problem poses a greater threat to research, preservation, and study than it does to the production and

37 38

Balto, supra note 34. See Lemley at 6. 39 Balto, supra note 34. 13

marketing of new copyrighted works.40 Clarifying the rights thus may serve a significant societal interest, preserving decaying works and enabling study and research efforts currently stymied by uncertainty, but the total economic gains are purely speculative. 3. Settlement Section 3.8(a) Some concerns have been voiced about Section 3.8(a) of the Settlement agreement, a provision referred to at times as a "most favored nation" provision. Economic literature on "MFN" clauses suggests they can have both positive and negative effects.41 Whatever effect this provision might have, it will be minimal. This provision applies only to unclaimed works, and only should the Book Rights Registry subsequently be authorized to license these works. This condition may never be satisfied within the 10-year period that it will operate. Even if the provision should operate at some future date, the clause pertains only to unclaimed works ­ works which evidence suggests have the least expected economic value. Further, this provision does not disable the BRR from striking better deals with Google's competitors for in-print works, which as noted previously, represent the overwhelming share of book sales. In addition to being very narrow, this obligation only requires equal treatment, not preferred treatment.42 Misconceptions surrounding this provision highlight an important observation, however: Even if they were valid, complaints that the settlement enables anticompetitive behavior should not raise Rule 23 concerns regarding fairness to the class. An anticompetitive environment works injustice, but to buyers, not the seller ­ which in this case is the plaintiff class. Assuming (absent any evidence) that monopoly rents could be extracted through the settlement, this would

Report on Orphan Works: A Report of the Register of Copyrights (Jan. 2006) available at <www.copyright.gov/orphan/orphan-report.pdf>. 41 See Hausman & Spivak at 429. 42 Requirements of equal treatment are rarely objectionable. See III.B Philip Areeda & Herbert Hovenkamp, ANTITRUST LAW ¶ 768a6, at 159-60 (3d ed. 2008). 14

40

actually yield more revenue over which the BRR could negotiate.43 One cannot argue that the settlement is unfair to the class by alleging it would permit rightsholders to act in anticompetitive ways that benefit them. For this reason, these claims are inappropriate for consideration in the context of class fairness. 4. The Book Rights Registry Like Google, the non-profit Book Rights Registry will not be a rightsholder per se. Given the BRR's unique position, this proposed entity merits separate consideration in the context of the fairness analysis. At present, however, competitive concerns about the BRR are not justified. Speculative claims about potentially anticompetitive misbehavior of a virtually non-existent entity are hardly a sound basis for penalizing the BRR before it exists, let alone disapproving the settlement. Indeed, this is another example of the fundamental tension in the various objections to the settlement: insofar as one entertains the pure speculation of objectors, the proposed BRR cannot simultaneously be unfair to the class and anticompetitive, since any anticompetitive conduct that the BRR engaged in would benefit class members, if anyone. Ultimately, the BRR should be judged before this Court and the law on the basis of its actions, as it will be judged by rightsholders, who if unsatisfied with BRR's performance, can still exclude books from sale or display and pursue a separate deal directly with a Google competitor, or cut BRR out entirely by pursuing a deal directly with Google. IV. Other Speculative Concerns An assortment of conflicting agendas is urged upon the Court. Some critics suggest the settlement should be disapproved because it constitutes a sort of legislative exercise, for For this reason, the settlement appropriately requires Google to ensure that the institutional subscription promotes "the realization of broad access to the Books by the public, including institutions of higher education." Settlement § 4.1(a)(i).
43

15

example, whereas others urge the Court to rewrite the parties' agreement to implement assorted public policies as if the settlement were in fact legislation. Some assert the deal is not good enough for certain interests; others claim it is too good. This lack of consensus among objectors may itself be evidence that an equitable compromise was struck between the parties. There appears to be little need to graft onto the parties' agreement a response to every contingency and hypothetical situation. To the extent that changes to the agreement are proposed, the litmus test should not be whether a proposed change would, by some subjective measure, improve the agreement, but rather whether absent the proposed change, would the problem be incapable of resolution via the settlement's arbitration provision or the continuing jurisdiction of this Court. V. Conclusion The Google Book Search product has offered unprecedented access to a wealth of information, and the proposed settlement before the Court offers considerable benefits to the public by expanding the scope of that project. The settlement also benefits the market by creating an open platform and legal mechanisms that will encourage investment and growth in book digitization. For the reasons discussed herein, allegations about competition provide no basis for disapproving the parties' settlement. Respectfully submitted, /s/ Matthew Schruers (pro hac vice) D.C. Bar No. 484839 Senior Counsel, Litigation & Legislative Affairs Counsel for Amicus Curiae Computer & Communications Industry Association (CCIA) 900 Seventeenth Street NW, Suite 1100 Washington, D.C. 20006 [phone] [e-mail] 16

17