• Topics
  • eCommerce
  • eDiscovery
  • CyberCrime
  • International
  • Media & Communications
  • Patents & Technology
  • Health & Biotechnology
  • Copyrights & Trademarks
• Submissions
• About DLTR
  • Contact Us
  • Masthead
• Subscriptions

PRIVATE ORDERING AND ORPHAN WORKS: OUR LEAST WORST HOPE?

Abstract

The political capture of copyright law by industry groups has inadvertently led to orphan works problems arising in less organized industries, such as publishing. Google Book Search (GBS) is a prime example of how private ordering can circumvent legislative inefficiencies. Digital technologies such as GBS can open up a new business model for publishers and other content industries, centered around aggregated rights holdings. However, the economic inertia that private ordering represents may pose a threat to the knowledge-oriented goals of copyright law.

Introduction

¶          Congress has been accused of many things, but initiative is generally not one of them. The current state of copyright law, with wildly longer term limits and automatic protection, is a result of continuous content-industry lobbying to protect their valuable, aging intellectual property. As a result, the problem of orphan works--where copyright holders are untraceable, and strict liability prevents the work’s further use--has arisen, particularly in less centralized industries, such as publishing. Private ordering offers an path around the Sisyphean task of legislative change, as parties seek to optimize economic benefit through private contracting or settlement.

¶          Google Book Search (GBS) is a prime example of such ordering at work. GBS holds promise as potentially being able to surmount the orphan works problem, leveraging Google’s financial might to assume the liability risks. Google also has the potential to redefine the publishing industry model, creating a payment modeled around small access payments from a large library of archived works. However, this model has potentially worrisome implications for the state of copyright law and for the future of the public domain.

I. Collective Action and Pressure Groups

¶          The emergence of so-called "pressure groups"--political organisms designed to promote particular economic interests--is arguably the most significant development in American politics in the last century.² Their emergence has called into question the independence of the "public interest" from private economic interests.³ The growth of government results in increased "differential or discriminatory impact on the separate and identifiable groups of the population," which incentivizes investment in obtaining favorable differential gains through political lobbying.⁴ Buchanan and Tullock explain the cyclical nature of this relationship: "[t]he organized pressure group thus arises because differential advantages are expected to be secured through the political process, and in turn, differential advantages for particular groups are produced because of the existence of organized activity."⁵ In other words, the success of one organized pressure group successfully disrupting the prior "general welfare" equilibrium incentivizes further collective action, as more groups look to capture their political interest niche.⁶ Economic theory suggests this will eventually lead to a new equilibrium, given full organization.⁷ This new balance--assuming it exists--does not mean that discriminatory legislation will cease to be adopted.⁸ Interest group activity is a function of net profit expected from the political process.⁹ Given unequal power balances within particular market segments, pressure groups can secure legislation that provides benefits that "are not applied generally to the whole population."¹⁰

¶          This power imbalance manifests in modern copyright law. The consolidation and growth of copyright-reliant industries has not been met with a suitably large political counterweight.¹¹ The result is a spiral effect, where copyright laws have become increasingly more beneficial to interested pressure groups.¹² 

II. Copyright Expansion and Orphan Works

¶          Over the past half-century, content industries have gained control of the legislative pipeline for copyright to a degree likely unsurpassed by any other modern cartel.¹³ The 1976 Copyright Act , for instance, was entirely drafted through a series of negotiations--at the urging of the U.S. Copyright Office--between authors, publishers, and other content owners.¹⁴ More recently, the 1998 Sonny Bono Copyright Term Extension Act (CTEA) pitted studios and publishing interests, led by Disney, against a coalition of "college professors, constitutional lawyers, librarians[,] and small town school teachers."¹⁵ Since the 1960s, Congress has expanded the copyright term eleven times, after expanding it just twice from 1790 to 1960.¹⁶ The economic size of the copyright industries should not be understated: At the time of the CTEA’s passage, they made up nearly six percent of American gross domestic product, and one of the U.S.’s largest exports.¹⁷ Disney in particular stood to lose control of billions of dollars worth of copyrights-- Mickey Mouse and Winnie-the-Pooh alone were valued at nearly $8 billion dollars each--if the CTEA was not passed.¹⁸ In simple terms, the cartel-like organization of the content industries, coupled with a lack of an economically significant lobbying opponent, has led copyright inexorably down the road of increased term protection.

¶          As copyright protection began to expand, and particularly as it expanded well past the death of the author, the problem of orphan works arose. Most significant to the explosion of the orphan problem was the dropping of the renewal requirement. This shifted copyright from an "opt-in" system, which required content creators to actively maintain copyright, to an "opt-out" system.¹⁹ Thus, works that might enter the public domain due to a low commercial value now remain covered by copyright.²⁰ With such a large population of copyrighted works, there are some works that have nigh-untraceable origins. However, the inability to locate a copyright holder does not protect users from the strict liability hammer of copyright law.²¹ Whatever value those works originally had as foundational materials for other works is then lost, as future use is chilled by the possibility of litigious parents returning to protect their (previously valueless) orphan work.²²

¶          The issue of orphan works arises largely as a side effect of content-industry efforts to expand copyright protection, and lingers largely because there is no group with a significant enough incentive to push for such a change. Large content-providers meticulously catalogue their copyrights, and in any event are fairly easy to track down for licensing.²³ Orphan works generally tend to be either below their radar or within their contingency budgets.²⁴ On the other hand, a creator of a derivative work does not know that she has encountered an orphan works problem until she actually does, and cannot find the current copyright holder of a work she wishes to use.²⁵ This uncertainty surrounding orphan works encounters inhibits the creation of an interest group to address the issue, as the potential members of such a group are not only nebulously defined and spread out, but also may lack the will or wherewithal to pursue the issue further. This creates a classic problem of collective action, in which "large collectivities with diffuse interests" (academics, librarians, etc) are "systematically disadvantaged in the political process as compared to smaller groups with more acute interests" (Big Content) because "larger groups face higher organizing costs and are affected more severely by incentives to free ride."²⁶ 

¶          Further, the number of orphan works actually in existence is virtually indiscernible.²⁷ The fluid social mobilization through orphan works-related obstacles further magnifies the collective action problem. While an orphan work issue may prevent some people from creating a work, others may move on and utilize a substitute work, and the orphan works problem encountered becomes untraceable. A calculation of the orphan works population would almost invariably be reliant on the number of unsuccessful or abandoned private searches by any and every potential user. Further, gauging the severity of the problem would require examining the preventive impact orphan works have on the ability to create a new work--whether users find a substitute or just accept the risk of suit. As such efforts to estimate the scope of orphan works problems generally dissolve into speculation.²⁸

¶          Further, the scale of orphan works problems vary widely across various content formats.²⁹ For instance, within the relatively consolidated music and movie industries, a serious effort has been made to create databases and clearinghouses which provide owner information for copyright protected works.³⁰ There is abundant economic incentive in these industries for such clearinghouses, especially in music, where the explosion of sampling-based production enables works to be revenue-producing long after release.³¹ In industries such as publishing, there is little incentive for clearinghouses, as out of print books or artwork are more difficult to locate and build upon.

¶          Due to the cross-industry reach of copyright law, pressure group lobbying from the music and movie industries has inadvertently spawned externalities in other industries. This drag-along effect of industry-neutral law is not unique to copyright. In patent law, lobbying efforts of multiple, discordant industry groups have resulted in a tangle of law that fails to appropriately account for the innovation models of any industry.³² In copyright, this effect is more unidirectional, specifically in the area of term extensions. Here, publishing has been dragged along into a new copyright regime, without the mechanisms in place to account for the orphan-works problem.

¶          Recent attempts to forge legislative compromise with regards to orphan works have been unsuccessful.³³ Graphic arts and photography groups have strong incentive to oppose such legislation, as more often than not, information about the creator of a work is not available on the work itself,³⁴ and the amount of effort required to show a so-called "good-faith" search would likely be quite minimal, even given modest publicity and monetization efforts on the part of the original creator. In short, there is little evidence that the legislative logjam will be broken, particularly given the sweeping applicability of copyright across media formats.

III. Orphan Works, Publishing, and Knowledge Maintenance

¶          Within the publishing industry, the orphan works problem is much more significant. The business of publishing is highly decentralized, and publishers do not typically maintain long continued relationships with the author of a given book, especially if printing is limited to a single run. There is little business incentive to maintain a rights database for out-of-print books, which are presumably out-of-print because they were not profitable enough to continue printing. The publishing industry has undergone several shifts in printing formats before making the shift to digital formats in the mid-1980s. Many of these pre-digital printing formats were proprietary to individual printers. As a result, not only would a publisher likely be unable to reprint books made before the shift to digital without significant effort, an older publisher may not even be able to confirm that they printed a particular book. Adding to the problem is an extremely fragmented publishing industry, raising the possibility that a particular publishing house (generally the most reliable way to find a rights-holder) may no longer exist.

¶          Due to the uncertain character of orphan rights distribution and a fragmented publishing industry, the most significant advocate for orphan works legislation continues to be libraries, particularly university libraries. From 1999 to 2001, Carnegie Mellon University (CMU) Libraries attempted a feasibility study of digitizing their library. CMU attempted to track down copyright owners for a random sample of books published since 1923, with the intent of gaining permission for digitization.³⁵ In their response comment to the U.S. Copyright Office’s 2005 inquiry on Orphan Works, they detail the results of their study.³⁶ 

¶          According to a rough estimation CMU performed based on WorldCat publication database numbers and studies done by the U.S. Copyright Office, only 7% of book copyrights were renewed between 1923 and 1963, when protection became automatic.³⁷ While the numbers are not directly comparable to today, copyright renewal rates provide an indication as to what proportion of the copyright-eligible population feel strongly enough about the continued profitability potential of their work to warrant the effort of preventing it from entering the public domain.³⁸ Renewal rates also provide a further explanation for the lack of incentives on the part of publishers to keep track of older works - if the works were worth copyrighting, someone would have done so.³⁹ 

¶          For their sample, Carnegie Mellon next attempted to track down the copyright owners of books that were not in the public domain. Despite the original claimant being listed at the U.S. Copyright Office, for 22% of books sampled, the publisher of a book simply could not be found.⁴⁰ A further 11% of books were eliminated from the study because third-party copyright ownership made tracking down the proper owner excessively complicated.⁴¹ Even when a publisher could be located, they were either unresponsive (36%), unaware of having published older books, lacked proper records regarding a particular work (dead-ending the search), or unsure what rights they possessed regarding a work.⁴² The transaction costs of tracking down copyright owners in the publishing world can be a significant deterrent,⁴³ as can the wildly variant restrictions and fees that publishers can demand, even for continued access to out-of-print books.⁴⁴ A simple lack of proper rights cataloguing can further deter users from a search--CMU averaged over 100 days per copyright owner before an answer was received, sometimes after multiple letters of inquiry.⁴⁵ A substantial portion of orphan works problems in publishing may thus be a function of poor industry organization.

¶          Concordant with the copyright concerns is the general policy goal of knowledge maintenance. Carnegie Mellon’s study suggests that over half of the books published in the United States since 1923 are now out of print.⁴⁶ Further, this wealth of knowledge is literally disintegrating. Older books printed on non-acid-free paper will eventually dissolve, and many out of print books are in advanced states of decay, a process the best preservation efforts can only slow.⁴⁷ As transaction costs and difficulty in finding copyright owners inhibit the ability of scholars and creators to build off of these works, the increasingly poor condition of these works will further deter their use. The fragile nature of books demands a solution not only in copyright law but also in digitization. Current copyright law stifles efforts to digitize, preserve, or use many of these rapidly expiring works. Ultimately, this magnifies orphan works problems to scales worthy of attracting attention. The digital era has arrived in time to allow preservation of the rapidly expanding printed record. Orphan works underscore the fact that this preservation effort’s largest obstacle is not technological, but legal; an ironic reality, given copyright’s constitutionally stated purpose of promoting knowledge.

¶          The digitization of the world’s printed knowledge, or even the libraries of a dozen major universities, eventually boils down to an issue of scale. Anything that was last printed before the mid-1980s--still some 10 million books--will likely need to be manually scanned, due to a lack of transferrable digital form.

IV. Enter Google

¶          Google hardly needs an introduction. Far and away the most successful company of the late-1990s tech boom, Google has a self professed mission "to organize the world’s information and make it universally accessible and useful."⁴⁸ All of this data is used as a driver for Google’s market-dominating internet advertising engine. In October 2004, the company launched Google Books--then called Google Print, and began digitally scanning books from publishers and university libraries. In October 2009, Google announced that they had scanned their 10 millionth book.⁴⁹ Depending on the copyright status, users are permitted to search within books and view excerpts of a few pages surrounding their search term. Google’s effort brought them a lawsuit from the Author’s Guild and five major publishing companies--McGraw Hill, Pearson, Penguin, Simon & Schuster, and John Wiley and Sons. As of this writing, the settlement is still being negotiated.

¶          In essence, Google Book Search allows copyright holders to collect revenue from digital access to a work. In some cases, digitization will no doubt reignite interest in a forgotten work, perhaps enough to warrant a reprinting. This new opportunity for revenue with little additional cost on the part of the creator will in theory incentivize copyright holders of orphan works to come forward. However, while publishing houses with a large cache of copyrights could stand to reap a tidy sum, it is unlikely that the majority of individual book-owners will see much in the way of revenue, due simply to a lack of queries involving their book.⁵⁰ However, the fact remains that the revenue regime constructed by Google is far more flexible than that of traditional publishing, even given the increasing ability of publishers to print smaller runs of books (even down to a run of one). In essence, GBS allows "the producer to capture more of the benefit from what he produces. . . . The higher the return to producing intellectual property the more intellectual property will be produced."⁵¹ In addition, a privatized database allows Google and copyright owners to price discriminate based on levels of access. Access to searches and excerpts may be free, access to an entire work may require a fee for individuals, and large commercial or academic organizations may be charged high prices for wide-reaching subscriptions.⁵² 

¶          Given the particulars of the parties involved, it is right to posit that Google made the strategic choice to provoke a lawsuit in order to force a unified settlement. As previously discussed, the transaction costs for investigating the copyright status of books is prohibitively high, a problem which is only exacerbated by the scale of the project. Once an actual lawsuit is filed, the incentive to settle for both sides is high.

¶          It has been argued that Google likely would have had a strong fair use argument in the vein of its victory over Perfect 10⁵³ but the actual probability of success Google might have had at trial is irrelevant, so long as the probability was significant to give the plaintiff publishers pause. Not pursuing a fair use argument at trial also allows Google to negotiate additional ways of monetizing its new digital book collection- such as getting a cut of eventual digital book distribution. Just as important as the perceived strength of its arguments at trial, Google’s size and cash reserves ensure that it has the financial means to pursue the lawsuit as far as necessary.⁵⁴ Thus, Google’s negotiating position, as compared to a fragmented publishing industry currently in a state of flux, is quite strong.

V. Market Optimization and the Long Tail

¶          The copyright status of a work does not guarantee profitability, absent some market force that demands access to the work in sufficient quantity to warrant further production. The renewal requirement of copyright was an expression of copyright’s supporting role in the monetization of a creative work. Absent this requirement, the public is prevented from accessing works that have no present commercial exploitative value, thus creating a dead-weight societal loss and subverting the knowledge dissemination aim of copyright. Along this vein, William Landes and Richard Posner have argued for a return to copyright renewal practices, but with the twist of allowing near indefinite renewal.⁵⁵ The appeal of this argument is strong: Renewals would thus be driven by the presence of sufficient economic incentive, allowing for the "optimal use of property rights." In the publishing world, despite the rise of publishing on demand and e-publishing, the cost of a new print-run for out-of-print books may still be prohibitively high⁵⁶ or at least higher than whatever additional revenue a publisher might be able to extract from the run. Compare this to other media, where the cost of pressing another CD or copying another song is close to zero.⁵⁷ Private ordering can help further hone this optimization by focusing on the long tail of out-of-print works.

¶          Chris Anderson argues that digitization has rendered the "hit-driven" economic model obsolete.⁵⁸ Cost-revenue analysis of physical media is calculated in groups of thousands: books are not printed unless they will make up for the cost of producing a run, and Wal-Mart will not sell a CD unless they believe they can sell 100,000 copies of it, in order to cover retail overhead.⁵⁹ Over half of Amazon.com’s revenue comes from books that are not available in a brick-and-mortar bookstore, or books that do not sell in sufficient quantities to warrant bookstore placement.⁶⁰ The bulk of Google’s advertising revenue comes from a swarm of smaller websites and search queries, rather than from a small group of rich repeat buyers.⁶¹ The implication for copyrighted content is clear: as digitization brings marginal cost close to zero and allows for increasingly fine-tuned price discrimination, the optimal approach for content owners is to maximize the quantity of content available for purchase, regardless of forecasted success. ⁶² 

¶          Long tail models reconfigure economies of scale, emphasizing the value of aggregation above all else. Digitization minimizes the marginal cost of production for "soft" content industries such as music, movies, and eventually books and art by obviating the need for products to take physical form.⁶³ Eventually, the only costs involved are the logistics of database maintenance and delivery.

¶          This economic model has two impacts. First, it lowers the barrier for continued profitability on "back catalogues," content that has already been through a primary market cycle. This incentivizes rights-holders not only to consolidate, but to advocate longer terms for rights exclusivity, in order maximize the economic gain realizable only through large aggregate rights holdings. Second, content creators can more cheaply distribute their works to a wider audience due to a lower initial overhead. However, initial costs of content creation may still be high, and the individual creators on balance will not accumulate significant revenue in the long tail from an individual work, especially not when compared with the initial market cycle.⁶⁴ Nonetheless, for content-industry corporations that hold wide back catalogues of works and rights, digitization opens the door to this new economic style. GBS represents a market-optimizing instance of private ordering, using near-zero marginal cost to allow publishers to take advantage of the long tail. Google is able to use its vast financial resources and technical expertise to implement a project that a fragmented, low-margin publishing industry would likely not be able to execute on its own. Perhaps just as importantly, Google is covering the primary capital and labor overhead: the physical act of scanning and digitizing several million books. In effect, the publishing industry is getting a revamp of the economic engine for next to nothing. Google’s ad-driven business model is contingent on as many users as possible being online for as long as possible. This model allows Google to essentially give away its services, so long as it keeps users online, and therefore on Google. This alignment of interests makes Google an ideal partner for a digitizing publishing industry.

VI. The Impact of Digitization

¶          Google's approach is not without problems. Concerns have been raised over granting Google a virtual monopoly over orphan works, or even book digitization, given the massive barriers to entry in the form of labor, money, and time. In essence, Google has leveraged its business strength in one area, web search, and used it to essentially force the cooperation of the publishing industry. These antitrust concerns, while worthy of discussion, are beyond the scope of this note. More pressing here is the potential for Google’s incentive program to lead to an ironic inversion of copyright law. This manifests in two arenas: fair use and the public domain.

¶          First, GBS can contractually circumvent fair use. By allowing content owners to determine the size of the excerpts that surround a search query, a potential user may be forced to pay for more complete access. This problem is most profound with regards to rarer works and scholarly research, especially ones for which paper copies may be in exceedingly poor condition. The transaction costs of securing access to a rare work are unfairly reallocated to the copyright holder. Further, the possibility of perfect price discrimination with regards to use and access restrictions to a work may eventually undermine the fair use defense, due to the presence of a readily available licensing market.

¶          Fair use exists as a result of the market’s failure to foster socially desirable uses of copyrighted works. Rights-holders have little incentive to selectively yield their copyright, particularly when these uses may paint their original work in a negative or undesirable light. Fair use is an attempt to prevent monopolies on certain expression from shrinking First Amendment protections, as well as to foster the practices of criticism and academic study, which ultimately maximizes a given work’s contribution to the overall knowledge base.

¶          Second, GBS threatens to further slow the flow of books into the public domain. While the settlement explicitly excludes public domain works, a significant tactical question arises for rightsholders when book copyrights expire and enter the public domain. It would be in the public interest for Google to then provide open access to that particular book, at no charge, and indeed it is questionable whether Google would have to continue paying the copyright holder at all for continued use of the work. While the benefits of this strategy are unclear, it is possible that publishing houses might try to leverage permission for books they do still control in order to maintain revenue streams on books that they once controlled. Such a deadlock could threaten to undermine the entire system. A compromise could be reached by opening up access and continuing to split the advertising revenue. However, this is reliant on Google pushing for open access as works enter the public domain, an uncertain proposition, although it seems likely that universities would advocate for an increase in the public domain. Otherwise, there is a social loss in the form of a smaller public domain.

¶          To a point, such harms are speculative. However, the question raised is significant: Does private ordering undermine copyright policy? If copyright law is merely a matter of market optimization by means of maximum renumeration for creators, surely a private ordering system is ideal, particularly with the advent of micropayments and electronic transaction costs that near zero. Indeed, there is a strong argument for reframing copyright in this context, particularly given the dependence of the American economy on copyright-protected exports.

VII. Private Ordering

¶          Private ordering manifests itself in a knowledge-based economy, where the boundaries between industry and market segment are fluid and easily surmounted. Apple, Google, and Microsoft are so successful precisely because of their ability to transcend these industry boundaries. This fluidity means that rivals in one market segment may be allies in another--Google’s YouTube and Maps services are prominently featured on the iPhone, even while Google launches a competing smartphone operating system. It also requires a new understanding of how these actors will interact across market segments. Public choice theory presumes that social actors have fixed interests, and that they do not need to make complex judgments to determine how their interests can be advanced. It also does not explain how nonmaterial motivations and resources affect collective action.⁶⁵ Private ordering in the information economy is then primarily an issue of framing.

¶          Outside of copyright, private ordering is already a highly prevalent phenomenon--most legal disputes in the U.S. are resolved without "intervention of the public legal process."⁶⁶ In intellectual property, private ordering is on the rise, most notably in the management of patent licensing.⁶⁷ Private ordering is simply good business.

¶          It is important to note that private ordering is not ignorant of the current state of the law, nor does it necessarily seek to change existing law. Law has a gravitational pull, and shapes the existing discourse. Mobilizations in response to law may serve to strengthen existing law, rather than unsettle it. Niva Elkin-Koren’s critique of Creative Commons is structured along similar lines, arguing that the use of a licensing tool, "of property rights, backed up by contract, is bound to entail a logic of exclusion," and strengthens the conception of information as private property.⁶⁸ Framing shifts are shifts in discourse, and in that sense, even the term "intellectual property" represents a shift, introducing the term property as an umbrella description of copyright, trademark, and patent. Property, unlike patent, trademark, and copyright, lacks an expiration date, and implies an entitlement in perpetuity. Any debate on copyright term extensions is now framed in a context of property protection. Thus, rather than a discussion beginning from the analysis of what is best for the ostensible knowledge advancement ends of copyright, the default shifts to center around the value of private property preservation. This shift is made easier because it aligns with already engrained market-based value frameworks. Discourse framing can alter the value criteria of copyright debate and legislation, thus changing the status quo bias in such a way that will benefit the debate’s winner. Content industries have used framing to their advantage, justifying term extensions with the rhetoric of preventing property loss, crucial to economic engines.⁶⁹ 

¶          Intellectual property battles are fought by shifting groups, defined generally as "users of information and as afflicted by a similar set of problems that could be expressed . . . in terms of intellectual property and information economics and structures of innovation."⁷⁰ Similar groups were formed by the so-called "content industries", who forged an originally uneasy alliance based on "forging a common identity as intellectual property industries, and by framing the use of their products without permission as ‘theft.’"⁷¹ The conceptual battle is then between framing intellectual property as a knowledge advancement issue or a freedom of speech issue, and framing intellectual property as an "indispensable tool of modern economic management."⁷²

¶          From a knowledge policy standpoint, Google’s financial might combined with the interests of its partner libraries has created a bloc strong enough to impact orphan works policy.⁷³ 

¶          If or when an eventual settlement does coagulate, it will be the most prominent recent example of the private ordering phenomenon in copyright law. In essence, "[p]rivate ordering operates when the rule-making process regarding the use of information is privatized, and the legal power to define the boundaries of public access to information is delegated to private parties."⁷⁴ Such a system resembles one of real property rights wrapped in contractual protection, and could serve to further weaken copyright as a tool of knowledge advancement in favor of a conception grounded in real property rules.

VIII. Digital Libraries and Digital Locks

¶          Private ordering is not a counterbalance to framing momentum. Rather, private ordering seeks to maximize economic benefit from within a given frame. In turn, this maximizes the economic activity dependant on that intellectual model. Ultimately, the cost of multiple changes in private alignment incentivizes pressure groups to lobby for legislative framings that help maintain benefits they currently enjoy. Legislative reluctance to upset the status quo of good business further entrenches such framing. Here, GBS threatens to cement existing conceptions of copyright law as primarily remunerative in function, and founded in the same roots as real property rights. GBS has the opportunity to seed an expansive digital public domain, but the rights-holders they negotiate with have incentive to maximize digital rights protection in order to leverage the long tail benefits of aggregated rights holdings. The resulting economic model has serious implications for copyright law.

¶          The long tail does not provide additional incentive to authors for work creation. Consider that the value of the long-tail is in aggregation. While digitization may marginally bolster the demand of some works over the long term (and in rare cases, reinvigorate demand), the model for individual works is still the same as it is in most content fields, with the overwhelming majority of revenue being made on the initial print runs. After all, if a book is not sufficiently in demand to warrant continued printing, there is little reason to think that GBS will suddenly result in a windfall for potential authors.⁷⁵ 

¶          Google is consolidating and assuming the risk of liability on orphan works through the use of an opt-out settlement, and hopes to mitigate that risk with the carrot of digital revenue. To that end, it is unclear if the long tail is sufficient on an individual level to induce owners of orphan works to come out of the woodwork. Surely, any positive amount of revenue is better than none at all, but owners of orphan works in some cases may not even know that they are the rightful copyright holder--for instance, small estates or since-merged publishing companies--and the cost involved in determining whether or not they are entitled to any revenue may exceed the actual revenue itself. This in itself may not be problematic for the public domain. Scholars can still access works whose owners have not shown up to remove them from the system, and an orphan work in GBS is virtually in the public domain if no one ever claims it, aside from Google having control of the digital copy.

¶          Thus, if digitization does not provide sufficient consistent revenue to induce the creation of additional works, the benefits are confined to remuneration. The danger of rent-seeking from a GBS settlement is extremely high. Given both the aforementioned ambiguity in old contractual arrangements between publishers and authors as well as the relatively low revenue for each individual author compared to the potentially high aggregate revenue of a large publishing portfolio, publishers are incentivized to claim copyright on as many works as possible, including works to which they have at best dubious claims. Again, the proceedings to prove the contrary may be more costly than any expected benefit for individual authors. This ambiguity is exacerbated by the complete lack of contractual language regarding digital and e-book rights for older books.⁷⁶ This is again endemic of private ordering, which rather than seeking to revise ambiguities or problems in current legislative structure, instead looks to each party to maximize their economic return through contract, for better or for ill. Here, publishers are able to advantage the ambiguous contractual market for out of print books, at the cost of author compensation and incentive.

¶          Private ordering is an economically optimizing phenomenon, and as such, is ignorant of the non-economic impacts such ordering may have. In essence, copyright law uses economic means to promote non-economic ends - the constitutionally enumerated "Progress of Science and useful Arts."⁷⁷ Thus, any additional economic controls are only useful to copyright to the extent that it continues to further promote progress. It is the duty of Congress to determine the optimal balance of inducement and open access in order to encourage new writings and discovery. This balance is justified by the understanding that copyrighted works are not created in a vacuum, and that progress nearly always builds upon previous works.

¶          Thus, it might be suggested that any additional protection beyond what is necessary to induce new work is detrimental to copyright and to progress. By maximizing economic remuneration for aggregate rights holders while failing to significantly contribute to the inducement of new work creation, private ordering may create social waste. The ripe opportunity for rent-seeking with digital projects such as GBS suggests that on the whole, copyright holders could take more from the public domain than they contribute.

¶          Private ordering is unable to mitigate an overbearing copyright scheme. While in the aggregate it may breathe new life into orphan works, this comes at the cost of increased incentives for large rights-holders to remonetize forgotten works in order to maximize their long-tail revenue. Further, publishers now have renewed incentive to join the pressure groups of the other large content-industries in pushing for increasing copyright term extensions, because digitization now enables them to continue benefitting from even unpopular older works.

¶          A possible counterweight here may be university libraries who are offering their collections for scan and use. While is possible that libraries will push for works to become digitally free for access by anyone once the copyrights expire, the amount of leverage they have is likely small, as libraries do not own copyrights. In fact, libraries may eventually be cut out of the project as publishers move to contribute digital versions of the books they both currently print and printed in the past. Rightsholders will also certainly balk at Google continuing to collect ad revenue from queries for books to their books that enter the public domain.

¶          Private ordering is incapable of addressing the underlying issues with the present state of copyright law, and in particular it is unable to resolve copyright’s non-economic goals. Here, although private ordering has almost inadvertently managed to work around orphan works, it has also incentivized rent-seeking and threatens the public domain.

IX. Is There An Alternative?

¶          Economic efficiency is ultimately a strong variable in determining appropriate policy. Landes and Posner suggest a novel approach to optimizing economic efficiency in copyright: shortening copyright terms, increasing fees for term renewal, and allowing near-indefinite renewal.⁷⁸ Basing their argument on historical data, they argue that the amount of renewed copyrights under this sort of system would be a "tiny fraction" of all copyrights.⁷⁹ Landes and Posner analogize to trademark in order to suggest that indefinite copyright renewals "need not starve the public domain."⁸⁰ In theory, this allows for a self-sorting market in that those works determined by their authors to still have exploitable value are renewed, while those that are not are free to the public domain.⁸¹

¶          Digitization has zeroed the cost of additional distribution. Consider that prior to a digital model, book copyright holders in Landes and Posner’s world would have had to account for the cost of publishing a new run of books--however small--,in addition to the cost of renewing the copyright, and whether the profit would outweigh these costs.⁸² With GBS, rights-holders essentially make the bet that they will collect at least the renewal fee in access and ad micropayments over the term of the copyright. Landes and Posner suggest ten year terms for copyright renewal, and even with a fee of several hundred dollars, this might be a feasible proposition for many more books. The question of projected profitability does not change, but the additional costs incurred to distribute the work are now next to nothing, meaning the potential range of copyright protected works may be much broader under this model, ultimately harming the public domain.

¶          Landes and Posner dismiss public-goods problems, essentially arguing that without any copyright protection, there is no incentive to maintain existing works, leading to inefficiencies such as congestion externalities.⁸³ Here, framing comes into play, as Google and its partner libraries actually do have incentive to maintain digital book databases that are inclusive of public domain works, for both search ad revenue and general institutional missions. Further, copyright over-protection is similarly inefficient, as the orphan works problem erects barriers preventing those who actually wish to exploit the work from doing so.

¶          Landes and Posner provide a potentially novel approach that emphasizes private ordering to achieve economically optimal results and prevent orphan works issues. However, the declining marginal costs of digitization still incentivizes publishers to use digital locks to control both works they own and works they do not. Regardless of the copyright model, the economic incentives involved raise the question as to the feasibility of a public domain in the digital era.

Conclusions

¶          Are the incomplete solutions of private ordering the best ones we can hope for? The overwhelming pressure group imbalance, coupled with an endowment bias, makes the reversal of existing term extensions an unlikely prospect. However, this does not preclude an evolving conception of intellectual property and its corresponding values. How then will the underlying problems of copyright law be solved? It may be that private ordering can indirectly provide a solution. GBS could prove to be a catalyzing event for critics of current intellectual property policy as it brings the negative impacts of excessive term limits to the political front burner.

¶          The so-called "access to knowledge" movement is pushing for new models to govern the flow of information-based goods, and has so far primarily gained steam in the international development context.⁸⁴ The sheer volume of academic and policy discussion generated as a result of the GBS settlement only makes it more likely that actors will coalesce into a strong movement for improved information policies. It may be then that private ordering provides a cognizable frame of reference for groups to react to. This is not to understate the level of resistance that this new potential movement would be facing. Private ordering is already causing publishing interests to reframe the direction of their advocacy, much like a budding social movement, and surely they will push hard to keep this new potential long-term source of revenue. Private ordering, then, is best seen as perhaps just another pull in the ongoing policy tug-of-war, albeit one that shakes more individuals into action.

¶          Is a public domain possible in the digital era? Surely the newfound ease of distribution would suggest an answer in the affirmative. However, in order to maximize the benefit of the information age, it is imperative that we succeed in preserving as much knowledge as possible before it disintegrates. In that process, we have the opportunity to undo the wasteful traps preventing orphan works from being fully utilized and built upon. It is critical that in taking advantage of this opportunity, we are careful not to erect larger, more complex barriers to access, and refrain from sacrificing the progress ends of copyright to digital revenue’s ever-growing long tail.

Footnotes