3. THE BALANCE OF THE COPYRIGHT – FAIR USE DOCTRINE
3.2 Market Failure Revised
3.2.2 Copyright Scope and Doctrinal Efficiency
For the purpose of assess specific doctrinal recommendations, namely Mar-ket Failure Approach and Cost-Benefit Approaches of fair use analyses,163
159 Id.
160 Indeed, as in case of refusal-to-license, “simply describing the copyright owner’s motives
as ‘noneconomic’ is not analytically useful by itself.” See Matthew J. Sag, Beyond Ab-straction: The Law and Economics of Copyright Scope and Doctrinal Efficiency, 81 TUL. L. REV. 187, 234 (2006).
161 Through out this new article, Professor Gordon doesn’t provide any example of excusable
cases other then “high transition cost”. See GORDON, supra note 143, at 3 nn.3, 4, 9, 22,
& 52.
162 Id. at 44-45.
163 In order to make the transition from abstract theory to practical implementation, Professor
Sag develops a set of metrics, see infra note 168-70, to assess three specific law and eco-nomics approaches to copyright’s fair use doctrine, which are exactly the three picked by
fessor Sag develops an economic model of copyright scope and doctrinal effi-ciency as a vehicle for evaluating the welfare implications of changes in the breadth of the rights vested in copyright owners.164
Briefly, the scope of an individual copyright can be defined, in economic terms, “as the extent to which its owner can use copyright law to impose costs on third parties or exclude them from certain markets altogether.”165 And the effi-ciency of an individual copyright doctrine “is determined by the extent that a change in its scope benefits first-generation authors more than it costs second-generation authors for a given level of copyright scope.166 Evaluating specific doctrinal recommendations, Professor Sag suggests, “we need to assess both the effect on copyright scope in general and the specific costs and benefits of the doc-trinal formulation in particular.”167
Analyzing the copyright scope and doctrinal efficiency generates a set of metrics, which, in turn, is useful for analyzing a doctrinal theory:168
1. Does the theory take account of the role of private ordering in determining the ideal scope of copyright?169
2. Is the theory doctrinally efficient?
this author: Gordon’s “Market Failure Approach,” Fisher’s “Efficiency Maximum Ap-proach,” and Lunney’s Balance Approach.
164 See Sag, supra note 160, at 192.
165 Id. at 199.
166 Id. at 219.
167 Id. at 218.
168 Id. at 226.
169 In Professor Sag’s analysis, the net welfare effects on a change in copyright scope are
dependant on the efficiency of private ordering. “The more efficiently the market reallo-cates rights through licensing or the consolidation of production into firms, the higher the optimum level of copyright scope will be.” Id. at 224.
3. Is the theory feasible in light of the expectation that there will be substan-tial variation, both within and between industries, in the welfare-scope relation-ship?170
3.2.2.1 Apply to Market Failure Approach
The evaluation of Professor Sag’s economic model would not be discussed for the present. Going through the set of metrics, Professor Sag observes some convincing weaknesses of Market Failure Approach. Nevertheless, he validates that Market Failure Approach, with some improvements, is an acceptable eco-nomic model for the determination of fair use.171
Under the first prong of the metrics, Sag identifies that “the requirement that the defendant prove the existence of market failure as a prerequisite for a finding of fair use tilts this apparently neutral framework decidedly in favor of the copy-right owner.”172 Besides, manifold obstacles to market perfection may simulta-neous exist, the defendant have to further prove that these obstacles “are of a suf-ficient degree to constitute a market failure.”173 This difficulty of proving the existence of market failure is much severer in the cases of less tangible causes of market failure such as externalities or noneconomic motivations: “the
170 Professor Sag argues: “The welfare-scope relationship is both complicated and subject to
substantial variation, both within and between industries. Doctrinal recommendations which simply assume that the welfare effects of a change in copyright scope are easily as-certainable are likely to be far too simplistic.” Id.
171 “In the final analysis, the market-failure approach to fair use performs well when assessed
against the metrics developed in this Article, but the metrics also highlight ways in which the market-failure test could be improved.” Id. at 237.
172 Id. at 231.
173 “[M]erely identifying the existence of one or more potential causes of market failure will
never be sufficient; the defendant must establish that these market imperfections are of a sufficient degree to constitute a market failure.” Id. at 232-33.
failure approach itself gives little guidance as to what degree of positive external-ities or noneconomic motivations might justify the application of fair use.”174
To overcome this difficulty, Professor Sag suggests that game theory, behav-ioral economics, and price theory can be useful in assessing the efficiency of pri-vate ordering.175 In sum, he concludes, “the market-failure approach to fair use does take account of the role of private ordering in determining the optimum scope of the copyright owner’s rights,”176 but a more nuanced approach to allo-cate the burden of proving the existence of market failure would improve the ap-proach.177
Because adopting market failure approach may largely reduce the opportuni-ties of finding fair use, which in turn, reduce the cost of administration, market failure approach passes the second prong of the metrics – doctrine efficiency.178 Yet, because market failure approach implicitly assumes that the copyright owner’s rights are absolute, unless modified with a more nuanced approach to allocate the burden of proof, adopting market failure approach would constitute a significant expansion of copyright scope.179 As to the third prong of the metrics, variation in the welfare-scope relationship, Professor Sag states, because market failure approach does account for the efficiency of private ordering – a substantial cause of the variation in the welfare-scope relationship – it passes the test.180
174 Id. at 233.
175 Id. at 234-35.
176 Id. at 235.
177 Id.
178 Id. at 236.
179 Profess Sag explains, “requiring the defendant to establish market failure implicitly
as-sumes that the copyright owner’s rights are absolute and that any deviation from those rights requires substantial justification.” Id.
180 Id. at 237.
By and large, Professor Sag suggests, with some improvements, market fail-ure approach is an acceptable method for analyzing fair use doctrine.
3.2.2.2 Inevitable Drawbacks
The most significant drawback of market failure approach, as identified by Professor Sag, is it puts the burden of proof on the defendant.181 Proving the ex-istence of market imperfections is difficult; proving these imperfections accumu-late enough to constitute a market failure is even more complicated – especially in cases of “externalities or noneconomic motivations.”182 In this regard, Profes-sor Sag improves market failure approach, by introducing game theory, behav-ioral economics, and prices theory,183 to be a more nuanced approach to allocate the burden of proof,184 however, he did not tell us how to apply these theories, and how a “more nuanced approach” can be processed.
Professor Sag cites a price theory explanation of fair use developed by Ben Depoorter and Francesco Parisi, which enumerates factors courts should take into
181 This is not the fisrt time for this drawback been identified, Professor Lunney also has
emphasized that: “By starting with an assumption that private markets are efficient, the market failure approach shifts the burden to the defendant to establish the existence of market failure as a threshold matter … Moreover, even where a defendant successfully establishes a market failure, the strong preference for market outcomes and the distrust of government intervention (that fair use, but not copyright itself, somehow represents) re-main, limiting the availability of the fair use ‘defense.’” See Lunney, supra note 139, at 992.
182 See supra text accompanying note 171-73.
183 All these three economic terminologies, as showed in Sag’s suggestion, refer to one
con-cept: anticommons. See Sag, supra note 160, at 234-35; See also Ben Depoorter & Fran-cesco Parisi, Fair Use and Copyright Protection: A Price Theory Explanation, 21 INT’L
REV. L & ECON. 453, 459-62 (2002). “The anticommons equilibrium pricing is in fact the outcome of a prisoner’s dilemma problem that the individual copyright sellers face when pricing their copyrights independently from one another.” Id. at 462.
184 See supra text accompanying note 175-77.
account when valuating fair use, including: “(a) the number of copyright holders;
(b) the degree of complementarity between the copyrighted inputs; [and] (c) the degree of independence between the various copyright holders.”185 However, in the original article, Fair Use and Copyright Protection: A Price Theory Explana-tion, those variables identified by Depoorter and Parisi were purposed to guide and constrain the application of fair use doctrines,186 not to illustrate how to al-locate the burden of proof in the fair use finding.
Depoorter and Parisi provide an explanation of “Tragedy of the Anticom-mons”: “when multiple owners have the right to exclude others from taking ad-vantage of a scarce resource, and no one has an enforceable privilege of use, the resource might be underutilized.”187 Because copyright holders each pricing their works independently and competitively, due to the strategic behavior, in the case of strict complementarity – ex. copyrighted works which are essential and irreplaceable to the success of a derivative work,188 copyright owners can im-pose external costs on the sellers of other complementary inputs, due to the cross-price effects between the goods.189 Conversely, in the case of perfect
185 See Sag, supra note 160, at 234-35 (citing Depoorter & Parisi, supra note 183, at 453,
453-54, 463-64).
186 There is a fourth factor lists in Depoorter & Parisi’s critical variables that should guide
the application of fair use doctrine: “[the copyright owner’s] ability to price discrimina-tion.” Id. at 453-54, 464.
187 Id. at 458. “Michelman (1982) coined the term anticommons …, defining it as ‘a type of
property in which everyone always has rights respecting the objects in the regime, and no one, consequently, is ever privileged to use any of them except as particularly authorized by others.’” (citing F. L. Michelman, Ethics, Economics and the Law of Property, in N O-MOS XXIV:ETHICS, ECONOMICS AND THE LAW 3, 3-40 (J. Roland Pennock & John W.
Chapman eds., 1982)).
188 Depoorter & Parisi, supra note 183, at 459.
189 “The greater the number of individuals who can independently price an essential input,
ability – ex. less essential sources without compromising the quality and success of the final product,190 the copyright owner is unable to impose any external cost on the owners of other copyrighted material, due to the Bertrand-type competition between the various sellers.191
In conclusion, Depoorter and Parisi state, the failure of the various copyright holders to coordinate prices resulting social deadweight loss.192 In the case of complementarity, “the competitive Nash equilibrium would generate anticom-mons pricing, making both society and the individual copyright sellers worse off.”193 On the other hand, in the case of substitutability, the competition be-tween copyright holders will draw their price to the marginal cost which would be detrimental for them.194 Therefore, “in light of the anticommons insight, fair-use doctrines retain a valid efficiency justification even in a zero transaction-cost en-vironment. Fair-use defenses can be regarded as justifiable and instrumental in minimizing the welfare losses occasioned by the strategic behavior of the
the higher the equilibrium price that each of these individuals will demand for his own li-cense. At the margin, as the number of copyright holders approaches very large numbers (or infinity), complete abandonment of valuable resources will result.” Id. at 461.
190 Id. at 459.
191 Id. at 461. Bertrand competition is a model of price competition between duopoly firms
which results in each charging the price that would be charged under perfect competition, known as marginal cost pricing. See “Wikipedia, the free encyclopedia”, http://en.
wikipedia.org/wiki/Bertrand_competition (last visited Nov. 19, 2007).
192 “If the copyrights are in a relationship of complementarity in the production of a
deriva-tive work, the competideriva-tive Nash equilibrium would generate anticommons pricing, mak-ing both society and the individual copyright sellers worse off.” Depoorter & Parisi, su-pra note 183, at 462.
193 Id. “As in a traditional prisoner’s dilemma game, the inability of copyright holders to
coordinate prices produces a result that is both privately and socially inefficient.” Id.
194 Id.
right holders.”195
In sum, although the propose that Professor Sag intruding price theory is to improve market failure to be a more nuanced approach to allocate the burden of proof, however, after surveyed the price theory provide by Depoorter and Parisi, one can find it does not work the way Professor Sag implied, thus it can not help market failure to surmount the most significant drawback identified by Professor Sag – puts the burden of proving the existence of market failure on the defendant.
Nevertheless, the price theory verifies another economic justification for the existence of the fair use doctrine: whenever anticommons costs are serious enough to undermine the viability of the transaction, fair use doctrine is a valu-able tool for mitigating the resulting deadweight losses.196 Furthermore, even in a zero transaction-cost environment such as digital network, “[i]f strategic behav-ior is not prevented by the ability of users of copyrighted work to ‘click and pay’
in order to obtain copyright licenses, sub-optimal equilibria may still result from the independent pricing of copyright licenses … In light of this, the defense of fair use retains an important, albeit residual, role in minimizing the deadweight losses.”197