Vol ##] THE PROBLEM OF UNIFORMITY COST IL. THE PROBLEM OF UNIFORMITY COST Economic analysts have long recognized that intellectual property rights impose social costs because they interfere with competitive distribution of information goods. But from a dynamic perspective, some distortion must be tolerated as the price to be paid for having information good created in the first place. The social costs that matter, then, are not all static deadweight losses, but only the distortions caused by rights that are more or less robust than necessary to have induced investments in innovation that deliver a net benefit to society. Uniform intellectual property rights necessarily impose such costs. Therefore, Uniform propert policymaking must manage. This section demonstrates this point by revisiting the standard economic justification for intellectual property ights and then by reorienting this analysis around the problem of formity cost This reorientation reveals that the theoretically optimal policy, if intellectual property rights are the only feasible response to derproduction of valuable information, is to fashion perfectly tailored ights rather than to promote perfect price discrimination, as some theorists suggest. While perfect tailoring is just as elusive as perfect price discrimination, the uniformity-cost perspective shows that the focus of policy analysis should be on how intellectual property rights can be rendered more context-sensitive. Further. this theoretical reorientation emphasizes the role of"law"in the law-and-economics of intellectual property because even after economic analysis identifies industries or technologies for which uniformity costs are particularly high, legal scholars must assess whether legal institutions can competently address this problem. Legal analysis demonstrates that while tailoring rights ex ante is an important strategy for reducing uniformity cost, it is not the only one. As is discussed in Section IV, infra, real options and flexible standards also can be used to inject context sensitivity into the application of formally uniform rights 8 See, e.g, WILLIAM M. LANDES RICHARD A POSNER, THE ECONOMIC STRUCTURE OF INTELLECTUAL PROPERTY LAW 74, 310(2003)(discussing social costs of copyright and patent law, respectively) See id See id. at 40(discussing perfect price discrimination)
Vol. ##] THE PROBLEM OF UNIFORMITY COST 5 II. THE PROBLEM OF UNIFORMITY COST Economic analysts have long recognized that intellectual property rights impose social costs because they interfere with competitive distribution of information goods.8 But from a dynamic perspective, some distortion must be tolerated as the price to be paid for having the information good created in the first place.9 The social costs that matter, then, are not all static deadweight losses, but only the distortions caused by rights that are more or less robust than necessary to have induced investments in innovation that deliver a net benefit to society. Uniform intellectual property rights necessarily impose such costs. Therefore, uniformity cost is the central problem that intellectual property policymaking must manage. This section demonstrates this point by revisiting the standard economic justification for intellectual property rights and then by reorienting this analysis around the problem of uniformity cost. This reorientation reveals that the theoretically optimal policy, if intellectual property rights are the only feasible response to underproduction of valuable information, is to fashion perfectly tailored rights rather than to promote perfect price discrimination, as some theorists suggest.10 While perfect tailoring is just as elusive as perfect price discrimination, the uniformity-cost perspective shows that the focus of policy analysis should be on how intellectual property rights can be rendered more context-sensitive. Further, this theoretical reorientation emphasizes the role of “law” in the law-and-economics of intellectual property because even after economic analysis identifies industries or technologies for which uniformity costs are particularly high, legal scholars must assess whether legal institutions can competently address this problem. Legal analysis demonstrates that while tailoring rights ex ante is an important strategy for reducing uniformity cost, it is not the only one. As is discussed in Section IV, infra, real options and flexible standards also can be used to inject context sensitivity into the application of formally uniform rights. 8 See, e.g., WILLIAM M. LANDES & RICHARD A. POSNER, THE ECONOMIC STRUCTURE OF INTELLECTUAL PROPERTY LAW 74, 310 (2003) (discussing social costs of copyright and patent law, respectively). 9 See id. 10 See id. at 40 (discussing perfect price discrimination)
Vol # THE PROBLEM OF UNIFORMITY COST A. Standard Economic Justification for Intellectual Property Rights Intellectual property rights are a second-best solution to an appropriability problem. The now-familiar utilitarian justification for ntellectual property law starts with Thomas Jefferson's observation that Informations is that no one possesses the less, because every other possesses the whole of it... [ H]e who lites his taper at mine. receives light without darkening me. For Jefferson. the capacity for information to"freely spread from one to another over the globe"is part of nature's "benevolent["design. For the economist however, the "benevolent"design of information poses a problem. To the extent that pecuniary motivation drives innovation, we should not expect to see useful information 4 produced unless the producer can recoup his or her investment. Because the distribution of valuable information cannot be controlled in the ways that distribution of scarce goods like tubes of toothpaste or radial tires might be- that is, information can be consumed nonrivalrously -the producer, acting alone, cannot rely on competitive markets to supply a sufficient return to make the investment in producing such information worthwhile. The government Letter from Thomas Jefferson to Isaac McPherson(Aug. 13, 1813), in 6 THE WRITINGS OF THOMAS JEFFERSON 180(HA. Washington, ed. 1861) s Information exhibits public goods characteristics. a public good can be consumed without depletion (non-rivalrous consumption) and can be withheld from nonpaying beneficiaries only at prohibitive cost (non-excludability ) See RoBert COoTER THOMAS ULEN, LAW AND ECONOMICS 40-41(2d ed. 1997); see also wilfried Ver eecl Public Goods: An Ideal Concept, 28 J OF SOCIO-ECONOMICS 139(1999). Information imperfectly excludable but it is because it is its non-rivalrous quality that makes it problematic The terms"valuable" anduseful" information in this article refer to information that is costly to produce and that members of our society find to be useful, informative enriching, or otherwise of value. Such information includes the ideas and expressions or embodiments of those ideas found in, for example, novels, movies, music, methods for manufacturing useful articles(medicines, computers, clothing, etc. ) computer software S Eg, Mark A. Lemley, The Economics of Improvement in Intellectual Property, 75 TEX. L REv. 989, 994(1997)[hereinafter Lemley, Economics of Improvement]("In a private market economy, individuals will not invest in invention or creation... unless they can reasonably expect to make a profit from the endeavor. ") Christian Koboldt Intellectual Property and Optimal Copyright Protection, 19 J OF CULTURAL ECON. 131 134(1995)(same) See, e.g, CARL SHAPIRo HAL VARIAN, INFORMATION RULES: A STRATEGIC GUIDE TO THE NETWORK ECONOMY 3(1999), F M. SChERER DAvID ROSs, INDUSTRIAL
Vol. ##] THE PROBLEM OF UNIFORMITY COST 6 A. Standard Economic Justification for Intellectual Property Rights Intellectual property rights are a second-best solution to an “appropriability problem.” The now-familiar utilitarian justification for intellectual property law starts with Thomas Jefferson=s observation that information=s Apeculiar character . . . is that no one possesses the less, because every other possesses the whole of it. . . . . [H]e who lites his taper at mine, receives light without darkening me.@ 11 For Jefferson, the capacity for information to Afreely spread from one to another over the globe@ is part of nature=s Abenevolent[]@ design.12 For the economist, however, the Abenevolent@ design of information poses a problem.13 To the extent that pecuniary motivation drives innovation, we should not expect to see useful information14 produced unless the producer can recoup his or her investment.15 Because the distribution of valuable information cannot be controlled in the ways that distribution of scarce goods like tubes of toothpaste or radial tires might be B that is, information can be consumed nonrivalrously B the producer, acting alone, cannot rely on competitive markets to supply a sufficient return to make the investment in producing such information worthwhile.16 The government 11 Letter from Thomas Jefferson to Isaac McPherson (Aug. 13, 1813), in 6 THE WRITINGS OF THOMAS JEFFERSON 180 (H.A. Washington, ed. 1861). 12 Id. 13 Information exhibits public goods characteristics. A public good can be consumed without depletion (non-rivalrous consumption) and can be withheld from nonpaying beneficiaries only at prohibitive cost (non-excludability). See ROBERT COOTER & THOMAS ULEN, LAW AND ECONOMICS 40-41 (2d ed. 1997); see also Wilfried Ver Eecke, Public Goods: An Ideal Concept, 28 J. OF SOCIO-ECONOMICS 139 (1999). Information is imperfectly excludable but it is because it is its non-rivalrous quality that makes it “problematic.” 14 The terms Avaluable@ and Auseful@ information in this Article refer to information that is costly to produce and that members of our society find to be useful, informative, enriching, or otherwise of value. Such information includes the ideas and expressions or embodiments of those ideas found in, for example, novels, movies, music, methods for manufacturing useful articles (medicines, computers, clothing, etc.), computer software, etc. 15 E.g., Mark A. Lemley, The Economics of Improvement in Intellectual Property, 75 TEX. L. REV. 989, 994 (1997) [hereinafter Lemley, Economics of Improvement] (AIn a private market economy, individuals will not invest in invention or creation . . . unless they can reasonably expect to make a profit from the endeavor.@); Christian Koboldt, Intellectual Property and Optimal Copyright Protection, 19 J. OF CULTURAL ECON. 131, 134 (1995) (same). 16 See, e.g., CARL SHAPIRO & HAL VARIAN, INFORMATION RULES: A STRATEGIC GUIDE TO THE NETWORK ECONOMY 3 (1999); F.M. SCHERER & DAVID ROSS, INDUSTRIAL
Vol # THE PROBLEM OF UNIFORMITY COST response has been to grant and administer rights under patent and copyright law. Such rights give the innovator the r to exclude or inhibit direct competition, which yields potential power over price. If demand is sufficient, the innovator can use that power to earn a positive return on investments in innovation While addressing underproduction, intellectual property rights also impose social costs. Professor Lemley nicely summarizes these as follows First, intellectual property rights distort markets away from the competitive norm, and therefore create static inefficiencies in the form of deadweight losses. Second intellectual property rights interfere with the ability of other creators to work, and therefore create dynar inefficiencies. Third, the prospect of intellectual property rights encourages rent-seeking behavior that is socially wasteful. Fourth, enforcement of intellectual property right research and development is itself distortionary > Stment in imposes administrative costs. Finally, overinve B The problem of Uniformity Cost Economic analysts generally agree that these social costs must be minimized, and intellectual property rights should be no more robust than necessary to induce the desired level of investment in cultural and technological innovation. However, in both the economic and the law- MARKET STRUCTURE AND ECONOMIC PERFORMANCE 622(3d ed 1990); See, e.g, Mark tellectual Property, and Fr 054-55(2005)[hereinafter Lemley, Free Riding lan E Novos Michael Waldman The Effects of Increased Copyright Protection: An Analytic Approach, 92 J. POL. ECON 236,237(1984) Lemley, Free Riding, supra note XX, at 1058-59 See, e.g., Lunney, Quiet Revolution, supra note XX, at 5(arguing that patent protection should be provided only to the precise extent necessary to secure each individual innovations ex ante profitability and acknowledging that this level will have to account for unsuccessful research efforts); William W. Fisher Ill, Property and Contract on the Internet, 73 CHI.-KENT L. REV. 1203, 1249(1998)(arguing that the goal of copyright law is to give creators enough entitlements to induce them to produce the works from which we all benefit but no more"). The canonical version of this argument was voiced by lord Macauley, who argued that a grant of copyright was a grant of an evil monopoly and that [for the sake of the good we must submit to the evil; but the evil ought not to last a day longer than is necessary for the purpose of securing the good. Thomas Babington, Lord Macaulay, A Speech Delivered in the house of Commons on the 5th of February, 1841
Vol. ##] THE PROBLEM OF UNIFORMITY COST 7 response has been to grant and administer rights under patent and copyright law. Such rights give the innovator the power to exclude or inhibit direct competition, which yields potential power over price. If demand is sufficient, the innovator can use that power to earn a positive return on investments in innovation. While addressing underproduction, intellectual property rights also impose social costs. Professor Lemley nicely summarizes these as follows: First, intellectual property rights distort markets away from the competitive norm, and therefore create static inefficiencies in the form of deadweight losses. Second, intellectual property rights interfere with the ability of other creators to work, and therefore create dynamic inefficiencies. Third, the prospect of intellectual property rights encourages rent-seeking behavior that is socially wasteful. Fourth, enforcement of intellectual property rights imposes administrative costs. Finally, overinvestment in research and development is itself distortionary.17 B. The Problem of Uniformity Cost Economic analysts generally agree that these social costs must be minimized, and intellectual property rights should be no more robust than necessary to induce the desired level of investment in cultural and technological innovation.18 However, in both the economic and the law- MARKET STRUCTURE AND ECONOMIC PERFORMANCE 622 (3d ed. 1990); See, e.g., Mark A. Lemley, Property, Intellectual Property, and Free Riding, 83 TEX. L. REV. 1031, 1054-55 (2005) [hereinafter Lemley, Free Riding]; Ian E. Novos & Michael Waldman, The Effects of Increased Copyright Protection: An Analytic Approach, 92 J. POL. ECON. 236, 237 (1984). 17 Lemley, Free Riding, supra note XX, at 1058-59. 18 See, e.g., Lunney, Quiet Revolution, supra note XX, at 5 (arguing that patent protection should be provided only to the precise extent necessary to secure each individual innovation’s ex ante profitability and acknowledging that this level will have to account for unsuccessful research efforts); William W. Fisher III, Property and Contract on the Internet, 73 CHI.-KENT L. REV. 1203, 1249 (1998) (arguing that the goal of copyright law “is to give creators enough entitlements to induce them to produce the works from which we all benefit but no more”). The canonical version of this argument was voiced by Lord Macauley, who argued that a grant of copyright was a grant of an evil monopoly and that “[f]or the sake of the good we must submit to the evil; but the evil ought not to last a day longer than is necessary for the purpose of securing the good.” Thomas Babington, Lord Macaulay, A Speech Delivered in the House of Commons on the 5th of February, 1841
Vol ## THE PROBLEM OF UNIFORMITY COST and-economics literature, the problem of social cost in intellectual property law often is discussed at a very high level of abstraction. The literature surrounding the optimal length of a patent is a typical example Neoclassical economic models concerning an optimal patent term often hold that optimality is conditional, implicitly recognizing that efficiency might dictate varying terms from patent to patent. Other analysts make the point more explicitly. But these economists offer no suggestion for how variable patent terms might be implemented, and those who contemplate the matter find the administrative difficulties intractable These economists apparently have recognized and despaired over, the problem of uniformity cost in intellectual property law. This despair is premature. Economic analysis can help identify situations in which uniformity costs are particularly high, but it will require pragmatic legal analysis to identify ways in which the legal system can competentI redress the problem 1. The Problem Legal scholars only recently have begun to analyze the social costs or uniform rights as a general problem in intellectual property law Analyzing U.S. patent law, Professor Lunney has advanced a formal economic model of uniformity cost that assesses the trade-offs between strictly uniform rights, rights tailored to individual innovations, an certain intermediate options. At bottom he shows"[e]ven where an innovative product represents the most valuable use of available resources in Speeches by Lord Macaulay with his Minute on Indian Education 156, 162(G M Professor John Duffy argues that the mobility of capital makes analysis causal connection between rights and investment levels unstable. See John F Intellectual Property Isolationism And The Average Cost Thesis, 83 TEX. L. REV 1078-89(2005). But, Professor Lemley rightly responds that because intellectual property rights distort the market away from competitive equilibrium, entry will not necessarily compete away supracompetitive returns. See Lemley, What is Different, supra note Xx at 1102-03 i9 See infra notes XX and accompanying text( discussing economic literature on patent See infra notes XX and accompanying text(discussing recognition of uniformity cost in the literature) 2I See Lunney, Quiet Revolution, supra note XX, at 6(stating that uniformity costs rise as gap between optimal uniform level of protection and level needed for individual innovation increases) 2 See Lunney, Quiet Revolution,, supra note XX, SIV
Vol. ##] THE PROBLEM OF UNIFORMITY COST 8 and-economics literature, the problem of social cost in intellectual property law often is discussed at a very high level of abstraction. The literature surrounding the optimal length of a patent is a typical example. Neoclassical economic models concerning an optimal patent term often hold that optimality is conditional, implicitly recognizing that efficiency might dictate varying terms from patent to patent.19 Other analysts make the point more explicitly.20 But these economists offer no suggestion for how variable patent terms might be implemented, and those who contemplate the matter find the administrative difficulties intractable. These economists apparently have recognized and despaired over, the problem of uniformity cost in intellectual property law.21 This despair is premature. Economic analysis can help identify situations in which uniformity costs are particularly high, but it will require pragmatic legal analysis to identify ways in which the legal system can competently redress the problem. 1. The Problem Legal scholars only recently have begun to analyze the social costs of uniform rights as a general problem in intellectual property law. Analyzing U.S. patent law, Professor Lunney has advanced a formal economic model of uniformity cost that assesses the trade-offs between strictly uniform rights, rights tailored to individual innovations, and certain intermediate options.22 At bottom he shows “[e]ven where an innovative product represents the most valuable use of available resources in Speeches by Lord Macaulay with his Minute on Indian Education 156, 162 (G.M. Young ed., 1935). Professor John Duffy argues that the mobility of capital makes analysis of the causal connection between rights and investment levels unstable. See John F. Duffy, Intellectual Property Isolationism And The Average Cost Thesis, 83 TEX. L. REV. 1077, 1078-89 (2005). But, Professor Lemley rightly responds that because intellectual property rights distort the market away from competitive equilibrium, entry will not necessarily compete away supracompetitive returns. See Lemley, What is Different, supra note XX, at 1102-03. 19 See infra notes XX and accompanying text (discussing economic literature on patent length). 20 See infra notes XX and accompanying text (discussing recognition of uniformity cost in the literature). 21 See Lunney, Quiet Revolution,, supra note XX, at 6 (stating that uniformity costs rise as gap between optimal uniform level of protection and level needed for individual innovation increases). 22 See Lunney, Quiet Revolution,, supra note XX, § IV
Vol ## THE PROBLEM OF UNIFORMITY COST an optimal uniform scheme of protection will provide protection that will leave some desirable innovative products unprofitable To illustrate the point, imagine four innovations, A, B, C, and D These could be musical compositions, types of business software biotechnological inventions, or chemical compounds with pharmacological uses. Society places a value of 50 on each of these if it is available for use and is free from any intellectual property rights. Assume that intellectual property rights apply uniformly to all covered forms of information and can be calibrated to yield levels of protection ranging from 0 to 3. Innovations B-D will require a level of protection above to be created and distributed. so that a alone will be created and distributed at 0. A and b will be created and distributed at level 1. and so on. as protection increases, however, social value decreases because some users are priced out of desired uses for which they would pay more than marginal cost. Assume that each increase in the level of protection reduces the social value of each innovation by 10. Sliding the protection lever upward yields the following distribution of social values Table 1 Level of Protection Innovations Created Total Social Value and Distributed B A,B.C A,B,C,D a policymaker interested in maximizing social value from intellectual property rights but bound by the uniformity condition would set the level of protection at 2, leaving innovation D unprofitable even though society places a net positive value of 20 on having it created. Within this highly stylized example, it is easy to see that if the uniformity condition could be relaxed, it would be possible to adjust rights to entice the creation of A-D by, for example, eliminating protection for A and reducing the scope or duration of rights granted to B See lunney, Quiet Revolution,, supra note XX, at 50-51 Some empirical data suggests that innovations such as A are more than hypothetical Edwin Mansfield interviewed research and development managers from 100 randomly selected firms to ask what percentage of the firms inventions would have been developed and brought to market in the absence of patent protection. See Edwin Mansfield, Patents and Innovation: An Empirical Study, 32 MGMT SCIENCE 173(1986)
Vol. ##] THE PROBLEM OF UNIFORMITY COST 9 . . . . an optimal uniform scheme of protection will provide protection that will leave some desirable innovative products unprofitable.” 23 To illustrate the point, imagine four innovations, A, B, C, and D. These could be musical compositions, types of business software, biotechnological inventions, or chemical compounds with pharmacological uses. Society places a value of 50 on each of these if it is available for use and is free from any intellectual property rights. Assume that intellectual property rights apply uniformly to all covered forms of information and can be calibrated to yield levels of protection ranging from 0 to 3. Innovations B-D will require a level of protection above 0 to be created and distributed, so that A alone will be created and distributed at 0, A and B will be created and distributed at level 1, and so on. As protection increases, however, social value decreases because some users are priced out of desired uses for which they would pay more than marginal cost. Assume that each increase in the level of protection reduces the social value of each innovation by 10. Sliding the protection lever upward yields the following distribution of social values: Table 1 Level of Protection Innovations Created and Distributed Total Social Value 0 A 50 1 A, B 80 2 A,B,C 90 3 A,B,C,D 80 A policymaker interested in maximizing social value from intellectual property rights but bound by the uniformity condition would set the level of protection at 2, leaving innovation D unprofitable even though society places a net positive value of 20 on having it created. Within this highly stylized example, it is easy to see that if the uniformity condition could be relaxed, it would be possible to adjust rights to entice the creation of A-D by, for example, eliminating protection for A and reducing the scope or duration of rights granted to B.24 23 See Lunney, Quiet Revolution,, supra note XX, at 50-51. 24 Some empirical data suggests that innovations such as A are more than hypothetical. Edwin Mansfield interviewed research and development managers from 100 randomlyselected firms to ask what percentage of the firm’s inventions would have been developed and brought to market in the absence of patent protection. See Edwin Mansfield, Patents and Innovation: An Empirical Study, 32 MGMT SCIENCE 173 (1986)