The uneasy case for patent races over auctions

Stanford Law Review

Volumn 60, Issue 3, 2007, Pages 803-862

  Abramowicz, Michael ^a  

^a GEORGE WASHINGTON UNIVERSITY   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 40749093601     PISSN: 00389765     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Review

References (264)

1. Under the Copyright Act, "Copyright protection subsists . . . in original works of authorship fixed in any tangible medium of expression." 17 U.S.C § 102(a) (2000). A caveat is that in theory, someone independently could originate the exact same work and be entitled to a copyright too. See Sheldon v. Metro-Goldwyn Pictures Corp., 81 F.2d 49, 54 (2d Cir. 1936) ("[I]f by some magic a man who had never known it were to compose anew Keats's Ode on a Grecian Urn," he would be entitled to copyright protection.). That, however, virtually never happens in practice, because of the large number of ways in which ideas can be expressed.
2. Under the Lanham Act, "The term 'trademark' includes any word, name, symbol, or device, or any combination thereof . . . used by a person, or . . . which a person has a bona fide intention to use in commerce and applies to register." 15 U.S.C. § 1127 (2000).
3. The Patent Act provides, "Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title." 35 U.S.C. § 101 (2000).
4. The most famous article urging a shift from the administrative approach to auctions is R. H. Coase, The Federal Communications Commission, 2 J.L. & ECON. 1 (1959), which included a brief version of what would later be known as the Coase Theorem. For a later analysis,
5. see Thomas W. Hazlett, The Rationality of U.S. Regulation of the Broadcast Spectrum, 33 J.L. & ECON. 133 (1990).
6. See, e.g., Richard L. Fullerton & R. Preston McAfee, Auctioning Entry into Tournaments, 107 J. POL. ECON. 573, 581-82 (1999) (showing that under certain conditions, an auction should be used to limit a research tournament to the two participants willing to pay the most for the right to participate).
7. The difference is that "in a research tournament, the terminal date is fixed, and the quality of innovations varies, while in an innovation race, the quality standard is fixed, and the date of discovery is variable." Curtis R. Taylor, Digging for Golden Carrots: An Analysis of Research Tournaments, 85 AM. ECON. REV. 872, 874 (1995).
8. A notable exception is Gideon Parchomovsky's proposal for using auctions to distribute domain names. See Gideon Parchomovsky, On Trademarks, Domain Names, and Internal Auctions, 2001 U. III. L. REV. 211. NO one appears to have considered auctioning rights to genres of copyrighted works, and such broad copyrights would almost certainly be unattractive on economic and free speech grounds.
9. See Michael Abramowicz, A Theory of Copyright's Derivative Right and Related Doctrines, 90 MINN. L. REV. 317, 356-57 (2005) (explaining why a "super-copyright" regime would be problematic). The possibility that the government might auction patent rights has received only two footnotes plus one sentence worth of attention in the literature.
10. See infra Parts I.A-I.B.
11. The Patent and Trademark Office already defines various invention classifications through the U.S. Patent Classification System. See, e.g., U.S. Patent and Trademark Office, Patent Classification Home Page, http://www.uspto.gov/go/classification;
12. see also 35 U.S.C § 8 (2000) (authorizing the classification system). A single patent conceivably might be classified in several classifications, however, and if rights to a particular subclass were to be auctioned, more attention might need to be paid to careful definition of the scope of the subclass.
13. Edmund W. Kitch, The Nature and Function of the Patent System, 20 J.L. & ECON. 265 (1977).
14. Id. at 267 (emphasizing the issues of patent scope and early patent application and issuance).
15. For example, Robert Merges and Richard Nelson conclude: Public policy, including patent law, ought to encourage inventive rivalry, and not hinder it. As the "race to invent" models show, a rivalrous structure surely has its inefficiencies. But such a structure does tend to generate rapid technical progress and seems a much better social bet than a regime where only one or a few organizations control the development of any given technology.
16. Robert P. Merges & Richard R. Nelson, On the Complex Economics of Patent Scope, 90 COLUM. L. REV. 839, 908 (1990);
17. see also id. at 877 ("[W]e have little faith in the imagination and willingness of a 'prospect' holder to develop that prospect as energetically or creatively as she would when engaged in competition."). Mark Lemley insists that "[t]he claim that an initial inventor is better suited to control research than the market is fundamentally an empirical one" and that this empirical claim "has been tested and found wanting."
18. Mark A. Lemley, Ex Ante Versus Ex Post Justifications for Intellectual Property, 71 U. CHI. L. REV. 129, 140 (2004). The ultimate question, however, is not simply whether an "initial inventor" will better control research than competitive firms, but whether an early granting of an intellectual property right may lead to earlier invention. There are strong theoretical reasons for believing that the answer is affirmative.
19. See infra Part I.C (considering Professor Duffy's demonstration of how the reduction in competition associated with patent races may accelerate innovation).
20. See Tim Wu, Intellectual Property, Innovation, and Decentralized Decisions, 92 VA. L. REV. 123 (2006) (arguing that patents may harm innovation by centralizing decisionmaking, but acknowledging that centralized information may be useful in some information contexts).
21. For an argument that centralized decisionmaking may reduce costs associated with duplication and delay, see Patrick Bolton & Joseph Farrell, Decentralization, Duplication, and Delay, 98 J. POL. ECON. 803 (1990). Wu disputes the relevance of Bolton and Farrell's analysis for intellectual property, arguing that "the relative quickness of centralized decisionmaking structures . . . seems less important in the intellectual property context."
22. Wu, supra note 12, at 130 n.18. But in innovation as in tax, timing is everything, since any feasible and potentially valuable invention presumably will be created someday. Bolton and Farrell note that if two firms have equal costs, they may enter into what eventually will be a natural monopoly market simultaneously, creating duplication, and if both have high costs, they may both wait, creating delay.
23. Bolton & Farrell, supra, at 805. These problems exist as well in the intellectual property context, when two potential inventors have roughly equal chances of winning a patent.
24. For overviews of the economics of patent races, see JEAN TIRÓLE, THE THEORY OF INDUSTRIAL ORGANIZATION 394-414 (1988);
25. Partha Dasgupta & Joseph Stiglitz, Uncertainty, Industrial Structure, and the Speed of R&D, 11 BELL J.ECON. 1 (1980).
26. See infra Part II.A.2.a.
27. The Federal Circuit has embraced "inventing around," insisting that it "is the stuff of which competition is made." State Indus., Inc. v. A.O. Smith Corp., 751 F.2d 1226, 1236 (Fed. Cir. 1985);
28. see also Paul N. Katz & Robert R. Riddle, Designing Around a United States Patent, 45 S. TEX. L. REV. 647, 649 (2004) (endorsing the Federal Circuit conclusion and claiming that designing around is "one of the ways in which the patent system works to the advantage of the public"). Some commentators, however, have emphasized the inefficiencies associated with inventing around.
29. See, e.g., Louis Kaplow, The Patent-Antitrust Intersection: A Reappraisal, 97 HARV. L. REV. 1813, 1869 (1984) (recognizing that if the new invention adds no functionality, the expense of creating it is wasted and the reward function of the patent system is dampened).
30. See infra text accompanying notes 106-07.
31. See Donald G. McFetridge & Douglas A. Smith, Patents, Prospects, and Economic Surplus: A Comment, 23 J.L. & ECON. 197 (1980).
32. See, e.g., Michael Abramowicz, Perfecting Patent Prizes, 56 VAND. L. REV. 115 (2003);
33. Steve P. Calandrillo, An Economic Analysis of Property Rights in Information: Justifications and Problems of Exclusive Rights, Incentives to Generate Information, and the Alternative of a Government-Run Reward System, 9 FORDHAM INTELL. PROP. MEDIA & ENT. L.J. 301 (1998);
34. Michael Polanvyi, Patent Reform, 11 REV. ECON. STUD. 61 (1944);
35. Steven Shavell & Tanguy van Ypersele, Rewards Versus Intellectual Property Rights, 44 J.L. & ECON. 525, 537-39 (2001). Conceivably, an auction system might be combined with a reward system, in which the government sometimes might take exclusive rights it had previously granted in exchange for just compensation. The reward system would thus be invoked only once the auction system had provided sufficient incentives for an auction winner to generate (and perhaps commercialize) socially valuable inventions.
36. John F. Duffy, Rethinking the Prospect Theory of Patents, 71 U. CHI. L. REV. 439, 476-80 (2004). Some earlier commentators recognized the similarity between a patent race and an auction.
37. See, e.g., Wolfgang Leininger, Escalation and Cooperation in Conflict Situations, 33 J. CONFLICT RESOL. 231, 233 (1989) (analogizing a patent race to an "escalation" or "all pay" auction in which all auction participants must pay but only the highest bidder wins). Duffy, however, appears to be the first commentator to recognize that patent racers bid not only by committing a particular amount of money to research and development but also by deciding when to commit that money.
38. See infra Part LC For example, Duffy rejects the classic prospect theory position that patents should be broad, but the possibility of explicit auctions helps crystallize the benefits of broad patents. The endorsement of relatively broad patents is critical to the Article's ultimate policy proposal.
39. See infra Part IV.B.
40. See infra Part III.A.
41. See infra Parts III.B.1, III.B.2 (considering and comparing patent term auctions and pure cash auctions).
42. Even where a patent might cover numerous products or services, a patent auction could approximate the low price approach by awarding the patent to the bidder who agrees to accept the lowest fraction of typical patent damages, compulsory license rates, or regulated prices. See infra text accompanying notes 179-82.
43. To the extent that the existing patent system functions as an implicit auction, it too may have this feature, as the winning inventors are those who offer the shortest effective patent terms. See infra Part LC (discussing Professor Duffy's insight). Explicit price auctions, however, may be more efficient, because the last increment of monopoly pricing tends to be the most socially costly.
44. See infra text accompanying notes 167-73 (discussing the implications of an analysis by Ian Ayres and Paul Klemperer). A slight price concession for a long patent term is likely more efficient than a large price concession for a short term, if the two provide equal value to the bidders.
45. All that is needed to implement such a design is an announcement of some conversion rate between the different components of an auction bid. Another alternative is to declare a fixed requirement for all components but one, which then becomes the auction currency. Indeed, as argued below, it will generally be necessary even for patent term and price auctions to insist that bids have at least some cash component, as a means of deterring frivolous bids. See infra text introducing Part III.
46. The Article will consider how the government might seek to harness private information in comparing bids. If such approaches could be perfected, the optimal approach might not be for the government to hold patent auctions, but for the government to hold auctions that would give bidders incentives to define the optimal structure of patent law in particular technological fields. See infra Part IV.B.
47. See, e.g., WILLIAM D. NORDHAUS, INVENTION, GROWTH, AND WELFARE: A THEORETICAL TREATMENT OF TECHNOLOGICAL CHANGE 76 (1969) ("[A] longer [patent] life increases invention and thus gives on balance a larger amount of output for a given level of inputs. . . . [But] a longer life means that the monopoly on information lasts longer and thus there are more losses from inefficiencies associated with monopoly.").
48. See infra note 103 and accompanying text (discussing the doctrine of equivalents);
49. see also Douglas Lichtman, Rethinking Prosecution History Estoppel, 71 U. CHI. L. REV. 151 (2004) (providing an analysis of the doctrine of equivalents and the related doctrine of prosecution history estoppel).
50. Yoram Barzel, Optimal Timing of Innovations, 50 REV. ECON. & STAT. 348 (1968).
51. Id. at 351. Zero is used here in the conventional economic sense of corresponding to a normal rate of return.
52. Id. at 349 (assumptions 1 and 3).
53. Id. at 352 n.11 ("A conceptual way to overcome this kind of misallocation is by granting (or by auctioning) monopoly rights on potential innovations before resources are committed to the innovating activity.").
54. Id. at 349 (assumption 4).
55. Id. at 350.
56. Id. at 355 tbl.2.
57. This is clear in the first sentence of his piece. See id. at 348 ("It is widely recognized that when innovators are unable to realize the full benefits generated by their innovations the profit motive may not provide an incentive strong enough for them to innovate at the socially optimal rate.").
58. For attempts to measure the extent of spillovers, see Timothy F. Bresnahan, Measuring the Spillovers from Technical Advance: Mainframe Computers in Financial Services, 76 AM. ECON. REV. 742, 753 (1986);
59. Edwin Mansfield et al, Imitation Costs and Patents: An Empirical Study, 91 ECON. J. 907 (1981);
60. and M. Ishaq Nadiri, Innovations and Technological Spillovers (Nat'l Bureau of Econ. Research, Working Paper No. 4423, 1993).
61. See, e.g., Janusz A. Ordover, Economic Foundations and Considerations in Protecting Industrial and Intellectual Property, 53 ANTITRUST L.J. 503, 509 (1984) (noting that governmental research subsidies may be needed when stringent property rights are not available to reduce the extent of inventive spillovers).
62. Kitch, supra note 9.
63. Id. at 265.
64. Id.
65. Id. at 265-66.
66. Id. at 266 n.4.
67. Id. at 276-78.
68. McFetridge & Smith, supra note 18, at 198.
69. See, e.g., Roger L. Beck, The Prospect Theory of the Patent System and Unproductive Competition, 5 RES. L. & ECON. 193, 194 (1983).
70. Kitch, supra note 9, at 285.
71. Id.
72. For a more detailed version of this argument, see Michael Abramowicz, The Danger of Underdeveloped Patent Prospects, 92 CORNELL L. REV. 1065 (2007).
73. See generally NORDHAUS, supra note 28, at 76 (exploring this trade-off).
74. See infra Part III.
75. Duffy, supra note 20.
76. Duffy's analysis also helped answer the second major criticism of the prospect theory, that our patent system is not a prospect system that genuinely gives a patentee the power to control innovation, because it allows improvers to receive patents within the claims of the granted patent. Id. at 443 n.22 (citing sources making this objection);
77. see supra note 47 and accompanying text (noting this objection). A true prospect system that did prevent improvements, moreover, might stifle innovation. Duffy notes that the power of a prospect patent holder to block holders of improvement patents from using their inventions provides some control over invention.
78. Duffy, supra note 20, at 485. More importantly, the inevitability of patent expiration provides prospect patent holders a continuing incentive to compete to obtain patentable improvements.
79. Id. at 488. Thus, even under a prospect system, inventors will compete in time to improve earlier, with the result that improvements also end up in the public domain earlier.
80. McFetridge & Smith, supra note 18, at 198. Later commentators suggested that the patent system might be seen as a mechanism that seeks to reduce the sum of the inefficiencies associated with patent races and competitive development of patented inventions.
81. See Mark F. Grady & Jay I. Alexander, Patent Law and Rent Dissipation, 78 VA. L. REV. 305, 317 (1992) ("[A] full accounting of the effects of the patent system must balance the savings in reduced follow-on investment against the losses from accelerated pioneering investment.").
82. Duffy, supra note 20, at 443-44, 464-75.
83. Id. at 475-80.
84. Harold Demsetz, Why Regulate Utilities?, 11 J.L. & ECON. 55, 63 (1968).
85. Demsetz's theory on natural monopolies does not necessarily depend on the government holding a centralized auction, though a centralized auction in effect usefully facilitates collusion among buyers. Id. at 57-58 (providing a simple example involving a natural monopoly in production of license plates).
86. Duffy, supra note 20, at 469-75 (offering a model in which inventors race to obtain a patent).
87. There may be some randomness in who obtains the patent, because an attempt to invent may or may not be successful. Nonetheless, this does not change the fundamental conclusions of Duffy's analysis. Id. at 480-82.
88. Id. at 470-72.
89. See supra note 11 (noting commentators' failure to recognize the point later made by Duffy).
90. Kitch recognized the possible inefficiency of inventing around, but thought prospects might help patentees enter into agreements with potential competitors that would help "forestall the wasteful investment." Kitch, supra note 9, at 279. Kitch, however, acknowledged that "the patent holder cannot offer discounts to everyone who threatens to look for a substitute."
91. Id.
92. Duffy, supra note 20, at 499-500.
93. See infra Part II.A. 1.
94. See Ian Ayres & Paul Klemperer, Limiting Patentees' Market Power Without Reducing Innovation Incentives: The Perverse Benefits of Uncertainty and Non-Injunctive Remedies, 97 MICH. L. REV. 985, 1031-32 (1999) (suggesting a "duopoly auction" in which the patentee would receive the revenues from an auction for a complete license to the patent, as a way of ensuring some degree of competition within the patent system);
95. Michael Kremer, Patent Buyouts: A Mechanism for Encouraging Innovation, 113 Q.J. ECON. 1137 (1998) (suggesting an auction that would be consummated only some of the time, with the information from the bids used the rest of the time to buy out the patentee and place the patent into the public domain).
96. Oren Bar-Gill & Gideon Parchomovsky, A Marketplace for Ideas?, 84 TEX. L. REV. 395, 419 (2005). Bar-Gill and Parchomovsky also suggested that under certain circumstances, it might be optimal instead to allow anyone to develop the idea, subject to a liability rule compulsory license regime.
97. See id. at 412-17.
98. For example, idea registrants would have to satisfy the traditional criteria of usefulness, novelty, and nonobviousness. Id. at 426. The inventors would also have to demonstrate "developability," i.e., that the invention "constitutes a material contribution to the innovative process," and does not fit within the category of "fanciful ideas (i.e., those deemed undevelopable by a person skilled in the technological field to which the idea belongs)."
99. Id. at 427. Although there is substantial ambiguity about the developability requirement, Bar-Gill and Parchomovsky make clear that it is not a high bar.
100. See id. ("Importantly, developability is more lenient than implementability. Ideas, as opposed to patents, need not be readily implementable. By its very nature, an idea lacks the detail required for implementation.").
101. Id. at 419 (considering an alternative auction design in which "developers would not bid on ideas in their entirety, but rather on individual applications of ideas they wished to pursue").
102. Id. at 418. Bar-Gill and Parchomovsky "believe that a relatively small percentage of ideas would be picked up by developers."
103. Id. This conclusion, however, depends on the additional stipulation that there would be a rule requiring some minimum bids.
104. Id. at 418 n.70. Bar-Gill and Parchomovsky suggest that this will prevent "idea conceivers [from] bidding token amounts on their own ideas just to prevent them from falling into the public domain."
105. Id. The problem, however, is more complicated. Someone might make a relatively low bid to obtain the option to develop, even recognizing the small likelihood that the option will turn out to be worth exercising.
106. See supra Part I.C. Thus, some care must be taken in setting the appropriate level of the minimum bid.
107. Bar-Gill & Parchomovsky, supra note 68, at 421. Bar-Gill and Parchomovsky do not explicitly say that failure to develop would lead to development rights being placed in the public domain. This is the most natural reading of their proposal, but they say only that the entitlement would be "strip[ped] away."
108. Id. In theory, the entitlement could be reauctioned if development did not occur.
109. Id. at 405. Bar-Gill and Parchomovsky argue that the possibility of entering nondisclosure agreements does not solve the problem, because of the paradox that the best legal developers will not want to risk signing nondisclosure agreements until they know about the invention.
110. Id. at 404.
111. Id. at 408-09.
112. Id. at 417.
113. Bar-Gill and Parchomovsky concede this point but respond that "the privately optimal auction design will generally diverge from the socially optimal auction design." Id. at 417 n.66. They do not, however, explain what the divergence would be, or justify the conclusion that a one-size-fits-all auction design chosen by the government will be more socially optimal than the design chosen by individual idea registrants.
114. Sometimes individualized negotiation might be more efficient than an auction because an auction requires all participants to research the invention, even though at most one will win the right to develop it. But cf. Jeremy Bulow & Paul Klemperer, Auctions Versus Negotiations, 86 AM. ECON. REV. 180 (1996) (offering a model showing that an auction generally will be more efficient than a negotiation with one less participant).
115. See supra note 50 and accompanying text.
116. For a discussion of the European approach to compulsory licensing, see Ronald E. Myrick, Influences Affecting the Licensing of Rights in a Unitary European Market, 4 FORDHAM INTELL. PROP. MEDIA & ENT. L.J. 81, 95 (1993).
117. Bar-Gill and Parchomovsky recognize the difficulty of enforcing this approach. Bar-Gill & Parchomovsky, supra note 68, at 421-22. ("[Verifying whether the development requirement has been satisfied might be costly, perhaps prohibitively so."). One problem is what should count as a patent or product sufficient to avoid the invention being placed in the public domain. A firm might produce a useless patent or product simply to avoid losing patent rights. That would not only be wasteful, but would also evade the Bar-Gill and Parchomovsky rule, which does not require continued development after a patent or product is initially produced.
118. Others make similar assumptions. For example, Mark Lemley argues that "once an intellectual property right expires, many companies can compete to make the good, and they will do so only so long as they can manufacture and distribute it for less money than people will pay to buy it." Lemley, supra note 11, at 136. Professor Lemley recognizes the counterargument that some form of intellectual property right may be necessary to provide "adequate incentives to improve on an existing work."
119. Id. at 138. He responds that in these cases, the proper response should be to provide intellectual property rights to the improver.
120. Id. at 139. Patents, however, may need to be developed in ways that, at least under current doctrine, are nonpatentable.
121. See Abramowicz, supra note 50, at 1098-103 (discussing development activities that do not themselves entitle developers to additional intellectual property protection). It is not easy to see, for example, even how a hypothetical intellectual property regime would provide a reward to someone who markets a product without giving some right over the product itself. In a world in which some development and improvement of patented technologies does not receive the benefit of independent intellectual property protection, the placement of inventions into the public domain has at least the potential to harm ex post efficiency. Indeed, Lemley concedes that there are some circumstances in which prospecting rights may be justified.
122. See Lemley, supra note 11, at 141 ("Prospect theory is needed when control over subsequent development is a necessary part of the incentive to produce the pioneering invention in the first place, as is arguably true with pharmaceuticals.").
123. Such a justification may not be straightforward. A threat to deprive a would-be developer of a patent right for failure to develop might inefficiently accelerate invention efforts. See Barzel, supra note 30, at 348 (showing that innovation may occur inefficiently early).
124. See supra Part I.A.
125. See F.M. Scherer, Nordhaus' Theory of Optimal Patent Life: A Geometric Reinterpretation, 62 AM. ECON. REV. 422, 424 (1972).
126. See, e.g., WILLIAM M. LANDES & RICHARD A. POSNER, THE ECONOMIC STRUCTURE OF INTELLECTUAL PROPERTY LAW 421 (2003) (noting that transaction costs may provide a justification for narrowing intellectual property rights).
127. Nonetheless, in the absence of exogenous shocks, Duffy's argument suggests that we should gradually transition to a longer and longer term. See John F. Duffy, A Minimum Optimal Patent Term (2003) (unpublished manuscript), available at http://www.ssrn.com/abstract=354282 (explaining that a dynamic model of invention suggests that the patent term should be no less than a particular length).
128. See supra Part I.C.
129. For example, the right might be to any process providing for positional control in chemical synthesis. See Ralph C. Merkle, Molecular Manufacturing: Adding Positional Control to Chemical Synthesis, CHEMICAL DESIGN AUTOMATION NEWS, Sept.-Oct. 1993, at 1.
130. See Mark A. Lemley, The Economics of Improvement in Intellectual Property Law, 75 TEX. L. REV. 989, 1000-13 (1997) (discussing patent doctrines related to improvement patents).
131. A rule of first possession also will tend to emerge because it is easy for a court to recognize. See, e.g., Richard A. Epstein, Possession as the Root of Title, 13 GA. L. REV. 1221, 1222-23 (1979). An auction regime requires some type of centralized administrative machinery.
132. The heterogeneity of potential prospectors will tend to reduce rent dissipation. See, e.g., Dean Lueck, The Rule of First Possession and the Design of the Law, 38 J.L. & ECON. 393, 401 (1995);
133. Wing Suen, Rationing and Rent Dissipation in the Presence of Heterogeneous Individuals, 97 J. POL. ECON. 1384 (1989). The result is that the best-situated prospectors can expect to receive a positive rent from a prospecting system. In an auction system, the best-situated party also may receive a rent. In a second-price auction, for example, the winning bidder enjoys a rent equal to the difference between the highest and the next-highest bid.
134. See generally William Vickrey, Counterspeculation, Auctions, and Competitive Sealed Tenders, 16 J. FIN. 8 (1961) (introducing the idea of a second-price sealed bid auction). Even in a first-bid auction, a party expecting to be the top bidder will bid somewhat less than its actual valuation to ensure that it will receive some benefit from winning the auction. Thus, there may be some incentive for well-situated potential bidders to lobby for the government to conduct an auction. Nonetheless, bidders are likely to be less heterogeneous with respect to their valuations at auction than with respect to their expected success in prospecting. A slight advantage in a prospecting race can lead to a large difference in the probability of victory, but only a small difference in the amounts that the participants would bid if the right were auctioned.
135. See generally MANCUR OLSON, THE LOGIC OF COLLECTIVE ACTION: PUBLIC GOODS AND THE THEORY OF GROUPS (1965) (noting the difficulty of organizing and mobilizing diffuse groups, and thus the difficulty these groups have in influencing the legislative process).
136. A significant reason for a legislature to enact an auction scheme is because auctions are likely to generate revenue, and thus substitute for taxes or reduced spending in helping to balance budgets. See, e.g., Ellen P. Goodman, Digital Television and the Allure of Auctions: The Birth and Stillbirth of DTV Legislation, 49 FED. COMM. L.J. 517, 526 (1997) (noting the billions of dollars the government received from spectrum auctions and observing that "the temptation to wring auction revenue from the broadcast spectrum was intense"). Non-cash auctions are less likely to be politically appealing than cash auctions.
137. See infra Part III.
138. See supra Part I.C.
139. Duffy, supra note 20, at 469-72 (assuming "N identical firms," where N is exogenous to the model).
140. Id. The provision of early patenting does introduce the danger of inefficient delay by the patentee in developing the invention.
141. See supra Part I.A. But, at least under some assumptions, the costs of this delay will be considerably less than the benefits of limiting duplicative effort.
142. See Abramowicz, supra note 50, at 1084-86 (elaborating on the Duffy model by allowing the appropriability of an invention to vary in a context in which two firms are competing for a patent).
143. An explicit auction system, however, could exacerbate the risk that inventions will not be commercialized. Suppose, for example, that the government auctions a twenty-year patent term in an emerging technological field, but an inventor is unlikely to be able to develop an invention until at least ten years of development. If the auction system is exclusive and replaces the patent system, then the inventor will have less of a patent term than the inventor otherwise would have had, and therefore not only might not commercialize the invention, but also might not have sufficient incentive to develop the invention in the first place. A nonexclusive auction system that complements the patent system will not produce this risk.
144. Suppose, for example, that a patent is expected to produce $1000 per year in profit and can be commercialized after ten years. Assume a 5% discount rate. With a twenty-year term, the patent in present value terms will be worth $4978. With a thirty-year term, the patent will be worth $9909, almost double.
145. See supra Part I.C Indeed, in a model offered by Duffy, longer nominal patent terms can lead inventions to be placed in the public domain earlier, because the indirect effect of the longer patent term on date of invention can more than compensate for the direct effect of term length.
146. See Duffy, supra note 86, at 4-5.
147. For arguments that a "patent thicket" or anticommons problem may slow innovation even in the existing patent system, see Michael A. Heller & Rebecca S. Eisenberg, Can Patents Deter Innovation? The Anticommons in Biomedical Research, 280 SCIENCE 698 (1998),
148. and Arti K. Rai & Rebecca S. Eisenberg, Bayh-Dole Reform and the Progress of Biomedicine, 66 LAW & CONTEMP. PROBS. 289 (2003).
149. The Supreme Court defines a "pioneer invention" as "covering a function never before performed, a wholly novel device, or one of such novelty and importance as to mark a distinct step in the progress of the art." Westinghouse v. Boyden Power Brake Co., 170 U.S. 537, 561-62 (1898).
150. See, e.g., John R. Thomas, The Question Concerning Patent Law and Pioneer Inventions, 10 HIGH TECH. L.J. 35, 37 (1995).
151. See, e.g., Dan L. Burk & Mark A. Lemley, Policy Levers in Patent Law, 89 VA. L. REV. 1575, 1656 (2003) ("The pioneer patent rule has not been invoked by the Federal Circuit in recent years, leading some to consider it moribund, but it provides at least one factor to consider in deciding how broadly to apply the doctrine of equivalents." (footnotes omitted)).
152. See, e.g., O'Reilly v. Morse, 56 U.S. (15 How.) 62, 86 (1854) (invalidating one patent claim, purporting to cover the use of
153. see Merges & Nelson, supra note 11, at 845-51.
154. This does not mean that there would not be boundary questions. For example, if the right to all antidepressant drugs were auctioned, then the question would arise as to how a drug with some other primary purpose but also some antidepressant effects should be treated. Conceivably, the holder of the antidepressant patent might have a blocking right to these drugs, or a compulsory license regime might allocate profits in cases in which rights overlap.
155. Although patent law includes a utility requirement, see 35 U.S.C. §101 (2000), the patent office and the courts have interpreted the requirement loosely, considering scientific plausibility rather than efficacy.
156. See, e.g., Nathan Machin, Prospective Utility: A New Interpretation of the Utility Requirement of Section 101 of the Patent Act, 87 CAL. L. REV. 421, 426-27 (1999).
157. See infra Part II.A.2.b.
158. See supra Part I.A.
159. Patentees sometimes may give away secrets through publication and obtain narrower patents because they want to encourage improvers to improve inventions without worrying about transaction costs. See Oren Bar-Gill & Gideon Parchomovsky, The Value of Giving Away Secrets, 89 VA. L. REV. 1857 (2003) (offering a theoretical and empirical case for this theory).
160. See sources cited supra notes 11-12.
161. Duffy, supra note 20, at 470-72;
162. supra text accompanying note 62.
163. The product differentiation literature shows that excessive entry can occur in competitive markets because each entrant fails to take into account the extent to which its product's sales come at the expense of products that already exist. See, e.g., Avinash K. Dixit & Joseph E. Stiglitz, Monopolistic Competition and Optimum Product Diversity, 67 AM. ECON. REV. 297 (1977) (developing a model in which excess or insufficient entry can occur);
164. Steven C. Salop, Monopolistic Competition with Outside Goods, 10 BELL J. ECON. 141, 142-45 (1979) (developing a model in which excess entry occurs). To the extent that monopoly rights may decrease the number of new inventions, they counterintuitively may enhance efficiency.
165. See, e.g., Wu, supra note 12 (offering an extended version of this argument).
166. We cannot learn much empirically about how patent auctions would work by considering the behavior of firms that have already profited from broad patents. See, e.g., Merges & Nelson, supra note 11, at 884-908 (providing specific examples of firms that slowed the pace of innovation after receiving patents). Inventors who have already covered research expenses many times over might be less efficient than inventors who will need to produce phenomenal results to avoid the costs of an auction.
167. See infra Part III.A.
168. See Bar-Gill & Parchomovsky, supra note 110.
169. See supra note 28 and accompanying text;
170. supra notes 51-52 and accompanying text.
171. See infra Part III.
172. Genevieve Oger, French Firm Hopes to Get PR Bounce out of Nanotubes in Tennis Rackets, SMALL TIMES, NOV. 7, 2001, http://www.smalltimes.com/document_display.cfm?document_id=2506 (discussing the Babolat tennis racket).
173. See, e.g., Nanobot.info, Report on Nanobots (Nanotechnology Robots), http://www.nanobot.info (trumpeting the possibility that "nanomedibots" will be able to interact with cells in the treatment of disease);
174. see also Mark A. Lemley, Patenting Nanotechnology, 58 STAN. L. REV. 601, 602 (2005). Lemley notes that patents in nanotechnology have been issuing at a rapid pace for a new technology with few immediate applications.
175. Id. at 604-05. This suggests that the patent system may well be contributing to incentives to innovate, but these incentives may still be suboptimal.
176. Cf. Lemley, supra note 121, at 628 ("Nanotech inventions will require substantial investment that will not be recouped for a long time, if ever.").
177. Another example is anti-aging research. Aubrey de Grey argues that such research might result in dramatic life extension for mice within 20 years, and for humans within 15 to 100 years after that. See Aubrey de Grey, Timeframe for Progress in Life Extension, http://www.sens.org/time.htm. Perhaps this is wildly optimistic. For our purposes, the key point is that even if scientists and entrepreneurs thought that the goal of eliminating aging might be achievable in fifty years, relatively little private capital would be aimed directly at this goal because of the short patent term. The field thus advances only to the extent that it is funded through basic research or as a result of spillovers from other medical research. Auctioning a very long patent on the envisioned anti-aging technologies could do little harm if the technologies seem unlikely to develop in the absence of such a patent anyway.
178. See supra note 101 and accompanying text. Lemley suggests that the patent thicket may be a greater problem in nanotechnology than in earlier emerging technologies because of the greater number of patents, some overlapping, on building block inventions.
179. See Lemley, supra note 121, at 618-21.
180. Some have argued that patents should be granted only on inventions that would not have been developed in the absence of patent law. See, e.g., A. Samuel Oddi, Beyond Obviousness: Invention Protection in the Twenty-First Century, 38 AM. U. L. REV. 1097, 1101 (1989). A problem with such proposals is that it is difficult to determine the extent to which patent law has accelerated an invention.
181. See generally LANDES & POSNER, supra note 85, at 306-08 (considering whether there is adequate funding for basic research and whether patent law should protect it).
182. See generally Thomas O. McGarity, Peer Review in Awarding Federal Grants in the Arts and Sciences, 9 HIGH TECH. L.J. 1 (1994) (critiquing the peer review system and considering the possibility of radical alternatives to peer review).
183. When a patentee through amendment narrows a patent application to meet the requirements of the Patent Act, the doctrine of prosecution history estoppel will limit the patentee's ability to claim infringement under the doctrine of equivalents. See Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co., 535 U.S. 722 (2002) (clarifying the prosecution history estoppel doctrine, and noting that amendment by itself does not conclusively bar application of the doctrine of equivalents).
184. Aditya Bamzai, Comment, The Wasteful Duplication Thesis in Natural Monopoly Regulation, 71 U. CHI. L. REV. 1525, 1546 (2004).
185. Id.
186. The patent scope auctions described below, however, avoid this problem. See infra Part IV.B.
187. See supra text accompanying notes 106-07.
188. See supra text accompanying note 111.
189. At least until Duffy's analysis, the modern scholarly consensus was that inventive rivalry accelerates innovation. See supra note 11 (noting arguments by Merges and Nelson, as well as by Lemley). This would seem to suggest that auctions, by dampening rivalry, would make things worse. Duffy's model, however, calls into question these analyses.
190. See Duffy, supra note 20. Once a particular technology exists, that technology is likely to be improved faster when it is in the public domain than when a single patent owner has exclusive rights to it. But exclusive rights may lead to faster development of a technology in the first instance by reducing the probability of failure, so auctions potentially can accelerate technological progress. The ultimate question in the auction context is whether auctioning prospects would increase useful inventions.
191. See supra Part II.A.2.a.
192. A significant caveat to Duffy's theory is that implicit auctions cannot work when unanticipated exogenous shocks suddenly make attractive an invention that previously seemed unattractive. See infra note 159 and accompanying text. The empirical question is whether the inventive process is ordinarily smooth and predictable, or whether possible inventions often suddenly become apparent, triggering destructive races.
193. See, e.g., Wu, supra note 12 (arguing that centralized decisionmaking about invention might be less efficient than decentralized decisionmaking).
194. For example, one might be more willing to race for a patent if the remainder of the field in the future will be auctioned, because the winner of the race may face fewer substitutes than in a world in which racing begins anew.
195. The sale, however, might be impeded by Arrow's disclosure paradox - that it is difficult to sell information without disclosing it or to disclose information without giving it away. See Kenneth J. Arrow, Economic Welfare and the Allocation of Resources for Invention, in THE RATE AND DIRECTION OF INVENTIVE ACTIVITY: ECONOMIC AND SOCIAL FACTORS 609 (1962).
196. This could be accomplished only if the auctioned right is for an improvement patent only. See supra note 89 and accompanying text.
197. A drawback is that an auction system might produce less disclosure of the results of invention. Some disclosure would still occur, however, to the extent that patents would be allowed within the scope of the patent grant. There is considerable doubt about how much of a benefit mandated disclosure provides. See Note, The Disclosure Function of the Patent System (or Lack Thereof), 118 HARV. L. REV. 2007 (2005) (arguing that the patent system is failing in the goal of disseminating information). If mandated disclosure is considered a significant benefit, the law might seek to require an auction winner to reveal the results of subsequent research within the patent term.
198. See Duffy, supra note 20, at 488-89 (showing that the owner of a blocking right has significantly greater incentive to conduct research within the scope of that right than third parties).
199. Rent seeking is an activity in which the competition for rents, for example from an exclusive government franchise, dissipates the benefits of those rents. See Anne O. Krueger, The Political Economy of the Rent-Seeking Society, 64 AM. ECON. REV. 291 (1974) (introducing the term);
200. Gordon Tullock, The Welfare Costs of Tariffs, Monopolies, and Theft, 5 W. ECON. J. 224 (1967) (offering an additional seminal treatment).
201. There may, however, be incentives for an inventor with a chance to win a patent race to lobby against an auction. A system of patent auctions thus might produce costs associated with rent opposition. See Robert D. Cairns & Ngo Van Long, On Opposition to Rent Seeking: Implications for Developing Countries, 2 REV. DEV. ECON. 282 (1998) (noting that the full costs of rent-seeking activity must include the costs of those opposing the granting of rents).
202. For an overview, see INTELLECTUAL PROPERTY AND INTERNATIONAL TRADE: THE TRIPS AGREEMENT (Carlos M. Correa & Abdulqawi A. Yusuf eds., 1998).
203. DAVID MAMET, THE WATER ENGINE: AN AMERICAN FABLE, AND MR. HAPPINESS (1978). For a plot summary of the televised adaptation, see http://www.imdb.com/title/tt0105788.
204. See, e.g., Kurt M. Saunders & Linda Levine, Better, Faster, Cheaper - Later: What Happens When Technologies Are Suppressed, 11 MICH. TELECOMM. & TECH. L. REV. 23 (2004), available at http://www.mttlr.org/voleleven/saunders.pdf.
205. See infra Part III.A.
206. See supra note 16.
207. The product differentiation literature relatedly shows that because of the fixed costs associated with entering a market, excessive entry from a social perspective can occur. See supra note 113 and sources cited therein.
208. For the classic argument that a target firm's management should be passive in the face of a hostile tender offer, see Frank H. Easterbrook & Daniel R. Fischel, The Proper Role of a Target's Management in Responding to a Tender Offer, 94 HARV. L. REV. 1161 (1981).
209. See, e.g., Ivar M. Kaardal, The American Inventors Protection Act, the Independent Inventor's Interest & Consumer Choice in the Market, 84 J. PAT. & TRADEMARK OFF. SOC'Y 503 (2002).
210. See supra note 89 and accompanying text.
211. See supra notes 11-16 and accompanying text (discussing the issue without taking a position on it).
212. See supra Part I.C.
213. But see Abramowicz, supra note 50, at 1091 (noting that a patentee might obtain a patent for its option value when the probability that commercialization will be worthwhile is low).
214. Arguably, it is desirable to provide a single party with an incentive to argue that research and development spending fell short of the promised amount. For example, the government could hold an auction for the right to sue the winning bidder for spending too little. The revenues from this auction would go to the winning bidder of the R&D auction, but a successful suit would result in the funds not spent being allocated to the party that wins the right to sue. In litigation, courts would need to evaluate the degree to which spending really was within the research field.
215. For a discussion of some of the difficulties in patent valuation, see F. Russell Denton & Paul J. Heald, Random Walks, Non-Cooperative Games, and the Complex Mathematics of Patent Pricing, 55 RUTGERS L. REV. 1175 (2003). One possible approach is to use a self-assessment mechanism. Third parties could be invited to make bonded offers to purchase the patents, with the highest offeror receiving a reward (some small percentage of the highest bid); this reward, combined with the possibility that the patentee might accept an offer, give incentives for honest third-party valuation.
216. See Saul Levmore, Self-Assessed Valuation Systems for Tort and Other Law, 68 VA. L. REV. 771, 783-88 (1982) (devising a similar competitive assessment system for valuing real property).
217. I am grateful to Sam Dinkin for this suggestion.
218. Duffy, supra note 20, at 505.
219. Petition for a Writ of Certiorari, KSR Int'l Co. v. Teleflex Inc., 127 S. Ct. 1727 (2005) (No. 04-1350), 2005 WL 835463.
220. Id. at *28-29 ("The Federal Circuit 'test' ignores that exogenous changes - i.e., economic, regulatory or technological changes not attributable to the work of the alleged inventor - can create new possibilities that can be exploited, or new needs that can be satisfied, with technological trivial combinations of existing technology."). The Federal Circuit had generally insisted that a combination invention would be obvious only when a motivation, suggestion, or teaching for that combination existed in the prior art.
221. See, e.g., In re Kotzab, 217 F.3d 1365, 1370 (Fed. Cir. 2000) (holding that "to establish obviousness based on a combination of the elements disclosed in the prior art, there must be some motivation, suggestion or teaching of the desirability of making the specific combination that was made by the applicant").
222. 127 S. Ct. 1727 (2007).
223. At least three difficult questions about the design of such an auction would remain. First, would the winning bidder be required to perfect the intellectual property right by subsequently filing a patent detailing the combination? An argument against such a requirement is that it would needlessly increase transaction costs. An argument for such a requirement is that it would ensure appropriate disclosure of the best means of achieving the combination. Second, would the winning bidder still be required to show that the combination is nonobvious? This might be difficult to establish, because the mere request for the auction could count as prior art. The danger of not requiring nonobviousness is that the explicit auctions might unnecessarily create patents of at least some finite length, given the requirement that a cash payment accompany the winning bid, leading to deadweight loss. The occurrence of an explicit least-patent-term auction, however, generally makes concerns about obvious inventions less salient, because obvious inventions that require little development will receive relatively short patent terms, and when inventions are obvious but second-mover advantages exist, some patent protection may be optimal. Third, would the auctioned right supersede the rights of inventors who have pending patent applications or who have invented but have not yet filed patent applications? A particular concern is that the auction system might lead inventors to hold off patenting until they have also patented combinations of the original invention and other inventions, lest someone else obtain rights to the combination.
224. The cash payment to the government, however, may reduce the need for other taxes, which themselves may cause economic distortions. See, e.g., Charles L. Ballard & Don Fullerton, Distortionary Taxes and the Provision of Public Goods, J. ECON. PERSP., Summer 1992, at 117 (discussing the economic distortion associated with taxation). Nonetheless, this distortion is generally thought to be less than the distortionary effect of monopoly pricing created by patents.
225. See Abramowicz, supra note 19, at 201-02.
226. See, e.g., F.P. Ramsey, A Contribution to the Theory of Taxation, 37 ECON. J. 47 (1927).
227. Ayres & Klemperer, supra note 67, at 989.
228. Id.
229. Id.
230. Id. at 990, 994-1000.
231. See id. at 990-93 (noting that this is an implication of Ramsey, supra note 165, applied intertemporally).
232. Ayres and Klemperer acknowledge this point implicitly, noting that a patent of infinite duration might not be possible as a "pragmatic or political matter." Id. at 1011.
233. Duffy, supra note 20, at 478.
234. Id. at 478 n.125
235. (citing Shlomo Yitzhaki, A Note on Optimal Taxation and Administrative Costs, 69 AM. ECON. REV. 475, 475, 480 (1979), as presenting an analogous optimal taxation problem).
236. For a careful analysis of perpetual patent terms that rejects the concern about the anticommons, see F. Scott Kieff, Coordination, Property & Intellectual Property: An Unconventional Approach to Anticompetitive Effects and Downstream Access 64-66 (Stanford Law Sch. John M. Olin Program in Law & Econ., Working Paper No. 323, 2006), available at http://ssrn.com/abstract=910656.
237. See supra Part II. A. 1.
238. See supra note 101 and accompanying text.
239. Such activities may become more attractive over time for the same reason that invention may become more attractive over time in Barzel's model - that the demand for an invention may increase over time. See Barzel, supra note 30, at 349.
240. See supra Part I.B.
241. For an analysis of the relevance of price discrimination to copyright policy, see Michael J. Meurer, Copyright Law and Price Discrimination, 23 CARDOZO L. REV. 55 (2001).
242. For an empirical analysis purporting to show that compulsory licenses do not deter innovation, see Colleen Chien, Cheap Drugs at What Price to Innovation: Does the Compulsory Licensing of Pharmaceuticals Hurt Innovation?, 18 BERKELEY TECH. L.J. 853 (2003). This may be an area, however, in which economic theory is more useful than tentative empirical results. If compulsory licenses lower the return to innovation, there will be less innovation. Even if compulsory licensing does not lower the return to innovation overall, it may change the distribution of innovation, and the case for compulsory licenses then depends on whether the government does better than the market in determining which innovations should be pursued.
243. There are two disadvantages of eliminating injunctive relief. First, some infringers might be judgment-proof. See Developments in the Law - Injunctions, 78 HARV. L. REV. 994, 1001-02 (1965) (arguing that judgment-proof defendants may provide one justification for injunctive relief). An antidote would be to require alleged infringers to obtain bonds ensuring that they will be able to obtain damages. Second, in a liability rule regime, it may be more advantageous to infringe than to negotiate whenever the infringer expects that the courts will underestimate damages or that the patentee will not be able to afford litigation.
244. Cf. Thomas F. Cotter, An Economic Analysis of Enhanced Damages and Attorney's Fees for Willful Patent Infringement, 14 FED. CIR. B.J. 291, 308-10 (2004) (noting several reasons that patentees may fail to enforce their rights). A patent regime without injunctive relief, however, might compensate for this by imposing a small damages multiplier fixed in advance, which would then be discounted in an explicit auction.
245. Ayres & Klemperer, supra note 67, at 1028-31.
246. A variant on this approach would be to award full damages but only with some positive probability, and the winning bidder would be the party that offers the lowest probability. The idea of "probabilistic patents" in which enforcement occurs with less than certainty is Ayres and Klemperer's primary proposal. Id. at 993-1013.
247. Conceivably, it might be more efficient than a regime of reduced damages, if it reduced litigation cost. If that is so, however, the litigation system in general should provide for only probabilistic enforcement of legal rights, with damages set at a multiplier of the usual damages level in the cases in which the rights are enforced. See, e.g., David Rosenberg & Steven Shavell, A Simple Proposal to Halve Litigation Costs, 91 VA. L. REV. 1721 (2005) (offering such a proposal).
248. A much more ambitious approach might be to use information markets to decide when auctions should occur and to identify the bidder most likely to maximize social benefits. See Michael Abramowicz, Information Markets, Administrative Decisionmaking, and Predictive Cost-Benefit Analysis, 71 U. CHI. L. REV. 933 (2004) (providing an overview of the emerging literature on information markets).
249. See supra notes 5-6 and accompanying text.
250. See supra Part II.A.1.b.
251. The assumption of no spillovers does not affect the case for the ultimate proposal. See infra text accompanying note 189.
252. Ayres and Klemperer suggest the possibility of such a "duopoly auction" as a means of reducing patent pricing and thus deadweight loss. See Ayres & Klemperer, supra note 67, at 1031-32.
253. See supra note 91 (discussing the mechanics and strategy of a second-price auction).
254. See supra note 38 and accompanying text.
255. See Abramowicz, supra note 50, at 1112-19 (introducing a proposal for patent term extension auctions).
256. Id. at 1111 (noting the undesirability of a policy lever that must be exercised at the beginning of the patent term).
257. Id. at 1115 ("It also assures that the difference between the original patentee's valuation and the third party's valuation derives from the benefit of development of the patent within the original patent term and not from some other factor that makes the patent particularly valuable to the patentee.").
258. Harold Demsetz, Toward a Theory of Property Rights, 57 AM. ECON. REV. (PAPERS & PROC.) 347, 350 (1967) ("Increased internalization, in the main, results from changes in economic values, changes which stem from the development of new technology and the opening of new markets, changes to which old property rights are poorly attuned.")
259. A principal criticism of the Demsetz approach is that it provides no means of aggregating quality and price. Richard Posner proposed that quality might be assessed in combination with price by initially allowing potential suppliers to solicit contracts with consumers and then awarding the contract to the producer who is able to obtain the highest dollar volume of commitments. See Richard A. Posner, The Appropriate Scope of Regulation in the Cable Television Industry, 3 BELL J. ECON. & MGMT. SCI. 98, 115 (1972) (suggesting that this might alleviate the difficulty of determining the bid offering the best combination of price and quality). Among the potential challenges of such a scheme is that preventing monopolists from reneging on vague quality commitments may be difficult.
260. See Oliver E. Williamson, Franchise Bidding for Natural Monopolies - in General and with Respect to CATV, 7 BELL J. ECON. 73, 80-82 (1976) (discussing the difficulties of ascertaining quality in a Demsetzian auction).
261. See, e.g., Robert P. Merges, Intellectual Property Rights and the New Institutional Economics, 53 VAND. L. REV. 1857, 1868 (2000) (noting that Demsetz "assumed that changed conditions would automatically translate into revised property rights," without considering the political process that would effect this translation).
262. Demsetz suggested that changes in property rights "have arisen in Western societies largely as a result of gradual changes in social mores and in common law precedents." Demsetz, supra note 193, at 350. This does not provide a direct explanation of how institutions for distributing property rights evolve.
263. Even if an auction is efficient, we should not necessarily expect a change from a prospecting system to a Demsetzian auction. See supra text accompanying notes 91-93;
264. see also David D. Haddock & Lynne Kiesling, The Black Death and Property Rights, 31 J. LEGAL STUD. S545, S564-65 & S565 n.51 (2002) (noting that Demsetz's analysis leaves the possibility that because of rent seeking, it will not always be the case that resources will be privatized once the benefits exceed the costs).