Do patent holdup and royalty stacking lead to systematically excessive royalties?

Journal of Competition Law and Economics

Volumn 4, Issue 3, 2008, Pages 535-570

  Elhauge, Einer ^a  

^a HARVARD LAW SCHOOL   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 54049104401     PISSN: 17446414     EISSN: 17446422     Source Type: Journal    
DOI: 10.1093/joclec/nhn027     Document Type: Article

References (139)

1. Mark Lemley & Carl Shapiro, Patent Holdup and Royalty Stacking, 85 TEX. L. REV. 1991 (2007)
2. Carl Shapiro, Injunctions, Hold-Up, and Patent Royalties (August 2006), http://faculty.haas.berkeley.edu/shapiro/royalties.pdf.
3. Lemley & Shapiro, supra note 1, at 1991, 2000-02, 2008, 2044
4. Shapiro, supra note 1, at 1, 12-17, 27-31.
5. Lemley & Shapiro, supra note 1, at 1991, 2003-05
6. Shapiro, supra note 1, at 21-22.
7. Lemley & Shapiro, supra note 1, at 2008.
8. Id. at 2046-48.
9. Lemley & Shapiro, supra note 1, at 1994, 2021-22
10. Shapiro, supra note 1, at 23-24, 32-33.
11. Lemley & Shapiro, supra note 1, at 1994, 2035-44
12. Shapiro, supra note 1, at 18-21
13. Mark Lemley, Ten Things to Do About Patent Holdup of Standards (And One Not To), 48 B.C. L. REV. 149, 153-54, 161-62, 166-67 (2007)
14. Carl Shapiro, Patent Reform: Aligning Reward and Contribution, in INNOVATION POLICY AND THE ECONOMY, 112-113 & nn. 5, 8, 120-22, 138-39 (Jaffe, Lerner & Sterns, eds. 2008).
15. See Robert Hahn & Hal Singer, Assessing Bias in Patent Infringement Cases: A Review of International Trade Commission Decisions 21 Harv. J.L. Tech. 457, 467 (2008)
16. Mark A. Lemley & Philip J. Weiser, Should Property or Liability Rules Govern Information?, 85 TEX. L. REV. 783, 795-800, 834 (2007).
17. eBay Inc. v. MercExchange, LLC, 547 U.S. 388, 396-97 (2006) (J. Kennedy, joined by Stevens, Souter & J.J. Breyer, concurring).
18. See John M. Golden, Commentary, "Patent Trolls" and Patent Remedies, 85 TEX. L. REV. 2111, 2113, note 12 (2007).
19. Shapiro, supra note 1, at 25
20. Lemley & Shapiro, supra note 1, at 2005, 2008.
21. Lemley & Shapiro, supra note 1, at 1996-2000
22. Shapiro, supra note 1, at 6-9.
23. Lemley and Shapiro actually use capital V and B, whereas the Shapiro article uses small v and the Greek letter β. Throughout this paper, where there are these differences in notation, I will use the notation used in the Shapiro article, which forms the mathematical basis for the conclusions in the Lemley and Shapiro piece.
24. Lemley & Shapiro, supra note 1, at 1995
25. Shapiro, supra note 1, at 11.
26. Shapiro, supra note 1, at 6-7, 13-14.
27. Lemley & Shapiro, supra note 1, at 2000-01
28. Shapiro, supra note 1, at 10, 13-14.
29. Shapiro, supra note 1, at 13-14, 31.
30. See also Lemley & Shapiro, supra note 1, at 2001 (using the same formula to calculate the percentage overcharge, where C is set to F/vX).
31. Shapiro, supra note 1, at 14, 31.
32. See also Lemley & Shapiro, supra note 1, at 2002 (using the same formula to calculate the percentage overcharge).
33. Shapiro, supra note 1, at 7-8.
34. Lemley & Shapiro, supra note 1, at 2004-05
35. Shapiro, supra note 1, at 21-22.
36. Shapiro, supra note 1, at 21-22.
37. For θ> θ*μ - βθ* Shapiro concludes the overcompensation is greater for the weak patents than for the strong patents. Shapiro, supra note 1, at 15.
38. Shapiro, supra note 1, at 1
39. Lemley & Shapiro, supra note, at 2044.
40. Lemley & Shapiro, supra note 1, at 2021-22
41. Shapiro, supra note 1, at 23-24, 32-33.
42. See Shapiro, supra note 1, at 33.
43. Id. at 9. Golden correctly criticizes Lemley and Shapiro's failure to justify their benchmark.
44. See Golden, supra note 10, at 2115-16, 2137-40. But this article goes beyond such critique to explain why their benchmark is too high and what the correct benchmark would be.
45. Joseph Farrell & Carl Shapiro, How Strong Are Weak Patents? 4, 8-10 (January 2007), forthcoming in AM. ECON. REV., http://faculty.haas.berkeley.edu/shapiro/weak.pdf.
46. In addition to the article co-authored by Professor Farrell in the last footnote, see Joseph Farrell & Robert P. Merges, Incentives to Challenge and Defend Patents, 19 BERKELEY TECH. L.J. 943, 958 (2004).
47. See Lemley & Shapiro, supra note 1, at 1999-2000
48. see also Mark A. Lemley & Carl Shapiro, Reply: Patent Holdup and Royalty Stacking, 85 TEX. L. REV. 2163, 2164-65 (2007) (hereinafter Lemley & Shapiro, Reply to Golden).
49. See Farrell & Shapiro, supra note 26. Below I explain some reasons to doubt this conclusion.
50. Nor does it work to focus on what the law provides without injunctive remedies, because the holdup power of the downstream monopolist is enforced by injunctive rights to prevent others from invading its own monopoly facility.
51. Lemley & Shapiro, Reply to Golden, supra note 28, at 2164.
52. Id. at 2166-67
53. Shapiro, Patent Reform, supra note 7, at 114-26.
54. Lemley & Shapiro, Reply to Golden, supra note 28, at 2166-67
55. Shapiro, Patent Reform, supra note 7, at 122-25.
56. Shapiro, supra note 1, at 7
57. Lemley & Shapiro, supra note 1, at 1996.
58. Shapiro, Patent Reform, supra note 7, at 122-25.
59. Lemley & Shapiro, Reply to Golden, supra note 28, at 2166-67
60. Shapiro, Patent Reform, supra note 7, at 115-17.
61. See Golden, supra note 10, at 2136.
62. Dasgupta & Stiglitz, Uncertainty, Industrial Structure, and the Speed of R&D, 11 BELL J. ECON. 1, 18 (1980)
63. Tandon, Rivalry and the Excessive Allocation of Resources to Research 14 BELL J. ECON. 152, 156-57 (1983) (patent races might result in underinvestment or overinvestment in research, depending on the ratio of patent reward to social value)
64. SUZANNE SCOTCHMER, INNOVATION AND INCENTIVES 112, note 7 (MIT Press 2005) (summarizing the economic literature as concluding that: "Patent races may lead to deficient incentives or excessive incentives")
65. Id. at 102 (using the Tandon model to conclude that the "Optimal number of participants [in research] can be either larger or smaller than the equilibrium of the race, ..., depending on the patent value" and how close it comes to the full social value).
66. Dasgupta & Stiglitz, supra note 38, at 21.
67. Id. at 26.
68. Id. at 21.
69. Id. at 19 (emphasis in original).
70. See Shapiro, supra note 1, at 6 ("We are interested in situations in which m [the profit margin] is relatively large in comparison with v [the value of the patented feature]").
71. Such a view seems suggested in other work by Shapiro. See Farrell, Hayes, Shapiro & Sullivan, Standard Setting, Patents, and Hold-Up 17 (December 2007), forthcoming ANTITRUST L.J., http://faculty.haas.berkeley.edu/shapiro/standards2007.pdf.
72. See 35 U.S.C. §§ 101, 103(a).
73. See Lemley & Shapiro, supra note 1, at 1999-2000.
74. Shapiro, supra note 1, at 1
75. Lemley & Shapiro, supra note 1, at 2044.
76. Shapiro, supra note 1, at 7.
77. Id. at 6. The odds of patent validity under these assumptions cannot be inferred from the empirical odds at which current patent claims are held valid and infringed because current patents involve many patents for which this strong assumption is untrue. Nor can one simply drop those assumptions because they are integral to establishing the patent holder's bargaining power under their model.
78. Profits would have the appropriate discounts to the extent the first set of profits comes in earlier. But even if there are serial negotiations, the profits may all come in the same time when the product is launched, in which case such discounting is unnecessary.
79. Avoiding the patented feature will require either designing the product not to include the patented feature or (in the case of a surprise patent) ceasing to sell the product pending redesign.
80. See Lemley & Shapiro, Reply to Golden, supra note 28, at 2170.
81. See Golden, supra note 10, at 2133.
82. Id. at 2132.
83. This is true even for litigation costs, because although both sides can equally estimate the costs of litigating patent validity, the user can better estimate the costs of litigating its infringement.
84. Lemley & Shapiro, Reply to Golden, supra note 28, at 2170.
85. JOSEPH HENRICH, ROBERT BOYD, SAMUEL BOWLES & COLIN CAMERER, FOUNDATIONS OF HUMAN SOCIALITY: ECONOMIC EXPERIMENTS AND ETHNOGRAPHIC EVIDENCE FROM FIFTEEN SMALL-SCALE SOCIETIES(Oxford University Press 2004)
86. Hessel Oosterbeek, Randolph Sloof & Gijs van de Kuilen, Cultural Differences in Ultimatum Game Experiments: Evidence from a Meta-Analysis 7 EXPERIMENTAL ECONOMICS 171 (2004).
87. See Bjorn Wallace, et al., Heritability of Ultimatum Game Responder Behavior, Proceedings of the National Academy of Sciences, vol. 104, no. 40, pp. 15631-34 (2 October 2007) (identical twin study showing that genes explain 40 percent of the variation of ultimatum game responder behavior).
88. Shapiro, supra note 1, at 6-7.
89. See EINER ELHAUGE, UNITED STATES ANTITRUST LAW AND ECONOMICS 189 (Foundation Press 2008) (explaining why "even an absolute monopolist would never set a price that leaves it on the inelastic portion of its demand curve").
90. Lemley & Shapiro, supra note 1, at 2046.
91. Lemley & Shapiro, supra note 1, at 2021-22
92. Shapiro, supra note 1, at 23-24, 32-33.
93. If we take into account the other reasons to think that the Lemley-Shapiro model overstates predicted royalties, like the fact that the predicted royalties are lower if D engages in repeat bargaining with patent holders, if there is informational asymmetry favoring D, if norms allow credible commitments not to pay excessive royalties, or if demand is not constant, then basing damages on past negotiated royalties would be even more undercompensatory.
94. Lemley & Shapiro, supra note 1, at 2000, 2021
95. Shapiro, supra note 1, at 23.
96. Shapiro, supra note 1, at 23-24, 32-33
97. Lemley & Shapiro, supra note 1, at 2022, note 79.
98. Shapiro, supra note 1, at 33.
99. Id. at 22
100. Lemley & Shapiro, supra note 1, at 2004-05.
101. See Henry Smith, Property and Property Rules, 79 N.Y.U. L. REV. 1719, 1729 (2004) (arguing that property rules are superior to liability rules because they reduce opportunism that results when owners cannot communicate to others the true value of the property).
102. See Louis Kaplow & Steven Shavell, Property Rules v. Liability Rules: An Economic Analysis, 109 HARV. L. REV. 713, 731 (1996)
103. Richard Epstein, A Clear View of the Cathedral: The Dominance of Property Rules, 106 YALE L.J. 2091, 2093 (1997).
104. Kaplow and Shavell analyze a similar problem in the property versus liability rules discussion and reach the contrary conclusion that it will lead to inefficient takings only if the property owner has some idiosyncratic value that generally makes takings undesirable. See Kaplow & Shavell, supra note 69, at 759-63.
105. Without such an idiosyncratic valuation, in their model any taking at underestimated damages would not be undesirable because the taker would gain the same value that the owner would lose. In my model, what makes takings at underestimated damages undesirable is not the idiosyncratic value the owner puts on its patent, but rather the resulting reduction in ex ante incentives to create the invention at all, which Kaplow and Shavell do not model. See generally Lucian Bebcuk, Property Rights and Liability Rules: The Ex Ante View of the Cathedral, 100 MICH. L. REV. 601 (2001) (observing that the property versus liability rules literature has only considered ex post effects, and thus wrongly failed to recognize that ex post distributive effects can reduce efficiency by altering ex ante incentives).
106. The unbalanced effect of balanced errors that vary for different industries or doctrinal areas does not appear to have been noticed in prior articles on property versus liability rules.
107. See Kaplow & Shavell, supra note 69, at 739-41 (noting that limited defendant wealth can make damages undercompensatory and inefficient compared with property rules).
108. In a similar vein, some prior commentators have argued that the reason to have property rules is that it creates greater certainty about outcomes. See, e.g., Epstein, supra note 69, at 2094
109. Carol Rose, The Shadow of the Cathedral, 106 YALE L.J. 2175, 2187 (1997)
110. Henry Smith, Exclusionary and Property Rules, 90 VA. L. REV. 965, 984 (2004).
111. See Lemley & Shapiro, supra note 1, at 2036
112. Shapiro, supra note, at 1, 24.
113. See Hahn & Singer, supra note 8, at 467-68.
114. See eBay Inc., 547 U.S. at 396.
115. See Golden, supra note 10, at 2113, note 12.
116. See DENNIS CARLTON & JEFFREY PERLOFF, MODERN INDUSTRIAL ORGANIZATION 195-96 (4th ed. 2005).
117. Shapiro, supra note 1, at 25
118. Lemley & Shapiro, supra note 1, at 2005, 2008.
119. Lemley & Shapiro, supra note 1, at 2005-08.
120. See W. KIP VISCUSI, JOHN M. VERNON & JOSEPH E. HARRINGTON, ECONOMICS OF REGUIATION AND ANTITRUST 836 (The MIT Press, 2nd ed. 1998)
121. Kamien & Tauman, Fees Versus Royalties and the Private Value ofa Patent, 101 Q.J. ECON. 471, 477, 482-83 (1986).
122. See Carlton & Perloff, supra note 78, at 200-34.
123. Lemley & Shapiro, supra note 1, at 2008.
124. See Elhauge, supra note 60, at 209-11.
125. See Farrell & Shapiro, supra note 26
126. Shapiro, Patent Reform, supra note 7, at 119.
127. See Farrell & Shapiro, supra note 26, at 5-6.
128. Id. at 18.
129. Lemley & Shapiro, supra note 1, at 2039, note 153.
130. Lemley & Shapiro, supra note 1, at 2046.
131. Id. at 2046-48.
132. See Lemley & Shapiro, supra note 1, at 2047.
133. We know this because P* - C = 0.5(A + v + C) - C = 0.5(A + v - C) which equals X.
134. See Shapiro, supra note 1, at 30.
135. Avoiding redesign and lost profits during the period of redesign are not joint gains from licensing in advance because those are equally obtainable by agreeing to a post-judgment license, and are taken into account when considering the royalty for such a license.
136. I am indebted to John Golden for this point.
137. We know this because P* - C = r = 0.5(A + v + C + r) - C - r = 0.5(A + v - C - r), which equals X.
138. 0.5.
139. One can solve both the strong and weak surprise patent equations for r, but the resulting quadratics are messy.