University of Pennsylvania Law Review FOUNDED 1852 Formerly American Law Register VOL. 150 JANUARY2002 NO. 3 ARTICLES UNRAVELING THE PATENT-ANTITRUST PARADOX MICHAEL A. CARRIERt INTRODUCTION ..................................................................................... 762 I. THE PROBLEM: A CONFLICT AND NO EASY RECONCILIATION ..... 766 A. The Patent-AntitrustC onflict. ....................................................... 766 1. Different Paths to W elfare .................................................... 766 2. An Example ........................................................................... 771 B. RepresentativeS ection 2 Cases ...................................................... 774 1. X erox ....................................................................................... 775 2. Kodak 1I .......................................7.78.......................................... 3 . In tel. ........................................................................................ 78 1 4. Microsoft. ................................................................................. 784 C. Proposed Solutions Deconstructed. ................................................. 787 t Assistant Professor, Rutgers University School of Law-Camden. B.A. 1991, Yale University; J.D. 1995, University of Michigan. I would like to thank Steve Calkins, Roger Dennis, Stacey Dogan, Jill Fisch, David Frankford, Heather Gerken, Peter Hammer, Tom Kauper, Marina Lao, Mark Lemley, Dennis Patterson, Arti Rai, Pat Ryan, Allan Stein, Lawrence Sung, Steve Teplinsky, and Ed Zimmerman for helpful comments on earlier drafts of this Article. • (761) 762 UNIVERSITY OFPENNSYLVANIA LAWREVIEW [Vol. 150:761 1. Scope of the Patent ............................................................... 788 2. M ultiple Markets ................................................................... 791 3. Monopolist's Intent ............................................................... 793 4. Baxter's "Comparability" Test ............................................... 795 5. Bowman's "Competitive Superiority" Test ........................... 796 6. Kaplow's "Ratio" Test ............................................................ 797 II. THE COMMON DENOMINATOR ....................................................... 800 A. Definition .................................................................................... 802 B. Innovation and the Patent and Antitrust Laws ............................. 803 1. Patent Laws ....................................8.0.3...................................... a. Statutes. ................................................................................ 803 b. Legislative history .................................................................. 805 2. Antitrust Laws ................................................................... 807 a. Statutes/legislativeh istory ...................................................... 807 b. Antitrustj urisprudencea nd economic efficiencies .................... 810 C. Different Paths in Different Industries. .......................................... 815 III. THE TEST ........................................................................................ 816 A. The Presumption. ......................................................................... 816 B. The Rebuttal. ............................................................................... 818 1. ExAnte Incentives ................................................................. 819 a. Nonpatent market-based rewards. ............................................ 821 b. Product creation. ................................................................... 823 c. Ease of imitation. ................................................................... 827 2. Ex Post Perspective ................................................................ 829 3. Operation of the Rebuttal .................................................... 831 C. The Surrebuttal. ........................................................................... 833 D. Applications ................................................................................ 834 1. The Bully Monopolist ........................................................... 835 2. The Biopharm aceutical Patentee ......................................... 836 3. The Internet Auctioneer ....................................................... 837 IV. COUNTERING THE OBJECTIONS ...................................................... 841 A. Application of the Test .................................................................. 841 1. The Rebuttable Presumption ............................................... 841 2. Antitrust Government Enforcement Agencies .................... 844 3. Role of the Courts Under Section 2 ..................................... 846 B. Reduction of Patent Incentives. ..................................................... 848 C. Readjustments to the Patent System ............................................... 850 CONCLUSION ......................................................................................... 853 INTRODUCTION The intersection of the patent and antitrust laws presents a formi- dable paradox. The patent laws increase invention and innovation by offering inventors a right to exclude. The antitrust laws foster compe- 2002] PA TENT-ANTITRUST PARADOX tition, sometimes through the condemnation of such exclusion. As patents become ever more important in our information-based econ- omy, the significance of the conflict between the patent and antitrust laws will only increase. Courts and commentators have struggled with this paradox for generations. They have experimented with an array of disparate tests to determine, for example, when a company's reliance on its patents should immunize it from the antitrust offense of monopolization. Courts have applied rebuttable presumptions, emphasized the "scope" of the patent, examined the defendant's intent, and questioned whether an "essential facility" was denied. These various tests not only have dashed any hopes of predictability but also have failed to wrestle with the fundamental tension between the patent and antitrust laws.' Moreover, the tests are both overinclusive and underinclusive in targeting activity that harms welfare.2 Some of the tests are overinclu- The courts also have ignored the most thoughtful literature in the field, refusing even to acknowledge the tests offered in the seminal approaches to the intersection: WARD S. BOWMAN, JR., PATENT AND ANTITRUST LAW: A LEGAL AND ECONOMIC APPRAISAL, at xii (1973); William F. Baxter, Legal Restrictions on Exploitation of the Patent Monopoly: An Economic Analysis, 76 YALE L.J. 267, 313 (1966); Louis Kaplow, The Patent- Antitrust Intersection: A Reappraisa4 97 HARv. L. REv. 1813, 1820 (1984). For a discus- sion of this literature, see infra Parts I.C.4-.6. This Article will refer to "welfare" to signify "total welfare," or the sum of con- sumer surplus and producer surplus. Stated most simply, producer surplus consists of an aggregation of the differences between the price at which producers would be will- ing to sell the product and the market price. See generally HAL R. VARIAN, INTERMEDIATE MICROECONOMICS: A MODERN APPROACH 256-57 (5th ed. 1999); Peter J. Hammer, Antitrust Beyond Competition: Market Failures, Total Welfare, and the Challenge of Intramarket Second-Best Tradeoffs, 98 MICH. L. REv. 849, 858 (2000). Consumer sur- plus (sometimes referred to as consumer welfare) refers to an aggregation of the dif- ferences between what consumers would be willing to spend for goods and the market price. See generally VARIAN, supra, at 247. The terms are often used imprecisely, in par- ticular as antitrust commentators refer to "consumer welfare" when, because they also are considering producer surplus, they mean "total welfare." See Robert H. Bork, Legis- lative Intent and the Policy of the Sherman Act, 9J.L. & ECON. 7, 7-10 (1966). This Article will refer most generally to the positive effects on welfare (or total welfare) of the anti- trust and patent laws, and will not trace the precise contributions to welfare attribut- able to consumer or producer surplus. Although the antitrust laws conceivably could promote other goals, such as "the political and social values of dispersed control over economic resources, multiple choices for producers and consumers free of the arbitrary dictates of monopolies or cartels, equal opportunity, and 'fairness' in economic dealings," PHILLIP AREEDA & LOUIS KAPLOW, ANTITRUST ANALYSIS: PROBLEMS, TEXT, AND CASES 49 (5th ed. 1997), this Article will focus on the objective of increasing total economic welfare. It concen- trates on this objective because of the difficulties in simultaneously promoting eco- nomic and noneconomic goals, the contribution to noneconomic objectives from the pursuit of economic ends, and the inconsistencies among the various noneconomic 764 UNIVERSITY OF PENNSYLVANIA LA W REVIEW [Vol. 150: 761 sive in condemning conduct that has no adverse effect on welfare. For example, a company's intent to exclude typically reveals nothing more than an unexceptional desire to defeat its competitors. By pe- nalizing the right to exclude, courts engaging in this type of inquiry take direct aim at the modus operandi of the patent laws. At the same time, some of the tests are underinclusive in their blind deference to the patent laws. To state that action within the scope of the patent should automatically be immune from antitrust scrutiny (so the incentives underlying the patent system are not dimin- ished) "solves" the patent-antitrust conflict only by according priority to the patent laws. This purported solution amounts to an assumption that the increase in welfare from safeguarding the patentee's right to exclude will always outweigh the increase that would have resulted from antitrust's enhanced competition. Such an approach ensures that only the patent, rather than the antitrust, path to innovation will be traversed. In short, courts' approaches to the patent-antitrust in- tersection fail to recognize the two independent paths to innovation and fail to articulate a framework that conceivably could be used to maximize (or at least to increase) welfare. This Article proposes a new reconciliation of the patent and anti- trust laws. It proffers a common denominator by which the laws can be measured and compared: innovation. It also recognizes that inno- vation is achieved through different routes in different industries and thus adjusts the antitrust analysis based on the industry. For example, it counsels courts 3 to defer to patents in industries in which patents are critical to innovation, such as pharmaceuticals. And it anticipates a more significant role for the antitrust laws in industries in which the market provides the incentives to innovate, such as computer soft- ware. Part I of this Article sketches the dimensions of the conflict be- tween the patent system and the antitrust laws, both in theory and objectives. See 1 PHILLIP E. AREEDA ET AL., ANTITRUST LAW: AN ANALYSIS OF ANTITRUST PRINCIPLES AND THEIR APPLICATION 100, at 5, 110, at 96, 111, at 97- 115 (2d ed. 2000) (stating that populist goals should not be considered in formulating antitrust rules since antitrust courts cannot promote these goals when they conflict with economic efficiency, and "even where there is no evident conflict... [they] would multiply legal uncertainties and threaten inefficiencies not easily recognized or proved"); see also, e.g., Joseph F. Brodley, The Economic Goals of Antitrust: Efficiency, Con- sumer Welfare, and Technological Progress, 62 N.Y.U. L. REv. 1020, 1021 (1987) ("[T]he purely economic goals of antitrust, properly defined, embrace most of what a progres- sive antitrust policy requires."). 3 As explained in greater detail infra Part lI.B.2.a, because of the indeterminate nature of the Sherman Act, courts have been the primary expositors of antitrust law. 2002] PA TENT-ANTITRUST PARADOX through a hypothetical example. Part I then surveys representative approaches to the intersection that courts have taken and rejects pro posed solutions to the conflict that courts and commentators have of- fered. Part II begins the process of reconciling the patent and antitrust laws by introducing the common denominator of innovation. Part II grounds this selection in the text and legislative history of the statutes and the relevantjurisprudence and economic theory. Part III sets forth the test that courts should apply when evaluating monopolists' patent-based activity under Section 2 of the Sherman Act. Introduced at its most general level, the test takes the form of a rebuttable presumption that proceeds in three steps: (1) a presump- tion that, as long as the monopolist has a justification for the patent- based action other than harming competitors, the conduct is lawful; (2) a rebuttal if competition (and not patents) is responsible for inno- vation in the industry; and (3) a surrebuttal by which the monopolist can demonstrate that the relevant market in the industry is character- ized by innovation. Courts are to determine whether the rebuttal ap- plies based on an evaluation of three ex ante factors-the presence of market-based incentives to innovate, the ease of creating the patented product, and the difficulty of imitating the product-and the ex post factor of the cumulative nature of innovation in the industry. If both the ex ante and ex post factors reveal the primacy of competition in attaining innovation in an industry, then the rebuttal will be met. Part III concludes by applying the test to three hypothetical patentee mo- nopolists: the Bully Monopolist, the Biopharmaceutical Patentee, and the Internet Auctioneer. Part IV responds to anticipated objections to the proposal. First, it contends that courts are able to apply the test. Second, it demon- strates that any reduction in the incentives underlying the patent sys- tem resulting from the application of the test would be minor and not cause for concern. Third, it explains why the proposal is superior to readjustments to the patent system alone. 766 UNIVERSITY OF PENNSYLVANIA LA W REVIEW [Vol. 150: 761 I. THE PROBLEM: A CONFLICT AND No EASY RECONCILIATION A. The Patent-AntitrustC onflict4 1. Different Paths to Welfare On their broadest level, the patent and antitrust laws both en- deavor to increase welfare.' But the paths by which they pursue this objective frequently diverge. The primary purpose of the patent system is straightforward.6 In- ventors and investors expend substantial resources in creating and de- veloping inventions. Conducting research and development and bringing an invention to market often are lengthy and expensive pro- 4 The elucidation of a conflict between the patent and antitrust laws requires cer- tain caveats. In many cases, the laws will not conflict. For example, a company's pure exclusion of competitors will not implicate the antitrust laws if that company lacks mo- nopoly power. Any "conflict" also must be situated in the context in which it will typi- cally arise: a continually developing antitrust common law that, in certain of its itera- tions, affects patents. Of course, the conflict also should be recognized as a consequence of the categorization-in fact, the creation-of laws of "antitrust" and "intellectual property." A different categorization might have avoided the conflict (while creating other difficulties). The present categorization nonetheless implicates the larger issue, one lying beyond the scope of this Article, of the manner in which an- titrust law might affect parties' property rights. SeeSherman Antitrust Act, 15 U.S.C. §§ 1-7 (1994) (prohibiting trusts in restraint of trade and monopolies); Patent Act of 1790, ch. 7, § 1, 1 Star. 109 (codified as amended at 35 U.S.C. §§ 100-376 (1994)) (granting patents to inventors and discover- ers of new and useful processes, machines, manufactures, or compositions of matter); see also BOWMAN, supra note 1, at I ("Both antitrust law and patent law have a common central economic goal: to maximize wealth by producing what consumers want at the lowest cost." (emphasis omitted)). 6 The justification advanced in the text is the standard "utilitarian" justification that courts and commentators have articulated and that the Constitution contem- plates. See U.S. CONST. art. I, § 8, cl. 8 (granting Congress the power "[t]o promote the Progress of Science and useful Arts, by securing for limited Times to Authors and In- ventors the exclusive Right to their respective Writings and Discoveries"); see also, e.g., F.M. SCHERER & DAVID Ross, INDUSTRIAL MARKET STRUCTURE AND ECONOMIC PERFORMANCE 621-24 (3d ed. 1990) (discussing the logic of granting protection from competition with patents). Other conceivable (but much less frequently voiced) justi- fications for the intellectual property system (though not the patent system) include the "moral rights" approach, see Martin A. Roeder, The Doctrine of Moral Right: A Study in the Law of Artists, Authors, and Creators, 53 HARV. L. REV. 554, 557 (1940) (describing a creative act as an extension of an individual's identity); the related "natural rights" approach, seeJOHN LOCKE, Two TREATISES OF GOVERNMENT (Peter Laskett ed., Cam- bridge Univ. Press 1988) (1690) (stating that individuals are entitled to the fruits of their labor, as long as others are not worse off as a result of the privatization); and the "personhood perspective," see MargaretJane Radin, Property and Personhood,3 4 STAN. L. REV. 957, 957 (1982) (stating that an individual needs control over resources in the external environment that take the form of property rights). 20021 PA TENT-ANTITRUST PARADOX cesses, with no guarantees of success at the end of the tunnel. And on those occasions when success is achieved, "free riders" who did not make any such investments might imitate the hard-earned innovation and appropriate its value for themselves. Such activity would tend to deter future inventors and investors, thereby reducing innovation.7 To prevent this, the patent laws promise inventors a right to exclude for a period of twenty years,8 a right that permits inventors to charge prices higher than their postinvention costs, thereby allowing them to recover profits in excess of the value of their front-end investments.9 The right to exclude thus is designed to increase appropriability and thereby the level of invention in society. The unique characteristics of intellectual property support the right to exclude and shed light on the tension between the patent and antitrust laws. First, intellectual property is a public good. That is, it is nonrival (consumption by one person does not leave any less of the good to be consumed by others) and nonexclusive (others cannot be excluded from consuming it).'° As a result of these characteristics, public goods tend to be underproduced and subject to free riders, who are tempted to imitate the invention after it has been devel- oped." Second, the value of intellectual property often is uncertain, 7 See Kenneth W. Dam, The Economic Underpinningso f Patent Law, 23J. LEGAL STUD. 247, 247 (1994) (describing the "appropriability problem" that occurs where a firm fails to recover invention costs due to the inadequate protection of information). For a discussion of the concepts of invention and innovation, see infta note 180. 8 For a discussion of the duration of the right to exclude, see infra note 370 and accompanying text. 9 See SCHERER & Ross, supra note 6, at 622 ("[A]n ... inventor ... must expect that once commercialization occurs, product prices can be held above postinvention production and marketing costs long enough so that the discounted present value of the profits ... will exceed the value of the front-end investment."). Firms patent not only to prevent copying and to recover their development costs, but also, for example, to establish bargaining positions in cross-licensing agreements and to block rival pat- ents on related innovations. See infra note 275 for a discussion of other reasons a firm may seek patents. 10 SeeYochai Benkler, A PoliticalE conomy of the Public Domain: Markets in Information Goods Versus the Marketplace of Ideas, in EXPANDING THE BOUNDARIES OF INTELLECTUAL PROPERTY: INNOVATION POLICY FOR THE KNOWLEDGE SOCIETY 267, 270-71 n.9 (Rochelle Cooper Dreyfuss et al. eds., 2001) [hereinafter EXPANDING BOUNDARIES] ("A pure public good is one that is nonexcludable and nonrivalrous."); see also DONALD S. CHISUM ET AL., PRINCIPLES OF PATENT LAW 58-59 (1998) (discussing the two charac- teristics of public goods); Tracy R. Lewis & Dennis A. Yao, Some Reflections on the Anti- trust Treatment of Intellectual Property, 63 ANTITRUST L.J. 603, 606 (1995) ("Unlike private property, it is difficult to control the use and dissemination of intellectual property once it is released."). 11 CHISUM ET AL., supra note 10, at 59; see also Nancy T. Gallini & MichaelJ. Trebil- cock, IntellectualP roperty Rights and Competition Policy: A Frameworkf or the Analysis of Eco- 768 UNIVERSITY OFP ENNSYLVANIA LAWREVIEW [Vol. 150: 761 most notably because of its novelty and information asymmetry.2 From an antitrust perspective, therefore, it is difficult for enforcers and courts to determine the effect of particular practices involving in- tellectual property on welfare. The antitrust laws, on the other hand, scrutinize activity that re- stricts competition. The rationale of the laws is that competition leads to lower prices, higher output, and more innovation, and that certain agreements between competitors or conduct by monopolists prevents consumers from enjoying these benefits.'3 Because, for example, mo- nopolists lack the constraints provided by competitive markets, they often reduce output, raise prices, limit innovation (so as not to intro- duce products that might dislodge their market position), and fail to allocate resources to the uses most highly valued by consumers.14 But many acts undertaken by patentee monopolists or agreements be- tween patentees and licensees restrict competition by their very opera- tion. For example, patentees may refuse to use or license their pat- ent or may impose quantity restrictions, royalty payments, grantbacks,16 territorial restrictions," or field of use restrictions"' on nomic and Legal Issues, in COMPETITION POLICY AND INTELLECTUAL PROPERTY RIGHTS IN THE KNOWLEDGE-BASED ECONOMY 17, 17 (Robert D. Anderson & Nancy T. Gallini eds., 1998) ("Intellectual property embodies information that is a public good: an inves- tor's consumption of the information does not preclude others from consuming it and so, in the absence of property rights, an innovation will be imitated."). 12 See Lewis & Yao, supra note 10, at 605 (stating that the novelty of intellectual property "implies that the uses and future development potential" of the property are not well-known and that information asymmetry results from the inventor knowing more about the characteristics of the property than do potential licensees or antitrust agencies). 13 See BOWMAN, supra note 1, at I ("[Mionopoly makes it possible to restrict output and raise prices so that consumers pay more for and get less of the things they want most."); Baxter, supra note 1, at 305 ("The effect of monopoly is to lessen output, raise prices, increase returns to producers and diminish social utility ... "). 14 BOWMAN, supra note 1, at 1; VARIAN, supra note 2, at 420-24; Baxter, supra note 1, at 305. But seeJOSEPH A. SCHUMPETER, CAPITALISM, SOCIALISM, AND DEMOCRACY 87- 106 (3d ed. 1950) (arguing that in cases in which monopolists have "superior meth- ods" available to them than those afforded to a "crowd of competitors," the theory that the monopoly price is higher and the monopoly output is lower does not hold). 15 35 U.S.C. § 271 (d) (4) (1994). 16 Grantbacks are arrangements by which a licensee agrees to extend to the licen- sor of intellectual property the right to use the licensee's improvements to the licensed technology. U.S. DEP'T OF JUSTICE & FED. TRADE COMM'N, ANTITRUST GUIDELINES FOR THE LICENSING OF INTELLECTUAL PROPERTY 5.6 (1995) [hereinafter INTELLECTUAL PROPERTY GUIDELINES]. 17 The patent statute permits the exclusive licensing of a patent to "the whole or any specified part of the United States." 35 U.S.C. § 261 (1994). Such a restriction limits the licensee's use of the patented invention to one or 2002] PA TENT-ANTITRUST PARADOX licensees. Activity that may be encouraged under the patent system frequently raises the suspicion of the antitrust laws by reducing com- petition. The provision of the antitrust laws that targets monopolies, Sec- tion 2 of the Sherman Act, exposes this tension most dramatically in focusing on the actions of a single firm. A court may view a company's refusal to share its patented product as predatory conduct justifying a Section 2 violation, even if the action is perfectly lawful under the pat- ent laws. And as courts employ tests that fail to acknowledge the beneficial purposes of patents, such as those focusing on the intent of the monopolist, a change in the market, or the denial of an essential facility, patents may get short shrift.19 more specified fields. See, e.g., Gen. Talking Pictures Corp. v. W. Elec. Co., 304 U.S. 175, 179-82 (1938) (holding that a licensee's leasing of amplifiers for use in theaters violated the patent's restriction against commercial use). 19F or several reasons, Section 2 of the Sherman Act will be the primary vantage point of this Article in exploring the patent-antitrust intersection. Section 2 punishes "[e]very person who shall monopolize, or attempt to monopolize, or combine or con- spire with any other person or persons, to monopolize any part of the trade or com- merce among the several States." 15 U.S.C. § 2 (Supp. 1997). First, the type of conduct at issue in a Section 2 case presents the conflict most di- rectly. Only under Section 2 may a firm's unilateral invocation of its right to exclude- the modus operandi of the patent laws-lead to condemnation. In contrast, Section 1 of the Sherman Act applies to agreements, in which the objectionable conduct relates to actions other than pure exclusion. Courts also must be careful not to condemn uni- lateral actions such as excluding competitors or raising prices that, while initially ap- pearing suspicious, may be consistent with vigorous patent-based activity. See Frank H. Easterbrook, The Limits of Antitrust, 63 TEX. L. REv. 1, 4 (1984) ("The tradition [of in- hospitality] is that judges view each business practice with suspicion .... If the defen- dant cannot convince the judge that its practices are an essential feature of competi- tion, the judge forbids their use."). Second, analysis under Section 2 proves insightful because of the provision's amorphous jurisprudence. Courts interpreting Section 2 all begin with the same test: an antitrust plaintiff must show "(1) the possession of monopoly power in the relevant market and (2) the willful acquisition or maintenance of that power as distinguished from growth or development as a consequence of a superior product, business acu- men, or historic accident." United States v. Grinnell Corp., 384 U.S. 563, 570-71 (1966). But the similarities among courts' Section 2 analyses end with the recitation of the test. As discussed infra Part I.B., courts have applied an array of tests, in some of which they defer to the patent, but in many of which they shoehorn the patent-based nature of the act into a generic Section 2 analysis. In contrast, Section 1, as currently interpreted, offers more specific and manageable tests for courts to apply. For exam- ple, tying has its four-part (sometimes five-part) test; exclusive dealing focuses on the degree of foreclosure in the relevant market; and price fixing is per se unlawful. See, e.g., Eastman Kodak Co. v. Image Technical Servs., Inc. ("Kodak 1'), 504 U.S. 451, 461- 62 (1992) (describing the elements of a tying claim as including (1) two separate products, (2) coercion, (3) market power in the tying product market, and (4) a not insubstantial amount of commerce in the tied product market); Fortner Enter., Inc. v. 770 UNIVERSITY OFP ENNSYL VANIA LA W REVIEW [Vol. 150:761 Another lens through which to view the tension between the pat- ent and antitrust laws involves the balance between static and dynamic efficiency. In interpreting antitrust law, courts have focused primarily on static efficiency-in other words, on increasing economic welfare through a reallocation of the existing supply of resources in a Pareto- optimal fashion (i.e., so that no individual's welfare could be im- proved by a resource reallocation without some other person's welfare being diminished). ° In particular, courts analyze allocative efficiency, striving for an optimal allocation of goods and services to customers.21 Patent law, on the other hand, attempts to increase dynamic effi- ciency, or the Pareto-optimal allocation of resources between the pre- U.S. Steel Corp., 394 U.S. 495, 498-99 (1969) (indicating the elements of a tying claim); Tampa Elec. Co. v. Nashville Coal Co., 365 U.S. 320, 334 (1961) ("The remain- ing determination [after finding an exclusive-dealing arrangement] ... is whether the pre-emption of competition ... tends to substantially foreclose competition in the relevant ... market."); United States v. Socony-Vacuum Oil Co., 310 U.S. 150, 218 (1940) ("[Flor over forty years this Court has consistently and without deviation ad- hered to the principle that price-fixing agreements are unlawful per se under the Sherman Act .. "). As a result, courts interpreting Section 1 do not have the flexibil- ity to fashion a series of tests differentially affecting patents as they do under Section 2. Third, courts are forced to confront the patent-based nature of the defendant's activity more frequently in Section 2 cases than under other antitrust provisions. Be- cause Section 2 focuses on the unilateral activity of the defendant, courts sometimes (even if less often than they should) analyze the nature of the activity-exploring, for example, the justifications for a patentee monopolist's refusal to deal with competitors. Under Section 1, in contrast, the court first applies unexceptional, nonpatent-based standards, such as the extent to which the market is foreclosed by an agreement. The patent-based nature of the activity comes into play, if at all, only when courts consider the procompetitive justifications for the agreement. This late consideration becomes dispositive when it never even is reached: most courts dispose of Section 1 cases by finding that the plaintiff has failed to prove an anticompetitive effect and by never reaching the issue of procompetitive effects. See Michael A. Carrier, The Real Rule of Reason: Bridging the Disconnect, 1999 B.Y.U. L. REV. 1265, 1268 (finding that courts have disposed of eighty-four percent of Rule of Reason cases in the modern era on the grounds that the plaintiff could not demonstrate an anticompetitive effect). In short, Section 2 offers the best perch from which to explore the patent-antitrust conflict. The unilateral activity at issue frequently calls for a direct focus on the patent- based nature of the conduct. The courts have created tests that either ignore or defer excessively to the incentives underlying the patent system. The exclusion promised under the patent laws is punished under the antitrust laws. And as the courts are tossed between the powerful conflicting waves of patent and antitrust, the dim light- house signal provided by Section 2 fades to black. 20W. KIp ViscusI ET AL., ECONOMICS OF REGULATION AND ANTITRUST 75-76 (3d ed. 2000); Merritt B. Fox, SecuritiesD isclosure in a GlobalizingM arket: Who Should Regulate Whom. 95 MICH. L. REV. 2498, 2551 n.99 (1997). 21 Antitrust courts could reduce the tension between the laws by focusing on dy- namic efficiency (for example, by endeavoring to increase innovation).