"Law, Peter A. Allard School of"@en . "DSpace"@en . "UBCV"@en . "Ramage, Ian"@en . "2009-08-14T19:19:19Z"@en . "2002"@en . "Master of Laws - LLM"@en . "University of British Columbia"@en . "In advanced industrial economies where, increasingly, intellectual assets are the principal source of\r\nvalue, productivity, and growth, strong intellectual property rights (IPRs)\u00E2\u0080\u0094conferred by patents,\r\ncopyrights, and penalties for misappropriation of trade secrets\u00E2\u0080\u0094are an important inducement to\r\ninvention and investment. For this reason, the extension and strengthening of IPRs in the United\r\nStates and elsewhere in the past twenty-five years were appropriate and probably necessary. It may be\r\nthat in some respects those processes should proceed further. On the other hand, there is growing\r\nfriction over the assertion and exercise of some IPRs, particular patents, and claims that in some\r\ncircumstances they may be discouraging research, its communication, and use. The question arises\r\nwhether in some respects the strengthening and extension have proceeded too far.\r\nIt is well known that the use of, reliance upon, and effects of patent protections vary across\r\nindustries and technologies, but until recently there has been remarkably little empirical research\r\ndocumenting these differences. Fortunately, this is beginning to change, and the effects of some of the\r\npolicy changes in the 1980s and 1990s are beginning to be investigated. Some evidence suggests that\r\nthe effort to strengthen patent rights has indeed increased their importance and may have contributed\r\nto the growth of industrial R&D funding. On the other hand, recent survey evidence indicates that\r\nU.S. manufacturing firms in most industries rely more heavily on trade secrecy, lead time, and other\r\ntechnological protections to recoup their R&D investments than they do on legal mechanisms such as\r\npatents.\r\nThis thesis examines the effects that a stronger, broader patent regime is having on today's\r\nindustries. The main issues that emerge are those of patent quality and scope, as caused by problems\r\nwith patent administration and litigation. Various solutions to these problems are then investigated,\r\nand recommendations made for future reform."@en . "https://circle.library.ubc.ca/rest/handle/2429/12210?expand=metadata"@en . "7004180 bytes"@en . "application/pdf"@en . "Patenting Innovation: Intellectual Property Rights in the New Economy by Ian Ramage B.A.Sc., Simon Fraser University, 1996 LL.B., University of British Columbia, 2000 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF T H E REQUIREMENTS FOR THE D E G R E E OF MASTER OF LAWS FACULTY OF G R A D U A T E STUDIES T H E UNIVERSITY OF BRITISH C O L U M B I A V A N C O U V E R , B . C . , C A N A D A Faculty of Law April 2002 accept this thesis as conforming to the required standard. \u00C2\u00A9 Ian Ramage, 2002 In presenting this thesis in partial fulfillment of the requirements for an advanced degree at the University of British Columbia, 1 agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Faculty of Law The University of British Columbia Vancouver, Canada Date Abstract ii ABSTRACT In advanced industrial economies where, increasingly, intellectual assets are the principal source of value, productivity, and growth, strong intellectual property rights (IPRs)\u00E2\u0080\u0094conferred by patents, copyrights, and penalties for misappropriation of trade secrets\u00E2\u0080\u0094are an important inducement to invention and investment. For this reason, the extension and strengthening of IPRs in the United States and elsewhere in the past twenty-five years were appropriate and probably necessary. It may be that in some respects those processes should proceed further. On the other hand, there is growing friction over the assertion and exercise of some IPRs, particular patents, and claims that in some circumstances they may be discouraging research, its communication, and use. The question arises whether in some respects the strengthening and extension have proceeded too far. It is well known that the use of, reliance upon, and effects of patent protections vary across industries and technologies, but until recently there has been remarkably little empirical research documenting these differences. Fortunately, this is beginning to change, and the effects of some of the policy changes in the 1980s and 1990s are beginning to be investigated. Some evidence suggests that the effort to strengthen patent rights has indeed increased their importance and may have contributed to the growth of industrial R&D funding. On the other hand, recent survey evidence indicates that U.S. manufacturing firms in most industries rely more heavily on trade secrecy, lead time, and other technological protections to recoup their R&D investments than they do on legal mechanisms such as patents. This thesis examines the effects that a stronger, broader patent regime is having on today's industries. The main issues that emerge are those of patent quality and scope, as caused by problems with patent administration and litigation. Various solutions to these problems are then investigated, and recommendations made for future reform. Contents 111 C O N T E N T S ABSTRACT ii FIGURES v TABLES vi I. INTRODUCTION 1 II. STRENGTHENING & EXTENDING THE U.S. PATENT SYSTEM 5 A. The Patenting of Publicly-Funded Research 5 B. The New Realm of Patentability 8 C. The Creation of the Federal Circuit // III. CONCERNING THE EFFECTS ON INNOVATION 16 A. The Patent Surge 19 B. The Patent Paradox 22 C. Semiconductor Patenting 24 D. Software Pa ten ting 28 E. Business Method Patenting 34 I V . IDENTIFYING THE ISSUES 46 A. Patent Theory 46 B. Patent Scope 51 1. THEORETICAL ANALYSES OF PATENT SCOPE 52 2. EMPIRICAL STUDIES OF PATENT SCOPE 55 C. Pa ten t A dministra tion -61 D. Patent Litigation 66 E. The Disconnect \"73 V . FORMULATING THE SOLUTIONS 76 A. Reform To Date 76 1. FIRST INVENTOR DEFENSE 76 2. N E W PUBLICATION RULES 78 3. INTER PARTES REEXAMINATION 80 4. OTHER AIPA CHANGES 82 5. T H E P T O \"ACTION PLAN\" '. 85 B. Business Method Reform 86 1. REVISITING THE BUSINESS METHOD EXCEPTION 86 2. SHORTER TERMS 89 3. T H E BUSINESS METHOD PATENT IMPROVEMENT A C T 90 Contents i y C. A dministra five Reform 94 1. N E W PRIOR ART DISCLOSURE SYSTEM 95 2. N E W FEE SHIFTING REGIME \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 97 3. OFFICIAL NOTICE 99 4. JEPSON CLAIMS 102 5. M O R E RESPONSIBILITIES FOR THE HEAVIEST USERS 104 D. Doctrinal Reform 105 1. EXPERIMENTAL USE 106 2. FAIR USE I l l 3. REVERSE ENGINEERING U S V I . C O N C L U S I O N 120 R E F E R E N C E S 127 Figures v F I G U R E S Figure 1: U.S. Patent Activity (Domestic) 17 Figure 2: U.S. Research & Development Spending 17 Figure 3: U.S. Patent Activity (Total) 61 Tables vi TABLES Table 1: Effectiveness Rankings of Various Appropriability Mechanisms 23 Table 2: Internet and Business Method Patents Granted 62 Table 3: Patent Filing Rates, by Technology Field and Time Period 69 Part I. Introduction 1 I . INTRODUCTION As the industrialized nations of the world forge into the twenty-first century, they are being propelled by the information-based technologies of tomorrow, not the resource-based businesses of old. For the software and telecommunications companies of the new economy, success depends far more on technical know-how than manufacturing might. Brainpower, not steam-power, now drives the wheels of economic progress. In a world where intellectual assets are fast becoming the principal source of productivity and growth, it is increasingly important to protect those assets with the rights conferred by patents, copyrights, trademarks and trade secrets. Usually by virtue of a statutory monopoly, such intellectual property rights provide penalties for the unauthorized use of proprietary ideas and inventions. Yet the real value of such rights lies not in the rightholder's power to protect against imitation or theft but in society's ability to benefit from that protection. Creative or innovative activity typically involves substantial expenditures of time, effort, and money. If artistic works or inventions are easily copied, there is a smaller prospect of profits with which to recoup those costs, and so some innovators will choose not to devote the resources required to produce them. The protection offered by intellectual property rights counteracts this effect, thereby increasing innovative activity and providing society with valuable inventions and works of art which might not otherwise exist. Technological progress is further enhanced by the disclosure incentives of copyright and patents. By enticing an innovator to publicly disclose his invention or discovery in exchange for the grant of exclusivity, the technical advance embodied in the invention is added to the store of public knowledge, thereby stimulating other ideas and leading to further innovation. Based on this simple model, the past twenty-five years has witnessed a dramatic expansion of the world's intellectual property systems. If the intellectual property laws create incentives for people Part I. Introduction 2 to invent, then stronger intellectual property protection should result in increased innovative activity and technological advance\u00E2\u0080\u0094more protection, more innovation. Nowhere was the expansion more evident than in the United States. In the midst of the most severe recession since World War II and' faced with the rapid emergence of Japanese and European competitors in the computer and semiconductors industries, U.S. policy-makers in the mid-1970s had become particularly concerned about the technological competitiveness and productivity of the country's economy. Seizing upon stronger intellectual property protection as a way to regain the comparative advantages they had once enjoyed, the United States embarked on a number of initiatives aimed at significantly strengthening and extending intellectual property rights, both at home and abroad. The following chronology highlights the most important of those events: \u00E2\u0080\u00A2 The 1976 Copyright Act, completely overhauling copyright law to include new forms of media. \u00E2\u0080\u00A2 The 1980 amendments to the Copyright Act, expressly permitting copyright protection for computer programs. \u00E2\u0080\u00A2 The 1980 Bayh-Dole Act, enabling universities and other non-profit federal contractors to obtain exclusive rights to their inventions. \u00E2\u0080\u00A2 The Supreme Court's decisions in Diamond v. Chakrabarty, which established the patentability of genetically-engineered organisms, and Diamond v. Diehr, which opened the door to the patenting of computer software. \u00E2\u0080\u00A2 The creation in 1982 of the Court of Appeals for the Federal Circuit (CAFC), a unified federal court with exclusive jurisdiction over patent appeals. \u00E2\u0080\u00A2 The Semiconductor Chip Protection Act of 1984, creating an entirely new intellectual property right for the protection of semiconductor designs. \u00E2\u0080\u00A2 The 1984 Hatch-Waxman Act, extending the patent terms on regulated pharmaceuticals. \u00E2\u0080\u00A2 The 1988 Process Patent Amendments Act, blocking the import to the United States of foreign products produced by methods infringing U.S. process patents. Part I. In troduction 3 \u00E2\u0080\u00A2 The 1994 Uruguay Round Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS) and various bilateral agreements, ostensibly strengthening foreign IPR protection and harmonizing national standards. \u00E2\u0080\u00A2 The Economic Espionage Act of 1996, creating new federal criminal and civil penalties for theft of trade secrets. \u00E2\u0080\u00A2 The Digital Millennium Copyright Act of 1998, further strengthening U.S. copyright protection. \u00E2\u0080\u00A2 The 1998 Federal Circuit decision in State Street Bank v. Signature Financial, upholding the patentability of business methods and processes. \u00E2\u0080\u00A2 The 1999 Anti-Cybersquatting Act, protecting the rights of trademark holders in cyberspace. \u00E2\u0080\u00A2 The Napster appeals court ruling in March 2001, denouncing the cyber-circumvention of copyright laws. Looking back, it seems incontrovertible that the strengthening and extending of intellectual property rights has translated into greater innovation and economic growth. The past two decades have played host to the greatest technological advances in the history of human civilization. The current rate of technological advance is simply amazing. Research and development expenditures have never been higher. Productivity numbers are way up. Yet there is growing friction over the acquisition and assertion of intellectual property rights, especially patents. Recent litigation has highlighted the existence of patents on inventions which seem patently obvious or overly broad, particularly in the new area of business method patenting. In the software and semiconductor industries, the proliferation of patents appears to be erecting significant barriers to innovation. Furthermore, there is limited evidence that a stronger patent system has actually led to increased spending on research and development in the United States. Indeed, a number of surveys Part I. Introduction 4 have revealed that patents are considered to be among the least effective means for recouping investments in research and development. Paradoxes abound, forcing us to seriously question the traditional incentive model. It may be that patents aren't actually fostering innovation, they're hindering it\u00E2\u0080\u0094either directly, with barriers that discourage companies from innovating, or indirectly, by funneling resources away from research and development and into patent portfolio management, patent applications, litigation costs and damages. Using recent economic data, this thesis takes a hard look at the link between patents and technological innovation and whether it is, in fact, just a disconnect. Is the traditional model upon which the entire patent system has been built\u00E2\u0080\u0094a theory so patently intuitive as to remain essentially unquestioned for over 200 years\u00E2\u0080\u0094so far off the mark? And could that simple assumption, relied on for decades of patent reform in the United States and elsewhere, be doing more harm than good? In Part II: Strengthening and Extending the Patent System, I'll trace the developments in U.S. patent law over the last twenty years and identify how patent rights have been strengthened and dramatically extended. Part III: Concerning the Effects on Innovation will examine the effects that the new patent rights are having on the various industries of the new economy, including semiconductors, software, and the Internet. In Part IV: Identifying the Issues, I will review the theory behind the patent system and its recent expansion and fortification, and then lay out the main issues, including patent scope, patent administration and patent litigation. Part V: Formulating the Solutions will survey the proposed reforms. What can be done to make the patent system work better? What has been done so far to address the problems? How can we reconnect the patent system with innovation? Part II. Strengthening & Extending the U.S. Patent System 5 I I . S T R E N G T H E N I N G & E X T E N D I N G T H E U . S . P A T E N T S Y S T E M Nineteen-eighty was a watershed year for the United States patent system. A series of legislative initiatives extended patent protection to a whole new set of players\u00E2\u0080\u0094universities and federally-funded laboratories. Then, in two landmark decisions, the Supreme Court extended the reach of the patent system into two entirely new areas of endeavour\u00E2\u0080\u0094biotechnology and software. Two years later, the creation of a new patent court\u00E2\u0080\u0094the Court of Appeals for the Federal Circuit\u00E2\u0080\u0094resulted in a significant reinforcement of those rights and their further expansion beyond the traditional frontiers of patentable subject matter. A. The Patenting of Publicly-Funded Research Traditionally, U.S. law regarding the patentability of federally-funded research provided that the government held title to any invention generated with its funds. As patent holder, it could then issue royalty-free non-exclusive licenses to any private manufacturers wishing to commercialize the technology. Post-war efforts to allow federal contractors and universities to assume title to their inventions were met with opposition by those arguing that the results of federally-funded research should inure to the benefit of the public, not a private monopoly under which the public is charged for the use of a technology it had funded. This debate continued through the 1950s, and, despite being the topic of over four special commissions and forty congressional hearings or reports, in 1979 there was still no comprehensive policy regarding the patenting of publicly-funded research in place.1 A few universities patented fairly actively, but their patents represented either the fruits of research performed with no federal funding, 1 Adam B. Jaffe, \"The U.S. Patent System in Transition: Policy Innovation and the Innovative Process\" (1999) National Bureau of Economic Research Working Paper No. W7280 at S n.3. Part II. Strengthening & Extending the U.S. Patent System 6 patents sought for public or professional prestige and not commercial development, or inventions that had received a title rights waiver from the federal agency funding the research. Some agencies routinely granted such waivers, a few others signed blanket agreements that permitted certain universities to patent the results of all research funded by that agency, but the majority of agencies kept full title and offered only non-exclusive licenses. By the late 1970s, it was obvious that such a system was almost completely ineffective at commercializing federal research. Private firms had little incentive to spend resources on developing federal research into marketable products if that research remained in the public domain for anyone else to develop as well.2 Quite rightly, businesses viewed the government's non-exclusive licenses as \"no patent protection at all,\" and stayed away in droves\u00E2\u0080\u0094of the 28,000 patents owned by the government at the time, a mere four percent had been licensed to private industry for further development.3 Faced with additional evidence that the failure to patent federally-financed discoveries had allowed foreign competitors to expropriate them instead, Congress began to pass various pieces of legislation that eventually allowed all publicly-funded research to be either patented or exclusively licensed. The most important of these was the Bayh-Dole Act of 1980,4 which allowed universities and other non-profit institutions to automatically retain title to patents derived from federal research, 2 A firm could be assured of profiting from its expenditures only i f it could secure a competit ive advantage\u00E2\u0080\u0094that is, only i f it could , through exclusive licensing of the patented invention, appropriate the discovery for itself. 3 See A r t i K . Rai , \"Regulating Scientific Research: Intellectual Property Rights and the N o r m s of Science\" (1999) 94 N w . U . L . Rev. 7 7 a t 9 7 n . l l 3 \u00E2\u0080\u0094 q u o t i n g statement from S. Rep. N o . 96-480 at 28 (1979). The extension of patent protection to federally-funded inventions, then, was deemed necessary \"not for the traditional reason that such rights wou ld provide an incentive to invent, but rather as an incentive for private firms to undertake the further investment necessary to translate the inventions into marketable products.\" Rai at 96. 4 Pub. L . N o . 96-517, 6(a), 94 Stat. 3015, 3019-28 (1980), codified as amended at 35 U . S . C . \u00C2\u00A7\u00C2\u00A7 200-212 (1994). In response to pressure from industry, the Bayh-Dole Ac t , originally known as the Universi ty and Small Business Patent Procedures A c t , also gave small businesses the right to seek patents on the results of their federally-funded research. In 1983, the right to retain patent ownership of federally-sponsored research was extended to large businesses as w e l l . See Rebecca S. Eisenberg, \" A Technology Policy Perspective on the N I H Gene Patenting Controversy\" (1994) 55 U . Pitt. L . Rev. 633 at 637. Part II. Strengthening & Extending the U. S. Patent System 7 thereby removing the need to get an explicit waiver from the funding agency. The legislation explicitly recognized that technology transfer to the private sector was a desirable outcome of publicly-funded research, and endorsed the principle that exclusive licensing of such research was sometimes necessary to achieve that objective.5 To further stimulate innovation, the Bayh-Dole Act also required that universities share patent royalties with individual inventors, thus giving both parties an incentive to be on the lookout for patentable inventions.6 In addition to opening up university patenting, the Bayh-Dole Act also allowed government-operated laboratories such as the National Institute of Heath to grant exclusive licenses on government-owned patents. The simultaneous passage of the Stevenson-Wydler Technology Innovation Act of 1980,7 made technology transfer a mission of all federal laboratories and created a variety of institutional structures and regulations to facilitate it. Over the course of the 1980s, a series of initiatives further expanded the possibilities for patenting by these labs,8 while a variety of implementing memoranda and legislative clarifications extended many of these provisions to those facilities owned by the government but operated by research contractors, including the National Laboratories of Los Alamos, Oak Ridge, Lawrence Livermore, and others. The wave of legislation did not completely resolve the debate over how much ownership of government-funded R&D ought to be transferred to private sector entities\u00E2\u0080\u0094in recent years, the controversy has impinged on the patenting activities of federal labs and raised questions about the 5 The stated policy objective of the Bayh-Dole Ac t was \"to use the patent system to promote the utilization of inventions arising from federally supported research or development.\" 35 U . S . C . \u00C2\u00A7 200 (1994). 6 See 35 U . S . C . \u00C2\u00A7 202(c)(7)(b) (1994). Four years later, the passage of the Patent L a w Amendments A c t further expanded the rights of universities by removing certain restrictions contained in Bayh-Dole regarding the kinds of inventions universities could o w n and the freedom with which they could assign their property rights to other parties. Patent Law Amendments A c t of 1984, Pub. L . N o . 98-620, codified at 17 U . S . C . \u00C2\u00A7 907 (1994). 7 Pub. L . N o . 96-480, 94 Stat. 2311 (1980), codified at 15 U . S . C . \u00C2\u00A7 3701 (1980). 8 See, e.g., the Federal Technology Transfer A c t of 1986\u00E2\u0080\u0094Pub. L . N o . 99-502, 100 Stat. 1785-87 (1986), codified as amended at IS U . S . C . \u00C2\u00A7 3710a( l ) , (b)(2)-(3) (1994)\u00E2\u0080\u0094which later amended the Stevenson-Wydler Ac t to authorize government-operated laboratories to enter into cooperative R & D agreements wi th industry and to agree in advance to assign patents on inventions made by federal employees to the collaborating f i rm. Part II. Strengthening & Extending the U. S. Patent System 8 desirability of at least some forms of university patenting. Nevertheless, the 1980s witnessed an extraordinary transformation of the U.S. patent system from one in which the patenting of publicly-funded research was the exception to one in which it is the norm.9 B. The New Realm of Patentability Concurrent with these changes in who could patent were momentous changes in what could be patented.10 In the landmark 1980 case of Diamond v. Chakrabarty, the Supreme Court was asked to decide whether a living bacterium designed to degrade crude oil was patentable subject matter.\" The Court ruled in the affirmative, establishing for the first time the patentability of genetically-engineered organisms, and by extension, biotechnology and genomics. For purposes of patentable subject-matter determination, said the Court, the relevant distinction is not between living and non-living materials, but rather \"between products of nature, whether living or not, and human-made inventions.\"12 Less than four months later, the Court heard Diamond v. Diehr, a case involving the patentability of a process for curing synthetic rubber using a well-known mathematical algorithm and a 9 The extension of patenting and licensing privileges to inventors in universities and government laboratories can be viewed as an extension of the larger trend toward corporatization of research in general. See Catherine L. Fisk, \"Removing the 'Fuel oflnterest' from the 'Fire of Genius': Law and the Employee-Inventor, 1830-1930\" (1998) 65 U. Chi. L. Rev. 1127; Robert P. Merges, \"The Law and Economics of Employee Inventions\" (1999) 13 Harv. J. Law & Tech. 1. 1 0 In order to be eligible for U.S. patent protection, an invention must fit into one of the four categories specified by section 101 of the Patent Act, namely, \"process, machine, manufacture, or composition of matter.\" 35 U.S.C. \u00C2\u00A7 101 (1994). See Kewanee Oil Co. v. Bicron Corp., 416 U.S. 470 at 483 (1974): \"[N]o patent is available for a discovery, however useful, novel and nonobvious, unless it falls within one of the express categories of patentable subject matter of 35 U.S.C. \u00C2\u00A7 101.\" As we'll see, the new technologies of the 1980s and 90s would seriously test these traditional boundaries. \" Diamond v. Chakrabarty, 447 U.S. 303 (1980). 12 Id. at 313. The decision led inevitably to the patenting of the Harvard Mouse on 12 April 1988. See Carrie F. Walter, \"Beyond the Harvard Mouse: Current Patent Practice and the Necessity of Clear Guidelines in Biotechnology Patent Law\" (1998) 73 Ind. L.J. 1025 at 1036-37; Michael D. Davis, \"The Patenting of Products of Nature\" (1995) 21 Rutgers Computer & Tech. L.J. 293. Part II. Strengthening & Extending the U.S. Patent System 9 computer for the calculation of the appropriate cure time.'3 An attempt to patent a similar computerized process in 1978's Parker v. Flook was rejected because it violated the \"mathematical algorithm\" exception,'4 yet the Court upheld the patent in Diehr, relying on the fact that the program acted in cooperation with an \"otherwise statutory\" industrial process (the curing). This distinction resulted in some rather inventive patent drafting, but the door was now open to the patenting of computer software.15 While neither Chakrabarty nor Diehr actually involved traditional research, both decisions were likely motivated by the new Congressional mandate to better exploit the fruits of basic research.16 In Chakrabarty, the Court looked to the committee reports accompanying the 1952 Patent Act to conclude that \"Congress intends statutory subject matter to 'include anything under the sun that is made by man.\"'17 In Diehr, the Court reiterated that bold statement, and by allowing a patent on a process based almost exclusively on a mathematical algorithm, suggested that it would permit patenting of all inventions that moved even slightly in the direction of applied work.1 8 13 Diamond v. Diehr, 450 U.S. 175 (1981). 14 Parker v. Flook, 437 U.S. 584 (1978). Throughout the 1960s and 70s, computer software had been deemed unpatentable on the basis that it was really just a concatenation of mathematical algorithms or mental steps. See Gottschalk v. Benson, 409 U.S. 63 at 67 (1972): \"Phenomena of nature, though just discovered, mental processes, and abstract intellectual concepts are not patentable, as they are the basic tools of scientific and technological work.\" This was based on longstanding judicial doctrine. Sec Rubber-Tip Pencil Co. v. Howard, 87 U.S. (20 Wall.) 498 at 507 (1874) (\"An idea of itself is not patentable...\"); O'Reilly v. Morse, 56 U.S. (15 How.) 62 at 116 (1854) (\"[T]he discovery of a principle in natural philosophy or physical science, is not patentable.\"); Le Roy v. Tatham, 55 U.S. (14 How.) 156 at 175 (1852) (\"A principle, in the abstract, is a fundamental truth; an original cause; a motive; these cannot be patented, as no one can claim in either of them an exclusive right.\"). 1 5 The legal strategy employed after Diehr might be called the \"Doctrine of the Magic Words\"\u00E2\u0080\u0094software was now patentable subject matter, but only if you spoke the magic words and pretended you were patenting something else entirely. In the 1980 and early 90s, patent attorneys did just that, claiming software inventions as various hardware devices or machines\u00E2\u0080\u0094or even as pizza ovens. See Julie E. Cohen & Mark A. Lemley, \"Patent Scope and Innovation in the Software Industry\" (2001) 89 Calif. L. Rev. 1 at 9. 1 6 As expressed by the Bayh-Dole and Stevenson-Wydler Technology Innovation Acts, supra. 17 Chakrabarty, 447 U.S. at 309\u00E2\u0080\u0094quoting S. Rep. No. 1979, 82d Cong., 2d Sess. at 5 (1952); H.R. Rep. No. 1923, 82d Cong., 2d Sess. at 6 (1952). 1 8 This softened considerably the requirement of practical utility laid down in the 1966 Supreme Court decision of Brenner v. Manson, 383 U.S. 519 (1966). Part II. Strengthening & Extending the U.S. Patent System 10 If the Supreme Court had unlocked the doors to biotechnology and software patenting, its successor kicked them wide open. In 1994, the Court of Appeals for the Federal Circuit essentially eliminated the \"otherwise statutory process or apparatus\" requirement from Diehr, and established that a mathematical algorithm becomes patentable subject matter merely by virtue of its being programmed into a \"general purpose\" computer.19 Then, in 1998's State Street Bank & Trust Co. v. Signature Financial Group, Inc., the Federal Circuit effectively ended what was left of the software debate, requiring only that the \"practical application of a mathematical algorithm, formula, or calculation\" produce a \"useful, concrete, and tangible result.\"20 And it opened another door\u00E2\u0080\u0094the patenting of methods of doing business.21 It's been a similar story for biotechnology. Although the Federal Circuit has retained the idea that \"products of nature\" per se are not patentable, it has routinely upheld patents on purified and isolated forms of molecules that occur in nature, including full gene sequences whose physiological function (i.e., the protein for which they code) has been identified.22 Today, genes and gene fragments are routinely patented by genomics companies such as Human Genome Systems, Inc. and Incyte Genomics, Inc.23 19 InreAIappat, 33 F .3d 1526 (Fed. C i r . 1994) en banc. 2 0 State Street Bank & Trust Co. v. Signature Financial Group, Inc., 149 F .3d 1368 at 1373 (Fed. C i r . 1998), cert, denied, 119 S.Ct . 851 (1999). The Federal C i rcu i t confirmed its rul ing the next year i n AT&T Corp. v. Excel Communications, 1998 U . S . Dist . Lexis 5346 (Cov . A . 96-434-SLR) ( D . D e l . 27 M a r c h 1998), r e v ' d , 172 F .3d 1352 (Fed. C i r . 1999): \"It is n o w clear that computer-based programming constitutes patentable subject matter so long as the basic requirements of [novelty and utility] are met.\" 2 1 Business method patents w i l l be discussed in more detail i n Section III.E, infra. 2 2 See, e.g., Genentech, Inc. v. Chiron Corp., 112 F .3d 495 (Fed. C i r . 1997). 2 3 The patenting of gene fragments, known as expressed sequence tags (ESTs), has been controversial because, unlike full-length sequences, their functions are typically unknown. Many of these EST patent claims are also notable for their broad scope\u00E2\u0080\u0094the applications claim not only the EST but also the full gene of which it is a part, as w e l l as future uses of the gene. See Ra i , supra note 3, at 103-04. Again , a detailed examination of the issues i n biotechnology patenting is outside the scope of this paper (and my l imited scientific proficiency). Part II. Strengthening & Extending the LI. S. Patent System 11 C. The Creation of the Federal Circuit In the late 1970s, the U.S. patent system was widely perceived to be weak and ineffective, or at best, too slow. The Patent and Trademark Office (PTO) was so overworked that for a while in 1979, they stopped granting patents altogether.24 Distrustful of monopoly power in general, the Justice Department and the Federal Trade Commission (FTC) took a rather dim view of patents, often colouring attempts to enforce them with shades of anti-trust.25 The judiciary was even less kind. The U.S. Supreme Court had been hostile to patents since the Great Depression\u00E2\u0080\u0094between 1931 and 1976, it had invalidated the patent at issue in fifty-one, or 84%, of the 61 patent cases argued before it. 26 Although the majority of lower courts were somewhat more receptive to patent holders, some were worse\u00E2\u0080\u0094during the same period, the Eighth Circuit invalidated 89% of the patents coming its way. 27 Such discrepancies had led to the further problem of \"forum shopping\"\u00E2\u0080\u0094the practice whereby patent holders would try to bring cases in court circuits sympathetic to patents, while alleged infringers would seek out jurisdictions believed to be hostile. This further increased the costs and delays associated with patent cases, and acted to undermine the uniformity of the federal patent laws.28 Adam B. Jaffe, \"The U.S. Patent System in Transition: Policy Innovation and the Innovative Process\" (1999) National Bureau of Economic Research Working Paper No. W7280 at 3. The Department of Justice had a special section of its Antitrust Division devoted to anti-patent advocacy and litigation, and to some it appeared that the antitrust noose would constantly be tightened. See Easterbrook, \"Is There a Ratchet in Antitrust Law?\" (1982) 60 Tex. L. Rev. 70S. The FTC, not to be outdone, attacked Xerox for building a great company through the judicious use of patents. See Xerox Corp., 86 F.T.C. 364(1975). Lawrence Baum, \"The Federal Circuits and Patent Validity: An Analysis of the Record\" (1974) 56 J. Pat. & Trademark Off. Soc'y 758 at 777. Id. at 762. See Rochelle Cooper Dreyfuss, \"The Federal Circuit: A Case Study in Specialized Courts\" (1989) 64 N.Y.U. L. Rev. 1. See also H.R. Rep. No. 97-312 at 20-21 (1981) (noting that \"some circuit courts are regarded as'pro-patent' and others 'anti-patent' and [that] much time and money is expended 'shopping' for a favorable venue.\"); S. Rep. No. 97-275 at 5 (1981) (indicating that Congress wanted to \"increase doctrinal stability in the field of patent law.\") Part II. Strengthening & Extending the U. S. Patent System 12 In 1980, the patent system started to get stronger. Shortly after the Chakrabarty decision, the Supreme Court ruled that since monopoly power was indeed the purpose of the patent grant, efforts to extract the monopoly rents generated by them were not violations of anti-trust law.29 The Department of Justice removed itself from the business of patent bashing.30 Congress enacted a series of laws directed at strengthening and streamlining the Patent Office,31 and then, with the passage of Federal Courts Improvements Act of 1982, created a new federal patent court\u00E2\u0080\u0094the Court of Appeals for the Federal Circuit (CAFC)\u00E2\u0080\u0094replacing the regional appeals courts as the court of exclusive jurisdiction for all patent appeals.32 Although ostensibly a procedural reform\u00E2\u0080\u0094it was expected that the Federal Circuit would not only solve the problem of forum shopping, but also bring a degree of specialization and in-depth knowledge to a complicated area of law, and contribute to more uniform decisions by the various district courts\u00E2\u0080\u0094the creation of the Federal Circuit had a rapid and profound impact on U.S. patent law. In only a few short years, the doctrine laid down by nearly 150 years of Supreme Court decisions was reconsidered, revised, and re-written.33 29 Dawson Chem. Co. v. Rohm & Haas Co., 448 U.S. 176 at 201-02 (1980). 3 0 See United States v. Studiengesellschaft Kohle, 670 F.2d 1122 (D.C. Cir. 1981); Lipsky, \"Current Antitrust Division Views on Patent Licensing Practices\" (1981) SO Antitrust L.J. SIS. 3 1 Act of Dec. 12, 1980, Pub. L. No. 96-S17, 94 Stat. 301S, codified as amended at 3S U.S.C. \u00C2\u00A7\u00C2\u00A7 301-307 (1982). The bill increased funding to the PTO and gave the agency the power to reexamine patents. 3 2 Federal Courts Improvement Act of 1982, Pub. L. No. 97-164, 96 Stat. 2S (1982), codified at 3S U.S.C. \u00C2\u00A7\u00C2\u00A7 141-46 (1988). Technically, the CAFC was created by merging the Court of Customs and Patent Appeals (CCPA) and the Court of Claims, and is the successor to the CCPA in hearing appeals from the PTO's Board of Patent Appeals and Interferences. See Martin J. Adelman, \"The New World of Patents Created by the Court of Appeals for the Federal Circuit\" (1987) 20 U. Mich. J.L. Ref. 979 at 982-83. 3 3 Though, technically, the Supreme Court retained a certiorari authority to hear an appeal from the Federal Circuit, it rarely does. See Mark D. Janis, \"Patent Law in the Age of the Invisible Supreme Court\" [2001] U. 111. L. Rev. 387 at 387: \" The Supreme Court has rendered itself well nigh invisible in modern substantive patent law. The Court of Appeals for the Federal Circuit, created in 1982, has become the de facto supreme court of patents. In those rare patent cases when the real Supreme Court has materialized, the Court has left behind a largely uninspiring jurisprudence.\" Part II. Strengthening & Extending the U.S. Patent System 13 The most significant of the Federal Circuit's changes was the dramatic lowering of patentability standards\u00E2\u0080\u0094particularly the \"nonobviousness\" requirement of section 103.34 Indeed, the standards came down so swiftly and so significantly that a \"patentability gap\" was created, as companies scrambled to obtain patents for research considered unpatentable a few years before.35 Coupled with a reinforcement of the presumption of validity itself,36 the result was no less than a new \"pro-patent\" 37 era. The numbers tell the story. During the first three years of its existence, the Federal Circuit invalidated only forty-four percent of the patents it adjudicated,38 nearly half the Supreme Court rate of 83%, and far less than the old appeals court rate of approximately 70%.39 Before 1980, a district court In pertinent part, section 103 requires that \"[a] patent may not be obtained . . . if the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art.\" 35 U.S.C. \u00C2\u00A7 103(a) (1994). Essentially, \"prior art\" is a measurement of the current state of technological advance. See infra, note 79, and accompanying text. The reduction of the nonobviousness standard was accomplished primarily by elevating the importance of secondary factors such as commercial success and long-felt need \"to a central place in the law of patentability.\" Robert P. Merges, \"Commercial Success and Patent Standards: Perspectives on Innovation\" (1988) 76 Calif. L. Rev. 803 at 834. See Quillen, Cecil D., Jr. \"Innovation and the United States Patent System Today,\" Presentation to the Continuing Legal Education Institute, Antitrust and Intellectual Property: Practice and Policy Issues for the 1990's (19 October 1992) at 1. Section 282 of the U.S. Patent Act says that a patent challenged in court should be presumed valid. 35 U.S.C. \u00C2\u00A7 282 (1994). Prior to 1982, a party could overcome this presumption with a mere preponderance of the evidence, but in Connell v. Sears, 722 F.2d at 1549 (1983), the Federal Circuit ruled that \"clear and convincing evidence\" of invalidity must be put forth by the patent challenger before the court finds the PTO's decision of validity erroneous. The emergence of the Federal Circuit was not only the most influential patent policy development of the past twenty years, it was the most important ever. See Robert P. Merges, \"One Hundred Years of Solicitude: Intellectual Property Law, 1900-2000\" (2000) 88 Calif. L. Rev. 2187 at 2215: \"Patent law, no less than copyright and trademark, has had to adapt to several major waves of innovation this century. These innovations have produced monumental changes in the fabric of the law. ... [T]he three I consider most important: (1) the 'corporatization' of patent law, following similar trends in industrial research and development (R&D) in the early years of the century; (2) the consolidation of eighty years' worth of doctrinal adjustment in the Patent Act of 1952; and\u00E2\u0080\u0094most significant of all, in many ways\u00E2\u0080\u0094(3) the creation of the Court of Appeals for the Federal Circuit in 1982.\" Merges, supra note 34, at 822, with data derived from D.R. Dunner, \"The United States Court of Appeals for the Federal Circuit\u00E2\u0080\u0094Its First Three Years\" (1985) 13 AIPLA Q.J. 185 at 187-88 (Tables 1-3); reconfirmed by D.R. Dunner, et al, \"Statistical Look at the Federal Circuit's Patent Decisions, 1982-1994\" (1995) 5 Fed. Cir. Bar J. 151. Baum, supra note 26, at 760. From 1935 to 1974, the validity rate for the federal appeals courts (including the CCPA) was approximately 30%. Id. Part II. Strengthening & Extending the U.S. Patent System 14 ruling of patent validity was upheld on appeal approximately 62% of the time. Between 1982 and 1990, however, the Federal Circuit upheld 90% of such cases; conversely, where pre-CAFC appeals courts had overturned only 12% of district court findings of patent invalidity, the Federal Circuit reversed over half.41 A more recent study from 1998 estimated that the overall probability a patent might eventually be invalidated now stands at 54%.42 Ten years earlier, expectations of patent invalidations hovered at around 70%.43 Perhaps more importantly, patent lawyers now believe the court favors patentees, and presumably, advise their clients accordingly. Thus, the new pro-patent sentiment of the courts was communicated to the potential applicants to the patent system\u00E2\u0080\u0094the high-tech companies of the late 1980s. The Federal Circuit strengthened patents in a couple of other important ways. Not only is it far more likely to have a the validity of a patent hold up in court, the Federal Circuit has also made the consequences of infringing a U.S. patent much more serious. It is now far easier for a patent holder to obtain a preliminary injunction halting the sale of the alleged infringing product, and, if successful, to win damages that far exceed those awarded prior to 1982. The Polaroid v. Kodak battle of the early 1980s is often singled out as exemplifying this trend. After claiming an astronomical $12 billion for Kodak's infringement of its instant photography patents, Polaroid managed to win a judgment for the still-stratospheric sum of $837 million, including $233 million in lost profits, $204 million in royalties, and another $436 million in interest.44 Other 4 0 Coolley, \"The Status of Obviousness and How to Assert It as a Defense\" (1994) 76 J. Pat. & Trademark Off. Soc'y 625. 41 Id. 4 2 John Allison & Mark Lemley, \"Empirical Evidence on the Validity of Litigated Patents\" (1998) 26 AIPLA Q.J. 185. 4 3 See Cecil B. Quillen, \"Innovation and the United States Patent System Today,\" Presentation to the Continuing Legal Education Institute, Antitrust and Intellectual Property: Practice and Policy Issues for the 1990's(\9 October 1992). 4 4 Kodak was also permanently banned from competing in the instant-camera business. Polaroid Corp. v. Eastman Kodak Co., 16U.S.P.Q.2d (BNA) 1481 (D.C Mass. 1990), corrected at 17 U.S.P.Q.2d (BNA) 1711 (D.C Mass. Part II. Strengthening & Extending the U. S. Patent System 15 enormous awards include $1.2 billion from Honeywell to Litton for ring laser gyroscopes, over $1 billion thus far to the Lemelson Trust for bar coding, $212 million from Steelcase to Hayworth for movable office panels, and $171 million from Mobil to Exxon for a plastic catalyst.45 1991). That judgment was appealed, and the 15-year fight was eventually settled for $925 million. See \"Kodak Settles with Polaroid\" N. Y. Times (16 July 1991) at D8. See F.M. Schcrer, Presentation to the National Academies' Board on Science, Technology and Economic Policy (STEP), Intellectual Property Rights: How Far Should They Be Extended? (l-i February 2000: Washington, DC). See also Gary Forger, \"Common sense patents\" Modern Materials Handling (1 May 2000) at 7. Part III. Concerning the Effects on Innovation 16 III. C O N C E R N I N G T H E E F F E C T S O N I N N O V A T I O N The growth of the United States patent system has been meteoric. Once a fragile right, valuable to few and enforced by even fewer, the patent now stands colossal, dominating entirely new areas of scientific endeavour, directing previously non-proprietary research settings and demanding adherence by formerly hostile regions of the world. Meanwhile, major modifications to U.S. patent doctrine have made patents more valid, and more valuable, than ever. But have these changes been beneficial? What effect has this expansion and fortification had on technological innovation, the raison d'etre of the patent system? Has all this patent reform actually encouraged more innovative activity, as predicted by the traditional incentive theory? At first glance, it seems the answer is yes. As shown in Figure 1, on the next page, the number of patents has exploded over the last twenty years. More or less constant at 40 to SO thousand per year for most of the century, and actually declining through the 1970s, utility patents granted to American inventors totaled over 85,000 in the year 2000, more than double the number issued in any year from 1979 to 1986.46 Applications, meanwhile, rose to nearly triple what they were in 1984, and seem to be on a lunar trajectory. Research and development spending has followed a similar path. As illustrated in Figure 2, R&D expenditures\u00E2\u0080\u0094measured in inflation-adjusted dollars\u00E2\u0080\u0094have never been higher. Though R&D spending by the public sector, while rising through the 1980s, has been declining through the 90s, spending by U.S. industry has doubled since 1980. Note that Figure 1 shows only domestic patents\u00E2\u0080\u0094that is, those patents applied for and granted to U.S. companies, institutions and individuals, typically about 55-60% percent of total patenting. As shown in Figure 3, on page 61, the overall trend is the same\u00E2\u0080\u0094from 315,015 applications, total patent grants numbered 175,983 in the year 2000, including 85,072 domestic utility patents, 72,425 foreign utility patents, 17,414 design patents, 548 plant patents, and 524 reissues. Part III. Concerning the Effects on Innovation 1960 1965 1970 1975 1980 198S 1990 1995 2000 Figure 1 : U .S . Patent Activity (Domestic) 4 19SS 1960 1965 1970 1975 1980 1985 1990 1995 2000 Figure 2: U . S . Research & Development Spending 4 47 Sources: United States Patent & Trademark Office, http://www.uspto.gov/; Jaffe, supra note 24, at 54, Fig. 1; Robert M. Hunt, \"Patent Reform: A Mixed Blessing for the U.S. Economy?\" (Dec. 1999) Federal Reserve Bank Philadelphia Business Review at 17, Fig. 2. 48 Sources: National Science Foundation, Division of Science Resources Studies, http://www.nsf.gov/sbe/srs/; Jaffe, ibid, at 54, Fig. 1; Hunt, ibid, at 16, Fig. 1. Part III. Concerning the Effects on Innovation 18 Thus, the increases in both patenting and R&D activity seem to be prima facie evidence that a stronger patent system does indeed foster innovation. Shortly after the policy changes of the early 1980s, patenting by U.S. inventors began to skyrocket. At about the same time, spending on research and development started its upward march. If innovation can be measured by patent activity or R&D spending, then surely the policy changes of the last twenty years have been successful. On closer examination of Figures 1 and 2, however, we see that while the surge in patenting began in the mid-1980s, the increase in R&D spending started in the mid-70s. This suggests that while the increase in patenting may be due to changes in the patent system, it seems that the increase in research and development spending\u00E2\u0080\u0094or at least the beginning of it\u00E2\u0080\u0094is unrelated to those changes. Furthermore, the decade-long lag between the upturn in R&D and the upturn in patenting is too long to be consistent with evidence showing that there is little or no lag between changes in R&D spending and changes in patent application rates.49 One might expect that it might take some time for money spent on research and development to manifest itself in the form of patentable inventions, but not that long. This suggests that the increase in R&D spending is not correlated to the increase in patenting. We need to take a closer look at what actually caused the surge in U.S. patenting, and how exactly patents are related to research and development spending. Bronwyn Hall, Zvi Griliches & Jerry Hausman, \"Patents and R&D: Is There a Lag?\" (1986) 27 Intl. Econ. Rev. 26S. Part III. Concerning the Effects on Innovation 19 A. The Pa ten t Surge The most thorough analysis of U.S. patenting trends to date is that of Samuel Kortum and Josh Lerner.50 In their 1997 National Bureau of Economic Research paper, the authors consider four possible explanations for the unprecedented surge in patenting. The first is what they call the \"friendly court\" hypothesis\u00E2\u0080\u0094that, as discussed above, the doctrinal modifications made by the Federal Circuit have made patents more valid, and hence increased the so-called propensity to patent of U.S. inventors. Not all inventions are patented, nor even all potentially-patentable ones. Inventors balance the time and expense of the patent process\u00E2\u0080\u0094and the possible loss of secrecy that results from patent publication\u00E2\u0080\u0094against the protection that a patent potentially affords to the invention. The \"friendly court\" of the Federal Circuit, posit Kortum and. Lerner, has increased the probability of success in obtaining and defending a patent and has therefore shifted that balance, causing a higher proportion of potentially-patentable inventions to be patented. A variant of the friendly court hypothesis sees the changes of the 1980s as an example of \"regulatory capture\"\u00E2\u0080\u0094that the large firms which dominated the research and patenting processes in the United States managed to induce the government to change the rules in their favour, making patenting easier. Under this theory, the increase in patenting would be dominated by those firms, taking advantage of the new environment they had created. The third hypothesis is that a shift in technological opportunity has made more invention possible. This \"fertile technology\" theory points particularly to the new areas of biotechnology and information technologies as the source of the patent surge. Surely, opening these new frontiers to patenting must account for some of the growth in patenting. Kortum & Lerner, \"Stronger Protection or Technological Revolution: What is Behind the Recent Surge in Patenting?\" (1997) National Bureau of Economic Research Working Paper No. W6204. Part III. Concerning the Effects on Innovation 20 The fourth, and perhaps least likely, supposition considered by Kortum and Lerner is that the process of research and invention has become more productive, either by changes in research technology, such as the application of computers and advanced statistics to problems that had previously been handled heuristically, or by changes in the management of the research process, including an emphasis on the kind of applied research that is likely to generate patents. To distinguish among these hypotheses, the authors look at the patent data in several ways. First, they posit that the \"friendly court\" theory suggests that U.S. patents should have become more valuable for all inventors, both domestic and foreign. Thus, we should see an increase in patenting in the United States by foreigners roughly equal to the increase by domestic inventors. This hypothesis also suggests an increase in the propensity to patent in the United States only, and not in other countries\u00E2\u0080\u0094where the patent systems have remained relatively unchanged\u00E2\u0080\u0094and so U.S. inventors should not necessarily have increased their rate of patenting in other countries. Neither of these implications, however, is borne out by the data. Patenting by foreign entities in the U.S. has increased, but it was increasing rapidly before the 1980s, and there was no sign of acceleration after the creation of the Federal Circuit in 1982. Furthermore, patenting by American inventors has increased abroad. An analysis of the United States and other industrialized countries as both \"sources\" and \"destinations\" for patent applications showed the U.S. has become a significantly greater source, but not a more important destination. Surprisingly, this clearly suggests that the brave new era of patenting ushered in by the Federal Circuit is not the primary explanation for the surge in U.S. patenting activity. Kortum and Lerner also show that the data doesn't support the \"fertile technology\" or the \"regulatory capture\" hypotheses either. After a comprehensive analysis based on the international patent classification (IPC) system, they found that biotechnology patents grew from 3% of all patents in 1969 to about 6% in 1991, and that software increased from 4% to almost 7% over the same period. Part III. Concerning the Effects on Innovation 21 The authors found, however, that this kind of growth wasn't confined just to the new technologies\u00E2\u0080\u0094 approximately 70% of all patent classes exhibited an increased rate of patenting. Similarly, the increase was not confined to those firms which had traditionally dominated the patenting process. On the contrary, the growth in patenting was skewed toward those firms which previously had relatively few patents. Through this process of elimination, Kortum and Lerner conclude that there must have been an increase in the productivity of the research process, at least in terms of its ability to produce the kind of innovation that leads to patents. The most unlikely of the causes set forth\u00E2\u0080\u0094the so-called \"managerial process\" theory\u00E2\u0080\u0094seems to be the primary reason for the patent surge of the past decade. The authors go on to ask whether an increase in the productivity of research is consistent with the observed behaviour of research and development expenditure, and develop a simple model of research-driven economic growth in which an unexpected permanent increase in the productivity of R&D is predicted to lead to a transitory increase in R&D as well as patents. The difficulty with reconciling this prediction with the data is that, as noted above, R&D began its increase so much sooner that patenting. If research productivity began to increase in the 1970s, there should have been an effect on patenting rates within a few years. Thus, the inference of Kortum and Lerner's careful analysis is that the explanation for the patent surge lies largely outside the patent system, despite the coincidence of timing with momentous changes in patent policy. Much of the increase in patenting can be associated with an increase in real R&D spending that began much earlier, and much of the rest can be attributed to \"improvements in the management or automation of the innovation process itself. \"S 1 The strengthening of the patent system did, presumably, reinforce these initiatives\u00E2\u0080\u0094i.e., it is possible that the R&D boom would not have been so large or lasted so long without this reinforcement\u00E2\u0080\u0094but there appears to be little or no 3 1 /t/at 4 . Part III. Concerning the Effects on Innovation 22 empirical evidence that strengthening and extending U.S. intellectual property protection resulted in any significant changes to the innovation process. Studies by other economists have also failed to find evidence that the strengthening of patent rights during the 1980s directly stimulated industrial spending in research and development. Robert Hunt of the Federal Reserve Bank of Philadelphia recently examined the market valuation of R&D investments made by a dozen U.S. semiconductors companies from 1976 to 1994 and found that while there was a significant increase in the market value of those investments, it occurred after 1989\u00E2\u0080\u0094more than five years after the significance of the changes in the patent system were widely known.52 A more recent study by James Bessen and Eric Maskin at the Massachusetts Institute of Technology showed that rather than increasing, real R&D intensity (defined as R&D spending relative to sales) in information technology-related industries\u00E2\u0080\u0094including semiconductors, telecommunications and computers\u00E2\u0080\u0094has fallen steadily since about 1982.53 B. The Patent Paradox This lack of correlation between stronger patent rights and increased R&D spending is reinforced by survey evidence. A 1983 study by Levin, Klevorick, Nelson and Winter at Yale University asked R&D managers across various manufacturing industries about the effectiveness of different mechanisms for appropriating the returns to research and development.54 Somewhat surprisingly, they found that the 5 2 See Hunt, supra note 47, at 24. 5 3 Bessen & Maskin, \"Sequential Innovation, Patents, and Imitation\" (January 2000) Working Paper No. 00-01, Department of Economics, Massachusetts Institute of Technology. Similarly, a study by economist Bronwyn Hall at the University of California, Berkeley found that the market value of R&D investment made by about 1000 publicly-traded companies increased through most of the 1970s, began to decline after 1983, and then rose again around 1990. Hall, \"Industrial Research During the 1980s: Did the Rate of Return Fall?\" Brookings Papers: Microeconomics (1993) 289. 5 4 Levin, Klevorick, Nelson & Winter, \"Appropriating the Returns from Industrial Research and Development\" (1987) 3 Brookings Papers on Economic Activity 783. The authors surveyed 1,478 R&D labs in 43 industries. In an earlier study, Schercr had also found that respondents did not consider patents to be particularly important in Part III. Concerning the Effects on Innovation 23 protection from imitation offered by patents was seen as a relatively unimportant source of incentives for technological innovation.55 Instead, lead time and quality of service were perceived as far more useful tools for recouping R&D expenditures, as shown below. Table 1: Effectiveness Rankings of Various Appropriability Mechanisms 1983 Yale Survey 1994 Carnegie Mellon Survey Number of Industries Number of Industries Ranking Mechanism as: Ranking Mechanism as: Mechanism 1st 2nd 3rd 4th Mechanism 1st 2nd 3rd 4th 5th Patents S 6 20 13 Patents 14 14 7 8 1 Secrecy 0 0 19 25 Secrecy 3 4 5 12 20 Lead Time 17 21 6 0 Lead Time 22 6 10 4 2 Sales & Service 24 19 1 0 Sales & Service 3 9 11 15 6 Manufacturing 4 14 13 7 6 Table 1 also shows the results of a 1994 follow-up study administered by Cohen, Nelson and Walsh at Carnegie Mellon University.56 Although patents were ranked higher, first-mover advantages were still seen to be the most effective way to appropriate returns on investment. Thus the so-called \"patent paradox\"\u00E2\u0080\u0094that despite the fact that firms are taking out so many more patents, R&D managers do not perceive patents to be much more effective than they did over a decade before. What could explain this paradox? Cohen and his co-authors suggest that the reconciliation of the exponential rise in patenting and the modest increase in their perceived effectiveness may lie in the multiple ways that firms use patents. In particular, their survey shows that in addition to protecting the returns to specific R&D decision making\u00E2\u0080\u0094except when patent lawyers prepared the responses. F.M. Scherer, The Economic Effects of Compulsory Licensing(N.Y.U., 1977) at 36-37. 5 5 See Scherer, supra note 45. 5 6 Cohen, Nelson & Walsh, \"Protecting Their Intellectual Assets: Appropriability Conditions and Why U.S. Manufacturing Firms Patent (or Not)\" (2000) National Bureau of Economic Research Working Paper No. W7552. Part III. Concerning the Effects on Innovation 24 inventions, firms also use patents to block their competitors' products, as bargaining chips in cross-licensing negotiations, and to prevent or defend against infringement suits. It is possible, says Gohen, that the respondents didn't consider these benefits of patents when evaluating the effectiveness of patents in protecting their investments in innovation. C. Semiconductor Patenting These suppositions are supported by the work of Bronwyn Hall and Rosemarie Ham-Ziedonis. Intrigued by the Levin and Cohen studies, they embarked on a detailed examination of the one industry where the contrast between the increase in patenting and perceived unimportance of patenting as a means of securing returns to innovation was the strongest\u00E2\u0080\u0094the semiconductor industry.57 Indeed, in both surveys, R&D managers in semiconductor companies consistently reported that patents were among the least effective mechanisms for appropriating returns to R&D investments. Driven by the rapid pace of technological change and short product life-cycles, semiconductor firms tend to rely more heavily on lead time, secrecy, and manufacturing or design capabilities than any other industry. And yet the surge in patenting was highest in the semiconductor industry. Even accounting for increases in research and development spending, Hall and Ham-Ziedonis found that the propensity for semiconductor firms to patent essentially doubled between 1982 and 1992.58 In order to illuminate the factors underpinning the semiconductor patent surge and the effects of stronger patent rights on the innovative activities of semiconductor firms, the two researchers 5 7 Hall & Ham-Ziedonis, \"The Patent Paradox Revisited: An Empirical Study of Patenting in the U.S. Semiconductor Industry, 1979-95\" (2000) Working Paper, Department of Economics, University of California, Berkeley. 5 8 Realizing that increases in research and development spending are correlated to patenting rates, Hall and Ham-Ziedonis normalized the patenting rate as a function of R&D expenditures to find patent yields (or propensities), i.e., the number of successful patent applications per million dollars of R&D spent (in constant 1992 dollars). Doing the same for other industries, they found that the patent surge in semiconductors was exceptional\u00E2\u0080\u0094while patent yields in computing and electronics rose slightly, yields for manufacturing as a whole was fairly stagnant and those for pharmaceuticals actually declined over the same period. Part III. Concerning the Effects on Innovation 25 employed a two-fold research strategy. First, they conducted interviews with intellectual property managers and executives from several U.S. semiconductor firms\u00E2\u0080\u0094both traditional manufacturers and the newer, specialized design firms, most of whom entered the industry during the period associated with stronger patent rights. Second, they performed a quantitative econometric analysis of a much larger sample of such firms, including the compilation of a detailed database of their patent portfolios from the period 1975 to 1998. Like Kortum and Lerner, Hall and Ham-Ziedonis found that much of the increase in patenting was due to managerial improvements. But whereas Kortum and Lerner concluded that the patent surge was caused by managerial change in the research and development process, Hall and Ham-Ziedohis found that the change was far more focused on managing the patenting process itself. The . large-scale manufacturers they talked to spoke of \"ramping up\" their patent portfolios and \"harvesting\" latent inventions to add to their stock of patents. One firm they interviewed had gone from owning a total of 30 patents in 1990 to filing over 300 patents in one year, with an internal goal in place to \"own 1000 patents by the year 2000.\"59 Many firms had established \"patent advocacy committees\" that work with the company's engineers, making sure they identified more of the patentable inventions and getting those inventions to the patent office, while minimizing the amount of time it took for engineers to be involved in the patent process. Not all patenting, however, could be traced to managerial improvements. Two events, both indirectly related to the more favourable judicial treatment of U .S. patent rights\u00E2\u0080\u0094the \"friendly court\"\u00E2\u0080\u0094were frequently mentioned as having played a pivotal role in reshaping the patent strategies of the firms they interviewed.60 First and foremost, interviewees emphasized the important \"demonstration effect\" of the huge damages award in the Polaroid v. Kodak case, and the realization 5 9 Id. at 11 60 The authors note that Kortum and Lerner's use of aggregate (international vs. domestic), rather than firm-level, data may have led them to dismiss the effects of the \"pro-patent\" (\"friendly-court\") shift too quickly, at least in the context of the semiconductor industry where the surge in patenting was most dramatic. Id. at 9. Part III. Concerning the Effects on Innovation 26 that courts were willing to take an aggressive stance against infringement by halting\u00E2\u0080\u0094either temporarily or permanently\u00E2\u0080\u0094production that used infringed technologies. For a high-volume fabrication facility costing billions of dollars, a preliminary injunction is prohibitively expensive, perhaps costing millions of dollars per week in lost revenue. A second widely-cited reason was referred to as the \"TI\" effect. During mid-1980s, Texas Instruments had successfully asserted various patents on integrated circuits and manufacturing methods and was thereafter able to charge higher royalty rates for their patents. Other companies such as AT&T, IBM and Motorola quickly followed suit, and now most large-scale firms use their patent portfolios as assets with which to generate significant revenues. Revenues from patent licenses in the United States soared from about $ 15 billion in 1990 to more than $100 billion in 1998.61 IBM alone took in well over $1 billion'from licensing in 1999, and received a record 2,756 new patents.62 Citing these effects, interviewees at the manufacturing firms noted that patents had become far more important as bargaining chips in negotiations with other firms. Because of the \"systems\" nature of semiconductor technology, it is virtually impossible to make products that do not incorporate the technology of other firms, and so everyone needs cross-licensing arrangements, both to gain entry into the market and to avoid infringement suits. One industry executive estimated that without a patent portfolio with which to bargain, a new manufacturer would need to spend $100 to $200 million just to license what are now considered basic manufacturing principles.63 Although such agreements had always been fairly common in the industry,64 the strengthening of patent protection in the 1980s brought new attention to the value of patent portfolios. 6 1 James Gleick, \"Patently Absurd\" N. V. Times (12 March 2000) at 6-44. 6 2 Id. 63 Id. at 13\u00E2\u0080\u0094citing Michael Rostoker, former vice president of strategic alliances at LSI Logic. 6 4 In 1981, Intel Chief Counsel Roger S. Borovoy was quoted as saying, \"In the electronics industry, patents are of no value whatsoever in spurring research and development. We use them because we have to. You can't be the only holdout against the angry hordes or else you pay everyone.\" See \"The Patent is Expiring as a Spur to Innovation\" Part III. Concerning the Effects on Innovation 27 Even where cross-licensing agreements were not in place beforehand, it appears that firms look at their portfolios as a kind of post facto protection from infringement suits. James Rose, former in-house counsel at both Sun Microsystems and programmable logic-chip maker Altera, tells the story of getting a phone call from an engineer asking, \"You know, I would like to use this circuit or technology, but I think some other company has a patent on it. What should we do?\" Surprisingly, most of the time Mr. Rose would answer, \"Don't worry about it.\"65 Though this may seem like an irresponsible response from a patent counsel, his stance seems justified. The fact that in the semiconductor field there are literally tens of thousands of patents and that \"any given product probably infringes 10, 20, 30, maybe more of those patents. If a company were to spend all of its time trying to figure out and design around patents, they would never get a product out the door.\" The other factor, explained Mr. Rose, was that \"because we were filing and protecting our own innovations [but] also filing in areas that were outside of our scope of technology . . . the hope or the expectation was, if this particular company did find out about the infringement, we would hopefully be able to go to our own portfolio\" and find a patent that the other company is infringing. That, he said, \"gave us the freedom to design the best product and not necessarily worry about other companies' patents.\"66 To that end, he spoke of the quest by semiconductor firms to build patent portfolios as a \"frenzy to file\" or an \"arms race\" with a \"mutually-assured destruction (MAD) mentality.\"67 The story was quite different for the smaller firms in the study, however. Hall and Ham-Ziedonis found that the period of increased patenting coincided with a time of significant entry by new firms into the industry, many of which were so-called \"fab-less\" manufacturers who design computer Business Week(\ 1 May 1981) at 44E. Apparently the ineffectiveness of patents as recouping returns on investment was a commonplace phenomenon even then. 6 5 James Rose, Presentation to the National Academics' Board on Science, Technology and Economic Policy (STEP), Intellectual Property Rights: How Far Should They Be Extended? {2-3 February 2000: Washington, DC). 6 6 Id. Part III. Concerning the Effects on Innovation 28 chips but contract the manufacturing to others. This separation of design from manufacture, and the resulting reduction in the barriers to entry into chip design, would not have been possible if the design firms could not protect their creations from appropriation by the large manufacturers. For such companies, patents also play a critical role in attracting capital, specifically venture capital. A number of design firms said that in order to secure funding, venture capitalists demanded that they protect their inventions with strong patent rights that would stand up in court. James Rose also underscored the importance of patents to VCs, saying that patents are particularly necessary in Silicon Valley, where there is so much start-up activity. Venture capitalists almost always look to see if a new company has patent applications, and frequently hire patent attorneys to determine if they are likely to issue. Investment decisions can often be based almost entirely on the strength of a start-up's intellectual property. D. Software Patenting The software industry shares many of the problems found in semiconductors. Martin Konopken, in-house counsel at the software firm Autodesk, Inc., the fourth largest PC software company in the-world, echoes the comments of James Rose when he lays out what he sees as the main problem with software patents. The typical software program has thousands of functions and hundreds of important features. With over 100,000 software patents now issued, Konopken estimates that large products, such as Autodesk's AutoCAD program, may infringe up to ten thousand of those. Searching all of them, he says, is impossible, especially since software patents are notoriously difficult to interpret, often requiring patent counsel to examine their prosecution histories. Add to that the typical 18-Id. Part III. Concerning the Effects on Innovation 29 month product cycle and the pressure to get a product out the door as soon as possible, and the rule for software companies, says Mr. Konopken, \"is not to search, because you simply can't.\"68 And so the end result is what he calls a game of \"gotcha\"\u00E2\u0080\u0094in which certain companies use patent litigation as a way to make money. They have a patent, and they look around for someone in the software industry who has inadvertently infringed it. Once a product is on the market, it's very hard to modify, since customers get used to certain features and don't want them changed. So most software companies fight the infringement suit. \"If you don't either slam dunk the other side, or settle, then you go to court,\" says Konopken, \"where a patent case costs $2 million to fight, plus your engineers' time.. It is really a game of extortion in many cases.\" For Mr. Konopken and others, the advent of software patenting has not been a good thing: \"I speak with my fellow in-house counsel in the software industry frequently. There is an amazing degree of unanimity about software patents. We all hate them.\"69 Unlike semiconductors, however, software has not always been patentable. Yet while the twenty-five year-long debate over software patentability is finally over, the effects of that controversy still linger. By focusing everyone's attention on the subject-matter issue, and giving lip service to the idea that software per se was unpatentable well into the 1990s, a climate was created in which the actual patenting of software was largely ignored.70 This has had a number of important consequences unique to the software industry. Martin Konopken, Presentation to the National Academies' Board on Science, Technology and Economic Policy (STEP), Intellectual Property Rights: Mow Ear Should They Be Extended? (2-3 February 2000: Washington, DC). Id. As discussed above, though the Supreme Court's decision in Diehr opened the door to software patenting, somewhat creative drafting was required to ensure that the software in question operated in the context of an \"otherwise statutory\" industrial process\u00E2\u0080\u0094see supra note 15. Part III. Concerning the Effects on Innovation 30 First, while volumes have been written about whether software should or should not be patentable,71 there has been a very limited discussion of other validity issues such as obviousness, enablement, or best mode.72 In addition to being novel, useful and nonobvious under sections 102 and 103, a valid patent must meet the disclosure requirements of the first paragraph of section 112: The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.73 As suggested by the emphasis, the above paragraph is generally interpreted as calling for three separate requirements: (1) the description requirement; (2) the enablement requirement; and (3) the best mode requirement. The description requirement requires all of the claims in a patent to be supported by an adequate description of the claimed invention in the patent's specification.74 The enablement requirement places an obligation on the patent applicant to provide a sufficiently clear explanation of the invention that would enable a person having ordinary skill in the art to make and use the invention . Bookshelves verily groan under the weight. One commentator estimates that from 1970-79, over 200 academic articles were devoted to the patenting of computer programs; from 1980-89, there were over SOO; and from 1990-2000, well over 1,000 articles were written. See James P. Chandler, \"Patent Protection of Computer Programs\" (2000) 1 Minn. Intell. Prop. Rev. 2 at4 n.15. For some of the more influential of these, see Pamela A. Samuelson, \"Benson Revisited: The Case Against Patent Protection for Algorithms and Other Computer-Related Inventions\" (1990) 39 Emory L.J. 1025; Richard H. Stern, \"Tales from the Algorithm War: Bensonto Iwahashi, It's Deja Vu All Over Again\" (1991) 18 AIPLA Q.J. 371; Jur Strobos, \"Stalking the Elusive Patentable Software: Are there still Diehror was it just a FlookT (1993) 6 Harv. J. of Law & Tech. 363; Maximilian R. Peterson, \"Now You See It, Now You Don't: Was It a Patentable Machine or an Unpatentable 'Algorithm'? On Principle and Expediency in Current Patent Law Doctrines Relating to Computer-Implemented Inventions\" (1995) 64 Geo. Wash. L. Rev. 90. Discussion came only after the Federal Circuit began deciding such issues. See, e.g., Lockwood v. American Airlines, 107 F.3d 1565 (Fed. Cir. 1997); In re Zurko, 11 F.3d 887 (Fed. Cir. 1997); Fonarv. General Electric Co., 107F.3d 1543 (Fed. Cir. 1997); In re Dossel, 115 F.3d942 (Fed. Cir. 1997). 35 U.S.C. \u00C2\u00A7 112 (1994). The second paragraph of 112 requires that \"the specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.\" Id. This imposes a requirement of definiteness for each claim in the patent\u00E2\u0080\u0094a claim will be considered definite if it reasonably describes the claimed invention to those skilled in the art and if it serves to distinguish the claim over the prior art. See Andrew Corp. v. GabrielElecs., Inc., 847 F.2d 819 at 821 (Fed. Cir. 1988). In rc Wertheim, 541 F.2d 257 at 262 (CCPA 1976). The description requirement usually becomes an issue when a claim is added by the patent applicant at some time after the original filing date of the application Part III. Concerning the Effects on Innovation 31 without undue experimentation.75 The disclosure of section 112 is complete when the inventor reveals the \"best mode\" he or she knows for practicing the invention.76 Disclosure, in a sense, is the price an inventor pays for patent protection. In exchange for its grant of monopoly, the government demands that an inventor reveal the inner workings of his invention so that others can understand it and possibly improve on it. Thus the patent system's classic quid pro quo\u00E2\u0080\u0094disclosure is the quo for which the inventor receives his quid, the monopoly grant. Unfortunately, however, it seems that the level of disclosure in software patents is severely lacking. As Pamela Samuelson, Professor of Law and Information Management at the University of California, Berkeley, has said, \"we are getting lots of quid but no quo.\" Despite years of contact with people in the software industry, she claims that she has \"never talked to anybody who, in the process of development, actually reads patents in order to get informed.\"77 Other problems are more critical. Since software was not officially patentable until the mid-908, the Patent Office only recently began to hire patent examiners qualified in computer software or related fields.78 During the 1980s and early part of the 90s, the lion's share of software patent applications were handled by examiners operating outside their areas of expertise. For similar reasons, the PTO's classification system has been ill-equipped to handle software patents. In the past, patents have tended to be classified according to the field in which the software would ultimately be used\u00E2\u0080\u0094for example, making pizzas or curing rubber\u00E2\u0080\u0094rather than the nature of the software invention itself. This The enablement requirement will be met even if some experimentation is required, so long as that experimentation is not undue. In re Barker, 559 F.2d 588 at 591 (C.C.P.A 1977) The analysis of the best mode requirement involves two questions: (1) did the inventor know of a better mode of carrying out the claimed invention than that disclosed in the specification?; and (2) did the inventor conceal that better mode? Chemcast Corp. v. Arco Indus. Corp., 913 F.2d 923 at 927 (Fed. Cir. 1990). Pamela Samuelson, Presentation to the National Academies' Board on Science, Technology and Economic Policy (STEP), Intellectual Property Rights: How Far Should They Be Extended? {l-l February 2000: Washington, DC). See Scott Thurm, \"A Flood of Web Patents Stirs Dispute Over Tactics\" Wall St. Journal'(9 October 1998) at BI \u00E2\u0080\u0094 asserting that the PTO didn't hire its first examiner with a degree in computer science until 1995. Part III. Concerning the Effects on Innovation 32 has made it much harder for examiners to find what prior art exists for a certain invention, and thus to determine whether it is novel. Under section 102 of the U.S. Patent Act, an inventor may not receive a patent if the invention already exists in the \"prior art\"\u00E2\u0080\u0094that is, if the invention is patented or described in a printed or online publication anywhere in the world before the date on which the inventor files his or her patent application.79 In practically every field that the Patent Office covers, examiners determine whether an invention is new by searching two kinds of computerized databases: the PTO's own Automated Patent System, which tracks the more than six million patents issued since 1790, and the so-called Electronic Information Center, which includes nearly every commercially-available database of scientific literature. For most technologies, this works fine. But unlike more established engineering fields such as semiconductors and even biotechnology, most software inventions are not described in industry publications. Instead, computer science academics commonly save up their best work for big textbook projects, and few publishers have allowed the full text of their books to be incorporated into online databases. And a lot of software prior art might not be written down at all. As the ex-commissioner of patents and trademarks, Bruce Lehman, once admitted, \"We search the patent database, both U.S. and foreign, and search every commercial database. But there are many concepts that have been done which are what I call folklore. They are out there, and people know about them, but we can't find any written documentation. The examination process requires that we have written documentation which we can point to which states a particular fact. Too often we can't find the documentation. Then when the patent is issued, some people say, 'Well, this is well-known, it has been in the industry for 35 U.S.C. \u00C2\u00A7 102 (1994). See also 37 C.F.R. 1.104(a) (1997), requiring the patent examiner to \"make a thorough investigation of the available prior art relating to the subject matter of the claimed invention.\" Simson L. Garfinkel, \"Patently Absurd\" WiredQuly 1994) 104 at 107. Part III. Concerning the Effects on Innovation 33 This is a main criticism of software patents\u00E2\u0080\u0094that the Patent Office has issued (and is issuing) patents which are far from novel. Examples of such patents are commonly pointed out by Gregory Aharonian, a vocal critic of software patents for the past ten years via his Internet Patent News Service. He cites U.S. Patent No. 5,865,827, describing a \"device and method to provide a gateway for the transfer of information between financial markets and customers.\" It was issued on 26 January 1996, despite the fact that a speaker at a computing conference in 1985 had presented a paper on the exact same thing.81 Likewise, on 6 July 1993, the Patent Office issued a patent for an \"Automated Health Benefit Processing System\" whereby the system asks patients a few questions and enters the answers into their records, despite the fact that throughout the 1980s there were yearly conferences with, lengthy proceedings on the very same topic.82 Other patents seem far too broad. Take U.S. Patent No. 5,173,051, issued to Optical Data Corporation of Warren, New Jersey on 22 December 1992, describing a system for \"curriculum planning and publishing\" using a computer and a videodisc player. If enforced, such a patent would make out millions of teachers to be infringers. Or consider Patent No. 5,105,184, the so-called \"Energizer Bunny Patent,\" issued in 1992 to Software Advertising Corp. It covers \"displaying and integrating commercial advertisements with computer software\"\u00E2\u0080\u0094in other words, any advertisement integrated into a screen saver. Another common criticism is that the companies getting all of the patents are not the ones doing the innovating. According to Aharonian, in 1998, IBM held 1200 software patents, far more than the runners-up, Motorola (with 360), Fujitsu (with 330), and Canon (330).83 None of these companies is considered an innovative software maker. Microsoft is next on Aharonian's list (with 8 1 Gregory Aharonian, Presentation to the National Academies' Board on Science, Technology and Economic Policy (STEP), Intellectual Property Rights: How Far Should They Be Extended?'(2-3 February 2000: Washington, DC). 82 Id. 8 3 Aharonian, \"1998 U.S. software patent statistics\" Internet posting (23 July 1999), available on-line at http://www.aful.org/pipermail/patents/1999-July/00004S.html. Part III. Concerning the Effects on Innovation 34 310) but only because they've been playing some serious catch-up\u00E2\u0080\u0094in 1992, they held only 13 patents.84 The companies that collectively constituted the microcomputer revolution, companies such as Borland, Novell, Adobe, Lotus, NeXT, Intel, Apple, Sun, and SGI, have all had relatively weak software patent portfolios. IBM, critics claim, has a very strong patent portfolio only because it has \"patented every single trivial idea every employee ever comes up with, rather than having any great propensity to be truly innovative.\"85 Indeed, IBM even has a patent, number 5,247,661, on a software application to permit employees to automatically document ideas for later patenting.86 And in 1999, nearly 30 years after: the production of the IBM 370, a minicomputer with rotary switches on the front for selecting applications and data entry functionality, IBM got a patent for sticking a rotary switch on the front of a PC\u00E2\u0080\u0094for choosing functions such as Program Manager, Notepad, Calculator, Paintbrush, Solitaire, or Minesweeper:87 E. Business Method Patenting Lately, however, the concerns of the semiconductor and software industries have been overshadowed by the controversy surrounding business method patenting. As mentioned, in the 1998 case of State 8 4 Gordon Irlam & Ross Williams, \"Negative Correlation of Innovation and Software Patents\" (January 1994) Revised version of Appendix D of the League for Programming Freedom's submission to the U.S. Patent Office, available on-line at http://www.base.com/software-patents/scoreboard.html. Apparently, Microsoft hasn't slowed its pace\u00E2\u0080\u0094a 27 November 2001 search of the PTO Web site for patents assigned to Microsoft revealed 1,785 patents (though a portion of those are for hardware inventions). 8 5 Id. 8 6 Id. 8 7 Aharonian, supra note 81. U.S. Patent Number 5,973,666, \"Method and Means for Controlling the Concurrent Execution of a Plurality of Programs on a Computer System.\" The first page of the patent actually shows a dial with 16 settings, from Program Manager (0) to MS-DOS Window (12)\u00E2\u0080\u0094positions 13 to 15 are left blank, presumably for additional applications. Part III. Concerning the Effects on Innovation 35 Street Bank v. Signature Financial Group, the Federal Circuit had not only eliminated what was left of the software debate, it had endorsed the patenting of so-called methods of doing business.88 Signature Financial Group held the patent disputed in that case.89 Directed to a \"Data Processing System for Hub and Spoke Financial Services Configuration,\" it described a data processing system for implementing an investment structure known as a \"Hub and Spoke\" system. This system allowed individual mutual funds (spokes) to pool their assets in an investment portfolio (hub) organized as a partnership. According to the patent, such an investment regime provided the advantageous combination of economies of scale in administering investments coupled with the tax advantages of a partnership.90 Following issuance of the patent, Signature entered into licensing negotiations with a competitor, State Street Bank, that ultimately proved unsuccessful. State Street then brought a declaratory judgment action against Signature, seeking the invalidity of the patent. The district court granted summary judgment in favor of State Street under two alternative grounds.91 The first was based on what the district court termed as the \"mathematical algorithm/physical transformation test,\" or the Freeman-Walter-Abele test.92 The district court found that Signature's patent claim failed this test because there was no physical transformation or reduction taking place, concluding that: 88 State Street Bank & Trust Co. v. Signature Financial Group, Inc., 149 F.3d 1368 (Fed. Cir. 1998). 8 9 United States Patent No. 5,193,056 (9 March 1993). 9 0 Id. 91 State Street Bank & Trust Co. v. Signature Financial Group, Inc., 927 F. Supp. 502 (D. Mass. 1996). 9 2 The test had been developed by the Federal Circuit's predecessor, the Court of Customs and Patent Appeals, following the Supreme Court judgment in Diamond v. Diehr. See supra note 13; In re Abele, 684 F.2d 902 (CCPA 1982); In re Walter, 618 F.2d 758 (CCPA 1980); In re Freeman, 573 F.2d 1237 (CCPA 1978). Under the first part of the test, a patent must recite, either directly or indirectly, a mathematical algorithm. Under the \"physical transformation\" part of the test, the court then must determine whether the claimed invention is applied to or limited by physical elements or process steps, or, in the words of the district court, \"regardless of whether the invention performs mathematical operations, if it transforms or reduces subject matter to a different state or thing, it is statutory under 101.\" State Street, 927 F. Supp. at 51 3. Part III. Concerning the Effects on Innovation 36 At bottom, the invention is an accounting system for a certain type of financial investment vehicle claimed as means for performing a series of mathematical functions. Quite simply, it involves no further physical transformation or reduction than inputting numbers, calculating numbers, outputting numbers, and storing numbers. The same functions could be performed, albeit less efficiently, by an accountant armed with pencil, paper, calculator, and a filing system.93 The district court then buttressed its holding by turning to \"the long-established principle that business 'plans' and 'systems' are not patentable.\"94 After reviewing the case law and treatises supporting this \"business methods exception,\"95 the court reasoned that to allow the patent in question would, in effect, grant Signature a monopoly on the idea of a multi-tiered portfolio-investment structure.96 The Federal Circuit reversed the district court on both the mathematical algorithm exception and the business method exception as to unpatentable subject matter. Writing for a three-judge panel, Judge Giles Rich first addressed the mathematicalalgorithrn exception, finding that the district court erred in applying the inconsistent Freeman-Walter-Abele analysis, explaining that \"after Diehr and i Chakrabarty;the Freeman-Walter-Abele test has little, if any, applicability to determining the presence of statutory subject matter.\"97 Rather, the court asserted that \"whether a claim encompasses statutory Id. at 515. Id. Id. at 515-16. See Loew'sDrive-In Theatres, Inc. v. Park-In Theatres, Inc., 174 F.2d 547 at 552 (1st Cir. 1949); Hotel Sec. Checking Co. v. Lorraine Co., 160 F. 467 at 469 (2d Cir. 1908) (\"A system of transacting business disconnected from the means of carrying out the system is not, within the most liberal interpretation of the term, an art.\"); E.B. Lipscomb, Lipscomb's Walker on Patents 2:17 (3d ed. 1984) at 171. The business methods exception developed as an extension of the prescription on patenting abstract principles. As early as 1868, the Patent Commissioner sensed that \"it is contrary to the spirit of the law . . . to grant patents for methods of book-keeping\": Ex parte Abraham, 1868 Dec. Comm'r Pat. 59. Nineteenth century courts also opined that \"a method of transacting common business\" and \"a mere contract\" were unpatentable: United States Credit Sys. Co. v. American Credit Indemnity Co., S3 F. 818 at 819 (CCS.D.N.Y. 1893); In re Moescr, 27 App. D.C. 307 at 310 (1906) The court judged that \"patenting an accounting system necessary to carry on a certain type of business is tantamount to a patent on the business itself. Because such abstract ideas are not patentable, either as methods of doing business or as mathematical algorithms,\" the patent was held invalid. Id. at 516. State Street Bank, supra note 88. Though Judge Rich was reputedly the principal drafter of the 1952 Patent Act and the longest-sitting federal judge at the time, this statement shows little sense of history\u00E2\u0080\u0094Diehr was, decided in 1981 Part III. Concerning the Effects on Innovation 37 subject matter should not focus on which of the four categories of subject matter a claim is directed to ... but rather, on the essential characteristics of the subject matter, in particular, its practical utility.\"98 Significantly, the Federal Circuit held that the transformation of data could constitute a practical application of an algorithm to qualify for a patent, and explained, \"today, we hold that the transformation of data, representing discrete dollar amounts, by a machine through a series of mathematical calculations into a final share price, constitutes a practical application of a mathematical algorithm, formula, or calculation, because it produces 'a useful, concrete and tangible result'.\"99 Signature's \"Hub and Spoke\" software satisfied the test under section 101 because it produced a final share price, momentarily fixed for recording and reporting purposes, and that final share price was considered a \"useful, concrete, and tangible result.\"100 Despite the presence of a mathematical algorithm, the usefulness of the software rendered it patentable under the Patent Act, reasoned the court, even though the useful result was \"expressed in numbers, such as price, profit, percentage, cost, or loss.\"101 Accordingly, in order for an invention employing an algorithm to be patentable, it is not necessary that the transformation be limited to physical matter. Overruling the district court's holding as to the business method exception, Judge Rich embraced \"the opportunity to lay this ill-conceived exception to rest,\" asserting that \"since the 1952 Patent Act, business methods have been, and should have been, subject to the same legal requirements for patentability as applied to any other process or method.\"102 Moreover, neither the Federal Circuit nor its predecessor court, the CCPA, had ever invoked the business method exception to deem an and Abele was decided in 1982. See John R. Thomas, \"The Post-Industrial Patent System\" (1999) 10 Fordham I.P., Media & Ent. L.J. 3 at 25-26, for other reasons why \"State Streetis a curious opinion on a number of fronts.\" Id. Id. at 1373. Id. at 1375. Part III. Concerning the Effects on Innovation 38 invention unpatentable.103 When this exception has been applied, said the court, it had been after a \"ruling based on some clearer concept of Title 35 or ... on finding a mathematical algorithm.\" The Federal Circuit found additional support in the PTO's position on the business method exception. It contrasted the 1994 Examination Guidelines, which instructed that \"though seemingly within the category of process or method, a method of doing business can be rejected as not being within the statutory classes,\" with the 1996 Guidelines, which omitted this exclusionary language and stated that methods of doing business should be treated like any other process claims.104 The1 court endorsed this interpretation, concluding that \"whether the claims are directed to subject matter within [section] 101 should not turn on whether the claimed subject matter does 'business' instead of something else.\"105 While a limited amount of business method patents had been issued before,106 the PTO experienced a veritable boom in patent applications after State Street, especially for business methods performed over the Internet\u00E2\u0080\u0094so-called \"e-commerce\" patents.107 The then-Commissioner of Patents and Trademarks, Q. Todd Dickinson, confirmed that the Patent Office received 2,600 \"business model\" patent applications and granted 583 of these patents in the 12-month period between October Id. at 1375-76. The court also pointed out that the Second Circuit in Hotel Security Checking v. Lorraine Co. did not rely on the business method exception,to invalidate the patent at issue in that case, but invalidated the patent because the invention lacked the requisite \"novelty.\" Id. at 1376-77, citing MPEP 706.03(a) (1994); Examination Guidelines, 61 Fed. Reg. 7478 at 7479 (1996). Id. Indeed, Signature Financial got one. In late 1998, one source reckoned that despite the business methods exception, there may have been over 2,000 patents for financial products already in existence. Jaret Seiberg, \"Docket: Ruling Threatens Banks with Patent Lawsuits\" American Banker (2 September 1998) at 3. Another source estimated in 1999 that more than 1,000 patents had been issued for software-based inventions in the areas of business management, finance and accounting. Paul. A. Beck, \"State Street Bank Case Causes Shock Waves in Banking and Financial Industry\" (1999) 147 Pitts. L.J. 7 at 9. See Tony V. Pezzano, \"State Street Court Case Opens News World for Tech Patents\" American Banker (18 January 1999) at 2. State Street's ringing endorsement of software patents caused a boom is computer-related patents generally\u00E2\u0080\u0094during 1999 it was reported that there had been a 45% increase in approvals for data processing and computer related inventions, and numerous sources report a drastic increase in the number of applications: John T. Acquino, \"Patently Permissive: PTO Filings Up After Ruling Expands Protection for Business and Net Software\" (1999) 85 A.B.A.J. 30. Part III. Concerning the Effects on Innovation 39 1 1998 and September 1999.1 0 8 The most notorious of these was U.S. Patent No. 5,794,207, entitled \"Method and Apparatus for a Cryptographically Assisted Commercial Network System Designed to Facilitate Buyer-Driven Purchase Offers.\" Issued less than one month after State Street to Priceline.com, the patent protects that company's on-line reverse-auction bidding system, whereby you can \"name your own price\" on airline tickets, home mortgages, hotel rooms, and automobiles.109 Upon receiving the patent, the Priceline's Chairman and Chief Executive Officer, Jay Walker, trumpeted that e-commerce patents are a tremendously positive next step for all U.S. companies involved in creating Internet-based applications. ... Traditionally, patents have been the bedrock on which inventors built long-term, thriving businesses. E-commerce is no-exception. The PTO's actions support our nation's policy that cyberspace innovators should be encouraged and rewarded for advancing America's lead in Internet-based applications.110 : ; Others, including a reader of Forbes magazine, were less enthusiastic: \"Cool! Jay Walker has apparently patented the 'business method' known as a Dutch auction\u00E2\u0080\u0094a method by which the U.S. Treasury sells hundreds of billions of dollars' worth of securities each year.\"111 This comment highlights the main criticism of business method patents\u00E2\u0080\u0094that the use of a computer to implement well known business practices on the Internet is obvious and should not result in a patent grant.112 1 0 8 W. Scott Petty, \"Business Model Patents: Here Today, Gone Tomorrow?\" Intel/. Prop. Today (May 2000) at 40. In 1997, only 31 pure business-method patents had been granted. William C. Smith, \"Patent This!\" (March 2001) 87 A.B.A.J. 49 at 51. 1 0 9 Peter H. Lewis, \"Web Concern Gets Patent for its Model of Business\" N. Y. Times(\ 1 August 1998) at D l . Two other patents protecting Internet technology received attention at about the same time: CyberGold, for its method of paying consumers to look at advertisements on the Internet, and NetDelivery, for its electronic delivery management (\"EDM\") technology. Id. 1 1 0 Don Sussis, \"Autumn Signals A Harvest of Opportunity On (and Off) the Net\" Silicon Alley Reporter (October 1998) at 88. 1 , 1 Byron L. Winn, \"Readers Say\" Forbes(31 May 1999) at 18. 1 1 2 James Gleick, \"Patently Absurd\" A^ . Y. Times (12 March 2000 ) at 6-44. Lawrence Lessig, \"Online Patents: Leave Them Pending\" Wall St. Journal (Ii March 2000) at A22. Tim Watts, \"The E-boom Causes A Troubling Test Of Patents\" Business Review Weekly (5 May 2000) at 42. Part III. Concerning the Effects on Innovation 40 Rather than fostering innovation, claimed critics, business model patents stifle creativity and represent an improper grant of a monopoly within the open framework of the Internet. Even more troubling, companies were litigating these patents very aggressively. In October 1999, Pricelihe sued Microsoft for using a reverse-auction method on its Expedia travel site, asserting that it turned over confidential and technical data to Microsoft during an eight-month period when the companies were negotiating a strategic partnership.113 Under a January 2000 settlement, Expedia was allowed to continue its price-matching service if it agreed to pay a royalty to Priceline.114 Amazon.com also made headlines in October 1999 with its patent infringement suit against-Barnes & Noble. On September 28, Amazon.com founder and CEO Jeff Bezos had received a patent for Amazon's \"1 - Click\" ordering system, whereby a customer who has already given his or his pertinent ordering information (i.e., name, address, and credit card number) is able to purchase an item over the Internet with a single \"click\" of the mouse button.115 As the Christmas shopping season approached, Barnesandnoble.com, the website for Barnes & Noble, began offering an \"Express Lane\" one-click ordering service much like Amazon's. A suit was filed, and after a five-day hearing the Federal District Court for the Western District of Washington issued an injunction barring Barnesandnoble.com from 1 1 3 Priceline asked the court to award both actual and punitive damages as well as a permanent injunction based on patent infringement and state unfair practice claims. Del Jones, \"Businesses Battle over Intellectual Property: Courts Choked with Lawsuits to Protect Ideas\u00E2\u0080\u0094And Profits\" USA Today {2 August 2000) at IB. According to Jay Walker, when the deal fell through, Microsoft Chairman Bill Gates said that \"many companies were suing Microsoft for patent infringement and that Microsoft had no intention of allowing patent rights to stand in its way,\" and that \"Priceline could, in effect, get in line.\" For its part, Microsoft denied receiving any confidential information and publicly reiterated its strong belief in information protection. Id. 1 1 4 See Smith, supra note 108. 1 1 5 U.S. Patent No. 5,960,411, \"Method and System for Placing a Purchase Order Via a Communications Network.\" According to Bezos, who was voted Time's Man of the Year for 1999, the invention was designed to facilitate on-line purchasing by overcoming a customer's tendency to abandon \"shopping carts\" prior to checkout. See Noel D. Humphreys, \"Patent Protection: Encouraging or Discouraging Innovation?\" (2000) 22 Pennsylvania Lawyer SA. This author submits, however, that such a one-click system probably surprises shoppers into inadvertent purchases. Amazon won another controversial method patent in February of 2000, Patent No. 6,029,141, which protects its \"affiliate\" system that lets websites collect a cut of sales when they refer a customer to another site. Id. Part UI. Concerning the Effects on Innovation 41 offering the one-click feature pending outcome of the suit.\"6 In her decision, Judge Pechman invoked a variety of the traditional justifications for intellectual property rights: The public is served by innovation on the Internet and in electronic commerce, particularly now while it is still developing rapidly. Competition to provide unique, effective and enjoyable consumer experiences will lead to innovation and diversity in on-line commerce. On the other hand, innovation will be discouraged if competitors are permitted a free ride on each other's patented inventions. Protection of intellectual property rights in innovations will foster greater competition and innovation. ... The public has a strong interest in the enforcement of intellectual property rights. The purpose of the patent system is to reward inventors and provide incentives for further innovation by preventing others from exploiting their work. Encouraging Amazon.com to continue to innovate\u00E2\u0080\u0094and forcing competitors to come up with their own new ideas\u00E2\u0080\u0094unquestionably best serves the public interest.\"7 With only weeks until Christmas, Barnesandnoble.com was forced to design around Amazon.corn's patent by creating a \"double-click\" ordering system in which two clicks of the mouse were required instead of one. The case caused a firestorm of controversy, fanned in large part by maverick Internet entrepreneur Tim O'Reilly, president of computer book publisher O'Reilly & Associates, Inc.\"8 O'Reilly was joined in his cause by MacArthur \"genius\" grant winner Richard Stallman, an ardent foe of software patents and founder of the Boston-based Free Software Foundation. Stallman was so disgusted with Amazon.com that he called for a boycott of the Internet bookseller. Although O'Reilly stopped short of supporting the boycott, he did encourage each visitor to his Web site to contact Jeff 1 , 6 Amazon.com v. Barnesandnoble.com, 73 F. Supp. 2d 1228 (W.D. Wash. 1999). 117 Id. at 1248-49. ns \"yjjg frcc_wheeling development environment that has given us both the Internet and open source tools ... has demonstrated convincingly that there is enormous power in an open, shared platform where ideas are given away as the foundation for further innovations,\" O'Reilly said. \"So far, entrepreneurs, stock market investors, and computer users have all benefited enormously from the open, non-proprietary approach historically used for developing Internet technologies.\" Bleys W. Rose, \"Patent debate hot topic on web\" The Press Democrat [Santa Rosa, CA] (10 March 2000) at E l . Part III. Concerning the Effects on Innovation 42 Bezos with concerns about the Amazon patent. Within a day or two, more than 10,000 had responded.119 Surprisingly, after participating in an online conversation with O'Reilly, Bezos started criticizing the patent system himself. In a three-page reform proposal posted on the Amazon.com website, Bezos agreed that \"meaningful (perhaps radical) patent reform\" was needed. \"I now believe,\" he said, that \"it's possible that the current rules governing business method and software patents could end up harming all of us.\"120 Among his suggestions for restructuring the patent system was a reduction of the patent term\u00E2\u0080\u0094now 20 years from the time of application\u00E2\u0080\u0094to three to live years for software and business-method patents, and the introduction of a one-month public comment period before a patent could issue.121 Despite all this, the \"1-Click\" litigation led to an array of new lawsuits and encouraged hundreds of U.S. companies to apply to patent their Web site features. In a move that caused another uproar from the tech community, British Telecom announced in June 2000 that it would consider enforcing its patent for \"hyperlinking,\" the method by which certain words or phrases can be cross-referenced with the click of a button.122 On August 9, Mercata announced that it had been issued a patent for its group-buying technology, and that it was going \"to probe for possible infringements by competitors.\"123 Similarly, on August 28, DE Technologies announced that it was about to be issued a broad patent covering \"a process for carrying out an international transaction ... using computer-to-\" 9 Victoria Slind-Flor, \"Bar reacts to Bezos patent reform plan: Adopting his changes would upset good system, they say\" The National Law Journal (21 March 2000). 1 2 0 R. Scott MacKendrick, \"Amazon 1 -Click decision\u00E2\u0080\u0094relief or invitation to litigate?\" The Lawyer's Weekly (13 April 2001). 1 2 1 Slind-Flor, supra note 119. Bezos even suggested that the shorter term for business-method patents be applied retroactively, to his own patents as well. I'll discuss the merits of this reform in Section V.B.2, infra. 1 2 2 SeeAndrewJ. Frackman & Robert M. Stern, \"An update from the patent wars\" New York Law Journal (29 November 2000) at 1. 1 2 3 See \"Mercata Wins Patent on Buying Technology\" N. Y. Times (9 August 2000). Such a broad business method patent conceivably could prevent any company from offering a group-buying Internet service. Part III. Concerning the Effects on Innovation 43 computer communication\" and that it was embarking on a program of enforcement against any company conducting international trade transactions over the Internet.124 Other, less-publicized patents were also causing a stir. After State Street, it seemed that you could get a patent on a method of doing practically anything, no matter how trivial. Inventor Edward Pechter of Valencia, Calif., for example, was granted a patent for a \"method of bra size determination by direct measurement of the breast.\"125 In pertinent part, Pechter's pectoral procedure gauges cup size \"by directly measuring with the tape the circumference of each unclothed breast from the beginning of the breast mound at one side laterally to the parasternal area medially.\" On 30 November i 1999, Kevin and George Repper were granted patent number 5,993,336 for a method of swinging a tennis racquet. As described in the abstract, their technique consists of wearing kneepads and swatting the ball \"either while the covered knee is on the tennis court surface or just prior to the knee contacting the tennis court surface.\" This \"innovation,\" the patent claim concludes, \"enables a player to successfully return balls that otherwise are out of effective stroking reach.\" Even before State Street, though, the PTO was issuing very curious patents. In 1995, for example, after finding that \"cats are not characteristically disposed toward voluntary aerobic exercise,\" Kevin Amiss and Martin Abbot of Virginia patented a \"method for inducing cats to exercise [by flashing a laser beam] onto the floor or wall or other opaque surface in the vicinity of the cat, then moving the laser so as to cause the bright pattern of light to move in an irregular way fascinating to cats and to any other animal with a chase instinct.\"126 Since April Fools' Day 1997, Dale D. Miller has had the rights to Patent No. 5,616,089, covering a \"putting method in which the golfer controls the speed of the putt See William M. Bulkeley, \"A Billion-Dollar Patent?: Software Developer Seeking To Protect Process Using Internet For Foreign Trade\" Wall St. Journal(28 August 2000) at Bl; \"Patent Application Could Evolve into Trouble for E-Commerce\" Wall St. Journal (li August 2000) at Bl . U.S. Patent No. 5,965,809 (12 October 1999). U.S. Patent No. 5,443,036 (22 August 1995). Part III. Concerning the Effects on Innovation 44 and the direction of the putt primarily with the golfer's dominant throwing hand, yet uses the golfer's nondominant hand to maintain the blade of the putter stable.\" And consider the following claim: A method for remodeling an existing building, said method comprising: cataloging design ideas that utilize predetermined building products; presenting the design ideas to a client; allowing the client to select a design idea...; [and] preparing a visual image ... representing the building remodeled with the design idea selected by the client.127 Wholly divorced from particular artifacts, it broadly appropriates an architectural services technique.128 Or consider Patent No. 5,761,857, which was recently issued to two architects for their configuration of residential housing. In place of a traditional hallway connecting the several rooms and apartments, the patentees came up with the ground-breaking idea of applying staircases external to the structure.129 Other patents are troubling for different reasons. Patent No. 6,025,810, issued in February: 2000, is for an antenna that sends signals faster than the speed of light\u00E2\u0080\u0094an impossible feat, if you believe decades of science based on Einstein's theory of relativity.130 Similarly impossible is Patent No. 5,533,051, issued in 1996, which covers a technique that purports to compress any data set by at least one bit without loss of information\u00E2\u0080\u0094a process that, if done recursively, could shrink the Encyclopedia Britannica to a single word from which the original could be flawlessly reconstructed.131 More worrisome than the issuance of such trivial and ridiculous patents is the effect that the new patents will have on the business environment in general. As pointed out by Leo Raskind, Professor of Law at Brooklyn University, business method patents were \"thrust into a vibrant, established process of competitive commercial rivalry, a process that has traditionally been governed by emulation and by customary practices. An added perverse result of this intrusion is the incentive for 1 2 7 U.S. Patent No. 5,668,736 (16 September 1997). 1 2 8 Thomas, supra note 97, at 30-31. 1 2 5 See Teresa Riordan, \"Architects Debate Concepts Behind Housing Design\" N. Y. Times (19 July 1999) at Cl 1. 1 3 0 Sec Philip E. Ross, \"Patently Absurd Technology and gamesmanship has overwhelmed the U.S. Patent Office. How to fix it?\" Forbes (29 May 2000) at 180. 131 Id. Part III. Concerning the Effects on Innovation 4S some entrepreneurs to become collectors of patent royalties, rather than to continue as active participants in the marketplace. If the boom in business method patents continues at its accelerating pace, the so-called superhighway of electronic commerce could be partially converted into a toll road/' 1 3 2 Leo J. Raskind, \"\"The State Street Bank Decision: The Bad Business of Unlimited Patent Protection for Methods of Doing Business\" (1999) 10 Fordham L P , Media & Ent. L.J. 61 at 66. Part IV. Identifying the Issues 46 IV. I D E N T I F Y I N G T H E ISSUES We've looked at the evidence from the various industries of the new economy, but it would be useful to now organize the problems along issue lines. What exactly are the issues facing the patent system, and in what areas are the problems most critical? Before we do that, however, we should step back and review the theory behind the patent system, and how it is intended to foster innovation. A. Patent Theory The origin of the United States patent system dates back to birth of the nation itself. Drafted in 1787, Article I, Section 8, Clause 8 of the U.S. Constitution authorizes the creation of both the patent system and the copyright system: The Congress shall have Power . . . To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries.133 The language of the clause hints at the two main philosophical justifications for granting patent and copyright monopolies\u00E2\u0080\u0094natural rights theory and utilitarianism. Generally attributable to the writings of John Locke, the theory of natural rights holds that individuals have an inherent right to the fruits of Art. I, \u00C2\u00A7 8, cl. 8. While contained in a single clause, the Patent and Copyright Clause of the U.S. Constitution is commonly treated as two distinct provisions, each with a distinct grant of authority and distinct subject matter: (a) For Patents: \"The Congress shall have Power ... To promote the Progress of ... useful Arts, by securing for limited Times to ... Inventors the exclusive Right to their ... Discoveries.\" (b) For Copyrights: \"The Congress shall have Power ... To promote the Progress of Science..., by securing for limited Times to Authors ... the exclusive Right to their ... Writings...\" See also Donald S. Chisum, Chisum on Patents(1998) at GI-23: '\"Useful Arts' is synonymous with the contemporary concept of technological arts and is contrasted with 'Science' the promotion of which is the object of the copyright system.\" Part IV. Identifying the Issues 47 their labour.134 Like other systems of property rights, an intellectual property system should protect individual rights to inventions so that others cannot appropriate them. Utilitarianism, on the other hand, proposes that the main justification for the grant of monopoly is a matter of public welfare. By promoting scientific progress, intellectual property systems serve all of society.135 In analyzing exactly how the patent system fosters innovation, traditional economic theory has emphasized the incentive function of patents.136 It usually costs more, in terms of effort and money, to discover something new than it does to duplicate someone else's discovery. Inventors may work on their discoveries for a variety of reasons, but so long as one of the motivations is the prospect of financial reward, inventors will be concerned about the possibility that others will imitate their discoveries. If an invention can be imitated quickly, the inventor will soon be forced to compete with those who did not incur the development costs that he did. This competition will reduce, or even eliminate, the profits an inventor can earn from his discovery, and so he may decide not to pursue it. By providing a temporary monopoly under which an inventor can recoup his initial expenditures, patents therefore provide an important incentive to engage in innovative activity. John Locke, Second Treatise of Government (1698). Adam Smith also strongly asserted the property right of an individual to his labour: \"The property which every man has in his own labour, as it is the original foundation of all other property, so it is the most sacred and inviolable.\" Adam Smith, An Inquiry into the Nature and Causes of the Wealth of Nations, Book 1, Chapter 1 (1776). While the two theories seem inherently opposed, the Framers of the Constitution saw little conflict. In the one comment regarding the intellectual property clause, James Madison appears to reconcile both justifications: \"The utility of this power will scarcely be questioned. The copyright of authors have been solemnly adjudged in Great Britain to be a right of common law. The right to useful inventions seems with equal reason to belong to the inventors. The public good fully coincides in both cases with the claims of individuals.\" The Federalist No. 43 (Clinton Rossiter ed, 1961) at 271-72 . See also John Shepard Wiley, Jr., \"Copyright at the School of Patent\" (1991) 58 U. Chi. L. Rev. 119 at 140\u00E2\u0080\u0094contending that the labour-desert theory and utilitarian perspective \"tend to converge as a practical matter, because law governing innovation policy usually must treat creators fairly in order to give them the incentives to act in ways that benefit consumers.\" Nevertheless, the tension between the rights of the individual and the public good has animated the copyright and patent debates ever since. See Frederic M. Scherer, IndustrialMarket Structure andEconomic Performance (Rand McNally, 1970) at 379-99; John S. McGee, \"Patent Exploitation: Some Economic and Legal Problems\" (1966) 9 J. L. & Econ. 135; Dan Usher, \"The Welfare Economics of Invention\" (1964) 31 Economica 279; Richard R. Nelson, \"The Economics of Invention: A Survey of the Literature\" (1959) 32 J. Bus. 101. Part IV. Identifying the Issues 48 Thus, in the words of the U.S. Supreme Court, the patent laws were enacted to promote innovation by offering a limited right of exclusion \"as an incentive to inventors to risk the often enormous costs in terms of time, research and development.\"137 Congress intended that \"the productive effort thereby fostered [would] have a positive effect on society through the introduction of new products and processes of manufacture into the economy, and the emanations by way of increased employment and better lives for our citizens.\"138 Economists have identified three main ways in which society benefits from the grant of patent rights.139 First, patent ownership enables a successful inventor to recoup sunk research costs, thereby, providing incentives for the initial investment in discovering the invention. Second, by allowing the owner of a new invention to exclude the \"free-riders\" who did not bear the costs of invention, patents encourage the inventor to develop that invention commercially. Finally, by requiring that an inventor publicly disclose the inner workings of his invention, patents foster the diffusion of new technology, and may lead to other inventions from competitors attempting to invent around the patent.140 Yet while it provides substantial benefits, the patent system also creates significant costs. Foremost among these are the output restrictions and correspondingly higher prices associated with monopolization. Since patent holders have a monopoly over the patented technology, they can charge a higher price than they could in a competitive market. In most cases, there will be some consumers Kewanee Oil Co. v. Bicron Corp., 416 U.S. 470 at 480 (1974). Id. Sec Scherer, supra note 136, at 380: \"Governments have elected to grant exclusive patent rights on inventions for three main reasons: to promote invention, to encourage the development and commercial utilization of inventions, and to encourage inventors to disclose their inventions to the public.\" The three different benefits have been called the \"incentive to invent\" theory, the \"incentive to innovate\" theory (innovation being the development of an invention into commercial form), and the \"incentive to disclosure\" theory. See Rebecca S. Eisenberg, \"Patents and the Progress of Science: Exclusive Rights and Experimental Use\" (1989) 56 U. Chi. L. Rev. 1017. Again, in the words of the Supreme Court, \"such additions to the general store of knowledge are of such importance to the public weal that the Federal Government is willing to pay the high price of seventeen years of exclusive use for its disclosure, which disclosure, it is assumed, will stimulate ideas and the eventual development of further significant advances.\" Kewanee Oil, supra note 137, at 481. Not only does disclosure promote greater knowledge, it eliminates secrecy which might lead to wasteful duplicative research. See Martin J. Adelman, \"Property Rights Theory and Patent Antitrust: The Role of Compulsory Licensing\" (1977) 52 N.Y.U. L. Rev. 977. Part IV. Identifying the Issues 49 willing to buy the product at the competitive price, but unwilling to pay the higher price charged by the patent holder.141 Other inefficiencies may result from distortions of economic activity. Patents may force competitors to waste time and effort finding redundant solutions to a problem that has already been solved.142 Competing firms hoping to make patentable inventions ahead of their rivals may spend too much money trying to develop inventions quickly, when the same result could be achieved at lower social cost through a less accelerated research effort.143 Companies may also divert resources from research in fields where patent protection is weak or unavailable to research in areas that are more likely to yield profitable patent monopolies.144 Thus, under the traditional incentive theory\u00E2\u0080\u0094also called the \"reward theory\"\u00E2\u0080\u0094patents are viewed as a classic tradeoff between the benefits of greater innovation and the costs of monopoly power. Later economic theory, however, has proposed that monopolies are actually conducive to innovation, and look at the patent model slightly differently. In the 1930s, Joseph Schumpeter argued that in a dynamic model of the capitalist system, monopoly conditions may promote innovation and growth more effectively than competition.145 He reasoned that in the rapidly-changing conditions of a capitalist economy, investment in innovation 1 4 1 The negative impact of monopoly power was forefront in the mind of the author of the original Patent Act. Thomas Jefferson, like other Americans, had an instinctive aversion to monopolies. After all, it was a monopoly on tea that sparked the Revolution and Jefferson did not favor an equivalent form of monopoly under the new government. His abhorrence of monopoly extended initially to patents as well. From France, he wrote to James Madison in July 1788 urging a Bill of Rights provision restricting monopoly, and against the argument that limited monopoly might serve to incite \"ingenuity,\" he argued forcefully that \"the benefit even of limited monopolies is too doubtful to be opposed to that of their general suppression.\" He subsequently modified his position, authored the Patent Act of 1793, and directed the patent board for several years. See Thomas P. Burke, \"Software Patent Protection: Debugging the Current System\" (1994) 69 Notre Dame L. Rev. 1115 at 1130. 1 4 2 See Donald F. Turner, \"The Patent System and Competitive Policy\" (1969) 44 N.Y.U. L. Rev. 449 at 455. 1 4 3 See Richard Posner, Economic Analysis of Law (Little, Brown, 1977) at 54 ; Yoram Barzel, \"The Optimal Timing of Innovations\" (1968) 50 Rev. Econ. & Stat. 348. 1 4 4 See Fritz Machlup, \"An Economic Review of the Patent System,\" Subcomm. on Patents, Trademarks, and Copyrights of the Senate Comm. on the Judiciary, Study No. 15, 85th Cong, 2d Sess. (GPO, 1958) at 45-50. 1 4 3 Joseph Schumpeter, Capitalism, Socialism and Democracy (Harper & Row 3d ed, 1950) at 87-106. Part IV. Identifying the Issues SO requires some sort of hedge against losses. Protection from competition also allows firms \"to gain the time and space for further developments.\"146 Finally, and perhaps most important, the prospect of earning more than an ordinary return permits innovators to secure the financial backing of capitalists and to bid productive resources away from their current uses.147 A monopoly position secured through patent protection thus may increase rather than restrict the use of known technologies by facilitating the commercial introduction of such technologies by innovating firms. In the 1970s, Edmund Kitch built on Schumpeter's work to develop what he called the \"prospect theory\" of patent protection.148 Kitch argued that granting broad patent rights in the early stages of innovation promotes efficiency in the further development of promising technological prospects, much like the award of exclusive mineral claims promotes efficiency in the exploration of natural resources. In his view, a new process or apparatus would more likely be commercialized by the inventor who had broad patent protection by insulating such patentee from competitive rivalry during the early stages of the commercial development of new technology. It would also allow the inventor to better coordinate its activities with those of potential imitators to reduce inefficient duplication of inventive effort. Broadest scope, said Kitch, should be afforded to \"pioneer patents\"\u00E2\u0080\u0094those patents which represented important advances and offered the greatest prospect of generating significant ancillary technology. Though many economists were skeptical of Kitch's prospect theory of patents,149 it appears to have influenced the policy debate at the time, lending credence to Congress' new mandate to better Id. at 89. ' Joseph Schumpeter, The Theory of Economic Development (Transaction reprint, 1983) at 67-74 . Edmund W. Kitch, \"The Nature and Function of the Patent System\" (1977) 20 J. L. & Econ. 265. The term \"prospect theory\" highlighted the analogy between the functions of patent monopolies and awards of exclusive mineral claims in government-owned lands in the American West. See Scherer, supra note 54, at 9-10; F.M. Scherer, Industrial Market Structure and Economic Performance (Rand McNally, 2d ed. 1980) at 447 n.30\u00E2\u0080\u0094stating that Kitch's views seem \"little influenced by any concern for reality.\" Kitch's views depend at least in part on an empirical assumption that many inventors acquire patent protection far in advance of the appearance of any commercial product. Empirical evidence contradicts Kitch on this point, Part IV. Iden tifying the Issues 51 exploit the fruits of basic research, and giving additional support for the strengthening the patent rights that occurred in the 1980s. In particular, it has been suggested that a tacit endorsement of the prospect theory of patents by the Federal Circuit led to its reduction of patentability standards and elevation of patent validity\u00E2\u0080\u0094making patents stronger and easier to get.150 B. Patent Scope Patent scope, is the basic measurement of the value of the patent right. At the most general level, patent scope\u00E2\u0080\u0094sometimes also referred to a patent breadth\u00E2\u0080\u0094can be described as the region of technology space from which a patentee may exclude others from operating. Clearly, the broader the patent is, the more valuable it is. It seems logical, therefore, that a patent system which confers broader scope makes inventions more valuable, and thereby provides greater innovation incentives. The problem, of course, is that a researcher has to worry about producing an invention that will be judged to infringe someone else's patent. A system which allows broader patents makes this more likely, and therefore makes research riskier and less valuable.151 Due to this tension, analysis of the consequences of patent scope involves more than just the classic tradeoff between the objective of encouraging innovation at the cost of creating monopoly power. The rate of innovation does not necessarily increase as patent scope increases, because of the constraints that possible infringement places on an inventor's expectation of return. however. Barkev Sanders, in a study of assigned patents issued in 1938, 1948, and 1952, found that of the estimated 10 percent of patented inventions ever put to commercial use, about 40 percent were first put to use before the patent application was filed, about 50 percent were first put to use while the application was pending, and only about 10 percent were first put to use after issuance of the patent. Barkev Sanders, \"Speedy Entry of Patented Inventions into Commercial Use\" (1962) 6 Pat., Trademark & Copyright J. of Research & Educ. 87. See Kevin Rhodes, \"The Federal Circuit's Patent Nonobviousness Standards: Theoretical Perspectives on Recent Doctrinal Changes\" (1991) 85 Nw. U.L. Rev. 1051. The infringing inventor may still be able to practice his invention by securing a license from the holder of the so-called \"blocking patent\"\u00E2\u0080\u0094but that, of course, increases costs. Part IV. Identifying the Issues 52 1. THEORETICAL ANALYSES OF PATENT SCOPE It is useful to distinguish between three types of patent scope issues, based on the relationship between a patented invention and others which might infringe it. The first situation is where the potentially infringing invention is developed independently of the patented invention. The second situation is that of \"cumulative invention,\" in which the potentially infringing invention builds upon the patented invention. The third situation is that of inventions which are research tools, used primarily for the specific purpose of developing other inventions. In the traditional context of independent, non-cumulative inventions, the issue of patent scope is viewed in terms of the tradeoff between providing monopoly profits to the inventor to stimulate innovation versus the losses in market efficiency resulting from that monopoly power.1 5 2 This tradeoff is similar to the one associated with granting a certain duration of patent protection, and thus the problem is to choose a combination of breadth and duration that minimizes welfare loss while still providing an adequate level of innovation incentive.153 Within this framework, a generalized increase in patent breadth or scope, although it may result in decreased efficiencies due to increased monopoly power, unambiguously increases the rate of innovation because it does not affect the incentives of subsequent, potentially infringing inventors\u00E2\u0080\u0094inventions are completely independent. The case of non-independent, cumulative invention\u00E2\u0080\u0094in which subsequent inventors stand metaphorically on the shoulders of the giants who came before them\u00E2\u0080\u0094has received considerable theoretical attention as of late.154 Edmund Kitch viewed this is a problem of optimal coordination 1 5 2 See M. Waterson, \"The Economics of Product Patents\" (1990) 80 Amer. Econ. Rev. 860; Paul Klemperer, \"How Broad Should the Scope of Patent Protection Be?\" (1990) 21 Rand J. of Econ. 113. 1 5 3 See Richard Gilbert & Carl Shapiro, \"Optimum Patent Length and Breadth\" (1990) 21 Rand J. of Econ. 106. 1 5 4 See Suzanne Scotchmer, \"Standing on the Shoulder of Giants: Cumulative Innovation and the Patent Law\" (1991) S J. of Econ. Persp. 29; Jerry Green & Suzanne Scotchmer, \"On the Division of Profit in Sequential Innovation\" (1995) 26 Rand J. of Econ. 20; Howard F. Chang, \"Patent Scope, Antitrust Policy, and Cumulative Invention\" Part IV. Identifying the Issues S3 among different researchers working on related technologies.1SS In the absence of coordination, there will be a wasteful duplication of effort, and possibly over-investment of research and development resources, as firms seek to beat each other to important results. As discussed above, Kitch proposed to grant broad patent rights to a pioneering inventor early in the development of a line of technology would ensure a more orderly development of that technology. To the extent that other inventors have ideas or capabilities that contribute to the advancement of the technology, the pioneering inventor would have an incentive to include them in the development, via cross-licensing or other contractual arrangements. Later work on cumulative invention incorporated the incentives of the potential follow-on inventors into the economic models. For researchers such as O'Donoghue, Scotchmer and Green, the question of scope can be characterized in terms of the magnitude of the improvement that an invention must represent before it will be granted a patent of its own, and/or held to infringe the patent of a previous inventor.156 Their line of research accorded with Kitch's view that broad patent protection should be afforded to the initial invention in a cumulative development line. Scotchmer even proposes that \"second-generation\" products should not be patentable at all. 1 5 7 If the first and second innovators,'. she argues, can negotiate the terms of a licensing agreement before the second inventor sinks any of its-research investments, the first inventor will have the incentive to license its technology to the.second whenever it is optimal to do so, under terms that do not prevent the development of the second invention. Her approach, however, presumes that the trajectory of innovation is known in advance. (199S) 26 Rand J. of Econ. 34; Ted O'Donoghue, \"A Patentability Requirement for Sequential Innovation\" (1998) 29 Rand J. of Econ. 654. The phrase \"standing of the shoulders of giants\" is generally attributable to Sir Isaac Newton. See Robert K. Merton, On The Shoulders of Giants (1965)\u00E2\u0080\u0094quoting a letter by Sir Isaac Newton. Kitch, supra note 148. O'Donoghue, supra note 154; Scotchmer & Green, supra note 154. Suzanne Scotchmer, \"Protecting Early Innovators: Should Second-Generation Products Be Patentable?\" (1996) 27 RandJ. of Econ. 322. Part IV. Identifying the Issues 54 As emphasized by evolutionary economics,158 it is more often the case\u00E2\u0080\u0094particularly early in the development of a new line of technology\u00E2\u0080\u0094that no one knows which directions of improvement are desirable, or even possible, and therefore such ex an te licensing would be unlikely.159 More recently, Hopenhayn and Mitchell explored the implications of the fact that inventions differ in the extent to which they spawn \"fertile\" lines of subsequent inventions.160 Ignoring the kinds of ex ante agreements that Scotchmer uses to ensure that infringing second-generation products will still be developed, they show that broad patent scope is more costly for inventions that are more fertile, because it may inhibit subsequent developments. On the other hand, it is important to provide good; incentives to develop such fertile inventions to begin with, because they are socially very, valuable. Hopenhayn and Mitchell show that overall innovation incentives can be improved by offering1 patentees a \"menu\" of combinations of patent duration and patent scope. Optimal construction of this menu induces patentees to reveal their private knowledge regarding fertility of their inventions, and thereby achieves a better balance between the incentives of the initial and subsequent inventors than; can be achieved with uniform patent scope. The authors suggest that a mechanism with properties such as theirs could be implemented by allowing patentees to choose different types of patents with different-durations and legal rights.161 The third patent scope situation, that of patented research tools, can be thought of as a special case of cumulative invention in which the initial invention in the sequence has no value except as a. 1 3 8 See Richard R. Nelson & Sidney Winter, An Evolutionary Theory of Economic Change (Harvard Univ. Press, 1982). . 1 3 9 In their historical review of U.S. industry, Merges and Nelson also show that ex ante licensing is unlikely and/or ineffective\u00E2\u0080\u0094see infra, notes 169-177 and accompanying text. 1 6 0 Hugo Hopenhayn & Matthew Mitchell, \"Innovation Fertility and Patent Design\" (1999) National Bureau of Economic Research Working Paper No. W7070. 1 6 1 As an example, they suggest that patentees might be required to accept shorter patent durations in order to receive protection against patent infringement under the doctrine of equivalents. Id. This approach presents difficulties, however\u00E2\u0080\u0094see infra, Section V.B.2. Part IV. Identifying the Issues 55 platform for future innovation.162 Unlike the cumulative invention, the research tool does not typically compete in the marketplace with the products developed using it, and therefore development of the downstream product does not reduce the profit stream of the research tool inventor. Indeed, if the patented tool has no direct commercial market, its owner can profit only to the extent it is used in the development of other inventions. This would seem to make the situation in some sense easier than that of cumulative invention, because the incentives of the research tool inventor and the research tool user are more in line with one another\u00E2\u0080\u0094they both want the downstream product to be manufactured and soldjthe only question is how much the inventor should receive in royalties. Schankerman and Scotchmer recently investigated the enforcement of patents for research tools from the perspective of maximizing the incentives to develop such tools.163 They considered whether it is necessary to grant the patentee the right to an injunction preventing the sale of products developed with the tool, or whether the availability of damages for patent infringement is sufficient. .\u00E2\u0080\u00A2 They showed that the current legal treatment of patent damages as a \"reasonable royalty\" is logically flawed, so that maximizing the incentive to develop research tools requires either that injunctions be available, or.that the damages rule be changed to award to the research tool patentee the profits earned by the infringer.164 2. EMPIRICAL STUDIES OF PATENT SCOPE There has been relatively little empirical analysis of the effects of different degrees of patent scope, primarily because it is very difficult to measure patent scope in a systematic way across large numbers 1 6 2 See Scotchmer, supra note 157; Michael Heller & Rebecca Eisenberg, \"Can Patents Deter Innovation? The Anticommons in Biomedical Research\" Science (1 May 1998) at 698. 1 6 3 Mark Schankerman & Suzanne Scotchmer, \"Damages and Injunctions in the Protection of Proprietary Research Tools\" (1999) National Bureau of Economic Research Working Paper No. W7086. 1 6 4 I'll discuss research tools in more detail in Section V.D, infra. Part IV. Identifying the Issues 56 of patents, and because there are few natural experiments in which different degrees of patent scope can be observed. Josh Lerner has examined whether patents that appear to have relatively broad scope are more valuable to patentees that narrower ones. Looking at biotechnology firms, whose values are closely tied to their intellectual property, he found that firms whose patents span more International Patent Classes (IPCs) are valued more highly by venture capitalists.165 While his finding does not directly address the question of patent scope expansion, it provides threshold support for the underlying idea that broader scope increases value and therefore increases innovation incentives. A more recent German study by Harhoff, Scherer and Vopel, however, failed to find that the number of IPCs is related to patent value, as measured in survey responses by patent owners.166 One of the few natural experiments on patent scope occurred in the late 1980s in Japan. Prior to 1988, the Japanese patent system essentially allowed only one claim per patent. A complex invention, or one with many distinct applications, had to be covered by many separate patents, all of which had to non-overlapping\u00E2\u0080\u0094a patentee would not be granted a patent whose claim covered any of the material of another patent, even if it was his own. 1 6 7 In 1988, Japan converted to a system much like the U.S. system, in which a single patent can have multiple, overlapping claims. In their study of the Japanese reforms, Sakakibara and Branstetter argue, based on discussions with Japanese companies and patent experts, that in many cases it is not possible to protect a complicated invention using a series of independent claims. Even the most complete set of independent claims leaves \"holes\" in technology space that can only be covered with overlapping claims. Therefore, the change to a multi-claim system effectively increased patent scope by permitting more effective protection of these inventions. 1 6 5 Josh Lerner, \"The Importance of Patent Scope: An Empirical Analysis\" (1994) 25 Rand J. of Econ. 319. 1 6 6 Dietmar Harhoff, F.M. Scherer & Katrin Vogel, \"Citations, Family Size, Opposition and the Value of Patent Rights\u00E2\u0080\u0094Evidence from Germany\" (1999) National Bureau of Economic Research Working Paper No. W7108. 1 6 7 The non-overlapping nature of the pre-1988 Japanese system caused it to be referred to as the \"sashimi\" system, in reference to the thinly-sliced sushi dish. Mariko Sakakibara & Lee Branstetter, \"Do Stronger Patents Induce More Innovation? Evidence from the 1988 Japanese Patent Law Reforms\" (1999) NBER Working Paper No. W7066. Part IV. Identifying the Issues 57 Sakakibara and Branstetter hypothesize, based on the standard innovation incentive model, that such an increase in patent scope would increase the return to inventive activity, an effect-which should be observable in two ways. First, research and development spending by Japanese firms should have risen. Second, Japanese firms should have produced more inventions, and since the incentives created by the U.S. patent system did not change in 1988, this should have lead to an increase in patenting by Japanese firms in the U.S. The data, however, shows that neither of these occurred. Although the late 1980s were a period in which both Japanese R&D and Japanese patenting in the U.S. were rising, there was no measurable increase in their rates of growth around the time the policy change occurred.168 It is possible that despite the impression of the managers interviewed for the study, the ability to file overlapping claims does not result in a significant increase in patent scope, or that there was an effect, but that it was too difficult to see in data series that are changing rapidly for other reasons. Otherwise, their results suggest that changes in patent scope do not have significant effects on research incentives. The role of patent scope in fostering innovation in the United States has been examined in detail by Robert Merges and Richard Nelson.1 6 9 Their evidence is not statistical, but rather an . investigation: of the development of several historically-important technologies, and whether, as asserted by Kitch, affording broad patent protection to the \"pioneer\" innovators is in fact desirable. In the case of the electric lighting industry, Merges and Nelson found that Edison's dominant patent on the carbon filament as a light source did not result in rapid commercial development, and conclude that \"the validation of Edison's broad patent slowed the pace of improvements considerably.\"170 They also point out that the acquisition of Edison's patent enabled General Electric to 1 6 8 Sakakibara and Branstetter did find, however, an increase in Japanese R&D spending in the early 1980s. Id. Thus, just as in the United States, the increase in R&D spending preceded rather than followed the change in patent policy, and therefore cannot be considered a result of the policy change. See supra note 49 and accompanying text. 1 6 9 Robert P. Merges & Richard R. Nelson, \"On the Complex Economics of Patent Scope\" (1990) 90 Colum. L. Rev. 839. 170 Id. at 886. Part IV. Identifying the Issues 58 obtain injunctions shutting down competitors, thereby increasing its market share from 40 to 75 percent. General Electric then used its market dominance to limit entry into the industry, .while making just minimal improvements to the light bulb. Merges and Nelson contrast the rate of technological improvement in electric lamps with the more rapid rate of development in other sectors of the electric industry, such as power generation and arc-lighting, where they found a causal relation between the absence of broad patent protection and the rapid entry of competing firms and accelerated product improvement.171 The authors then turn to the automobile and airplane industries, which, like electric lighting, were also marked by dominant patents at their.inception. The Selden patent on the light-weight internal combustion engine and the Wright patent on the steering and stabilization system of aircraft contained broadly drawn claims covering a variety of embodiments. Again, Merges and Nelson found that these industries did not conform to the theoretical prediction that the dominant patentee would \ consider broad patent protection as an incentive to develop the subject matter. To the contrary, rather than commercializing his patent, Selden sought only to license the technology and collect royalties. The automobile industry then developed through the efforts of manufacturers who paid royalties to his licensing entity. The Wrights, on the other hand, did undertake to produce airplanes and to improve their design, but they refused to license others whose design ideas were at odds with theirs. Even worse, both parties further impeded technological process by engaging in extensive litigation against companies that did not recognize their patents.172 The problems caused by the Selden and Wright patents were compounded when different companies finally managed to patent improvements on their designs. The existence of additional broad patents on complementary technologies meant that different companies could block each other from using key components. In fact, the situation in the aeronautics industry during World War I was so Id. at 887. Part IV. Identifying the Issues 59 serious that the Secretary of the Navy had to intervene with the establishment of an automatic cross-licensing arrangement.173 Merges and Nelson cite the radio industry as a further example of the potential for gridlock in the development of a technology. In that industry, Marconi had acquired the patent rights to the basic two-element vacuum tube, or diode, while AT&T had acquired the fundamental patents on the triode vacuum tube. But because Marconi's diode patent was held to dominate AT&T's triode, and neither party would license the other, no one could commercialize the admittedly revolutionary triode. The conflict was ultimately resolved by the formation of a single entity, the Radio Corporation of America (RCA), which took in the owners of all of the potentially blocking patents as major shareholders in the 174 corporation. \u00E2\u0080\u00A2 < \u00E2\u0080\u00A2 .. \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 , , The development of the semiconductor and computer industries might have been similarly impeded but for a fortunate sequence of circumstances. The first involved AT&T's initial patent rights on the transistor. Although they were also broad enough to have given their holder control over a large \"prospect,\" AT&T was barred,by an antitrust decree from enforcing them. The second instance, involved the parallel inventions of the integrated circuit (by Texas Instruments) and the Planar process for producing them cheaply (by Fairchild Instruments). Both of these companies obtained patents on their own inventions, which meant that each had to license the other to produce integrated circuits \u00E2\u0080\u00A2 effectively. Fortunately, the Department of Defense was the principle purchaser of semiconductors at the time and had a strong interest in seeing these important technologies become broadly available throughout the industry. As a result, the government quickly stepped in to impose a cross-licensing arrangement on the companies.175 172 Id. at 888-89. 173 Id. at 891. 174 Several other companies had important patent positions in radio, including General Electric and Westinghouse. Id. at 892. Id. at 894. Part IV. Identifying the Issues 60 Computer technology also benefited from the lack of broad initial patents. Although pioneering inventors Eckert and Mauchley did file for and receive a patent on their ENIAC computer design, the patent was ruled invalid because of a judgment that the prior art included much of what they claimed. Since this ruling, Merges and Nelson find, patents have played only a very minor role in the computer industry, and where patents are concerned, cross licensing is common. As a result, the pace of technical change has been rapid. 1 7 6 Based on their review of these and other industries, including chemical and biotechnological, the authors state their conclusion as to the effects of broad patent protection as follows: Our, general conclusion is that multiple and competitive sources of invention are socially preferable to a structure where there is only one or a few sources. Public policy, including patent law, ought to encourage inventive rivalry, and not hinder it. ... [A] rivalrous structure surely has its inefficiencies. But such a structure does tend to generate rapid technological progress and seems a better social bet than a regime where only one or a few organizations control the development of a given technology.177 Thus, the' findings of Merges and Nelson, while not definitive, seriously question the-conclusion of Kitch\u00E2\u0080\u0094and implicitly, the later work of Scotchmer, Green, O'Donoghue and others\u00E2\u0080\u0094 that strong property rights for pioneering inventors maximize innovation. They also raise significant doubts about the assumption underlying that conclusion\u00E2\u0080\u0094that different inventors will license their technologies to each other if it is efficient to do so. As Merges and Nelson suspected, uncertainty and disagreement among competitors often make licensing agreements and other coordination mechanisms unlikely and/br ineffective. Id. at 894. Id. at 908. Part IV. Identifying the Issues 61 C. Patent Administration As discussed above, a lot of patents are being issued that are neither novel nor nonobvious, or even useful. Inevitably, much of the blame for the lack of patent quality can be traced to the sheer volume of patent applications. Intuitively, as the number of patent applications skyrockets, and the number of patent examiners stays relatively constant, the amount of review given to each patent goes down\u00E2\u0080\u0094and bad patents slip through. Let's look again at the surge in patents over the last twenty years. As illustrated in Figure 3, patent applications and issuances have nearly tripled in that time. The PTO examined over 315,000 patent applications in 2001, granting nearly 176,000 patents. Growth has been especially fast in software, where the number of grants has gone from hundreds to tens of thousands in just the last ten years. 350,000 300,000 A . 250,000 H 200,000 \u00C2\u00A3 150,000 oo D 100,000 A 50,000 - Applications - Grants - Software Patents 25,000 20,000 15,000 (j 10,000 5,000 1980 1985 1990 1995 2000 Figure 3: U.S. Patent Activity (Total)' Part IV. Identifying the Issues 62 As shown in Table 2, for the subset of Internet and business method patents, the trend is even more dramatic. Table 2: Internet and Business Method Patents Granted179 Internet- Related Year Internet Business Methods Business Methods 1995 163 200 2 1996 379 273 13 1997 659 390 56 1998 2217 739 178 1999 3992 999 399 2000 5745 1054 527 Unfortunately, the patent bureaucracy's budget hasn't kept pace with the exploding workload. From fiscal 1998 to 1999, patent applications rose 11%, but the PTO endured a 16% decline in how much fee revenue it was allowed to spend on operations.180 Part of this reduction was due to the fact that PTO revenues can be diverted to other government programs. In 1999, over $214 million was siphoned off, and in 2001, the federal budget is scheduled to divert $268 million, or nearly one quarter, from the anticipated PTO income of $1,152 billion.1 8 1 And the PTO's hands are tied. While a private business could just raise prices to increase its budget, the PTO may not do so without the consent of Congress.182 8 Sources: USPTO, Internet Patent News Service (Software numbers from 1990 through 1995 are actual, while 1996 through 1999 are estimates; they include the subset Internet and Internet-related business method patents of Table 2). 9 Source: Statistical Mapping of Business Method Patents: A Research Report to the National Academy of Sciences, Board on Science, Technology, and Economic Policy (9 July 2001). 0 Philip E. Ross, \"Patently Absurd Technology and gamesmanship have overwhelmed the U.S. Patent Office. How to fix it?\" Forbes (29 May 2000) at 180. 1 See J.M. Hostcny, \"The Sky Is Falling: Or, Over-Reaction to the Anecdote\" Intell. Prop. Today (May 2000) at 8. 2 See 35 U.S.C. \u00C2\u00A7 41 (1994) (providing the schedule of congressionally mandated PTO fees). Part IV. Identifying the Issues 63 A lack of funding means a lack of examiners. In 1999, 3,200 examiners awarded 161,000 patents, or 50 apiece.183 In software, some 375 examiners pored over 29,000 applications and anointed 13,900 new patents. One software reviewer reportedly approved two hundred patents, an average of four new awards per week, a frequency that defies doing exhaustive research of prior art to weigh a claim's uniqueness.184 \"We would have liked to hire another 750 examiners this year and next,\" said Brigid Quinn, the Patent Office's spokeswoman at the time, \"but budget constraints will probably preclude that.\"185 Even if they had the money, the pay given to PTO examiners is hardly lucrative, ranging from $35,000 to $55,000 a year.186 After 10 or 15 years, an examiner who has reached \"Primary Examiner\" status may earn $70,000 to $80,000.187 In contrast, the average salary for entry-level chemical engineers was $49,150 in 1998, and for management-level engineers, between $95,000 and $120,000.188 Such discrepancies make it extremely difficult for the PTO to attract and retain the skilled technical minds that it needs to examine patents properly. As admitted by the Commissioner himself, most examiners don't see the PTO as a career but rather as a stepping stone to private practice after 3 or 4 years.189 Lack of training is also a problem\u00E2\u0080\u0094junior examiners often complain that they receive very little effective training. This is primarily because the most effective trainers\u00E2\u0080\u0094the senior examiners\u00E2\u0080\u0094 do not have any incentive to spend any time training. The patent compensation system, a combination 1 8 3 Ross, supra note 180. 184 Id. 185 Id. 1 8 6 Jenna Greene, \"Staking A Claim: How State Street Has Spurred A Rush On the PTO\" Legal Times(\0 April 2000) at 14. 1 8 7 Robert P. Merges, \"As Many as Six Impossible Patents Before Breakfast: Property Rights for Business Concepts and Patent System Reform\" (1999) 14 Berkeley Tech. L.J. 577 at 607. 1 8 8 Agnes Shanley, \"You and Your Job: Shifting Career Gears Can Open New Door\" Chemical Engineering (December 1998) at 141. 1 8 9 Statement by PTO Commissioner Q. Todd Dickinson in S. Zeller, Gov. Exec. (February 2000) at 39. Part IV. Identifying the Issues 64 of base salary and bonuses, directs their efforts heavily towards their own examining activities. There are no bonus points for training younger examiners. But as Brian Kahin, Director of the Center for Information Policy in Washington, D . C , points out, there may be more important institutional problems with the PTO. Although the agency is charged merely with the administration of the patent and trademark systems, he submits that it has been taking a much more active role, and now sees itself as a kind of crusader for stronger and more widely available patent protection. Consider the goals laid out in its Corporate Plan \u00E2\u0080\u0094 2000: Strategic Goal: Play a leadership role in intellectual property rights policy . . . Performance Goal: Help protect, promote and expand intellectual property systems 1 throughout the United States and abroad.190 Not only does such a position conflict with the PTO's role as administrator of the patent system, it seriously undermines its job of advising the executive branch on intellectual property rights policy\u00E2\u0080\u0094 especially since that policy strives for a utilitarian balancing of interests between public and private good. Also troublesome is the PTO's mission statement with respect to its Patent Business: \"The primary mission of the Patent Business in to help customers get patents.\"191 This contrasts dramatically with the mission statement of the Trademark Business, which states: \"Our primary mission is to apply the provisions of the Trademark Act of 1946 in the examination and registration of trademarks.\"192 In the PTO Today, an internal circular, the agency even offers testimonials from its customers: \"I am pleased with the customer approach to processing patent applications as opposed to the previous, sometimes adversarial approach.\" 1 9 0 U.S. Patent and Trademark Office, Corporate Plan \u00E2\u0080\u0094 2000zX. 17. 191 Id. at 20. 192 Id. at 38. Part IV. Identifying the Issues 65 \"Examiners seem flexible and interested in working with applicants to allow patentable subject matter to grant. Even more disconcerting is the fact that the PTO budget is directly tied to application and maintenance fees.194 Lower standards for granting patents induce more applications, which generates more fees. And more grants result in more maintenance fees. With this professed self-dedication to its fee-paying customers and expressed expansionist philosophy, asks Kahin, can the PTO be expected to examine patent applications energetically and even-handedly?195 Kahin also argues that the ex parte nature of the patent application process results in marked information asymmetries which further limit the quality of issued patents. A patent examiner deals only with the applicant, who has the motivation and resources to convince the examiner to issue the patent. And because of the nature of the prosecution procedure, the examiner is motivated to allow the patent, unless there is unequivocal prior art. A large part of an examiner's pay comes from bonus points, which are accumulated only for \"dispositions\"\u00E2\u0080\u0094i.e., final allowances or rejections of patents. \"Final\" rejections, however, do not always result in the end of the examination\u00E2\u0080\u0094post-\"final\" action amendments and the like are often permitted\u00E2\u0080\u0094and so the only way for an examiner to earn bonus points with confidence is to allow a patent application.196 The nature of the patent system means that patents are, in effect, rules drafted by the applicant, approved by the PTO, and then privately enforced. The private applicant is allowed considerably more leeway in formulating and expressing the rules (patent claims) than would a public body, so that vocabulary and terminology may vary substantially from patent to patent, especially in the case of non-technical subject matter. Even worse, the primary motivation of the patent applicant is 1 9 3 Commissioner's Page, PTO Today, Vol. 1, No. 1 Qanuary 2000). 1 9 4 35 U.S.C. \u00C2\u00A7 4 2 (1994). 1 9 5 Brian Kahin, Presentation to the National Academies' Board on Science, Technology and Economic Policy (STEP), Intellectual Property Rights: How Far Should They Be Extended?(2-3 February 2000: Washington, DC). 1 9 6 See Brenda Sandburg, \"Patent Applications Flow Freely\" Legal Times(22 February 1999) at 12. Part IV. Identifying the Issues 66 not to make the information accessible and clear but to maximize tactical options in making use of the rules. Greg Aharonian of the Internet Patent News Service points out one of these tactical moves, what he calls one of the \"games patent lawyers play.\"'97 They will send in some patent claims, he says, knowing that the examiner is going to reject them at the first office action, using up the limited art that is sent in. They will then rewrite their claims to get around the first office action, often adding additional claims. The examiner doesn't have any more time to find more prior art or to analyze the new claims,\u00E2\u0080\u00A2\u00E2\u0080\u00A2 since the PTO's production system discourages continued searches\u00E2\u0080\u0094there is strong encouragement to either issue or make a final rejection. Without more time to find new artand.make a rejection, the examiner's hands are tied, and they are pretty much forced to issue the claims atthe second office^action.198 Unfortunately, the PTO has little incentive to worry about these issues. Although it does makes perfunctory attempts at quality control, the basic fact is that once a patent is granted, the PTO rarely sees it again. Any bad patents it issues are then a problem for the private sector and the courts to resolve^ofteh via litigation. D. Pa ten t Litiga tion By all accounts, patent litigation is on the rise. During the period 1978 to 1999, the number of patent suits rose by almost tenfold, with much of the increase occurring during the 1990s, and concentrated in the areas of pharmaceutical, biotechnology, computers, and electronics.'99 Judges in the District of 1 9 7 Greg Aharonian, Presentation to the National Academies' Board on Science, Technology and Economic Policy (STEP), Intellectual Property Rights: How Far Should They Be Extended}'(2-3 February 2000: Washington, DC). 198 Id. 1 9 9 Jean O. Lanjouw & Mark Schankerman, \"Enforcing Intellectual Property Rights\" (2001) Presented to the National Academies' Board on Science, Technology and Economic Policy (STEP), New Research on the Operation and Effects of the Patent System (22 October 2001: Washington, DC) at 2. Part IV. Identifying the Issues 67 Delaware, the district in which most U.S. companies are incorporated, recently conducted a study of trends in civil litigation and found that 13.5% of all new civil cases filed in their district are patent litigation suits.200 The costs of litigation are not insubstantial. The American Intellectual Property Lawyers Association recently estimated that the average cost to litigate a patent in 1999 (for each side) was $2.1 million. 2 0 1 Nor is patent litigation quickly resolved. Estimates suggest that the average District Court patent case lasts 31 months.202 If you add prosecution time and appeals, it now takes approximately 12.3 years from the-filing of a patent application to the final resolution of the lawsuits associated with i t : 2 0 3 All of this has raised fears that patent litigation, or the threat of it, is impeding the ability of firms to conduct research and development effectively. A 1995 study by Josh Lerner showed that small firms avoid R&D areas where the threat of litigation from larger firms is high. 2 0 4 Likewise, in a ; forthcoming.study, Lerner and Jean Lanjouw of Yale University and the Brookings Institute argue that the use of preliminary injunctions by large firms can discourage R&D by small firms.2 0 5 Even if parties can settle their patent disputes without resorting to suits, they posit, the effective threat of litigation will influence settlement terms and, ultimately, the incentives to undertake R&D. 2 0 0 Mark A. Lemley, \"Reconceiving Patents in the Age of Venture Capital\" (2000) 4 J. Small & Emerging Bus. L. 137 at 138. 2 0 1 James Pooley, Presentation to the National Academies' Board on Science, Technology and Economic Policy (STEP), Intellectual Property Rights: How Far Should They Be Extended?'(2-3 February 2000: Washington, DC). 2 0 2 Stuart J.H. Graham, Bronwyn H. Hall, Dietmar Harhoff & David C Mowery, \"Post-Issue Patent 'Quality Control': A Comparative Study of U.S. Patent Re-examinations and European Patent Oppositions\" (2001) Presented to the National Academies' Board on Science, Technology and Economic Policy (STEP), New Research on the Operation and Effects of the Patent System (22 October 2001: Washington, DC) at 8. 2 0 3 Allison & Lemley, supra note 42. Prosecution time itself takes between 2 and 3 years. Id. 2 0 4 Josh Lerner, \"Patenting in the Shadow of Competitors\" (1995) 38 J. L. & Econ. 463. His study also raised serious questions about the ratio of litigation costs to R&D expenditures: \"The patent litigation within USPTO and the federal courts begun in 1991 will lead to total legal expenditures (in 1991 dollars) of about $1 billion, a substantial amount relative to the $3.7 billion spent by U.S. firms on basic research in 1991.\" Id. at 470 2 0 5 Jean O. Lanjouw & Josh Lerner, \"Preliminary Injunctive Relief: Theory and Evidence from Patent Litigation\" (2002) forthcoming in J. L. & Econ. Part IV. Iden tifying the Issues 68 In October 2001, Lerner and Lanjouw completed the most comprehensive study of patent litigation ever undertaken.206 Using the LitAlert database provided by Derwent, a private vendor, they found that 13,625 patent cases were filed during the period 1978-1999. Using only the \"main\" patent identified by each suit, and accounting for the fact that one patent could be involved in more than one suit, their data set included 9,345 patents. Using the PTO's patent classification system, these patents were separated into eight broad technology groups: Chemical, Mechanical, Electronics (excluding computers), Computers (including hardware and software), Biotechnology, Pharmaceutical,; Other Health, and Miscellaneous. They then matched the patent data with information from the U.S. court database organized by the Federal Judicial Center (FJC) indicating the progress or resolution of suits\u00E2\u0080\u0094 i.e., whether the case is settled and at which stage of the proceedings this occurs, whether the case proceeds to trial, and the outcome of the trial. 2 0 7 Looking at the data, Lanjouw and Lerner found that, because of the sharp increase in patent activity, filing rates for patent suits have been roughly constant over the last two decades\u00E2\u0080\u0094i.e., the growth in litigation has been comparable to the growth in patenting itself. The authors found that the total number of patent applications grew by 71% over the period, with patenting nearly tripling in Drugs, Biotechnology, and Medical Instruments, and quadrupling in Computers.208 Table 3, on the next page, shows the filing rates (in cases per thousand) for the various groups, broken down into three sub-periods: 1978-84, 1985-90, and 1991-95. Notice that the filing rates are much lower for Chemical, Mechanical, and Electronics, and much higher for Computers, Biotechnology, and Other Health. Lanjouw & Lerner, supra note 199. The authors set included all cases coded 830 by the FJC. However, since some patent-related cases may have been coded under other categories by the court {e.g., because the patent issue was part of a broader contractual dispute), the authors admit that a small percentage of cases identified in Derwent are not in the FJC database. In addition, the database runs only through the end of 1997. Id. Part IV. Identifying the Issues 69 Table 3: Patent Filing Rates, by Technology Field and Time Period2' Technology Field 1978-84 1985-90 1991-95 Total: 1978-95 Chemical 11.6 10.9 13.0 11.8 Mechanical 17.7 14.5 18.7 16.9 Electronics 16.2 13.1 16.8 15.4 Computers 32.6 21.2 25.9 25.6 Biotechnology 33.3 27.6 25.5 27.9 Pharmaceutical 22.S 18.9 24.3 22.2 Other Health 48.2 35.2 27.3 34.6 Miscellaneous 32.4 28.9 40.7 34.2 Total 19.3 16.6 21.1 19.0 Lanjouw and Lerner then looked at the number of claims and the number of citations per claim for the patents in their data set, and compared them to a control set of randomly selected patents.210 They found that the litigated patents had far more claims than the control set, and also more forward citations per claim (i.e., more patents cite them than cite the control group) and fewer backward cites per claim (i.e., they cite less patents on average than the control group, an indication that the technology area is not well-developed and innovation is less likely to be derivative). Both of these findings indicate that valuable patents are more likely to be involved in litigation. The authors then perform a detailed analysis of three primary litigation metrics\u00E2\u0080\u0094the probability of filing a suit, the probability of settlement after a suit is filed (including the timing of such settlements), and the plaintiff win probability for cases at trial\u00E2\u0080\u0094as related to the type of the firm involved. They found that publicly-listed (i.e., large) companies are far less likely to file suits on their patents\u00E2\u0080\u0094their mean filing rate was 10.4 suits per thousand patents, as compared to about 45 suits for 209 Source: Id. The authors note that these filing rates may be understated, however, since they were calculated using only the main patents in each suit, when in fact there may be several patents per suit, especially in areas like biotechnology. 2 1 0 For each litigated patent, a \"matched\" patent having the same application year and primary patent classification code was chosen from at random from the set of all U.S. patents (both litigated and unlitigated). Part IV. Identifying the Issues 70 unlisted companies and individuals.2\" Moreover, filing rates for foreign patentees (mostly unlisted firms) were much lower than for their domestic counterparts. Lanjouw and Lerner explain this difference in filing rates by looking to the firm's size, and hypothesizing that size presents two advantages. First, firms with large patent portfolios may be better able to settle disputes through trading (i.e., cross-licensing) intellectual property, without resorting to suits. Second, assuming that smaller firms are less able to finance expensive patent litigation, relatively large firms may be better able to settle because they pose greater litigation threats when confronting smaller firms. And when large firms have disputes with each other, they are likely to have many points of interaction other than trading intellectual property, especially through competition in product markets, and this interaction in other dimensions should promote settlement. , Theciata confirmed these hypotheses. For a small (unlisted) company with a small portfolio of 100 patents (cumulative over a 10-year period), the. average probability of litigating a given patent is 7.8 percent. , For a similar company but with a moderate portfolio of 500 patents, the figure drops to. only 2.0%.212 The authors found that the marginal effect of patent portfolio size is stronger for smaller companies (as measured by employment), which indicates that for smaller firms, having a portfolio of patents to trade is likely to_be the key mechanism for avoiding litigation. Lanjouw and Lerner's most important finding, however, was that the post-suit outcomes\u00E2\u0080\u0094the probability of settlement and the plaintiff win rates at trial\u00E2\u0080\u0094are almost completely independent of firm or portfolio size. From a policy perspective, this is good news because it means that enforcement of patent rights depends on the effective threat of court action more than on extensive, post-suit legal proceedings which consume court resources. This conclusion was reinforced by the two additional findings: first, that about 80 percent of all patent suits filed are settled by the parties before trial; and 211 Id. The authors assume that the listed companies are generally bigger than the unlisted ones. They also measure size more directly by looking to how many people the companies employ. 2 1 2 Id. Part IV. Identifying the Issues 71 second, that most settlement occurs soon after the suit is filed, often before the pre-trial hearing is held. 2 1 3 The bad news, of course, is that individuals and small firms, as we saw, are much more likely to be involved in suits\u00E2\u0080\u0094and no more likely to resolve those disputes quickly in post-suit settlements. In the end, it seems that-patent litigation only tells a small part of the story of how the patent system works as a whole. As we saw in Table 3, only 19 out of every 1,000 patents are ever involved in litigation, or about 2 percent. Of those 19, only about 4 ever go to court\u00E2\u0080\u009480% are settled beforehand. That means that out of the 175,983 patents granted last year, only about 670 will ever see court time. Mark Lemley, Professor of Law at the University of California, Berkeley, calls the other 99.6% of patents the \"missing patents/'214 The overwhelming number of patentees are spending the money to get patents\u00E2\u0080\u0094a sum he estimates to be almost $5 billion a year\u00E2\u0080\u0094and then not enforcing them. Of course, a number of phenomena we've already looked at go a long way to explaining most of the, so-called.missing patents. As observed by Lanjouw and Lerner, one of the primary \"new uses\" for patents is-cross-licensing between firms. Although such patents are not enforced in the traditional sense, they do end up being extremely valuable for firms which might otherwise be forced to litigate instead. Another related use for the missing patents is simply defensive. The threat of litigation can quickly be negated if a firm has a patent portfolio with which to threaten back. A third main use, one which is more valuable to smaller firms lacking significant patent portfolios, is that of financing tool. Although venture capitalists may not like the risks associated with the actual use of patents in litigation, they love patents as a concrete indicator of the value of a start-up company. Another recent study by Kortum and Lerner found that there is a strong positive relationship between venture capital financing 213 Id. 2 1 4 See Lemley, supra note 200. 2 1 5 He estimates that at a prosecution cost of $25,000 per patent, $5 billion per year is paid to patent lawyers and the PTO. Id. at 1 38. He refines these numbers in his next essay\u00E2\u0080\u0094see infra note 387 and accompanying text. Part IV. Identifying the Issues 12 and patenting\u00E2\u0080\u0094rthat is, the amount of venture capital activity in an industry significantly increases its rate of patenting.216 In addition, inventors may choose to prosecute patents for other reasons. Some companies patent just so they have a registry of their ideas. After inventing something, they patent it, in the same way that scientists might, in other fields, publish it. The company then has an official government record saying'that they invented this thing on this date. Similarly, particularly in academic settings, patents may be used as internal yardsticks for progress in research and development, with companies or research labs measuring how innovative they are by how many patents they have. In this sense, a patent may be an end to itself for a scientist who will be rewarded for getting it or just wants to put it on his resume. \u00E2\u0080\u00A2 We've seen the emergence of licensing shops, like Walker Digital, whose sole business purpose is not to create innovative products, but merely to accumulate and license patents. Other individuals may just be \"paper patenting\"\u00E2\u0080\u0094inventing something, but rather that building it themselves, licensing out. the technology, or even worse, suing others who try to build it. Or there may be companies at the end of their useful technological life, so to speak, whose major asset is not new technology, but patents they acquired ten or fifteen years ago when they were a more productive entity. It also seems that patents are to be used simply as indicators of product differentiation or branding, or to enhance the image of the patenting entity with an aura of creativity and technological proficiency.217 Consider the 1999 release of the Oral-B CrossAction toothbrush. In a television commercial announcing its entry into the market, a baritone-voiced narrator touted its uniqueness and high-tech nature and advised viewers that there had been three years of development, three hundred 2 1 6 Samuel Kortum & Josh Lerner, \"Does Venture Capital Spur Innovation?\" (1998) National Bureau of Economic Research Working Paper No. W6846. Part IV. Identifying the Issues 73 scientific studies, and over twenty-three patents filed in relation to its innovative tooth brushing properties.218 The discovery that 339 patents have been issued in the past 25 years relating to hair-brushes\u00E2\u0080\u0094heated, flocked, vented, bendable, extendable, foldable, and self-cleaning hair-brushes, brushes for straightening, curling, colouring and cutting hair, at least one combination brush and shoehorn, and a birth-control pill dispenser in the form of a hair-brush\u00E2\u0080\u0094begs the question as to how much money was spent securing those patents, what returns were generated, and how they could possibly enable monopoly hair-brush profits that were not attainable through effective marketing and efficient distribution.219 \u00E2\u0080\u00A2 All in all, much of the focus of academic inquiry, especially among legal scholars, needs to be aimed away from the small number of cases that turn up in court toward the large number of patents revealed in these broader uses. Attention should be paid to the new players, as well\u00E2\u0080\u0094the venture capitalists, the company researchers, the university scientists, and those who patent just because they can. E. The Disconnect Starting in 1980, there began a series of radical changes to patent policy and practice in the United States that have had the generally-perceived effect of strengthening the protection that patents provide and extending the applicability of that protection both institutionally and technologically. Roughly coincident in time with these changes, there was an unprecedented surge in both patenting and R&D spending. Despite the significance of the policy changes and wide availability of detailed patent data, however, robust conclusions as to their effect on technological innovation are disappointingly few. 2 1 7 See Ann Bartow, \"Separating Marketing Innovation from Actual Invention: A Proposal for a New, Improved, ^ Lighter, and Better-Tasting Form of Patent Protection\" (2000) 4 J. Small & Emerging Bus. L. 1. 2 , 8 Id. at 3. 2 l 9/ "Thesis/Dissertation"@en . "2002-05"@en . "10.14288/1.0077445"@en . "eng"@en . "Law"@en . "Vancouver : University of British Columbia Library"@en . "University of British Columbia"@en . "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en . "Graduate"@en . "Patenting innovation : intellectual property rights in the new economy"@en . "Text"@en . "http://hdl.handle.net/2429/12210"@en .