4
Seven Recommendations for a 21st-Century Patent System

The committee supports several steps to ensure the vitality and improve the functioning of the patent system.

  • An open-ended, unitary, flexible patent system. The system should remain open to new technologies with features that allow flexibility in protecting new technologies. Among the features that should be exploited is the United States Patent and Trademark Office’s (USPTO) development of examination guidelines for new or newly patented technologies. The office should seek advice from a wide variety of sources and maintain a public record of the submissions in developing such guidelines, and the results should be given appropriate deference by the courts. The Court of Appeals for the Federal Circuit (“Federal Circuit”) also should ensure its exposure to a variety of expert opinions by encouraging submission of amicus briefs and by exchanges with other courts. In addition to qualified intellectual property professionals, appointments to the Federal Circuit should include people familiar with innovation from a variety of perspectives—management, finance, and economics, as well as nonpatent areas of law affecting innovation.

  • Non-obviousness standard. The requirement that to qualify for a patent an invention cannot be obvious to a person of ordinary skill in the art should be assiduously observed. In an area such as business methods, where the common general knowledge is not fully described in published literature that is likely to be consulted by patent examiners, another method of determining the state of general knowledge needs to be employed. Given that patent applications are examined ex parte between the applicant and the examiner it would be difficult to bring in other expert opinions at that stage. Nevertheless, the Open Review procedure



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 81
A Patent System for the 21st Century 4 Seven Recommendations for a 21st-Century Patent System The committee supports several steps to ensure the vitality and improve the functioning of the patent system. An open-ended, unitary, flexible patent system. The system should remain open to new technologies with features that allow flexibility in protecting new technologies. Among the features that should be exploited is the United States Patent and Trademark Office’s (USPTO) development of examination guidelines for new or newly patented technologies. The office should seek advice from a wide variety of sources and maintain a public record of the submissions in developing such guidelines, and the results should be given appropriate deference by the courts. The Court of Appeals for the Federal Circuit (“Federal Circuit”) also should ensure its exposure to a variety of expert opinions by encouraging submission of amicus briefs and by exchanges with other courts. In addition to qualified intellectual property professionals, appointments to the Federal Circuit should include people familiar with innovation from a variety of perspectives—management, finance, and economics, as well as nonpatent areas of law affecting innovation. Non-obviousness standard. The requirement that to qualify for a patent an invention cannot be obvious to a person of ordinary skill in the art should be assiduously observed. In an area such as business methods, where the common general knowledge is not fully described in published literature that is likely to be consulted by patent examiners, another method of determining the state of general knowledge needs to be employed. Given that patent applications are examined ex parte between the applicant and the examiner it would be difficult to bring in other expert opinions at that stage. Nevertheless, the Open Review procedure

OCR for page 81
A Patent System for the 21st Century described next provides a means of obtaining expert participation after a patent issues. With respect to gene-sequence-related inventions, a low standard of non-obviousness results from Federal Circuit decisions making it difficult to make a case of obviousness against a genetic invention (for example, gene sequences). In this context the court should return to a stricter standard, which would also be more consistent with other countries’ practices in biotechnology patenting. Open Review procedure. Congress should seriously consider legislation creating a procedure for third parties to challenge patents for a limited period after their issuance in an administrative proceeding before administrative patent judges of the USPTO. The speed, cost, and design details of this proceeding should make it an attractive alternative to litigation to determine patent validity and be fair to all parties. USPTO capabilities. To improve its performance the USPTO needs additional resources. These funds should enable hiring additional examiners, implementing a robust electronic processing capability, and creating a strong multidisciplinary analytical capability to assess management practices and proposed changes. In addition, the funds should be used to provide early warning of new technologies being proposed for patenting, and to conduct reliable, consistent, reputable quality reviews that address office-wide as well as subunit and examiner performance. The current USPTO budget does not suffice to accomplish these objectives and to administer an Open Review procedure. Research liability for patent infringement. In light of the Federal Circuit’s 2002 ruling that even noncommercial scientific research enjoys no protection from patent infringement liability, and in view of the academic research community’s belief in the existence of such an exemption, and behavior accordingly, there should be some level of protection for noncommercial uses of patented inventions. Congress should consider appropriately narrow legislation, but if progress is slow or delayed the Office of Management and Budget and the federal government agencies sponsoring research should consider extending “authorization and consent” to grantees as well as contractors, provided that such rights are strictly limited to research and do not extend to any resulting commercial products or services. Either legislation or administrative action could help ensure preservation of the “commons” required for scientific and technological progress. Litigation elements. Three provisions of patent law that are frequently raised by plaintiffs or defendants (rarely by the courts) in infringement litigation depend on determining a party’s state of mind, and therefore generate high discovery costs. These provisions are (1) “willful infringement,” which if proven, exposes an infringer to possible triple damages; (2) the doctrine of “best mode,” which addresses whether an inventor disclosed in an application what the inventor considered to be the best implementation of the invention; and (3) the doctrine of “inequitable conduct,” concerning whether the applicant’s attorney intentionally misled the USPTO in prosecuting the original patent. To reduce the cost and

OCR for page 81
A Patent System for the 21st Century increase the predictability of patent infringement litigation outcomes, and to avoid other unintended consequences, these provisions should be modified or removed. International harmonization. The United States, Europe, and Japan should further harmonize patent examination procedures and standards to reduce redundancy in search and examination and eventually achieve mutual recognition of results. Differences that among others are in need of reconciling include application priority (“first-to-invent” versus “first-inventor-to-file”), the grace period for filing an application after publication, the “best mode” requirement of U.S. law, and the U.S. exception to the rule of publication of patent applications after 18 months. This objective should be pursued on a trilateral or even bilateral basis as well as a multilateral basis. Although some of our recommendations parallel those of previous commissions and reports, the most relevant comparison is with the proposals of the Federal Trade Commission (FTC) in its report released in October of last year. Although we approached the operation of the patent system from different perspectives, addressed somewhat different topics, and employed quite different methodologies, there are several areas of agreement. The USPTO and the Court of Appeals for the Federal Circuit should broaden their consideration of relevant economic and technical analysis. The non-obviousness standard should be more vigorously applied, at least in some technological fields. Congress should create a review procedure for challenging and reviewing issued patents. The financial resources of the USPTO should be increased. All patent applications should be published after 18 months. The legal doctrine subjecting “willful” infringers to enhanced damages should be modified or eliminated. PRESERVE A FLEXIBLE, UNITARY, OPEN-ENDED PATENT SYSTEM Innovation processes differ markedly from one industrial sector to another. There is ample evidence that development lead times, product cycles, the relative dominance of cumulative or interoperative or stand-alone innovations, capital investment requirements, and even sources of innovation all vary greatly. We know, too, that firms in different industries acquire, value, and exercise patents differently. Accordingly, the optimal number, coverage, and division of patent rights to encourage innovation may vary. These circumstances, some might argue, call for designing a formal (that is, statutory) system in which patent standards, strength, duration, and other features vary from technology to technology and, conceivably, certain technologies are excluded from patenting altogether.

OCR for page 81
A Patent System for the 21st Century Historically, there has been strong resistance to a differentiated patent system and to subject matter exclusions and fairly consistent adherence to a relatively open-ended unitary system. Exceptions, although more common recently, are relatively few and narrow and usually in the nature of limited exceptions rather than sui generis systems of intellectual property protection. For example, in 1996 Congress exempted medical practitioners and related health care entities using patented medical procedures from infringement liability rather than bar surgical procedure patents altogether.1 It lengthened terms for some pharmaceutical patents to compensate for regulatory delays2 and protected certain experimental uses of pharmaceuticals by generic suppliers from liability.3 Recently, Congress was persuaded that the advent of business method patents might snare longtime users of newly patented business procedures in infringement suits; but, rather than curtail the issuance of such patents or limit their terms, legislation made prior use a defense available to accused infringers of that class of patents.4 A special obviousness provision deals with concurrent process and composition-of-matter claims on biotechnology patent applications.5 The Plant Patent Act of 19526 and the Plant Variety Protection Act of 1970,7 representing modified patent regimes, and the 1984 semiconductor mask protection legislation8 and the 1988 Vessel Hull Design Protection Act,9 representing modified copyright regimes, are the only examples of new statutory classes of intellectual protection designed for particular technologies. Apart from the very recent congressional ban on human organism patents,10 clearly a special case, there have been no successful legislative attempts to circumscribe patenting. Some members of this committee are concerned about patent incursions on the public domain of ideas and information, particularly in the realm of scientific research results. Even so, they believe that the proper approach is on a case-by-case basis through a mechanism for review of issued patents for conformity with the statutory standards and associated case law rather than an attempt to draw the line more sharply. The entire committee endorses such a postgrant review procedure, described later in this chapter. The committee also agrees that given the state of our knowledge there are strong reasons to preserve a formally unitary system. For one thing, we do not 1   35 U.S.C.§ 287(c) (2000); P.L. 104-208, 1996 HR 3610. 2   35 U.S.C. §§ 155, 156 (2000). 3   35 U.S.C. § 271 (2000). 4   35 U.S.C. § 273(b)(1a)(3) (2000). 5   35 U.S.C. § 103(b) (2000). 6   35 U.S.C. §§ 161, 164. 7   7 U.S.C. § Sec 2321 et seq. 8   17 U.S.C. § 901-14 (2000). 9   Part of the Digital Millennium Copyright Act, P.L. 105-304. 10   It has been USPTO policy since 1987 not to issue any human organism patents.

OCR for page 81
A Patent System for the 21st Century know enough about innovative processes to advise Congress on the optimal characteristics of different classes of patents in different circumstances. Legislative tailoring of the patent system to each major industry would be a prime opportunity for interest-group politics to influence the results, which could be quite resistant to change or adjustment. Even if Congress were able to get it “right” in economic terms, technological change and industries’ structural evolution might render the specifications obsolete and possibly counterproductive. That appears to have been the case with semiconductor mask protection. Although the industry lobbied vigorously for the legislation, and there have been a number of filings under it, few in the industry still view it as an important way to protect proprietary chip design, primarily because the underlying technology evolved rapidly, obviating the perceived need. The fact that such instances are rare suggests that Congress has no great appetite for crafting industry- or technology-specific patent policies. In any case, it has largely tied its hands by ratifying the TRIPS agreement of 1994, which prohibits signatory states from discriminating in the grant of patents based on the technology involved.11 The committee realizes that there may appear to be some contradiction between this position and our belief in the importance of exploiting the mechanisms and doctrines that reflect differences among technologies or allow for some deliberate discrimination among them by the USPTO, by the courts, and by patent holders themselves. These include subtle differences in the application of the common legal standards of obviousness, enablement, and written description, and the various other policy levers described by Burk and Lemley (2003a). The difference is that these mechanisms, in contrast to legislation, allow for incremental adjustments that are more easily made. In particular, the committee endorses the USPTO’s development of examination guidelines, outlining how it will apply the statutory standards to emerging technologies. In the case of the utility and written description guidelines for genetic inventions and earlier in the case of computer programs, this was accomplished through a notice and public comment process not unlike that employed by federal regulatory agencies in formal rulemaking proceedings. This is not only a means of achieving some degree of standardization in USPTO practice involving a new technology or newly patented technology well before a number of validity cases are decided by the courts. It is also a means of obtaining advice from a variety of sources in a way that is open to all interested parties.12 The USPTO should solicit comments from legal scholars, economists, and independent experts as well as stakeholders and maintain an open record of the submissions. Further, the guidelines and the record behind them should be part of 11   Article 27(1). 12   The USPTO periodically requests comments on issues other than those in proposed rulemaking, and these are publicly available.

OCR for page 81
A Patent System for the 21st Century the record in any appeal to a court, where they could be used to inform judicial decisions. There are other ways to expose the courts to a wide range of opinion and analysis. The Court of Appeals for the Federal Circuit is in most instances the final arbiter of patent law. Both students of and practitioners before the court are in general agreement that the 1982 centralization of patent appeals in the Federal Circuit has been a vast improvement over adjudication in the circuit courts of appeals. It reduced forum shopping, focused attention and thought on neglected issues of patent law, produced innovations at the trial court level, and in general yielded greater consistency. At the same time, specialized institutions have insular tendencies.13 For example, the Federal Circuit appears to rely less on independent scholarly analysis, even legal scholarship, than the regional generalist appeals courts. Nard (2002) found that the circuit courts cite scholarly work roughly four times as often as the Federal Circuit. He acknowledges that the Federal Circuit is more familiar with patent law than regional courts are with, say, copyright or trademark law; but he suggests that the disparity is such that the “court verges on the abstract by failing to give adequate weight to empirical and economic scholarship.”14 We recommend some modest steps to ensure that the Federal Circuit, despite its specialization, has broader exposure to legal and economic analysis in all areas of innovation-related law and to the impact of its decisions on the lower courts and on the Patent Office. Briefs. The Federal Circuit should encourage the submission of briefs that draw upon insights from other judicial decisions, legal scholarship on the patent system, and the growing body of patent-related economics literature. In particular, 13   The definitive analysis of specialized adjudicative bodies and their biases is by Marver Bernstein (1955). Interestingly, some of the early patent appeals court proposals did not contemplate permanent appointments to the court. For example, the National Research Council (1919) recommended selection by the chief justice from the district and circuit court benches, with service limited to a six-year term unless reappointed. The 1936 National Research Council report recommended permanent appointment, but of judges with diverse scientific and technical backgrounds as well as experience in the trial of patent cases. 14   Judge Paul Michel, speaking at the University of California, Boalt Hall, Conference on Patent System Reform, March 1, 2002, made a similar point about the recent litigation on the issue of the doctrine of equivalents in FestoCorp. v. Shoketsu Kinzoku Kogyo Kabushiki Co., 234 F.3d 558, 56 U.S.P.Q.2d 1865 (Fed. Cir. 2000) (en banc), overruled in part by 535 U.S. 722, 122 S. Ct. 1831, 152 L. Ed. 2d 944 (2002): “Now you might have thought … where there was a concern about the relative needs to promote adequate incentives—or you could say fairness to inventors—on the one hand with the need for competitors to have adequate predictive value and certainty on the other hand, that somebody at least amici and, one would hope, also the parties would have given us some very meaningful data about that. Now I read all the briefs … and I can’t remember anything that I would consider empirical data…. If you trace back the pedigree I suspect that you will find that in a great many cases there never was any meaningful economic or quantified data.”

OCR for page 81
A Patent System for the 21st Century as the court has done in some recent notable cases,15 it should welcome amicus briefs, as these tend to raise broader issues and cite a wider range of literature than do the briefs of parties to cases. Appointments. Because the Federal Circuit’s docket extends to diverse cases far removed from patent or, more broadly, innovation law, the few appointments intended to support the court’s expertise in that area should be made with particular care. They should not be confined to intellectual property practitioners and academics. Rather, the court’s perspective should be broadened by appointing judges familiar with innovation more generally, including men and women with backgrounds in antitrust or finance law or, in addition to their legal training, in economics or economic history. Furthermore, some appointments might be through the elevation of regional district court judges, a routine practice with respect to the regional appeals courts but not the Federal Circuit. The addition of one or more district judges with patent litigation experience would give the court perspective not only on problems at the trial court level but also on economic issues outside its jurisdiction. Designations. Trial court judges are often asked to sit by designation in other courts of appeals, helping the system as a whole keep track of jurisprudential trends. Federal Circuit judges have rarely participated in this practice.16 The committee suggests that the Federal Circuit invite regional judges to sit on its panels and regional circuits invite Federal Circuit judges to sit by designation. This would give Federal Circuit judges a better sense of how patent law fits in with other laws influencing innovation and how other courts incorporate economics into their decision making. REINVIGORATE THE NON-OBVIOUSNESS STANDARD Non-Obviousness and Business Method Inventions The non-obviousness determination is necessarily a judgment, not something that can be resolved through a bright line test. Fundamentally, it assumes that an invention is novel and the decision maker must determine whether the hypothetically skilled person in the art would nonetheless have considered the novel invention something within the routine skill of the field. The USPTO and the reviewing courts are concerned that an invention that was genuinely non-obvious before it was made may often look obvious in retrospect. The courts have been vigilant in 15   For example, the Federal Circuit recently granted an en banc review to consider the willfulness doctrine discussed below and actively solicited amicus briefs on a wide range of issues to aid in its deliberations. See Knorr-Bremse Systeme Fuer Nutzfahrzeuge Gmbh v. Dana Corp. 344 F.3d 1336 (Fed. Cir. 2003). 16   See Dreyfuss (forthcoming) for suggestive data on the extent to which circulation of judges into and out of the Federal Circuit occurs.

OCR for page 81
A Patent System for the 21st Century preventing hindsight based on the inventor’s patent disclosure from leading to an “obvious” determination. The doctrines that protect against hindsight, while well founded in most other contexts, assume that the USPTO will have access to the state of the art at the time the invention at issue was made. That assumption has not been true in the business methods area. The prior art by which inventions are judged is defined by 35 U.S.C. § 102. The most common forms of prior art in conventional areas of technology are printed publications, including scientific journals and patents. The prior art also includes prior information “known or used by others.” In foreign patent systems this is known as the “common general knowledge,” which describes the concept somewhat better. Under U.S. and foreign law it is not enough that one or a few people in the field have the information alleged to be prior art. The information must be generally known to be patent defeating. In conventional technologies the published literature in fact represents a fairly good catalog of the common general knowledge in the field, and the USPTO can therefore access it. That is not the case for business methods, which may not be written in any of the places likely to be consulted by examiners.17 Several examples illustrate how the conventional approach to obviousness breaks down in the business methods context. For example, one common form of patent application involves use of a new component or ingredient to replace an old one for a particular function. In patent law this is an issue of “equivalence,” that is, whether the new component is recognized in the prior art as an equivalent of the old component. Section 2144.06 of the Manual of Patent Examining Procedure (MPEP) addresses this issue, relying on three Federal Circuit cases. It states, “In order to rely on equivalence as a rationale supporting an obviousness rejection, the equivalency must be recognized in the prior art, and cannot be based on applicant’s disclosure or the mere fact that the components at issue are functional or mechanical equivalents.” In other words, use of a new ingredient, even a familiar ingredient in other circumstances, can lead to a patentable invention unless the USPTO can determine that such a substitution was part of the state of the art. This understanding in the art can be based on the common general knowledge apart from the published literature.18 In an area like business methods, where the published literature does not fully describe the state of the art, the USPTO is 17   For example, relevant prior art, such as for the patent on the inverse elasticity rule, appears in economics texts. But much of the prior art for business methods is embodied in services and processes not described in any literature (Laurie and Beyers, 2001). 18   Graham v. John Deere Co., 383 U.S. 1 at 10 (1966) quoted Jefferson that “[A] change of material should give no right to a patent. As to the making a ploughshare of cast rather than of wrought iron; a comb of iron instead of horn or ivory …” T. Jefferson, Letter to Isaac McPherson (Aug. 1813), VI Writings of Thomas Jefferson, at 181. Presumably this statement was not based on a learned treatise suggesting cast-iron ploughs but the common understanding at the time that cast iron was a general substitute for wrought iron in the latter’s many applications.

OCR for page 81
A Patent System for the 21st Century severely handicapped in its ability to make rejections based on the obviousness of a substitution. A corollary to the common-general-knowledge principle involves the combining of ideas from different sources. Every invention at some level is a combination of old elements. Again, experience has shown that truly non-obvious inventions will be denied patent protection unless the decision maker guards against the use of hindsight.19 USPTO policy reflects this concern. Based on Federal Circuit and Court of Customs and Patent Appeals (CCPA) holdings, MPEP 2143.01 states that “[t]he mere fact that references can be combined or modified does not render the resultant combination obvious unless the prior art also suggests the desirability for the combination.” Again, the poverty of the published literature on business methods makes it difficult for the USPTO to make obviousness rejections. If the problem of the non-obviousness standard’s application to business method inventions is lack of access to published information regarding the common general knowledge in the field at a particular time, that knowledge is best provided through testimony (affidavit or live) by those active in the field at the relevant time. Consideration should therefore be given as to how the USPTO could obtain such testimony. Patent applications are examined ex parte in secret between the applicant and the USPTO. The USPTO does not employ experts to provide evidence that might support an obviousness rejection. One approach would be to change this practice to admit the testimony or written opinion of the USPTO-appointed experts. This would, however, prolong the patent prosecution and make it more expensive. Perhaps more importantly, there are significant concerns regarding the USPTO’s ability to maintain the impartiality it should have respecting the merits of an application. The examiner already plays the dual role of adversary and “judge” during the examination process. It is likely that the USPTO’s testimonial evidence will be contradicted by testimony submitted by the applicant. It may not be wise to have the USPTO act as the decision maker resolving the competing evidence where one of the sources is an expert retained by the USPTO itself. This concern is particularly compelling in view of the recent Supreme Court decision, Dickerson v. Zurko,20 which held that on appeal the courts will have to affirm any 19   An example is In re Fine, 837 F.2d 1071 (Fed. Cir. 1988), which involved a system for detecting and measuring certain nitrogen compounds by using a gas chromatograph, a converter to oxidize the nitrogen compounds into nitric oxide, and a nitric oxide detector. Two previous references were relevant; one disclosed an analogous approach to monitoring certain sulfur compounds, and the other described nitric oxide detectors. Although the examiner and the Board of Patent Appeals thought it obvious to substitute the nitric oxide detector in the system, the Federal Circuit found that there was no support for such a conclusion. The USPTO had not demonstrated that the idea of the combination of teachings was within the state of the art. 20   527 U.S. 150 (1999) which reversed and remanded In re Zurko, 142 F.3d 1447 (Fed. Cir.1998).

OCR for page 81
A Patent System for the 21st Century USPTO finding of fact (for example, what was part of the common general knowledge) unless there was no substantial evidence to support the finding. Merely disagreeing with the USPTO on how the evidence should have been weighed will not be a basis for reversal. The testimony of the USPTO’s witness would almost always meet the substantial evidence test.21 Thus, the use by the USPTO of retained experts during examination does not appear to be a workable solution. Elsewhere in this report the committee has proposed the implementation of an efficient process in the USPTO by which third parties could challenge the validity of an issued patent. This process, which we call Open Review, would be a preferable method of bringing testimony regarding the common general knowledge to the attention of the USPTO. The parties affected by the patent will likely be in the best position to obtain testimony from those working the field at the relevant time. The USPTO will not have a vested interest in either side’s experts. Further, since obtaining and evaluating testimony requires more resources than conventional patent prosecution, the Open Review forum will tend to confine those costs to cases in which the patent is of some market importance. It may be argued that the case of business methods is not only unique but transitional and therefore of little broader significance. We are not confident on either score. Some of the apparently obvious patents listed in Chapter 3 appear to have the same characteristic and therefore may have been approved not carelessly but under the prevailing rule that references should not be combined for the purpose of proving non-obviousness unless the examiner can point to a specific piece of prior art that says the references should be combined. The business methods arena lacks a publication culture but even where such a culture exists, scientists, artisans, and creative people generally speaking strive to publish non-obvious information. So if it is obvious to those of skill in the art to combine references, it is unlikely that they will publish such information. It is therefore difficult to imagine that another class of patent applications will not pose the same issue in the future. In the meantime, with business methods patent grants, it is true that there will be a steady accumulation of patent prior art. But even it will be limited if the United States remains the only major country issuing business method patents. Moreover, given the great variety of business method applications, the lack of nonpatent published information may remain a significant handicap in assessing obviousness. 21   USPTO reliance on expert witnesses raises additional concerns. One concern is whether the office would have to disclose to the applicant any contrary expert opinions it has obtained; in much the same way the applicant must disclose adverse information to the USPTO under the duty of disclosure. It is also questionable whether the USPTO would have access to those persons most knowledgeable about the state of any art. Such individuals may be competitors of the patent applicant. Thus, it will be problematic to retain them for the secret examination process. These are the types of individuals the applicant wishes to exclude.

OCR for page 81
A Patent System for the 21st Century Non-Obviousness and Gene Sequence-Related Inventions One basis for rejecting an invention for obviousness has been to allege that the invention was “obvious to try.” This test has been rejected by the courts because it penalizes those who devise a sensible plan of research by exploring the paths most likely to succeed. Particularly when success of the chosen path was not assured, eliminating the patent incentive in such a circumstance was recognized as contrary to the purpose of the patent system. The courts, therefore, have held that an invention is only obvious and unpatentable when the obvious route to try is coupled with a “reasonable expectation of success.” One of the earliest Federal Circuit decisions in the field of biotechnology, In re O’Farrell,22 found that an invention related to gene expression was unpatentable under the above test even though success was not assured. The O’Farrell court dealt with how high the “reasonable” bar should be set and set it quite low. The court found that the inventors’ own prior publications with similar systems expressing “nonsense” sequences in E. coli provided a reasonable expectation that actual gene sequences from an exogenous source would be expressed into functional proteins. This decision made the obviousness standard easier to use in rejecting applications.23 The USPTO has applied this standard for biotechnology in general.24 If the early technical advance of O’Farrell in the relatively unpredictable period during which recombinant gene expression was still being worked out was obvious under the above standard, it is a fair question to ask why are not most of the gene sequences from the human genome project obvious and therefore unpatentable. After all, the technical question presented by the genome project was not whether the human genome could be sequenced, but which group would finish first. The sequencing of the genome, and other collections of mass sequence-related data (for example, expressed sequence tags [ESTs]) would appear to be obvious to try with a reasonable expectation of success.25 The reasonable expec- 22   853 F.2d 894, available at 7 U.S.P.Q.2d 1673 (BNA) (Fed. Cir. 1988). 23   The Supreme Court’s decision in Dickerson v. Zurko, 527 U.S. 150 (1999) should further strengthen the USPTO’s position in making obviousness rejections under the reasonable expectation test. Several of the key issues underlying the obviousness test are factual issues, namely the content of the prior art, level of skill in the art, expectation of success, and motivation to combine prior art. Thus, applicants will have a difficult time overturning adverse USPTO decisions if there is any reasonable evidentiary basis to support an expectation test. 24   For example, the obviousness standard has defeated patentability for novel monoclonal antibodies prepared against known antigens using routine techniques. Ex parte Erlich, 22 U.S.P.Q.2d 1463 (BNA) (Bd. Pat. App. & Int. 1992). 25   The law permits an applicant to overcome a presumption of obviousness if an applicant can demonstrate an unexpected or superior property of the claimed invention not shared by the prior art. In re Papesch, 315 F.2d 381, 137 U.S.P.Q. (BNA) 43 (CCPA 1963); In re Dillon, 919 F.2d 688, 16 U.S.P.Q.2d (BNA) 1897 (Fed. Cir. 1990). Thus, patentability for a new gene might lie in the discovery of an unexpected or superior property of the gene, or more likely, the protein it encodes. Many

OCR for page 81
A Patent System for the 21st Century of lawyers providing such opinions at a cost ranging from $10,000 to $100,000 per opinion.62 Worse, in some business sectors, exposure to claims of willful infringement has led to a practice of deliberately avoiding learning about issued patents, a development sharply at odds with the disclosure function of patent law. Willfulness creates a strong disincentive to read patents, irrespective of whether any infringement allegations are made. The mere existence of the doctrine in its current form means that any time an individual or company learns of a patent that might bear on its products, the company is at risk. Regardless of how the patent comes to light, the company must spend tens of thousands of dollars to obtain an opinion that it is not infringing. And then it foregoes some or possibly all of its attorney-client privilege in the evaluation of the patent.63 To avoid this situation, in-house counsel and many outside lawyers regularly advise their clients not to read patents if they can avoid it (Lemley and Tangri, 2003; Taylor and Von Tersch, 1998). In this respect patent law stands in contrast with trademark law, which premises willfulness in part on a failure to search for prior marks. Other collateral issues that enter into infringement litigation and raise the complexity and cost of pretrial discovery are the competence of the exculpatory opinion and the propriety of opinion counsel appearing as trial advocates. Against these costs, complications, and uncertainties there has been no empirical demonstration that the availability of enhanced damages provides substantial additional deterrence over and above that associated with the usual costs and risks of defending an infringement claim, the threat of pretrial injunction relief (rare but potentially devastating to an enterprise), and post-trial award of attorney’s fees against deliberate infringement. Thomas Cutter (2004) has analyzed the deterrent effects of enhanced damages for patent infringement generally and concluded that in many circumstances the criteria courts now employ in determining willfulness have an overdeterrent effect—discouraging marginally lawful behavior and taking advantage of the patent disclosure—and therefore undesirable social costs. Lacking evidence of its beneficial deterrent effect but with evidence of its perverse antidisclosure consequences, the committee recommends elimination of the provision for enhanced damages based on a subjective finding of willful infringement; but we recognize that this is a matter of judgment and that there are a number of alternatives short of elimination that merit consideration.64 A modest 62   The process of preparing an opinion is described in Poplawski (2001). 63   Reliance on the advice of counsel may provide a basis for a successful defense, but by choosing to inject its attorney’s advice into the case the defendant waves attorney-client privilege for at least all circumstances and documents relating to that advice and possibly for all advice given before the suit was filed. 64   As noted above, the Federal Circuit has taken for en banc review a willful infringement case, Knorr-Bremse v. Dana, 344 F.3d 1336 (Fed. Cir. 2003), and signaled its intention to consider many aspects of the doctrine. Whether the outcome will address the committee’s concerns remains to be

OCR for page 81
A Patent System for the 21st Century step is to abolish the effective requirement that accused infringers obtain and then disclose a written opinion of counsel. Another possibility is to limit inquiry into willful infringement to cases in which the defendant’s infringement has already been established. A third alternative that preserves a viable willfulness doctrine but curbs its adverse effects is to require either actual, written notice of infringement from the patentee or deliberate copying of the patentee’s invention, knowing it to be patented, as a predicate for willful infringement (Federal Trade Commission, 2003; Lemley and Tangri, 2003). If some form of willfulness doctrine is retained, there is the question by how much should damages be enhanced. One answer is by the least amount needed to deter deliberate copying and make the victims whole. Lemley and Tangri suggest that in most instances awarding successful plaintiffs their attorney fees will suffice as an adequate penalty. Finally, modification or elimination of willful infringement raises questions about the status of the “duty of care” to avoid patent infringement. This is a matter we did not address that merits further consideration. “Best Mode” Defense65 Section 112 of the Patent Act requires that an application “set forth the best mode contemplated by the inventor of carrying out his invention.” As interpreted by the Federal Circuit, this requirement is judged by a two-part test; first, did the inventor, at the time of filing, know of a mode of practicing the invention that the inventor believed was preferable to others; and second, was the best mode adequately disclosed, in light of the scope of the claimed invention and the level of skill in the art. The first test is inherently subjective, focusing on the inventor’s state of mind; and although the second test is objective, it is not precise. The best-mode standard is different from Section 112’s “enablement” requirement, which goes to the sufficiency of the disclosure to teach one of ordinary skill to implement the invention. Best mode requires in addition that if the inventor knows of particular materials or processes for implementing a claimed invention that the inventor believes are most effective, they must be revealed. For example, consider a claim that states a range of temperatures for operation of a method. If at the time of filing the application the inventor believed that a particular temperature or narrower range was optimal, failure to disclose it—even if unintentional—may result in a finding that the claim is invalid. As with other invalidity defenses, establishing a best-mode violation requires “clear and convincing” proof; and the defense is applied only on a claim-by-claim basis. However, if     seen. The case raises the question of whether willful infringement should be presumed when an infringer failed to obtain an opinion of counsel before infringing or invokes attorney-client privilege or the work product doctrine to avoid disclosing an opinion obtained. 65   For a general description and background of the best mode defense, see Chisum (1997) and Hofer and Fitzgerald (1995).

OCR for page 81
A Patent System for the 21st Century intent to deceive is shown, the same proof can establish “inequitable conduct,” and the entire patent may be unenforceable. Only the United States imposes a best-mode requirement. Its goal is to motivate more extensive disclosure to the public by increasing the risk of withholding related information as a trade secret. As explained by the Federal Circuit’s predecessor court, the purpose of the requirement is to “restrain inventors from applying for patents while at the same time concealing from the public preferred embodiments of their invention which they have in fact conceived.”66 Analysis of the best-mode defense is made as of the time the inventor filed the original application; there is no obligation to “update” the application with information discovered during prosecution of the patent. Moreover, the defense applies only to information and belief personal to the inventor, and cannot be established by imputation of knowledge of others in the inventor’s company or working group. Therefore, this doctrine as applied gives only limited assurance that the best mode will be disclosed. Because the defense depends on historical facts and because the inventor’s state of mind usually can be established only by circumstantial evidence, litigation over this issue—especially pretrial discovery—can be extensive and time-consuming. Foreign patent applicants also criticize the doctrine as unfair, since their previously filed foreign applications cannot simply be translated for filing in the United States without attending to this unusual additional requirement. Given the cost and inefficiency of this defense, its limited contribution to the inventor’s motivation to disclose beyond that already provided by the enablement provisions of Section 112, its dependence on a system of pretrial discovery, and its inconsistencies with European and Japanese patent laws, the committee recommends that the best-mode requirement be eliminated. Inequitable Conduct Defense67 Even when a patent claim is valid, if obtained through fraud, it is deemed unenforceable. This concept is codified in Section 282 of the Patent Act. The defense of inequitable conduct applies when the patent applicant has made a material misstatement or omission with intent to deceive the USPTO. Like invalidity, unenforceability through inequitable conduct must be proved through “clear and convincing” evidence. However, unlike questions of invalidity, inequitable conduct is decided by a judge, not a jury. Inequitable conduct requires proof of both materiality of the information and intent to deceive. Materiality has been measured by a standard similar to that applied in cases of securities fraud: whether there is a substantial likelihood that a 66   In re Gay, 309 F.2d 769 at 772 (C.C.P.A. 1962). 67   For a general description and background on inequitable conduct, see Chisum (1997).

OCR for page 81
A Patent System for the 21st Century “reasonable examiner” would have considered the information important in deciding whether to issue the patent. A 1992 USPTO rule change appears to have raised the bar on materiality so that now, arguably, a defendant must prove that the information, properly disclosed, would have led the USPTO to reject the relevant claim.68 Intent is, of course, a subjective issue, directed at the state of mind of the patent applicant or the applicant’s attorney. Once a judge has determined that a threshold level of materiality and intent have been proved, the court will consider all of the circumstances surrounding the applicant’s conduct and balance the level of materiality and intent to determine if inequitable conduct exists. In other words, a high level of one might offset a low level of the other so that in some cases inequitable conduct is found despite very little evidence of deliberate misconduct. Examples of behavior punished as inequitable conduct include failure to cite a known prior art reference unless it is merely “cumulative” to those already cited, “burying” a material reference in a stack of irrelevant information, submitting false or misleading declarations related to dates of invention or enablement, and failure to disclose offers for sale and public uses that would make the claim invalid under Section 102(b) of the Patent Act. The consequences of a finding of inequitable conduct can be severe. First, the entire patent, not just the relevant claim or claims, is rendered unenforceable. Second, other patents in the same family may be deemed “infected” by the fraud. Third, the defendant may be awarded attorneys fees under the “exceptional case” standard of Section 285 of the Patent Act. Finally, the patentee may be exposed to an antitrust claim.69 As with willful infringement and the best-mode defense, discovery is more complex and expensive. Moreover, because the level of disclosure to the USPTO usually involves choices made by the patent attorney, issues of the scope and waiver of attorney-client privilege are implicated. Another major complaint is that the defense is asserted too freely. One judicial opinion commented: “[T]he habit of charging inequitable conduct in almost every major patent case has become an absolute plague.”70 If invalidity, disciplinary action, and reputational concerns are not sufficient deterrent to misconduct, other civil and even criminal remedies exist—antitrust, unfair competition, common law fraud, and tortuous interference. Moreover, since the creation of the inequitable conduct doctrine by the courts, other safeguards 68   The effect of the USPTO rule change has not yet been decided by the Federal Circuit. In Molins PLC v. Textron., 48 F.3d 1172 at 1179 n.8 (Fed. Cir. 1995), the court merely observed that because administrative rules are not retroactive, and the case arose before the rule change, it did not have to address the issue. However, in Semiconductor Energy Laboratory Co., Ltd. v. Samsung Elecs Co., 204 F.3d 1368 at 1374 (Fed. Cir. 2000), the court applied the new rule to a patent issuing on an application filed after 1992, without any discussion of whether the old standard should apply. 69   See Walker Process Equipment, Inc. v. Food Machinery & Chemical Corp., 382 U.S. 172 (1965). 70   Burlington Industries, v. Dayco Corp., 849 F.2d 1418 at 1422 (Fed. Cir. 1988).

OCR for page 81
A Patent System for the 21st Century have been adopted by Congress and the USPTO to support the integrity of the patent system. These include third-party- and USPTO-initiated re-examination on withheld prior art, publication of pending applications, and third-party access to pending prosecution papers and the ability to submit material information. The Open Review process we propose would also contribute to the integrity of the system. In view of its cost and limited deterrent value the committee recommends the elimination of the inequitable conduct doctrine or changes in its implementation. The latter might include ending the inference of intent from the materiality of the information that was withheld, de novo review by the Federal Circuit of district court findings of inequitable conduct, award of attorney’s fees to a prevailing patentee, or referral to the USPTO for re-examination and disciplinary action. Any of these changes would have the effect of discouraging resort to the inequitable conduct defense and therefore reducing its cost. Our recommendations would almost certainly simplify litigation and curb unproductive discovery and thereby reduce its expense, but to what extent? And if only one or two rather than all three elements of patent litigation areas of law were reformed, which would yield the largest litigation cost saving? We are not certain, but we have benefited from the opinions of a large group of highly experienced patent litigators, most of them in private practice. Knowing the committee’s interest in these questions, the fellows (senior members and former officers) of the American Intellectual Property Law Association conducted an informal survey of their colleagues on the three elements of litigation considered. A substantial minority of 93 respondents supported the modification or elimination of one or more of these rules even though they are beneficiaries of the complexity and cost of patent litigation. A slight majority considered them significant cost drivers, although some of them suggested that uncontrolled discovery was a problem of civil litigation generally. Respondents split evenly in identifying willful infringement versus inequitable conduct as the main cost factor. Best mode ranked a distant third. HARMONIZE THE U.S., EUROPEAN, AND JAPANESE PATENT EXAMINATION SYSTEMS As early as 1966 a presidential commission appointed by President Johnson recommended that the United States adopt the otherwise universal first-inventor-to-file basis for determining patent priority as a step toward making the major industrial countries’ patent systems compatible. The Carter administration’s policy review on innovation and Commerce Secretary Mosbacher’s commission on the patent system in the first Bush administration also urged progress toward harmonization. The adoption of the 20-years-from-filing-patent term and publication of most patent applications at 18 months, both under the TRIPS agreement, were important recent steps in that direction. The United States is currently

OCR for page 81
A Patent System for the 21st Century engaged in negotiations under the World Intellectual Property Organization (WIPO) aimed at substantive patent law harmonization. These developments are a recognition that in an increasingly integrated global economy, differences in patent law create redundancy and inconsistencies that raise the cost of doing international business. The committee did not consider the thorny issues associated with reconciling differences in intellectual property protection between developing and industrialized countries but is primarily concerned with differences in patent examination among the latter, especially the United States, Europe, and Japan. In that context greater harmonization has taken on more urgency with the increase in patent filings generally and the increase in multinational filings in particular (see Figure 4-1). Each of the three major patent offices has had difficulty coping with the surge in applications, yet work sharing is minimal. For most commercially important inventions, technically and legally skilled patent examiners in each office analyze the same application, search more or less the same prior art, and perform similar examinations, sometimes with identical results, sometimes with results dictated by differences in law. The committee believes that the United States, Europe, Japan, and other countries should continue to harmonize substantive laws regarding patentability, application priority, rules of prior art, and standards of examination with the objective of establishing systems of reciprocity or mutual recognition of the results of searches and examinations. This goal will require changes in law and practice on all sides. The committee members agree that the following are among the principal differences that need to be reconciled, and we agree on the preferred terms of an agreement on patent system harmonization. First-to-Invent Versus First-Inventor-to-File Priority The United States should conform its law to that of every other country and accept the first-inventor-to-file system. There are several reasons for this shift. First, the discrepancy means not only that in some cases different people will own patents on the same invention in different countries but also that there are radical differences in procedure. The United States has an elaborate legal mechanism, both in the USPTO and in the courts, for determining who was the first to invent. Because the rest of the world has no analogous process, foreign patent applicants are subject to uncertainty and perhaps challenges that are entirely unfamiliar. The governments tend to view U.S. acquiescence to the first-to-file as the cornerstone of international harmonization. Second, U.S. inventors also file their applications in ignorance of whether they are the first or second to invent and when an opponent might be expected to file. For those subject to challenge under first-to-invent, the proceeding is costly and often very protracted; frequently it moves from a USPTO administrative pro-

OCR for page 81
A Patent System for the 21st Century FIGURE 4-1 USPTO, EPO, and JPO patent application filings, 1982-2002. SOURCE: OECD (2003).

OCR for page 81
A Patent System for the 21st Century ceeding to full court litigation. In both venues it is not only evidence of who first reduced the invention to practice that is at issue but also questions of proof of conception, diligence, abandonment, suppression, and concealment, some of them requiring inquiry into what an inventor thought and when the inventor thought it. A third reason to adopt the first-inventor-to-file priority basis is that for the overwhelming majority of applicants, that is the system the United States has. Of the more than 300,000 applications the USPTO receives each year only about 200 to 250—less than 0.1 percent—end up in interference proceedings because a second filer claims to be the first inventor. There are, nonetheless, three concerns that merit attention in considering whether to abandon first-to-invent. The first concern is how often first inventors would be unfairly deprived of their inventions by second inventors who happened to file with the Patent Office first? The answer, it turns out, is a nontrivial number or at least a non-negligible proportion of applicants involved in interferences. Lemley and Chien (2003) examined two sets of interference cases—first, 76 final adjudications by the Board of Patent Appeals and Interferences (BPAI) between 1990 and 1991 that were decided by determining who was the first inventor; and second, a random selection of the few hundred interference proceedings reported on the BPAI web site between 1997 and 2003. They concluded that second filers won approximately 43 percent of the cases. Nevertheless, in a large proportion of these cases first- and second-filers’ invention dates were so close as to be nearly simultaneous. A second concern is the inducement inherent in a first-to-file system to file early and perhaps before the invention is fully characterized, which could be a source of patent quality deterioration. The incentive for early filing surely exists but is mitigated by two factors. First is provisional application filing whereby inventors who file a complete technical disclosure secure priority rights without a major expenditure of resources for legal services. This allows the applicant a year to characterize, refine, consider claims for, and assess the commercial value of an invention before submitting a formal application. The second mitigating factor is that inventors already have significant incentives to file applications early, for instance any inventor who seeks protection outside the United States competes in a first-to-file system. Most important from a fairness and a political point of view, first-to-file is claimed to disadvantage individual inventors and small business, who may not have the resources to be as fast as large companies. This has been the premise of very effective “independent inventor” opposition to first-to-file and harmonization generally for a very long time. To illuminate the issue Gerald Mossinghoff (2002) studied all 2,848 interference decisions between 1983 and 2000 to determine whether small inventors were more likely to prevail in priority disputes. He found that the first-to-invent system did not benefit small inventors on average. Of that number, 203 were decided in favor of a small entity filing second, but in 201 other cases small-entity first filers lost. For another perspective on the small-

OCR for page 81
A Patent System for the 21st Century entity issue, Lemley and Chien (2003) examined on whose behalf the interference cases in their study were initiated. Strikingly, of the 94 initiators for which status data were available, 77 percent were large firms while only 18 percent were small entities. Of the responding parties 43 percent were individuals or small businesses while 53 percent were large entities. Both sets of evidence support the conclusion that the first-to-invent system is not working to the benefit of small entities; rather, in the preponderance of cases large firms are ensnaring small companies in complex, costly interference proceedings. Even if this were not so, small businesses increasingly oriented toward international markets might prefer harmonization as a way to reduce the total costs of multimarket protection. According to a 2002 General Accounting Office survey, 70 percent of small business respondents agreed with that objective. Grace Period The United States should retain and seek to persuade other countries to adopt a grace period, allowing someone to file a patent application within one year of publication of its details without having the publication considered prior art precluding a patent grant.71 This provision encourages early disclosure and is especially beneficial for dissemination of academic research results that may have commercial application. As other countries try to accelerate the transfer of technology from public research organizations to private firms through patents and licensing, the idea of a grace period is likely to become more widely accepted. Germany recently adopted such a provision. Best-Mode Requirement The “best mode” requirement, having no analog in foreign patent law, imposes an additional burden and element of uncertainty on foreign patentees in the United States. This, in addition to its dependence on discovery aimed at uncovering inventor records and intentions, justifies its removal from U.S. patent law. Prior Art In the interest of arriving at a uniform definition of prior art, the United States should remove its limitation on non-published prior art and its rule that foreign patents and patent applications may not be recognized as prior art as of their filing dates. In connection with moving to a first-inventor-to-file system, the foreign patent prior art rule for unpublished prior patent applications should also 71   35 U.S.C. § 102(b).

OCR for page 81
A Patent System for the 21st Century be adopted. A common misconception about the EPO and other foreign systems like that of the EPO is that they are winner-take-all systems similar to the U.S. interference proceeding. A difference in prior art treatment, however, prevents this from occurring. Abroad an unpublished prior patent application is available for prior art purposes only under the novelty standard. It cannot be used in a non-obviousness (or equivalent) rejection. This allows the later filing applicant to obtain claims to a disclosed aspect of the invention that is novel with respect to the prior application even if it would have been obvious. This has the affect of giving some reward to near simultaneous inventors. Where the second to file is first with a commercially important embodiment of the invention, the foreign rule increases cross-licensing and enhances competition in the marketplace. Application Publication The United States should abandon its exception to the rule of publication after 18 months for applicants not intending to patent abroad. This, too, would promote the disclosure purpose of the patent system. Eliminating the non-publication option would minimize the uncertainty associated with submarine patents, which remain a problem as a consequence of the continuation practice, enabling an applicant to abandon one application and file a continuation or pursue an application to issue while maintaining a continuation on file—in either case in the hope of winning a better patent eventually. Moreover, universal publication would extend to all patentees the provisional rights under 35 U.S.C. Sec. 154(d) (2000) that give a patentee a reasonable royalty for infringement that occurs after publication but before patent issuance under certain conditions (Lemley and Moore, 2004). Other Issues There are other differences regarding the scope of patentable subject matter and the standards for non-obviousness and utility that we have not examined in detail and for which we therefore have no precise prescriptions. Given that patent laws are part of historically evolved national legal systems there may be limits to harmonization, but these are likely to recede over time as the international economy becomes more integrated and enterprises more dependent on global markets. Reconciling patent system differences will be challenging but would make the outcome of this so-called “deep harmonization” more rewarding. The committee supports the pursuit of harmonization through WIPO but recognizes the difficulty of achieving agreement among 180-odd countries with widely divergent views of intellectual property protection generally and the patent system in particular. There is a risk that harmonization of the three major patent systems could be sidetracked by disagreements between the developing and developed countries. We believe that harmonization should and can be pursued in

OCR for page 81
A Patent System for the 21st Century trilateral or even bilateral negotiations or on selected issues whose results, if there is agreement, will have a beneficial demonstration effect on other countries. In the meantime, the practice of convening international panels from the three patent offices to explore common approaches to search and examinations in new technological areas should be continued. This practice is helpful not only in identifying issues for negotiation but, more immediately, in informing patent applicants how their inventions are likely to be treated in each of the patent offices. The committee recognizes that its proposals, apart from foreign adoption of a grace period, would represent U.S. conformity with other patent systems and may be subject to the charge that we favor “Europeanizing” the U.S. patent system. That is a narrow view. It presumes that only the items enumerated are part of a negotiated package. It implies that the U.S. system features we propose changing are important to its integrity. We disagree. Most important, it ignores what we expect to be the benefits of harmonized priority and examination procedures for U.S. inventors, whether large or small entities—first, faster, more predictable determinations of patentability; second, simplified, less costly litigation; and third, less redundancy and much lower costs in establishing global patent protection.