Mr. President, I rise today to comment on S. 3600, the Patent Reform Act of 2008. This bill is based on, but makes a number of changes to, S. 1145, a patent reform bill that was reported out of the Judiciary Committee in 2007 but that was…
Mr. President, I rise today to comment on S. 3600, the Patent Reform Act of 2008. This bill is based on, but makes a number of changes to, S. 1145, a patent reform bill that was reported out of the Judiciary Committee in 2007 but that was never considered by the full Senate.
S. 1145 proposed several salutary and uncontroversial reforms to the patent system, but also included provisions that would rewrite the formula for awarding damages in patent cases and that would create new administrative proceedings for challenging patents. These and other provisions of that bill would have made it much more expensive to hold and defend a patent, would have extended the time for recovering damages for infringement, and would have substantially reduced the amount that the patent holder would ultimately recover for infringement. The changes proposed by S. 1145 went so far that under that bill's regime, it may have proved cheaper in many cases to infringe a patent and suffer the attenuated and reduced consequences of doing so, rather than to pay a license to the holder of the patent. Once such a line is crossed, the incentive to invest in research and development and the commercialization of new technology in this country would be greatly reduced. Such a change would do enormous harm to the U.S. economy in the medium-to-long term. Reputable economists estimate that historically, between 35 and 40 percent of U.S. productivity growth has been the result of innovation.
My bill makes substantial changes to those sections of S. 1145 that address damages, post grant review, venue and interlocutory appeals, applicant quality submissions, and inequitable conduct. This bill will not be considered in this Congress. I nevertheless thought that it would be useful to propose alternative approaches to these issues now, to allow Senators and interested parties the time to consider these alternatives as we prepare for the patent reform debate in the next Congress. I hope that my colleagues will work with me in a bipartisan and deliberative manner to construct a bill that will be considered in the next Congress. With those thoughts in mind, allow me to describe the significant changes that this bill makes to S. 1145.
I believe that S. 1145 goes too far in restricting a patent owner's right to recover reasonable royalty damages. On the other hand, I also believe that there is room for improvement in current law. Some unsound practices have crept into U.S. patent damages litigation. My staff and I spent several months at the end of last year and the beginning of this year discussing the current state of patent damages litigation with a number of seasoned practitioners and even some professional damages experts. I sought out people with deep experience in the field who had not been retained to lobby on pending legislation.
A substantial number of the experts with whom I spoke said that there is nothing wrong with current damages litigation and that Congress should not change the law. Others, however, identified a number of unsound practices that they believe have led to inflated damages awards in a significant number of cases. Different attorneys and experts repeatedly identified the same valuation methods and criteria as being unsound, subject to manipulation, and leading to damages awards that are far out of proportion to an invention's economic contribution to the infringing product. Examples of problematic methodologies that were identified to me include the so-called rule of thumb, under which an infringed patent is presumptively entitled to 40 percent or some other standard portion of all of the profits on a product, the use of the average license paid for patents in an industry as a starting point for calculating the value of a particular patent, and a formula attributed to IBM whereby every high-technology patent is entitled to 1 percent of the revenues on a product. A number of experts also criticized the use of comparables, whereby the value of a patent is calculated by reference to the license paid for a supposedly comparable patent.
The views of those experts who were critical of current damages law find
some support in the macro evidence. Data collected by PricewaterhouseCoopers and FTI Consulting indicate that the majority of the largest patent-damages awards and settlements of all time have been entered only since 2002. Also, the inflation adjusted value of awards entered since 2000 is more than 50 percent higher than it was during the early 1990s. And it also appears that jury awards tend to be about ten times higher than the average damages award entered by a judge, and that results vary markedly by jurisdiction. These facts suggest that the problems that sometimes lead to inflated damages awards are to some extent systemic.
The task of reforming substantive damages standards presents a very difficult legislative question. Damages calculation is an inherently fact-intensive inquiry and requires legal flexibility so that the best evidence of a patent's value may always be considered. Any proposed changes to the law must be evaluated in light of the kaleidoscope of factual scenarios presented by the calculation of damages for different types of patents.
I have largely given up on the idea of developing a unified field theory of damages law that solves all problems at once. I also oppose proposals to require a prior-art subtraction in every case. Most measures of a reasonable royalty, such as established royalties, costs of design-arounds, comparisons to noninfringing alternatives, or cost savings produced by use of the patented invention, already effectively deduct the value of prior art out of their estimate of the patented invention's value. To mandate prior-art subtraction when using such measures would be to double count that deduction, effectively subtracting the prior art twice and undervaluing the invention.
And for reasons mostly explained in my minority views to the committee report for S. 1145, S. Rep. 110-259 at pages 64-65, I also disagree with the argument that defendants should be allowed to revisit validity questions, such as a patent's novelty or nonobviousness, during the damages phase of litigation. To those comments I would simply add that, if Congress were to desire that patents be defined more specifically and narrowly, then it would need to provide express guidance as to how to do so. Simply using adjectival phrases such as ``specific contribution'' or ``inventive features'' will not suffice. These terms merely express a hope or objective. But legislation needs to be about means, not ends, particularly if it is intended to achieve its results by altering the practices and outcomes of litigation. I should also add that although I have consulted with many neutral experts in the field of patent damages, and many of those experts described to me what they believed to be serious problems with patent damages litigation, none of those experts told me that insufficiently specific claim construction is causing excessive damages awards. If overly broad claim constructions were a major source of problems with damages litigation, I undoubtedly would have come across at least one neutral expert who expressed that view.
Discussions that I have had with several proponents of S. 1145 indicated that they understand the principal evil of current damages litigation to be the award of damages as a percentage or portion of the full price of the infringing product. It also appears that some proponents of S. 1145 believe that a statutory instruction to define the invention more narrowly and clearly would prevent parties from seeking damages based on the entire value of the infringing product. The linkage between claim construction and the damages base is not clear to me. Even a concededly limited invention could be fairly valued by using the full product's price as the damages base, so long as the rate applied to that base was appropriately small.
Many unjustified and excessive awards certainly do use the full value of the infringing product as the damages base. Indeed, awards that are derived from the rule of thumb almost always are based on the entire value of the infringing product, as is the typical industry averages award. Precluding or sharply limiting the use of net sales price as a damages base certainly would block the path to many of the bad outcomes that are produced by the use of these methodologies.
The problem with a rule that bars the use of net sales price as the damages base when calculating a reasonable royalty is that in many industrial sectors, net sales price is routinely used as the damages base in voluntary licensing negotiations. It is favored as a damages base because it is an objective and readily verifiable datum. The parties to a licensing negotiation do not even argue about its use. Instead, they fight over the rate that will be applied to that base. Even if the net sales price of the product is very large and the economic contribution made by the patented invention is small, net sales price can still serve as the denominator of an appropriate royalty if the numerator is made small.
Thus in these industries, the initials, NSP, appear frequently and repeatedly in licensing contracts. A legal rule that precluded use of net sales price as the damages base would effectively prevent participants in these industries from making the same royalty calculations in litigation that they would make in an arm's length transaction. Such an outcome would be deeply disruptive to the valuation of patents in these fields. Evidence and techniques whose use is endorsed by the market via their regular use in voluntary negotiations are likely to offer the best means of valuing a patent in litigation. After all, what is an object in commerce worth, other than what the market is willing to pay? We simply cannot enact a law that bars patentees from using in litigation the same damages calculation methods that they routinely employ in arm's length licensing negotiations.
The bill that I have introduced today uses what I call an enhanced gatekeeper to address problems with damages awards. The bill strengthens judicial review of expert witness testimony, provides greater guidance to juries, and allows for sequencing of the damages and validity/infringement phases of a trial. The bill also codifies the principle that all relevant factors can be considered when assessing reasonable royalty damages, while adopting guidelines and rules that favor the use of an economic analysis of the value of an invention over rough or subjective methodologies such as the rule of thumb, industry averages, or the use of comparables. Allow me to provide a subsection- by-subsection summary of the bill's revisions to section 284, the basic patent damages statute.
Subsection (a) of the bill's proposed section 284 copies and recodifies all of current section 284, including its authorization of treble damages and its admonition that compensatory damages shall ``in no event be less than a reasonable royalty for the use made of the invention.''
Subsection (b) codifies current Federal circuit precedent defining a reasonable royalty as the amount that the infringer and patent owner would have agreed to in a hypothetical negotiation at the time infringement began. It tracks the language of the Rite-Hite case, 56 F.3d 1538 (Fed. Cir. 1995), and follow-on decisions. Some supporters of S. 1145 are critical of the hypothetical negotiation construct and believe that it leads to bad results. Not only is this test established law, however, but it is also inherent in the concept of a ``reasonable royalty.'' That standard requires the trier of fact to determine what would have been--i.e., what the parties would have agreed to. As long as the patent code requires a ``reasonable royalty,'' courts and juries will need to engage in a hypothetical inquiry as to how the invention reasonably would have been valued at the time of infringement. Indeed, it is not apparent by what other means the factfinder might approach the calculation of a reasonable royalty. And in any event, the source of occasional bad results in damages trials is not the mental framework used for approaching the question of a reasonable royalty, but rather the particular evidence and methods used to value some inventions. It would be a noteworthy omission to avoid mention of the hypothetical negotiation concept in a bill that regulates damages analysis to the degree that this one does. This subsection thus codifies the Federal circuit's jurisprudence on the hypothetical negotiation.
Subsection (c) simply makes clear that, despite subsection (d), (e), and (f)'s codification and modification of several of the Georgia- Pacific factors,
the rest of the Georgia-Pacific factors--as well as any other appropriate factor--may be used as appropriate to calculate the amount of a reasonable royalty.
Subsection (d) is probably the most important subsection in the bill's revised section 284. It bars the use of industry averages, rule- of-thumb profit splits, and other standardized measures to value a patent except under particular circumstances. Standardized measures are defined as those methods that, like rule of thumb and industry averages, do not gauge the particular benefits and advantages of the use of a patent. Instead, they are relatively crude, cookie-cutter measures that purport to value all patents--or at least all patents in a class--in the same way, without regard to a particular patent's economic value. These back-of-the envelope methods are occasionally used in arm's-length, voluntary licensing negotiations, as are things such as gut instinct and intuition. But they are rough methods that can produce wildly inaccurate results. Subsection (d) disfavors their use.
This subsection restricts the use of Georgia-Pacific factor 12, which largely describes the rule of thumb. Subsection (d)'s general rule cites the rule of thumb and industry averages as important and illustrative examples of standardized measures. But it also expressly applies to other methods that are ``not based on the particular benefits and advantages'' of an invention, to ensure that variations on these examples and other methods that consist of the same evil also are brought within the scope of subsection (d)'s main rule.
An example of a standardized measure other than profit splits and industry averages that is also currently in use and that also falls within subsection (d)'s scope is the so-called IBM 1-percent-up-to-5 formula. This formula apparently was used by IBM in the past to license its own portfolio of patents. Under this methodology, each patent receives 1 percent of the revenues on a product until a 5 percent ceiling is reached, at which point the whole portfolio of patents is made available to the licensee.
I have heard more than one representative of a high-technology company describe the use of this formula in litigation against his company. Apparently, there exists a stable of plaintiff-side damages expert witnesses who will testify that this formula is appropriate for and is customarily used to calculate the value of any patent in the computer or information-technologies sectors. These experts start at 1 percent and then adjust that number based on the other Georgia-Pacific factors, supposedly to account for the particular aspects of the patent in suit, though these adjustments almost always seem to push the number higher.
Obviously, 1 percent of revenues or even profits is a grossly inflated value for many high-technology patents. It is not uncommon for high-technology products to be covered by thousands of different patents, which are of greatly differing value. Not every one of those patents can be worth 1 percent of revenues. Some patents inevitably will be for features that are trivial, that are irrelevant to consumers, or that could be reproduced by unpatented, off-the-shelf noninfringing substitutes. One percent of the sales revenue from, for example, a laptop computer is an enormous sum of money. Many patents are worth nothing near that, and any methodology that starts at that number is likely to produce a grossly inflated result in a large number of cases.
It bears also mentioning some of those common methodologies that clearly are not standardized measures. In addition to established royalties, which are afforded an express exemption from this subsection by paragraph (2), there are the methods of calculating the costs of designing around a patent, drawing comparisons to the experience of noninfringing alternatives, or calculating the costs savings produced by use of the invention. All of these factors gauge the benefits and advantages of the use of the invention and therefore are outside the scope of subsection (d).
Paragraph (1) of subsection (d) allows parties to use a standardized measure, such as a rule-of-thumb profit split, if that party can show that the patented invention is the primary reason why consumers buy the infringing product. If the patented invention is the primary reason why people buy the product, then the patent effectively is the reason for the commercial success of the product, and its owner is entitled to a substantial share of the profits, minus business risk, marketing, and other contributions made by the infringer.
Some have advocated a lower standard than ``primary reason'' for allowing use of profit splits and other standardized measures--for example, using a ``substantial basis'' standard. I rejected the use of a lower standard because a profit split should basically award to the patent owner all of the profits on the product minus those attributable to business risk. Thus the test for allowing such profit splits must be one that only one patent will meet per product, since the bulk of the profits can only be awarded once. If the test were ``substantial basis,'' for example, multiple patents could meet the standard and multiple patent owners could demand all of the profits minus business risk on the product.
Paragraph (2) of subsection (d) makes established royalties an express exception to the bar on standardized measures. In earlier drafts, I did not include this exception in the bill because I thought it obvious that an established royalty is based on the benefits and advantages of the use of the invention and is thus outside the scope of the subsection (d) rule. Some parties who reviewed those earlier drafts, however, found the bill ambiguous on this point, and in any event the lack of an exception would have forced parties to litigate the question whether an established royalty was, in fact, based on the benefits and advantages of the use of the patent. Since established royalties are widely considered to be the gold standard for valuing a patent, we should avoid making it harder to use this method. It is thus expressly placed outside the scope of subsection (d)'s restrictions by paragraph (2).
Paragraph (3) of subsection (d) allows industry averages to continue to be used to confirm that results produced by other, independently allowable methods fall within a reasonable range. The paragraph speaks of ``independently'' allowable methods in order to make clear that an industry average cannot be used to confirm an estimate produced solely by reference to a ``comparable'' patent. Subsection (e) requires that comparables only be used in conjunction with or to confirm other methods, and thus under this bill comparables are not a method whose use is allowed ``independently'' of other methods.
A brief explanation is in order as to why this bill regards industry averages as a potentially unreliable metric and restricts their use. An industry average often will reflect a broad range of licensing rates within a technological sector. Even a licensed patent whose value is included in the calculation of such a range may fall at a far end of that range, producing highly inaccurate results if that average is used as a starting point for calculating the value of that patent. Moreover, many existing patents, though valid and infringed by a product, disclose trivial inventions that add little to the value of the product. But the types of patents that typically are licensed--and that therefore would be a source of available data for calculating an industry average--are the ones that are substantial and valuable. Trivial patents don't get licensed, and their value does not enter into industry average calculations. Thus particularly in the case of a minor patent that has never been and likely never would be licensed, an industry average would provide an inflated estimate of the patent's value. This is because the industry average is not the average licensing rate of all patents in a field, but merely the average of those that have been licensed and for which data is publicly available.
Paragraph (4) of subsection (d) creates a safety valve that allows parties to use standardized measures if no other method is reasonably available to calculate a reasonable royalty, and the standardized method is otherwise shown to be appropriate for the patent. Over the course of drafting this bill, I have consulted with a number of experts with broad experience in patent damages calculation. Only a few believed that they had ever seen a case where use of a standardized measure
was necessary--that is, where a more precise economic analysis was not feasible. I thus anticipate that this safety valve may almost never need to be used, but I nevertheless include it in the bill, because it is impossible to say with certainty that no situation will ever arise in the future where parties will be unable to calculate a reasonable royalty without use of the rule of thumb or other standardized measures. Suffice to say that if one party to a suit presents appropriate evidence of a patent's value and that evidence falls outside the scope of subsection (d) or within one of the other exceptions, then that method is ``reasonably available'' and paragraph (4) could not be invoked.
A word about the need for substantive standards: some critics of S. 1145 have made the argument to me that any problems with damages litigation can be cured through procedural reforms, and that changes to substantive legal standards such as those in subsections (d) through (f) are unnecessary. These parties also have made the related, though different argument that to the extent that litigants are using unreliable evidence or methodologies, this problem should be addressed through cross examination and advocacy.
Though I share these critics' displeasure with S. 1145, I do not think that problems such as the overuse of rule of thumb and industry averages will be completely solved through purely procedural reforms. The most likely mechanism for excluding these methodologies would be rule 702. But the use of some of these methods for valuing patents is endorsed by multiple experts. These methods, while ultimately unsound, represent a significant minority view that is backed by some published commentary, albeit sometimes only commentary in journals that are exclusively written by, subscribed to, and read by plaintiff-side damages expert witnesses. In such circumstances, it is no sure thing that a party will be able to exclude under Daubert the testimony of an expert employing these methodologies. These metrics are sufficiently entrenched that the only way to ensure that the courts will disallow them when their use is not appropriate is for Congress to tell the courts to disallow them.
As to the second point, it is true that it is the lawyer's duty to identify the flaws in the other side's arguments and to debunk unsound theories. But the reality is that because of the limited expertise and experience of many jurors and the limited time allowed to argue a case at trial, often the trier of fact will not divine the truth of the matter. And some unsound damages methodologies are particularly likely to be appealing to those untutored in the field. An industry average analysis, for example, employs the one statistical concept that is understood by virtually everyone, and this method's use may amount to no more than a simple back-of-the-envelope calculation that requires only one expert to give you the industry average licensing rate and another to calculate the gross revenues on the product. When a complex economic analysis that focuses on noninfringing alternatives to the patented invention or the costs of a design-around is forced to compete for the jury's favor with a simple average-rate-times-sales calculation, many jurors may find the simpler and readily understandable method more intuitively appealing, even if it is less accurate. And of course, when two different and even slightly complex damages calculations are presented to a jury, there always exists a risk that the jury will resolve the dispute by splitting the difference between the two methods. In a high-value case where the patent owner uses an unsound method that produces a wildly inflated number, the risk that the jury will pick the wrong method or even split the difference may easily be unacceptable from a business perspective.
In the end, it is the premise of the rules of evidence that some types of evidence are so unsound, so prejudicial, or so likely to produce an unjust result that we do not require the other side's lawyer to debunk this evidence, but rather we require the judge to bar it from the courtroom altogether. If we find that particular methodologies routinely produce inaccurate and unjust results, it is appropriate that we amend the law to directly restrict the use of those methodologies.
Subsection (e) restricts and regulates the use of licenses paid for supposedly comparable patents as a means of calculating the value of the patent in suit. The use of comparables is authorized by Georgia- Pacific factor two and can generate probative evidence of a patent's value. Nevertheless, such use is regulated and restricted by this subsection. Comparables are a valuation method that is often abused, both to overvalue and to undervalue patents. When an infringer is sued for infringing an important patent, he often will cite as evidence of a reasonable royalty the license paid for a patent that is in the same field but that is much less valuable than the patent in suit. Similarly, a plaintiff patent owner asserting a trivial patent may cite as ``comparable'' other patents in the same field that are much more valuable than the plaintiff's patent. The fact that another patent is licensed in the same industry should not alone be enough to allow its use as a comparable in litigation.
Comparability is a subjective test. By definition, every patent is unique and no two patents are truly comparable. Subsection (e) thus requires that comparables be used only in conjunction with or to confirm the results of other evidence, and that they only be drawn from the same or an analogous technological field. I chose the latter term rather than ``same industry'' because the term ``industry'' is too broad. Parties might define ``industry'' so expansively that every patent in the universe would fall into one of only two or three ``industries.''
Paragraph (2) of subsection (e) sets out guideposts for determining whether a patent is economically comparable to another patent. It suggests requiring a showing that the supposed comparable is of similar significance to the licensed product as the patent in suit is to the infringing product, and that the licensed and infringing products have a similar profit margin. Obviously, a patent that makes only a trivial contribution to a product cannot accurately be valued by reference to a comparable that makes a critical and valuable contribution to its licensed product, or vice versa. And similarity in the profitability of the licensed and infringing products will also generally be important to establishing the economic comparability of two patents. As an economic reality, when the profits on a product are high, the manufacturer will be more generous with the royalties that he pays for the patented inventions that are used by the product. This economic reality is undergirded by the fact that it will typically be the patented inventions used by a product that make that product unique in the marketplace and allow it to earn higher profits. Even if two patents are the principal patent on products in the same field, if one patent's product has a 2-percent profit margin and the other's has a 20-percent profit margin, that first patent evidently is doing less to distinguish that product in its market and to generate consumer demand--and thus has a lower economic value.
A thorough analysis of comparability, of course, likely will depend in a given case on many factors beyond those listed here. Subparagraphs (A) and (B) are simply guideposts that describe two factors that are likely to be relevant to comparability. The bill only provides that these two factors may be considered. It does not preclude consideration of other factors, nor does it require that these two factors be considered in every case. A party asserting the propriety of a comparable may be able to show that one or even both of these factors are not appropriate to establishing economic comparability in a given case.
Subsection (f) bars parties from arguing that damages should be based on the wealth or profitability of the defendant as of the time of trial. Some lawyers have been known, after making their case for an inflated royalty calculation, to emphasize how insignificant even that inflated request is in light of the total revenues of the defendant infringer. Such arguments do not assist the jury in gauging a reasonable royalty. Rather, they serve to reduce the jury's sense of responsibility to limit a reasonable royalty to the actual value of the use made of the invention. This subsection does not bar all
consideration of the financial condition of the infringer. It may be appropriate to consider the infringer's finances at the time of infringement especially if there is some evidence that such information is considered when licensing patents in the relevant industry. But in no case should a court allow such information to be presented when the evident purpose of doing is to tell the jury that the defendant has deep pockets and will not be burdened by an inflated award.
Subsection (g) gives either party a presumptive right to demand that validity and infringement be decided before the jury hears arguments about damages. Currently, some plaintiffs will force a premature debate over damages in order to color the jury's view of validity and infringement. For example, in some cases, the same defense witness who testifies as to validity and infringement will also know facts relevant to the patent's value. This may allow the plaintiff's lawyer to question that witness about damages, forcing the defendant to begin arguing about the amount of his liability before the jury has even heard all the arguments as to whether the patent is valid and infringed. A defendant who is already arguing about what a patent is worth will tend to look as if he has already conceded that he owes something, and that the dispute is simply over the amount.
This tension also exists even when all validity and infringement arguments are presented before damages are argued. Current law routinely allows the defendant to be forced to argue in the alternative to be made to argue in one breath that he is not liable and in the next that if he is liable, then this is the amount for which he is liable. A presumptive right to have one issue resolved before the other is addressed would cure this tension. This subsection allows only sequencing of the trial, not full bifurcation. It does not require the use of a second jury, and allows all pretrial activity, including that related to damages, to be completed before the validity and infringement case is presented and decided. The jury would decide validity and infringement and then proceed immediately to hear the damages case, if still needed.
Subsection (h) requires an expert to provide to the opposing party his written testimony and the data and other information on which his conclusions and methods are based, and to also provide the written testimony to the court. This subsection supplements current law, codifying and enforcing the better interpretation of what is currently required by the rules of procedure. It is necessary because those current rules are sometimes not fully enforced, and experts sometimes are allowed to testify, for example, as to what is customary in an industry without providing the facts and figures or evidence of actual events that are the basis for the expert's view that something is customary. Rule 702 exists to ensure that expert witnesses are not simply allowed to argue from authority. It allows opposing counsel to challenge the expert's methods as unsound, but that right becomes illusory if the expert is allowed to testify without ever disclosing an objective foundation for his conclusions. Requiring the expert's written testimony to also be provided to the judge should allow the judge to prepare himself to consider motions regarding the relevance and admissibility of the expert's testimony.
Subsection (i) codifies and reinforces current law allowing a party to seek summary judgment or JMOL on damages issues. It also requires a court to instruct the jury only on those issues supported by substantial evidence, a requirement which, when appropriate motions have been made, should prevent the court from simply reading the laundry list of all 15 Georgia-Pacific factors to the jury. The court's identification of those factors for which there is substantial evidence not only will provide better guidance to the jury, but should also clarify the record and give form to the factfinder's decision, thereby providing a better foundation for an appeal.
Section 299A creates a patent-specific and expanded Daubert rule. First, it makes Rule 702 specific to the Federal circuit and patent law. Currently, rule 702 is regarded by the Federal circuit as a procedural rule, and thus in each case the Federal Circuit simply follows the Daubert jurisprudence of the regional circuit whence the district court decision came. Since the regional courts of appeals do not hear patent cases, this system retards the development of a rule 702 jurisprudence that thoroughly considers some of the unique issues presented by patent law and particularly patent-damages law. The current situation also requires the district courts to look only to rule 702 precedent that is based only on nonpatent cases. By embedding rule 702 in the patent code, section 299A will force the development of more consistent and thorough jurisprudence regarding what kinds of reasonable royalty damages calculation methodologies are reliable and what kinds are not. Like subsection (h) above, this section supplements rather than replaces current law.
Section 299A also codifies the four indicia of reliability that were announced in the original Daubert v. Merrell Dow Pharmaceuticals decision, 509 U.S. 579 (1993), as well as two other indicia that are not described in Daubert. These two additional reliability indicia, at paragraphs (5) and (6), are based on standards announced in court of appeals decisions that apply Daubert. These decisions are discussed in footnote 30 of section 6266 of Wright and Miller's Federal Practice and Procedure. The first new factor, whether a theory or technique has been employed independently of litigation, should be useful in flushing out methodologies that exist only in litigation expert witness' testimony and are never employed in actual licensing negotiations. Use of this reliability indicator should inject more honesty into the hypothetical negotiation. It should force parties to use methodologies that actually would have been used had the infringer and claimant negotiated a license, rather than metrics that are only ever employed in an expert's imaginary parallel universe.
The second new reliability indicator, whether the expert has accounted for readily available alternative theories, should exclude the expert who ignores precise and objective metrics of value in favor of subjective and manipulable methodologies that allow him to produce the result that happens to most favor his client. If there is clear evidence, for example, of the market price of a noninfringing alternative to the infringing product, of the costs of noninfringing substitutes for the invention or the costs of a design-around, or of the cost savings produced by use of the invention, an expert witness should not be allowed to ignore that evidence. He must consider that evidence or at least provide a persuasive account as to why it should not be considered. One common sign of a bad or biased expert witness is his disregard of readily available alternative theories or techniques. Paragraph (6) will help to ensure that Federal courts exercise their gatekeeper role and bar such witnesses from misleading the jury.
Finally, subsection (c) of proposed section 299A requires district courts and circuit courts to explain their Daubert determinations, which should facilitate appeal of those decisions.
Section 5 of the bill authorizes the creation of post grant review proceedings for challenging the validity of patents. It allows both first- and second-window review of a patent, with procedural restrictions that will limit the time and expense of these proceedings and protect patent owners. The bill uses a procedural model that is favored by PTO and is calculated to allow quick resolution of petitions. Importantly, the bill also imposes procedural limits on when a second-window proceeding may be sought after civil litigation has commenced, and restricts duplicative or second and successive proceedings, preventing infringers from using post grant review as a litigation or delaying tactic.
Section 5(a) of the bill repeals the procedures for inter partes reexam effective 1 year after the date of enactment of the bill, while allowing requests for reexam that are filed before that effective date to continue to be considered by the office. Director-initiated reexam is also repealed, out of concern that in the future political pressure may be brought to bear on PTO to attack patents that are a nuisance to politically important businesses.
The bill's proposed section 321 authorizes two types of post grant review
proceedings, a first-period proceeding in which any invalidity argument can be presented, and a second-period proceeding that is limited to considering arguments of novelty and nonobviousness that are based on patents or printed publications. The first-window proceeding must be brought within 9 months after the patent is issued. The second window is open for the life of the patent after the 9-month window has lapsed or after any first-period proceeding has concluded.
The bill uses an oppositional model, which is favored by PTO as allowing speedier adjudication of claims. Under a reexam system, the burden is always on PTO to show that a claim is not patentable. Every time that new information is presented, PTO must reassess whether its burden has been met. This model has proven unworkable in inter partes reexam, in which multiple parties can present information to PTO at various stages of the proceeding, and which system has experienced interminable delays. Under an oppositional system, by contrast, the burden is always on the petitioner to show that a claim is not patentable. Both parties present their evidence to the PTO, which then simply decides whether the petitioner has met his burden.
If we expect post grant review proceedings to be completed within particular deadlines, I think that it is obligatory that we consult with the agency that is expected to administer the proceedings. In this case, PTO has expressed a strong preference for an oppositional model, and it believes that it can comply with reasonable deadlines if that model is adopted. The bill's use of an oppositional system thus allows proposed section 329(b)(1) to mandate that post grant review proceedings be completed within one year after they are instituted, with a possible 6-month extension for good cause shown or in the event of second-window joinder.
Section 5 also imposes a number of procedural limitations on post grant review proceedings. Proposed section 321 applies a standing requirement that petitioners must have a substantial economic interest adverse to the patent. This is a relatively low threshold that simply requires a showing that some substantial economic activity of the petitioner's is hindered by the express or implied threat of the patent's monopoly. Nevertheless, the requirement does give patentees a measure of control over when they might be forced to defend themselves in a post grant review proceeding.
Proposed section 322 includes a number of provisions that are designed to limit the use of post grant review proceedings as a delaying tactic and to mitigate these proceedings' negative impact on efforts to enforce a patent. Subsection (a) provides presumptive immunity from post grant review proceedings to a patent that is enforced in court within three months of its issue. A patent asserted in court this early in its life likely is already the subject of a well-developed commercial dispute. A delay in resolution of the case under these circumstances probably would do unjustified and irreparable harm to one or another party's market share. Such disputes should be resolved as soon as possible, which means hearing all of the case in the one forum capable of hearing all claims, the district court.
Paragraph (1) of subsection (b) bars a party that has filed a declaratory-judgment action challenging the validity of a patent from also challenging the patent in a post grant review proceeding. And paragraph (2) requires a defendant in an infringement action who seeks to open a second-window proceeding to do so within 3 months after his answer to the complaint is due. I think that this is a better rule than one requiring that a petition for a second-window proceeding be filed before an infringement action is filed. Such a restriction might cause parties who think that they may be sued but who are not otherwise inclined to seek post grant review to file defensive petitions for second-period review, lest they later be sued and lose the right to request post grant review.
Subsection (c) of section 322 bars a party that has already sought a post grant review proceeding against a patent from subsequently seeking another post grant review or a reexam with regard to the same patent.
Subsection (d) of section 322 estops a party that has brought a post grant review proceeding against a patent from raising in any subsequent PTO or ITC proceeding or civil action any claim against that patent that it did raise in a post grant proceeding or that it could have raised in a second-window proceeding.
A word about privity: subsections (b)(2) and (d) of section 322 bar second-window proceedings from being instituted or claims from being raised if particular proceedings or claims were pursued by privies to the party now seeking to start proceedings or raise claims. The concept of privity, of course, is borrowed from the common law of judgments. The doctrine's practical and equitable nature is emphasized in a recent California Court of Appeals decision, California Physicians' Service v. Aoki Diabetes Research Institute, 163 Cal.App.4th 1506 (Cal. App. 2008), which notes, at page 1521, citations omitted, that:
The word ``privy'' has acquired an expanded meaning. The
courts, in the interest of justice and to prevent expensive
litigation, are striving to give effect to judgments by
extending ``privies'' beyond the classical description. The
emphasis is not on a concept of identity of parties, but on
the practical situation. Privity is essentially a shorthand
statement that collateral estoppel is to be applied in a
given case; there is no universally applicable definition of
privity. The concept refers to a relationship between the
party to be estopped and the unsuccessful party in the prior
litigation which is sufficiently close so as to justify
application of the doctrine of collateral estoppel.
It bears noting that not all parties in privity with a would-be petitioner for other purposes or by way of various contracts would also be in privity with the petitioner for purposes of estoppel--that is, for purposes of section 322. This limitation on estoppel privity is usefully highlighted in a decision of the Federal circuit, International Nutrition Co. v. Horphag Research, Ltd., 220 F.3d 1325 (Fed. Cir. 2000), which notes, at page 1329, that:
One situation in which parties have frequently been held to
be in privity is when they hold successive interests in the
same property. See, e.g., Litchfield v. Crane, 123 U.S. 549,
551, 8 S.Ct. 210, 31 L.Ed. 199 (1887) (defining privity to
include a ``mutual or successive relationship to the same
rights of property''). Thus, a judgment with respect to a
particular property interest may be binding on a third party
based on a transfer of the property in issue to the third
party after judgment. See Restatement (Second) of Judgments
Sec. 43 (1982) (``A judgment in an action that determines
interests in real or personal property . . . [h]as preclusive
effects upon a person who succeeds to the interest of a party
to the same extent as upon the party himself.''). A corollary
of that principle, however, is that when one party is a
successor in interest to another with respect to particular
property, the parties are in privity only with respect to an
adjudication of rights in the property that was transferred;
they are not in privity for other purposes, such as an
adjudication of rights in other property that was never
transferred between the two. See 18 Wright et al., supra,
Sec. 4462. Put another way, the transfer of a particular
piece of property does not have the effect of limiting rights
of the transferee that are unrelated to the transferred
property. See Munoz v. County of Imperial, 667 F.2d 811, 816
(9th Cir.1982) (concluding that non-parties were not in
privity with a party to litigation because ``[t]he right
which the [third parties] seek to litigate is not one which
they obtained through contractual relations with [a party to
the previous litigation]. It is a completely independent
right[.]'').
Proposed section 327 also imposes important limits on post grant review proceedings. Its requirements are designed to protect both patent owners and the PTO. Section 327 establishes a substantial evidentiary threshold for bringing any post grant review proceeding, and it imposes a further elevated threshold against the bringing of a second-period proceeding for a patent that already has become the subject of such a proceeding. Subsection (a) requires that any petition present evidence that, if unrebutted, would show that a claim in the patent is unpatentable. This threshold is designed, among other things, to force a petitioner to present all of his best evidence against a patent up front. His petition itself must present a full affirmative case. It thus reinforces the front-loaded nature of an oppositional system, which is critical to the efficient resolution of proceedings by PTO. This threshold is considerably higher than ``significant new question of patentability,'' and thus, particularly in combination with the mandates of section 329(c), should provide the PTO with sufficient discretion to protect itself against being overwhelmed by a deluge of petitions.
Subsection (b) of section 327 is designed to allow parties to use first-window proceedings to resolve important legal questions early in the life of such controversies. Currently, for example, if there is debate over whether a particular subject matter or thing is really patentable, parties who disagree with PTO's conclusion that it is patentable must wait until a patent is granted and an infringement dispute arises before the question can be tested in court. In such a situation, subsection (b) would allow parties with an economic interest in the matter to raise the question early in its life. If PTO is wrong and such a thing cannot be patented, subsection (b) creates an avenue by which the question can be conclusively resolved by the Federal circuit before a large number of improper patents are granted and allowed to unjustifiably disrupt an industry. Obviously, subsection (a) alone would not be enough to test the view that PTO has reached an incorrect conclusion on an important legal question, because subsection (a) requires the petitioner to persuade PTO that a claim appears to be unpatentable, and PTO is unlikely to be so persuaded if it has already decided the underlying legal question in favor of patentability. Subsection (a) is directed only at individual instances of error that PTO itself appreciates, while subsection (b) allows PTO to reconsider an important legal question and to effectively certify it for Federal circuit resolution when it appears that the question is worthy of early conclusive resolution.
Subsection (c) of section 327 applies a successive-petition bar of sorts to second or successive petitions for second-period review. It is a rare patent that should be twice subjected to second-window proceedings. Nevertheless, Congress ought not preclude such review entirely. It is possible, for example, that a second-period proceeding may be resolved in a way that suggests that there was some collusion between the petitioner and the patent owner. And PTO may over time identify other circumstances in which even a second or third second- period proceeding is appropriate. Subsection (c) requires that such latter circumstances be exceptional, however.
Lengthy and duplicative proceedings are one of the worst evils of other systems of administrative review of patents. During the pendency of such proceedings, a patent owner is effectively prevented from enforcing his patent. Subsection (c) should ensure that second or successive second-period proceedings are few and far between.
It would be desirable that, when the Director grants petitions, he identify for the parties those issues that he found to be sufficiently established and those that were not. Such a practice would help to expedite proceedings in many cases, as it would limit the issues, and it would also give the patent owner a sense of what issues are important to the board and where he ought to focus his amendments. Ultimately, though, I decided against requiring such practice in the text of the bill. If a mandate were in the statute, it would create problems for the board in the rare but inevitable case where the board initially identifies one issue as the basis for granting the petition, but it later becomes apparent that a different issue is really the central issue in the case. It is better that these proceedings not become as formal as is certiorari practice in the Supreme Court. Nevertheless, it would be helpful to the process and to the parties if the board were to adopt a practice in the ordinary case of identifying the issues that formed the basis of its grant of the petition.
A few words about joinder: section 325 mandates that multiple first- period proceedings be consolidated, and allows multiple second-period proceedings to be so joined. There is no provision in the bill for successive first-period proceedings, so any additional first-period petition that is worthy of being instituted must be joined with the first one. The threshold imposed by section 327, in combination with the mandates of section 329(c), gives the Director the discretion to reject additional first-period petitions that do not add anything new to the case. This section is not intended to make first-period review operate like a notice-and-comment proceeding, in which everyone gets his say and the agency may be buried under an avalanche of repetitive comments.
In the case of both first and second-period proceedings, additional petitions can be joined only if, among other things, they are properly filed. The words ``properly filed'' are a term of art that is also employed in section 2244 of title 28 and that has been given content no less than three times during this decade by the U.S. Supreme Court, see Artuz v. Bennett, 531 U.S. 4 (2000), Pace v. DiGuglielmo, 544 U.S. 408, and Allen v. Siebert, 128 S.Ct. 2 (2007). The gist of these decisions is that a petition is properly filed when it is delivered and accepted in compliance with applicable rules governing filings, though particular claims within filings be barred on other procedural grounds, and that time deadlines for filing petitions must be complied with in all cases.
Where possible, I have sought to make the intended operation of these provisions clear and evident on their face, but the interaction between sections 325(b), 327, and 329(b)(2) requires some explanation. Under 329(b)(2), a request to join a second-period proceeding must be made within a time period to be set by the Director. If the request is so made, the additional second-period petition may be joined to a pending proceeding at the discretion of the Director if he has determined that the additional petition satisfies the threshold set in section 327(a). If the 329(b)(2) deadline is not met, however, the additional second- period petition can still be joined to a pending proceeding at the discretion of the Director if he determines that the additional petition satisfies the threshold set in section 327(c). Section 325(b) requires that a petition be procedurally in order if it is to be considered for joinder, but there is no time deadline that applies to petitions for second-period proceedings, other than that they not be filed before first-period proceedings are concluded. The deadline set pursuant to 329(b)(2) applies only to the motion for joinder, not to the filing of the additional petition itself, and 327(c) expressly contemplates that successive petitions will be filed outside the 329(b)(2) deadline for seeking joinder. Thus a procedurally proper successive petition for second-period review may be joined to a pending proceeding at the discretion of the Director, even if the 329(b)(2) deadline has not been met, so long as the Director determines that the petition satisfies the threshold set in section 327(c).
This is by design. Such a rule encourages petitioners to seek timely joinder to a pending second-period proceeding, but gives the Director discretion to join petitions that meet the successive petition bar even if the request for joinder is untimely. Since an additional petition that satisfies 327(c) would be entitled to its own successive proceeding in any event, it makes sense to allow the Director to join that petition to the pending proceeding, even though joinder was not timely sought.
Section 325(c) gives the PTO broad discretion to consolidate, stay, or terminate any PTO proceeding involving a patent if that patent is the subject of a postgrant review proceeding. It is anticipated, for example, that if a second-period proceeding is instituted and reexam is sought, the Director would be inclined to stay the postgrant review during exhaustion of the reexam. On the other hand, if a postgrant review is near completion, the Director may consolidate or terminate any other PTO proceeding that is initiated with regard to that patent.
Section 329(a)(5) prescribes discovery standards for first-window proceedings, and section 329(b)(3) sets standards for second-period discovery. The standard for allowing second-period discovery is more limited, out of recognition of the fact that the issues that can be raised in that proceeding are few and thus the need for discovery is less. Also, because a second-period proceeding can be instituted long after the patent has issued, it is more burdensome for the patent owner. Limiting second-window discovery limits that burden. Subparagraph (A) of section 329(b)(3) thus allows depositions of witnesses submitting statements, and subparagraph (B) allows further discovery as necessary in the interest of justice. This latter standard restricts additional discovery to particular limited situations, such as minor discovery that PTO finds to be routinely useful, or to discovery that is justified by the special circumstances of the case. Given the time
deadlines imposed on these proceedings, it is anticipated that, regardless of the standards imposed in section 329, PTO will be conservative in its grants of discovery.
Let me comment on two arguments and concerns with regard to second- period review that are not addressed in the text of this bill. First, many parties have made the case to me that any postgrant review of a patent should be limited to a first window that can only be opened within a limited period of time after the grant of a patent. There are strong arguments to be made for this view. Any type of second-period proceeding, whether an opposition or inter partes reexam, invariably interferes with and delays litigation. There is simply no avoiding this result. District judges, many of whom do not enjoy adjudicating patent cases, almost always will stay litigation when a second window has been opened and has the potential to terminate the patent.
I have decided, however, that it would be too radical a step to try to repeal inter partes reexam and not offer any other type of second- period review in its place. As a political and legislative reality, this decision was made in 1999 and probably cannot be undone. To address some of the concerns about a second window, this bill limits such review to the issues that can be raised in inter partes reexam, and includes provisions that are designed to preclude the kinds of tactical and abusive uses of second-period proceedings that are currently seen in inter partes reexam. Though it does not attempt to put the second-period genie back in the bottle, the bill should be an improvement over current law's inter partes reexam. I would welcome a debate about the desirability of second-window review during the next Congress.
Second, a number of parties have expressed concern to me about the current could-have-raised estoppel standard, which I have carried over to second-period proceedings in section 322(d)(2). It is arguable that applying could-have-raised estoppel to the second window does not actually protect the interests that it is designed to vindicate. This estoppel standard's main purpose appears to be to force a party to bring all of his claims in one forum--everything that he ``could have raised''--and therefore to eliminate the need to press any claims in other fora. In this bill, however, the issues that can be raised in the second window are so sharply limited that the goal of flushing out all claims is unattainable. Only 102 and 103 arguments based on patents and printed publications can be raised in the second window. Accused infringers inevitably will have other challenges and defenses that they will want to bring, and those arguments can only be raised in district court. Regardless of the estoppel standard that is applied, the patent owner will almost always be forced to fight in two fora, and the intended goal of could-have-raised estoppel will remain beyond reach.
The real reforms in this bill that would protect patent owners from abusive and duplicative proceedings are the various restrictions imposed in section 327 and in subsections (a), (b), and (c) of section 322. These provisions, I think, would be more useful and valuable to patent owners than could-have-raised estoppel. I welcome a broader debate on this issue. At the very least, it would be helpful to me to more clearly understand the interests that proponents and opponents believe are protected or injured by could-have-raised estoppel.
Section 8 of the bill addresses venue. It adopts an activities-based test for determining whether a particular district is an appropriate locale for a patent-infringement suit. Under section 8's proposed amendments to 28 U.S.C. section 1400, some significant activity involving either the patent or the infringing product must take place in the district in order for venue to be proper there. This section aims to limit patent litigation to districts with some reasonable connection to the patent, but without generating substantial preliminary litigation over venue. Of course, any change to the venue statute will result in a period of litigation over the new statute's meaning. To the extent possible, section 8 uses terms of art that have a settled meaning in the venue context.
Paragraph (2) and subparagraphs (B) and (C) of paragraph (6) refer to acts of infringement and to a product or process that embodies an invention, events or facts whose existence likely will be the subject of the litigation. I considered whether the word ``allegedly'' should be added before ``infringement'' or ``embodies,'' since those facts will not yet have been proven at the time when venue is being determined. Current section 1400(b), however, refers simply to ``acts of infringement.'' I am unaware of any courts that, when applying the current law, have required the plaintiff to demonstrate that infringement has in fact occurred before allowing themselves to be persuaded that venue is proper. I would expect courts and litigants to also use common sense when applying paragraphs (2) and (6), and to not construe the language to require that the merits of the case be litigated before a threshold question may be determined.
Paragraph (4) refers to the place where an invention was conceived. This can, of course, be more than one place and can involve collaborative activities.
Paragraphs (5) and (6)(A) refer to ``research and development.'' Other patent venue reforms that have been proposed in this Congress have referred to research or development, treating the two words as if they were separate concepts. In most circumstances, however, research and development are treated as one thing and no effort is made to distinguish research from development. Although theoretical distinctions are possible, they become very difficult to apply to actual practical situations. Thus section 8 treats research and development as a unified concept.
Paragraphs (5) and (6)(A) also refer to ``significant'' research and development. This bill uses the word ``significant,'' rather than the word ``substantial,'' which is a word that has been used in other legislative proposals made in this Congress. Having reviewed judicial constructions of both terms, it appears to me that ``significant'' means something like ``legitimate,'' and that the significance of an activity can be evaluated on the face of that activity, without reference to the whole of which it is a portion. The word ``substantial,'' on the other hand, appears to measure an activity in light of the whole of which it is a part. Arguably, one cannot know whether particular research-and-development activity is substantial without knowing all of the research-and-development activity that has taken place with regard to the patent in suit. Using the word ``substantial'' here or elsewhere in this section likely would in many cases require discovery to determine just what is the whole of which the activity in question is alleged to be a substantial part. Since the last thing that I would want to be responsible for is a patent law that made discovery and a 2-day evidentiary hearing a routine feature of establishing venue in patent litigation, my bill uses the word ``significant'' rather than ``substantial.''
Paragraph (7) allows venue at the place where a nonprofit organization managing inventions for colleges and universities, including the patent in suit, is principally based. These organizations manage inventions by, among other things, helping the schools to commercialize them. Whether such an organization acts on behalf of a university should not be construed to turn on whether there is an agency relationship between the organization and school. Even an independent contractor acts on behalf of the party that has retained it.
A few words about interlocutory appeals: I expressed skepticism in the committee report to S. 1145 about requiring the Federal circuit to accept interlocutory appeals of claim constructions. I noted that such a rule risked allowing a district judge who is insufficiently enthusiastic about his duty to decide patent cases to rid himself of a case by certifying an interlocutory appeal to the Federal circuit, in the hope that the case would go away and never come back. Not only would such an event waste the Federal circuit's resources, it would also force that circuit to decide a claim construction on the basis of what may be an inadequate evidentiary record. And no matter how thin that record may be, once the claim construction was before the Federal circuit and that court were forced to decide it, whatever came back to the district court would be the law of the case. The Federal circuit's claim construction could not be
changed by the district court on remand, no matter how obvious it later became in light of a more complete record that the Federal circuit had gotten it wrong.
I have heard from more than one patent lawyer that claim construction often is a rolling process. Even when a court holds a Markman hearing and attempts to definitively construe a patent early in a trial, frequently new information comes forward over the course of the trial that sheds new light on claim terms, or it becomes clear that different claim terms constitute the heart of the dispute and must be construed. An interlocutory appeal would prove to be a large waste of time if it later became clear that different claim terms formed the heart of the dispute. And such an appeal could prove to be an utter disaster if the Federal circuit were forced to construe the key claim terms without having all of the necessary information before it and, as a result, that court misconstrued those claims. Because of the great risk of such undesirable outcomes, and the delay that interlocutory appeals would inject into trials, I have not included a proposal to require interlocutory appeals in this bill.
Section 10 of the bill addresses applicant quality submissions. PTO believes that all applicants for a patent should be required to conduct a search of prior art and a patentability analysis before they submit their patent application. Such a requirement not only would improve the quality of applications, it would also persuade many would-be applicants not to file in the first place, since they would discover that their invention already is disclosed in the prior art.
PTO presents a strong case that the patent system currently is buckling under the volume of applications, and that if present trends continue, in 10 years the system could be brought to the point of collapse. Today, many applications provide little useful information to examiners and are filed without any awareness of the prior art. Some have suggested that PTO simply needs to hire and retain more examiners, but there are natural limits to PTO's ability to hire, train, and assimilate new examiners into the culture of PTO. Already PTO is hiring a significant percentage of every year's graduating class in particular fields of engineering. If something does not change, Congress may find it necessary to mandate across-the-board search-and patentability requirements in the future.
PTO urged the adoption of search-and-patentability requirements during this Congress. The ability of such proposals to secure acceptance from the relevant interests ultimately foundered, however, on our inability to answer several key questions about how such a system would function and how much it would cost. The types of searches that PTO performs, for example, are rather specialized. Many patent applicants would want to hire a search firm to conduct such searches rather than learn how to conduct PTO searches themselves. Currently, however, no market exists for such services and no firms exist that offer to conduct searches that would meet PTO's specifications. It is thus impossible at the moment to say with certainty how much patent applicants can expect to pay to have a private firm conduct a search that meets PTO's requirements.
It also is unclear exactly what kind of patentability analysis PTO might want. It will probably be necessary for PTO to launch such a system and to adjust it over a period of years before PTO itself discovers what kinds of requirements produce information that is useful to the Office.
And finally and most importantly, under the current system, in which statements made by the applicant during prosecution are used to construe the claims of the patent in district court, any requirement that the applicant make additional statements about patentability during prosecution would prove to be very expensive to the applicant. Under the current litigation regime, applicants who can afford to do so would be wise to hire expensive patent lawyers to think through how every statement made to PTO during a patentability analysis might later affect claim construction in an infringement suit. In other words, a patentability analysis requirement likely would result in heavy legal costs for patent applicants.
Rather than mandate that all applicants submit a search report and a patentability analysis, section 10 of the bill authorizes PTO to offer incentives to parties who do so, and it makes the prosecution record of a patent that is secured through such a program inadmissible to construe patent claims in later proceedings. This last requirement is both an essential prerequisite to the palatability of a voluntary search-and-patentability program, and is also expected to be a powerful draw to applicants to participate in the program. By effectively providing immunity in later litigation against all information that is in the file wrapper of the patent's prosecution history, this provision allows applicants to speak freely with examiners, without having to constantly think through--or rather, have their lawyers think through-- how each statement might later affect claim scope in subsequent litigation. I also anticipate that the prospect of being able to assert a patent based solely on its claims, without having to litigate over the meaning of every action and statement in the prosecution record, will be a strong inducement to many patent applicants to try to comply with the PTO's voluntary search-and-patentability program.
Proposed section 123(b) also authorizes PTO to issue regulations identifying material submitted in an attempt to comply with the search- and-patentability program that also shall receive file-wrapper immunity. Such regulations should encourage applicants to try PTO's system who might otherwise be deterred by fear that if they try to comply with PTO's program and abort the attempt or are unsuccessful and later secure the same patent by the conventional route, the possibly substantial record produced during the failed attempt will later be used in litigation to limit claim scope. And of course, even ultimately successful users of the search-and-patentability program who are not confident that they will complete the program likely would, in the absence of the immunity tendered by such regulations, engage in the very type of defensive and overlawyered discussions with the examiner that the prospect of file-wrapper immunity is designed to prevent.
Proposed section 123(a) authorizes PTO to offer various other incentives to parties who participate in a search-and-patentability program. Subsection 10(b) of the bill is intended to preclude a negative implication that because the bill authorizes PTO to offer such incentives, PTO must currently lack the authority to offer incentives to applicants who submit additional information. I should also note that PTO may continue to offer incentives to applicants under existing pilots and programs without issuing regulations.
Section 10 of the bill is designed to allow a substantial trial run of a search-and-patentability program. It is my hope that if the incentives offered are powerful enough and if PTO's search-and- patentability demands are reasonable, eventually a major portion of all patent applicants will choose to prosecute their patents under such a system. A well-functioning and heavily used search-and-patentability program not only would help PTO to process its backlog of applications, it also would answer some of the questions that we were unable to answer this year, such as how much would private prior-art searches cost, and will file-wrapper immunity operate as intended in court?
I hope that the gathering patent-application storm that PTO perceives will be diverted by the program authorized in this section and by the reforms to the inequitable-conduct doctrine in section 11 of the bill, both of which should encourage applicants to be more frank with PTO and to provide information that is more useful to the Office. If present filing trends continue for another decade, however, and Congress is forced to consider applying search- and patentability-analysis requirements across the board to all applications, it likely will have proven useful to have had a substantial trial run of a search-and- patentability program.
Section 11 of the bill addresses the doctrine of inequitable conduct. Under current law, this doctrine allows an accused infringer to have an entire patent declared unenforceable if he can demonstrate that when the patent was
prosecuted, the patent applicant intended to deceive the examiner by misrepresenting information that the court deems material under one of a variety of tests, such as whether the information would be important to a reasonable patent examiner in deciding whether to allow the application. See, e.g., Digital Control, Inc. v. Charles Machine Works, 437 F.3d 1309, 1313-14 (Fed. Cir. 2006). This doctrine, which is applied in the course of infringement litigation, is a court-made doctrine that is designed to force patent applicants to be forthcoming and to not mislead the PTO when prosecuting their patents. In practice, however, the doctrine does not fulfill this purpose and instead generates a variety of undesirable consequences.
There are two aspects of the current inequitable conduct doctrine that I find particularly troubling. The first is that it is asserted in a majority of all patent lawsuits. As much as one might think ill of the ethics of particular industries, it is simply inconceivable that fraud and other misconduct infects anything close to half of all of the patents issued in this country.
One explanation that a number of lawyers have given to me for the high rate at which inequitable conduct is asserted in litigation is that the doctrine gives the accused infringer an opportunity to examine the inventor--often in the jury's presence--and to paint him as deceptive and dishonest. Even the most upright and honest inventor can be made to look sly and shifty under aggressive examination as to why exactly he chose not disclose particular facts or documents to the PTO. And thus even an infringer who has no reasonable hope of prevailing on an inequitable-conduct claim will assert the doctrine simply because it offers an opportunity to cast the inventor and his work in a negative light. This tactic tends to increase the odds that the jury will find the invention obvious and to decrease the jury's estimate of the damages to which the inventor is entitled.
The doctrine also carries high transaction costs. It typically is grounds for exhaustive discovery of the inventor's files and for depositions directed at his state of mind at the time of the prosecution--for questioning him as to what did he know and when did he know it, and what was his motive for not disclosing particular pieces of information. The doctrine adds substantially to the expense of litigation.
The other aspect of the current doctrine that I find problematic is that it applies a draconian penalty to instances of misconduct whose materiality often appears to be doubtful. Jon W. Dudas, the Under Secretary of Commerce for Intellectual Property and Director of the United States Patent and Trademark Office, commented on this aspect of the doctrine in his testimony before the Judiciary Committee on June 6, 2007:
Under existing case law, courts must hold all of a patent's
claims invalid if they find inequitable conduct in any aspect
of prosecuting a patent application even if the claims are
completely valid and/or the inequitable conduct was
irrelevant to prosecution of the claims. Thus, the only
remedy available is complete loss of the patent. Inequitable
conduct can be found if the applicant deliberately withholds
or inaccurately represents information material to patent
prosecution. Anything the court deems that a reasonable
examiner would find important can be material and the
evidence necessary to show intent varies according to the
nature of the omission. Accordingly, the inequitable conduct
standard is uncertain and the potential penalties severe. For
example, any misstatement in an affidavit, or even a failure
to disclose a possible source of bias, has been held to be
capable of rendering all claims of the patent unenforceable.
Because inequitable conduct is a court-enforced doctrine, the assessment of what is material--of what would have been important to a reasonable patent examiner--is made by a U.S. district judge. But district judges very rarely have any firsthand knowledge of the patent- prosecution process or the workings of the PTO and are not in a position to accurately assess what information actually would have been important to a reasonable examiner.
The Federal courts' sometimes hair-trigger assessments of materiality are a substantial injustice to those patent owners who lose the right to enforce what is an otherwise perfectly valid patent. This injustice can be particularly acute when the current owner of the patent is a good-faith purchaser who is not even alleged to have engaged in any type of misconduct himself.
Judicial enforcement of the doctrine of inequitable conduct also has led to consequences that are of a more general concern. The doctrine's severe penalty, combined with the unpredictability of its application, has led applicants to adopt extreme tactics that are designed to eliminate the risk that their patent will ever be held unenforceable on the ground of inequitable conduct. These tactics, while perhaps effective at minimizing such risk, are inconsistent with sound prosecution practice. They constitute the exact opposite of providing PTO with the information that it needs in order to be able to assess whether a claimed invention is patentable, and they make it harder for PTO to do its job. Under Secretary Dudas commented on this phenomenon in his June 6, 2007 Judiciary Committee testimony:
In some other cases, applicants or their attorneys fear
that the legal doctrines of inequitable conduct and
unenforceability may unfairly punish them with draconian
penalties for innocently omitting information. The theory is
that, if one does provide information, it must be perfect.
Otherwise, the consequence may be loss of the patent and/or
disciplinary action (for the applicant's attorney). By way of
contrast, failure to share or disclose information has
absolutely no adverse legal consequence.
* * * * *
While the risk of an inequitable conduct finding is low, it
is frequently alleged. When alleged, inequitable conduct
assertions add substantially to litigation costs and
malpractice claims. The ``all or nothing'' result of an
inequitable conduct finding understandably has a perverse
effect on the actions of applicants and their attorneys with
respect to ``risking'' a proper search in the first place. As
a result, the doctrine results in counterproductive behavior
before the USPTO. It discourages many applicants from
conducting a search and leads others to be indiscriminate in
the information they submit. In a review two years ago, we
found that over 50 percent of submitted applications
contained either no information disclosure statement or that
such submissions included more than 20 references.
The Under Secretary's testimony is consistent with what has been described to me by a number of attorneys and patent applicants. The current state of inequitable conduct enforcement leads applicants to adopt one of two tactics: either they flood the Office with prior-art references but offer no explanation of how the invention is distinguished from that prior art or which prior art is most relevant, since by providing the reference they cannot be accused of concealing it, and by providing no explanation they cannot be accused of misleading the Office or mischaracterizing the information, or applicants provide no information at all with their applications, since providing some information would inevitably mean not supplying other information in the universe of existing information and thus could open the applicant to charges of having concealed something in that universe of information not provided. Both tactics impede the PTO's examination of patent applications.
Professor John F. Duffy of George Washington University Law School has made a persuasive case that inequitable conduct that occurs during patent prosecution should be addressed in proceedings before the PTO itself. He notes that the 1940s decisions that are viewed as giving the Supreme Court's imprimatur to judicial enforcement of the doctrine are much more limited in their rulings than the expansive approach to inequitable conduct that has been developed by the Federal circuit. He also points out that the patent system's use of civil litigation to enforce good conduct in dealings with an agency is unique to the patent system. In the case of every other Federal administrative agency, the agency itself polices misconduct and fraud committed in agency proceedings.
Professor Duffy also notes that in other administrative contexts, the Federal courts themselves have predicted that judicial supervision of agency proceedings would produce the very consequences that judicial intervention has produced in the PTO. Though Buckman Co. v. Plaintiffs' Legal Committee, 531 U.S. 341, 351 (2001), is a case about the FDA, it might as well be describing the impact of the inequitable-conduct doctrine on patent prosecutions:
[F]raud-on-the-[agency] claims inevitably conflict with the
[agency's] responsibility to
police fraud consistently with the Administration's judgment
and objectives. As a practical matter, complying with the
[agency's] detailed regulatory regime in the shadow of [the
courts' varying fraud standards] will dramatically increase
the burdens facing potential applicants * * *.
Conversely, fraud-on-the-[agency] claims would also cause
applicants to fear that their disclosures to the [agency],
although deemed appropriate by the Administration, will later
be judged insufficient in * * * court. Applicants would
then have an incentive to submit a deluge of information that
the Administration neither wants nor needs, resulting in
additional burdens on the [agency's] evaluation of an
application. As a result, the [agency certification] process
could encounter delays, which would, in turn, impede
competition * * * and delay [innovation].
Section 11 of the bill that I have introduced proposes a new approach to addressing misconduct in proceedings before the PTO. It effectively shifts enforcement of the doctrine of inequitable conduct from civil litigation to administrative proceedings before the PTO. Under the procedures authorized in proposed sections 298 and 299, PTO will reissue patents if needed to remove any invalid claims, will assess the culpability of any misconduct, and will impose sanctions on any parties that have engaged in inequitable or fraudulent conduct before the Office.
I believe that the administrative framework proposed in section 11 is consistent with the principles outlined in the Supreme Court cases that the Federal circuit relies on as the basis for its own inequitable conduct jurisprudence, Precision Instrument Manufacturing Co. v. Automotive Maintenance Machinery Co., 324 U.S. 806 (1945), and Hazel- Atlas Glass Co. v. Hartford-Empire Co., 322 U.S. 238 (1944). Section 298 would require district courts to order patents that are infected by fraud to go into reissue proceedings, where invalid claims would be removed. Limiting patents to their proper scope serves important public interests. As the court noted in Precision Instrument, at pages 815 to 816, citations omitted:
The possession and assertion of patent rights are issues of
great moment to the public. As recognized by the
Constitution, [a patent] is a special privilege designed to
serve the public purpose of promoting the ``Progress of
Science and useful Arts.'' At the same time, a patent is an
exception to the general rule against monopolies and to the
right to access to a free and open market. The far-reaching
social and economic consequences of a patent, therefore, give
the public a paramount interest in seeing that patent
monopolies spring from backgrounds free from fraud or other
inequitable conduct and that such monopolies are kept within
their legitimate scope.
Proposed section 299 would authorize procedures whereby the PTO can receive and assess complaints about misconduct committed by parties to its matters or proceedings, assess the materiality of the misconduct and the mens rea of the malfeasant, and levy appropriate sanctions, including civil fines and, in severe cases, unenforceability of the patent. This section is animated by the principles expressed in Precision Instrument, at page 818, where the court emphasized that:
Those who have applications pending with the Patent Office
or who are parties to Patent Office proceedings have an
uncompromising duty to report to it all facts concerning
possible fraud or inequitableness underlying the applications
in issue. * * * Public interest demands that all facts
relevant to such matters be submitted formally or informally
to the Patent Office, which can then pass upon the
sufficiency of the evidence.
A few provisions of proposed section 299 deserve some commentary and explanation. Subsection (a) authorizes the PTO to issue regulations accepting complaints from any source. It is anticipated, based on preliminary discussions with the Office, that the PTO will accept complaints from a broad range of parties, including those that are third parties to any commercial disputes involving the patent. The scope of such regulations, however, ultimately remains within the Office's discretion, and PTO may later decide to limit who may file a complaint should it discover that allegations of misconduct that originate from particular types of sources are burdensomely voluminous or otherwise unproductive.
Though any person may file an allegation of misconduct under section 299, that section only allows such complaints to be filed against individual and entities that are parties to matters or proceedings before the Office. This limitation excludes examiners and other PTO personnel. Prosecutions occasionally become contentious, particularly when examiners fail to appreciate an inventor's revolutionary genius. If section 299 were not limited to complaints against parties, we would run the risk that such proceedings might come to be regarded by a subset of applicants as their final means of appealing an examiner's rejection.
Section 299 is not limited, however, to entertaining complaints against applicants and patentees. A party that engages in intentionally deceptive and material misconduct while challenging a patent during a postgrant review proceeding, or even while requesting such a proceeding, also may be sanctioned pursuant to section 299.
Some parties have criticized the fact that the proceedings authorized by section 299 will be prosecuted by the PTO alone, without the participation of parties adverse to the patent. PTO prefers it this way. If misconduct has resulted in the grant of claims that are invalid, that patent can still be challenged in court if its owner attempts to enforce it. And to the extent that alleged misconduct has not resulted in the grant of claims that are invalid, the interests principally affected by any misconduct are those of PTO. The primary injury in such a case is to PTO's interest in ensuring that parties are honest and forthcoming in their dealings with the Office and its general interest in the integrity of its proceedings. In such circumstances, it is appropriate that PTO control the prosecution of the misconduct.
Subsection (b)(3)(C) of section 299 permits PTO to sanction a patent owner by rendering his patent unenforceable. That penalty, however, is reserved by subparagraph (C) for particularly egregious misconduct that was committed by the current beneficial owner of the patent.
This elevated standard is consistent with the standards for unenforceability set in Precision Instrument and Hazel-Atlas Glass, the foundational Supreme Court cases of the modern inequitable-conduct doctrine. In Precision Instrument, an applicant ``gave false dates as to the conception, disclosure, drawing, description and reduction to practice of his invention.'' When his fraud was discovered by the other party to an interference proceeding, the applicant colluded with that other party to assign the false application to the party. The Supreme Court held the patent unenforceable, concluding that ``[t]he history of the patents and contracts in issue is steeped in perjury and undisclosed knowledge of perjury'' and that ``inequitable conduct impregnated [the patentee's] entire cause of action.'' Pages 809, 816, and 819. Similarly, in Hazel-Atlas Glass, the court rendered a patent unenforceable upon ``conclusive proof'' of a ``deliberately planned and carefully executed scheme to defraud not only the Patent Office but the Circuit Court of Appeals.'' The court also emphasized in that case that ``no equities have intervened through transfer of the fraudulently procured patent or judgment to an innocent purchaser.'' Pages 245 and 246.
I should also comment on a few other significant changes that this bill makes to S. 1145. My bill's proposed section 102(a)(1) amends the novelty condition of patentability by eliminating public use and the on-sale bar as independent bases of invalidity and instead imposes a uniform test of whether art has been made available to the public. By eliminating confidential sales and other secret activities as grounds for invalidity and imposing a general standard of public availability, this change will make the patent system simpler and more transparent. Whether a patent is valid or not will be determined exclusively on the basis of information that is available to the public. As a result, at the outset of any dispute over a patent, the patentee and potential infringer can develop a full and complete understanding of the information that will determine the novelty and nonobviousness of the claimed invention. This change not only will provide greater certainty and predictability--it should also substantially reduce the need for discovery in patent litigation, since defendants will no longer need to uncover evidence of private sales or offers for sale or other nonpublic information in order to determine whether the patent is valid.
It bears mention that the extent of what is deemed to be publicly available
is defined in important respects by the doctrine of inherency. Under that doctrine, once a product is sold on the market, any invention that is necessarily present or inherent to the product and that would be recognized as such by a person skilled in the art is itself deemed to be publicly available. Such an invention becomes publicly available art and cannot be patented. See generally Rosco, Inc. v. Mirror Lite Co., 304 F.3d 1373, 1380-81 (Fed. Cir. 2002).
To address the possible concern that a uniform available-to-the- public standard might allow secret commercialization of a product followed by belated patenting, I should note that a manufacturer who embarked on such a course would run the risk that, under the first-to- file system, someone else might patent the invention out from under him. Perhaps for this reason, among others, industrialized countries that currently employ this standard do not appear to have experienced significant problems with manufacturers attempting secret commercialization and late patenting of their products.
The bill also includes other provisions that would make the patent system more objective and transparent. Section 3(c) eliminates current law's best-mode requirement, and section 15 strikes several provisions of title 35 that require inquiry into a patentee's subjective intent. Any useful information that might be supplied by describing a patent's best mode generally also will be provided while satisfying the written description and enablement requirements. And because the best-mode requirement turns on the patentee's subjective intent, rather than on objective facts, it often becomes grounds for deposition of the inventor and other discovery. Eliminating that requirement will make patent litigation less burdensome.
My bill also strikes S. 1145's elimination of the exception to the 18-month publication requirement. Small-patent-owners' groups have persuaded me that the current exception should be preserved. That exception, although used only about 40,000 times annually, is invoked heavily by small-business applicants. These smaller applicants believe that the opt-out of 18-month publication allows them to preserve the market advantage generated by their ingenuity, and prevents their inventions' being appropriated in foreign countries, in the event that their application is not granted or is only granted on a second attempt. Under Secretary Jon Dudas, in his June 6, 2007, Judiciary Committee testimony, also expressed doubt about the wisdom of eliminating the current exception. He noted that serious concerns had been expressed ``by independent inventors and small entities that large entities and foreign interests may misappropriate their inventions upon disclosure and prior to issuance of a patent.''
Sections 12 and 13 of the bill are carried over from S. 1145 as reported by the Judiciary Committee. I have included additions to those sections that I understand that their supporters had intended to adopt and have also made an addition of my own to section 12. The new subsection (c) in that section converts various day-based deadlines in title 35 into month-based deadlines. Month-based deadlines are easier to calculate. The use of months should make it easier to avoid the type of ministerial mistake that apparently is the cause for section 12. It should also save the patent system hundreds of billable hours over the years.
Section 2(b) of the bill includes a minor modification to the CREATE Act, Public Law 108-453. This change more closely aligns the text of that act to the PTO's current and uncontested interpretation of that act with regard to who must own the prior art that is regarded as jointly owned by the parties to a joint research agreement pursuant to the CREATE Act.
And last, but certainly not least, section 14 of the bill consists of the Coburn amendment, which would create a revolving fund for PTO fees. Under that amendment, all fees paid by patent and trademark applicants and owners to the PTO would remain in the PTO and could not be diverted to unrelated Government programs.
According to Senator Coburn, the fees collected by PTO are more than adequate to pay for the costs of all patent examinations and other PTO proceedings. But PTO is not allowed to keep those fees. Instead, the fees are deposited into the U.S. Treasury, and PTO's operations are funded by a congressional appropriation. It is that appropriation that effectively determines on an annual basis what portion of the fees that PTO has collected it will be allowed to keep and use.
Since 1992, Congress has diverted over $750 million in PTO fees to other governmental programs. As recently as 2004, over $100 million was diverted from the PTO.
Fee diversion unquestionably has a negative impact on the patent system. In recent years, it has hampered PTO's ability to hire an adequate number of examiners. Multiple studies and multiple witnesses at congressional hearings have concluded that fee diversion contributes to the growing backlog and lengthening pendency of patent applications. It currently takes nearly 3 years to get a patent, and 786,000 applications are pending. That means that large numbers of businesses, universities, and other inventors are waiting to learn if they will receive a patent for their invention.
Because of recent public outcry over lengthy patent-application pendency periods, the administration and Congress have abstained from diverting PTO fees since 2004. As a result, PTO has been able to hire a record number of new examiners and begin to address its backlog of applications. Unless the Coburn amendment is enacted into law, however, Congress and the administration could easily begin diverting PTO fees again in future years. Certainly, any bill that aspires to deserve the title ``Patent Reform Act'' should include a revolving-fund provision.
I thank all of the individuals who have assisted my attempts to understand and find answers to the difficult questions posed by efforts to improve the patent system, and I look forward to next year's congressional debate on patent reform legislation.