Prof. Doug Lichtman, UCLA School of Law
December 6, 2010
The Patent Act provides that an issued patent “shall be presumed valid,” but the statute fails to specify exactly how strong that presumption should be, and it fails to specify against exactly what evidence that presumption should weigh. The Supreme Court last week granted cert in Microsoft v. i4i in order to address one part of that puzzle. Specifically, the Court agreed to consider whether an accused infringer must prove patent invalidity by “clear and convincing evidence” in a case where the relevant prior art was not considered by the Patent and Trademark Office prior to the original issuance of the asserted patent.
I will have more to say about i4i in the weeks ahead; and, indeed, my own view is that cert was improvidently granted in this case, because i4i presents a terrible fact pattern from which to evaluate this important legal issue. For now, however, I want to put aside the details of the case and instead stake out a basic and broad proposition: Courts should not much defer to the Patent Office’s initial determination that
a particular invention qualifies for patent protection, because patent review cannot be particularly accurate at the application stage.1
One issue is the budget. Several hundred thousand patent applications are filed every year, and those applications cover the full range of technologies – from breakthroughs that involve the human genome to innovative new designs for consumer electronics. Patent examiners who are assigned to evaluate those applications are chosen because they have background roughly related to the technology at hand, but examiners are rarely experts on the precise details of the relevant inventions.
Thus, to evaluate an application, an examiner not only has to read the typically voluminous documentation submitted by the applicant, but the examiner also must use computerized databases and other available sources to learn about the state of the art. The examiner in addition has to interact with the applicant’s lawyers and document any decisions ultimately made. Yet, strikingly, examiners are asked
to do all this in what turns out to be an average of less than 20 hours total. Worse, those hours are spread over what is often a several-year period.
A lack of information is a second significant impediment to Patent Office review. Patent applications are evaluated early in the life of a claimed technology, and thus at the time of patent review there is typically no publicly available information about (say) how well the technology has been received by experts in the field, or whether consumers have deemed the technology to represent in some way an advance over existing alternatives. Patent examiners cannot solicit these sorts of credible outsider opinions either, not only because for many technologies it is unclear at the early stages who the relevant experts and customers might be, but also because patent evaluation is for the most part a confidential conversation between
applicant and examiner, designed to keep an applicant’s work secret just in case the patent application is ultimately denied.
Given all this, it is hardly a surprise that the Patent Office makes mistakes during the initial process of patent review, granting patents that, on the merits, should never have been issued. The real surprise is that the presumption of patent validity makes these issuance mistakes unnecessarily difficult to reverse.
The theoretical justification is that patent examiners have expertise when it comes to questions of patent scope; thus, if patent examiners have decided that a given invention qualifies for protection, judges and juries should not second-guess the experts. But the reality is that Patent Office expertise is brought to bear under such poor conditions that any advantages associated with expertise are fully overwhelmed by the disadvantages associated with insufficient funding and inadequate outsider information.
Contrast that to court review, where information is a natural product of the adversarial process, and where financial constraints are reduced because only a tiny fraction of all issued patents end up sufficiently valuable and contentious to warrant litigation. Thus, the presumption of validity backfires. Rather than
protecting accurate initial decisions from inefficient later meddling, the presumption precludes what would often be a worthwhile second look.
The consequences are significant. When the patent system works properly, a patent holder earns a living first by first patenting a genuine invention and then telling potential customers about the technology. The patent in this instance protects the inventor from having his idea stolen, but the patent is worth nothing unless and until the associated inventor can find customers for his idea. The system thus encourages both the creation of new ideas and their dissemination.
Patents that are issued wrongly, however, do not remotely follow this pattern. A patent holder whose patent covers a technology that was already obvious to those skilled in the art has a strong incentive to sit quietly after the patent is issued, knowing full well that other parties will stumble into that same obvious technology in time. When that happens, the patent holder can step forward, threaten litigation, and in the end extract royalties from infringers who neither knew of nor remotely benefited from the patent holder’s
work. Sadly, a large and growing number of “patent trolls” today play this exact strategy, using patents on obvious inventions quite literally to tax legitimate business activity.
If the courts were to discard the current “clear and convincing” standard and replace it with something appropriately more modest, patent examiners would still play their customary role in terms of evaluating claim language and ensuring that applicants comply with the patent system’s many rules about the form and content of patent disclosures. Patent examiners would also continue to weed out the most egregious applications and, in various ways, force inventors to commit up-front to details about their claimed accomplishments. Patent examiners, however, would no longer themselves make a definitive ruling with respect to validity. An examiner would document his reasons for allowance, and then the substance of that argument would be asked to carry the day – rather than the mere fact of the prior decision.
The primary beneficiaries of this reform would not be patent holders. Patent holders would benefit, sure – the underbrush of undeserving patents today undermines the value of well-earned patents – but the primary beneficiaries would be the countless firms that, in the course of putting out some product or service, might inadvertently infringe a patent. These firms need the patent system to ensure that only genuine inventions are awarded patent protection, because these firms are the ones that end up paying royalties or in other ways having their businesses disrupted every time some obvious idea is nevertheless allowed to fall within a patent holder’s exclusive rights.