Stop Using PTAB Invalidation Rates as a Metric for Measuring Patent Quality

Every experienced patent attorney knows what is written below is true, but if you want the quiet parts to be spoken loudly, keep reading. This article isn’t revolutionary, but it screams a truth that everyone with an interest in patents needs to understand.

Patent invalidation rates by the Patent Trial and Appeal Board (PTAB) are often used to attack the quality of examination by the U.S. Patent and Trademark Office (USPTO). While I agree that there is room for improvement in examination quality, patent invalidation rates should not be used as a barometer for how well the USPTO is examining patent applications.

Plain and simple – invalidation rates are misleading. To understand why, let’s look at the practicalities of patent examination and the realities of the patents that are challenged at the PTAB. In short, to increase certainty of patentability to a degree that would substantially lower invalidation rates at the PTAB, the USPTO would have to spend hundreds of hours examining each patent application. This is unrealistic and contrary to public interest.

Patent Examination at the USPTO

For those not familiar with patent examination at the USPTO, the Office is divided into nine different technology centers. Each tech center includes hundreds of individual patent examiners. All patent examiners have a college degree in science or engineering and most have a technical background matching the subject matter of the tech center. Some patent examiners have worked in the public or private sector, while others came straight to the USPTO after graduation from college.

For example, the examiners in tech center 1600 examine patent applications related to biotechnology and organic chemistry fields and those in tech center 2100 examine patent applications related to computer architecture software and information security. Each tech center is further divided into art units, and each examiner generally examines applications in one or two art units. Art units are specific sub-categories of the technology of the tech center, and are designed such that, over time, examiners become experts in particular subject matters.

Limited Resources Allocated to Patent Examination

Each examiner is assigned a specific amount of time to examine a patent application and issue an action in response to the patent application. Depending on the experience level of the examiner, the complexity of the technology and other factors, it is my understanding that the examiner will spend between 10 and 45 hours examining an application in the first instance to determine whether the claimed invention should issue into a patent. This action is usually a rejection, but a skilled patent attorney can often turn that rejection, or the next rejection, into a patent. On average, utility patents issued in the first half of this decade averaged 1.7 rejections per patent, according to Juristat. Thus, to make this analysis simplistic, we will estimate that an average patent was subject to approximately 17 to 77 hours of examination time.

Which Patents are Invalidated by the PTAB?

Whether or not a granted patent’s validity matters to the public depends on whether the technology is valuable. For illustrative purposes, let’s compare (1) a patent that does not cover any product in the market, (2) a patent that covers one or more products generating $500,000 in annual revenue, or (3) a patent that covers one or more products generating $20 million in revenue.

Patent (1) is often narrowly tailored and has no value. Whether it is “valid” does not matter at all because its validity will never be even considered.

Patent (2) might be relevant within a niche industry and it might be the subject of a cease and desist letter sent by the patent owner to a competitor. There may be a license agreement, the competitor may design around the patent, or the accused infringer may just say “screw you, sue me.”  But there is almost zero chance that an infringement lawsuit is filed or that the patent is challenged at the PTAB. Thus, the validity is also never analyzed by the PTAB or a district court.

Patent (3) will likely be litigated if it is not licensed, and if it is litigated, the defendant will file a petition at the PTAB challenging the validity of the patent.

A vast majority of patents are like patents (1) and (2), and exceedingly few patents are in the same sphere as patent (3). In the first half of this decade, around 330,000 utility patents have been issued each year, and around 1,360 utility patents are challenged each year at the PTAB. Accordingly, very few, yet valuable, utility patents are challenged at the PTAB.

What Kind of Resources are Thrown at Patents That are Being Litigated? 

As I mention above, an average patent is subject to approximately 17 to 77 hours of examination time. Do you think an accused infringer, who could be on the hook for millions of dollars, says to their patent lawyers, “Spend less than 100 hours seeing if we can invalidate this patent”? Hell no, they don’t!

They throw substantial resources into finding prior art that the Examiner might have missed. They direct their lawyers to spend hours crafting reasoning as to why it would have been obvious to have combined the prior art references. They hire expensive expert witnesses and pay their lawyers to brainstorm on how to fill in any evidentiary gaps in the prior art. Almost all inventions are a combination of using previously existing technology in a new way, so skilled lawyers, especially with the help of hired-gun expert witnesses, can always craft solid arguments that claims of a patent are obvious.

Well, is it Obvious?

I am not suggesting that the behavior of challenging the validity of a patent is necessarily ill-intentioned or wrong. Obviousness, which is usually what is being considered in a validity challenge, is inherently subjective and subject to hindsight bias. If a patent was granted years ago on a technology that is now valuable, the natural instinct is to skeptically view its inventiveness. If a party is accused of infringement, their inclination is that the claims of the patent are obvious, especially if there are economic incentives to challenge the patent versus paying a licensing fee.

I also view obviousness as a sliding scale. At the ends of the spectrum are inventions that are clearly obvious and inventions that are clearly nonobvious. In between, it can depend quite a bit on advocacy abilities of the parties, and like in sports, good offense usually beats good defense. It is often the case that the accused infringer has deeper pockets than the patent owner, which means the accused infringer has access to more expensive lawyers and more expensive expert witnesses. While more expensive does not always mean better, capitalism often rewards skilled professionals in the form of higher fees. It would thus be hard to argue that this monetary imbalance does not skew in favor of patents being invalidated.

In patent examination, the Examiner is supposed to be a neutral party in determining patentability. The patent applicant usually hires an advocate – either a patent agent or patent attorney – to help convince the Examiner that claims are patentable. While examiners have varying views on obviousness (as illustrated by the widely varying rates at which examiners grant patents), there is no advocate for unpatentability when patents are being examined by the USPTO.

We Should Consider Changing PTAB Trials to More Align with Examination Instead of Trying to Align Examination with PTAB Trials

There is no reasonable reality in which a patent examiner can spend as much time analyzing a patent as an alleged infringer will if the patent is asserted in litigation. And there is no reasonable reality in which the neutral role of patent examination should be transformed to include an aggressive and skilled advocate armed with an expensive and persuasive expert witness. Such a skilled advocate can look to the thousands of PTAB decisions where patents were at least partially invalidated, craft convincing hindsight reasoning as to how the prior art would have rendered obvious a claimed invention based on these decisions, and prep a persuasive expert witness to support the convincing hindsight reasoning. In such a situation, the deck is stacked against the patentee, essentially eviscerating the presumption of validity that is supposed to accompany a patent granted by the USPTO and allowing the USPTO a second chance at establishing unpatentability after patentability was established by the Examiner.

If we want to strengthen U.S. patents and add certainty to patent rights, we should reconsider how PTAB trials are conducted. If I was redesigning the PTAB, I would start by removing the expert witnesses from PTAB trials. This would help to lessen the imbalance resulting from accused infringers having deeper pockets by causing the prior art to speak for itself when it is judged against a valuable patent. If a patent is truly of such poor quality that it is an undeserved burden on the public, an expert witness is unnecessary to prove invalidity.

I would also require more objective evidentiary support for proving obviousness by changing the standard for PTAB trials to the teaching, suggestion, motivation (TSM) test that was the obviousness standard before the Supreme Court’s decision in KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398 (2007). Sure, the TSM test was imperfect, but it placed more objective guiderails on the obviousness analysis than the overly flexible standard set forth in KSR. Specifically, the TSM test requires prior art evidence to support reasoning alleging that it would have been obvious to have combined prior art references to obtain a claimed invention. If the USPTO is taking away a patent right that has considerable value in the market, based on assertions by an aggressive and skilled advocate, the evidentiary burden should be higher than that required for a neutral patent examiner.

I think both of these changes would align PTAB trials with their original mission of invalidating “bad patents,” and would decrease the likelihood that a patent is challenged at the PTAB simply because it is valuable. They would help to restore the presumption of validity, which is lost in the present system that combines an aggressive and skilled advocate, an expensive and persuasive expert witness and an overly flexible obviousness standard.

Essentially, the present system provides a do-over for the USPTO – the PTAB panel second guesses the examiner and concludes the claim is obvious if they would have reached a different decision than the Examiner.  Such a do-over removes the presumption of validity from a valuable patent based on subjective argumentation from a highly incentivized party. It does not result in the post-grant review system embodied by the version of the AIA that was sold to the legislature and the public as an error correction mechanism for examination mistakes. The PTAB should fix objective examination mistakes made by the USPTO and should not simply decide subjectively after the fact that the examiner could have rejected the claims. If the post-grant review system was more objective, then the PTAB’s invalidation rates would make more sense as a barometer for how well the USPTO is examining patent applications.