In a post-truth world, it is more tempting than ever to evaluate data based on gut instinct, intuition, and anecdotal evidence. It is thus refreshing when results of a robust statistical analysis are published, even if the response to the ultimate outcome is, "Yeah, we knew that."
Case in point, patent eligibility. Most patent practitioners are likely to agree that the Supreme Court's Alice Corp. v. CLS Bank decision made it more likely to receive 35 U.S.C. § 101 rejections during prosecution before the U.S. Patent and Trademark Office (USPTO). Many of these practitioners would also likely concede that Alice made it less predictable as to whether a given claimed invention would run into eligibility issues. These same practitioners would probably concur that the Federal Circuit's Berkheimer v. HP case as well as the USPTO's 2019 patent eligibility guidance (PEG) made it less likely to receive such a rejection and resulted in the eligibility inquiry being more predictable.
But most of this is anecdotal. It is easy to find data supporting the premise that § 101 rejection rates went up after Alice and down after each of Berkheimer and the PEG. But does this data withstand a rigorous examination? Thanks to the USPTO's economists, the answer is yes.
Last week the U.S. Patent and Trademark Office published a study of how the patent eligibility landscape has changed within its art units over the last six years, with emphasis on the inflection points of Alice, Berkheimer, and the PEG.
The USPTO identified Alice-affected technologies by considering patents litigated in the Supreme Court and Federal Circuit for which eligibility was at issue. The USPTO's U.S. Patent Classifications for the associated applications were used to identify the Alice-affected technologies, and all other U.S. Patent Classifications were used as the control group.
Two main metrics were considered -- percentage of first-action § 101 rejections and § 101 examination uncertainty. The former is a simple quotient of first-action § 101 rejections over all first office actions in the Alice-affected technology areas within a defined time period. The latter is a bit more complicated. First, the rate of first-action § 101 rejections for each examiner is determined over the time period. Then, the variance of these per-examiner rates was calculated. The § 101 examination uncertainty metric was the average of these variances. Similar metrics were found for the control group.
The Alice Decision
In Alice, the Supreme Court set forth a two-part test to determine whether claims are directed to patent-eligible subject matter under § 101. One must first decide whether the claim at hand involves a judicially-excluded law of nature, a natural phenomenon, or an abstract idea. If so, then one must further decide whether any element or combination of elements in the claim is sufficient to ensure that the claim amounts to significantly more than the judicial exclusion. But, elements or combinations of elements that are well-understood, routine, and conventional will not lift the claim over the § 101 hurdle.
From the data, it is clear that Alice had a statistically significant impact on § 101 rejections. In the Alice-affected technology areas, the likelihood of receiving a first-action § 101 rejection increased by 31% in the 18 months following the opinion's publication. Also, Alice increased the § 101 examination uncertainty metric by 26% in these technology areas during the same period.
The Berkheimer Decision
While the Alice two-part inquiry is generally carried out as a matter of law, factual issues can come into play when determining whether something is well-understood, routine, and conventional. Berkheimer stands for the notion that when such factual issues are present, a patent application cannot be found ineligible without further review.
Berkheimer resulted in the first-action § 101 rejection rate in the Alice-affected technology areas dropping from about 35% during March of 2018 to about 28% in early January of 2019. Further, the uncertainty metric also fell from about 0.12 (its highest point since Alice) to about 0.9 in this time frame. As a point of reference, in the months before Alice, the uncertainty metric was slightly below 0.9.
In a nutshell, the PEG breaks the first part of the Alice test into two sub-steps. In the former, one determines whether the claim recites a judicial exception, such as an abstract idea. In the latter, one determines further "whether the recited judicial exception is integrated into a practical application of that exception." To focus the analysis, the PEG indicates that all abstract ideas should fall into one of three categories: mathematical concepts, certain methods of organizing human activity, and mental processes. Think of these as the "Three M's" -- math, money, and mental steps.
Similar to Berkheimer, the PEG reduced both the first-action § 101 rejection rate and the uncertainty metric. The first-action § 101 rejection rate fell steeply in the Alice-affected technology areas, from about 28% to about 18%. The uncertainty metric dropped from about 0.9 to about 0.5, the latter its lowest point in the decade since Bilski v. Kappos. Thus, the overall impact of the PEG was greater than that of Berkheimer.
Given the lack of surprising results in this study, it is tempting to just yawn and say, "So what?" To some extent, doing so would not be unwarranted -- the PEG caused a huge reduction in § 101 rejections and uncertainty -- but the Federal Circuit has explicitly stated that it is not bound by the USPTO's guidance. The USPTO can pat itself on the back for its efforts in providing more consistent and predictable examination with the PEG, but that ultimately may not matter when the federal courts will apply a less patent-friendly test when determining whether to invalidate patents that the USPTO granted.
But perhaps the most telling aspect of the study is not the data, but instead in the casual way that the authors describe the uncertainty surrounding § 101 and its potential impact on the economy. Particularly, they take it as given that Alice "created uncertainty in the business and legal communities" and that "[a]mbiguity in the language . . . and in the scope of technologies involving abstract ideas made it difficult to predict how and where the standard would be applied."
These statements reflect those of federal judges, former federal judges, certain members of Congress, the inventor community, patent professionals, and (of course) USPTO Director Iancu. There is widespread consensus that Alice's unclear language is problematic and has made it riskier for small and medium-sized businesses to invest to protect their inventions. Under Alice, patent applications can be rejected (and granted patents can be invalidated) based on subjective and conclusory reasoning. But perhaps Alice is yet another cynical by-product of the post-truth narrative, where opinions and gut feelings trump objective facts.
 It is not clear from the study whether this average was weighted by the number of first actions per examiner. Arguably, it should have been so that more active examiners had more influence over the metric than less active examiners.