An Empirical Study of Low Allowance Rate Examiners

By Michael Borella --

Any patent attorney who has been in the business for more than a few years understands from experience that some USPTO examiners are tougher than others.  This should not be surprising, as each examiner is an individual who is applying their own experience and knowledge during the examination process, which inherently includes some degree of subjectivity.

Nonetheless, we all have experienced examiners from time to time that appear to be exceptionally difficult.  These examiners are more likely than most to reject claims on multiple grounds and be less responsive to applicants' rebuttals.  Some are unwilling to engage in productive conversations during interviews.  Most have low allowance rates.

With analytics tools now readily available to applicants and their attorneys, we can obtain concrete statistics about these examiners, and then analyze their prior prosecution histories in order to better understand their individual proclivities and nuances.  It has become generally accepted in the field that the examiner that is assigned to an application can be as determinative on its outcome as the application itself.  If an application is assigned an unusually tough examiner, there often is relatively little the applicant can do to obtain an allowance short of an appeal or filing a continuation in hope of obtaining a more reasonable examiner.

In an effort to better understand these tough examiners, we undertook a study to identify and analyze what we call the low allowance rate examiner (LARE).  These are examiners who not only have a low allowance rate, but also have had an allowance rate significantly below that of their art unit for a non-trivial period of time.  In other words, we wanted to identify a handful of examiners who are true outliers in terms of their examination practice.  The goal is to determine just how difficult these examiners are and in which art units these examiners are found.

For our purposes, a LARE is defined as an examiner who meets all of these criteria:

• Has an overall allowance rate below 10%;
• Has an overall allowance rate more than 25% below the average of their current art unit; and
• Has examined more than 100 applications.

These criteria help us identify examiners are not only outliers in general, but also outliers among their own art unit peers.  Further, rookie examiners who may not have examined enough applications to establish a pattern of behavior are omitted.  Additionally, we only considered applications that were filed on January 1, 2010 or later in this study to exclude considering too many examiners who have since retired.

Notably, these criteria exclude many examiners who have low allowance rates.  In searching the Juristat USPTO database, we identified a large number of examiners with allowance rates below 25% (the current overall average allowance rate at the USPTO is 73%).  Further, many examiners with allowance rates below 10% are in art units with low allowance rates (more on that later), so these examiners are not exceptions when compared to their peers.  We also noticed many newer examiners with allowance rates below 10% but without the requisite number of applications examined.

We used Juristat's database and its analytical tools to identify LAREs.  After manually eliminating some examiners whose statistics were extremely skewed based on what appeared to be administrative nuances,[1] we identified 24 LAREs.[2]

We have to be careful here -- just because an examiner is identified as a LARE does not mean that they have engaged in any malfeasance.  An examiner may have a low allowance rate because they specialize in examining subject matter that has a low allowance rate.  Further, the Juristat data is a reflection of the USPTO public database and is necessarily not up to date (e.g., it will only reflect Office actions mailed in 2023 or 2024 that are not subject to the 18-month publication delay).  Thus, some of these LAREs might not qualify if this unpublished data were considered.  Nonetheless, we are convinced that these LAREs are true outliers because we intentionally made the selection criteria extremely restrictive.

First, some demographics.  The average number of examined applications for LAREs is 211.5 and two-thirds have 10 or more years of experience as examiners at the USPTO.  All LAREs have at least 4 years of experience.  This clearly establishes that most LAREs are highly-experienced examiners and very few are junior examiners.

The average allowance rate for LAREs is 6.4%.  In other words, if your application is assigned to a LARE, there is a 93.6% likelihood that it will not be allowed -- a devastating number.  Further, the average art unit allowance rates for these LAREs is 52.73%.[3]  So, even though these LAREs tend to be in art units with average allowance rates that are lower than the overall USPTO average, LAREs are still quite anomalous.

As noted, all LAREs have an allowance rate that is at least 25% below that of their art unit average.  The difference between LARE allowance rates and their art unit averages ranged from 27% to 77%.  At the high end of this spectrum, we found an examiner with an 8% allowance rate in an art unit with an average allowance rate of over 85%.

One LARE had an allowance rate below 1%, which was 50% lower than their art unit average.  Another LARE had an allowance rate between 1% and 2%, which was 38% lower than their art unit average.  Yet another LARE had an allowance rate between 3% and 4%, which was over 57% below their art unit average.  These were the most extreme examples in terms of low allowance rate.

It should surprise nobody that 21 of 24 LAREs were in Tech Center 3600.  This area of the USPTO is notorious for low allowance rates as it is where most business method and many software applications are routed.  Since these types of applications are unusually susceptible to § 101 rejections, we see a rather high rate of such rejections from Tech Center 3600 LAREs.

To add some color the § 101 issue, we considered the last 20 substantive Office actions from each LARE and calculated the rate of § 101 rejections for each.  Unsurprisingly, it was 70% for Tech Center 3600 LAREs and much lower (0%-5%) for the other 3 LAREs.  Two of the LAREs gave § 101 rejections 100% of the time and five more gave § 101 rejections 90%-95% of the time.

One of the reasons for such high § 101 rates may go beyond the nature of the subject matter being examined.  Anecdotally, in conversations with Tech Center 3600 examiners, some have told me that they feel pressured by their supervisors to give § 101 rejections for every single application unless they have a really good reason to do otherwise.  Further, experienced examiners in Tech Center 3600 may understand that the PTAB is highly unlikely to reverse a § 101 rejection, especially for business method and software inventions.[4]  Thus, an appeal to the PTAB is not a viable option for applicants and the examiner has a very low likelihood of being reversed on appeal.

Moreover, the USPTO has utterly failed to police the slop in some examiner' § 101 rejections, including conclusory reasoning, ignoring claim elements, not understanding the concept of the prima facie burden on examiners, employing vigorous hand waving over dependent claims, ignoring technical improvements in the claims and the specification, and erroneously thinking that concrete tangible inventions can be "abstract."[5]  These factors may incentivize examiners to provide § 101 rejections regardless of whether such rejections are truly warranted.

Given that an application examined by a LARE is almost certainly not going to be allowed, what can applicants do?  While one can file a continuation or continuation in part with a different claiming strategy, this approach typically results in the continuation being assigned to the same examiner.  A better strategy is to do everything you can to front-load the application so that it does not get routed to art units known for having LAREs.  This involves avoiding business language in the specification, repeatedly describing the invention's technical benefits, and using art unit steering tools to avoid such art units.

Notably, the LAREs identified in this study were all in the 1620 (organic chemistry), 2860 (printing, measurement and testing), 3620 (business methods), 3640 (aeronautics, agriculture, fishing, trapping, plant and animal husbandry, weaponry, nuclear systems), 3670 (construction), 3680 (business methods), 3690 (business methods), and 3710 (amusement and education) art units.  These art units should be avoided at all costs through careful up-front drafting.

But the elephant in the room is why the USPTO allows LAREs to continue as examiners for years or decades.  The USPTO appears to have very little oversight of examiners, especially senior examiners.  If the USPTO is unable or unwilling to provide this oversight and to at least explain why so many LAREs and other examiners with low allowance rates exist, an independent oversight board should be established to conduct such a study and provide recommendations to the Department of Commerce.

Finally, if the USPTO takes issue with any of this data or our methodology, it should conduct its own study of LAREs and make the results available to the public.  As noted above, the USPTO has access to examination data that is non-public and may be able to shed more light on this issue.

[1] For example, we omitted an examiner with an allowance rate of 0.1% who appeared to have been associated with hundreds of applications that they did not examine.

[2] We are going to be careful not to provide information that clearly identifies individual examiners.  Thus, while we have precise numbers for all of the metrics discussed herein for each LARE, we will apply some rounding up or down in order to make such identification harder. In all cases, however, LARE behavior is so extreme that any fuzziness that we add to their statistics is not significant and does not impact any of our conclusions.

[3] This number is an average of an average and therefore is not a robust statistic.  Nonetheless, we present it for informational purposes and because producing a more accurate measure would require an excessive amount of manual effort.

[4] In 2022, the PTAB affirmed examiner § 101 rejections from Tech Center 3600 at a blistering rate of 97%.  See https://www.patentdocs.org/2023/01/ptab-remains-hostile-to-section-101-appeals.html.

[5] Much blame should be assigned to the Supreme Court's legislating from the bench, the Federal Circuit's self-contradictory lines of § 101 case law, and USPTO administration for not providing examiners with clear § 101 examination guidance.

Comments

[Mike Borella]

To add some perspective, a 6.4% allowance rate is just a little higher than the chance of an applicant getting into an Ivy League university.

[Max Hensley]

I retired 20 years ago so I can safely say publicly that fish rot starting from the head first. Brought back many happy memories of my now far departed prosecution days, when I used to call these the immaculate rejections because they required no substance at all. For the examiners it was just “101 and go home”

[Ernest "Buddy" Cusick]

I would put forth to the readers, what is preferred.

1. An easy examination and allowance. Minimal PA cited, no per se file wrapper, no real distinguishing reasons why allowed
OR
2. A "tough" examination, good search and PA of record, comprehensive distinguishing features noted in the record, and the like from a "LARE"?

As a former Primary and current corp Counsel, I believe that #1 opens you up to a lengthy PGR/IPR if that patent has any teeth, while #2 clearly identifies what is patentable.

Concerning the author's inference that LARE's are uncooperative, as a former Primary and Petitions Examiner at the PTO, an initial call from the atty to the examiner without an attitude goes a long way. Too many attys fight during the interview request, do not provide reasons why the rejections are in error, and are not willing to understand the Exm position.

I believe that if attys spent time examining, they understand the examiner's role, pressure, and job. Otherwise, too many attys have NO idea what it is like to examine, under time constraints that define their "quality", where one just cannot bill the client another 2.4 hours.

And when was it that you as an atty sent kudos to the Examiner and their SPE for a job well done?

[Max Drei]

You write:

"It has become generally accepted in the field that the examiner that is assigned to an application can be as determinative on its outcome as the application itself."

and I (looking in from Europe) am appalled. My first thoughts are that i) Examiners at the EPO (coming to the EPO from 40 or so different EPC Member States) have at least as wide a spectrum of personal opinion about what is patentable and what not as Exrs at the USPTO but that ii) the established law of patentability at the EPO is clear enough to render those disparate opinions irrelevant to the examination outcome. Basically, at the EPO, it makes no difference to the substantive outcome, which Examiner is allocated to the case.

[Extraneous Attorney]

@ Max Drei,

I would argue that some (if not most) of this falls down to the EPO working in three-examiner Examining Divisions. As far as I know, you do not get this setup at the USPTO until you use pre-appeal brief review.

[Max Drei]

Hi Extra. A 3-member Examining Division was necessary in the early days of the EPO, because the Examiners were arriving at the EPO froma multitude of different national Patent Offices all with their own different ideas about patentability, and quality depended on consolidation on a single line to be held by all Examiners. Now, 50 years later, no EPO Examiner has experience of any other ideas of patentability other than those set out in the EPO's own MPEP. EPO management pushes remorselessly for more "productivity". No longer does your European patent application receive scrutiny from more than one Examiner. But you may well be right, that the clarity of the EPO's law on patentability comes from that history.

Another reason for present day quality is that there is no supervisory court of law that can instruct the EPO what to allow and what to refuse. You might agree with me that there is nothing more effective than such a supervisory court, to render the Patent Office's job ever more difficult.

[Karl P Dresdner Jr PHD, Patent Agent No. 63,319]

It’s easier for the PTO management to hand over responsibility to examiners and have easier jobs themselves. There is also a failure of enforcement of inequitable conduct. PTO staff who do the filing work now seem untrained to see or report inequitable conduct so it’s under the radar. I have had 2 examiners with huge egos who played games until RCEs were filed. Blame should also be directed to the USPTO fees not being increased enough as that underfunds the PTO so I believe they have insufficient staff workers. Also the AIA modifications to patenting are a disaster that makes patenting overly complicated and dysfunctional. The PTO websites are also very over complicated now and user unfriendly.

It’s no longer fun to do patenting.