Do You Really Want to Make PTAB Judges ‘Inferior Officers’? —Think Again!

“PTAB judges that received relatively higher bonus awards were also those who, on average, had a larger fraction of decisions that were adverse to patent owners—creating the appearance of pecuniary bias in favor of petitioners.” a recent Supreme Court decision in Arthrex v. Smith & Nephew, the Court held that the unreviewable authority wielded by Administrative Patent Judges (APJs) at the Patent Trial and Appeals Board (PTAB) of the U.S. Patent & Trademark Office (USPTO) exercises authority of a “principal officer” and is incompatible with their appointment by the Secretary of Commerce to an “inferior office.” Instead of declaring their appointment unconstitutional, the Court’s remedial ruling was aimed at making PTAB judges “inferior officers.” It did so by a ruling interpreting 35 U.S.C. § 6(c) as enabling the USPTO Director to “review decisions rendered by APJs,” subordinating them to the Director’s full supervision.

The USPTO’s position in the case was that APJs are already “inferior officers” because they are subject to “significant oversight, direction, and control” by the Director, who is a Senate-confirmed “principal officer.”  The notion of PTAB judges’ subordination to the Director’s “significant oversight, direction, and control” as aspects of their “inferior officer” position was previously bolstered by the Federal Circuit’s Arthrex decision below, observing that “[n]ot only does the Director exercise administrative supervisory authority over the APJs based on his issuance of procedures, he also has authority over the APJs’ pay.” The utility and efficacy of that financial lever—APJs’ base salary—and the Supreme Court’s requirement for Director’s review of APJs’ decisions, are touted as key solutions for making APJs “inferior officers.” However, these are not nearly as powerful, direct, and agile as another lever for “significant oversight, direction, and control.” It is a lever which neither the USPTO, the Federal Circuit, nor the Supreme Court ever acknowledged is actually at play in greater force—the PTAB bonus system, with discretionary awards of up to 20% of the PTAB judge’s base salary.

A Profound Conflict

Ironically, the very framework the USPTO and the courts heralded as saving the APJ appointment statute from Constitutional infirmity—the Director’s supervisory levers making APJs “inferior officers” (thereby dictating the outcome of PTAB cases)—is also the framework that likely undermines another Constitutional principle. That is the requirement that APJs’ adjudications be independent and impartial, in keeping with the due process clause of the Fifth Amendment of the U.S. Constitution. This is a substantive and profound conflict arising out of the artificial constraint to preserve Congress’ failure to protect the due process rights of patent owners.

Administrative Law Judges (ALJs) presiding over adjudications in administrative tribunals operated by most other Federal agencies may not be rated on job performance, nor receive performance appraisal reviews or related bonus awards. This ensures the appearance of their independence from their agency’s influence and impartiality in adjudication. In contrast, APJs received bonus awards of up to 20% of their base salary (reaching $33,910) based on a Performance Appraisal Plan (PAP) used by the USPTO to measure their job performance. The relative size of the bonus awards related to adjudication per se may raise due process concerns. I studied the features of the PTAB bonus plan and recently reported on my empirical analysis of the relationship between the bonus awards paid to APJs in fiscal year 2016 and the number and type of decisions they made that year.

I did this by obtaining the list of all America Invents Act (AIA) trial cases in which the PTAB made decisions that year on whether or not to institute a proceeding, and Final Written Decisions (FWD) on whether to uphold all patent claims or to cancel some claims. For each participating APJ, I tabulated in these four decision categories the total number of decisions in which the APJ participated in deciding to deny institution, grant institution, uphold all claims, or cancel claims. I also obtained from the bonus awards received for that fiscal year by each APJ participating in those decisions. I found that PTAB judges that received relatively higher bonus awards were also those who, on average, had a larger fraction of decisions that were adverse to patent owners—creating the appearance of pecuniary bias in favor of petitioners.


Pinpointing Apparent Pecuniary Bias

To quantify these trends, the empirical regression analysis described further in my study found that in FY 2016, APJs participating in AIA trials appeared to have earned an average bonus of $255 per decision when granting institution, but only an average of $208 per decision when denying institution. They also appeared to have earned an average bonus of $314 per Final Written Decision when canceling patent claims, but only an average of $2 per Final Written Decision when upholding all patent claims. The difference for each decision pair is the apparent APJ pecuniary bias. Aggregating across all such decisions, I found that this resulted in an annual average apparent APJ pecuniary bias of $1,307 in institution decisions and $4,453 in FWDs. This amounted to a total apparent APJ pecuniary bias against patent owners of $5,760 out of an average annual APJ bonus of $21,166.

The APJs’ PAP does contain a “Production” element that weighs 35% in overall rating, based on the total number of Decisional Units (DUs) they earn for all decisions they made in the period. The larger the DU credit, the higher the rating towards the bonus award. However, APJs received identical number of DU credits deciding to institute an AIA trial as to deny institution. Similarly, they earned an identical number of DU credits per FWD to cancel patent claims as to uphold all patent claims. Therefore, APJ rating under the “Production” element itself must be presumed neutral. What then could account for the empirically-observed bias in the bonus awards? The answer must lie in the structure and operation of the PAP’s subjective elements or other factors that are outside the objective “Production” element.

Other subjective critical elements in the APJ PAP, namely “Quality,” “Supporting the Mission of the Board,” and “Internal/External Stakeholder Interactions,” weigh in the aggregate 65% of the overall PAP rating. Rating APJs under these elements apparently opened the less transparent door for PTAB officials to indirectly and tacitly reward APJs for outcomes of their decisions (whether consciously or not). No causal relation need be proven here—mere correlation of any such non-production ratings with the number of decisions in any of the four categories is sufficient to create the appearance of pecuniary decisional bias.

For example, under the critical element weighing 20% called “Internal/External Stakeholder Interactions,” the APJ PAP elaborates: “Outstanding performance in this element includes, … appropriately, promptly, and courteously addressing any questions, comments, or requests from internal and external stakeholders.” But who makes these rating evaluations, and who supplies the factual record upon which such rating determinations are made? Are “external stakeholders” interviewed for that purpose? My study discusses plausible scenarios wherein APJ’s interactions with “external stakeholders” inherently and without intention renders a subjective measure of such interaction prone to pecuniary bias in favor of patent challengers, who are the heaviest users and repeat “customers” of the PTAB.

With what the USPTO called “gain-sharing bonuses” in an APJ recruitment brochure, it would hardly be a surprise if APJs that participated disproportionately in higher numbers of decisions to grant institution received higher ratings for “Supporting the Mission of the Board” element of their PAP. This is particularly plausible when such decisions are perceived to help “gain” workflow for the PTAB and where its top officials that rated the APJ were themselves incentivized to “ensure the most valuable PTAB employees are rewarded.” I can understand what a “fair,” “efficient,” or “impartial” judge means; but what is a “valuable” judge? “Valuable” to whom?

In yet another example, APJs were informed in the PAP Support Document that a secret extra-panel review committee of the PTAB called the AIA Review Committee (ARC) is of relevance to their rating under the “Quality” element of the PAP. According to this guidance, APJs must circulate their draft decisions to the ARC at least 6 business days in advance of the deadline to enable the ARC to review and comment. The guidance explains that “ARC comments are not binding, but instead suggestions that a panel may consider in preparing decisions.” Where the mention of ARC comments is made in the “Quality” element of the PAP, a reasonable APJ would draw the inference that earning higher “Quality” rating inevitably follows from greater conformity with the ARC recommendations. Coupling this with the fact that 85% of FWD’s were adverse to patent owners, and that by default, PTAB judges earn no DUs for any dissents they author, leads to the likelihood that higher “Quality” ratings of judges were correlated with a greater number of their decisions that were adverse to patent owners.

As Federal Circuit Judge Pauline Newman recognized in addressing this issue recently, the introduction of these potential due process infirmities in AIA trials is particularly difficult to reconcile with a tradition of due process of law that historically governed patent validity challenges.  Here, a quasi-judicial structure is created essentially to replace judicial proceedings, and in fact to provide an estoppel in judicial proceedings, to nonetheless conduct adjudications subject to questionable judge compensation procedures which would not be available to the Judicial Branch of government.

Further Analysis

I address the implications of these and other findings (including “stacked” panels) on the due process rights of patent owners in my recent report and analysis.


Image Source: Deposit Photos
Image ID:429683268


Warning & Disclaimer: The pages, articles and comments on do not constitute legal advice, nor do they create any attorney-client relationship. The articles published express the personal opinion and views of the author as of the time of publication and should not be attributed to the author’s employer, clients or the sponsors of Read more.

Join the Discussion

32 comments so far.

  • [Avatar for Ron Katznelson]
    Ron Katznelson
    July 9, 2021 01:52 pm

    IamI, you are welcome and thanks for raising that issue. I should add that had my data consistently underestimated the bonus levels, your observation may not have been correct as to the actual effect of larger bonuses, as suggested here by two commenters, but it would have been correct for my regression analysis and the derived coefficients. This is because errors in either direction in the bonus inputs to the regression analysis undermines the reliability of the derived coefficients. So thanks to your input, I will include my explanation above in the next revision of my paper.

  • [Avatar for IamI]
    July 9, 2021 09:04 am

    Ron, thanks for that detailed response. I appreciate you walking through that for me.

  • [Avatar for Anon]
    July 9, 2021 08:39 am

    Pro Say,

    You are aware that your “Ugh” translates to “Wah” because you are refusing to acknowledge the content that I have provided, eh?

  • [Avatar for Pro Say]
    Pro Say
    July 8, 2021 08:15 pm

    Ugh, Anon, just ugh.

  • [Avatar for Anon]
    July 8, 2021 04:00 pm

    Pro Say,

    The generalization FITS you in this instance (based on the substance provided).

    You seek ‘insult’ when you should be noting the content. Quite in fact, you live up to (down to) your chosen moniker in this instance, and thus the generalization is simply apt. You may not like that it fits, but fits, it does.


    The opposite of “blowing up” as the figures that are not available can only make Ron’s argument’s stronger (not weaker). The data provides a floor of ill repute, not a ceiling.

  • [Avatar for Ron Katznelson]
    Ron Katznelson
    July 8, 2021 02:26 pm

    IamI @24: “the bigger issue here is Ron basing his entire argument based off of bonus figures reported by, which is inaccurate with bonus figures because the site itself states that performance bonuses aren’t listed.”

    Julie Burke @26: “what criteria (since its presumably not performance-rating based) is being used by the USPTO to award some but not other APJs up to $33K additional annual compensation in 2016?”

    The answer is that IamI @24 misses the reality that those statements are made generically with respect to a large number of agencies with varied types of employees and methods of compensation. The portion of the statement applicable to APJs at the PTAB is “[the OPM] reports the adjusted base salary of all employees covered by this dataset, as well as any award amounts paid to each employee.” As my paper explains, the appraisal period in the APJ PAP is the full fiscal year (unlike situations of shorter review periods in the examiner corps that Julie mentioned @13), and therefore the additional “award amount” the OPM data provides for APJs is their total annual bonus cash award. This is corroborated by a computational “checksum” on individual APJs bonuses by the fact that under this “award” category, the limit in 5 U.S.C. § 4505a(a)(2) for a “cash award” not exceeding 20% is remarkably met by the numerical fraction of 19.9% precisely for individual APJs (see Figure 4). Given that this precise numerical limit is met, for any misreporting of an individual APJ bonus award, there must also be a misreporting of that APJ’s base salary in exact proportion so as to maintain the precise 19.9% ratio. That is simply implausible. In conclusion, while OPM data on bonuses may be partial or inaccurate with respect to other government employees, it is accurate for the bonuses of APJs.

  • [Avatar for Julie Burke]
    Julie Burke
    July 8, 2021 11:29 am

    @10, if the Federal Pay data excludes performance-rating based awards, then doesn’t one wonder what criteria (since its presumably not performance-rating based) is being used by the USPTO to award some but not other APJs up to $33K additional annual compensation in 2016?

  • [Avatar for Pro Say]
    Pro Say
    July 8, 2021 10:24 am

    “pro se advocates simply are not savvy to the nuances of the trade”

    Anon, generalizations steal individuality.

    Thank you for no longer trying to steal mine.

  • [Avatar for IamI]
    July 8, 2021 09:44 am

    Julie @13: I appreciate your comments. While what you said may be true (we have no real way of knowing what their exact bonus structure is), the bigger issue here is Ron basing his entire argument based off of bonus figures reported by, which is inaccurate with bonus figures because the site itself states that performance bonuses aren’t listed. It kind of blows up a lot of his argument before it even gets going.

  • [Avatar for Anon]
    July 8, 2021 09:08 am

    Pro Say,

    It is less a caste system and more a reflection on the well known (and universally accepted) notion that pro se advocates simply are not savvy to the nuances of the trade.

    Thus, the comment is very much about content.

    I invite you to go back and read the content that I provided to you.

  • [Avatar for Pro Say]
    Pro Say
    July 7, 2021 11:12 pm


    Your view is, well, typically Anon.

    Long past time to discontinue your use of “pro se” as a term of denigration.

    As if pro ses are due any less respect than attorneys are.

    Everyone knows something (and often many somethings) worth knowing. I do. You do. Large majority of folks who comment here do as well.

    Accordingly, please pay attention to what is being said . . . and not who is saying it.

    Content over caste, my friend.

    Content. Over. Caste.

  • [Avatar for Anon]
    July 7, 2021 06:06 pm

    Pro Say,

    Your view is, well, typically pro se.

    Please consider that for legal matters (actually, for most all contested matters), one wants to know as much as possible about ‘the other side.’

    Most times (such as here), the duplicity will be evident. Further, a number of us ‘regulars’ are likely to note the foibles of the other side.

    Lastly, think of this as a way to avoid a “bubble” situation in which you are only exposed to those viewpoints that you already agree with. You don’t want that. Hint: that is one of the supreme problems with Academia.

  • [Avatar for Pro Say]
    Pro Say
    July 7, 2021 02:38 pm

    Josh Landeau: Bought. And. Paid. For.

    Surprising . . . and disappointing that Eli would even give this joker the time of day; let alone provide a platform to spew his b.s.

    One doesn’t give the Devil equal time, “in the name of balance,” or some such misplaced rationale.

    You’re better than this, Eli.

  • [Avatar for Anon]
    July 7, 2021 01:29 pm

    The comments were not turned on in that Landeau piece – and I suspect I know full well why.

    I had the distinct impression while listening to the interview that the interview had been a virtual one, as if it were in person, I suspect that Mr. Landeau may have been punched in the face.

    I am all for such interviews and am thankful that IPWatchdog ‘reaches to the other side.’

    Landeau is clearly speaking in half truths. It may well be true (the half true part) that attorneys ARE upset because they do not like the outcomes of the malfeasance of the Supreme Court in regards to patent eligibility jurisprudence, but one cannot stop with that statement as it misleads. One must continue to delve into the WHY attorneys are upset – and THAT part DOES include the plain fact that “unclear” does not even begin to describe the Gordian Knot created by the Supreme Court.

  • [Avatar for Erfinder]
    July 7, 2021 12:42 pm

    PTAB judges invalidate patents
    Bill Gates gives vaccine advice
    “The foxes are guarding the hen house!”

  • [Avatar for Night Writer]
    Night Writer
    July 7, 2021 10:34 am

    Look, all of this is a joke. Just listen to Josh Landeau on the other post. He is clearly being paid big money to weaken patents and will say and do anything for his money from Silicon Valley. That is who the enemy is. There are no ethics and there billions and billions of lobbying money to stop any patent reform. There is an endless stream of negative ridiculous nonsense from huge monopolies to weaken patents.

    Just think about the AIA. Anyone that knew anything about patents knew that IPRs was the death knell of patents and yet it passed and there wasn’t even that much push back from the patent community. I remember the senior partner at my law firm saying that we should all consider switching to a different area of the law if the AIA passes as IPRs will greatly reduce patent value (down 80 percent since the AIA passed).

    Josh Landeau >>>I will actually push back on the idea that section 101 is particularly unclear. The problem is not that it’s unclear. The problem is that in many cases or some cases, the judges of the Federal Circuit or attorneys don’t like what the outcome is.

  • [Avatar for Dose O. Reality]
    Dose O. Reality
    July 6, 2021 01:57 pm

    What if I told you the PTAB did away with productivity bonuses years ago. And a study of current data shows no appreciable difference in these numbers.


  • [Avatar for Jacek]
    July 6, 2021 01:54 pm

    I wish somebody would write a similar article about the “Ethic standards” PTAB judges must adhere to.
    Why they are different than any other lawyers in the US?

  • [Avatar for Anon]
    July 6, 2021 12:53 pm

    Miss Burke @ 13,

    That is not a cheerful thought.

    I know of at least one regular patent blog commentator — known by his actions as the IPR Cheerleader — that continues to turn a blond eye to any indication of bias when it comes to the judges of the PTAB. To think then that matters may be even worse than the data indicates is a bit chilling.

  • [Avatar for Julie Burke]
    Julie Burke
    July 6, 2021 11:11 am

    @10, I agree that Federal Pay bonus data is incomplete, at least for patent corps employees.

    My additional compensation was routinely under reported on the Federal Pay website. Extra compensation labelled as “awards” or “gainsharing” appeared to be left out. Also some bonuses are given out once a year, following annual performance appraisals. Others, like the Docket Management awards, which, at least during my tenure were given out quarterly, are likely missing from the Federal Pay data.

    If the PTAB bonus/award data are also incompletely reported to Federal Pay, then the calculations in this article underestimate the APJ payouts. The appearance of bias could be much worse.

  • [Avatar for erfinder]
    July 6, 2021 10:40 am

    IMHO, the PTAB judges, along with the entire corrupt PTAB, MUST be eliminated STAT

  • [Avatar for AAA JJ]
    AAA JJ
    July 6, 2021 09:01 am

    “There’s a culture of opaqueness and deceit, that’s actually promoted and compensated for such behaviors.”

    That’s not just at the Office of Petitions.

    And it’s a feature. Not a bug.

  • [Avatar for IamI]
    July 6, 2021 08:47 am

    From “Salary / Compensation – The Office of Personnel Management reports the adjusted base salary of all employees covered by this dataset, as well as any award amounts paid to each employee. Performance-rating based awards are not included, and some agency-specific incentive pay may also not be listed.”

    With that in mind, how can you *know* the bonuses listed on Federal Pay’s website are a) correct and b) for what your article claims them to be? I am not an APJ, but my bonus amount on Federal Pay’s website is always incorrect.

  • [Avatar for BP]
    July 5, 2021 09:36 pm

    Excellent work, and the bias is condoned, being a given throughout the agency, especially at the Office of Petitions, which was headed by the acting Director. There’s a culture of opaqueness and deceit, that’s actually promoted and compensated for such behaviors.

  • [Avatar for Pro Say]
    Pro Say
    July 5, 2021 09:28 pm

    p.p.s. Indeed, Ron’s startling, jaw-dropping findings form a new unconstitutionality basis for the elimination of the PTAB.

    Something which needs to be brought to the attention of the CAFC via appeal; and, should it prove necessary, to SCOTUS.

  • [Avatar for Pro Say]
    Pro Say
    July 5, 2021 09:07 pm

    p.s. If and until such time that the Patent Office does the good, right, and moral thing by eliminating these bias-heavy, justice-denying bonuses; and with Ron’s research in hand; it’s time for patent owners who’ve had their patent/s wiped out (or prevented from being issued) to file complaints with the respective state bars against any attorney’s who’ve ruled against them — whether in an IPR / PGR or in ex parte appeals.

    Enough is enough.

    For CA attorneys, start here:

    For DC attorneys, start here:

  • [Avatar for Ron Katznelson]
    Ron Katznelson
    July 5, 2021 08:22 pm

    For some reason, the SSRN site blocks downloads of the PDF of my article. An alternative download link for the full paper is at

    Sorry for the inconvenience.


  • [Avatar for BP]
    July 5, 2021 06:33 pm

    Thank you Ron. Gorsuch’s dissent points to many problems. Quiet title for a patent? That simply does not exist because the Director, a political appointee, can take as pleases. Incentivizing the taking process, that drives a viscous loop. Kill more patents and you’ll get more customers. But doesn’t somebody have to balance all two or three of those “patent troll” friendly Article III district court judges (that aren’t incentivized financially for trying to maintain a level playing field between defendants with trillions and plaintiffs with pennies)? Billion dollar infringement judgments for digital tech, starting mid-late 1990s, simply could not be tolerated. The solution, the political route, lobby, lobby, lobby. Paid for legislation is seldom coherent, poison pills and landmines are baked-in, in this instance, to weaken/confound the agency viewed to be at the heart of the “patent troll” problem and waste millions in fees paid by independent innovators.

  • [Avatar for George]
    July 5, 2021 05:57 pm

    They are PART of the PTO, so how can they be trusted?! They aren’t independent judges! This was a terrible idea! They are also not experts in the technology areas they are judging and so can’t understand them! What do they know about what is really new and what isn’t?! Just because of ‘terminology arguments’? Screw ‘terminology’! Focus on ‘content’ and ‘teachings’. If it’s new & ‘significant’ (therefore not obvious) – it deserves IP protection!

  • [Avatar for Pro Say]
    Pro Say
    July 5, 2021 05:16 pm

    Thanks Ron.

    Yet another powerful, fact-based indictment of the structure and operation of the PTAB.

    Big bonuses for killing innovation.

    Let’s call these illicit bonuses what they really are, shall we:

    Bribes. Graft. Payola.

    The attorney’s taking these phony “judge” Death Squad positions should be ashamed of themselves.


    Indeed, knowing what they now know given Ron’s and others research, how can the respective state bars permit their members to engage in this activity?

    How? Can? They?

    How much longer . . . before Congress ends this experiment-gone-bad?

    How? Much? Longer?

  • [Avatar for ipguy]
    July 5, 2021 03:59 pm

    “I did this by obtaining the list of all America Invents Act (AIA) trial cases in which the PTAB made decisions that year on whether or not to institute a proceeding, and Final Written Decisions (FWD) on whether to uphold all patent claims or to cancel some claims.”

    I wish a similar analysis would be done for ex parte PTAB appeals in terms of the appearance of pecuniary bias in favor of affirming rejections. If it already has been done, I’d appreciate a link.

  • [Avatar for Julie Burke]
    Julie Burke
    July 5, 2021 03:51 pm

    Thank you, Ron, for this important deep dive into the APJ’s compensation plan.

    WRT rewarding APJs with gain sharing bonuses “for supporting the mission of the Board” to ensure the “most valuable PTAB employees are rewarded,” I wonder how the APJ PAP defines the “mission of the Board.”

    And I look forward to your next report to be published tomorrow.