‘A Study in Scarlet’—Powers of Attorney and USPTO Rulemaking, Part I: A Hidden Guidance Document

“The USPTO’s ‘Apparent Authority’ guidance is a classic case of a rule that has no reason to exist at all. It creates $30 to $40 million per year excess burden on the public because the USPTO skipped all requirements of administrative law.”


This two-part article explains the United States Patent and Trademark Office’s (USPTO’s) practices with respect to powers of attorney. The pattern of neglect of administrative law identified here with regard to powers of attorney alone imposes a $30 to $40 million per year excess burden on the public. For the USPTO’s rules as whole, the costs are about $2 billion per year. Over the last 18 months, about 100 patent attorneys signed on to letters to ask the USPTO to do the simple right thing: conform its practices to the rule of law. (Editor’s note: the linked letter was submitted in an ongoing Paperwork Reduction Act renewal that the USPTO is still in the process of finalizing).

By Freedom of Information Act (FOIA) requests, I obtained a trove of hidden rules that bear the legend “For Internal Use ONLY.” These rules are totally irrational. Because the USPTO doesn’t follow the law, powers of attorney get bounced for irrational and undisclosed reasons, which creates immense and unnecessary costs. Part I of this article will explain these secret rules—they’re still irrational, but once you know about them, you can reduce costs and frustration.

This is one more example that demonstrates a pattern of disdain for administrative law among USPTO senior career staff. In an article about four years ago, I reported results of a survey that suggest that the USPTO’s neglect of rulemaking law imposes costs on the public of about $2 billion per year. The USPTO evades rulemaking law often by just skipping required steps, often by making statements to ex parte regulatory review tribunals that no knowledgeable person would accept as truthful. This pattern has been demonstrated in every recent rulemaking, including the fee-setting rule, the DOCX rule, the trademark “domicile address” rule, the IPR/PGR “ordinary meaning” rule, the Director review rule, and PTAB precedential opinions (my amicus brief is here, and the Federal Circuit’s adoption of the position I advocated, and rejection of the USPTO’s, is here). Even the Office of Enrollment and Discipline—the USPTO’s chief ethics office—in promulgating its CLE rule evaded multiple requirements of law, and made statements in an ex parte proceeding that are objectively false, and must have been known to be false (see AIPLA’s letter and in mineEditor’s note: OED did not finalize the proposed CLE guidelines and the CLE program has been postponed, thus the USPTO has not responded to any comments).

The problems with the USPTO’s rulemaking operations have been known for years—public comment letters laid them out clearly when the USPTO sought comment on its regulatory practices in 2011.

Solutions are well within reach: Judge Plager of the Federal Circuit recommended that every patent lawyer would do well to read one of my articles on rulemaking. I am told by friends who practice before other agencies, a regulatory oversight officer retired from the Executive Office of the President, and a friend who had senior executive service roles in both the USPTO and another agency, that every other agency in the Federal Government takes this stuff seriously. To my knowledge, the USPTO is the sole and extreme outlier. This article suggests reforms for powers of attorney and offers the first few items from a long list of long-overdue reforms that should apply throughout the USPTO.

Background: Powers of Attorney and LLCs

Powers of attorney are governed by 37 C.F.R. § 1.32. Rule 32 says that a power must “be signed by the applicant for patent (§ 1.42) or the patent owner,” and that the document is to be signed by a “principal.” § 1.32(a)(2) and (3). But most patent applications are owned by organizations, and organizations can’t sign papers. Only humans can sign papers. So what organizational titles typically communicate authority of the principal of a limited liability company (LLC) to appoint patent attorneys and agents?

By 1996, all 50 states had adopted an LLC statute, and today LLCs are a popular form of organization for business entities. State LLC statutes call the participants “members” (example: the Delaware LLC statute). The formal title of the head of any LLC is specified in the LLC’s operating agreement (the equivalent of a corporation’s charter). The overwhelming majority of LLCs designate the chief officer as “managing member” or “managing director.” For most LLCs, the managing member holds powers equivalent to a corporation’s president, CEO, and Chairman of the Board.

Under basic principles of administrative law, because the agency hasn’t acted by regulation to limit those options, there’s no standard that binds against the public, so a patent applicant can comply with § 1.32(b)(4) in any way that’s convenient and rational.

You’d think that, by 2021, the USPTO would have adjusted its practice for powers of attorney to accommodate law that’s nearly uniform among the 50 states. Certainly, no regulation authorizes the USPTO to refuse a power of attorney signed by an LLC’s “managing member.”

Nonetheless, the attorney email lists regularly include questions of bafflement when the USPTO bounces a power of attorney signed by “managing member” or some other title that any lawyer recognizes as sufficient to appoint an attorney. Typically, the USPTO’s bounce notice includes a counterfactual expression that the power of attorney hadn’t been signed, or that the signatory didn’t have authority to appoint power of attorney. When the attorney phones the USPTO’s “Application Assistance Unit” phone line to ask why “managing member” is not sufficient to appoint power of attorney, and what to do instead, the answer is less-than-uninformative.

The USPTO’s Guidance Document for Bouncing Powers of Attorney

In March 2021, I filed a Freedom of Information Act (FOIA) request asking for all of the USPTO’s “secret rules” regarding signatures on powers of attorney. FOIA requires the agency to produce either documents or an explanation in 20 business days. After eight months of silence, I sued. The USPTO responded to my suit by producing one page. The one page has a breathtaking footer: “Internal Use ONLY.” This content of that one page is a staggering display of ignorance of modern corporate America, and defiance of the rule of law.

The document specifies titles that the USPTO will accept to sign a power of attorney, and those it won’t. The titles the USPTO will accept and not accept are just silly:

  • The column “Not Acceptable Apparent Authority” includes “Managing Member,” “Principal,” “Owner,” “Partner,” and “Delegation on behalf of” the company’s CEO. Any competent lawyer knows that “Managing Member” of an LLC is roughly the equivalent of “CEO and Chairman of the Board” of a corporation. “Principal” is even more astonishing: “principal” is the word used in the operative regulation, 37 C.F.R. § 1.32, and yet the guidance says it’s not enough. Agencies are not permitted to issue guidance that directly conflicts with, or attenuates rights that exist in, agency regulations.
  • Conversely, the “Acceptable” column of the USPTO’s guidance document allows an appointment signed by Vice President for Sales, Treasurer, Treasurer of Sales, Assistant to the President. Even a “Secretary” can be a signatory. How many companies give these officers authority to appoint lawyers? Likewise, according to the USPTO, “Director,” “Dean” of a university, and other similar titles are sufficient authority to appoint power of attorney. But any observer of modern organizations knows that persons with these titles rarely have authority to appoint attorneys.

As President Obama observed, the starting point for any successful policy is “Don’t do stupid stuff.” The law of rulemaking allows agencies to propose really stupid rules. In fact, the law expects agencies to propose really stupid rules. But the whole point of various statutory procedural requirements is to ensure that really stupid proposals don’t get very far or last very long or create very large costs. Because the USPTO habitually ignores or intentionally subverts procedural law, a really stupid rule has been creating really stupid costs for at least a decade:

  • The Administrative Procedure Act (APA) forbids agencies from keeping secrets—rules, interpretations, statements of policy, instructions to staff, and the like, must be made public. 5 U.S.C. § 552(a)(1) and (2). The statute is crystal clear: “A person may not in any manner be required to resort to, or be adversely affected by, a matter required to be published [in the manner required by § 552(a)(1) or (2) respectively] and not so published.” The USPTO has never bothered with statutory publication—indeed, the USPTO’s guidance document is labeled “Internal Use ONLY.” Even though it’s secret, the USPTO enforces it, with no known way to obtain an exception.
  • The First Amendment to the Constitution gives the public the right to “seek redress” to get really stupid rules corrected. But they can’t be corrected if the USPTO keeps them secret.
  • The Paperwork Reduction Act and its implementing regulations require that agency rules must be “plain, coherent, and unambiguous terminology … understandable to those who are to respond.” 44 U.S.C. § 3506(c)(3)(D); 5 C.F.R. § 1320.9(d). Secret rules are not “plain” or “understandable.” Really stupid rules that are kept secret “Internal Use ONLY” are even less so, and can’t be corrected though the Paperwork Reduction Act’s procedures (which we’ll discuss in Part II).
  • An agency can use subregulatory guidance documents as informative, nonbinding “rules of thumb,” but may not enforce them as if they had force of law. So directs the Supreme Court, the President, and the Department of Commerce. As anyone with experience with the “Application Assistance Unit” can confirm, the USPTO enforces this really stupid document as a binding rule, no exceptions.

The USPTO’s “Apparent Authority” guidance is a classic case of a rule that has no reason to exist. Any lawyer with basic understanding of corporate law understands that delegation of authority within a company is entirely a matter of the organization’s own rules, and no federal regulatory agency has authority to say who can or can’t do corporate acts.

Under Executive Order 12866 § 1(a) and § 1(b)(1), agencies shouldn’t regulate where there’s no identifiable problem to solve. For powers of attorney, all the private sector incentives are aligned to achieve the right result—the USPTO has nothing to add by issuing this guidance document. The appointed patent attorney/agent has every interest in making sure that the appointment is valid—the attorney/agent wants to be darn sure he/she will get paid! Likewise, the person signing the appointment has every interest to make sure the appointment is valid—wrongful signature might or might not be a firing offense, but it sure wouldn’t go over well. (There could be a problem of fraud, but the USPTO’s one-page guidance document does nothing to reduce that problem.)

The USPTO’s  rules create a quandary: if “Managing member” is in fact the legal title of the person signing the document, and every lawyer knows that this title is sufficient actual authority for an LLC, does the practitioner use the actual title and fight it out with the USPTO, or does one guess at an alternative title that might satisfy the intake clerks? Those who have tried the former path have found it totally fruitless—despite multiple directives from the President and Department of Commerce that agencies are to train their employees not to enforce subregulatory guidance as binding, the USPTO has never done so. A few weeks ago, I had a phone call with an attorney in the Office of Petitions—even the folks most directly responsible for knowing which rules are enforceable and which aren’t don’t understand that the MPEP cannot be cited as a binding rule. If an applicant’s lawyer has to misrepresent the signatory’s title in order to get the USPTO to accept a power of attorney, is this a misrepresentation that will trigger an investigation by the USPTO’s ethics office? If no one in the USPTO understands the legal limits on guidance documents, how does the public deal with errors in that guidance?

No reasonable person could predict the irrational basis on which the USPTO will accept or refuse powers of attorney. This has gone on for years, simply because the USPTO keeps the guidance document hidden, and doesn’t train its employees that subregulatory guidance is not binding. The public can’t request correction of an irrational guidance document that the USPTO holds secret.

For the Patent Bar: How to Work Around Irrational Rules

First, for powers of attorney, you need to know the titles that the USPTO accepts as “apparent authority” for appointment of power of attorney. Thanks to my FOIA request, I can let you in on the current rules here and here. They’re absurd, I can’t explain them, but I can share them with you.

Until the USPTO begins to follow the law, and stops imposing requirements that have no grounding in properly-promulgated regulation, you can rely on another “secret” rule. I can’t find it in any regulation or the MPEP, but it seems to exist in uniform intra-USPTO folklore: if you include words like “authorized to act on behalf of [the assignee company]” or “authorized representative of [the assignee company]” near the signature block of the power of attorney (even if you don’t include all the documentary evidence required by § 3.73(c)), the USPTO will accept the power—most of the time. The title of the signatory—most of the time—won’t matter. But because consistency is so low-priority at the USPTO, the attorney email lists discuss cases where an individual USPTO employee overrides that folklore, especially where the stated title is “managing member” or another title from the right column of the one-page guidance document.

Director Vidal: Implement the Recommendations

Unlike other rules, where the USPTO has an economic rationale for shortcutting, for powers of attorney there’s no motivation or upside for cheating. The only visible explanation is force of habit, a culture in which scofflawry is the norm. In our letter of March 2021, 91 attorneys offered several pages of recommendations—we identified ways in which the USPTO’s handling of powers is unpredictable, and ways to fix the unpredictability. Through four rounds of notice-and-comment, these recommendations have now had thorough vetting by the USPTO. The USPTO has never explained any reason to believe these recommendations can’t be implemented or would be costly to the USPTO, and has given no reason to disagree that they’d reduce burden on the public by tens of millions of dollars per year. Director Vidal, you have the authority to direct your staff to implement them. In fact, under the Paperwork Reduction Act, 5 C.F.R. § 1320.5(d)(1)(i), implementation is nondiscretionary.

In Part II, we’ll follow this rule through the dark side—the USPTO’s legal errors are not confined to sins of omission. Instead, when this rule came up for regulatory review by the Executive Office of the President—the triennial review where agency good faith errors are supposed to be corrected—the USPTO did everything possible to avoid self-correction or observance of law. In papers to an ex parte regulatory review at the Executive Office of the President, the USPTO gave a number of factual and legal averments that have no objective support, and certifications that are counter to the evidence before the agency, and for which the USPTO apparently neglected any investigation. Those statements apparently were signed off by USPTO lawyers. Any reasonable person would have had to know they were incorrect, especially after the errors had been pointed out in earlier comment letters. When the falsifications were pointed out, the USPTO doubled down on them. It’s hard to come away from the record of this regulatory review with any inference other than intentional avoidance of candor, by USPTO lawyers, made under certification, to influence a decision of an ex parte tribunal.

IPWatchdog reached out to the USPTO for comment on the points made in this article. The Office provided the following written response:

  • The 2016 document is just a training resource to explain prior examples where the authority to sign and prosecute patent applications on behalf of the applicant (“apparent authority”) has not been found, and best practices for when it has, consistent with MPEP sections 324, V; 325 V; and 1820. 
  • It does not apply when filers signing on behalf of an organization make an assertion of actual authority, which would avoid potentially ineffective signatures. 
  • USPTO forms include this “actual authority” assertion (see USPTO’s Transmittal for Power of Attorney form here).
  • We always welcome comments or suggestions to improve the MPEP and encourage stakeholders to utilize our online discussion tool here.
  • We believe in a robust conversation about our procedures and are happy to answer questions about them and their importance to our stakeholder community.



Warning & Disclaimer: The pages, articles and comments on IPWatchdog.com 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 IPWatchdog.com. Read more.

Join the Discussion

5 comments so far. Add my comment.

  • [Avatar for Julie Burke]
    Julie Burke
    July 14, 2022 12:37 pm

    It was June 1995. I was at the USPTO patent training academy. It was our first day on the job.

    The instructors made their view of agency’s mission crystal clear:

    The most important thing you new patent examiners need to know- no matter what else goes on, NEVER EMBARRASS THE OFFICE.

    From my experience as a TC 1600 Quality Assurance Specialist, every single thing in David Boundy’s article rings true.

    Nothing in my two decade career has changed my initial impression: USPTO career management then and now values protecting the institution and their positions over anything else.

    If that mindset is not a form of ‘Deep State,’ what is?

  • [Avatar for Anon]
    July 14, 2022 11:18 am

    While certainly new and important (not merely of itself, but to show the continuing disregard for the patent community); the “Deep State**” of the USPTO has long been known to violate 37 CFR 1.2 (to wit: The action of the Patent and Trademark Office will be based exclusively on the written record in the Office. is NOT merely binding on applicants).

    As has been noted many times in relation to continued shadow programs (like SAWS was), when the Office was busted trying to deny that SAWS even existed, they publicly stated that SAWS was but one of many such programs.

    ** I recognize that the term “Deep State” may set some otherwise intelligent people scurrying for the “That’s Right Wing Conspiracy talk” line of defense, but if the Office does not want that type of label, then they should stop taking actions that support that type of label.

  • [Avatar for Pro Say]
    Pro Say
    July 13, 2022 03:26 pm

    Follow laws, rules, and regulations like the other government agencies do?

    Chuckles. We’re way, way past April Fools day!

    Way past.

    — Notorious P.T.O.

  • [Avatar for Julie Burke]
    Julie Burke
    July 13, 2022 11:41 am

    Thank you, David, for this in deep dive into the USPTO’s parsing of authorized officer titles they deem acceptable for powers of attorneys.

    Interesting that in the FOIA-obtained internal email from June 2017, Deputy Director of International Patent Legal Administration alerted MPEP staff of certain “clarifications” pertaining to acceptable and unacceptable titles variations not specifically set forth in MPEP 325(V) and 1820.

    The FOIA-obtained 2016 document, identified now by the USPTO as “just a training resource” includes lengthy list of titles that USPTO considers as acceptable or not. USPTO employee and contractors likely rely on this list when deciding if a title is acceptable or not.

    Five years later, this extensive list of acceptable and unacceptable titles (whether legally appropriate, I have no opinion) is still not included in the MPEP or otherwise made available to the IP community. Til this article. At a cost of $30 to $40 million a year, that’s a lot of expensive confusion and uncertainty being imposed onto the IP community.

    And from a quick look, contrary to the USPTO’s responder, the information in the 2016 document is not consistent with the MPEP. For examples:

    The 2016 document lists “Chairman” as an acceptable title, while the MPEP limits this to “Chairman of the board of directors.”

    The 2016 document lists Dean of University as acceptable- a title not mentioned in the MPEP.

    The 2016 document permits the use of non-English title Geschaffsfuhrer, which is not mentioned in MPEP.

    Again, the USPTO operates from a different set of rules than that made available to patent practitioners.

    Given differences between the 2016 internal document and MPEP, it’s not surprising that USPTO employees and contractors, too, are confused as to whether a title is deemed acceptable or not.

    Applicants are left in the dark to play guessing games as to whether an individual’s actual title will be accepted.

    Looking forward to Part II.

  • [Avatar for David Boundy]
    David Boundy
    July 13, 2022 09:37 am

    Dear anonymous PTO person —

    I have no doubt that there are individuals in the USPTO that individually and subjectively believe “We believe in a robust conversation about our procedures and are happy to answer questions about them and their importance to our stakeholder community.” That view is not universal within the USPTO. Part II will be about actual agency behavior in response to comments raised in formal notice-and-comment proceedings.

    Likewise, I am certain that individuals hold the aspirational belief that “We always welcome comments or suggestions to improve the MPEP and encourage stakeholders to utilize our online discussion tool here.” Actual agency behavior is otherwise — the observation of user community is that IdeaScale is “where good ideas go to die.” https://blog.oppedahl.com/?p=6123

    The two-part article is symptoms and diagnoses. Part II concludes with an invitation to contact me to discuss a treatment plan.

Add Comment

Your email address will not be published.