USPTO Institutes Additional Fees for Large Information Disclosure Statement Filings

By Michael Borella and Joshua Rich --

The advice to practitioners faced with marginally relevant prior art has long been "when in doubt, cite it."  There was a small cost for the applicant (or practitioner) to cite such art by filing an information disclosure statement (IDS), but substantial potential risks that could arise if the art was not cited.  On the other hand, examiners had to consider tangentially-related art along with the most relevant art.  The USPTO has now decided to institute a fee for filing large IDSs, which may very well have unintended, adverse consequences for the patent system as a whole.  It will create greater ethical challenges for practitioners and may degrade patent quality.  That is, the USPTO's cure may be worse for everyone involved than the problem has been for examiners.

The duty to disclose prior art on an IDS is unique to U.S. patent law practice.  While some other jurisdictions have prior art disclosure procedures (for example, in Canada, applicants can identify prior art voluntarily or do so in response to an examiner's request) or require identification of co-pending applications in other jurisdictions (such as in India, where the art cited in these applications may be used for evaluation of patentability), the U.S. requirements are both obligatory and more extensive.  As set forth in 37 C.F.R. § 56(a), the duty of disclosure requires that:

Each individual associated with the filing and prosecution of a patent application has a duty of candor and good faith in dealing with the Office, which includes a duty to disclose to the Office all information known to that individual to be material to patentability as defined in this section [emphasis added].

§ 56(b) adds some clarity on whether art is "material", stating that "information is material to patentability when it is not cumulative to information already of record or being made of record in the application."

The duty applies to a surprisingly large number of individuals -- all patent practitioners (whether in house or outside counsel) working on the application, all inventors, and any other individuals "associated with the filing and prosecution of a patent application."

A failure to discharge the duty of disclosure can have severe consequences for both the patentee (a finding of inequitable conduct can render the patent unenforceable) and practitioners (who can face professional discipline).  That is, the cost of failing to disclose a material, non-cumulative reference is potentially devastating.  Accordingly, it is critical for an applicant to thoroughly disclose all such references.

But the rub with the duty to disclose is that there is no clear definition of what is material to patentability or cumulative.  Both of these terms have a degree of subjectivity, and will be fiercely litigated based on facts unique to each case.  Given that, the aforementioned individuals are incentivized to disclose all known art when filling out IDSs because the cost of an incremental disclosure is far lower than that of even just being accused of inequitable conduct.

The USPTO contends, however, that "[in] many instances, these large IDS submissions contain clearly irrelevant, marginally relevant, or cumulative information."  To battle this perceived problem, on November 20, 2024, the USPTO published its final rule for "Setting and Adjusting Patent Fees During Fiscal Year 2025" in the Federal Register.  While the final rule includes a number of fee increases and new fees, the IDS-specific fees are set forth in Table 10 and excerpted below.

Notably, while Table 10 provides the large entity (undiscounted) fees, the final rule makes no mention of small or micro entity discounts for these fees.  Thus, it appears that the new IDS payment rules are as follows for all applicants:

a first amount ($200) in § 1.17(v)(1) for a cumulative number of applicant-provided or patent owner-provided items of information in excess of 50;

a second amount ($500) in § 1.17(v)(2) for a cumulative number of applicant-provided or patent-owner-provided items of information in excess of 100 but not exceeding 200, less any amount previously paid under § 1.17(v)(1); and

a third amount ($800) in § 1.17(v)(3) for a cumulative number of applicant-provided or patent owner-provided items of information in excess of 200, less any amounts previously paid under § 1.17(v)(1) and/or (v)(2).

The USPTO justifies these fees by noting that "[a]bout 5% of applications contain 51 to 100 applicant-provided items of information, about 4% of applications contain 101 to 200 applicant-provided items of information, and only 4% of applications contain more than 200 applicant-provided items of information."  While this means that the new IDS fees are expected to only impact about one of every eight applications, the USPTO states that it has provided examiners with "approximately 80,000 additional hours each year to consider large IDS submissions in applications, costing the agency $10 million annually."  The USPTO further asserts that "this cost is subsidized generally by patent fees, primarily maintenance fees collected for patents that resulted from applications that did not contain large IDS submissions."

At first blush, this seems to be a reasonable justification for the fees.  However, upon deeper reflection it is not clear how the USPTO has calculated these numbers and whether they reflect actual costs in practice.  For example, it is common practice for applicants to file continuation applications with an IDS citing all art from all parent applications.  This can easily break the new 50-reference limit on free IDSs, especially if the application family has foreign counterparts where art has been cited or is perhaps the 3rd or 4th continuation (or more).

This is less of a burden than meets the eye because the USPTO typically assigns the same examiner to all U.S. applications in a family.  So the majority of references in an IDS for a continuation may have already been considered by the examiner, and the examiner can proceed through the cited references much more quickly than references that are actually new.  If the USPTO's analysis does not take this factor into account, it is misleading and could dramatically overestimate the actual cost of handling large IDSs.

Regarding continuations and the like, the USPTO writes:

The cumulative count is determined for each application or patent separately.  That is, the count from an application does not carry over to any continuing applications, CPAs, reissue applications, or any post-issuance proceedings such as supplemental examinations or reexamination proceedings.  Instead, continuing, CPA, and reissue applications and post-issuance proceedings will start with a count of zero.[1]

This means that references cited in a parent application do not count against any of its child applications unless those references are re-cited in the children.  But the USPTO reiterates its policy that examiners must "consider items of information that were considered in a parent application when examining a child application (e.g., a continuation, continuation-in-part, or divisional application) without any action required on the applicant's part."  It continues by clarifying that "applicants who wish to avoid paying the IDS size fees in a child application for items of information considered in a parent application may do so by not resubmitting the items."  That is, for the examiner, an applicant citing the same references makes essentially no difference in the scope of art that must be considered, but the applicant may face substantial fees for re-citation of those references.

The USPTO's statements suggest a fee-reducing solution for applicants -- do not cite any references in child application if those references were cited and considered in any of its parents.  The applicant will still need to cite any references that were not cited in the parent applications, but that number is likely to be significantly smaller.  The downside to this strategy is that the uncited references from the parent application will not appear on the face of the any patent that issues from the child application.

The value of the cited references being so clearly set forth in the patent is debatable, but it is clear that it provides some benefit.  On one hand, whether references are printed on its face has no material impact on the validity of a patent.  On the other hand, it may be helpful to show a jury (and perhaps certain judges) this crystal clear evidence that the examiner found the issued claims patentable over the references.  And for the public, it provides immediate notice that the cited art was considered by the examiner without requiring a dive into the prosecution history.

To simplify its enforcement of the IDS fee plan, the USPTO is also requiring that "an IDS contain a clear written assertion by the applicant and patent owner that the IDS is accompanied by the appropriate IDS size fee or that no IDS size fee is required."  While "[t]here is no specific language required for the written assertion, . . . it should be readily identifiable on the IDS and clearly convey the applicable IDS size fee by specifying the particular paragraph in § 1.17(v) that applies (e.g., ''the fee due under 1.17(v)(2)''), if any."

The USPTO will update its standard IDS form to incorporate this language, but there is a trap for the applicant.  Unlike other fees, the applicant cannot generally authorize the office to charge the appropriate IDS fee to a deposit account.  Instead, "authorization to charge fees to a deposit account is not a compliant written assertion under the new § 1.98(a) requirement, unless the authorization clearly identifies the particular IDS size fee that should be charged for submission of a particular IDS."

This means that the applicant needs to maintain an accurate count of applicant-submitted references per application.  Doing so is an onerous task that the USPTO is pushing on to applicants.  While there may be software tools to help an applicant do this bookkeeping, these tools are not free and the labor cost of using them is not cheap.  So, for the privilege of paying the USPTO more fees, applicants have do more work.  And the burden will fall most heavily on small and micro entities and lower-cost practitioners who have less ability to spread the cost of calculation over numerous applications.

The USPTO states, to the contrary, that "the fee should not disproportionately impact small and micro entities."  We disagree.  All USPTO fees disproportionately impact small and micro entities.  The largest entities are multi-billion and trillion dollar companies.  While their patent budgets are finite, they are often quite large and can be adjusted for inflation and fee increases.  On the other hand, small and micro entities are often startups, small businesses, or individual inventors.  These entities are self-funded or funded through loans, angel investors, and/or venture capital, and every dollar counts.  Patent filing and prosecution (not to mention inter partes reviews) are significant costs that are harder to absorb than for larger entities.  As a general premise, bureaucracy favors the powerful.  Here, the USPTO's increase in the bureaucratic process required to comply with the duty to disclose will disproportionately impact those with fewer resources.

In any event, what is an applicant to do in light of the final rule and its new IDS fees?  Here are a few concrete suggestions to reduce the impact of these fees and the time-cost of associated labor.

• As noted above, refrain from filing previously-cited art in IDSs in continuations and other forms of child applications.  While doing so results in the references not appearing on the issued patent, they are still of record in the file wrapper of the parent(s).

• The new fees are based on the number of applicant-cited references per application, but not the size of these references.  To the extent that the applicant can cite to longer, more comprehensive references in lieu of multiple shorter references containing the same content, the applicant should do so.[2]  For example, a product manual (if it qualifies as prior art) is often one of the most detailed and relevant references available to applicants and could (in principle) replace several other references that are cumulative.

• Many applicants use third-party prior art search firms.  Applicants may wish to instruct these firms to provide search results for no more than the very most relevant references found.  Applicants have no duty to disclose references of which they are not aware and do not have in their possession.  Of course, this may make the examiner's job more difficult and decrease patent quality, but it is a problem the USPTO is buying for itself.

• Applications with corresponding academic publications may be the type that most easily run afoul of the new fees.  Modern academic papers routinely cite dozens or even hundreds of non-patent literature references.  Some of these are quite lengthy (e.g., book chapters or entire books).  Moreover, it may be difficult for patent practitioners to reasonably determine the relevance or cumulative nature of such references in highly sophisticated, complex sciences and technologies.  Current practice is to cite all references in a corresponding paper in an IDS, and there may not be a way to avoid doing so without incurring more actual cost than the USPTO fees themselves.

• While applicants should avoid citing references of marginal relevance, they need to remember the subjective nature of such an analysis.  When push comes to shove, it is better for both the applicant and practitioner to pay the USPTO a few hundred dollars than risk any eventual patent being found unenforceable due to a known but uncited reference.

In short, there may be steps that applicants can take to minimize the effect of the new fee rules.  But we are going to see prosecution practices change in unpredictable ways that may not benefit applicants, practitioners, examiners, or the public.

[1] Given that the cost of Requests for Continuing Examination (RCEs) are going up significantly, the cost of an IDS citing additional references may factor heavily into whether an applicant files a continuation application or an RCE.  This is another unintended consequence of the USPTO’s fee changes, but it is unclear how it will change prosecution practice.

[2] It should go without saying that the applicant must never engage in gamesmanship, such as combining multiple references into one file and submitting it as a single reference or citing a later-dated, more comprehensive document instead of an earlier publication.

Comments

[Mike Borella]

One other thing we forgot to mention. The final rule does not provide a grace period for applications filed prior to the rule going into effect. In other words, if you have a pending application with 50 references cited pre-rule, your first post-rule IDS would trigger IDS fees.

[Examiner so and so]

I see your concerns from Applicants/client side, but there are serious concerns on the examiner side that I don’t think you see (and I’m confident the Office has not relayed well). I have had a good amount of experience with this exact issue at the Office (large IDSs) if you want to chat offline.

[Paul F. Morgan]

These are not unreasonable fees, but if it is of real concern to a client, here are two ideas [for responsive adverse comments?] to avoid these excess prior art citation fees?: (1) When ordering prior art searches, set reasonable limits for the searcher on the number of reported items. (2) Cite unknown claim materiality prior art collections in the spec background section rather than on the IDS form.

[Dan Feigelson]

These IDS fees, like most of the other fees imposed by the USPTO recently (e.g. "non-docx filing fee") are nothing but a money grab. Everyone knows that examiners (a) often don't look at the things on IDSes, and (b) are not given time to look closely at every item on even a relatively small IDS, let alone an IDS with 50 references or more. So as it stands currently, IDS practice is in many if not most cases a legal fiction - and it's certainly a fiction in cases of large IDSes: the Examiner says "all the art was considered", even though it wasn't, and in a licensing or acquisition negotiation, or in litigation, everyone pretends that the examiner actually considered everything in the IDS and that the patent is therefore stronger. The new fees won't change a thing about that.

Rather, IDSes will continue to be primarily a cover-your-rear business for applicants and their attorneys, because of Rule 56, and because of the court-created doctrine of inequitable conduct (which creates obligations that overlap with but are not the same as those of Rule 56). And the PTO will rake in money that does nothing to improve examination, because it will still be cheaper to pay for a longer IDS than it will be to later on litigate the omission of some publication from an IDS. That cost disparity will be further exacerbated if the applicant or attorney spends time trying to sort out which references are relevant and which references are cumulative of one another. (I don't know any practitioners who will do that - even if a client will pay for it, the malpractice risk is too high. Just pony up the fee).

Yet another present from Kathi Vidal. I sincerely hope the person Trump picks to replace her undoes the lousy things she's done and cleans up the mess she's left.

[Max Drei]

Is there any rebuttal to what Dan Feigelson so trenchantly writes? It makes total sense to me. You?

How shall Vidal's successor clear up "the mess"? Merely to rescind the new rule doesn't solve anything. Does a solution to the problem exist? If so, what is it?

[Karen Oster]

When I file a reference in a foreign language, I often will provide a translation as an NPL. This doubles the number of references. Any suggestions on how to handle this? Can they be filed as one document? Or is that "gaming the system"?

[James Demers]

Karen- I provide just the translation. It conveys the content of the art (I include legible reproductions of photos, drawings, charts, etc.), and the publication date, which are all the Examiner needs.
(I don't worry about failing to disclose some detail due to a poor translation, since I'm only aware of what the translation says.)

[Dan Feigelson]

Karen and James: why not merge the original foreign language document and the translation into a single pdf file, and submit it that way, listing it as single document on the IDS?