At a high level, patent examination is a straightforward process: The applicant files a specification and claims. The examiner issues an office action setting forth the basis for rejecting the claims. The applicant responds with amendments and/or arguments, and the examiner issues a next office action, etc. This turn-taking continues until the application is allowed, abandoned, or appealed.
However, delving into the details of the process reveals a surprising amount of administrative complexity, much of which centers on the classification of each office action as “non-final” vs. “final.” While the names suggest a distinction of significant gravity (as in “tentative” vs. “terminal”), the option of overcoming the “finality” of an office action by filing an RCE – and even an unlimited sequence of RCEs for a sequence of “final” office actions – negates any such distinction.
These observations present the basis for today’s Mystery of Patent Law: Why does U.S. patent examination involve unnecessary distinctions between “non-final” and “final” office actions?
The characterization of office actions as “non-final” and “final” has no basis in Title 35 of the United States Code (which is not surprising, since this body of laws defines substantive rights and the overall structure of patent examination, but not the finer details). 1 The characterization is amply discussed in Title 37 of the Code of Federal Regulations, and any patent examiner or practitioner can list a number of procedural differences:
- Between the non-final office action and the final office action, the applicant can have claim amendments entered as a matter of right. After the final office action, claim amendments are only permitted at the discretion of the examiner.
- A reply to a non-final office action, filed within six months of the office action, satisfies the applicant’s obligation and keeps the application pending. Such a reply places the case back on the examiner’s docket, and the applicant doesn’t need to do anything further to keep the case pending. However, a final office action requires the applicant to appeal or file an RCE to keep the case alive. 2
- According to MPEP 706.07(a), an office action that introduces a new ground of rejection that is not necessitated on a claim amendment or IDS must be non-final. If the examiner identifies it as a final office action, the applicant can request to have the finality withdrawn.
- Following a final office action, an application can be enrolled in the After Final Consideration Pilot (AFCP 2.0) Program to give the examiner additional production credit for conducting a follow-up search and interview. Applications that are only non-finally rejected are not eligible for the program.
- While examination production hours are generally awarded for each round of prosecution, the actual events that trigger examiners’ credit are (1) the First Office Action on the Merits; (2) the first office action following an RCE, which is typically non-final); and (3) the final disposition of the application (abandonment, allowance, or appeal). In other words, examination credit is awarded for non-final office actions, and no credit at all is awarded for final office actions. Accordingly, examiners are inclined to do significantly more searching and reconsideration while preparing the (first) non-final office action than the final office action.
And yet, to most applicants 3, the primary difference between “non-final” and “final” office actions is:
- A final office action requires paying an RCE fee in order to continue prosecuting the application before the examiner. A non-final office action doesn’t.
This discrepancy therefore raises the question: What is the point of the “non-final” and “final” categorization of office actions?
The legislative motivation that prompted the creation of this characterization is lost to history. While 37 CFR and the MPEP extensively discuss the procedural differences, neither text indicates the *purpose* of such characterization. I did a bit of casual research into the legislative history of 37 CFR (including a trip through the microfiche of my law library), but couldn’t find anything more than a mechanical recitation of which words had been changed, and when.
Also missing is any identifiable advantage upon the patent process. To put it simply – if an application requires six office actions over the course of three years to reach a disposal, why should it matter whether each office action is “non-final” or “final?”
However, distinctly *not* missing, and readily apparent, are the significant disadvantages of this approach. In addition to the needless administrative complexity, the characterization itself is a major source of inefficiency in the USPTO’s limited examination resources. Because the production system specifically credits non-final office actions and does not credit final office actions, the latter are typically a rehash of the former. That is, approximately half of issuing office actions are largely a rehash of the other half. 4 And it’s widely acknowledged that the production count system itself encourages examiner gamesmanship, such as issuing a final rejection and then promptly allowing the same claims when the applicant files an RCE. And despite not conferring progress in the examination, final office actions take time for the examiner to prepare and send – i.e., final office actions are a waste of the examiner’s time, the practitioner’s time, the applicant’s budget, and the patent term.
So it seems that our attachment to the “non-final” / “final” examination system is like the QWERTY keyboard: a peculiar artifact of history with more disadvantages than advantages, and that persists because of its ubiquity (and vice versa). However, given the trendiness of “patent reform,” including the America Invents Act revision of Title 35 of the U.S. Code (thereby replacing some truly horrid legislative language with a much more readable version), perhaps a rewriting of 37 CFR is achievable.
In an ideal rewrite, every office action is “non-final,” unless the application is allowed, abandoned, or appealed – much the same requirements as the definition of a “disposal” today. Examiners receive two counts for the first action on the merits (to account for the process of getting up to speed with the application), and one count for each successive office action. RCEs would be entirely eliminated – rather, applicants would pay a single flat fee with every reply (e.g., 50% of the current RCE fee), which would cover the cost of the examiner’s updated search, reconsideration, an interview (always available for free at the applicant’s request), and the next office action. 5
I believe that a streamlined process like this would confer several advantages on our patent system:
- Reducing the administrative complexity in the examination process.
- Giving examiners the resources, during every office action, to perform a substantive reevaluation of the case. (Also, providing a structure whereby examiners can be held accountable for failing to do so, such as with cursory, “copy and paste” office actions that just “kick the can down the road” toward the RCE.)
- Improving the efficiency of the USPTO’s use of resources.
- Maintaining applicants’ USPTO fees at their approximate current levels, but giving applicants more bang for the buck out of every office action by reducing poorly considered office actions.
- Tying the administrative costs of patent examination, and the collection of fees, more closely to the actual allocation and use of USPTO resources. (The USPTO’s finance managers should appreciate this.)
- Making the patent process easier for applicants to understand (and for practitioners to explain)
- The closest reference in 35 USC to “finality” of an office action is 35 USC 134, which provides that a claim that has been “twice rejected” can be appealed to the PTAB. However, “twice rejected” is irrespective of whether such rejections are “non-final” or “final” – and 37 CFR 41.31 explicitly says so. ↩
- The finality can be tolled by filing an after-final reply within two months, but the examiner’s response – which can be as simple as “applicant’s arguments aren’t persuasive, and amendments aren’t entered because they require a new search” – places the onus back on the applicant. ↩
- The other effect that’s relevant to applicants is the patent term adjustment, since the USPTO’s calculation only applies to any delays occurring before the first RCE. However, this aspect only tends to be relevant to patents with strong value remaining at the end of the patent term – specifically, pharmaceutical companies. ↩
- Indeed, the Code of Federal Regulations appears to condone this tactic: 37 CFR 1.113 indicates that “in making such final rejection, the examiner shall repeat or state all grounds of rejection then considered applicable to the claims in the application, clearly stating the reasons in support thereof.” ↩
- This fee structure could perhaps be modified to account for office actions that really are just a rehash of the previous one: “applicants’ arguments have been considered but are unpersuasive; former rejection maintained.” Because such office actions didn’t require the expenditure of USPTO resources that the applicant paid for, the reply to *that* office action could have a 50% discount. ↩