Evidence Based Prosecution I: Sensitivity to Claim Fee Variation


The graph above is a cumulative frequency chart of the total number of claims per application for a large sample of recently filed patent applications.  You can see that almost half of new applications are field with fewer than twenty-claims. Then there is a jump — twelve-percent of new applications have exactly twenty claims.  Above twenty claims, the numbers begin to drop-off dramatically, with only nine-percent of applications filed with more than forty-claims.

Of course, the issue here is cost. The USPTO charges a fixed thousand-dollar fee for applications filed with twenty-claims or fewer. Each additional claim is fifty bucks. (Fees are halved for small-entities).

Interestingly, this data spans across a change in the Patent Office fee structure.  Before December 2004, fees for extra claims were only eighteen dollars each.  At first glance, the thirty-two dollar jump from eighteen to fifty is seemingly de minimis when compared to the thousands of dollars in attorney-fees associated with the filing (not to mention R&D costs).  As it turns out, a significant number of patentees responded to the change in fee structure by filing fewer claims. Prior to the rule change, for instance, about sixty-one percent of applicants stayed within the twenty-claim guideline. After the rule change, sixty-five percent filed no more than twenty claims.

This is one example of how patent applicants are sensitive to even small changes in administrative fees. The data is important because it gives credence to suggestions that the USPTO fee structure can serve as a successful policy instrument.  For instance, rather than limiting the number of claims or continuations to hard numbers, the PTO might want to simply apply a high surcharge and let the market go.  The data also gives empirical support for ideas such as Lemley, Lichtman, & Sampat Gold Plating.

Cite as Dennis Crouch, “Patent Prosecution: Sensitivity to Claim Fee Variation,” Patently-O (Sept. 21, 2006), available at https://patentlyo.com.

13 thoughts on “Evidence Based Prosecution I: Sensitivity to Claim Fee Variation

  1. 13

    Stephen, if you look at the file history for that case, you will see that they filed the preliminary amendment with payment of the fees, and did not pay for the excess claims that were removed by the preliminary amendment.

  2. 12

    As a corollary to the RCE point (yes, I know they were excluded): the allowance/rejection of claims is extremely examiner dependent. Even if you spend thousands on researching all references and sculpting a hypothetical claim with the perfect scope, the examiner assigned to the case may not allow the claim. On the other hand, you may have an examiner who allows a claim which should never have been allowed.

    The Office must know this but for probably political reasons cannot even acknowledge this. Thus, it is often a total crapshoot as to whether claims will be allowed, and helps to increase the number of RCEs filed.

    Most applicants, in my humble experience, would also like to get something for their money. Thus, whether they agree or not with the rejection, they like to take allowed claims and file a continuation(especially as appeals can get expensive quickly).

    Back to the main point, of course most applicants are sensitive to administrative costs. If you are a larger company filing hundreds or thousands of applications, you are extremely cost conscious to a slight change as it adds up quickly. If you are an individual inventor or smaller company, you are unlikely to have the overall budget. Yes, it is a powerful policy lever. However, it should be used wisely.

  3. 11

    But there are always going to be a few outliers. My favorite is a whole series of patent applications filed by KLA-Tencor several years ago. Published Application 2002/0107650 is typical, and was filed with 6,632 claims. The story gets curiouser and curiouser — all but 83 claims were then cancelled, and U.S. Patent 6,782,337 issued with a grand total of 83 claims. According to PAIR, no continuing application was filed. Thus some 6,549 claims were filed (and I presume paid for), and then dedicated to the public use by failing to ever prosecute. I’m sure that there must be some logic to this tale, but confess it entirely eludes me. But if nothing else, I’m sure this series alone skews statistics for the relevant period.

  4. 10


    I am, unfortunately, acutely aware of the problem you describe: crappy first search, overcoming amendment, new search, crappy final rejection, RCE, repeat.

    That’s why I excluded RCEs from my definition of continuation above.

  5. 6

    Steve, Great comments. I agree with you that this can be complicated. My main point here is that patentees are swayed by changes in the PTO fee structure — thus making it a potentially powerful policy lever.

  6. 5

    You note that “[p]rior to the rule change, for instance, about sixty-one percent of applicants stayed within the twenty-claim guideline. After the rule change, sixty-five percent filed no more than twenty claims.”

    One thing that might bolster your argument is a statistical analysis of these numbers. It would also be cool to do a graphical overlay from a previous time period (say, pre-rule). Overall, your idea is quite interesting, though.

  7. 4


    Unfortuantely is some very dense art areas it takes an RCE or two (or three) to get a claim to the point where an Examiner will issue it. It is an unfortuante fact of life, and one that the Office doesn’t seem to understand.

  8. 3

    Dennis, your analysis assumes we file just one application with all of the claims we want. It doesn’t have to be that way. As the number of claims grows, there are points where it is cheaper to file additional applications. The analysis has some complexity because the cost depends on both total claims and independent claims. The downstream costs need to be considered, too. The rule of thumb is that when you exceed 40 claims, you should start thinking about splitting the claims into two applications.

  9. 2

    That is a good idea. The majority of continuations (excluding RCE’s and CIPs) I believe are filed to avoid the onerous procedures of reissue and to provide copendency for correction of errors in prosecution. If the cost goes up, reissue does not look so onerous and may cause applicants to forego the continuation application, which currently is a bargain compared to reissue.

  10. 1

    I think this is a much better approach to changing Applicant behavior than the proposed changes.

    RCE practice might be a good example: It would still allow those companies/inventors that need the current RCE practice (assuming the RCE cost goes up for every RCE filed) to balance it against cost. GREAT idea, lets hope the patent office is listening.

Comments are closed.