One of the issues often cited in discussions of continuation reform is the so-called 'wearing down of examiners'. The theory is that by taking advantage of (or abusing) the unlimited continuations available to applicants, they can wring an undeserved allowance from an examiner simply by refusing to abandon the application.
The latest incarnation of this argument is in the June 2008 edition of the JPTOS, in an article (this is a link to an abstract only) by James W. Beard, a JD candidate at the University of California. Here's what the article has to say, although it's nothing that we haven't heard before:
"The danger posed by repeated continuation application lies in the nature of the incentive structure at the USPTO. Examiners are only rewarded for the initial response to a patent application and for finally disposing of an application, and get no additional incentive processing "a patent that cites 150 pieces of prior art and has 120 claims than a patent that cites two pieces of prior art and has three claims." [Comments by Mark Lemley at the Symposium, "Ideas Into Action: Implementing Reform of the Patent System", 19 Berkeley Tech L.J. 1053, 1112.] Disposing of an application requires granting a patent, and because an applicant can continually refile amended claims and continuation applications in response to an examiner's rejection or criticism, this creates a weakness in the process. Examiners have no incentive to spend additional time on more difficult applications; by the time it is on their desk they have already been rewarded for taking the case, and the only means to get it off their desk is to grant it. Therefore, if an applicant were sufficiently stubborn in filing continuation applications, the examiner's incentive would be to both spend less time on each successive continuation application and, finally, to grant the application even if the claims were overly broad. Effectively, the current incentive structure of the USPTO leads to a tendency to give decreased vigilance to successive applications. In sum, this allows a determined applicant to obtain a patent monopoly on something that would otherwise be unpatentable, resulting in significant costs to future inventors and potential infringers."
But wait, there's more...
"The proposed rule limiting applicants to two continuation applications and one request for continuing examination would have mitigated the most serious instances of applicants repeatedly filing continuation applications to wear down examiners, but it does not completely eliminate the issue. Somewhat counter-intuitively, there is almost no reduction in examination time for applications containing continuations relative to those unamended. [Stuart Weinberg, Inventor Sues U.S. Patent Office Over New Rules, Dow Jones Newswire, Aug. 23, 2007] There is therefore no incentive, in terms of efficiency, to having the same examiner prosecute each filing."
The author then goes on to posit that since there is no increased efficiency in having the same examiner assigned continuations, then having them assigned randomly might help with this whole 'wearing down examiners' thing.
Of course, it's all bull, and anyone who knows anything about how the count system actually works knows it. So let's look at a few of the author's points.
"Examiners are only rewarded for the initial response to a patent application and for finally disposing of an application..."
Wrong. Every time an applicant files an RCE (or a straight continuation), the examiner receives a count for the express abandonment, and another count for the first action after RCE. If they are doing their job right, the examiner will get a count for every single action they write (excluding advisory actions), the abandonment count acting as a 'deferred' count for the final rejection.
"[Examiners] get no additional incentive processing a patent that cites 150 pieces of prior art and has 120 claims than a patent that cites two pieces of prior art and has three claims."
Well, no argument there, but this doesn't really have anything to do with 'wearing down examiners' with excessive continuations. Different issue altogether.
"Disposing of an application requires granting a patent..."
Wrong. Abandonment and appeal will also dispose of an application.
"...by the time it is on their desk they have already been rewarded for taking the case, and the only means to get it off their desk is to grant it."
Wrong. I have to assume that by 'off their desk' the author actually means 'off their docket'. Well, yes, the only way that I can take unilateral action to get an application off my docket is to allow it. All of the other ways for applications to leave my docket [abandonment or appeal] are at the sole discretion of the applicant. Even so, as discussed above, there is nothing that I do that I will not get 'rewarded' for, assuming that I'm making good rejections that do not necessitate writing second non-finals.
"...if an applicant were sufficiently stubborn in filing continuation applications, the examiner's incentive would be to both spend less time on each successive continuation application and, finally, to grant the application even if the claims were overly broad."
Believe me, working under the requirements of the current production system, the examiner's incentive is always to spend as little time on an application as possible. Nonetheless, taking the time to find the best art up front and make a good first action is fully consistent with that goal, since that will minimize the time spent on subsequent actions, and will tend to shorten prosecution history by getting the issues focused sooner.
"Somewhat counter-intuitively, there is almost no reduction in examination time for applications containing continuations relative to those unamended...There is therefore no incentive, in terms of efficiency, to having the same examiner prosecute each filing."
Uh, yeah. Unfortunately, I couldn't find a copy of the original Dow Jones Newswire article, but I'm curious as to whether Mr. Weinberg spoke to any examiners before he made this assertion.
Speaking for myself, examining a continuation really does give me a serious head start. Usually, I'll have a good enough recollection of the parent application that it is much more efficient to work the child application. Even if it's been long enough that I don't recall the parent very well, a quick perusal of the file wrapper of the parent will bring me up to speed in no time at all.
I'm certainly not saying that having a discussion over current continuation practice is not a worthwhile exercise. Given the current backlog problems, it's probably something that should be looked at. But really, the debate should be limited to facts, and the tired old 'wearing down examiners' argument is pure fiction (to my experience).
Yet, like a bad penny, this argument has been used again and again in scholarly papers as evidence of the need for continuation reform. (It is somewhat reminiscent of that 97% allowance rate figure that Quillen and Webster came up with in 2002; the figure has been thoroughly debunked [even by the original authors!] yet the figure keeps rearing its head as evidence of the abuse of continuations.)
That particular fiction is especially problematic, though, and I would say detracts from the productiveness of the debate, because to those who have no real knowledge of how the count system works, it sounds very convincing.
But it's simply not true.