Thursday, August 02, 2007

More on patent grant rate; the USPTO is NOT a rubber stamp

Further to the IPBiz post Do the published applications of 2001 tell us about patent grant rate?, someone directed LBE to a manuscript on SSRN entitled Is the Patent Office a Rubber Stamp? by Lemley and Sampat.

The patent grant rate numbers disclosed therein are more in accord with those of Robert Clarke in JPTOS in 2003 than with those of Quillen and Webster, or with those of John Thomas quoting Crouch/Patently-O in 17 Fordham Intell. Prop. Media & Ent. L. J. 875 (2007)).

Nevertheless, the manuscript has a puzzling sentence at page 25: Nonetheless, it is notable that quite a few patents -- significantly more than half of those issued--issued without any amendment, as Table 11 shows. IPBiz suspects that the idea that more than half allowed patents issue WITHOUT any amendment might be questioned. [There are other puzzling statements in the manuscript, but that's another story.]

Of the prior issue of whether the 100 published applications relied upon by Thomas are representative, IPBiz suspects now that they are not representative of a random set of applications and that the 90%+ grant rate suggested by this set of 100 is not extrapolatable to all applications.

On the matter of a "rubber stamp" -->


Notably, review by BPAI judges, who are familiar with the relevant law and frequently have some skill in the relevant science, is far from a rubber stamp for examiner denials. To the contrary, in fiscal year 2004, the BPAI affirmed examiner denials in only [p. 317] 37.1% of cases. 251 Moreover, even those denials that were actually reviewed by the BPAI represent only a very select subset of all patent denials. In fiscal year 2003, examiners from most technology classes reopened the case, or simply allowed the application, more than 50% of the time after an appeal brief to the BPAI had been filed. 252

n251 See United States Patent & Trademark Office, Fiscal Year 2004, offices/dcom/bpai/docs/receipts/fy2004.htm (last visited Oct. 22, 2006) (showing that 37.1% of denials were affirmed by the BPAI).

n252 In total, 59.4% of appeals resulted in either allowance or reopening of the case. Examiner Actions after Submission of Appeal Brief (on file with authors). In Part IV, we discuss trends in BPAI reversals and appeals to the BPAI as a function of changing standards of review by the Federal Circuit.


Matthew Sag and Kurt Rohde in Patent Reform and Differential Impact, 8 Minn. J.L. Sci. & Tech. 1 (2007) state

The FTC reports that the Patent Office's approval rate might be as high as 98%. 80 This is compared with a 67% approval rate in Europe and 64% in Japan. 81 The Patent Office argues that the true figure is more like a 75% approval rate or even a 60% rate. 82 However, other studies that account for the effect of continuations, continuations-in-part, and divisional applications estimate the Patent Office's approval rate to be much higher. 83 Whatever the reasons for the Patent Office's high rate of allowance might be, the fact that persistent applicants are almost always successful does not indicate a high threshold of quality control.

The important policy question that flows from the imperfect nature of patent examination is whether it is worth spending money to make patent examination any better. Patent lawyers and a number of academics have called for better funding for the Patent Office to improve patent examination. 84 In contrast, somewhat like Dr. Strangelove, 85 a few commentators have suggested giving up worrying about the issuance of invalid and uncertain patents and learning to love a system of quick-and-dirty examination at the Patent Office - the theory being that poor quality patents will either be ignored by the market or dealt with through litigation. 86

Footnote 80 is to at page 217. One notes that page 217 of the pdf file contains footnote 39 which attributes to Quillen/Webster the idea of a 95% grant rate in the year 2000. One notes that the same page presents footnote 42 which cites to Clarke’s work finding a rate of 74-75%, but Sag and Rohde did not include that fact within footnote 80.

n83. C. Quillen and O. Webster, Continuing Patent Applications and Performance of the U.S. Patent Office, 11 Fed. Cir. B.J. 1, 5 (2001) (finding that the Grant Rate for the PTO in fiscal years 1993-1998, corrected for continuing applications, ranges from 80% to 97%). [IPBiz: hmmm, ranges from x to y is not a synonym for is in the range between x and y.]

PDF of version of 17 Fordham Intell. Prop. Media & Ent. L. J. 875 (2007), which is generally about KSR v. Teleflex (notwithstanding Professor Thomas' hyperbole on patent grant rate).


Post a Comment

<< Home