Ten Years of Inter Partes Reexamination Data

Guest Post by Eric Rogers. Eric is a recent graduate of Chicago Kent's JD program and also holds a PhD in molecular genetics. Eric has conducted an extensive analysis of inter partes patent reexaminations that will soon be published in the Santa Clara Computer & High Technology Law. I asked Eric to provide an overview of his results from the study. Read the full article here: http://ssrn.com/abstract=2121917.

+ + + + +

The Inter Partes patent reexamination (IP reexam) playing field is tilted in favor of rejecting patent claims, as shown by comparing third party requestor appeals to patent owner appeals. On appeal, third party requestors are more likely to have a proposed ground of rejection affirmed and/or to add a new ground of rejection than patent owners are likely to have an adopted ground of rejection reversed. In addition, the vast majority of examiner rejections in IP reexams are upheld upon appeal. Of course, readers should be aware of selection bias in bringing an appeal.

An empirical analysis was performed of the first ten years of decisions by the Board Patent Appeals and Interferences (BPAI) regarding appeals of IP reexams. The analysis of 101 cases focused on answering three broad questions: 1) How accurate are the specialist patent examiners of the Central Reexamination Unit (CRU) of the USPTO? 2) Do patent owners or third party requestors fare better in appeals of decisions in IP reexam proceedings? 3) Which types of appeals are more or less likely to be successful?


The average IP reexam appeal involved 7.54 grounds of rejection, and the average completed IP reexam proceeding containing an appeal lasted 4 years and 10 months, as measured from filing the IP reexam request to the issuance of the reexam certificate. About half of all IP reexams were patent owner initiated appeals, and the remainder was divided equally among those initiated by the third party requestor and cross-appeals filed by both the patent owner and the third party.

Examiner Accuracy/Error Rate

The examiners' determinations were fully affirmed 76-78% of the time. This is slightly higher than the affirmed percentage for examiners in general, which over the same time period was around 61-76%. This suggests the CRU specialist examiners might be slightly less likely to be reversed by the BPAI.

Appealing Party Differences

Third party requestors tended to be more successful than patent owners at winning appeals. The reversal percentage observed differed from 14-27% for patent owners and 30-36% for third party requestors. Third party requestors' appeals received the most reversals at 30% to 33% while patent owner appeals received reversals in only 14% to 21% of their appeals. Similarly, third party requestors' appeals received the most reversals of grounds of rejection at 36% while patent owner appeals received reversals of grounds of rejection in only 22 to 27% of their appeals.

Regardless of third party participation, the IP reexam proceeding favors third party requestors. This is shown by third party requestors' statistically significant difference in successful appeals of about 14 percentage-points in maintaining patent claim rejections and adding new grounds of rejection as compared to patent owners' tendency to successfully have patent claim rejections reversed on appeal. Based on a comparison to the general ex parte appeal reversal rate, this bias is probably caused by disadvantages to patent owners rather than any special advantage conferred to third party requestors.

Type of Appeal Differences

Appeals solely by third party requestors of unadopted grounds of rejection were the most successful type of appeal, with a reversal rate of 33%. Appeals by patent owners of adopted grounds of rejection were the least successful, with a reversal rate of 14%.


These historical observations can be used to predict the success rates for appeals of Patent Trial and Appeal Board decisions in Inter Partes Review (IPR) proceedings, which will be implemented in September 2012. Appeals of IPR decisions by third party requestors are predicted to have a better success rate, by about 10-15 percentage points, at maintaining patent claim rejections and adding new grounds of rejection than appeals by patent owners will have at receiving reversal of patent claim rejections. Patent owners will probably have a reversal percentage closer to 15-20% while third party will probably have a reversal percentage closer to 30-35%.

8 thoughts on “Ten Years of Inter Partes Reexamination Data

  1. bad, BRI is a big problem, but as you noted, the estoppel provisions prevent anyone from even considering an IPR unless they have great new art.

    In contrast, tossing new art at the patent in an ex parte reexamination is a low cost way of piling huge costs on the patent owner, disrupting his license program, harassing him with BRI to thereby strip him of past damages, etc. Even if the reexam results in a victory for the patent owner, the damage imposed on him through attorney fees, delay and harassment is worth the few shekels it costs to file the necessary papers.

  2. I’ll go ahead and hedge that. Sheesh and Coe bring up great points about litigants only heading to reexam when they have great prior art that was not presented in the original prosecution.

    The appeals differential between reexam and regular examination shows a bit of “defending our own and preventing appeals.”

  3. The author’s characterization of the IP reexam process being “tilted” toward the 3d pty requestor is, I believe, based on an incorrect evaluation of the process. Given the number of companies, universities, independent research facilities, and individuals active in any given technology area, it doesn’t seem surprising that “good” prior art exists against virtually any patent. The trick, of course, is finding it.

    In my experience, IP reexams brought by 3d pty requestors are more successful because more effort (sometimes a LOT more effort) is put into finding prior art. Given this effort, it should not be surprising to find 3d party requestors are more successful in IP reexams. This is not due to any bias or “tilt” in the process, but rather the effort put into identifying prior art. To use a sports metaphor (my apologies), a team may win a disproportionate number of their competitions NOT because the referees are biased or tilted in their favor, but because they simply have better players.

    Based on the “self-selecting” phenomenon of 3d pty requestor initiated IP reexam proceedings, the reported data, outside of their pure descriptive value, do not seem to support any conclusions as to bias.

  4. What’s all this talk about tilted playing fields and disadvantages in IPR process to patent owners? There is an enormous selection bias in the patents that are put into IPR. TPR’s will not risk the estoppel unless they have a great case.

Comments are closed.