Guest Post by Christopher Cotropia on Existing Deference in Patent Claim Interpretation

Christopher Cotropia is a Professor of Law and the Austin Owen Research Fellow at the University of Richmond School of Law.  Below he highlights a few important conclusions from an empirical study on patent claim construction that he recently completed.  The full study is available here.

Is Patent Claim Interpretation Review Deference or Correction Driven?

Christopher A. Cotropia

The Federal Circuit’s now fifteen-year practice of reviewing lower court claim interpretation de novo is being reconsidered en banc in Lighting Ballast Control LLC v. Philips Electronics North America Corporation. Although the question of whether review should be de novo can be answered under a number of criteria, in patent cases it is commonly viewed as a tradeoff between certainty and correctness.  

With regards to certainty, the current belief is that de novo review results in great uncertainty in patent cases because litigants cannot predict the controlling claim construction in a case until the Federal Circuit makes its independent determination on appeal.  Giving district courts greater deference would, in turn, allow certainty as to claim construction to come earlier in patent cases.

In contrast, appellate review is in place, at least in part, to correct mistakes made by lower courts and ensure the “correct” result is reached in a given case. The more deference provided to lower courts, the less opportunity the Federal Circuit has to correct (or incorrect, depending on your view of the Federal Circuit) the decision.  Standard of review presents the question of who—that is which tribunal—is more likely to reach the correct answer regarding claim interpretation and, given claim meaning’s central nature to the patent dispute, the ultimate question of who should win the case.   

While many aspects of these issues need to be explored to determine the optimum standard of review (how much is greater certainty worth?, who is more likely to reach the correct answer?, and what is the correct answer we want to achieve?), there is value in establishing a baseline on both issues. Assuming the Federal Circuit’s claim construction is correct, it follows that de novo review results in greater correctness, albeit at the expense of possible increased uncertainty.  But both of these assumptions rest on the belief that the Federal Circuit currently engages in de novo review, currently provides no deference in practice, and arrives at the “correct” result in appealed cases, at least more often than district courts do on first pass.

In a new paper, I try to establish this baseline by collecting and analyzing Federal Circuit claim construction reviews over the last three-plus years.

At first blush, the results show there is a high level of certainty in claim construction appeals. The rate of affirmance of district court claim interpretations has been increasing since 2010.  And, in the first quarter of 2013, the affirmance rate has reached over 75%.   This data confirms that observed by Jonas Anderson and Peter Menell in 2011 for an earlier period.   


Cotropia Table 1



However, a closer look as to which claim interpretations are affirmed and which are reversed shows little certainty at the Federal Circuit for particular types of cases.  Lower court decisions where the patentee ultimately wins are more likely to be subject to a claim construction reversal that, in turn, causes the alleged infringer to ultimately win (or at least gives them another shot on remand).

Cotropia Table 2


And these claim construction reversals are even more likely in appeals of cases involving electronic, information technology, or business method patents where the patentee won below. The opposite is true in cases involving patentee wins in the biological and chemical technologies—the Federal Circuit affirms these constructions, and results, more often. 


Cotropia Table 3



What appears to be certainty in all claim construction reviews is much more complex, with the Federal Circuit performing corrections in specific categories of cases.  And given that claim constructions are appealed in patentee wins and patentee loses at about the same rate, 46.67% to 45.94%, these findings about claim construction review are not likely driven by selection bias. Put simply, the Federal Circuit’s claim interpretation review is not driven solely, or perhaps at all, by deference, but instead is correction focused.

What does this data mean for Lighting Ballast Control?  A change in standard of review would have an impact—greater certainty, and less change on appeal, for particular types of cases.  This certainty would come at a cost—losing the current correction going on.  Answering whether this is truly a cost, specifically one that outweighs any certainty gained by changing the standard of review, is left for others, as is the meta-question of what the “correct” answer is in any patent claim construction case. 

This data also describes a Federal Circuit that is very different then commonly portrayed.  The Federal Circuit, when reviewing a district court’s claim interpretation, is not as patent-friendly as some would think.

A full version of the paper this post is based upon can be found here.

10 thoughts on “Guest Post by Christopher Cotropia on Existing Deference in Patent Claim Interpretation

  1. 10

    Actually, Judge Rader has justified his 101 jurisprudence, at least in part, on his view that certain subject matter should be patentable. He works backwards from this assumption seeking to justify it in the precedents.

  2. 9

    MM, indeed.  A lot of the law seems to come from judges first decide what the result should be based upon ethics or morality and then work backwards to justify that result in there precedent.  As you suggest, what seems to be going currently is that the Federal Circuit seems to be rewriting claims to correspond to what they think the invention is rather than what the claims actually say. If the claims are too broad, such that they read on something which they do not consider to be the invention, they narrow the claims. If the District Court  claim construction results in a finding of noninfringement where clearly the defendant is appropriating the invention, they broaden the construction. But both of these depend upon their assumptions about what the invention is and whether they think the defendant is infringing.


  3. 8

    link to

    That’s funny, just the other day I was hearing about how hiring in engineering was totally down and software hiring was totally up. Perhaps that was simply because there weren’t enough qualified, capable engineers to go around lol.

  4. 7

    To the authors’ credit, Ned, they do recognize that there are differences between the tech areas and the types of claims that are prosecuted and litigated. And they try to control for those differences. I’m not just convinced that it’s possible to eliminate all the other complicating factors and arrive at the “realist” conclusion that they seem to arrive at, i.e., that Federal Circuit judges are biased against tech/it/biz patentees (relative to chem/bio) patentees and manipulate claim construction accordingly. Maybe I am reading the quote (in my July 24 1:50 pm comment) incorrectly?

    I also wonder: in the cases where the district court’s claim construction is affirmed and the patentee wins, is there a technology-correlated difference in the frequency with which the Federal Circuit reverses or remands on other grounds related to enforceability or patentability?

    And: are there tech-correlated differences in the rates at which plaintiffs lose at the Federal Circuit on issues that are completely unrelated to the patent (e.g., procedural issues, standing)?

    As to the larger question noted by Prof. Cotropia, I don’t believe that changing the standard of review of claim construction will accomplish much. The problem with the law regarding claim construction right now is the Federal Circuit’s confusing tendency of late to ignore its best and clearest precedent (Phillips) and re-write an invalid claim to accord with some “invention” allegedly described in the summary or description of the specification. It won’t do anybody any good for the Federal Circuit to “defer” to a district court that is trying to wrestle with the Federal Circuit’s increasingly contradictory “rules” for interpeting claims.

  5. 5

    MM, we all know what is going on here and why patentee wins in business methods are likely suspect. In many cases, the patents are asserted by the so-called “trolls” who want to read their claims broadly and want to avoid the specification as much as possible.

    In contrast, the chem and bio fields are not dominated by trolls, but by folks trying to protect their monopolies in real products. This normally does not require aggressive claim construction.

  6. 4

    The issues regarding bias in the data I raised in my earlier comment are addressed by the authors in their full paper … although perhaps less convincingly in certain respects. It’s an interesting paper, though, and well-written. Here’s the passage I find a bit weak:

    the [difference in rates observed for winners and losers in IT/biz cases] could be explained as an exercise in deference if the errors are more egregious in appeals from patentee wins in electronic, information technology, and business method cases and much less egregious in appeals from patentee wins in biological and chemical technology cases. And this could supported by the fact that biological and chemical claims are traditional clearer and easier to construe as compared to electronic, information technology, and business method claims. But if this was the reason, it would hold true regardless of whether the patentee won or lose below. This is not the case, however—the results vary depending on who wins

    Given all the other factors built into the decisions to appeal or not appeal a particular claim construction (including the data in papers such as this one) the observed difference between claim construction reversal rates when the chem/bio “patentee wins” versus when the chem/bio “patentee loses” doesn’t appear to be a very substantial one. So this bias isn’t adequately addressed, in my opinion.

    The paper mentions that the Eastern District of Texas has a “high affirmance rate” of claim construction (84%) so the forum preferences of NPEs in the IT/biz area by itself appears not to explain the data.

    The authors are more clear in the paper about their conclusion than in the blurb above:

    The Federal Circuit appears to be reviewing [claim construction] with a certain “correct” result in mind—generally anti‐patentee focused mainly at electronic, information technology, and business methods patents with an opposite, pro‐patentee mindset for biological and chemical technologies.

    I don’t disagree with the statistics but I don’t buy this “difference in attitude” explanation for the data. There are just too many complicating facts (impossible to address, statistically) to justify this particular leap into the collective jurist mind. That’s not to say that the theory is wrong. I just find it unlikely and don’t see that it necessarily follows from the data.

    It’s interesting, though. Assuming the litigants are rational actors and assuming they are aware of and believe in the data, will this lead to more or less skewing down the road?

  7. 3

    As Hal Wegner has often noted, a serious concern with Fed. Cir. de novo claim interpretations is demonstrated by the number of split decisions. That is, decisions in which even the Circuit judges on the panel cannot agree among themselves on proper claim interpretation, and a change of one judge on the panel would have given an opposite result. At least one prior study demonstrated frequently displayed strong differences of opinon between certain Fed. Cir. judges as to how narrowly a claim should be interpreted with reference to its specification examples, or even applying 112 rejections for overclaiming relative to the specification in some cases, not just in chemical cases, as in Lizardtech, et al.

  8. 2

    Unfortunately (for you), nomenclature is not an issue here.

    Right, because the use of nomenclature has nothing to do with the certainty of claim construction. Man, you are on a ROLE today!

  9. 1

    these claim construction reversals are even more likely in appeals of cases involving electronic, information technology, or business method patents where the patentee won below. The opposite is true in cases involving patentee wins in the biological and chemical technologies—the Federal Circuit affirms these constructions, and results, more often.

    This isn’t unexpected if one recognizes that chem/bio practitioners typically tend to use established nomenclature when describing the structures of their new compositions and/or methods, and also tend to include definitions for key terms. In contrast, the biz and IT practioners especially seem to rely more on nomenclature created in the marketing department or created de novo specically for the application. The same practitioners seem to prefer to define key terms in prosecution (or better yet, in court) (e.g., that “one skilled in the art would understand the definition to include …”) rather than provide express definitions with defined boundaries.

    I say “seem” because this is my impression from reviewing Federal Circuit decisions over the time period of this study (and going back further). It would be interesting to learn of an empirical study that finds differently.

    Another possibility, of course, is simply that the typical patentee in a chem/bio case is less inclined to be trolling compared to the elec/it/biz sphere and therefore less likely to be appealing from a “banana republic” district court.

Comments are closed.