As an initial disclaimer, Irwin IP LLP is privileged to be lead counsel for LKQ Corporation and Keystone Automotive Industries, Inc. (collectively, “LKQ”) in several design patent validity disputes, including this case against GM Global Technology Operations and by extension General Motors Co. (collectively, “GM”). LKQ neither requested nor paid for preparation of this article, however, and the views expressed herein are those of the authors alone.
Last month, the Court of Appeals for the Federal Circuit (“CAFC”), held an en banc oral argument for LKQ’s appeal of the Inter Partes review decision by the Patent Trial and Appeal Board (“PTAB”) which held that U.S. Design Patent No. D797,625 (“the ’625 Patent”) was not unpatentable as obvious. During the US Government’s argument, Judge Chen broached a central issue of the appeal when he asked whether it was “true that only about one to two percent of all design patent applications get a prior art rejection during examination?” The Government indicated that their rate is “a little higher,” but admitted that only “about 4%” of design patent applications receive a prior art rejection. This low rejection rate for design patents has received the attention of design patent scholars. See Sarah Burstein, Is Design Patent Obviousness Too Lax?, 33 BERKELEY TECH. L. J. at 608-610, 624 (2018) (positing that design patent validity precedent makes it nearly impossible for the United States Patent and Trademark Office (“USPTO”) to reject design patent claims); Dennis Crouch, A Trademark Justification for Design Patent Rights, 24 HARV. J. L. TECH. at 18-19 (2010) (citing the “vastly different” rejection rate between design and utility patents as evidence that the USPTO has abdicated its role in examination for design patents). For reference, by some estimates, the amount of total rejections for utility patents under prior art (35 U.S.C. §§ 102, 103) is as high as 68%.1 An important component of this explanation is that design patents have not received the flexible and expansive obviousness inquiry mandated by the US Supreme Court’s KSR decision.
In an effort to explain the low rejection rate, GM referred to an amicus brief submitted by Hyundai Motor Company and Kia Corporation (“Hyundai-Kia”), arguing that “often these, these design patents, because it’s a singular claim, you will see, for example, a shoe will be patented once and then you’ll see, you know, more patents off that similar initial concept. So, there’ll be a lot of patents that come out of the same invention because you can’t have, you know, 25 claims.” GM also argued that “the narrowness of design patents is another reason why we see a difference in the allowance rate as compared to utility.” Neither Hyundai-Kia nor GM cited any support for their theories. This article shows that neither argument adequately explains the alarmingly low rate at which design patents receive prior art rejections.
Regarding the argument that multiple design patents are filed on the same invention, there is no explanation as to why that would drive down the rejection rate. Logically, if there are groups of design patents for a particular invention (a “design patent family”) one would expect that prior art applicable to one patent in the family would be applicable to all of the patents in the family. And, one would expect that the number of design patent families that receive a prior art rejection as compared to the total number of design patent families would mirror the rate individual design patents receive a prior art rejection as compared to the total number of individual design patents. As such, the fact that there may be design patent families should not impact the overall rejection rate. Further, dissecting a whole design into many component elements (“design patents on fragments”) would likely seem to create a higher probability that any of these elements would have been known in the art to an ordinary designer. And, as such design patents on fragments would logically seem to drive up the rejection rate.
Regardless, it is appropriate to stress test the effect “duplicate design patents” could have in skewing the prior art rejection rate. By randomly selecting tranches of 50 sequential design patents granted between 2013 and 2023, we can create a statistically significant subset of design patents (that demonstrates the same 2-4% prior art rejection rate overall) to identify the percentage of design patents that were, as GM and Hyundai-Kia have argued, multiple, narrowed claims of the same invention. By removing these patents from the analysis altogether, we can determine an adjusted rejection rate for just individual design patents. For example, assuming no patent applications that received a prior art rejection were issued, if we were to hypothetically apply a 4% rejection rate to our sample of 500 patents, we can determine that about 21 patents were rejected to result in 500 allowed patents. If within those 500 patents, there are 50 duplicate design patents, that would leave 450 individual design patents. Dividing the rejected patents by just the individual design patents leads to an adjusted rejection rate. In our example, the adjusted rejection rate (21/450) is 4.67%. In performing our analysis, to be as generous as possible to GM and Hyundai-Kia’s argument, we also calculated an adjusted percentage by including similar designs issued to the same entity.
Applying this methodology, only 8.6% of design patents are duplicate patents in the way that GM and Hyundai-Kia argue. Removing these from the randomized subset, 91.4% of the design patents remain. Applying the Government’s 4% prior art rejection rate against the sample that excludes the duplicate patents, we arrive at an adjusted prior art rejection rate of 4.37% (still “about 4%”). If we expand the definition of duplicate design patents beyond GM and Hyundai-Kia’s definition in the above-mentioned manner, there are a total of 15.2% duplicate design patents. Once these are excluded and we are left with 84.8% of the randomized subset, applying the Government’s 4% prior art rejection rate results in an adjusted prior art rejection rate of 4.7%. At bottom, the exclusion of duplicate design patents in the way that is most generous to GM and Hyundai-Kia’s argument does not raise the prior art rejection rate of design patents by even one percentage point. The existence of duplicate design patents fails to explain the alarmingly low rate at which design patents receive prior art rejections.
Although not specifically argued by GM or Hyundai-Kia, the scholarly community should consider whether a claim-by-claim analysis of utility patents would result in a prior art rejection rate of about 4% for individual patent claims. That is beyond the scope of this article, but based on the readily available data we hypothesize that such a prior art rejection rate would still be “vastly different.”
Finally, GM’s appeal to design patent “narrowness” does not provide design patents with any kind of conceptual armor. The adage that “design patents have no scope” due to their visual claim style, has been practically rejected by modern precedent. See, e.g., Crocs, Inc. v. Int’l Trade Comm’n, 598 F.3d 1294, 1304-06 (Fed. Cir. 2010) (finding that several products with noticeable differences read on a single design patent, effectively expanding its scope beyond the literal claim depicted). Many courts have found that design patents have expansive scope in terms of infringement and damages.
The CAFC’s attention to the issue of the alarmingly low design patent prior art rejection rate at oral argument was well placed. As shown above, the disparity of this rate with that of utility has not been reconciled by any explanation or argument and is probably indicative of an inability or acquiescence to rigorous examination on the part of the USPTO.
- https://projectpq.ai/patent-rejections-103 ↩︎