Data

Counting and Discounting Patents – The USPTO Study on Patenting Activity in 5G

USPTO released a study on February 15, 2022 entitled “Patenting Activity among 5G Technology Developers” (the “Report”). According to the USPTO press release, the Report uniquely “examines overall patenting trends as well as trends in patent filings and value indicators in the four most-patented 5G-related technologies. The Report shows that six companies—Qualcomm, Ericsson, Huawei, LG, Nokia, and Samsung—consistently filed more 5G-related patent applications in the last decade. The report also examines a subset of applications filed in 2010-2021 in four critical 5G technologies: (1) management of local wireless resources; (2) multiple use of transmission path; (3) radio transmission systems; and (4) information error detection or error correction in transmission systems. In these four areas, Qualcomm and LG emerged as the most active, followed by Samsung, Huawei, Ericsson, and Nokia.”  The Report further concludes that it “supports the finding that the United States remains highly competitive in 5G innovation.”

This Report adds to the increasing corpus of data that supports a more objective understanding of the threats posed by China’s emergence as a major patent filer. The Report accomplishes this goal through consideration of multijurisdictional “triadic” patent families (US, European, Japan). By focusing on multijurisdictional families, it ignores filings where “the patent owner files disproportionately in its home jurisdiction to the exclusion of filings in major foreign markets”. This approach is consistent with the analysis of many others who have looked at quality issues in China’s domestic patenting.  Jonathan Putnam has noted that China’s relatively low level of overseas filing, coupled with its high level of domestic filing, suggests that there is a quality gap, that is likely attributable to domestic subsidization of applications. The USPTO report of January 2020 on “Trademarks and patents in China: The impact of non-market factors on filing trends and IP systems,” similarly showed that Chinese patent applicants are much less likely to seek foreign protection than are U.S. inventors and that there is a low commercialization level of Chinese patents.  In a recent article I co-authored, I also looked at “top-down directives and goals for patenting induced Chinese applicants to game the system, introducing distortions and inefficiency” that are manifested in China’s patent data, particularly in end of year patent filings.  Moreover, this planned and subsidized domestic environment for SEPS is also likely to be of continuing significance to Chinese inventors, with China’s recently released National Informatization Plan (2021-2025) calling for increasing the number of patents per 10,000 people in “new-generation information technology industry” from 2.7 patents in 2020 to 5.2 in 2025, or over 700,000 patents.  

Notwithstanding the significant concerns about China’s domestic patent quality, the Report’s failure to directly consider filings in the world’s largest patent office may be viewed as a deficiency.  In fact, the studies that USPTO cites for using triadic data date primarily to 2001, 2009 and 2011, which are well before China’s recent technological emergence (Report at fn. 13).   I believe that a preferable approach to ignoring the Chinese patent office would have been to take a discounted approach to Chinese patent filings, much as Jonathan Putnam did in his study (referenced above).   Had USPTO committed the time and resources to a thoughtful analysis of domestic Chinese patent filings, that data might also have been useful for courts, regulators and others who must consider the quality and value of Chinese patents.

Nonetheless, the Report does offer criteria for judging quality of a patent portfolio.  To ensure that quality is being appropriately measured, the Report considers such factors as market coverage (size of market where patents are filed), technical relevance (forward citation), radicalness (low level of citations in the patent), legal breadth (words used in shortest independent claim) and scope (numbers of distinct CPC classes).  On the “legal breadth” metric, Qualcomm is first, and Ericsson is second. Ericsson and Nokia generally score high on the “radicalness” measure, whereas Qualcomm and Samsung are at or near the top on the “technical relevance” scale. No strong differences emerge on the “scope” and “market coverage” metrics.  Although Huawei filed the most ETSI-declared filings in the study, it did not score among the highest in the above categories.  Moreover, ZTE’s filed very few triadic families and as a result the company appeared only selectively in the Report.  

I am not certain that “radicalness” criteria adopted by USPTO fully applies to patent-dense IT technology such as 5G.  The study that the USPTO relied upon for its “radicalness” approach examined patenting in tennis rackets, where two patents involving oversized rackets and stiffer frames were widely viewed as path-breaking innovations. The authors of that study looked at 581 tennis racket patents over a 30-year period.  This cohort is considerably smaller and less complex than the 106,00 ETSI-declared patents that read on 5G devices noted by the Report.  Moreover, I suspect that radical innovation may also be occurring outside of the seven companies studied in the Report, including in universities and research organizations.  “Radicalness” in 5G standardization is also worthy of further study.

1 reply »

  1. There was an interesting Footnote 14 in which Chinese patenting was considered, but apparently didn’t alter the results. Text pasted here: “With relatively few exceptions, the economic literature has, to date, generally defned “triadic” as consisting of these three markets. In the preparation of this report, USPTO economists additionally examined a data set limited to what might be termed “quadratic” patent families, including those with flings in the four markets that the most active declarers call home (the United States, EU, China, and South Korea). The results are not presented here because they difer little from those generated by the more widely embraced triadic approach. When using the quadratic approach, we still fnd that Qualcomm has the most families represented, followed by Samsung, LG, Huawei, Nokia, Ericsson, and ZTE. “

    Like

Leave a Reply to John Kolakowski Cancel reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s