Quality versus quantity.

Because of the limitations of patent counts, some researchers have taken to use patent quality-based indicators. But assessing patent quality is far easier said than done.

For one thing, the concept of patent quality’ will vary according to whom you talk to. For patent attorneys and engineers, a high-quality patent can be a well-written patent, whose content is clearly described, or a patent protecting a major invention rather than an incremental step or technology. Legal scholars tend to interpret quality as the ability of a patent to stand scrutiny in a court situation without being invalidated while economists regard a good patent as one that rewards and incentivizes innovation while enabling diffusion and further technological development.

The issue of finding indicators in the indicators – or how to interpret patents.

Then there is the problem of finding patent indicators that effectively capture both the technological and the economic value of innovations.

Patents and citations.

Some maintain that the number of times subsequent patents cite a patent (these are referred to as ‘forward citations) is a good indicator of quality and even market value, stock price, or the value of innovation. However, the absolute number of citations is not, by itself, a clear indicator of value. Indeed, patents in different technologies tend to be cited at different rates, and older patent tend to have more citations than patents that were issued more recently which highlights the problem of citation lag – the time elapsed between the publication of the patent application and the first forward citation it receives. Such lag complicates counting and affects the timeliness of any analysis.

Patents and claims.

Given the aforementioned limitations of citations, others have sought to employ other metrics such as patent claims on the presumption that patents with more claims are expensive to draft and prosecute, involve higher filing fees but offer greater operational freedom for the patentee. But claim counts are irrelevant if, for example, the claims do not cover a commercially viable product or a product that could become commercially viable.

Patents and litigation.

Others have proposed alternative indicators including patent infringement litigation. Studies have shown that during a patent infringement action, the company initiating the legal action experiences a positive stock return while the defendant firm experiences a negative stock return and that patent litigation is a more direct indicator of market growth than other patent activities.

When Ericsson sued Samsung for infringement of patents involving technology for clear voice transmission, network efficiency, and touchscreen functions in 2012, won the lawsuit and settled with Samsung, Ericsson’s stock price gained 5.01% while the stock prices of Samsung fell by 1.2%.

But given that only a small percentage of patents are actually litigated (one study noted that there were 4,537, 5819, and 6114 patent infringement suits filed in the US in 2016, 2015 and 2013 respectively, while USPTO figures show 9,226,437, 8,925,112, and 8,341,762 were granted during those years – meaning that statistically, less than 0.07% of issued patents are litigated), patent litigation as an indicator is statistically questionable at best.

Market mechanisms.

Others have proposed using ‘market mechanisms’ to determine patent value, for instance, by looking at transaction activities in the markets for intellectual property transfer and seeing how much a patent might trades or assessing the impact of the loss of a patent (through invalidation or expiration) to a company’s stock price or revenue stream.

However, in practice, attempting to use a market mechanism to assess patent value is difficult given the paucity of reliable IP exchanges and the fact that data regarding IP transactions is not readily available in many cases, for example in dispute resolution cases where the costs of settlement, which may involve a transfer of patents, is not revealed.

When Lipitor went off patent at the end of 2011, revenues dropped from nearly US$9.6 billion in 2011 to $3.95 billion in 2012. From this, we can roughly assess the value of the Lipitor patent yet such data is rarely available.

Using patents: more limitations.

It is clear that there is interest in how patents can impact the value of a firm, whether by contributions or additional protections to its IP asset base, direct impact on its revenues or stock price or a combination thereof. But despite their apparent attractiveness, one must remember that patents have several inherent limitations.

First, there is the time it takes to get a patent (a process referred to as ‘patent prosecution’) which makes patent protection unsuitable for inventions in fast-moving areas though some patent offices have introduced accelerated patent examinations in certain technological areas. The Intellectual Property Office of Singapore, for example, recently granted a patent to Alibaba in artificial intelligence (for a modified metric to improve the accuracy of object detection in areas such as image and text recognition) in just three months, compared to an average of two to four years.

Second, one must disclose technical details in a patent and patents have limited lifespans, making patents unsuitable for protecting things like long-lasting proprietary know-how. Consider Gorilla glass whose roots trace back to the 1960s; Corning managed to keep its formulation a secret allowing them to benefit when applications for Gorilla glass materialized decades later (e.g. with smartphones). Had Corning gotten a patent several decades earlier, the Gorilla glass formula would now be in the public domain and Corning’s rivals would be able to reap the rewards of its development work.

Third, there are some things one cannot patent. For example, one cannot patent algorithms and, in some jurisdictions, one cannot patent medical procedures.

Fourth, patents are sometimes combined with other patents to form patent pools. Thus, assessing the value of an individual patent in such situations is nearly impossible.

Fifth, as patents are invention focused, they do not offer much insight into non-invention innovations. For example, an innovative new music form would not enjoy patent protection (though it may be possible to patent processes or inventions associated, for instance, with the creation of the new music form).

Back to commercialization.

Yet the most serious limitation of using patents as indicators alludes back to an earlier point, that other associated activities such as marketing and product positioning, play an important role in the success of an innovative product and these are things that patents cannot, by their very nature, capture. But other IP rights can. Thus, there is somehow room for a better IP-related indicator – to be continued…

Further reading:

• Mr. Gary N. Keller, “Guide On Intellectual Property (IP) Commercialization”, WIPO, November 2015.
• Phillipp G. Sandner, Joern Block, “The Market Value of R&D, Patents and Trademarks”, Research Policy, January 2011.
• Mafini Dosso, Antonio Vezzani,”Firm market valuation and intellectual property assets”, JRC Working Paper on Corporate R&D and Innovation – No. 07/2017.
• Pol Steinbusch, Daan Vodegel, “Influence of the Number of Patents on Stock Performance: A quantitative analysis focusing on IBM”, 2015.
• Mariagrazia Squicciarini, Hélène Dernis and Chiara Criscuolo, “Measuring Patent Quality: Indicators of Technological and Economic Value”, OECD Science, Technology and Industry Working Papers, 2013/03, OECD Publishing.
• Bronwyn H. Hall, Adam B. Jaffe, Manuel Trajtenberg, “Market Value and Patent Citations: A First Look” (June 2000). NBER Working Paper No. w7741.