We keep seeing data on what we call Canada’s Patent Puzzle. The prevalent Canadian narrative is that as a country, we struggle to compete in the global innovation economy. This conclusion is based on three frequently cited and well-known metrics that have shaped our industrial research and development (R&D) strategy and policy conversation for decades:
- expenditures on research and development,
- the number of patents granted, and
- multifactor productivity.
In this Brief, we look more closely at Canada’s performance in the numbers of patents granted. When blended with other metrics, patents are thought to be an effective measure of a country’s ability to convert knowledge into novel inventions that allow it to reap the commercial benefits of the newly protected intellectual property (IP). But Canada has fared poorly on this specific indicator: the Conference Board of Canada has given us a “D” grade, placing Canada 14th out of 16 peer countries in the number of triadic patents (Conference Board of Canada’s Patents Index, 2010). Triadic patents are defined as patents issued for the same patent family in three major jurisdictions: the US, Europe, and Japan.
Although there are drawbacks with using triadic patent counts in the assessment of innovation, they are generally considered the gold standard for IP-related metrics and continue to be used liberally.
But, by examining individual firm behaviour, comparing patenting practices of small and large firms, and the issuance of patents in Canada’s largest market (the US), we can demonstrate that the way we have looked at patents so far has been fundamentally flawed.
Triadic patents largely reflect the patenting of multinational firms with operations in all three sectors of the globe. We found a high degree of correlation between the number of large technology companies headquartered in a particular country (expressed in terms of business revenue as a percentage of the country’s gross domestic product [GDP]) and a country’s triadic patent ranking. Therefore, triadic patents are in part an indication of industry structure and scale—rather than innovativeness. Since this finding also points to a shortage of world-class Canadian companies that are capable of pursuing markets and IP protection around the world, it suggests a problem of commercialization and scale, and not a problem of R&D.
Canadian Patents in the US
Canada’s success in obtaining patent grants in the US has improved by 143% over the last ten years. The number of patents with one or more Canadian inventors climbed from 3,661 in 2005 to 8,903 patents in 2015, placing us eighth on a per GDP basis against competitor countries in 2015 and in terms of our growth rate over 10 years.
However, of the patents granted to Canadian inventors by the US Patent Office in 2016, 58% were assigned to companies domiciled in other countries. This is up from 45% in 2005. This means that Canada earns a return through commercialization for less than half of the patents granted in the US to Canadian inventors. Therefore, Canada’s critical issue is not an inability to turn invention into innovation. Our challenge is to ensure that Canada retains some of the economic and social benefits from our innovation activities.
Role of Leading Canadian R&D Firms
Of the top 50 R&D spenders in Canada in 2015, 17 were subsidiaries of foreign companies (including Ericsson, IBM, Cisco, and PMC Sierra.) The subsidiaries represented 32% of all patents granted in the US to the 50 leading Canadian R&D firms. However, 96% of the patents granted to the foreign subsidiaries were assigned to parent companies in another country.
Subsidiaries of foreign companies conducting R&D in Canada are also eligible for a scientific research and experimental development (SR&ED) tax credit equal to 15% of eligible expenditures (subject to stringent rules.) But if the benefits of Canadian R&D are transferred to other countries through patenting practices, are Canadian taxpayers subsidizing research whose long-term impact is felt elsewhere?
Patents as a Metric
Using granted patents as a measure of innovativeness is also exacerbated by the fact that patents are an input metric; they do not correlate well with results-oriented measures such as revenue. Patents also range significantly in the quality and nature of the underlying invention (i.e., process versus product IP) that must be accounted for in the analysis.
Although using patents as a metric for how well we are performing internationally may have worked in the industrial economy, it does not work in an internationalized knowledge economy. Patenting is an international, not a local activity; and the nuances of the process must be considered before the numbers are aggregated into a single indicator for the purpose of policy making. If we are serious about improving our ability to compete internationally using R&D as a base for international growth, then we need to ensure that we use appropriate metrics rooted in strong and valid assumptions. Otherwise, our efforts will most likely remain misdirected or ineffective.