How innovative is Canada? How do our regions differ in their rates of innovation productivity? How do we compare with our international peers in innovation output? These questions are all more difficult to answer than they need to be because the federal government unnecessarily restricts access to data on Canadian innovation. This lack of free and open access hampers research on Canadian innovation, impedes informed policy-making and ultimately slows Canadian research and development, holding back economic growth.
The federal government has repeatedly expressed its interest in promoting Canadian innovation, diversifying the national economy and engaging in evidence-based policy-making. However, the policies on how Canadian innovation data are made available — or how they are not made available — obstruct research into Canadian innovation and leave us near the back of the pack in innovation data policy within the G8.
By innovation data policy I mean the set of policies that dictate how governments license and distribute data on the innovation system. In Canada, the most obvious examples of innovation-related data are generated by the Canadian Intellectual Property Office (CIPO): records of patents, copyright and industrial design. Among many other important results, the datasets that CIPO generates can tell us who invented what and where they are employed. These data provide valuable insight into the innovation system and can contribute to better policy-making.
However, despite the potential value of these data, they remain difficult to access, to improve upon and to share. Notwithstanding the trend toward open access to public data, CIPO datasets remain walled off and accessible primarily to those willing to pay fees on the order of thousands or tens of thousands of dollars. Furthermore, although CIPO does offer some exemptions that allow researchers to request data for research purposes, it requires them to sign non-disclosure agreements and forbids them from distributing the data they receive.
Limiting data access to those who can pay for it prevents entrepreneurs from repurposing the data to start new ventures. Similarly, restrictions on researchers keep them from improving upon the data and sharing those improvements with other researchers. Sound research is iterative: researchers need to build upon the work of their predecessors.
In contrast to Canadian practice, the United States Patent and Trademark Office (USPTO) freely shares well-structured and machine-readable datasets, with no constraints on use or sharing. The USPTO has also seen the value in sharing previously internal datasets that chronicle its own procedures, with the hope that doing so will result in insights that will help the office improve efficiency. While the European Patent Office is somewhat less inclined toward open access than the USPTO, it too provides greater access to innovation data than its Canadian peer. Canada’s hanging back in this regard has led to a pronounced American-European focus in the body of empirical innovation research.
Furthermore, the current data-sharing policy at CIPO is in contravention of Canada’s declared belief in the value of open data. When it adopted the G8 Open Data Charter in 2013, Canada committed itself to establishing the foundations for a transition to an open data regime by the end of 2015. One of the core principles of the Open Data Charter is to release “data for innovation” that will “empower future generations of innovators.”
CIPO policies are also at odds with the basic tenets of intellectual property theory. One of the primary reasons to have an intellectual property protection system is to provide an incentive for innovators to disclose their innovations, allowing future innovators to build upon them. By disallowing free access to bulk datasets, CIPO puts an unnecessary barrier between potential innovators and the ideas already disclosed. These are ideas that the public has in essence “paid for” by providing 20 years of monopoly rights to the patent owners. CIPO is depriving Canadians of the benefit of the intellectual property bargain.
Happily, there is a simple solution: open the data. CIPO should follow the lead of the USPTO by providing free and open access to all its bulk data, including full access to its back files as well as regular updates with newly generated data. Opening Canadian innovation data would require little work on the part of CIPO and would provide great advantages.
One of the few possible reasons to oppose sharing data arises from concern about the costs of doing so. After all, CIPO currently charges third parties for data and gets income this way. To determine how great this loss of revenue would be, I submitted access-to-information requests seeking information about CIPO’s data-selling revenues. Records show that in the last five years, CIPO has received an average of about $24,000 per year from selling patent data. During that time CIPO had an average of 15 patent data clients per year. Together with fees for other data products — including datasets on copyrights and industrial design — average revenue remains a relatively small $49,000, from an annual average of 31 clients. In the context of CIPO’s total annual revenues of over $150 million, the amount CIPO makes from selling innovation data products is insignificant.
The value society would derive from open innovation data would far exceed the lost data-fee revenues. If these data were open, they would be freely available to CIPO’s current data clients, but more importantly they would also be freely available to researchers, startup firms specializing in data provision and innovators themselves.
CIPO does already provide free access to some innovation data via its Web-based search engine, but this site does not allow for macro-scale data collection or analyses. Many of the most pressing questions about Canadian innovation require analyzing hundreds of thousands or even millions of patent documents, which the current user interface does not permit. Furthermore, any improvement or repurposing of Canadian innovation data requires access to complete datasets, not the handful of records one can retrieve manually from the CIPO website.
In summary, Canadian innovation data should be freely available. The costs of opening these datasets are minor and will be offset many times over by the social benefits of their increased usage. Opening these data accords with our government’s professed interest in empirically informed policy-making, our G8 Open Data Charter obligations and the most fundamental principles of the intellectual property system. Innovation, Science and Economic Development Canada — CIPO’s parent organization — should adopt a policy mandating open access to all of the innovation data that it generates.
Do you have something to say about the article you just read? Be part of the Policy Options discussion, and send in your own submission. Here is a link on how to do it. | Souhaitez-vous réagir à cet article ? Joignez-vous aux débats d’Options politiques et soumettez-nous votre texte en suivant ces directives.