Ownership of Intellectual Property in Canadian Post-Secondary Institutions

A clear intention behind IP ownership is a prerequisite for effective tech transfer and economic benefit from commercialization of publicly funded research

It’s not controversial at this point that Canada is terrible at commercializing its research outputs. From the Jenkins Report, to the Bouchard Report, to discussions of Canada’s hilariously misnamed Productivity Paradox, it is well established that Canada outputs a high volume of excellent research, and mostly fails to take the process any further.

Before I started writing through CanInnovate, I wrote a short article on IP ownership policies among Canadian post-secondary institutions. The upshot of that piece was that institutional policies that favor inventors correlates with higher degrees of entrepreneurship, albeit in a fairly limited sample size given that Canada has so few universities that crack the global rankings for entrepreneurship activity. What was less clear is whether or not this relationship is causal.

In this article, I explore this correlation in greater detail, focusing on the role of the intention behind technology ownership as the causal factor relating IP policies and entrepreneurial outcomes. In this framework, the correlation between inventor ownership and entrepreneurship is an emergent property that can be exploited to improve tech transfer (and eventually economic) outcomes irrespective of the IP ownership policy used. I finish the article with practical suggestions for institutional policy development to address systemic deficiencies in Canadian tech transfer practices.

Correlation is not Causation

The institutional exceptions to the rule of poor tech transfer performance are interesting. Among entrepreneurial Canadian universities, Waterloo is in a league of its own with respect to entrepreneurship and translation. Unlike most Canadian institutions, their IP policy is entirely inventor-owned: the university only gets involved if explicitly requested by inventors, who are otherwise free to patent and commercialize (or not) their inventions, at their own expense. McGill, with a jointly-owned IP policy that puts much of the IP-related decision making in the hands of inventors, is not far behind on a per capita basis, followed by Queen’s and University of Toronto, with inventor-owned and jointly-owned IP, respectively.

Far down the list of international rankings, we find only four more Canadian institutions, all with fully University-owned IP policies, performing at about a third of the level of the first four in terms of number of spinouts per capita and average capital raised. No other Canadian institutions make the list, but a review of publicly available IP policy documents will quickly make clear that institutional ownership of IP is more common than inventor ownership among the big names.

In a vacuum, it is tempting to conclude that university ownership of IP is the problem, a trap I fell into early in my exploration of the process. This conclusion does not hold up. It is clear from a review of American tech transfer policies, among other examples, that institutional ownership of IP does not by itself prevent entrepreneurship. As I wrote recently, the Bayh-Dole Act requires institutional ownership of IP, and American research commercialization is practically formulaic.

The factor that connects institutional policies and commercialization outcomes is not ownership, but intention.

Intention is the Seed of Causation

Consider the case of inventor-owned IP. An individual researcher with a patentable idea is free to seek protection for that idea, a process that can easily cost tens of thousand of dollars. A patent filed without intent to commercialize is a waste of time and money, and so the only circumstance in which they proceed is if they intend to use it. In cases like Waterloo, where there is a well-established route by which the researcher can request assistance from the university to protect the IP in question (see section 7.C in their IP policy), it is still the case that this option would only be exercised if there was a clear plan for what follows. Whether or not that plan is successful in the long run, there is a high likelihood that any patent filed at Waterloo will be the basis for at least an attempt at commercialization.

Institutional ownership generally correlates with worse economic outcomes in Canada, but not in the United States. The difference lies in the Bayh-Dole Act, discussed in detail previously, that essentially imposes intent to commercialize patents on academic institutions at the level of federal policy, on penalty of giving the government the right “march in” and do it anyway. By ensuring that intent to commercialize is central to IP policies in publicly funded research institutions, the US enjoys a steady stream of commercialization irrespective of local variations in institutional IP policy that has contributed to establishment of a large number of thriving entrepreneurship hubs all over the country.

In Canada, funding agencies are hands-off with respect to the intellectual property arising from funded research, leaving it to the institutional tech transfer offices (TTOs) to negotiate with research partners within the scope of their IP policies, which were in turn developed without any overarching intention for their target impact. As a direct result of this lack of leadership, while most Canadian universities have publicly available IP policy documents, few, if any, have a clear mission and vision for the outcome of applying it.

At a macro level, this public policy failure to provide guidance in institutional policy development has resulted in development of a patchwork of disconnected institutional IP policies, very few of which come even close to creating self-sustaining TTOs, let alone local ecosystems. At a micro level, TTOs are often (through no fault of their own) left without clear guidance on the intended outcome of their activities, and a severe lack of funding to execute in any case. The net result is that most Canadian TTOs patent technology portfolios without a clear intention for what comes next, and as a result are forced to favor licensing arrangements that allow them to quickly recover costs on filed patents and meet short-term performance metrics that are uncorrelated to long-term economic benefit to Canada.

Intentional Tech Transfer Policy Development

The open conversation around creation of a capstone agency for management of Canadian research funding provides a possible opportunity for top-down reform to incentivize universities to address these challenges, but given that we are heading into an election I am not holding my breath. While I believe that top-down reform led by sound public policy development is required to address these problems in the long term, bottom-up, institution-led policy development will play a key role as well.

As a prerequisite to everything that follows, every Canadian university with a TTO should have a publicly available answer to the following question:

What benefit does the university seek to derive from its tech transfer and entrepreneurship activities?

In my experience, very few Canadian TTOs have an answer for this question, and fewer still are empowered to act on it. How can they, given that neither the institutional central administration nor the funding bodies provide any practical guidance? While I would prefer to see this dictated at the level of federal research funding policy, there is a role for the individual institutional TTOs and the central administrations that control them to take the necessary steps in the continued absence of top-down leadership.

The answer to the question should be “to be the epicenter of a self-sustaining entrepreneurial ecosystem based on technologies derived from institutional research.”

Note carefully that this answer makes no reference to licensing revenues or the direct value of licensing deals to the university. This is a challenging prospect for TTOs that are structured as separate cost centers with their own P&L, and requires that universities that seek to support entrepreneurship change how they structure their TTOs from a budgeting perspective.

Waterloo makes almost no money from IP licensing revenues (barely $55,000 in 2017), not even breaking the top ten for licensing income, yet the tech ecosystem build up around it stands at a valuation of more than 3% of Canada’s entire GDP (and 8% of Ontario’s), delivering $71B to the local economy in 2023, growing more than 50% in the last two years. The direct cash value of this activity to the institution is low, but the publicity, good will, and stellar international reputation of Waterloo as a hub for entrepreneurship offsets any lost licensing revenues by orders of magnitude and are enormously valuable to the country. Given that it is a taxpayer-funded institution, this is precisely as it should be for all publicly Canadian research institutions.

In comparison, in spite of the focus on licensing revenues, Canadian universities collectively made only about $75M in revenues in 2017, about 1.3% of cost of the research conducted in that same period. Cost recovery and licensing revenues may make for easy P&L reports, but they are of effectively no value to Canada, or arguably even to the individual institutions, in the long run.

Seeding an Ecosystem

As with so many issues in Canadian innovation, a lack of top-down leadership from funding bodies on IP management best practices continues to be a major roadblock to achieving necessary system change. University administration is at the end of the day driven by funding, and without tying long-term outcomes to funding, there is no impetus to change. Ideally, funding for management of the IP resulting from publicly funded research would be part of any academic grant that could possibly generate IP. However, even in the continued absence of top-down leadership on this issue, there is an opportunity for Canada’s universities to learn from the example of Waterloo and to begin seeding their own entrepreneurial ecosystems.

This process begins by clearly setting intentions for the outcomes of tech transfer in the form of a public mission and vision that is embodied by the approach taken to commercializing the outputs of research. This point is addressed above. The suggestion that I have for an overarching mission and vision for Canadian tech transfer is not the only viable path, but any viable path must begin with intention.

Second, changing the way that TTOs are managed within the context of a larger university budget will be required. Investment in low-TRL IP by nature will almost always involve running a deficit for a few years, and any system that seeks to support commercialization of IP must be tolerant to that fact by design if it is to have an impact. As we see research funding reform unfold I hope to see the costs of IP management explicitly added to research grant budgets; absent that, universities should budget a portion of the overheads charged on every grant for IP management and accept that any return on that investment will take years to materialize.

Finally, there is a need to standardize the approach taken to tech transfer to favor inventor-led small businesses over large multinationals, and to build standardized and transparent tools to enable it. Creation of these is an ongoing project at CanInnovate.

Given that the time delay between beginning this process and actual establishment will require at least a decade of operating in the red as early-generation startups build to the point that they can support the next generation of companies, these policy reforms will need to be combined with internal metrics of success that are tolerant to the time delays involved in doing so and which are directly connected to long-term economic growth, both at the level of the local ecosystem and nationally. These metrics, like those I have suggested in other related contexts, should focus on control over IP assets in the long run and favor licensing to inventor-led startups and small businesses over large ones, and should focus on long-term control over and flow of access to IP assets over licensing revenues or annual TTO P&L.

Whether inventor-owned or institution-owned, the IP resulting from Canadian research can be made into a productive asset. The only difference between Waterloo and the rest of Canada is the intentionality behind patenting activity, and with effective tools for tech transfer, even institutions that insist on owning the IP output of taxpayer funded research can become effective shepherds of technology beyond the lab if they can get past the short-term thinking that dominates Canadian tech transfer currently.