Ottawa, qui a fait une nouvelle mise de fonds de 400 millions de dollars, doit éviter de répéter l’erreur d’utiliser cet argent pour atteindre des objectifs politiques.
The lacklustre performance of venture capital in Canada and elsewhere over the last decade should focus our attention on how we can improve public policy in this area of the innovation ecosystem. In Canada, the federal government plays a central role in providing venture capital, through the labour-sponsored venture capital (LSVC) tax credit and direct investments by government-backed funds such as the Business Development Bank of Canada. As Ottawa now considers how to spend an additional $400 million in the industry, it is essential to examine whether government support to venture capital is a stimulant or hindrance to greater private-sector investment, if it can fuel the desired expansion in innovation and entrepreneurship and how it can be made more effective.
As with any government intervention, the most critical test is whether public money acts as a complement to, or substitute for, existing behaviour. If government sponsorship simply replaces private venture capital that would have been invested anyway, then nothing has been accomplished and public resources have been wasted. Answering this question is challenging, partly because of limited Canadian data but also because it requires creating an alternative record of what would have happened without government support.
Our research has tracked the survival, monetization, investment and patent history of thousands of Canadian startups funded over the last two decades. The results tend to indicate that the presence of government-sponsored venture capital does have the effect of “crowding out” some private capital, although it also generates some additional funding for emerging enterprises. We find that firms funded entirely by government-sponsored venture capitalists, such as LSVCs, perform poorly, with lower chances of successful exit (acquisition, IPO, etc.), less total investment and less patent activity than firms funded exclusively by private venture capital funds. However, enterprises with mixed sources of funding — private and government venture capital — do well. Enterprises with a combination of majority private venture capital and minority government sponsored venture capital (VC) participation perform particularly well.
This research suggests that while there is a discernible role for government policies to facilitate access to financing, the design of these programs remains very important. Injecting more public capital alone won’t necessarily solve the underlying problems.
Canada’s LSVC programs are the primary building blocks of government support for venture capital. Under this model, investments made through a fund formally sponsored by a union or an organization affiliated with organized labour are eligible for a federal refundable tax credit of 15 percent on investments up to $5,000, in addition to parallel tax credits available in certain provinces (Ontario began to phase out its own program in 2005). For many reasons, this structure is unwieldy and ineffective.
In particular, there is no overriding policy interest we can see as to why government-sponsored venture capital should be tied to organized labour. Although this may have little or no practical effect, the structure appears awkward and cumbersome.
Of even greater concern is the requirement that investors in government-supported funds are limited to small, passive retail investors, primarily among the general public. Handling such investors is costly, and they rarely have the expertise necessary to become well informed.
Furthermore, the LSVC program is based on special licences that do not seem to be allocated or reallocated on the basis of market performance. Poor performing funds can survive and retain their licences, while high-quality funds are often prevented from benefiting from government support because they cannot obtain a licence. In British Columbia, for example, a 2010 analysis of the province’s own VC support program, which includes tax-credits for both labour- and non-labour-sponsored funds, found that funds supported by the program posted five-year losses of between 11 and 57 percent when the tax rebate was excluded.
The fact that tax credits are linked to a limited number of exclusive licences has created a polarized VC market, where private and government-sponsored funds operate under very different rules. The result is an uneven playing field that is costly and inefficient. Entrepreneurs, individual investors and taxpayers are poorly served by an environment in which a policy allows continued support for capital formation that is attractive only on the basis of a continuing tax credit.
But to get to the right endpoint — a system that is more nimble, competitive, and compact — we need to think bigger than just the $400 million.
We suggest three broad principles that should guide reform of the LSVC program. First, any tax credit program should have an inclusive structure that does not give preference to any particular group or sector. Any program should be accessible to all VC funds that satisfy transparent eligibility criteria. We also see no reason to exclude other early-stage investors, most notably angels or corporate VC funds, from these programs, and note provinces are making some headway here.
Second, we believe that instead of relying primarily on tax credits as a policy lever, the federal government could benefit from making greater use of matching funds. Consider the difference between a 50 percent tax credit and a 1-to-1 matching fund. In both cases, the government pays for half of the funds raised by an enterprise. However, with a tax credit the investor gets all of the shares of the company that are purchased. With matching funding, the investor gets half and the government the other half. With a tax credit the investor is subsidized and the government receives no direct return (other than the indirect effect of collecting higher tax revenues if the company performs well).
A system of matching funds does not mean that governments should be involved as the investment manager, however. As shown by notable international examples such as the Small Business Investment Company model in the United States or the Innovation Investment Fund in Australia, governments can successfully enable co-investment using a series of competitive tools. One possible mechanism is for a government to auction management licenses to a variety of private-fund managers who, in turn, would be responsible for selecting transactions. Alternatively, professional program managers could allocate licences or funding using performance-based assessments of venture fund managers.
The primary feature of any allocation method should be to make allocations based on economic performance, not on political grounds or as instruments of social policy. It is important to draw on the private sector’s comparative advantage in investment selection and treatment, while providing clearer performance incentives than the current subsidy model. Ensuring sufficient turnover in licenses is an important lesson to be drawn from the LSVC experience.
We should note that several provinces have recently established their own investment funds that aim to move government-sponsored venture capital away from direct investing and toward this kind of more passive system. The Ontario Venture Capital Fund and Quebec’s Teralys Capital (see Gilles Duruflé’s article in this issue) are both examples of this. We would emphasize that while the empirical evidence on these changes is still emerging, in general governments need to think critically about how capital and policy can be used to promote greater competition and performance among funds.
Given the significant amount of funding Ottawa announced in the March budget, now is the time for experimentation and transformative thinking. There is a perception that entrepreneurial finance in Canada is particularly lacking at very early stages (“incubation” and “angel finance” stages) and at very late stages, the final phase before an enterprise has its initial public offering or is acquired by a third party. This is an area where pilot initiatives could be particularly useful.
Our third element of reform focuses on the relationship between foreign-direct investment and Canadian venture capital. Currently, LSVC and related VC support programs essentially limit funds to Canadian investors and Canadian enterprises. While this restriction may have been relatively harmless in the past, the trend in venture capital is away from being “local generalists” toward becoming “global specialists.” For a Canadian venture capitalist to remain competitive, even within Canada, it may require a global (or at least North American) investment reach. The current tax-credit structure is ill-suited to support such an investment model.
Public policy must reflect globalization. For example, a Canadian firm that needs to outsource a significant part of its production in order to remain competitive now has a choice between forgoing Canadian tax credits and settling for a suboptimal business model. Canadian firms would also benefit greatly by attracting more foreign investors that provide different expertise and networks.
Whether the LSVC system is retained or not, it is imperative that Canadian VC policy be flexible and open toward foreign direct investment. Relaxing geographic restrictions is bound to raise political concerns and will therefore require special attention to political sensitivities. There are, nonetheless, several options to consider. In the context of a tax-credit system, governments can relax the restriction of how many employees or assets have to remain in Canada for an enterprise to be considered Canadian. Alternatively, if a Canadian enterprise raises funds from a mix of Canadian and foreign investors, foreign investors should be allowed to share of a portion of tax credits up to a specified limit. This might allow Canadian companies to attract the best investors globally. It should also be possible to allow Canadian investors to obtain tax credits on an investment portfolio that contains foreign investments within certain parameters.
Similar flexibility toward foreign investment should be shown if the government opts for a system of matching funds instead of tax credits.
We believe the first proposal is relatively easy to implement, the second would be quite valuable and might well be politically feasible, while the third is the most politically challenging.
The government is to be commended for its foresight in wanting to ensure that Canadian innovation policy doesn’t repeat the mistakes of the past. There are many entrepreneurs who would like more money, and many financial intermediaries (like LSVCs) who would like additional government support. That is not surprising. But there is no strong evidence to indicate that, without transformation of the current system, higher government funding levels will improve our performance.
Instead, the investments announced in the 2012 federal budget should be viewed as an opportunity to reset and refresh the policy framework on venture capital in Canada. Doing so will require active engagement with the provinces so reforms are effectively coordinated and harmonized, particularly within the integrated tax-credit system. But to get to the right end point — a system that is more nimble, competitive and compact — we need to think bigger than just the $400 million.