The Canadian Council of Innovators releases their SR&ED review recommendations

A close look at the CCI report for SR&ED reform

The Canadian Council of Innovators (CCI) represents a group of high-growth Canadian companies and is leading the charge in lobbying for effective innovation policy in Canada. Alongside Canadian innovators from all over the country, the CCI recently released a set of five recommendations for cost-neutral SR&ED overhaul, entitled Cost-Neutral Steps Toward a 21st Century Scientific Research and Experimental Development Tax Credit.

Generally speaking, the report is excellent and if implemented would represent a significant improvement over what exists currently. In this article, I provide an overview and commentary on their recommendations. I encourage you to read the CCI report fully beforehand.

1. Rebalance innovation incentives for larger firms towards outcomes that matter

The CCI recommendation is to maintain the current level of SR&ED refund for small business to reduce the non-refundable rate for larger companies, with the intention of offsetting the difference through the use of a patent box regime.

This is a very clever approach. It achieves several goals at once: reductions in refunds for larger firms frees up capital to maintain cost-neutrality while implementing other suggested reforms without hurting the small end of the SME scale. As the CCI points in their report, it also shifts the focus from inputs to outcomes by ensuring that larger companies can potentially even increase their aggregate benefit, provided they actually monetize Canadian-owned IP instead of sitting on it as a moat.

This point also highlights a major issue with Canadian innovation funding supports generally, which is that the definition of SME is far too broad to be useful in designing support structures for all of the companies that fall within it. By distinguishing between different scales in their suggested approach to SR&ED refund rates, CCI indirectly makes clear that within the umbrella of SR&ED-eligible companies there exist a spectrum of stages of development, that all require a different set of supports and incentives. By adopting a more granular breakdown of company stages to implement this suggestion, the CCI is laying the groundwork for a more nuanced approach to categorizing companies that is a prerequisite to building effective innovation policy generally.

2. Incentivize early investments into valuable intangible assets and allow for the inclusion of commercialization (i.e. patenting) and continuous improvement activities

The CCI report here points out that the narrow focus of SR&ED on R&D limits the impact of the funding, given that extracting value from IP developed during R&D requires far more than just its initial development. They report also highlights the harm done by ambiguity in eligible expense definitions.

The core of the recommendation is that SR&ED-eligible activities be expanded to include commercialization activities that follow patent filing, and to include the filing costs of patents themselves, both domestically and in foreign jurisdictions such as the US.

While I would have liked to see the report more explicit about the activities that should be included, I am strongly in agreement generally with this approach. At a bare minimum, the eligible activities should be expanded to include engineering around productization of IP, software development in support of products enabled by Canadian-owned IP, and filing of patents in foreign jurisdictions to support Canadian companies in establishing themselves as exporters. Just like in the academic world where support ends when a patent is filed, SR&ED limiting support to R&D activities without covering anything relating to subsequent commercialization is in part responsible for the exodus of IP from Canada via innovators who cannot find the funding to take it beyond the R&D stage.

3. Modernize key program parameters such as expenditure and phase-out thresholds

The CCI points out that the thresholds that define benefit cutoffs in the SR&ED program have not been updated in decades, and are far behind inflation as a result, calling for an update to bring the program back into line with the actual costs of developing and commercializing R&D outputs.

This point I would deprioritize in favor of recommendation 5. While I agree that the SR&ED program needs to be modernized, I am less convinced that increasing the thresholds is the answer, at least in the short term. As I pointed out previously and as noted in the CCI report itself, as much as 30% of the value of the SR&ED incentive is diluted through the cottage industry of SR&ED consultants and lenders that have sprung up in response to the complexity and subjectiveness of the administration surrounding the program. In the interest of maintaining cost-neutrality in a challenging budget environment, administrative overhauls that eliminate this waste may be preferable to an update to the thresholds that increases the overall spend under the program. Administrative changes to program delivery that recoup the current loss of value to third party SR&ED consultants and lenders would have the effect of increasing the value to innovators by almost the same amount as the suggested increase in thresholds, without increasing the overall cost of the program.

4. Ensure SR&ED supports long-term growth in Canada by implementing new transparency practices and evaluation metrics

The CCI here calls for improved tracking of metrics relating to the long-term impact of SR&ED credits, specifically looking at tracking the long-term impact of IP assets, and call on SR&ED to ensure that companies that export IP must repay public funding received to support its creation. I’m encouraged to see alignment across to many different levels of economic development on this point.

In designing such a tracking program, it is important to remember that there is a time delay, often measured in years, between the R&D that leads to an IP asset and commercial benefit. Consideration of this delay is conspicuously missing from most existing innovation support initiatives, which usually consider only outcomes on shorter timescales. Any system that seeks to truly capture the long-term impact of innovation support must track outcomes for up to a decade to get a complete picture, and must ensure that reporting requirements are inherited by all entities who ever gain access to that IP.

5. Make delivery more efficient through a mix of administrative streamlining and targeted training for personnel

The direct value of SR&ED credits that is lost to administrative friction, consultants, and lenders is enough to fund an entire industry in Canada that needs to be eliminated.

The CCI calls for a real-time submission system and pre-approval of project expenses, which are the key points to achieving the desired elimination of lost impact to third parties. While they do not suggest is explicitly, the monthly payroll tax submission system could easily expanded to include SR&ED claims, which are already based in most cases on a percentage of eligible salaries. The report also calls for better education of auditors reviewing claims, which is also an important aspect of reducing uncertainty around the possibility of a disallowed claim. Consistent, real-time program delivery would have an enormous impact incentivizing development of IP by Canadian firms armed with the confidence that their cashflow will not be negatively impacted by a disallowed claim months after the fact.

General commentary

Overall, the CCI SR&ED recommendations are well thought out and would be highly effective if implemented. Since I feel like I should come up with at least a few criticisms just on principle, I have two broad suggestions that I hope will inform some of the discussions with policymakers that CCI will no doubt be holding over the next few weeks and months.

First, while the focus is clearly on intellectual property and how best to extract its long-term value, the report lacks a clear definition of IP. As written, the implication is that it is focused on patents. I would like to see datasets that can be used for AI/ML included in the list of IP activities covered by SR&ED under an expanded mandate that takes a more holistic view of R&D and commercialization. The increasingly ironically named OpenAI, for example, has no patent (or even copyright, in most cases) on the text they used to train GPT-4 and other LLMs, but it is difficult to argue against the value-add achieved through their curation and structuring of that data.

CCI is already leading the innovation community in educating policy-makers as to the importance of IP that goes beyond just patents, and while that discussion has implications well beyond SR&ED, this consultation call is a golden opportunity to cement that education into policy to provide a foundation on which to incorporate it into other areas of policy development.

Second, while the report does call for administrative streamlining and has a few suggestions for how it could be done, I would have liked to see more explicit detail on suggested methods for implementation. Specific recommendations for consideration can be found here.

I look forward to hearing their detailed thoughts on the proposed patent box regime in the next few weeks.