Canada Keeps Trying to Fix Productivity. Why Aren’t the Results Changing?
We measure funding. We measure activity. We measure participation. Perhaps it’s time we measured value creation.
Canada has a productivity problem.
That is hardly news.
For decades, economists, business leaders and governments have warned that Canadian businesses invest less per worker than many of their international competitors — particularly those in the United States.
The problem has persisted long enough that it is no longer simply an economic statistic. It is affecting our standard of living.
In November 2025, the Bank of Canada estimated that if Canada’s productivity growth since 2000 had simply matched that of other G7 countries, Canadian GDP today would be approximately 9% higher — almost $7,000 more per person.
That is an extraordinary amount of unrealized prosperity.
But Canada certainly hasn’t spent the past 25 years doing nothing.
Successive governments of different political stripes have introduced tax incentives, innovation strategies, investment funds, superclusters, accelerated depreciation measures, digital programs, skills initiatives and commercialization programs.
Businesses have participated. Universities have produced world-class research. Billions have been deployed.
And here is what makes the problem interesting:
Many of these initiatives can legitimately demonstrate success.
So why does the national outcome remain so stubbornly weak?
When successful programs don’t produce a successful system
Consider Canada’s Global Innovation Clusters.
As of the end of 2025, the clusters had approved 676 projects involving 3,500 partners, more than half of them SMEs.
Those projects generated more than $3.39 billion in co-investment, including $2.11 billion from industry and other partners. The program achieved $1.60 of partner investment for every federal dollar — exceeding its target.
More than 6,000 licences to foreground intellectual property have been granted to third parties.
Those are meaningful results.
Or consider the Strategic Innovation Fund.
An early federal evaluation found that 60% of surveyed recipients said their project would not have occurred without SIF funding. Funding also affected project timing, scope and scale.
During the first year of SIF-funded projects, R&D intensity among recipient firms increased from approximately 6.1% to 8.4% of revenues.
That is meaningful additionality.
And it illustrates something important.
This is not an argument that government innovation programs don’t work.
Some clearly do, according to their stated objectives.
That is precisely why Canada’s persistent productivity problem deserves a deeper question.
If individual initiatives can demonstrate success while the national outcome remains persistently weak, perhaps the missing link lies somewhere between program execution and value creation.
“Additional” to what?
Economists often evaluate programs such as R&D incentives by measuring additionality.
The idea is straightforward:
Did the incentive cause an organization to undertake activity it otherwise would not have undertaken?
Suppose a company would have spent $1 million on R&D without an incentive but spends $1.2 million because the incentive exists.
We can reasonably identify $200,000 of additional R&D spending.
That matters.
But what happened next?
Did that additional spending create new technological capability?
Did it generate useful intellectual property?
Did it improve the company’s productivity?
Did it create a commercially valuable product?
Did it improve margins, generate exports or enable the company to scale?
Or did it simply produce more technically eligible R&D activity?
“Additional” describes a difference. It doesn’t tell us whether that difference matters.
In my own experience, I have seen perfectly legitimate R&D credits attached to projects that were technically eligible but strategically peripheral to the business.
The work was legitimate. The claim was legitimate. The people involved were genuinely performing qualifying R&D.
The credit reduced the effective cost of salaries devoted to that work.
That’s what the program is designed to do.
But eligibility and strategic value are different questions.
This does not mean every publicly supported R&D project should generate immediate revenue.
Innovation is uncertain by definition. Experiments fail.
Technologies can take years to mature. Fundamental research can generate enormous spillovers that no single company will ever capture.
The relevant question is not whether every project succeeds commercially.
It is whether, across a portfolio and over time, publicly supported innovation creates capabilities and economic value that would otherwise not have existed.
That is a much higher bar than measuring activity.
What happens after the IP is created?
There is another gap in the innovation conversation that deserves more attention.
We often treat intellectual property creation as an outcome.
A patent is filed.
A technology is developed.
A licence is issued.
A spin-off is created.
Success.
Except intellectual property creates little economic value simply because it exists.
Eventually, someone has to use it.
The OECD describes Canada as performing well in fundamental research while identifying persistent difficulties in translating research into large-scale commercial applications.
It also points to an uncomfortable phenomenon: Canadian-created intellectual property and promising startups can migrate abroad through the movement of inventors, foreign ownership and acquisitions.
That should make us reconsider what we call an innovation outcome.
More patents do not necessarily mean more Canadian productivity.
And this issue extends well beyond corporate R&D.
Canadian universities, colleges, hospitals and public research institutions generate enormous amounts of knowledge and intellectual property, much of it supported directly or indirectly by public funding.
Canada has built institutions specifically to move those discoveries toward the market: technology-transfer offices, research centres, incubators, college technology-transfer centres and commercialization organizations.
In Quebec, Axelys was created specifically to help transform public research into technologies, licences and businesses.
These organizations play an essential role.
But they also reveal how many transitions must occur before research becomes economic value.
A discovery must become an invention.
The technology may need intellectual-property protection.
It must be validated.
Someone must determine whether it solves a sufficiently important problem.
It may need to be licensed.
Someone must finance it.
A company must adopt it.
Customers must buy it.
And if it succeeds, someone must scale it.
Each transition can fail.
Which means the innovation chain is considerably longer than we sometimes acknowledge:
Funding → Research → IP/Capability → Validation → Transfer → Adoption → Scale → Value Creation → Productivity
That raises a deceptively simple question:
How much of the knowledge Canada pays to create actually becomes productive economic capacity in Canada?
Statistics Canada has only recently begun rebuilding a national picture of this process. Its new survey of research activities and intellectual-property commercialization in higher education measures R&D, IP management, technology transfers, spin-offs and barriers to commercialization.
That is an important development.
Because what we don’t measure systematically is difficult to orchestrate.
What leading innovation economies are learning
Canada is not alone in confronting this problem.
In fact, some of the world’s strongest innovation economies are explicitly working on the same transitions.
Sweden is one of Europe’s leading innovation economies and has an exceptionally strong academic research base.
Yet Sweden’s innovation agency, Vinnova, acknowledged in 2026 that too many knowledge-intensive ideas still fail to reach the market.
Its response is revealing.
Rather than simply funding more research, Sweden is strengthening early verification and validation — testing technological feasibility, market potential and business viability before deciding whether an innovation should move further toward commercialization and scale.
Denmark has developed programs deliberately designed to bridge different parts of the innovation chain: research commercialization, academic entrepreneurship, industrial researchers working simultaneously with universities and companies, and national efforts to turn university research into spin-outs.
Norway similarly funds explicit qualification and verification stages intended to move publicly financed research toward commercial exploitation or societal use.
Japan offers another variation.
Its long tradition of collaborative industrial R&D has evolved into structures capable of bringing companies, universities and research institutions together around common technologies, shared capabilities, licensing, standards and commercialization.
Japan has also developed mechanisms under which certain publicly supported intellectual property can, under specific circumstances, be made available for third-party use rather than remain indefinitely unexploited.
None of these countries has solved innovation.
Sweden itself says too many promising ideas still get stuck between academia and the market.
And that may be the most important lesson.
Leading innovation systems don’t eliminate failure.
They identify where value is being lost and intervene at those transitions.
The difference may therefore not be that these countries possess programs Canada doesn’t have.
Canada has many of the same pieces.
The more interesting question is:
Are their systems better connected than ours?
Perhaps the program is the wrong unit of measurement
This brings us to what may be the central issue.
Government naturally manages programs.
Programs have budgets, administrators, eligibility criteria, objectives, dashboards and evaluations.
But the economy doesn’t experience programs.
It experiences the cumulative result of thousands of decisions and transitions across an innovation system.
A research grant may produce a discovery.
Another organization may protect the IP.
A commercialization centre may license it.
A startup may acquire it.
An investor may finance the startup.
A corporation may test the product.
Customers may adopt it.
Another company may eventually acquire it.
The technology may then scale — perhaps in Canada, perhaps elsewhere.
Every organization along that chain can meet its own objectives.
And Canada can still capture surprisingly little of the ultimate economic value.
That leads to a different way of thinking about innovation policy:
Perhaps the real unit of innovation policy shouldn’t be the program. Perhaps it should be the value-creation chain.
Instead of asking only whether individual programs work, we would also ask where value is being lost between them.
Where does research stop?
Where does IP sit unused?
Where does validation fail?
Where does financing disappear?
Where does adoption stall?
Where do promising companies stop scaling?
Where does Canadian-created IP leave the country?
And critically:
Who owns the outcome across those transitions?
Everyone may be doing their job
This is where Canada’s productivity problem starts looking less like an economics problem and more like a management problem.
Government identifies a priority.
A department designs a program.
Researchers perform research.
Universities generate knowledge.
Technology-transfer organizations seek partners.
Businesses rationally use the incentives available to them.
Advisers help companies navigate eligibility.
Investors pursue returns.
Regulators fulfil their mandates.
Program administrators measure their outputs.
Nobody needs to be incompetent.
Nobody needs to be acting irrationally.
Nobody even needs to be doing anything wrong.
Everyone can successfully execute their part while the system still fails to deliver the intended outcome.
Anyone who has managed a large organization has seen a version of this.
Every business unit meets its KPI.
Every project dashboard is green.
Every leader can demonstrate progress.
Yet the enterprise fails to achieve the strategic result those activities were supposed to produce.
The issue isn’t necessarily execution within the individual parts.
It is orchestration across them.
Canada’s internal market tells the same story
Innovation is only one example.
Internal trade provides another.
The Canadian Free Trade Agreement came into force in 2017.
Yet it took until June 2025 for the federal government to remove all 53 of its remaining exceptions from that agreement.
Eight years.
Meanwhile, hundreds of billions of dollars in goods and services move across provincial and territorial borders every year, and estimates of the economic opportunity associated with reducing remaining internal trade barriers run into the hundreds of billions of dollars over the long term.
Yet implementation requires federal, provincial and territorial governments, regulators, professional bodies and industries to align around an outcome no single participant controls.
Again, everyone owns part of the problem.
Who owns the outcome?
Perhaps it’s time we measured value creation
For years, we have measured dollars deployed, firms supported, projects launched, partnerships created, patents filed and jobs associated with programs.
Those are legitimate measures.
They tell us whether activity occurred.
Additionality tells us something more:
Did our intervention change behaviour?
But perhaps we need to keep measuring further down the chain.
Did the technology get adopted?
Did the intellectual property get used?
Did companies become structurally more productive?
Did capital investment per worker increase?
Did Canadian businesses scale?
Did exports grow?
Did new capabilities remain and compound inside the Canadian economy?
And ultimately:
Did the outcome we were trying to change actually change?
Not every program can or should be evaluated against national productivity.
Not every innovation should be commercialized.
Not every patent should become a company.
And government innovation policy is only one of many forces affecting Canada’s productivity performance.
Those caveats matter.
But they shouldn’t prevent us from asking a larger systems question.
Canada has spent decades building programs designed to improve innovation, investment and productivity.
Many produce legitimate results.
Yet the aggregate outcome remains disappointing.
Perhaps we have become very good at measuring whether the pieces work.
Perhaps it’s time we measured whether the system creates value.
From execution to orchestration
Canada does not appear to suffer from a shortage of ideas.
We have world-class researchers.
We have entrepreneurs and sophisticated companies.
We have capital.
We have universities and research institutions.
We have innovation agencies, tax incentives, commercialization organizations and investment programs.
In many cases, we can demonstrate successful execution.
What we have not yet demonstrated is that all of those pieces consistently combine into the productivity outcomes we need.
That suggests a question worth asking:
Is Canada measuring successful execution too close to the beginning of the value-creation chain?
If so, the solution isn’t necessarily another program.
It may be better orchestration of the programs, institutions, capital and incentives we already have.
Clearer outcomes.
Measurement across institutional boundaries.
Longer-term tracking.
Better visibility into where publicly supported intellectual property ultimately goes.
Stronger connections between research, commercialization, adoption and scale.
And accountability not simply for delivering programs, but for understanding whether value actually moves through the system.
Because successful execution at every individual stage does not guarantee a successful outcome.
Most organizations execute. The best orchestrate.
Perhaps the same is true of countries.
—————-
Sources & References
Bank of Canada. Canada’s weak productivity: Reversing course, November 2025. The Bank estimates that matching the productivity growth of other G7 economies since 2000 would have left Canadian GDP roughly 9% higher, or almost $7,000 more per person.
Bank of Canada. Time to break the glass: Fixing Canada’s productivity problem, March 2024. Discussion of Canada’s longstanding weakness in business investment and capital investment per worker.
OECD. OECD Economic Surveys: Canada 2025 — Raising business sector productivity. Analysis of Canada’s R&D intensity, commercialization challenges, business investment, innovation performance and international ownership of Canadian intellectual property.
Innovation, Science and Economic Development Canada. Global Innovation Clusters — Program Results. Project, partnership, co-investment and intellectual-property licensing results through December 2025.
Innovation, Science and Economic Development Canada. Evaluation of the Strategic Innovation Fund. Analysis of investment additionality, R&D intensity and early commercialization outcomes among SIF recipients.
Statistics Canada. Survey of Research Activities and Intellectual Property Commercialization in Higher Education, 2023. National measurement of research activity, intellectual-property management, technology transfer, spin-offs and barriers to commercialization.
Government of Quebec / Axelys. Public reporting and evaluations relating to the valorization and transfer of intellectual property arising from Quebec public research.
Vinnova — Sweden’s Innovation Agency. 2026 initiatives relating to early verification, validation, testing and scaling of research-intensive innovations.
Innovation Fund Denmark. Innoexplorer, Innofounder, Industrial Researcher and Spin-outs Denmark programs supporting transitions from research through commercialization and entrepreneurship.
Research Council of Norway. 2026 commercialization programs supporting qualification and verification of publicly funded research.
METI / NEDO, Japan. Collaborative Innovation Partnerships and intellectual-property frameworks governing collaborative and publicly supported industrial R&D.
Government of Canada. Canadian Free Trade Agreement and removal of federal exceptions, 2017–2025.
Comments