Editor’s note: See also the accompanying story, “Federal government failed to learn previous policy lessons in creating the innovation clusters program,” in today’s issue of Innovation This Week.

Ottawa’s creation of five innovation clusters was flawed due to the federal public service’s lack of “policy memory” and an inability to learn from a previous national cluster program, say policy experts.

The federal government established the Innovation Superclusters Initiative in 2017 – now called the Global Innovation Clusters program – using an inaccurate and confused concept of clusters and what a clusters program is, said David Wolfe (photo at right), professor emeritus of political science and co-director of the Innovation Policy Lab and at the Munk School of Global Affairs & Public Policy at the University of Toronto.

This federal clusters program, amounting to $1.7 billion over 10 years, and the government’s $15-billion annual spending on science, technology, and research and development programs, haven’t improved Canada’s innovation performance, which has deteriorated over the last decade, Wolfe said in an interview with Research Money.

Canada has more than 120 distinct federal programs and initiatives offering grants, loans and tax credits to support business innovation, with some policy analyses tracking up to 134 separate initiatives across different government departments and agencies.

“There’s no recognition or acknowledgement [by government] that ‘Maybe this isn’t the right policy mix, or maybe it’s not targeting the right things,’” Wolfe said.

Moreover, the government still lacks a central mechanism to examine the incrementality of all its innovation programs and coordinate them and their effectiveness across federal departments and agencies, he noted.

Wolfe and Travis Southin (photo at right), then a PhD candidate at the University of Toronto, co-authored a study, published in Canadian Public Administration, that looked at policy learning and development in innovation policy, in Canada’s innovation clusters initiative.

The clusters initiative was announced in 2017 with little reference to a previous cluster program delivered by the National Research Council (NRC) in the early 2000s, Wolfe said. “There was no policy learning from the experience with the NRC cluster initiative.”

Southin said the current federal clusters program “undoubtedly supported some positive R&D collaborations that will continue to benefit Canadian innovation long into the future.”

“Still, it’s clear from our interviewees’ responses [in the study] that the program didn’t make the most of the lessons from previous federal cluster initiatives,” he said in an email to Research Money.

“Empowering more autonomous cluster-level decision-making, involving provincial and regional entities earlier in the policy development process, and ensuring resources aren’t spread too thinly across regions, sectors and technologies would all contribute to a more efficient and effective program,” Southin said.

Wolfe pointed out that a clusters program, according to all the scientific literature, involves geographically constrained clusters that are regionally constrained and localized.

But the federal government, in setting up the Innovation Superclusters Initiative/Global Innovation Clusters (ISI/GIC) program, never made that distinction, he added. “There’s just a tremendous amount of conceptual confusion in what the program is and what it was supposed to do.”

The government used three to five different conceptual terms for clusters, all of which mean different things in the academic and the policy literature, Wolfie said. “But they used them almost interchangeably to describe the [ISI/GIC] program.”

“So it was really clear to me that the people evaluating the program don't have a clear idea of what it is. They don't have a clear idea of what the conceptual framework is that it's grounded in, other than [it being] an industry-led research program.”

Rather than being a government-led program, the ISI/GIC is an industry-led research program, Wolfe said. Instead of being geographically constrained, each of the five clusters has a network of companies, postsecondary institutions and not-for-profit organizations from across Canada.

Even though the government said the ISI/GIC program was different than the NRC national clusters program in the early 2000s, both programs focused on research funding, he said.

Both cluster programs were predicated on the idea that if you provide government money to – government labs in the NRC-run cluster, and an industry-led consortia in the ISI/GIC – that this funding will generate clusters and improve productivity and the pace of innovation, Wolfe said.

The scientific literature about clusters indeed says that companies located in clusters do more research and development and are more productive, and the cities that have higher concentration of clusters outperform in innovation and productivity, he said.

“But those are geographically constrained clusters,” where there is support that goes to the clusters themselves, he noted.

The advantage of actually working with geographically constrained clusters is the benefits of geographic proximity and the ability to deliver policy benefits simultaneously to a cross section of firms, which the government “seem to me to be completely oblivious to,” Wolfe said.

Evaluations of clusters program aren’t using meaningful metrics

In addition, Innovation, Science and Innovation Development Canada (ISED) isn’t using the right metrics to evaluate the performance of Canada’s five global innovation clusters, he said. ISED uses several output metrics such as jobs created, funding for clusters, amount of research performed, and amount of industry spending leveraged, he said.

“But it’s not a real evaluation,” Wolfe said. ISED’s evaluation, he added, doesn’t answer the fundamental questions such as:

• Is this spending that these companies wouldn’t have undertaken on their own? What would these companies have done without ISI/GIC funding?
• How is this R&D spending interacting with other government innovation programs?
• How are companies stacking different federal innovation programs and which programs specifically, and where and how do companies use the different federal funds for different parts of their research agenda?
• Has ISI/GIC funding made a difference to the companies’ overall performance in terms of their success in scaling up? What is the companies’ patent, revenue and export performance?

The data, including the federal Business Innovation and Growth Support statistics, and the methodology exist to do that kind of meaningful evaluation and answer those questions, but it’s not being done, Wolfe said.

The ISI/GIC program has done some good things, such as connecting multinationals with Canada’s small and medium-sized businesses, he acknowledged.

John Knubley, former deputy minister of ISED who was responsible for creating the Innovation Superclusters Initiative, told Research Money in 2021 that the clusters should be assessed by a broad innovation policy framework and with a longer time frame than their initial five-year mandate.

“Above all, the superclusters should be judged by how they move the needle on Canada’s innovation performance,” Knubley said.

Wolfe argued that the superclusters haven’t improved the country’s innovation performance in nearly a decade, based on the November 2025 Council of Canadian Academies (CCA) expert panel’s report on the state of science, technology and innovation in Canada.

The CCA’s report found that Canada still lacks an effective approach to support the development and commercialization – across the continuum from research to deployment – of the most promising areas that could improve national competitiveness and provide greater overall economic and social benefits.

“Canada is not translating upstream strengths in research and talent to downstream economic benefits,” the report said.

Wolfe said when it comes to the federal innovation clusters program, “We don't have a really clear understanding of what it has done, where the real benefits are, where the limitations are.”

Moreover, the federal government doesn’t have a good understanding of the critical policy question – “‘Why did all these new measures introduced since the Skills and Innovation Plan [part of the 2017 budget] not fundamentally move the dial on our innovation performance?’” Wolfe said. “What are we missing?”

Government lacks a central body to coordinate, deliver and assess innovation programs

When it comes to improving innovation performance and productivity, Canada doesn’t know what it’s missing because it lacks a central, integrated delivery mechanism or unit that controls all of the country’s innovation programs and can assess how these programs are impacting individual companies, Wolfe said.

There is no central unit responsible for coordinating innovation policy across ISED, the Department of National Defence, Natural Resources Canada, Environment and Climate Change Canada and several other departments and agencies, he said.

“When you've got five to 10 different federal government programs all being managed by separate program managers, all being funded on a one-off basis, you don't have that overall integrated perspective.”

“So [the problem] is both the lack of policy learning, the lack of continuity, but also the lack of coordination at a central agency level,” he added.

The government needs to create a coordinating unit within the Privy Council Office, Wolfe said.

The Canada Innovation Corporation (CIC), proposed in federal Budget 2022, was supposed to function as such a coordinating unit.

Both a new business innovation agency (named the Canadian Innovation and Investment Agency) and a new academic research funding management organization, or capstone agency (named the Canadian Knowledge and Science Foundation), were recommended by the federal Advisory Panel on the Federal Research Support System, chaired by Frédéric Bouchard, Dean of the Faculty of Arts and Sciences at the Université de Montréal, in its 2023 report.

However, Budget 2025 offered no update on the CIC and provided no new details on the proposed new capstone research funding organization to provide better coordination across the federally funded research ecosystem.

Wolfe noted that one of the limitations of Canada’s parliamentary government based on the Westminster model is that there is presumption that if someone is a good manager, the content expertise that person has is a secondary consideration.

The thinking is “that if you’ve demonstrated your ability to manage effectively in one department, you can manage well in any department without being a sector content expert,” he said.

That leads to constant turnovers of deputy ministers and associate deputy ministers in the federal public service, he added. “Just as people are building up serious expertise in their area, they get moved to a totally different department [and] a totally different set of files.”

“That’s one of the reasons, I think, why there is lack of institutional memory [in the federal public service],” because senior ministry officials aren’t kept in one department long enough to build and retain sector expertise, Wolfe said.

As a result, Canada’s government ends up not having built-in institutional mechanisms for policy memory, to ask: ‘What have we tried in the past? Why did it work? Why did it not work? What lessons can we learn for what we want to do today?’” he said.

Policy-making capacity of the federal bureaucracy “has atrophied significantly”

In a 2024 op-ed, Konrad Yakabuski, a columnist for The Globe and Mail, pointed out that between 2015 and the first quarter of 2024, the ranks of the public service increased by more than 40 percent to about 368,000, yet “the policy-making capacity of the bureaucracy has atrophied significantly.”

The role of outside consulting firms has exploded as in-house expertise withers and senior bureaucrats and departmental managers look to consultants for advice, “in essence abdicating their own responsibility for formulating policy recommendations to their political bosses,” Yakabuski wrote.

At the time, Donald Savoie, Canada Research Chair in public administration and governance at l'Université de Moncton and an expert on the federal public service, called for a royal commission to examine the state of the public service and recommend changes to its structure and mandate.

The aim was to restore the public service’s central role in the policy process, recalibrating the balance of power between the Prime Minister’s Office and the bureaucracy and ensuring the public service is able to attract the talent needed to address public policy challenges in the 21^st century.

“More than ever, in our darkening age of political polarization, we need a neutral and non-partisan public service to guide major policy decisions. And we need competent public servants to implement them without fear or favour,” Yakabuski wrote.

Kevin Lynch, former clerk of the Privy Council, and James Mitchell, an adjunct professor at Carleton University and former senior federal public servant, warned in their book, A New Blueprint for Government: Reshaping Power, the PMO and the Public Service, that “business as usual” is not a winning strategy for any country.

“Their opening premise is stark: Canada’s public sector has lost its capacity to think, plan and deliver in a world that is moving far faster than its institutions,” Colin Robertson said in a review of the book.

Lynch and Mitchell describe a machinery of government that has become inward-looking, reactive and structurally incapable of managing sustained change.

Much of New Blueprint for Government reads as an indictment of policy incoherence, Robertson said. “The authors point to a proliferation of goals without priorities, and announcements without implementation.”

“They describe departments more focused on ‘managing narratives’ than on solving problems, and central agencies consumed by risk aversion rather than innovation. The result is a policy ecosystem defined by slogans and silos.”

Lynch and Mitchell call for a return to cabinet government, where ministers are accountable for results and supported by policy-capable departments.

Said Robertson: “The key takeaway from A New Blueprint for Government is clear: In an age of populism, protectionism and policy fatigue, government must once again learn to think long term, act collectively and deliver competently.”

R$