Cluster Dynamism: A Longitudinal Study of Innovation and Research in the Canadian Auto Industry

Publication Type:

Conference Paper


Greig Mordue


Gerpisa colloquium, Paris (2017)


Research Questions

Can jurisdictions lacking an indigenous OEM leverage local OEM operations to bring significant R&D investment to their jurisdiction? What can we learn about research creation – purportedly a significant characteristic of cluster dynamism – from a case study of the Canadian automotive manufacturing industry?


Policy makers in Canada have long sought to leverage the presence of Original Equipment Manufacturers (OEMs) to generate R&D. Various programs have been devised. In the 1980s, Canada and Ontario even extracted a commitment from Chrysler to locate its R&D operations in Canada in return for their participation in the bailout package designed to save the company. Also in that decade, the auto industry became a target of policy makers seeking to gain global product mandates with all of the research and innovation such authorization would engender.

Fast forward to the 2000s. Several events and trends have conspired to cause policy makers to agitate about the future of automobile manufacturing in Canada, many of which are described by Mordue and Sweeney in a forthcoming edition of the International Journal of Automotive Technology and Management. These challenges have intensified efforts by Canadian policy makers to sustain the auto industry with the fulcrum of their strategy pivoting on perceived strengths in the country’s research and innovation capacity. This is evidenced by the fact that key automotive programs a) target large projects and therefore large companies (i.e. OEMs or large Tier 1s), and b) assess individual applications more favourably when they include commitments to research spending.


Our project employs mixed methodology. Conventional sources like Statistics Canada document the status and trends the automotive industry in Canada is experiencing in terms of production, employment, earnings and capital investment. Additional information is created from a database prepared by Canada’s Automotive Policy Research Centre (APRC), providing a richer, more robust overview of the industry in Canada.

Data on research intensity is derived from a variety of sources including both private and public sector databases on patents. Additionally, data from various Statistics Canada sources is deployed to study research spending by size of employer, ownership and product. The result is an overview of research intensity (over time) of the Canadian auto industry. Through this, new insight is provided into an important, and until now, insufficiently understood aspect of the Canadian auto industry.

An industrial policy analysis is also conducted. It considers the degree to which R&D-oriented tools are targeted to a specific firm or sector (e.g. vertical industrial policy) or across the economy overall (e.g. horizontal policy).

Beyond that, a series of interviews is conducted with industry actors. This includes executives from OEMs, suppliers and policy makers. By combining quantitative data with qualitative perspective, greater context is provided and a fuller, more nuanced understanding of the programs, policies and motivations result.


Findings to date include:
• Various challenges have caused Canada’s automotive and industrial policy makers to seek new ways for the industry in Canada to remain viable.

• Policy makers have long sought to position Canada as a compelling location for research. Historically, that process has been top-down and prescriptive, manifest by insisting upon “investments” in research in conjunction incentives for manufacturing. A multi-year analysis of spending and patent filings demonstrates that efforts, overtures and programs designed to invoke a top-down, OEM-levered approach have generated minimal automotive research breakthroughs by OEMs in Canada.

• Even though vertical industrial policies centred on automotive R&D have targeted OEMs, automotive-relevant research in Canada has originated disproportionately from non-OEM segments. Most of the country’s automotive R&D success (as evidenced by patent filings) has been generated by Canadian headquartered suppliers or other organizations having no direct connection to the automotive industry
Several explanations are offered for the non-OEM centred results:

• Auto manufacturing in places like Canada and auto research are unconnected. To explain: even in Canada, most automotive patents are focused on improving product, not processes. Process improvements that might be created at an assembly plant do not result in new or different automotive products. Further, assemblers have not historically sought patent protection for internally generated process improvements.

• OEMs in Canada, with few exceptions, have not been mandated to conduct, coordinate or promote research in Canada.

• Covenants included in contracts are, by definition, legal undertakings. Therefore, covenants in incentive packages dealing with research expectations, by their nature, describe the bare minimum expectation of their signatories.

• OEMs in Canada are, without exception, subsidiaries with headquarters outside of Canada. The research investments that policy makers covet are not made by assembly plant management. A big body of literature has accumulated over a long period of time about why research amasses near headquarters. Another discusses the matter of embeddedness, inferring that subsidiaries do not deeply ingrain themselves in the broader “community” or cluster that might exist. Canada’s auto industry validates those findings.

Policy Implications

1. Policy makers should avoid the proclivity to tie manufacturing incentives to undefined research. Doing so begs the question, what is so different about Canada that it could reasonably expect to use automotive assembly to build an automotive research / technology driven cluster?

2. For OEMs, geographic proximity to headquarters remains a more compelling rationale for R&D investment than the individual or combined effects of incentives for engaging in such work (carrots) or the application of pressure from the non-headquarter OEM jurisdiction (sticks). In other words, location still matters for automotive R&D.

3. If policy makers insist on attaching research undertakings to production incentives, those incentives should support well-articulated mandates for leadership in areas that are global or regional priorities.

4. It is unclear whether policy makers recognize the consistent futility of top-down strategies hinged on the existence of a foreign-owned assembly plant. However, evidence exists that automotive policy makers in Canada are increasingly willing to assist more organic approaches. In that regard, cluster development strategies that leverage underlying capabilities (e.g. technology) may be more successful than those hinged on a non-indigenous – in this case, OEM – anchor.

  GIS Gerpisa /
  4 Avenue des Sciences, 91190 Gif-sur-Yvette

Copyright© Gerpisa
Concéption Tommaso Pardi
Administration Juan Sebastian Carbonell, Lorenza MonacoGéry Deffontaines

Powered by Drupal, an open source content management system