Canadian and provincial governments have had science, technology and innovation policies since 1968 (Doern et al, 2016), but with mixed results in spurring business innovation and commercialization (Jenkins et al, 2011). Furthermore, Canada’s strengths in the primary and resource sectors, and the non- metropolitan regions that host them, have been overlooked by innovation policy in favour of sectors usually associated with urban centres (e.g. ICT, biotechnology) (ibid). Non-metro regions represent one third of Canada’s population and employment, and over one third of GDP (FCM, 2016). Non-metro regions face challenges in terms of demographic change and decline, price fluctuations on primary goods, and declines in manufacturing and agricultural employment (CRRF, 2015). Such regions would benefit from tailored innovation strategies to update existing sectors and create new economic niches, countering outmigration and the view that rural areas are now more places of consumption rather than production, and to expand production beyond single-commodities (OECD, 2007). This paper will propose the theoretical basis and justification for a framework to design innovation strategies for non- metro communities The framework will aim to deal with the social and environmental challenges these communities are facing, as well as rejuvenate/diversify their local economies.
Non-Metro Regions: Europe and Canada
In both the EU and Canada, and in contrast to urban regions, non-metropolitan regions face a series of limitations that make I tmore difficult to have the type of economic growth associated with innovation and technology sectors, and which make the regenation of traditional industries different. They do not have the agglomeration of services and activities that enhance economies of scale, nor is there proximity between a larger number of actors to facilitate knowledge exchange (Ashton et al, 2016; Culver et al, 2015; Doloreaux and Dionne, 2008; Hall et al, 2014; Naldi et al, 2015). Lower levels of capital and fewer experienced entrepreneurs exist in non-metro regions, while these same regions can be buffeted by global market forces due to the prominence of primary/resource sectors (Doloreaux and Shearmur, 2006). Furthermore, while the literature generally portrays research intensive Post-secondary institutions (PSIs) as important catalysts of innovation (Doutriaux, 2003; Wolfe, 2005), non-metropolitan regions in Canada tend not to have the most research intensive PSIs in their community (i.e. U15). Where PSI institutions do conduct R&D, they must balance between conducting research with interests from outside the local community for reputational gain with providing knowledge for a community that may have lower levels of absorptive capacity/abilities to use the research produced (Cohen and Levinthal, 1990; Kempton, 2015). What non-metropolitan regions require, and would lead to smart innovation policies, is a localized evolutionary and co-generative approach, which should be better able to engage local stakeholders than past policy attempts (e.g. EU LEADER programme; Dargan and Schucksmith, 2008).
Innovation, RIS3 and Regional Development
Innovation is the application of new products, processes, services and organizational methods for commercial or social purposes (OECD, 2005). Research and Innovation Strategies for Smart Specialization (RIS3) is a policy approach developed in Europe (EC-IIPTS, 2011; Foray at al, 2009; OECD, 2013) to foster regional development in a way that: a) leverages the R&D strengths in science and technology across multiple regions; and b) applies them in contextually appropriate ways to enhance local socio-economic productivity. As initially developed, however, RIS3 would not be able to deal with the likely gaps in entrepreneurial skill and low innovation system development in many non- metropolitan regions (Teras et al, 2015); this paper will recommend the necessary adjustments.
RIS3 was developed to address uneven regional development in Europe with the idea that cutting edge General Purpose Technologies (GPTs – e.g. nanotech, biotech, information tech) would be developed in regions with strong R&D (e.g. Cambridge, UK; Basel, Switzerland); and regions with less advanced R&D capacity could then develop GPT applications for sectors in their local economy, leading to more efficient use of research and innovation investments (Barca, 2009; Foray et al, 2009). Further, RIS3’s focus on local knowledge strengths, and pulling knowledge into the region for local needs, would aid Canadian non-metro regions.
I have critiqued RIS3 and propose a dynamic, evolutionary and socially inclusive institutional approach (Mastroeni, 2016; Mastroeni et al, 2013; Rosiello et al, 2015). I frame market changes and policy implementation as a series of consecutive events, with system evolution shaped either by agent activity or institutional influence (Carlsson and Stankiewicz 1991; Mastroeni et al, 2013; Teubal 1997). I further describe four features:
The first is the framework’s ability to deal with complexity and uncertainty. The complexity stems from regional economies having context-specific challenges different to each other, and the wide variety of stakeholders across different industries interacting and exchanging knowledge. Adding an evolutionary approach would keep system-complexity and regional specificity at the forefront of analysis.
The second is in contrast to the original RIS3’s assumption that private sector entrepreneurs will be present locally to identify opportunities for innovative application of knowledge and GPTs (Foray et al, 2009). In non-metro communities, skilled entrepreneurs and the support-structures that aid them may not be fully developed (McCann and Ortega-Argiles, 2011). Equally, public sector ability to enact change cannot be assumed (e.g. lack of resources, competence, will). Instead, all community members are potential entrepreneurs, emphasising multi-stakeholder dialogue (Mastroeni et al, 2013).
The third looks to avoid too much specialization (Asheim et al, 2011; Saviotti and Pyka, 2008) as too little economic variety reduces a system’s resilience to change (Cooke, 2009). The adjusted framework instead encourages “related variety,” new economic activity formed by new (re)combinations of knowledge held in the community (Asheim and Grillitsch, 2015; Mastroeni et al, 2013).
The fourth promotes strengthened communication and trust relationships between stakeholders, improving knowledge exchange and collaboration to address the uncertainty in the innovation process (Gertler and Wolfe, 2004; Langlois and Robertson, 1995; Morgan, 2007).
The theoretical framework will include two heuristics/tools and a foresight method respectively: an
Evolutionary Life Cycle (ELC), Innovation Matrix (IM) and Three Horizons foresight method.
The ELC heuristic frames innovation systems as moving through different phases: a background/pre-entrepreneurial phase, a pre-emergence/launching phase, and an emergence phase (i.e. critical mass) (Avnimelech and Teubal 2008; Rosiello et al, 2013).
Movement through phases is determined by the region’s ability to provide a set of functions such as research and development, skill building and access
to training, commercialization, and finance (see figure 1). Since an innovation strategy would not be able to realistically correct all weak system functions simultaneously, the ELC helps determine what the system weaknesses are, and the timing as to what functions to address first.
The IM was used in the region of Bavaria by Bayen Innovative, a governance agency for regional development (Cooke and Eriksson, 2011). The IM was used to identify “adjacent possible” areas of innovative activity, i.e. new niches for economic activity combining the knowledge needs with the knowledge strengths of different regional stakeholders in unexpected ways. It does so by laying out different economic activities that are important to the region along the y axis, and the knowledge that is being developed and/or available to the region along the x axis). Community stakeholders, including local PSIs, add their insights to the matrix (not limited to a 2×2 structure) through round-table discussions or interviews, and in the process have their contributions scrutinized by other stakeholders. In our framework, it will help analyze knowledge complementarities in non-metro regions, combining that of participants in established industries, local PSI, and some extra-regional participants to identify potential knowledge complementarities and synergy leading to innovation.
The Three Horizons foresight method will also be applied (figure 2). It maps shifts from the established, status quo patterns of social and economic activity (Horizon 1), towards new ways of doing things to better fit the changing regional conditions (Horizon 3), through a transitional stage that responds to the short-comings of the present (Horizon 2) (Sharpe, 2013; Sharpe and Hodgson, 2014). The Three Horizons approach helps identify potential areas of conflict in the transition from H1 to H3, and helps to facilitate positive rather than conflicting interactions amongst these interests. Three Horizons can be used for a variety of forward looking timelines (5 yrs, 20 yrs, etc.), can help to plan action attempting to reach a desired future, and is relatively easy to explain and teach to community stakeholders/participants.