Julian Wood

By Julian Wood - 08/04/2019

Julian Wood

By Julian Wood -

Like what you're reading?
Share it around.

Or just highlight the part you like...

Smart Specialisation


Across the OECD many have turned toward what has become known as a “smart specialisation” investment strategy. “The concept is now a key element of the EU 2020 innovation plan;”[1] it is set as a conditionality[2] for the R&I and ICT target objectives of the EU’s Cohesion Policy; and the OECD and the World Bank are launching activities to promote and measure smart specialisation.[3] In terms of regional development investment strategies smart specialisation is shaping up to be the best practice model.

As an investment strategy smart specialisation aims to facilitate the emergence of new activities rich in knowledge spillovers, diversify a regional economy through the generation of new options, and help build a critical economic mass through networks and clusters.[4] It is built on four key components: clear objectives, entrepreneurial discovery, granular interventions, and time-limited funding. We will discuss each in turn, highlighting the key components of a well-functioning regional development investment strategy.

Clear objectives

A smart specialisation strategy hinges on having clear objectives, agreed through a multi-level governance process[5] whereby the various levels of authority, be they central, regional, or local government, as well as local entrepreneurs and social institutions, work together and are funded in such a way that supports this approach.[6] This multi-level agreement fosters two important outcomes: collaboration and synchronisation of innovation investments, and good governance and transparency of public money through evaluation and monitoring tools.

Collaboration and synchronisation: Smart specialisation addresses a key challenge for regional investment: “the need to synchronise, on one hand, the national innovation strategies with the regional strategies, and on the other hand, the different regional strategies among themselves to make them consistent with each other.”[7] With a limited pool of investment money, synchronising this spend means different places work together for a common development goal rather than fighting each other for a slice of the investment pie. This working together also means that when economic activity within a functional economic region and regional administrative boundaries don’t align duplication and waste can be minimised. This requires that leading and lagging regions[8] (and even leading and lagging firms) see themselves as part of a wider innovation strategy. A smart specialisation strategy, for example:[9]

transforms less advanced regions into good followers: a region in transition which is building capabilities and is agglomerating knowledge resources in a certain domain of application, so that it will be able to capture knowledge spillovers from the leaders (those who are inventing the basic technology), to attract further knowledge assets and to develop an ecosystem of innovation with the prospect and the realistic hope of becoming a leader!

While individual interventions may be regional in nature the investment logic needs to be national and multi- level in structure. “Multi-level” governance co-ordination is required to synchronise both the horizontal policy foundations to the vertical or place based initiatives.[10]

Monitoring and evaluation: Secondly, clear goals enable evaluation and monitoring to occur over the policy cycle. As academics Philip McCann and Raquel Ortega-Argilés outline, these “monitoring and evaluation aspects have to be built into the policy design right from the beginning.”[11] This limits the ability of politicians, firms, and communities to seek rents, and helps governments to abandon support for non-compliant activities or activities that fail to meet their objectives.[12] [13]

McCann and Ortega-Argilés advocate that the following mix of evaluation types be used across the timeline of an intervention:

  • Ex-ante evaluations: these occur at the start of the policy implementation process, often as part of the decision-making process for the intervention. They can set in place targets, conditionalities (regarding when and how funding will be distributed), expected outputs from the intervention and how the implementation will be evaluated over its life cycle.
  • Interim and ongoing evaluations: give insight into how the policy is “progressing over time” and “help to manage the intervention and to ensure that there is warning if targets are not going to be met.”[14]
  • Terminal evaluations: are carried out when the intervention ends and aim to ensure the institutional memory surrounding the implementation of the intervention is captured. These can be qualitative and quantitative, and once again, aid in implementing similar interventions in different places or over time.
  • Ex-post evaluations: take place after the intervention has ended with a view to giving “a more detailed view of the impact of particular measures and whether the actions delivered the expected results effectively and efficiently.”[15] Knowing what has worked and what has not worked helps to better design future interventions.

The monitoring and evaluation process affords those making governance decisions the “opportunity to learn from good practice examples but also failures.”[16] This moves the risk-taking dial from one that punishes risk-taking decision makers if any failure occurs, to one that would only punish risk-takers if repeated failures for the same reason occur. This allows policy makers and decision makers to become part of a wider learning strategy.

Entrepreneurial discovery

Entrepreneurial discovery dictates that entrepreneurs within local communities should play the central role in making investment decisions. Rather than the government “picking winners,” decisions should instead be “informed by the local knowledge, ‘of time and place’, and the entrepreneurial knowledge about opportunities, constraints and challenges.”[17] Because government doesn’t have a crystal ball by which it can discern the future, their role under smart specialisation is one of coordination: to bring together local knowledge that is dispersed, decentralised, and sometimes hidden.[18] As described by the OECD, entrepreneurial discovery policy includes:

  • rewarding entrepreneurs who discover new activities or so that agglomeration or scale effects materialise;
  • building intra- and inter-regional linkages and co-operation (e.g. staff exchange, participation in conferences, joint research programmes) to improve absorptive capacity and to facilitate the mobility of the highly skilled;
  • supporting experimentation as a way to incentivise firms to reveal and share information on their experiments and explorations; and
  • educational programmes aimed at updating people’s skills.[19]

Part of the above is also about the government helping to create and harness knowledge spillovers, where the knowledge capital of one place “spills over” into complementary activities in others.[20] To promote innovation, governments encourage the formation of “networks of complementarity and comparative advantage” where firms and systems create and absorb new ideas or ways of working together.[21] As can be seen in the box on the next page on knowledge spillovers, fostering something new often occurs when different actors in the economy and society are brought together in a way that enables these spillovers to occur.

Knowledge spillovers explained

There are four major theories of knowledge spillovers, illustrated in Figure 1. The solid arrows highlight direct market relationships, and the dotted lines represent potential types of knowledge spillovers, labelled 1-4.

Source: Author adaptation from Richard Grey et al. (2005)[22] and Stuck et al. (2016)[23]

  1. Within-industry financial spillovers occur when innovations by firms upstream or on the supply chain financially benefit downstream firms in the same industry. Firms will tend to cluster to take advantage of the above- normal profits gained through this process. In the example above, a grass research firm innovates, resulting in cheaper grass production by the grass producers, which then results in lower milk input prices that benefit the milk firms in the downstream industry. Here, the milk firms still sell the output product for the same price as they previously did but enjoy the spillover benefit of cheaper input prices without having borne the cost of innovation. These are known as pecuniary “Marshall, Arrow and Romer” (MAR) spillovers in the literature.[24]
  2. Within-industry non-financial spillovers occur when firms in the same industry are able to exploit a shared knowledge pool. The more firms co-locate, the more likely these spillovers will take place. In the example above, workers from one grass-growing firm will flow to another (and vice versa), taking their expertise and knowledge with them. In the literature these are known as non- pecuniary or non-financial MAR spillovers.[25]
  3. Across-industry spillovers occur when firms from different industries learn from one another. In the example above, innovations in local pest control in the grass-growing industry may spill over into the grape-growing industry which shares the same pest problems. In New Zealand, carbon fibre hull and sail technology from ship-building is now being used for fuel tanks in the space rocket industry.[26] These are known as Jacobian spillovers, named after Urban Studies expert Jane Jacobs who showed how the automobile industry learned heavily from the shipbuilding industry.[27]
  4. Across-society spillovers occur when the different parts of the regional economic environment work together to form what is known as known as a Regional Innovation System (RIS). Unlike the previous MAR and Jacobian spillovers that occur within firms or across industries, a RIS “focuses on the effectiveness of the wider regional environment to generate synergies…”[28] These synergies that are illustrated as number four above comprise how well entrepreneurs and firms interrelate within the wider regional system of institutions, including:
    • Education institutions: higher and secondary education, including industry training organisations;
    • Workforce institutions: Work and Income, Immigration New Zealand, WorkSafe New Zealand, unions, and employer organisations;
    • Public and private funding organisations: publicly-funded Research and Development grants, university research funding, and philanthropic private funders of research;
    • Regulatory and political institutions: local government, central government, economic development agencies, and legal institutions; and
    • Social and cultural institutions: local iwi, hapū, charities, and not-for-profit organisations.

For example, participants in a RIS will be seeking to align education and industry training to local firm needs and then allowing others to learn from this process and build on the networks needed for this to occur. Similarly, firms will be seeking to engage with social and cultural institutions to enable better outcomes for their employees and build relationships that enable knowledge and innovation diffusion to and from other firms.

Granular interventions

The third element of a smart specialisation strategy is a focus on granular interventions: investing in innovative “activities” rather than industries or sectors. As outlined by Professor Foray, “smart specialisation is not a planning doctrine that would require a region to specialise in a particular set of industries… …but instead asks questions about whether a region would benefit from and should specialise in certain R&D and innovation projects….”[29] While investing in industries or sectors like forestry or tourism is too broad and investing in particular firms too narrow, smart specialisation finds the “granular” middle ground by directing funding to groups of firms or research partners doing something new.[30] Trialling something new and “hitherto unknown” as Professor Foray puts it, is a key concept for smart specialisation,[31] as it helps to avoid the risks and inefficiencies of top- down investment. This approach promotes both regional diversity (exploring areas for research and development that form a region’s comparative advantage) and the creation and development of knowledge chains and spillovers across the economy. For example smart specialisation would not dictate a region invest in the “rocket industry” or the “boat-building industry,” but instead focus on supporting clusters of firms or even a single firm to explore the potential of composite materials in their area of expertise, be it building rockets, building boats, or growing trees.

Further to this, Foray also outlines how regions should invest in activities that help push them toward diversification and structural change by following the principle of related variety.[32] Related variety means investing in activities “where innovative projects complement existing productive assets,”[33] but also diversifying and linking the region to knowledge that exists outside the region. So, taking composites once more, if a region has a natural advantage in the fishing industry, it should invest in exploring how composites or other innovations like nano-technology could be applied within that industry. Again, this would rely on strong knowledge chains and collaboration between firms and regions.

Time-limited funding

Finally in regard to funding, there is the need at the outset for “‘a self-destruction mechanism’ associated with all priorities, through which support will be discontinued after a certain period of time has elapsed.”[34] This aspect of smart specialisation is a necessary measure to ensure that investment continues to be focussed on activities that are new and innovative. Even successful interventions should cease to receive smart specialisation funding after a few years, for example, as what they are doing is no longer new—there are no exceptions to this rule.[35] This helps minimise the potential for firms to become dependent on external funding and keeps investment on the cutting edge.

This is an extract from Julian’s research paper “A Risk Worth Taking | Ensuring the Provincial Growth Fund is Fit For Purpose” Policy Paper. (Released 2019) 






[1] Erik Berglof et al., ‘Transition Economics Meets New Structural Economics’, Journal of Economic Policy Reform 18, no. 3 (3 July 2015): 217, https://doi.org/10.1080/ 17487870.2015.1018691.
[2] A conditionality in this context means that any funding is conditional on the initiative being part of a smart specialisation strategy. See Dominique Foray, ‘The Economic Fundamentals of Smart Specialisation’, Ekonomiaz. Revista Vasca de Economia 83, no. 02 (2013): 55–82.
[3] Foray, 56.
[4] Foray, 65.
[5] David Bailey, Paul Hildreth, and Lisa De Propris, ‘Mind the Gap! What Might a Place-Based Industrial and Regional Policy Look Like?’, in New Perspectives on Industrial Policy for a Modern Britain, ed. David Bailey, Keith Cowling, and Philip Tomlinson (Oxford University Press, 2015), 287–308, https://doi.org/10.1093/ acprof:oso/9780198706205.003.0015.
[6] The Innovation Policy Platform, ‘Multi-Level Governance’ (World Bank), accessed 14 December 2018, https://www.innovationpolicyplatform.org/content/multi- level-governance.
[7] OECD, ‘Innovation-Driven Growth in Regions: The Role of Smart Specialisation’, 41.
[8] Lagging regions are regions with below average growth, incomes or productivity measures. Lagging firms are firms that are not operating at the technology frontier for their industry.
[9] Foray, ‘The Economic Fundamentals of Smart Specialisation’, 66.
[10] Bailey, Hildreth, and Propris, ‘Mind the Gap! What Might a Place-Based Industrial and Regional Policy Look Like?’, 17.
[11] Philip McCann and Raquel Ortega-Argilés, ‘Smart Specialisation, Entrepreneurship and SMEs: Issues and Challenges for a Results-Oriented EU Regional Policy’, Small Business Economics 46, no. 4 (April 2016): 9, https://doi.org/10.1007/s11187-016-9707-z.
[12] OECD, ‘Innovation-Driven Growth in Regions: The Role of Smart Specialisation’, 15.
[13] McCann and Ortega-Argilés, ‘Smart Specialisation, Entrepreneurship and SMEs’, 9.
[14] McCann and Ortega-Argilés, 8.
[15] McCann and Ortega-Argilés, 8.
[16] OECD, ‘Innovation-Driven Growth in Regions: The Role of Smart Specialisation’, 47.
[17] Foray, ‘On the Policy Space of Smart Specialization Strategies’, 1433.
[18] Foray, 1433.
[19] OECD, ‘Innovation-Driven Growth in Regions: The Role of Smart Specialisation’, 45.
[20] Wood, Taking the Right Risks, 14.
[21] OECD, ‘Innovation-Driven Growth in Regions: The Role of Smart Specialisation’, 18.
[22] Richard Gray, Stavroula Malla, and Kien Tran, ‘Pecuniary, Non-Pecuniary, and Downstream Research Spillovers: The Case of Canola’ (XIth Congress of the EAAE (European Association of Agricultural Economists), Copenhagen, Denmark, 2005), 10, http://ageconsearch.umn.edu/record/24776/files/cp05gr01.pdf.
[23] Jérôme Stuck, Tom Broekel, and Javier Revilla Diez, ‘Network Structures in Regional Innovation Systems’, European Planning Studies 24, no. 3 (3 March 2016): 425, https://doi.org/10.1080/09654313.2015.1074984.
[24] MAR spillovers are named such as they are built on the foundational theory of spillovers by economists Alfred Marshall (1890), Kenneth Arrow (1962) and Paul Romer (1986) as summarised by Edward L. Glaeser et al., ‘Growth in Cities’, Journal Of Political Economy 100, no. 6 (1992).
[25] Edward L. Glaeser, Giacomo AM Ponzetto, and Kristina Tobio, ‘Cities, Skills and Regional Change’, Regional Studies 48, no. 1 (2014): 7–43.
[26] Tim Dodd (Everyday Astronaut), Sit down Interview with Rocket Lab Founder and CEO Peter Beck, 2018, https://www.youtube.com/watch?v=Nj9BncsgvuQ.
[27] Jane Jacobs, The Economy of Cities, Vintage Books 584 (New York: Vintage Books, 1970).
[28] Steve Talbot, ‘Creating a Smart Rural Economy through Smart Specialisation: The Microsphere Model’, Local Economy 31, no. 8 (December 2016): 897, https://doi. org/10.1177/0269094216678601.
[29] Foray, ‘The Economic Fundamentals of Smart Specialisation’, 55, 56.
[30] Foray, 57,59.
[31] Wood, Taking the Right Risks, 14.
[32] “Related Variety is an evolution and transformation of regional productive assets” based on the work of Koen Frenken, Frank Van Oortand Thijs Verburg, ‘Related Variety, Unrelated Variety and Regional Economic Growth’, Regional Studies 41, no.5 (July 2007): 685-697, accessed 20 February 2019, http://dimetic.dime-eu. org/dimetic_files/Boschma_Regional-Economic-Growth.pdf, cited in Foray, ‘The Economic Fundamentals of Smart Specialisation’, 60.
[33] Foray, 60.
[34] Foray, ‘On the Policy Space of Smart Specialization Strategies’, 1434. Different types of investments may well have different sunset timeframes and the cessation of smart specialisation funding might not preclude alternative forms of funding becoming available dependent on the type of investment.
[35] Foray, ‘The Economic Fundamentals of Smart Specialisation’, 64.

Post Tags:
Julian Wood

By Julian Wood -

Like what you're reading?
Share it around.

Or just highlight the part you like...

Want to know more about Maxim Institute and what we do?

Find out more


FORUM (monthly eNews)Event Invitations