Smart specialization

Smart specialisation is an important concept for better and more targeted innovation policy. It developed as a result of analysis that found that research and innovation efforts in Europe are often too fragmented and subcritical, and tend to duplicate one another too much. Smart specialisation was included in post­2013 Cohesion Policy as an ‘ex ante conditionality for European Structural and Investment Fund support’. This meant that all EU Member States and regions had to develop National and Regional Research and Innovation Strategies for Smart Specialisation (RIS3 strategies). These strategies set out the agendas for integrated, place­based economic transformation that would guide Member States’ and regions’ innovation­related investment. They: • focus policy support and investment on key national/regional priorities, challenges and needs for knowledge­based development; • build on each country/region’s strengths, competitive advantages and potential for excellence; • support technological as well as practice­based innovation and aim to stimulate private sector investment; • get stakeholders fully involved and encourage all forms of innovation and experimentation; and • are evidence­based and include sound monitoring and evaluation systems. Methodological guidance for policy­makers and implementing bodies on how to prepare for and design, draft and implement a National or Regional Research and Innovation Strategy for Smart Specialisation (RIS3) is provided by the Guide to Research and Innovation Strategies for Smart Specialisation, available from the Smart Specialisation Platform at: http://s3platform.jrc.ec.europa.eu/­/guide­on­research­and­innovation­strategies­for­smartspecialisation­ris3­guide

Clusters are important both as the building blocks of designing smart specialisation strategies and as a means for implementing these strategies. The evaluation of the previous European Regional Development Fund and Cohesion Fund programming period has, for example, shown that the promotion of networking through clusters has been ‘among the
 most successful instruments’ for supporting innovation in SMEs and nurturing their development, even if instruments of this type represented only a small proportion of those introduced during the programming period.11 Both cluster policies and smart specialisation strategies are important tools for generating growth in the EU by unlocking new business opportunities for SMEs in new value chains and creating synergies through interregional
 cooperation.

Clusters are not synonymous with specialisation. While clusters reflect the specialisation of locations in specific economic activities, they capture two important additional aspects. First, clusters reflect specialisation in groups of related industries, not just one narrow activity. Much of the dynamism of clusters derives from these ‘economies of scope’ rather than static ‘economies of scale’. Clusters are thus much more a reflection of cross­industry linkages and ‘related diversification’ than of narrow specialisation. Second, clusters reflect the dynamic interaction between a large number of firms. Specialisation through the presence of one large firm or plant does not constitute a cluster. Clusters and cluster initiatives are not merely networks. Clusters are geographic concentrations of firms active in related fields. Some of the linkages that exist between these firms and that contribute to the dynamism of the cluster result from the firms being in networks; others, such as access to a common labour market, do not. Networks, especially if they are formalised through an organisation, are more similar to cluster initiatives. But cluster initiatives not only connect firms, they organise joint actions among them and provide services to firms in the cluster. Networks, conversely, are not bound to any particular geographical location in the way cluster initiatives are. They are often a ‘closed shop’ of firms and other institutions that collaborate on a specific project or in a narrow field.13 Clusters and cluster initiatives are an important part of innovation ecosystems. Innovation eco­systems are similar to clusters, but do not have the same focus on specific sets of related industries. They tend to encompass all activities in a given location that are connected to innovation. In practice, this perspective can easily lead to a focus on research­driven innovation and the linkages between academia and business. It lacks the specificity of a distinct cluster, and often tends to have more of a supply­driven rather than a market­driven perspective. 2.4 What is the economic  relevance of clusters and cluster policy? Research in the area of clusters has over the last decade and a half become much more quantitatively oriented. ‘Cluster­mapping’ initiatives such as the European Cluster Observatory and similar projects in the US and other countries have created a transparent, data­rich foundation on the basis of which to evaluate the economic size and impact of clusters. Increasing attention has also been given to the need to evaluate and demonstrate the impact that cluster policies and programmes can have. This section discusses some of the key facts that have emerged in relation to Europe.
2.4.1 Clusters matters Economic activities that are located in clusters account for about 39 % of European jobs and 55 % of European wages.14 The cluster mapping methodology classifies industries as either ‘traded’, i.e. industries based in agglomerations in specific locations that also serve other markets, or ‘local’, i.e. industries that are dispersed and present everywhere as they mainly serve local markets – such as local retail and other services. Local industries are therefore not considered by the cluster mapping methodology as they are viewed neither as being exposed to direct competition across regions nor as tending to cluster together. Clusters are groups of related ‘traded’ industries, connected through multiple types of local linkages. Both types of industries are important for the health of the European economy. But they play different roles and are exposed to different types of dynamics. Traded industries located in clusters are the innovation and growth drivers of an economy. They account for more than 87 % of all patents,15 report much higher productivity, wages, and productivity growth,16 and their growth potential is not limited by the size of any local market. Local industries in turn are the basis of a large and growing proportion of employment. Activities with relatively limited productivity growth such as healthcare, retail, and other local services account for a large part of the aggregate job creation in the overall economy. Their productivity performance means that they have a significant impact on local prices and thus on living standards. Europe is home to some 2 500 strong clusters,17 i.e. statistically defined regional concentrations of related traded industries that achieve above average performance for employees, firms, and regions. Cluster effects become visible when the presence of related industries in a specific location reaches critical mass. Roughly 45 % of all employment in traded industries is located in strong clusters. Employees in strong clusters earn on average 11 % higher wages than their colleagues in the same industries but located outside of clusters. This reflects the higher productivity that companies can achieve in clusters. Strong clusters have reported job growth of 0.2 % annually in the post­crisis period (2008­2014), while traded industries outside of strong clusters have lost 1.7 % on average. Research in the US has shown that new business formation is higher in strong clusters, and that new firms are more likely to succeed and grow if located in strong clusters.18 Finally, regions that have a higher proportion of their employment in strong clusters register higher overall levels of prosperity.19  The presence of clusters also shapes locations’ opportunities for structural change, a key priority of many smart specialisation strategies. Recent academic research has provided evidence that locations diversify via a process of related diversification, i.e. new activities emerge in new fields that are related to existing fields in which a location is already strong.20 The European Cluster Panorama draws on this thinking to identify emerging industries as fields of related traded industries that go beyond an individual cluster. Locations can use this data to evaluate in which of these emerging industries they have the best
opportunities for growth given their existing cluster portfolio.21 The research undertaken over the last few decades has provided increasingly robust evidence that clusters are an important feature of modern economies. It also points strongly to a positive link between the presence of clusters and the economic performance of the companies in these clusters and the regions in which they are located.
2.4.2 Cluster policies matter The argument for cluster policy must, however, move beyond these observations. Clusters emerge in a market­driven process, reflecting the decisions taken by companies as to their location and differences in success across locations. Cluster policy is motivated by the recognition that this market process is subject to market failures and is strongly affected by policy choices.22 Governments are already active in many of these areas. The motivation to organise these policies around clusters is that such an approach offers benefits in terms of effectiveness. Relative to firm­level interventions, i.e. supporting activities undertaken by individual companies, cluster­level actions can create greater leverage and reach a larger number of firms. Relative to industry­level interventions, action taken by a cluster avoids many of the distortions relating to suppliers along the value chain that otherwise often emerge. And relative to economy­wide policies, cluster­level action can be better targeted at the specific issues faced by companies in a set of related industries, thus increasing the effectiveness and reducing wasted effort. Clusters and cluster initiatives will need to play an important role in system innovations, in particular in the transition to the circular economy, but also in any other industrial transformations that radically change business models and the organisation of value chains. Many Member States have national/ regional policies and programmes in support of clusters. A survey of the European Cluster Observatory in 2008, for example, identified 69 national cluster programmes and 88 regional cluster programmes. Some countries have a long experience of 20 years in supporting cluster efforts, while others started only recently. Differences can also be seen in the amount of public financial support allocated to cluster policy and related programmes and initiatives. An updated survey carried out in 2015 by the European Cluster Observatory23 identified 16 national cluster programmes in 15 EU countries, with other countries being in the process of revising their national cluster policies and programmes and a small number not having cluster policies as such in place at national level. Some of the recent evaluations of national cluster programmes, for example, of the French Programme ‘Competitiveness clusters’ (Pôles de compétitiveté)24 (2012) and the German ‘Leading Edge Cluster Competition’25 (2014) suggest that not only are clusters beneficial, but also that investment from cluster policy and public support for clusters pays off. The 2012 evaluation of the French national initiative ‘Competitiveness clusters’, implemented between 2006 and 200926 found, for example, that the ‘Competitiveness clusters’ initiative gave a significant boost to R&D investment (EUR 2.5 billion of public support saw almost 1 500 projects launched and generated more than EUR 6.5 billion of R&D expenditure), increased collaboration between industry and research considerably and benefited SMEs (as 80% of the 9 700 firms covered by the initiative were SMEs, which also received over 50 % of the funding allocated by the Competitiveness Cluster Fund). More than six out of ten firms created
jobs thanks to the ‘Competitiveness clusters’ and 84 % maintained jobs. In addition, at least 1 000 patents and 200 start­ups were created. The 2014 evaluation of the German ‘Leading Edge Cluster Competition’ also documented the positive effects of the cluster measure on the availability of resources and the improvements it led to in the quantity and quality of human capital in the selected clusters. The initiative also had a significant effect on the density and size of networks and led to increased cooperation between SMEs and large businesses. Slight positive effects on product and process innovation could also already be observed. The cluster measure was praised by the Expert Commission for Research and Innovation (EFI) in their 2015 report on Germany’s research, innovation and technological performance.27 Both evaluations confirm that these two programmes are good examples of cluster programmes and that they provide various examples of best practice for cluster programme development and implementation. Policy­makers and programme agencies are increasingly focusing on using evaluations of cluster programmes as a way of to improve the effectiveness and efficiency of the support provided. More needs to be done though to share experiences and practices on how to structure an evaluation and what can be learned from it. In parallel, research and information on cluster initiatives and cluster organi– sations is also being advanced. The inventory of cluster organisations at the European Cluster Collaboration Platform is being made more systematic in order to allow better matching. This platform is being strengthened in order to help implement forward­looking strategies and promote more European Strategic Cluster Partnerships, as a way of unlocking the growth potential by strengthening crossregional collaboration and creating synergies in the implementation of smart specialisation strategies.

Modern cluster
 policy in practice:
 tools for action
Cluster­based economic development is a comprehensive approach to directing government policies towards creating the foundations for growth and prosperity. The alignment between modern cluster policies and smart specialisation strategies will allow countries and regions to further sharpen their strategic focus in order to drive economic and industrial transformation and competitive advantage in their priority areas. Therefore, the policy approach needs to combine different instruments and tools and both ensure that the country or region is strategically positioned in international value chains and facilitate innovation­driven diversification and the development of future strengths. Modern cluster policy includes a number of specific policy instruments, in particular the use of cluster initiatives, but takes a much broader perspective on how to design and implement ongoing policies in fields such as innovation, internationalisation, SME support, workforce development and attracting investment. A number of characteristics are particularly important:
First, the objective of economic development policies needs to be increasing competitiveness, where competitiveness is understood as the qualities of a location that enable firms to succeed in national and global markets while supporting a high standard of living for local communities. To achieve this objective, government policies need to focus on the two bases, productivity and innovation, as the key drivers that support both aspects of this ambition. Clusters are an important part of this underlying capability base. A short­term focus on, for example, lowering costs or job creation only, is not sustainable if it is not achieving this through higher productivity. Second, each location has its unique set of economic opportunities and challenges. Policies need to be aligned with these local conditions, and they need to be delivered in ways that are consistent with the realities of the location. Learning from others is possible in terms of inspiration and at a narrow technical level. But locations need to set policies that work specifically for them. Clusters are an important element of understanding the local economic base and can play a key role in identifying the entrepreneurial and innovation opportunities for a region and its potential to tap into new and emerging value chains. Third, the modern economy is characterised by multiple linkages across economic activities. Effective economic development policies recognise and leverage these linkages. They overcome the inefficiencies of policies that rely only on cross­cutting measures affecting all businesses. And they avoid firm­ and industry specific interventions that create a high degree of distortion in markets and value chains. Clusters are a useful level at which to apply government action, because they capture the strongest linkages across industries, as have been revealed in the existing patterns of economic geography. The strongest level of competition exists within clusters and can be further increased by cluster­oriented policies. Fourth, modern economic development policies require both competition, and a true partnership and effective collaboration between the private and the public sector and the many relevant institutions that exist in both. The knowledge necessary to make appropriate choices
about economic policy priorities is dispersed across these many actors, and collaboration is therefore key for having the full picture. Successful implementation requires the actions of many actors to be aligned, again driven by a dialogue and a common view of the area’s objectives. Clusters are important focal points for the relevant public­private dialogue – it is at this level that interests are most aligned, and that the interdependence of choices is the strongest. The remainder of this section discusses different tools and examples that policy­makers should consider when developing cluster policy and cluster programmes. 3.1 Clusters as a tool throughout the policy process A cluster­based approach towards economic development uses clusters and cluster data at all stages of the policy process. As a simplification, this process can be described as a cycle (see Figure 1), moving from analysis to strategy to action and then back to analysis.34  This section discusses the use of clusters within these three broad process steps.