Digital meets smart: towards a technology-enhanced approach to Smart Specialisation Strategy development

ABSTRACT

Conceived in the framework of regional studies on Smart Specialisation Strategy (S3) development, this special issue strengthens research efforts oriented towards assembling a technology-enhanced approach to S3 policymaking. First, it sheds light on fundamental methodological limitations that affect S3 development and reports on the digital support tools currently available. Second, it offers a knowledge base for producing novel digital applications that align with the overlooked needs of S3 orchestrators. Third, it complements this practical input with theoretical advancements in regional innovation studies by addressing intellectual questions and policy issues of pivotal importance for the S3 debate.

KEYWORDS:

• Smart Specialisation
• regional innovation
• policy development
• methodological gaps
• digital technology
• European regions

JEL:

• O33
• O38
• R11

1. INTRODUCTION

In March 2000, the European Council set a new strategic goal: to make Europe ‘the most dynamic and competitive knowledge-based economy in the world, capable of sustainable economic growth with more and better jobs and greater social cohesion, and respect for the environment’ (Rodriguez et al., 2010, p. 11). This goal is at the core of the Lisbon Strategy and affirms the political ambition and commitment of the European Union’s (EU) member states to undertake the structural improvements needed to support the transition to a knowledge-based economy and society (European Council, 2000).

To accelerate this transition, the European Commission set up the Knowledge for Growth (K4G) Expert Group. Comprising European and American economists, this group operated as an independent advisory body and provided high-level recommendations on how to develop the research and innovation policies required to move Europe towards a competitive knowledge-based economy. Although the K4G Expert Group is no longer active, its recommendations are still embedded in a series of reports and policy briefs that were published between 2005 and 2009 (David & Metcalfe, 2007; Foray, 2006; Foray et al., 2009; Foray & Van Ark, 2007; Giannitsis & Kager, 2009; Hall & Mairesse, 2009; Marimon & de Graça Carvalho, 2008; O’Sullivan, 2007; Veugelers & Mrak, 2009). According to these publications and their recommendations, EU countries are required to address a set of key policy challenges to pave the way for a competitive knowledge-based economy: reducing the deficit in research and development (R&D) activities and innovation capacity, establishing a stronger governance framework for sustaining science and technology systems in a globalized world, strengthening the interrelation between technology production and diffusion, and reinforcing the relationship that higher education institutions have with industry. In addition, these advisory documents introduce the concept of Smart Specialisation, which emerges as one of the leading ideas of the K4G Expert Group (Foray et al., 2009; Foray & Van Ark, 2007).

The central premise of the group is that Europe is losing ground as a centre for research and innovation (European Commission, 2008); consequently, EU regions must create ‘R&D hubs which can compete with foreign hubs to attract more research capacities and other knowledge resources’ (p. 13). To support these developments, the K4G Expert Group calls for national and regional stakeholders across Europe to engage in the so-called Smart Specialisation process, which entails the identification and subsequent development of the most promising research and innovation domains through a prioritization logic. These domains are considered areas of specialization, and their identification is based on a process of entrepreneurial discovery: an evidence-based, bottom-up, place-based and government-led collaborative learning process that brings local entrepreneurial actors together. In these collaborative ecosystems, entrepreneurial actors form mutually reinforcing connections and pool their knowledge to collectively identify the strategic areas of development that can best sustain growth within EU countries and regions (del Castillo Hermosa et al., 2015; Foray et al., 2009; Martínez-López & Palazuelos-Martínez, 2019; Mieszkowski & Kardas, 2015; Periáñez-Forte et al., 2016; Santini et al., 2016). Entrepreneurial actors can be members of any organization, such as universities, industry, government or civil society; their entrepreneurial knowledge allows public sector organizations and innovation stakeholders to acquire a systematic understanding of ‘the most promising areas for future regional development’ (Foray et al., 2012, p. 12).

Recent developments in regional studies describe Smart Specialisation Strategies (S3) as ‘prioritisation agendas for regional innovation policy’ (McCann & Ortega-Argiles, 2013a, p. 206), which result from embedding the debate on non-spatial innovation policy into European Cohesion Policy (McCann & Ortega-Argiles, 2013b). Moreover, they show that the paradigm shift underpinning S3 lies in the combination of two key elements. First, a bottom-up and place-based approach to regional innovation paths, which ‘strongly depend on territorial elements rooted in the local society, its history, its culture, and its typical learning processes’ (Camagni et al., 2014, p. 72). Second, a policy-prioritization logic that ‘explicitly avoids automatically prioritizing high-technology sectors by taking a broader systems perspective’ (McCann & Ortega-Argilés, 2015, p. 1293).

The European Commission has identified S3 adoption as a main priority for successfully attaining smart, sustainable and inclusive growth across Europe (European Commission, 2010a, 2010b). In enacting this European growth agenda, the Council of the European Union formally endorsed a set of common rules for governing the European Structural and Investment Funds (ESIF) (European Commission, 2014b). This new legislative framework defines S3 as:

the national or regional innovation strategies which set priorities in order to build competitive advantage by developing and matching research and innovation […] strengths to business needs [and by addressing] emerging opportunities and market developments in a coherent manner while avoiding duplication and fragmentation of efforts.

(European Union, 2013, p. 338)

In addition, this legislative framework has also proposed that S3 development by member states is a prerequisite for any future allocation of structural funds related to research and technological development.

This requirement has triggered the scientific debates on Smart Specialisation and pushed national and regional governments to address the design and implementation challenges posed by S3 development. However, hopes of developing these large-scale innovation policies into instruments able to deliver a ‘true economic renewal’ (Boschma, 2014, p. 64) have been undermined by the very limited theoretical and practical understanding of how S3 development should be approached (Balland et al., 2019; Capello, 2014; Capello & Kroll, 2016; Gianelle et al., 2016; Kroll, 2015). As Foray (2015) pointed out, Smart Specialisation provides a classic example of where ‘policy runs ahead of theory’ (Varga et al., 2020, p. 48); in strategic terms, this gap has left many policy design and implementation issues unresolved.

To overcome this shortcoming, the European Commission has released a set of guidance notes and a handbook (Foray et al., 2012; Gianelle et al., 2016) that set out how policymakers should manage S3 policy design and implementation. However, the insights these documents offer have proven insufficient to advise national and regional innovation actors on optimal S3 development (Balland et al., 2019; Cooke, 2012; Iacobucci & Guzzini, 2016) – especially with respect to translating S3 principles and methodologies into the actionable knowledge needed to foster scientifically driven decision-making. Consequently, ‘the challenges, strengths, and risks associated with the best design and implementation of [S3] are still much debated’ (Capello, 2014, p. 5), and a comprehensive set of supporting tools fostering an evidence-based approach to S3 development is lacking (Balland et al., 2019).

After the S3 guide and handbook were released in 2013 and 2016, respectively, several studies highlighted their shortcomings (Balland et al., 2019; Iacobucci, 2014; Iacobucci & Guzzini, 2016; Komninos et al., 2018a). For example, an analysis of S3 developed by a group of Italian regions exposed the limited consideration given to the relatedness of technological domains, both in terms of the potential intra-regional links between specialization domains and the research and innovation priorities of other European regions. The guide and handbook recommend that European regions conduct systematic analyses that capture the presence of connectivity between technological domains. However, a theoretically sound methodology is still missing (Iacobucci & Guzzini, 2016). Similarly, the investigation conducted by Balland et al. (2019) into the ‘diversification dilemma’ (p. 1254) – which builds on Boschma and Gianelle’s (2014) note on regional diversification – reflects upon the importance of combining technological relatedness, knowledge complexity and diversification of technological trajectories in S3 development, recognizing that regional and national governments struggle to embed these critical components in their strategies.

Additional examples of practical issues currently challenging the design and implementation of Smart Specialisation policy statements include: measuring the economic impact of interventions (Varga et al., 2020); managing regional cooperation between organizations, which represents ‘an important determinant of regional diversification’ (Santoalha, 2019, p. 1269); shaping and effectively governing the multi-actor, multi-sectoral and multilevel collaborative environments where entrepreneurial discovery processes take place (Crescenzi et al., 2020); aligning the views and expectations of these heterogeneous collaborative environments with insights sourced from the analyses of regional contexts and the research and innovation potential they possess (Kleibrink & Magro, 2018); evaluating whether the specialization areas that are selected represent those with the greatest potential to sustain regional growth (D’Adda et al., 2019; Muscio & Ciffolilli, 2018); understanding how governance approaches to S3 development adapt to local conditions and geographical differences (Sörvik et al., 2019); and mapping regional assets to identify opportunities for innovation through existing and emerging activities (Foray et al., 2012).

In addition to highlighting the relative under-development of S3 research, these studies have been instrumental in emphasizing the difficulties that national and regional innovation actors experience when attempting to transition into the evidence-based, bottom-up and place-based approach to policymaking that S3 formulation requires. Moreover, they demonstrate that effective S3 policy design and implementation should be considered as the outcome of well-coordinated interactions between heterogeneous groups of stakeholders, whose collaborative decision-making can be made more efficient by leveraging digital technology – especially big data collection methods, advanced computational capacities and large-scale visual data exploration (Etzkowitz & Ranga, 2010; Komninos et al., 2018b; Sacco, 2017).

2. STRENGTHENING THE SYNERGY BETWEEN S3 DEVELOPMENT AND DIGITAL TECHNOLOGY

A growing body of literature acknowledges the key role that advanced technological solutions – particularly digital tools – can play in helping frame these complex interactions and overcoming the methodological gaps currently undermining the evidence-based, highly collaborative and place-sensitive nature of S3 development (European Commission, 2017; Fabińska, 2018; Griniece et al., 2017a; Kleibrink et al., 2014; Komninos et al., 2018b; Mackiewicz et al., 2018; Mariussen et al., 2016; McCann & Ortega-Argilés, 2016; Ortiz Cebolla & Navas, 2019; Özbolat & Harrap, 2018; Sörvik & Kleibrink, 2016). However, while calling for a technology-enhanced approach to S3 policymaking, this literature clarifies that significant research efforts are still needed to assemble it.

Recognizing that investments in digital technologies and data infrastructure assets can help governments enhance the quality of their policymaking processes (European Commission, 2014a), the European Commission (2016) has committed to boosting the digital transformation of government functions and tackling the problematic fragmentation generated by the independent attempts of EU member states to modernize public administrations. Developing a shared portfolio of digital solutions to common policymaking issues has been perceived as pivotal to the construction of a technology-enhanced approach to regional policy formulation in Europe (European Commission, 2015). Accordingly, the European Commission has invested significant resources to support the development of digital tools that regional and national governments across Europe can freely implement during S3 development.

Some of these digital tools have been released through the Smart Specialisation Platform. Hosted by the Joint Research Centre (JRC) in Seville, Spain,¹ this platform provides European countries and regions with policy advice, methodologies, and supporting tools for S3 formulation and implementation. These tools include a set of web applications that facilitate the identification, selection, and prioritization process of specialization areas while supporting interregional cooperation and the creation of strategic partnerships. Examples of tools include: the Eye@RIS3 database, which offers an overview of EU regions’ priorities, helping them ‘find their unique niches and to seek out potential partners for collaboration’ (European Commission, 2014c, p. 63); a monitoring tool focused on the information and communication technology (ICT) sector, which allows users to search ESIF data on ICT-related investments at the regional level (Sörvik & Kleibrink, 2016); a regional benchmarking tool jointly developed with Orkestra – the Basque Institute of Competitiveness (Fabińska, 2018; Kleibrink et al., 2014); the EU Trade Tool, a web-based application for the analysis and visualization of interregional trade flows (Mariussen et al., 2016); and the R&I Regional Viewer, which compares research and innovation investments in EU regions (Özbolat & Harrap, 2018).

By partnering with the Directorate General for Regional and Urban Policy (DG REGIO) of the European Commission, the JRC has also collaborated in launching the Urban Data Platform. This online platform provides access to numerous spatial indicators and facilitates their visualization and analysis. If correctly integrated into the S3 process, this digital tool can help decision-makers, policy analysts and other regional innovation stakeholders obtain the contextual knowledge needed to support the identification of Smart Specialisation areas.²

Additional digital support tools for S3 development have also emerged from EU-funded research projects such as OnlineS3, SSST-BD and the open innovation platform developed by the Lombardy region, Italy. Built in the framework of a multimillion-euro Horizon 2020 project, OnlineS3 is a web-based environment with 29 digital applications designed to help national and regional governments manage the main phases of S3 development processes (Komninos et al., 2018b). With a focus on big data foresight tools, SSST-BD has produced new automated knowledge-based management tools for S3 management operations.³ Finally, by leveraging a €5.6 million investment, the Lombardy region has assembled an open innovation platform that functions as a regional forum on research and innovation. After selecting one of the S3 areas that the region has expressed an interest in, users can deploy the suite of digital tools that the platform offers to discuss regional development priorities and collaborate in generating project proposals. Used by over 7000 members of the local research and innovation community, this platform has become a critical pillar in the S3 governance system of the Lombardy region and has already supported the development of more than 200 project proposals.⁴

In addition, Griniece et al. (2017b) have mapped hundreds of possible digital systems and tools that show potential for S3 development enhancement. Most of these digital solutions build on open-source software and can be freely accessed to unlock advanced functionalities, such as big data analytics, real-time data collection, crowdsourcing elements, and advanced visualization and data mining techniques, helping national and regional innovation actors cope with S3 design and implementation tasks.

3. AIM OF THIS SPECIAL ISSUE AND AN OVERVIEW OF ITS CONTRIBUTIONS

This special issue facilitates the understanding of how to assemble the technology-enhanced approach to S3 development championed by Europe over the past decade. Despite the many digital applications currently available to support S3 development and the significant investment of the European Commission in such technologies, their practical use remains limited (Gianelle & Kleibrink, 2015; Griniece et al., 2017b; Kleibrink & Magro, 2018; Komninos, 2014; Komninos et al., 2018a; McCann & Ortega-Argilés, 2016). Academic research acknowledges this gap but struggles to provide a clear understanding of how the array of stakeholders involved in S3 development can take advantage of available digital applications and systems. Moreover, the S3 literature is unable to explain whether the digital tools currently available meet the requirements of the sector or if new solutions that provide additional functionalities need to be developed.

The six articles in this special issue respond to the call for an improved understanding of how digital technology can sustain S3 policy design and implementation. First, they shed light on some fundamental limitations that affect currently available digital tools for S3 development. Second, with empirical analyses and experiments in different European regions, they generate the knowledge base required to produce novel digital applications that meet overlooked needs of S3 orchestrators. Third, in addition to this practical input, these articles also contribute to advancing theory in regional innovation studies by addressing intellectual questions and policy issues of pivotal importance within the Smart Specialisation domain.

In the first article, Mikko et al. (2022, in this issue) present an unsupervised machine-learning technique that regional innovation stakeholders can adopt during S3 policy design phases. This technique is built on the concept of cognitive proximity and based on a topic-modelling algorithm whose functioning has been tested by using Arctic Scandinavia as an application case. Expanding the set of currently available digital tools for S3 development, the authors show the potential of their digital application to facilitate the formulation of shared visions and the identification of potential Smart Specialisation areas – especially in the framework of cross-border innovation, where they expose the critical need for mutual understanding and the creation of common languages required by cross-border innovation systems.

In the second article, Natalicchio et al. (2022, in this issue) present a new patent-based methodology for examining regional technological capabilities. Like the initial contribution, among the objectives of this methodology is the need to help European regions detect technological competitive advantages and opportunities for knowledge recombination, assess S3 priorities against regional innovation performance measures, and conduct benchmarking activities to enhance cross-border innovation mechanisms. However, this methodology leverages the latent knowledge embedded in patent data, whereas the machine-learning technique proposed by Mikko et al. (2022, in this issue) rely on research data repositories.

With the third article, Panori et al. (2022, in this issue) propose a four-stage methodology that combines technological relatedness measures with regional sectoral characteristics. Patent data are introduced once again to measure technological relatedness, which becomes a prioritization tool. Furthermore, cognitive capabilities are used to support S3 development, in a methodology which makes it possible to expand ‘existing theoretical approaches that try to enhance spatially independent relatedness spaces with regional perspective measures’.

In the fourth article, Ruhrmann et al. (2022, in this issue) reposition the Mikko et al. (2022, in this issue) (Leydesdorff & Park, 2014) in the framework of S3 development processes. By examining the German context, the authors showcase the capability of this indicator to function as a quantitative, data-driven tool for assessing synergy levels in regional innovation systems. Building on lessons learned while observing the extensive decentralization of German innovation systems, Ruhrmann et al. invite policymakers to ensure that S3 accounts for decentralized innovation activities, and they provide the following recommendations: pay more attention to multilevel governance mechanisms and the coordination–autonomy balance, intensify cross-boundary collaborative efforts in S3 development to try to reduce inequalities among neighbouring regions, and ensure ‘that research and innovation communities in the regions can effectively coordinate above the regional level on mission-oriented innovations’.

Similar to Natalicchio et al. (2022, in this issue), who establish methodological foundations for the development of a novel set of advanced digital solutions for S3 policymaking, the fifth article by Rocchetta et al. (2022, in this issue) focuses on the non-linear effect of technological diversification on regional productivity. However, their attention shifts from synergy to entropy and coherence; the authors combine entropy-based measures of technological variety – a measure of technological co-occurrence – and multilevel modelling to jointly analyse the impact of technological diversification and diversity on European regional productivity.

Finally, the special issue closes with Barbero et al. (2022, in this issue), who propose that RHOMOLO – a dynamic and multi-regional computable general equilibrium model developed by the JRC – is introduced in S3 evaluation processes. The authors report on the first application of the RHOMOLO model for the ex-ante evaluation of S3. Their results highlight great potential for creating ‘meaningful models leading to estimations that can be a realistic benchmark against which to compare the reality of the intervention’.

Within the theoretical and practical contributions that these six articles collectively offer, knowledge and skills development, data access, and scale-up feature prominently and emerge as some of the most critical thematic areas, opening paths for future interdisciplinary research. Whether Europe will succeed in assembling a technology-enhanced approach to S3 policymaking strongly depends on the capability of regional innovation scholars to capture the needs of S3 developers and translate such demands into appropriate technological solutions. This translation process will require interdisciplinary efforts linking social sciences to engineering and technology subjects. The articles in this special issue contribute to showing that some high-priority needs have been overlooked. As a result, more efforts are needed to continue transforming these searches and translation stages into technological developments.

Connecting the demand to a proper offer, however, is insufficient; practical input and theoretical advancements are also needed in the framework of capacity-building for region-specific institutional settings, where local authorities may not be equipped to adopt digital tools for S3 development. To varying extents, all articles in this special issue implicitly contribute to unveiling this important requirement. The methodological approaches and digital tools that our authors have introduced, together with the existing offering, help frame a technological environment whose potential to enhance S3 policymaking depends upon other socio-technical changes. If improperly planned, pushing for digital transformations in current policymaking practices may cause more problems than solutions.

To make the most of this digital transformation process, there are critical questions that regional studies, as an interdisciplinary field of research, will need to help address. For example, recognizing the leading role of public sector organizations in S3 development and the common implementation challenges they face when approaching digital innovation, how can barriers such as knowledge and skills gaps, legal restrictions, and financial constraints be overcome in the context of new digital technologies for evidence-based policymaking (Demircioglu & Audretsch, 2017)? How can these technologies be integrated within existing legacy systems and siloed public administrative practices (Roberts, 2011; Esposito et al., 2021)? How do restrictions in the right to access and reuse commercial databases affect the capability of government officials to implement digital solutions to S3 development challenges (Himanen et al., 2019)? How can we move from experimental digital practices to the scale-up operations required to trigger wider adoption (Bason, 2010; Mora et al., 2021)? And how can we ensure that disadvantaged European regions are not left behind in this technological evolution (Matteucci, 2020)? Considering this uncertainty, when looking at the current state of the technology-enhanced approach to S3 development championed by European regions, we cannot help but wonder what hides beneath the water: undoubtedly, as of today, we are only seeing the (technological) tip of the iceberg.

ACKNOWLEDGEMENTS

We thank Professor Arnoud Lagendijk and the editorial office of Regional Studies for their continuous support throughout the editorial process. This special issue would have never been completed without their help, which has been as invaluable as the commitment of the reviewers and authors who have contributed to transforming our abstract idea into a concrete outcome. We are extremely grateful for all their efforts.

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the European Commission through the Horizon 2020 project FinEst Twins [grant agreement number 856602].

Notes

1. JRC is the science hub of the European Commission. Its main mission is to conduct independent research and generate scientific knowledge that can be leveraged to support European policy development.

2. See https://urban.jrc.ec.europa.eu/.

3. See https://cordis.europa.eu/project/rcn/221614/factsheet/en/.

4. For additional information, see https://www.openinnovation.regione.lombardia.it/.