Practical Implications of Smart Specialization Strategy: Barriers to Implementation,Role of the Public Sector,and Benefits for Entrepreneurs

Abstract

Smart specialization is one of the major public policies aimed at stimulation of innovations. The aim of this paper is to examine the interactions between implementation of innovation strategy in the form of RIS3, barriers and bottlenecks of innovation diffusion, and development of innovativeness in enterprises on a regional level. The analysis was based on 30 individual in-depth interviews with representatives of institutions implementing the assumptions of smart specialization in the regions, as well as cross-sectional survey questionnaires on a sample of 250 enterprises engaged in innovative activities in Poland sample. The results of this study show that the applied forms of innovation support, implemented within the framework of RIS3, are not adapted to the realities of the operation of enterprises. Improving this situation requires understanding the actual needs of entrepreneurs. It is proposed to increase the level of stakeholder interaction and communication. Improved measures should be tested and evaluated based on their ability to stimulate business innovation.

Keywords

innovation policy smart specialization strategy regional development innovativeness innovation development barriers public sector enterprise innovation

Introduction

The innovativeness of any country depends on many interacting factors, including demand conditions, knowledge resources, institutions, market competitiveness, and political environment (Fabrizio et al., 2017). The goal of the innovation policy is development of innovativeness in the economy. The realities of formulating any policy are always complicated, and they depend on the context and the goals being at its focus (Veldhuizen, 2020). In line with the European Commission’s development policy (Komisja Europejska, 2014) smart specializations have been introduced that offer an integrated set of principles to manage investments in innovations via setting priorities, concentration of public resources as well as mobilization of local resources and entrepreneurship potential.

To organize the terminology used in this text, it should be emphasized that smart specialization strategies are a broad concept referring to EU-wide activities, in individual regions they are implemented through research and innovation strategies for regional smart specialization (RIS3). This topic is particularly relevant due to the fact that the cohesion policy has created the largest pan-European program for financing the industrial policy, which enables application of such principles (Ahner & Landbaso, 2011). On the other hand designing a new innovation governance policy does not take place in the void, and it should account for the regional and national framework of its functioning, as well as the context of its stakeholders.

Public institutions are believed to contribute to or contain problems that are typically associated with place-based policies such as smart specialization (Grillitsch, 2016 see: Asheim i Coenen, 2006; Asheim i Isaksen, 2002; Breschi, 2000; Cooke, 2001; Edquist, 2005; Malerba, 2002, 2005; Tödtling i Trippl, 2005).Uyarra et al. (2020) suggest there is a need for more bottom-up (or focused on individual locations) approach to innovation policies, at the same time underlining the significance of structural factors in shaping the scope within which national or regional economies may develop and diversify. The European literature indicates that each region needs to have sufficient regulatory and administrative capabilities to enable the implementation of the required policy frameworks (Weller & Rainnie, 2022). This topic is particularly relevant given that the smart specialization framework covers all EU regions, and its implementation is one of the prerequisites for accessing structural funds from 2014.

The process of identifying, implementing, and monitoring smart specializations is already widely described in the literature (e.g., in: Foray, 2014; Foray & Europäische Kommission, 2012; Forayet al., 2011; Kleibrinket al., 2016; Lopeset al., 2019; McCann & Ortega-Argilés, 2016; Meyer, 2020; Midtkandalet al., 2014; Morgan, 2016). Researchers analyze the level of innovation of regions in the context of RIS3 (Brodny & Tutak, 2022; Kudełko et al., 2022) and the level of innovation of companies in the regions (Fu, 2021; Gianelle & Kleibrink, 2015). They also note the lack of integration of regional assets to RIS3 design to overcome societal challenges and missing political capabilities to utilize social innovation governances under RIS3 (Meyer, 2022).

In this context the barriers to implementation of innovations are particularly relevant. Despite the fact they are quite well described in the literature on the subject, most often the descriptions boil down to boosting development of innovativeness in its broad sense, rather than implementing any smart specializations. With reference to supporting innovativeness development, barriers to introduction of innovations were analyzed by i.a., Stornelli et al. (2021), Veldhuizen (2020), as well as Bourke and Roper (2016), and Baldwin and Lin (2002). Trippl et al. (2020) and Cortinovis et al. (2017) pointed out to barriers and constraints in an approach that comprised all the actors of the innovation system, particularly in relation to interactions taking place between them. A similar approach was presented by Papamichail et al. (2019) who examined how smart specialization barriers were connected with the lack of absorptive and networking capabilities, and pointed out the lack of strong cooperative relationships between universities and entrepreneurial businesses, this research, however, has a fragmented character. Constraints in implementation of smart specialization were also the subject of evaluations made by Uyarra et al. (2018) and Balland and Boschma, (2021), according to whom the lack of complementary interregional connections significantly decreases the probability of developing new technological specializations in the regions (particularly in the case of peripheral regions). Similar studies were carried out by Grillitsch and Asheim (2018) and Malanowski et al. (2021). Nevertheless, the literature review has not shown any relationships between the impact of barriers to implementation of smart specialization on the actual capability of implementing innovations in enterprises. There is also a lack of studies showing how actions taken by regional decision-makers (local government activities) translate into mitigation of the said barriers.

High competitiveness of the region testifies to the effective organization and optimization of economic processes (Palinchaket al., 2021). Poland, is a country classified as emerging innovator in the European Innovation Scoreboard report (2021). In the Global Competitiveness Report 2020 published by the World Economic Forum Poland took the 30th place in the ranking of 37 countries. The assessment of regional innovativeness indicators for the whole of Poland carried out with the use of the synthetic measure of regional innovation development (SMRID) (described in more detail in: Kogut-Jaworska and Ociepa-Kicińska, 2020), confirms the innovation paradox, according to which there is a conflict between the relatively higher needs for promoting innovations in backward regions and their smaller ability to absorb the available funds and effective investing in innovative operations, compared to more advanced regions. The analyses have led us to consider the significance of barriers to innovativeness for development of regional innovativeness. In view of the contradictory theoretical assumptions regarding the significance of public institutions in the process of creating and developing an innovation policy, it seems worthwhile to examine the impact of measures taken by public sector stakeholders at the regional level, and how smart specializations implemented in the regions affect innovativeness of enterprises.

In our analyses we have made the assumption that on the one hand the activity of public sector institutions is to contribute to development of innovations in specific areas, and on the other hand entrepreneurs operate within their best practices, following the assumptions ensuing from the priorities held by company owners and board representatives. Thus, public intervention is not always effective. We notice a wide range of external conditions which may constitute barriers to implementation of an innovation policy. We identify not only the lack of coordination of the processes of RIS3 designing and implementation, but also their inconsistency with the expected impact on innovativeness of enterprises. In our opinion, even the best coordination at the EU level cannot replace regional stakeholders’ ability to cooperate effectively, which should be both the cause and effect of the worked-out solutions.

In view of the fact that the studies encountered in the literature usually focused on selected areas connected with RIS3 implementation or on assessment of regions’ innovativeness, but without any cause-and-effect analysis, we decided to fill the identified research gap with regard to the three links of the process: smart specialization design and implementation, along with their impact on innovativeness of enterprises (Figure 1).

Figure 1.

Areas comprising the process of practical application of RIS3.

We based our assumptions on the following three theses:

T1: RIS3 policy is of key importance for developing innovative activities in enterprises.

T2: Barriers to innovativeness determine the level of development of enterprises’ innovativeness.

T3: Local government activities contribute to development of enterprises’ innovativeness.

Selection for the study 5 regions covered by the supra-regional Development Strategy of Western Poland allows for the analysis of the part of the country within which (at least theoretically) consistent goals are implemented aimed at the effective use of the regions’ potential, and at the same time the study does not include regions that are typically peripheral in terms of EU borders, the characteristics of which are not representative of all regions of Poland (see the results of the study Kogut-Jaworska & Ociepa-Kicińska, 2020). In-depth interviews (IDIs) were held in each of the regions surveyed with representatives of the key institutions responsible for RIS3 implementation in the respective region, and with representatives of enterprises that run innovative operations. The comprehensive approach was supplemented with the results of surveys conducted among entrepreneurs running innovative activities. This structure was aimed at assessment of areas connected with the innovation policy in the regions in the context of its key stakeholders. The contributions of this research are as follows, in some places there is a lack of coherent activities and expectations between the public and private sectors. Our study reveals a mismatch between forms of support and the realities of business functioning and a different perception of the needs of businesses by authorities, business support institutions, and entrepreneurs themselves. The approach used constitutes a unique contribution to the literature regarding the practical effects of smart specialization implementation. The research procedure developed may be applied in subsequent stages in all regions. On this basis, regional decision-makers, should modify the solutions and assumptions used in their regions in the direction of implementing solutions, not only resulting from the assumptions of the RIS3 concept, but also effective from the point of view of entrepreneurs.

The remaining part of the paper is organized as follows: the first section discusses the research procedure, the next section discusses the idea of an innovation policy and barriers to its implementation in the regions. The third section presents the study’s conceptual framework. Next, our research study concept is presented along with the description of the applied research methods. Section five describes the research results, then we present the conclusions, practical implications, limitations, and directions for future research.

Overview of the Research Process

In view of these interdependencies, we designed a multi-step research study aimed at explore the interactions between implementation of innovation strategy in the form of RIS3, barriers and bottlenecks of innovation diffusion, and development of innovativeness in enterprises on a regional level. The study design of the analysis of interactions between these elements is presented in Figure 2. It is worth noting that this article does not analyze the process of selecting smart specialization in regions, but it verifies their value for entrepreneurs.

Figure 2.

Applied study design.

The general scheme of the study includes three main research phases, that is, exploration, diagnosis, and verification and conclusion. The research scheme presents the objectives of the study, research techniques, which consequently allowed to arrive at the results and conclusions of the entire research project.

The initial stage of the research procedure is a literature review and a review of strategies and policy documents on issues such as smart specialization, innovation policy, and support for its implementation by public sector entities, as well as barriers to the implementation of innovation management policies in the regions. Its purpose is to check the relevance of these issues, help set the context, and identify the circumstances of their appearance in the literature. The analysis of strategies and documents in a given area and at a given time, as well as familiarization with the assumptions of national and European policies, provides the background for the entire research process. The exploratory stage showed how the research issues should be approached to enrich the existing discussions and literature on the subject.

The second stage of the research procedure was the diagnostic stage. Designed and implemented in two parts, it helped to bring out new, bottom-up, and often previously non-existent information on the progress of the implementation of smart specialization strategies, to assess the achieved effects within the framework of the conducted public sector support policies, as well as to define guidelines and recommendations to increase the effectiveness of innovation management in the regions within the framework of the implementation of smart specialization strategies. Two research techniques were used at this stage, that is, IDI—Individual in-Depth Interviews (qualitative research) and CATI—Computer Assisted Telephone Interview (quantitative research) questionnaire interviews.

The empirical material obtained on the basis of the IDI technique was used to define the research problem at the next stage of the research procedure carried out using the CATI technique. This is primarily to identify the most relevant issues that are the subject of the study and to prepare the questionnaire for the CATI survey. The diagnostic function of in-depth interviews was also used as part of the research conducted. This provided empirical data for the development of conclusions and findings.

The third stage produces empirical results in the form of a Bayesian model of the impact of smart specialization on business innovation. Its interpretation leads to conclusions and inferences, and makes it possible to formulate recommendations for future innovation policy conducted on the basis of smart specialization strategies. The model was built on the basis of variables whose relationships were characterized by significant correlation. The choice of the model was dictated by the fact that models based on Bayes’ theorem, are particularly suitable for problems with very many dimensions in the input. Despite the simplicity of the method, it often performs better than other very complex classification methods; secondly, given the assumption of independence, Bayesian models perform better compared to other models such as logistic regression and require less training data; thirdly, here we can predict the probability of multiple categories of the target variable.

The survey was conducted in the Polish regions covered by the supra-regional Development Strategy of Western Poland (five regions), in which IDIs were conducted with representatives of key institutions responsible for the implementation of RIS3 in the region and with representatives of enterprises engaged in innovative activities. The comprehensive approach was complemented by the results of surveys conducted among enterprises engaged in innovative activities. This design was intended to assess areas related to innovation policy in the regions in the context of its key stakeholders. Its results show in some places a lack of consistent actions and expectations between the public and private sectors.

Based on the survey results, which provided 20,278 input data, the indicators credibility was verified using Cronbach’s alpha coefficient. The computed indicators were used in order to build a model of the impact of smart specialization on innovativeness of enterprises, based on the Bayesian linear regression analysis, with Student’s t-distribution as residual probability distribution. The main objective of the analyses was to present the interdependencies and relationships that emerge between measures that facilitate innovation, benefits of RIS3, and measures taken by self-government decision-makers (regional authorities responsible for RIS3 implementation) and barriers (the regional level). Based on the analyzed theoretical framework, our analysis reveals the need to assess the impact of one factor on the others, whereby the factors are boiled down to indicators which are subsequently analyzed in relation to the data collected during the multi-stage study. Investigating the relationships between barriers to implementation of innovations and other factors within the framework of our concept will make it possible to analyze the results, draw conclusions and provide evidence of their significance for smart specialization policy management.

Theoretical Background

Public sector innovation policy—A dynamic institutional context

In line with the European Commission’s development policy (2014) integrated smart specialization strategies respond to complex development challenges by adapting the policy to the regional context. Public innovation policies evolve and change along with new trends, political decision-makers and changes to other policies, thus making up its totality. Each new political attempt must take into account the previously existing policies and find the way to create productive layers of both the existing and the new policies (Kern et al., 2019). Particularly important is the political stability of the implemented solutions, its absence is consider as building block to inhibit economic and innovation activities in many economies Lewis et al. (2018) examining factors that have an impact on innovation capacity emphasizes the need to understand innovation in public sector environments is growing. The RIS3 concept is based on the business and organizational dynamics, evolving as the framework for designing, implementing and evaluation of regional innovation policies (Coenen et al., 2017). The dynamic co-evolution of an innovation ecosystem is instigated by relationships between interdependent actors (Liuet al., 2022).

Barriers to innovation implementation—Regional context

Identification of barriers to implementation of innovations (bottlenecks) is a key factor of development of innovation activity, as a correct evaluation of the innovation system condition may help indicate activities that facilitate the determinants of development in the regions. Barriers to implementation of smart specialization strategies are often directly connected with the capability of key stakeholders to build inter-organizational networks, which corresponds to their strategic needs and enterprises’ absorptive capacities of utilizing scientific knowledge. Papamichail et al. (2019) demonstrated that the main barriers to RIS3 implementation resulted from the local firms’ limited capabilities of obtaining university knowledge and playing a leading role in developing action plans for smart specialization implementation. This argument was based on the fact that companies are unable to utilize scientific knowledge, which is of key importance for implementing smart specialization goals in a collective and systematic manner. Despite the significance of network relations, the analysis made by Ghinoi et al. (2021) has shown a significant lack of entrepreneurial activity in that regard. Papamichail et al. (2019) underline that lack of cooperation culture at the institutional level, diverse understanding of strategic networking between the key RIS3 entities, lack of relations based on trust, constitute the three major barriers to RIS3 implementation. According to De Noni et al. (2021), the barriers that prevent regions from entering more complex innovation trajectories depend on absorptive capacities required to assimilate and integrate the cognitively remote sources of knowledge. Lack of cooperation and reluctance to build partnerships are also another major barrier within international connections being the framework for smart specialization strategies (Ascani et al., 2020).

Mueller-Using et al. (2020) pinpoint that small and medium enterprises, striving to build capabilities of internationalization rather than direct support to win customers, expect establishment of appropriate support programs for innovativeness and competitiveness. Gianelle, Guzzo, et al. (2020), in turn, provide evidence for the need to refine the financial aid mechanisms within the incentive system established at the European Union level.

Barriers to implementing smart specializations—Organizational context

Barriers to implementation of smart specialization strategies are also identified in relation to enterprises themselves. Firm’s own innovation activities have a big impact (Sarkkinen & Kässi, 2013). Even though over the recent years there have been many changes having a positive effect on innovativeness in enterprises, i.a. investments in new R&D infrastructure, enhancing the competences of human resources engaged in innovative projects, technological progress, developing proprietary solutions based on R&D works results, they are still insufficient and there are still many challenges related to implementation of innovations. In particular, it is observed that entrepreneurs still exhibit limited inclination for pro-innovation activities, enterprises show low capability of risk-taking and cooperation, and there is lack of trust, low level of competence, lack of ability to manage innovation among employees, poor competitiveness of companies on a global scale.

Stornelli et al. (2021) analyzed the facilitating and inhibiting factors in the process of adopting advanced production technologies, and their relationship with various types of innovations being the outcomes of that process, and in particular they showed the relationship between the categories of the facilitating and inhibiting factors associated with the outcomes of the innovation types: innovations in the area of products, processes, services, or business models. The ability to successfully introduce innovations is becoming a key corporate capability, strongly dependent on companies’ access to a knowledge capital (Bourke & Roper, 2016). The studies completed by Swamidass and Winch (2002) and Sjödin et al. (2018) have shown that implementations in companies may go wrong when managers do not perceive the barriers connected with the employees’ defiance and the need for them to have specific skills and knowledge. The studies of Voss (1992) and Baldwin and Lin (2002), point out to the organizational system in relation to various corporate value systems, which, if ineffective, also constitutes a constraint. According to Müller et al. (2018), a significant barrier to raising innovativeness of companies can be improperly coordinated connections of the implementation systems with possible innovation types. Kuhlmann and Ordóñez-Matamoros (2017) and Malanowski et al. (2021) state that in such situations it is additionally necessary to discuss any failures in innovation management from the past, so as to better understand the possibilities offered at present.

Conceptual Framework

Data and methods

In the previous section we indicated three theses regarding interactions between innovativeness policy activity in the form of RIS3 and enterprises’ innovativeness development in the regions. The goal of this section is to discuss the research methodology drawn up by us, aimed at designing a model for examine the said theses.

Our analyses were based on a triangulated study design at the level of the regions that implement regional smart specialization strategies. Data triangulation, being a research method applied in social sciences, was selected as one that ensures better quality of our evaluations and conclusions. Via application of the key methodological tool in studies involving mixed methods, following Lewis-Beck et al. (2012), we reinforce both qualitative and quantitative analyses that reflect the findings about the relationships between the identified factors from more than one point of view. Next, we compared and consolidated the outcomes. Via utilization and combination of several data sets, at the same time verifying the qualitative and quantitative methods for potential source errors (Denzin, 2009), we strived to test the same research hypotheses, at the same time reducing the error bias. We triangulated various data sources pursuant to the existing standards in mixed methods, so as to provide the ultimate research regime.

Analysis of smart specialization strategies in regions and guidelines resulting from national and EU documents

The analysis was based on the RIS3 policies for regions located in Western Poland, and also on documents outlining the European profiles of the studied regions in terms of smart specializations, including the voivodeships/regions: Zachodniopomorskie (European Commission (JRC-IPTS), 2015d), Wielkopolskie (European Commission (JRC-IPTS), 2015c), Lubuskie (European Commission (JRC-IPTS), 2015a), Dolnośląskie (European Commission JRC-IPTS, 2015), Opolskie (European Commission (JRC-IPTS), 2015b), as well as on the currently binding regulations regarding the principles of RIS3 application in Europe (European Commission, 2018; European Union, 2013). The area covered by the research study is presented in Figure 3.

Figure 3.

Regions covered by the study.

The RIS3 regional strategies were analyzed by means of a qualitative analysis of the structural content. In particular, the qualitative analysis applied the testing approach recommended by Rich et al. (2018) mainly in applying text mining tools (Iramuteq), and its practical use involved an analysis of the content of the documents which were then sorted into thematic categories and formulated as themes related to our research questions. The analysis aimed at identifying the ways of implementing smart specialization in the regions, benefits that may be obtained by stakeholders upon such implementations, measures to be taken in order to complete the implementations.

Interviews with key stakeholders—Qualitative studies

To obtain and transcribe the interview data, the interview content analysis technique was applied (Rich et al., 2018). The empirical material comprised 30 individual in-depth interviews (IDIs). The individual in-depth interviews were held at the turn of November and December 2020, and they were preceded by identification of interviewees using the targeted sampling method which is a frequently applied technique for selecting surveyees in qualitative studies (Frankfort-Nachmias & Leon-Guerrero, 2016). The sampling of the surveyees made it possible to obtain a research sample that most fully reflected the studied population. Due to the COVID-19 pandemic and ensuing restrictions regarding the traditional face-to-face interviews, we decided to conduct on-line interviews (including recordings and transcripts) or phone conversations (using semi-structured telephone scripts). The survey was conducted among the respondents representing four types of entities:

(1) regional (voivodeship) self-government (officers in organizational units in Marshal’s Offices, who were directly engaged in the process of preparation or implementation of smart specialization strategies);

(2) local self-government—representatives of Municipal Offices of the metropolises, capital cities of the voivodeships covered by the study (officers in organizational units whose scope of responsibilities was related to the issues of smart specialization strategies implementation);

(3) business environment institutions—representatives of the institutions operating in the area covered by the study;

(4) enterprises—officers in enterprises provided with financial aid under regional operational programs in the voivodeships covered by the study.

The interviews (which were archived) took the total of nearly 28 hours, of out which over 16.5 hours were the interviews held with the representatives of the offices and business environment institutions (in that case the average length of interview exceeded 1 hour 6 minutes), whereas in the case of enterprises the total interviewing time was nearly 11.5 hours (slightly above 45 minutes per one interview). Pursuant to our approach to the qualitative analysis, the interview scenario was based on the RIS3 strategic documents of the Western Poland regions, and the guidelines resulting from the national and EU regulations. The interview structure covered questions regarding RIS3 management practices in the area of (1) process of selecting smart specializations in the region, (2) stakeholders’ role in RIS3 implementation process, (3) RIS3 implementation dilemmas, (4) barriers to RIS3/innovation activities implementation, (5) benefits resulting from RIS3 implementation, (6) RIS3 implementation tools, (7) RIS3 monitoring and evaluation, (8) effects of RIS3 implementation in the region, and its overall evaluation.

Questionnaire for entrepreneurs—Quantitative studies

The survey was conducted using the CATI method. The questionnaire covered the enterprises which in the years 2016 to 2020 supported their innovative operations with public/EU funds at the national (Intelligent Development Operational Program) and regional level (R&D under Regional Operational Program). The research study was carried out at the turn of January and February 2021. The questionnaire for the CATI survey was constructed on the basis of the obtained data about RIS3, and the structure of the research questions was the outcome of our analyses derived from the earlier IDIs, pursuant to the RIS3 evaluations and the approach utilized at the second stage. The survey was completed using the random-quota sampling method among 250 enterprises running their innovation activities in the territory of the five voivodeships. The rationale and size of the survey sample are included in Supplemental Appendix 1. The sample distribution reflected the general population distribution, that is, of the innovative companies in Western Poland (Central Statistical Office, 2020), which provided the grounds for generalization of the outcomes to the whole population (of innovative companies from Western Poland). The maximum measurement error for the sample of 250 was 6% (which was a satisfactory result), at a confidence level of 94%.

The pretested questionnaire comprised five parts which generally provided data regarding general information to classify the respondents, RIS3 strategy implementation progress, evaluation of obtaining expected results, and effectiveness of innovativeness management in the regions under RIS3. It contained 16 multiple-choice questions, and also a section regarding data collection and utilization. Two thousand eight hundred sixty-two records were used for the response rate, some of which were refusals to participate in the survey, most often due to absence of competent persons (n = 326), being uninterested in taking part in the survey (n = 123), lack of possibility of contact on the phone (n = 122).

Descriptive Statistics and Empirical Methodology

Indicators

The indicators analyzed under the study were selected on the basis of the literature review presented in the first sections of the article, and on the conclusions drawn from the IDIs held with the major stakeholders engaged in creation and implementation of smart specialization strategies in the studied regions. The first group of indicators regarded the measures that to the largest extent create/develop innovations in the analyzed enterprise. It is important, that enterprises take innovative steps even when they do not know the innovation policy assumptions. Within this category, 10 indicators were distinguished, which covered both the bottom-up employees’ initiatives and those on a broad scale, for example, cooperative activities: R&D team operation, initiatives of the executives/the board, employees’ initiatives, customers’ expectations, suppliers’ initiatives, actions taken by competitors, new market opportunities, purchase of technologies, licenses, patents, and know-how, cooperation with university research centers, obtaining external funding. The data show the areas of activity of enterprises and their main stakeholders, in which the enterprises identify the significance of innovation. They also provide information about whether or not the enterprise introduces/develops innovations.

The second group of indicators focused on perception of benefits coming from regional smart specialization strategies, such as: increased research and innovation activities of the enterprise, impact on development of scientific areas and the enterprise’s innovativeness, modernization of the economy, internationalization, and cooperation within the framework of cooperative links, an increase in direct investments in the region, enhancing the system of innovation financing. The information shows not only the entrepreneurs’ awareness of the structure and assumptions, but also the scope of influence RIS3 has on the innovativeness from the point of view of enterprises.

The third group contains indicators that describe barriers to innovativeness development, as perceived by enterprises themselves. The following issues have been taken into account: lack of mutual trust between entities, insufficient financial potential, low level of technical infrastructure, insufficient human resources/human capital, limited conditions for local economy development, low level of integration of local communities, insufficient institutional potential, unfavorable conditions of the natural environment, no need to run innovative activities. The data make it possible to conclude which of the areas constitute the most vital constraints from the point of view of the entrepreneurs.

The fourth group of indicators regarded the perception of self-governments’ activities taken within the framework of smart specialization support system. The focus was on the undertakings which to the greatest extent help to develop innovativeness in enterprises, that is: creating favorable administrative conditions, co-generating demand for smart specialization products, building diverse networks of cooperation (R&D), information campaigns and promotional support, arranging R&D cooperation centers, creating cyclical forms of R&D cooperation, commissioning and conducting research studies regarding smart specialization, running programs related to those issues, and subsidizing with budgetary (self-government) funds. Information provided by this group shows how entrepreneurs perceive measures taken by institutions from the public sector. This is particularly relevant in view of the fact that it is to them that the activities are dedicated, on the other hand, the ways of perceiving the needs of entrepreneurs by themselves and by public administration bodies are not always consistent.

Statistical analyses

The collected research material was used in order to build a model of the impact of smart specialization on innovativeness of enterprises. The statistical analyses and necessary operations on input data covering the above described four groups of indicators were performed using the R 4.0.2 statistical software package. The first step was to validate the credibility (internal consistency) of the indicators computed with regard to the individual question groups, using Cronbach’s alpha coefficient. Due to the ordinal nature of the answers to the questions, Cronbach’s alpha coefficient was calculated on the basis of polychoric correlation coefficients (Gadermann et al., 2012). Conventionally, Cronbach’s alpha coefficient value >.7 indicates the scale of credibility that is sufficient for research studies. In the second step, values of the indicators were calculated by averaging the answers provided by a given respondent to the questions within the group. Answers such as “I don’t know” or “Undecided” were treated as lack of data.

In the next step, the computed indicators were used in the Bayesian linear regression analysis, with Student’s t-distribution as the residual probability distribution. The regression coefficients estimations in this model are considerably less susceptible to outliers, compared to the classical linear regression (Kruschke, 2014). In the Bayesian analysis, the a posteriori probability distribution of the model parameters is estimated, which is calculated by means of credibility integration and a priori distribution. Conclusions about the statistical credibility of a parameter (e.g. difference between measurements) are made by computing the mean and the 95% credibility interval (CI). A statistically reliable outcome is the one for which the 95% CI does not contain z zero (Kruschke, 2014). This method was chosen primarily because it provides accurate inference, without relying on asymptotic approximations, inference on a small sample proceeds in the same way as if the sample you have was large. From the perspective of this study, it was important that predictive distributions allow in-depth testing of any particular aspect of the model, and thus simulated data from these distributions can be compared with real data (see more in: Donovan & Mickey, 2019). The level of data fitting was assessed by means of Bayesian R2 (Gelman et al., 2019), whereas comparison between the alternative models was made by means of the cross-validation method, based on the LOOIC (leave-one-out information criterion) statistics (Vehtari et al., 2017).

Estimation of a posteriori distributions was made by means of the BRMS software package (Bürkner, 2017) applying a sampling algorithm implemented in the Stan language (Carpenter et al., 2017). Each reported model was computed by means of four parallel chains based on 8,000 samples. The first half of the samples constituted the adaptation period and was not subject to analysis. Only 1 in 10 samples from the second half was registered, which led to registration of the total of 1,600 samples for each model. The sampling procedure was efficient, which was confirmed by the lack of autocorrelation in the chains, chain convergence (all values R-hat < 1.01) and visual inspection of a posteriori distributions and chains (Kruschke, 2014). Additionally, the fitting of the probability distribution to the dependent variable was assessed by means of a series of graphs comparing the empirical distribution of the variable to the distribution simulated by means of the model parameters (a posteriori predictive check).

Empirical Results

Findings of the qualitative study

The concept of regional smart specializations was evaluated quite differently by respondents, with rather not very positive assessments dominating. It was emphasized that the introduction of regional smart specializations looked like a not fully developed idea on the part of the European Commission, which was repeatedly modified and clarified already during the implementation process itself. It was also pointed out that the concept itself might work well in the conditions of highly developed regions.

Among the main barriers to the implementation of innovative activities, there were some discrepancies in assessment on the part of respondents representing offices and business environment institutions and interviewees representing companies. Among the former group, the following are indicated as the most serious barriers: entrepreneurs’ conviction that there is no need to implement innovations and fear of investing or entering a higher level of activity (a certain level of complacency, lack of need for continuous improvement of, for example, the production process, etc.), unfavorable size structure of enterprises (mainly micro and small enterprises), low availability, and low quality of services offered by business environment institutions. In the case of enterprises, the most frequently indicated barriers were inadequate and unsuited to the realities of the operation of enterprises, the offer of scientific entities and the lack of support from business environment institutions.

There are also divergent assessments of ways to eliminate barriers to innovation between the group of respondents representing offices and business environment institutions and those representing enterprises. In the first group of respondents, the need to intensify cooperation between the administration and enterprises and the scientific sector was quite often indicated as the most important postulate. In turn, many entrepreneurs expressed their dissatisfaction with the lack of sufficient cooperation and the lack of effects of such activities.

Findings of the quantitative survey

The research theses were tested by means of multi-step analyses, leading to emergence of the model of the impact of smart specialization on innovativeness of enterprises. Table 1 contains the descriptive statistics along with the credibility analysis results for the computed indicators. Each of the computed indicators reached at least a high level of internal consistency (α > 0.8).

Table 1.

Descriptive Statistics of Indicators Included in the Study.

	M	SD	Me	IQR	Min.	Max.	α	M [r]
Activities developing innovation	2.98	0.56	3.05	0.6	1	4	0.85	0.35
Benefits of RIS3	3.11	0.72	3.17	0.83	1	4	0.91	0.63
Barriers to innovativeness	3.2	0.84	3.22	1.22	1	5	0.85	0.38
Local government activities	3.92	0.79	4	1	1.29	5	0.91	0.54

Note. M = mean; SD = standard deviation; Me = median; IQR = interquartile range; Min = minimum value; Max = maximum value; α = Cronbach’s alpha coefficient; M [r] = mean value of polychoric correlation coefficient between questions within the indicator.

In the next step, Pearson correlation coefficients between the indicators were computed (Graph 1). On that basis, it was found that Activities developing innovation were moderately strongly correlated to perceiving Benefits of RIS3 and at the same time weakly correlated to the level of perception of Barriers to innovativeness and Local government activities. Also, some weak positive correlations between predictors of Activities developing innovation were observed.

Graph 1.

Matrix with dispersion graphs and corresponding Pearson correlation coefficients. The graphs on the diagonal axis are histograms. Indicators are presented in the normalized 0 to 1 scale.

In order to analyze the relationship between a given indicator and Activities developing innovation, during the next step, in the course of checking the impact of the other indicators, we performed a Bayesian linear regression analysis with Student’s t-distribution as the residual probability distribution, with variables in the standardized scale:

y_{i} ~ t (X_{i} β, σ, ν), dla i = 1, \dots, n,

(1)

where n is the number of observations, $X_{i}$ is a four column matrix, where the first one is the column of ones (to estimate the value of the intercept), and the three other columns are predictors of Benefits from RIS3, Barriers to innovativeness and Local governments activities, $y_{i}$ are Activities developing innovation. The parameter $β$ is a vector with the length of 4, where the first element is an intercept, and the three subsequent elements are standardized regression coefficients of predictors. The parameter $σ$ is t-distribution scale parameter, whereas the parameter $ν$ is Student’s t-distribution normality parameter (the higher the values, the more convergence between t-distribution and normal distribution, the distributions are the same when $ν = + \infty) .$ The value of the parameter $β = 0$ concludes that correlation coefficient between the predictor and the dependent variable is not significant, the values $β > 0$ indicate a positive correlation, whereas the values $β < 0$ show a negative correlation. The absolute values of $β$ in the 0.2 to 0.5 range are described as weak correlations, values in the 0.5 to 0.8 range may be described as moderately strong, whereas values >0.8—are considered strong correlations. In this modeling, the weakly informative a priori distributions were applied, which do not have a significant impact on estimation of standardized regression coefficients: $β ~ N (0, 0.5)$ . This means a distribution which assumes a priori that the most probable relationships between the predictor and the dependable variable are positive and negative, weak to moderately strong correlations. The modeling results in the form of a summary of the a posteriori distributions of the model parameters are presented in Table 2, whereas the model predictions are presented in Graph 2.

Table 2.

Model of the Impact of Smart Specialization on Innovativeness of Enterprises (Results of the Linear Regression Analysis, Where the Dependent Variable is “Activities Developing Innovation”).

Parameter	Me	SE	LI	UI
β
Constant	0.07	0.05	−0.04	0.17
Benefits of RIS3	0.45	0.06	0.32	0.57
Barriers to innovativeness	0.1	0.05	0	0.2
Local government activities	0.22	0.06	0.09	0.33
σ	0.7	0.06	0.59	0.81
ν	11.33	7.97	3.81	33.51
R ²	0.33	0.04	0.25	0.41

Note. Me, SE, LI, and UI are: median, standard error, and lower and upper limit of the 95% credibility interval for the a posteriori distribution of the parameter. β = standardized regression coefficient; σ = Student’s t-distribution scale parameter; ν = Student’s t-distribution normality parameter

Graph 2.

The linear relationships between the predictors and activities developing innovation, obtained from the a posteriori distributions of the intercept and the regression coefficient corresponding to the predictor (blue lines) against the data background (grey points). The variables are presented in the standardized scale.

In the next step, the following categorical variables: voivodeship (five categories) and enterprise size (three levels) were introduced into the model. Introduction of those variables did not improve the predictive power of the model, as the change in LOOIC was positive and it did not exceed the two-fold value of the standard error of the difference, ΔLOOIC = 8.30 (SE = 4.41), which means that the variables did not differentiate in a statistically reliable manner the medium levels of Activities developing innovation. Similarly, the type of activity (introduced as a random effect due to the large number of categories) also did not improve the predictive power of the model, as the change in LOOIC was positive and it did not exceed the two-fold value of the standard error of the difference, ΔLOOIC = 1.53 (SE = 1.48).

It was noted that the strongest predictor of Activities developing innovation was the perceived Benefits of RIS3, this correlation was positive and weak. This means that the entrepreneurs who noticed higher benefits of RIS3 more often implemented innovation activities in their companies. The a posteriori distribution of the coefficient for Barriers to innovativeness included a zero, which suggests that when checking Barriers to innovativeness and Local government activities, this indicator is not connected with Activities developing innovation. This can be interpreted as follows: the entrepreneurs who introduce innovations in their operations do not focus on identification of barriers that hinder the operations. Local government activities proved to be a statistically reliable predictor for the dependent variable in the analyzed statistical model. The correlation was positive and weak, more than twice weaker than the correlation between Benefits of RIS3 and innovative activities. This provides grounds to believe that Local government activities to only a small extent contribute to developing innovativeness in enterprises.

Even though the designed model of the impact of smart specialization on innovativeness of enterprises in fact disclosed the correlations between Activities developing innovation, perception of Benefits of RIS3, and Local government activities, the low level of the identified correlations causes all the formulated theses to be verified negatively. Although the research assumption was connected with an expectation of receiving a strong positive proof of the posed theses, the weak and very weak correlation leads to another important conclusion that the solutions based on smart specialization which have been designed and implemented in the present form have not been bringing expected results. The model developed has not proved the theses we have posed and which are coherent with the theoretical assumptions of the smart specialization strategy. The reasons for that could be, inter alia, diverse expectations and priorities of the public (self-government) sector stakeholders and entrepreneurs. The first group of the surveyed respondents indicated, as their major proposal, the need to intensify the cooperation between administration, enterprises, and the science sector. At the same time, entrepreneurs expressed their disappointment with hours-long meetings aimed at establishing such cooperation due to the ineffectiveness of this kind of actions.

Conclusions

This study contributes to the growing field of research in fostering regional innovation by demonstrating how the concept of smart specialization is perceived by its key stakeholders. This paper examines how regions’ efforts to foster innovation are evaluated by innovation entrepreneurs. Using a triangular research scheme, we provide evidence that despite the considerable efforts made by regions to implement the RIS3 objectives, their results are not consistent with entrepreneurs’ expectations. Our conclusions are consistent with the results obtained by Gianelle, Guzzo and Mieszkowski indicating that regional decision-makers lack capabilities and resources necessary for managing numerous priority goals (Gianelle, Guzzo, et al., 2020). Our results underscore the need for better management of smart specialization strategies so as to eliminate the gap we identified between RIS3 designing, implementation, and its actual impact on innovativeness of enterprises.

On the basis of the research results it is possible to conclude that there is only a small degree of interaction between the introduced institutional framework of smart specialization functioning and development of innovativeness in enterprises in Polish regions. The knowledge about RIS3 is insignificant—entrepreneurs knew the concept of RIS3 usually only in the situation when they sought funding for their projects from EU subsidies.

Our study offers implications for policymakers and economic practitioners. On our opinion, smart specialization has so far supported the more inclusive forms of management, whereas we are now noticing the need for coordination and changes on inter-institutional levels. Decision makers should consider the usefulness and possible practical implications of the forms of support they offer. They should also take more account of entrepreneurs’ expectations. Interaction and good communication in the relationship between key stakeholders is essential to increase the effectiveness of smart specialization. Our findings are consistent with Aranguren et al. (2019), according to whom the competences of the regional self-government constitute a merely initial issue—they must be combined with a possibility of their proper utilization, where institutional quality and learning capability gain key importance. We confirm Gianelle, Kyriakou, et al. (2020) suggestion that political measures should be tested and evaluated based on their capability of stimulating private entrepreneurs to make efforts to search for and devise new kinds of activities that are able to generate added value and meet the challenges of contemporary innovativeness. Political measures should be designed in such a way so as to mobilize entrepreneurial powers and innovative potential. An approach based on smart specialization makes it possible to develop concrete political solutions based on unique features and behavior of local actors, and in order to implement such truly territorial dimensions of the policy, the bodies responsible for the policy designing should be open to creation of cyclic forms of research & development cooperation and better appropriation of budgetary/ self-government funds, which our research study has identified as the most important from among self-government activities taken up within the framework of the RIS3 management system. The model we designed may be one of the first elements in an analysis of the current situation in any given region or country. In a situation where in any given area the theses we have posed are negatively verified, it is probably the whole structure of innovation policy implementation tools and its assumptions that requires remodeling.

At the same time, the entrepreneurs who noticed benefits of the RIS3 policy introduced innovative activities in their companies to a larger extent. This demonstrates the huge potential of the policy—its propagation might be a big step toward increasing the level of innovativeness in Polish regions. It was surprising to find a very weak correlation between barriers to innovativeness and the level of development of innovativeness in enterprises. The conclusions that may be drawn from the model of the impact of smart specialization on innovativeness of enterprises are that any hindrances, considered in the literature to be the barriers, are in practice noticed mainly by enterprises which in their operations do not create innovative activities, but make manifestations of innovative activities only when this is a precondition for obtaining financial support. The RIS3 policy is not as important for development of innovativeness in enterprises as it should be, and the barriers to implementation of innovativeness are perceived differently by decision-makers responsible for RIS3 implementation, and beneficiaries of the policy as such. The identified weak correlation between actions taken at the regional level and its effect on development of innovativeness in enterprises demonstrates an inappropriately utilized potential. The conclusions drawn from the created model were not specific to any concrete region, which shows that the problem is universal on the national scale. It is particularly important in the context of the findings of Magro and Wilson (2019), according to whom a new innovation policy becomes a collective intelligence process which is of key importance for avoiding mistakes in coordination between various policy levels, and also for effective management of decisions related to the strategic orientation. Even though most RIS3 strategies are implemented in accordance with their initial plans, it is necessary to take steps so that the policy is not perceived mainly as a tool for absorbing EU funds, and that it becomes a concept that in fact enhances innovativeness in the regions. The new innovation policy for the regions must focus on changes in the approach to understanding “smart specialisation,” in which innovations should not blindly follow the logic of regional competitiveness, but they should correspond to social challenges on a broader scale, and be an intermediate step in the direction of long-term goals to support sustainable growth.

This study has several limitations. First, it is a cross-sectional study and does not represent the entire population of innovative companies. It does not include comparisons between the level of innovation of companies that are beneficiaries of public/EU funds (included in the survey) and innovators who do not benefit from these funds. Second, Poland is not among the leaders in innovation. Covering regions from countries with a higher level of innovation with an analogous survey could show conclusions, which is done differently where entrepreneurs are better at evaluating activities implying a smart specialization strategy. Further research may also attempt to assess the relationship between the support tools used, the way they are perceived by entrepreneurs, and the level of innovation of countries/regions.

Supplemental Material

sj-docx-1-sgo-10.1177_21582440231180520 – Supplemental material for Practical Implications of Smart Specialization Strategy: Barriers to Implementation, Role of the Public Sector, and Benefits for Entrepreneurs

Supplemental material, sj-docx-1-sgo-10.1177_21582440231180520 for Practical Implications of Smart Specialization Strategy: Barriers to Implementation, Role of the Public Sector, and Benefits for Entrepreneurs by Magdalena Kogut-Jaworska and Elżbieta Ociepa-Kicińska in SAGE Open

Footnotes

Author Contributions

MKJ and EOK were responsible for conceptualization, literature review, methodology, investigation, and writing an original draft. MKJ and EOK were responsible for the project administration, selection of variables, development of databases, and their initial processing. MKJ and EOK were responsible for writing—review and editing.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The project is financed within the framework of the program of the Minister of Science and Higher Education under the name “Regional Excellence Initiative” in the years 2019 to 2022; project number 001/RID/2018/19; the amount of financing PLN 10,684,000.00.

ORCID iDs

Magdalena Kogut-Jaworska

Elżbieta Ociepa-Kicińska

Supplemental Material

Supplemental material for this article is available online.

References

Ahner

Landbaso

(2011). Regional policies in times of austerity. European Review of Industrial Economics and Policy. http://revel.unice.fr/eriep/index.html?id=3238

Aranguren

M. J.

Magro

Navarro

Wilson

J. R.

(2019). Governance of the territorial entrepreneurial discovery process: Looking under the bonnet of RIS3. Regional Studies, 53(4), 451–461. https://doi.org/10.1080/00343404.2018.1462484

Ascani

Bettarelli

Resmini

Balland

P.-A.

(2020). Global networks, local specialisation and regional patterns of innovation. Research Policy, 49(8), 104031. https://doi.org/10.1016/j.respol.2020.104031

Baldwin

Lin

(2002). Impediments to advanced technology adoption for Canadian manufacturers. Research Policy, 31(1), 1–18. https://doi.org/10.1016/S0048-7333(01)00110-X

Balland

P.-A.

Boschma

(2021). Complementary interregional linkages and smart specialisation: An empirical study on European regions. Regional Studies, 55(6), 1059–1070. https://doi.org/10.1080/00343404.2020.1861240

Bourke

Roper

(2016). AMT adoption and innovation: An investigation of dynamic and complementary effects. Technovation, 55–56, 42–55. https://doi.org/10.1016/j.technovation.2016.05.003

Brodny

Tutak

(2022). Assessing the level of innovation of Poland from the perspective of regions between 2010 and 2020. Journal of Open Innovation Technology Market and Complexity, 8(4), 190. https://doi.org/10.3390/joitmc8040190

Bürkner

P.-C.

(2017). BRMS: An R package for Bayesian multilevel models using stan. Journal of Statistical Software, 80(1), 1–28. https://doi.org/10.18637/jss.v080.i01

Carpenter

Gelman

Hoffman

M. D.

Lee

Goodrich

Betancourt

Brubaker

Guo

Riddell

. (2017). Stan: A probabilistic programming language. Journal of Statistical Software, 76(1), 1–32. https://doi.org/10.18637/jss.v076.i01

10.

Central Statistical Office. (2020). https://stat.gov.pl/obszary-tematyczne/nauka-i-technika-spoleczenstwo-informacyjne/nauka-i-technika/dzialalnosc-innowacyjna-przedsiebiorstw-w-polsce-w-latach-2019-2021,14,9.html

11.

Coenen

Asheim

Bugge

M. M.

Herstad

S. J.

(2017). Advancing regional innovation systems: What does evolutionary economic geography bring to the policy table? Environment and Planning C: Politics and Space, 35(4), 600–620. https://doi.org/10.1177/0263774X16646583

12.

Cortinovis

Xiao

Boschma

van Oort

F. G.

(2017). Quality of government and social capital as drivers of regional diversification in Europe. Journal of Economic Geography, 17(6), 1179–1208. https://doi.org/10.1093/jeg/lbx001

13.

De Noni

Ganzaroli

Pilotti

. (2021). Spawning exaptive opportunities in European regions: The missing link in the smart specialization framework. Research Policy, 50(6), 104265. https://doi.org/10.1016/j.respol.2021.104265

14.

Denzin

N. K.

(2009). The research act: A theoretical introduction to sociological methods. Transaction Publishers.

15.

Donovan

T. M.

Mickey

R. M.

(2019). Bayesian statistics for beginners: A step-by-step approach (Illustrated edition). Oxford University Press.

16.

European Commission. (2018). Regulation of the European parliament and of the council laying down common provisions on the European regional development fund, the European social fund plus, the cohesion fund, and the European maritime and fisheries fund and financial rules for those and for the asylum and migration fund, the internal security fund and the border management and visa instrument.

17.

European Commission JRC-IPTS. (2015). Stairway to Excellence Dolnośląskie (PL51) Facts & Figures (Numer June).

18.

European Commission (JRC-IPTS). (2015a). Stairway to Excellence Lubuskie (PL43) Facts & Figures (Numer June).

19.

European Commission (JRC-IPTS). (2015b). Stairway to Excellence Opolskie (PL52) Facts & Figures (Numer June).

20.

European Commission (JRC-IPTS). (2015c). Stairway to Excellence Wielkopolskie (PL41) Facts & Figures (Numer June).

21.

European Commission (JRC-IPTS). (2015d). Stairway to Excellence Zachodniopomorskie (PL42) Facts & Figures (Numer June).

22.

European Innovation Scoreboard report 2021. (2021). https://research-and-innovation.ec.europa.eu/statistics/performance-indicators/european-innovation-scoreboard_en

23.

European Union. (2013). Regulation (EU) No. 1303/2013 of the European parliament and of the council of 17 December 2013 laying down common provisions on the European regional development fund, the European social fund, the cohesion fund, the European agricultural fund for rural development and the European maritime and fisheries fund and laying down general provisions on the European regional development fund, the European social fund, the cohesion fund and the European maritime and fisheries fund and repealing council regulation (EC) No 1083/2006.

24.

Fabrizio

K. R.

Poczter

Zelner

B. A.

(2017). Does innovation policy attract international competition? Evidence from energy storage. Research Policy, 46(6), 1106–1117. https://doi.org/10.1016/j.respol.2017.04.003

25.

Foray

(2014). Smart specialisation: Opportunities and challenges for regional innovation policies. Routledge. https://www.routledge.com/Smart-Specialisation-Opportunities-and-Challenges-for-Regional-Innovation/Foray/p/book/9781138923652

26.

Foray

David

Hall

B. H.

(2011). Smart specialisation from academic idea to political instrument, the surprising career of a concept and the difficulties involved in its implementation. https://www.semanticscholar.org/paper/Smart-specialisation-From-academic-idea-to-the-of-a-Foray-David/29ad6773ef30f362d7d3937c483003d974bc91c5

27.

Foray

& Europäische Kommission. (2012). Guide to research and innovation strategies for smart specialisation (RIS 3). Publications Office of the European Union.

28.

Frankfort-Nachmias

Leon-Guerrero

(2016). Social statistics for a diverse society. Sage.

29.

(2021). Innovation efficiency and the spatial correlation network characteristics of intelligent-manufacturing enterprises. Complexity, 2021, 4299045. https://doi.org/10.1155/2021/4299045

30.

Gadermann

A. M.

Guhn

Zumbo

B. D.

(2012). Estimating ordinal reliability for likert-type and ordinal item response data: A conceptual, empirical, and practical guide. Practical Assessment, Research & Evaluation, 17, 3.

31.

Gelman

Goodrich

Gabry

Vehtari

(2019). R-squared for Bayesian regression models. The American Statistician, 73(3), 307–309. https://doi.org/10.1080/00031305.2018.1549100

32.

Ghinoi

Steiner

Makkonen

Hassink

(2021). Smart specialisation strategies on the periphery: A data-triangulation approach to governance issues and practices. Regional Studies, 55(3), 402–413. https://doi.org/10.1080/00343404.2020.1791321

33.

Gianelle

Guzzo

Mieszkowski

(2020). Smart specialisation: What gets lost in translation from concept to practice? Regional Studies, 54(10), 1377–1388. https://doi.org/10.1080/00343404.2019.1607970

34.

Gianelle

Kleibrink

(2015). Monitoring mechanisms for smart specialisation strategies (JRC Technical reports). European Commission.

35.

Gianelle

Kyriakou

Mccann

Morgan

(2020). Smart specialisation on the move: Reflections on six years of implementation and prospects for the future. Regional Studies, 54(10), 1323–1327. https://doi.org/10.1080/00343404.2020.1817364

36.

GISCO: Geographical information and maps. (2020). Global competitiveness report 2020. W World Economic Forum. https://ec.europa.eu/eurostat/web/gisco/gisco-activities/map-generator

37.

Grillitsch

(2016). Institutions, smart specialisation dynamics and policy. Environment and Planning C: Government and Policy, 34(1), 22–37. https://doi.org/10.1177/0263774X15614694

38.

Grillitsch

Asheim

(2018). Place-based innovation policy for industrial diversification in regions. European Planning Studies, 26(8), 1638–1662. https://doi.org/10.1080/09654313.2018.1484892

39.

Kern

Rogge

K. S.

Howlett

(2019). Policy mixes for sustainability transitions: New approaches and insights through bridging innovation and policy studies. Research Policy, 48(10), 103832. https://doi.org/10.1016/j.respol.2019.103832

40.

Kleibrink

Gianelle

Doussineau

(2016). Monitoring innovation and territorial development in Europe: Emergent strategic management. European Planning Studies, 24(8), 1438–1458. https://doi.org/10.1080/09654313.2016.1181717

41.

Kogut-Jaworska

Ociepa-Kicińska

(2020). Smart specialisation as a strategy for implementing the regional innovation development policy—Poland case study. Sustainability (Switzerland), 12(19), 1–21. https://doi.org/10.3390/su12197986

42.

Komisja Europejska. (2014). Strategie innowacji krajowych/regionalnych na rzecz inteligentnej specjalizacji (RIS3). European Commision. https://knowledge4policy.ec.europa.eu/publication/nationalregional-innovation-strategies-smart-specialisation-ris3_en

43.

Kruschke

(2014). Doing Bayesian data analysis: A tutorial with R, JAGS, and stan (2nd ed.). Academic Press.

44.

Kudełko

Żmija

(2022). Regional smart specialisations in the light of dynamic changes in the employment structure: The case of a region in Poland. Equilibrium, 17, 133–171. https://doi.org/10.24136/eq.2022.006

45.

Kuhlmann

Ordóñez-Matamoros

(2017). Governance of innovation in emerging countries. Research handbook on innovation governance for emerging economies: Towards better models (pp. 1–34). Edward Elgar Publishing Ltd.

46.

Lewis

J. M.

Ricard

L. M.

Klijn

E. H.

(2018). How innovation drivers, networking and leadership shape public sector innovation capacity. International Review of Administrative Sciences, 84(2), 288–307. https://doi.org/10.1177/0020852317694085

47.

Lewis-Beck

Bryman

Futing Liao

(2012). The SAGE encyclopedia of social science research methods. Sage Publications, Inc.

48.

Liu

Shao

Y.-F.

Liu

(2022). An evolutionary analysis of relational governance in an innovation ecosystem. SAGE Open, 12(2), 21582440221093044. https://doi.org/10.1177/21582440221093044

49.

Lopes

Ferreira

J. J.

Farinha

(2019). Innovation strategies for smart specialisation (RIS3): Past, present and future research. Growth and Change, 50(1), 38–68. https://doi.org/10.1111/grow.12268

50.

Magro

Wilson

J. R.

(2019). Policy-mix evaluation: Governance challenges from new place-based innovation policies. Research Policy, 48(10), 103612. https://doi.org/10.1016/j.respol.2018.06.010

51.

Malanowski

Tübke

Dosso

Potters

(2021). Deriving new anticipation-based policy instruments for attracting research and development and innovation in global value chains to Europe. Futures, 128, 102712. https://doi.org/10.1016/j.futures.2021.102712

52.

McCann

Ortega-Argilés

(2016). The early experience of smart specialization implementation in EU cohesion policy. European Planning Studies, 24(8), 1407–1427. https://doi.org/10.1080/09654313.2016.1166177

53.

Meyer

(2020). Reinforcing comparative monitoring of smart specialisation performance across European regions. Journal of Entrepreneurship and Sustainability Issues, 8(2), 1386–1400. https://doi.org/10.9770/jesi.2020.8.2(81)

54.

Meyer

(2022). Social innovation governance in smart specialisation policies and strategies heading towards sustainability: A pathway to RIS4? Social Sciences, 11, 150. https://doi.org/10.3390/socsci11040150

55.

Midtkandal

Periañez Forte

Nauwelaers

(2014). RIS3 implementation and policy mixes. S3 Policy Brief Series, 07/2014. https://core.ac.uk/reader/38628832

56.

Morgan

(2016). Collective entrepreneurship: The Basque model of innovation. European Planning Studies, 24(8), 1544–1560. https://doi.org/10.1080/09654313.2016.1151483

57.

Mueller-Using

Urban

Wedemeier

(2020). Internationalization of SMEs in the Baltic Sea Region: Barriers of cross-national collaboration considering regional innovation strategies for smart specialization. Growth and Change, 51, 1471–1490. https://doi.org/10.1111/grow.12439

58.

Müller

J. M.

Buliga

Voigt

K. I

. (2018). Fortune favors the prepared: How SMEs approach business model innovations in Industry 4.0. Technological Forecasting and Social Change, 132, 2–17. https://doi.org/10.1016/j.techfore.2017.12.019

59.

Palinchak

Tsalan

Brenzovych

Kucher

Kajánek

Grešš

(2021). Competitiveness as the basis of EU regional policy: Smart specialization and sustainability. European Journal of Sustainable Development, 10, 227–239. https://doi.org/10.14207/ejsd.2021.v10n4p227

60.

Papamichail

Rosiello

Wield

(2019). Capacity-building barriers to S3 implementation: An empirical framework for catch-up regions. Innovation: The European Journal of Social Science Research, 32(1), 66–84. https://doi.org/10.1080/13511610.2018.1537844

61.

Rich

R. C.

Brians

C. L.

Manheim

J. B.

Willnat

(2018). Empirical political analysis. W empirical political analysis. Routledge. https://doi.org/10.4324/9781315109664

62.

Sarkkinen

Kässi

(2013). Characterization of innovation situation in a remote rural region. International Journal of Engineering Business Management, 5, 12. https://doi.org/10.5772/56510

63.

Sjödin

D. R.

Parida

Leksell

Petrovic

(2018). Smart factory implementation and process innovation. Research Technology Management, 61(5), 22–31. https://doi.org/10.1080/08956308.2018.1471277

64.

Stornelli

Ozcan

Simms

(2021). Advanced manufacturing technology adoption and innovation: A systematic literature review on barriers, enablers, and innovation types. Research Policy, 50(6), 104229. https://doi.org/10.1016/j.respol.2021.104229

65.

Swamidass

P. M.

Winch

G. W.

(2002). Exploratory study of the adoption of manufacturing technology innovations in the USA and the UK. International Journal of Production Research, 40(12), 2677–2703. https://doi.org/10.1080/00207540210137549

66.

Trippl

Zukauskaite

Healy

(2020). Shaping smart specialization: The role of place-specific factors in advanced, intermediate and less-developed European regions. Regional Studies, 54(10), 1328–1340. https://doi.org/10.1080/00343404.2019.1582763

67.

Uyarra

Marzocchi

Sorvik

(2018). How outward looking is smart specialisation? Rationales, drivers and barriers. European Planning Studies, 26(12), 2344–2363. https://doi.org/10.1080/09654313.2018.1529146

68.

Uyarra

Zabala-Iturriagagoitia

J. M.

Flanagan

Magro

(2020). Public procurement, innovation and industrial policy. Research Policy, 49(1), 103844. https://doi.org/10.1016/j.respol.2019.103844

69.

Vehtari

Gelman

Gabry

(2017). Practical Bayesian model evaluation using leave-one-out cross-validation and WAIC. Statistics and Computing, 27(5), 1413–1432. https://doi.org/10.1007/s11222-016-9696-4

70.

Veldhuizen

(2020). Smart specialisation as a transition management framework: Driving sustainability-focused regional innovation policy? Research Policy, 49(6), 103982. https://doi.org/10.1016/j.respol.2020.103982

71.

Voss

(1992). Successful innovation and implementation of new processes. Business Strategy Review, 3(1), 29–44. https://doi.org/10.1111/j.1467-8616.1992.tb00021.x

72.

Weller

Rainnie

(2022). Not so “smart”? An Australian experiment in smart specialisation. Geographical Research, 60(2), 300–313. https://doi.org/10.1111/1745-5871.12499

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.01 MB