The entrepreneur in the regional innovation system. Acomparative study for high- and low-income regions

The entrepreneur in the regional innovation system. A comparative study for high- and low-income regions Fernández-Serrano, J. Department of Applied Economics. Universidad de Sevilla. Faculty of Economics and Business Sciences, Av. Ramón y Cajal, 1, E-41018, Sevilla (Spain) Phone: +0034-954557551. E-mail: [email protected] Martínez-Román, Juan A. Department of Applied Economics. Universidad de Sevilla. Faculty of Economics and Business Sciences, Av. Ramón y Cajal, 1, E-41018, Sevilla (Spain) Phone: +0034-954554344. E-mail: [email protected] Romero, Isidoro (Corresponding author) Department of Applied Economics. Universidad de Sevilla. Faculty of Economics and Business Sciences, Av. Ramón y Cajal, 1, E-41018, Sevilla (Spain) Phone: +0034-954554486. E-mail: [email protected] ABSTRACT This paper investigates the influence of entrepreneurs’ characteristics on innovation in regions with different levels of development. By doing so, this work seeks to contribute to a better understanding of the role of entrepreneurs in the functioning and performance of regional innovation systems. The influence of entrepreneurs’ personal characteristics and their perceptions of the business environment on firm innovation are investigated via a survey of companies carried out in six Spanish regions. The results allow the identifying of significant differences in the main determinants of innovation in the high-income regions and low-income regions studied. Entrepreneurs’ generalised trust stimulates innovation only in high-income regions, where necessity motivation has also a negative effect on innovation. Growth ambition seems to play a highly positive role only in the case of low-income regions. Human capital and infrastructure are perceived by the entrepreneurs as the main bottlenecks for innovation in low-income regions, whereas in the case of high-income regions the legal, fiscal and financial systems are considered the key institutional barriers. These differences in the entrepreneurial factor should be taken into account in order to design and implement policies to stimulate and foster innovation in different regional contexts. JEL Codes: O31; R11; L26; M21. Keywords: innovation, entrepreneur, regional innovation systems, Spain, business environment. Introduction The intensity of innovation varies across regions with different levels of development due to the economic, socio-cultural and institutional characteristics of the territories. In this respect, the ‘innovation system’ concept (Edquist 1997; Lundvall 1992; Nelson 1993) is a powerful conceptual tool for the spatial analysis of innovation (Asheim, Lawton, and Oughton 2011; Uyarra 2010). Soon this evolutionary approach was adapted from its original national level to the study of innovation at a regional level for the purpose of analysing the interactive and cumulative process of knowledge creation in a more homogeneous and specific context (Braczyk, Cooke, and Heidenreich 1996; Cooke, Gómez-Uranga and Etxebarría 1997). Research on Regional Innovation Systems (RIS) has contributed to a better understanding of the complexity of innovation at a regional level (Asheim, Lawton, and Oughton 2011; Doloreux and Porto-Gomez 2016). However, this literature shows the existence of significant challenges, dilemmas and knowledge gaps (Doloreux 2002; Porto Gómez, Otegui Olaso, and Zabala-Iturriagagoitia 2016; Tödtling and Trippl 2005). In this sense, some authors have pointed out the essential role of micro-agents in the performance of the innovation systems (Uyarra 2010), in line with the Schumpeterian theory that emphasises the role of the entrepreneur in the process of creation and dissemination of innovations (Schumpeter 1939). Innovation literature at the micro-level of analysis has shown how the entrepreneur’s personal characteristics condition a firm’s organisation and management (Koellinger 2008; Marvel and Lumpkin 2007; Romero and Martínez-Román 2012). Yet, the entrepreneur has been overshadowed by the influence of the institutional and regulatory framework in the dominant literature on innovation systems (Acs, Autio, and Szerb 2014). New approaches have expanded the traditional view of the innovation process with a focus on institutions, emphasising the role of entrepreneurs in innovation systems. Recent contributions explore the systemic interactions between the entrepreneurial factor and the institutional context, often applying the concept of entrepreneurial ecosystem (Acs, Autio, and Szerb 2014; Stam 2015; Alvedalen and Boschma 2017). The current paper aims at contributing to a better understanding of the functioning and performance of RIS exploring two main ideas. Firstly, it postulates that the personal characteristics of the entrepreneurs and their perceptions regarding the business environment influence innovation and condition the way in which the different actors (other entrepreneurs, companies and institutions) interact in RIS. For instance, necessity entrepreneurs who do not trust other people and are risk-adverse will rarely get engaged in R&D projects with other companies in the region. Likewise, entrepreneurs with a negative perception of public administration will probably reject the possibility to collaborate with Universities, public research centres and public offices. Secondly, the paper defends the idea that the same entrepreneurs’ characteristics can produce different results in terms of innovation in regions with different structural characteristics. There are manifold possible criteria to classify regions in terms of their contextual conditions, but the level of development can capture many relevant dimensions of the economic and institutional structure of the regions. In general, high-developed regions present a better economic and institutional environment for innovation than less developed ones (human capital, financial resources, infrastructures, high quality public services and regulation, etc.). The previous literature has shown that the nature and characteristics of RIS and entrepreneurial activity vary based on the regional conditions and levels of development (Cooke, Gómez-Uranga, and Etxebarría 1997; Asheim, Lawton, and Oughton 2011; Tödtling and Trippl 2005; Fernández-Serrano and Romero 2013; Gumbau Albert 2017). The current paper postulates that the organisation and functioning of RIS in high-income regions differ from those in low-income regions. This work explores these differences, particularly regarding the role of entrepreneurs. In order to test these ideas, the paper analyses the determinants of innovation in companies located in two groups of Spanish regions, which have very different levels of development. Spain is a moderate innovator with great regional disparities (European Commission 2017), ranking among the three most heterogeneous countries in Europe regarding regional innovative performance (European Commission 2009). The first group of regions considered in this studyis made up of Andalusia, Extremadura and Murcia, which are ranked among those regions with the lowest per capita income in Spain, while the other group is composed of the Basque Country, Madrid and Navarre, which are the three richest regions in the country. This dataset allows us to investigate and test the hypotheses proposed using regression analysis techniques. The results show that the entrepreneurs’ personal characteristics –such as their motivations, growth ambition, risk-taking propensity and trust– and their perceptions of the business environment influence firms’ innovation results. The analysis also confirms that the influence of the entrepreneurs’ characteristics presents particularities in regions with different levels of development. Furthermore, significant differences are observed in the way that entrepreneurs perceive the business environment in the high- and low-income regions, affecting the innovation activities of their companies. The results of the analysis carried out have straightforward implications from a policy perspective. Policy makers should take regional particularities regarding the entrepreneurial factor into account in order to improve the effectiveness and efficiency of the actions implemented to stimulate and foster innovation in different regional contexts. Literature review and theoretical framework Regional Innovation Systems in different contexts The RIS literature presents innovation as the result of a complex process of continuous interaction between private and public agents involved in the creation, dissemination and exploitation of new technological knowledge at the regional level. From a general perspective, an innovation system is formed by the interaction between the production system, composed of a wide range of companies, the science system –comprising technology centres and research institutions– the education system –consisting of universities and official training centres– and workforce mediating institutions –such as official recruitment centres and other intermediaries between labour and companies (Autio 1998; Kaufmann and Tödtling 2001). The interactions between those systemsand agents take place in a specific institutional-regulatory and socio-cultural environment and are shaped by the characteristics of this framework (see Figure 1). The specific characteristics of the institutional context greatly influence the economy, the companies and the individuals’ entrepreneurial activity (Autio et al. 2015; Acs et al. 2016; Leyden 2016). Previous studies have comparatively analyzed the RIS of European countries (Clarysse and Muldur 2001; Tödtling et al. 2013) and between regions of the same country (Blažeket al. 2013; Martin and Moodysson 2013), as in the case of Spain (González-Pernía et al. 2012; Gumbau Albert 2017). Empirical research has made it clear that there are significant structural differences between RIS in different contexts, especially in core and peripheral regions (Oughton, Landabaso, and Morgan 2002). Some authors emphasise the importance of economic issues (Tödtling and Trippl 2005), such as the sectoral structure and specialisation (Fritsch and Stephan 2005), in the formation and characterisation of RIS. Other research stresses the influence of socio-cultural characteristics in significant aspects, such as collective learning (Hauser, Tappeiner and Walde 2007), technological change (Dohse 2000), the level of cooperation (Fritsch and Lukas 2001) and the processes of creation and diffusion of technology and innovation (Leydesdorff and Fritsch 2006). Based on those differences, RIS typologies have been developed that classify the regions according to the innovation modes, types of learning and external knowledge linkages (Isaksen and Trippl 2017). Thus, core regions, traditional manufacturing regions and peripheral regions constitute innovation contexts with differentiated economic, sociocultural, scientific-technological and business characteristics (Brown, Gregson and Mason 2016; Isaksen and Trippl 2017; Tödtling et al. 2013) and specific barriers to innovation (Tödtling and Trippl 2005). High developed regions represent favourable environments for innovation due to their greater organisational and institutional thickness (Isaksen and Trippl 2017), while the innovative activity faces greater obstacles in less developed regions (Doloreux and Dionne 2008; Kuştepeli, Gülcan and Akgüngör 2013). In this respect, the current paper postulates that the characteristics of the process through which entrepreneurs and companies pursue and achieve innovation can differ in different environments. Insert Figure 1 around here The role of entrepreneurs in Regional Innovation Systems Entrepreneurship is considered today an essential element of a competitive and sustainable RIS in the global economy. The traditional top-down view, focused on the creation and exploitation of new technologies from the close cooperation between large industrial enterprises and R&D centres (Cooke and Leydesdorff 2006), has evolved into a bottom-up perspective. According to this view, the role of entrepreneurs, small businesses and entrepreneurial culture (Beugelsdijk 2007; Fernández-Serrano and Romero 2013; Yoon et al. 2015) is becoming increasingly recognised. Entrepreneurs are at the origin of the formation of clusters and the exploitation of local resources (Feldman, Francis and Bercovitz 2005), their collaboration with local authorities being vital in the development of sustainable innovation policies (Gasbarro et al. 2017) or in the diversification and reactivation of RIS based on traditional industries (Brown et al. 2016), among other situations. The study of innovation systems at the micro level (Kaufmann and Tödtling 2001, 2002; Martínez-Román et al. 2011) leads the researcher’s attention to the influence of the entrepreneur’s personal characteristics on the management of innovation (Koellinger 2008; Marvel and Lumpkin 2007), as Figure 2 illustrates. However, the entrepreneurs’ personal characteristics and the characteristics of companies are conditioned by the institutions and economic environment in each region. The regional entrepreneurial culture (Beugelsdijk 2007, Gonzalez-Loureiro, Sousa, and Pinto 2017), the institutional setting (Yoon et al. 2015) and the regional economic structure (Fritsch and Stephan 2005) thus interact in shaping the RIS’s functioning and performance. As a result, the possibility of combining the varied range of knowledge available in core regions reduces the uncertainty of entrepreneurs and facilitates their entry into new industries (Isaksen and Trippl 2017). The thick and diversified RIS of those regions offer the entrepreneur a wide range of resources (high labour mobility, knowledge spillovers, business networks, university-business cooperation and spin-offs) to exploit business opportunities and introduce radical innovations. On the contrary, in old, less developed industrial regions, there is a lack of complementary knowledge to be sought outside the region (Isaksen and Trippl 2017). In peripheral regions, the entrepreneur faces unfavourable conditions, which are characteristics of thin RIS and hinder interactive learning and innovation, especially those of small companies (Doloreux and Dionne 2008; Dubois 2016, Isaksen and Trippl 2017). In these contexts, entrepreneurs should strive to strengthen extra-regional knowledge links (Dubois 2016, Isaksen and Trippl 2017) and links with large foreign companies located in the region (Salamonsen 2015) to increase their capacity for technological absorption. The previous considerations have led to the developing of a comprehensive systemic vision of entrepreneurship, innovation, the institutional context and their influences on economic performance (Acs, Autio and Szerb 2014; Acs et al. 2016). From this perspective, entrepreneurship is a systemic phenomenon in which institutional and environmental variables interact and have a moderating effect on the entrepreneur's decision-making (Acs, Autio and Szerb 2014, Autio et al. 2015). The entrepreneur emerges therefore as a key actor in the creation and development of entrepreneurial ecosystems (Alvedalen and Boschma 2017; Stam 2015). Insert Figure 2 around here Hypotheses development The previous literature has related the personal characteristics of entrepreneurs to several key organisational aspects in the innovation process. The entrepreneur’s personal characteristics influence the identification of business opportunities and the company’s vigilance of the market (Renko, Shrader, and Simon 2012; Wright et al. 2007). The entrepreneur’s growth ambition conditions the size of the company (Morrison, Breen, and Shameem 2003; Stam and Wennberg 2009) and the influence of firm size on innovative activity has been very often shown in the literature (Amara et al. 2008; Martínez-Román Table 1. Some economic indicators (2011) Madr id Basq ue Coun try Nav arre High - Inco meR egio ns Extrem adura Murc ia Anda lusia Low - Inco meR egio ns SPAI N Population (number of inhabitants) 6,394 ,239 2,183 ,134 637, 099 1,102,3 00 1,459 ,076 8,332 ,087 46,66 7,175 Population (Spain=100) 13.7 4.7 1.4 2.4 3.1 17.9 100 GDP per capita (euros) 31,04 1 29,84 2 28,5 33 29,8 05 15,916 18,64 6 17,31 8 17,2 93 22,90 3 GDP per capita (Spain=100) 135.5 130.3 124. 60 130. 0 69.5 81.4 75.6 76.0 100.0 Shadow economy (over GDP)* 14.9 18.6 17.5 17.0 29.8 29.4 33.2 30.8 24.4 Agriculture (over GDP) 0.1 0.7 3 1.3 6 3.7 4.9 4.9 2.6 Industry (over GDP) 9.6 25 27.8 20.8 11.7 15.9 11.4 13.0 17.2 Construction (over GDP) 6.2 8.1 7.7 7.3 10.9 9.6 9.3 9.9 8.8 Services (over GDP) 84.1 66.2 61.5 70.6 71.4 70.8 74.4 72.2 71.4 Number of enterprises (including self-employed without workers) 501,6 69 165,4 96 41,5 41 65,103 90,85 6 492,3 41 3,250, 576 Number of enterprises (per 1,000 inhabitants) 78.5 75.8 65.2 73.2 59.1 62.3 59.1 60.2 69.7 Employment in high and medium-high-technology sectors (% of total employment) 9.9 12.5 11.5 11.3 2.3 3 3.2 2.8 6.6 Total R&D expenditure (over GDP) 2 2.2 2.1 2.1 0.9 0.9 1.2 1.0 1.4 Business R&D expenditure (over GDP) 1.1 1.6 1.5 1.4 0.2 0.3 0.4 0.3 0.7 Public R&D expenditure (over total R&D) 45.0 27.3 28.6 33.3 77.8 66.7 66.7 70.0 50.0 Population aged 16 years old and over literate/no schooling (% population) 8 3.1 3.2 4.8 17.4 15.4 15 15.9 10.6 Population (16 years old and over) with tertiary education 33.3 34.8 30.6 32.9 17.8 19.6 20.3 19.2 24.3 Early school leavers ratio 19.8 13 12 14.9 29.6 30.7 34.7 31.7 28.4 Source: Calculated from data of the National Institute of Statistics (http://www.ine.es) * González-Fernández and González-Velasco (2015). The survey was conducted in the last quarter of 2010. All types of companies were included in the study with the exception of self-employed people without employees. The survey was addressed to the person having the role of entrepreneur. The entrepreneur was defined as any business owner (or co-owner) who also carried out managerial functions within the firm. The stratified sample –with quotas for firm size groups and sectors– was representative of the business population of every region included in the study with an error of ±6.0% at a confidence level of 95.5 percent. Ten pilot interviews were done before starting the fieldwork to test the questionnaire and detect possible mistakes. The companies participating in this survey were randomly selected using public information from the Spanish Official Company Register. The surveying technique used was that of computerassisted telephone interviewing (CATI). A response rate of 20.8 percent was obtained in the fieldwork. The final dataset is made up of 1,595 observations –797 cases in high-income regions and 798 in low-income ones. No bias was detected between respondents and nonrespondents. The questionnaire used in this research included queries about the innovative activities of the businesses and about different possible explanatory variables for them, considering the entrepreneurs’ personal characteristics and perceptions and the characteristics of the firm’s management. Dependent variable: Innovativeness The business owners/managers were asked whether in the previous three years they had introduced four different types of innovation and, if so, they were asked to assess the level of innovativeness of these developments using a 7-item Likert scale. Value 0 was assigned to those companies that had not carried out the innovation activities and value 7 to those companies that had introduced the most ambitious developments in each area. The four types of innovation activities were the following: a) Introduction of new products (prod_innov). b) Introduction of new production processes (proc_innov). c) Entrance into new markets (mark_innov). d) Changes in internal organisation (org_innov). Table 2. Descriptive indicators for innovation variables High-income regions Low-income regions Value* S. D. Value* S. D. Innovation prod_innov 2.59 2.71 2.40 2.71 proc_innov 1.27 2.28 1.25 2.28 mark_innov 1.14 2.22 1.20 2.22 org_innov 1.39 2.45 1.53 2.45 (*) Value = average value. S.D. = Standard Deviation. Table 2 shows some descriptive indicators for these innovation variables. A significant correlation exists between these innovation indicators. Hence, the principal component analysis technique was employed in order to capture the latent dimensions associated with the innovation variables. This methodological approach has often been used in previous research on innovation (Gatignon et al. 2002; Martínez-Román and Romero 2017).The first principal component accounts for 56.15% per cent of the total variance. Appendix 1 presents the variable-factor correlations, the communalities and other details of this analysis. The econometric model presented in the following section seeks to explain this first principal component (innovation), capturing the companies’ level of innovativeness, using different explanatory variables with a linear regression specification. Explanatory variables The following control variables, regarding general characteristics of the companies, were included in the econometric models: - Number of employees (emp) as an indicator of firm size. - Two sectorial dummiesfor manufacturing industry (ind) and services (serv). Furthermore, a set of variables related to internal business practices favouring innovation were also considered as controls: - Innovation expenditure (inn_exp). This dummy variable takes value 1 for those businesses that made some expenditure in innovation-oriented activities in the previous three years and 0 in the negative case. - Training activities for workers (train). The entrepreneurs were asked whether they regularly carried out training activities in their businesses in order to improve the capabilities of their workers and develop new skills. The answers were coded using a binary variable. - Formal business planning of the different management areas (finance, marketing, logistics, human resources, etc.) (plan). This dummy variable takes value 1 for the companies which undertook formal business planning on a regular basis, and 0 in the negative case. - Alertness and identification of new markets and business opportunities (alert). This dummy variable takes value 1 for the companies which developed specific activities oriented at identifying new markets and business opportunities and 0 in the negative case. Three groups of variables were included in the model to test the hypotheses proposed in the previous section. The first group of variables captures entrepreneurs’ personal characteristics as follows:1 - Intrinsic entrepreneurial motivation (int_mot). The entrepreneurs interviewed were asked about their level of agreement with the following statement: “I became an entrepreneur because this was the best option for my personal and professional development”. The answers were coded as ordinal variables which take values from 1 to 7 –7 meaning full agreement and 1 complete disagreement. - Necessity entrepreneurial motivation (nec_mot): The following statement was proposed to capture necessity motivation: “I became a self-employed person because I did not have another option (I was unemployed)”. The answers were coded as in the previous variable. - Ambition for growth (ambition): The entrepreneurs interviewed were asked about their level of agreement with the following statement: “The idea that my business grows is very attractive to me”(7-item Likert scale). - Risk taking (risk): The entrepreneurs interviewed were asked about their level of agreement with the following statement: “In general, a tendency to undertake highly-risky projects exists in my business” (7-item Likert scale). - Trust (trust): The entrepreneurs were asked about their perception on the level of compliance with informal agreements in the local environment (7-item Likert scale). Secondly, a set of variables approaching the entrepreneurs’ perceptions on the business environment was considered. The managers/business owners were asked to assess the importance of the following obstacles (7-item Likert scale): - Tax burden (tax) - Difficulty in finding qualified workers (qual_work) - Administrative and legal obstacles (adm_leg) - Difficulty in getting funding (funding) - Poor infrastructures (infrast) Finally, four variables were also considered to capture the access of external information and knowledge that can favour innovation activities as a result of interactions within the different agents participating in the RIS: - R&D cooperation with other companies (coop_bus). The managers/business owners interviewed were asked to assess the intensity of the R&D cooperation activities (if any) with other companies, using a 4-item scale (from no cooperation to intense cooperation). - R&D cooperation with universities and research centres (coop_univ). The managers/business owners interviewed were asked the same question but regarding their possible collaboration with universities or other technological centres (4-item scale). - Suggestions or petitions from clients (client). The entrepreneurs interviewed were asked whether the suggestions or petitions from their clients were used in their businesses as sources to increase their knowledge or information. The answers were coded using a binary variable. Table 3. Descriptive indicators for the explanatory variables. High-income regions Low-income regions Value* S. D. Value* S. D. Organisation general characteristics emp 9.97a 0.44 9.04a 0.37 ind 11.2b 0.32 10.4b 0.31 serv 50.2b 0.50 44.9b 0.50 Organisation internal factors inn_exp 69.4b 0.46 62.5b 0.48 plan 44.9b 0.50 43.7b 0.50 train 70.0b 0.46 72.4b 0.45 alert 66.2b 0.47 64.4b 0.48 Entrepreneurs’ personal characteristics int_mot 4.28a 2.43 4.39a 2.37 nec_mot 3.02a 2.45 3.19a 2.51 ambition 5.05a 2.08 5.66a 1.87 risk 3.13a 1.83 3.22a 1.95 trust 5.34a 1.69 4.80a 1.92 Entrepreneurs’ perception of the business environment tax 5.00a 2.19 5.42a 2.01 qual_work 3.66a 2.37 3.52a 2.38 adm_leg 3.84a 2.36 4.10a 2.35 funding 3.87a 2.45 4.32a 2.52 infrastr 2.40a 1.90 2.64a 2.02 Companies’ relationships with other actors coop_bus 1.38a 0.81 1.40a 0.85 coop_univ 1.22a 0.60 1.23a 0.64 client 50.4b 0.50 50.3b 0.50 supply 45.2b 0.50 49.1b 0.50 (*) Value = (a) average value (b) percentage of 1 –in the case of binary variables. S.D. = Standard deviation. - Suggestions or petitions from suppliers (supply): The entrepreneurs were asked the same previous question but regarding their suppliers (dummy). Some descriptive indicators for the independent variables in the analysis are presented in Table 3. Next, the effect of the independent variables in the innovativeness variable derived from the principal component analysis was observed by means of estimating linear regressions. For this analysis, companies were grouped depending on the level of development of the regions in which they were settled. Thus, separated regressions were estimated for high-income and low-income regions. The condition indices are lower than 20 and the variance inflation factors are around 1. Therefore there is not a serious problem of multicollinearity in the analysis. Results and Discussion The econometric results for high- and low-income regions are presented in Table 4. Regarding the control variables, firstly, industry (ind) is statistically significant, with a positive sign in both types of regions. This indicates that the level of innovativeness is higher in the industry sector, as could be expected. In contrast, the sign of the coefficient for services (serv) –though it is not significant– is negative. In the case of firm size (emp), the variable is not statistically significant and the sign of the coefficient is uncertain (around zero). Table 4. Regression results. Dependent variable = Innovation. High-income regions Low-income regions B S.E. St.B Sig. B S.E. St.B Sig. constant -2.010 .166 *** -1.700 .197 *** Organization general characteristics emp .000 .001 .007 .000 .001 .012 ind .418 .009 .132 *** .295 .111 .089 *** serv -.019 .063 -.010 -.032 .069 -.016 Organisation internal factors inn_exp .405 .067 .188 *** .443 .069 .213 *** plan .327 .064 .164 *** .194 .068 .096 *** train .057 .068 .026 .162 .075 .072 ** alert .303 .065 .144 *** .163 .071 .077 ** Entrepreneurs’ personal characteristics int_mot .009 .012 .021 .022 .014 .052 nec_mot -.024 .012 -.060 ** -.005 .013 -.012 ambition .021 .015 .043 .040 .018 .074 ** risk .056 .016 .102 *** .039 .017 .076 ** trust .035 .017 .060 ** .009 .017 .018 Entrepreneurs’ perception of the business environment tax .027 .014 .060 ** .031 .017 .061 * qual_work .024 .013 .058 * .053 .014 .124 *** adm_leg .028 .013 .065 ** -.030 .015 -.069 * funding .026 .013 .064 ** -.013 .014 -.033 infrastr -.021 .016 -.040 .042 .017 .084 ** Companies’ relationships with other actors coop_bus .212 .041 .173 *** .055 .042 .046 coop_univ .068 .054 .040 .108 .056 .069 * client .088 .066 .044 .244 .073 .121 *** supply .149 .064 .074 ** .004 .073 .002 Note: S.E.= Standard Error. St. B = Standard coefficient. * Significant at the 0.10 level. **Significant at the 0.05 level. ***Significant at the 0.01 level. N = 798 in the case of high-income regions and 797 in the case of low-income regions. R2 adjusted = 0.355 in the case of high-income regions and 0.205 in the case of low-income regions. Both models are significant at the 0.000 level. Regarding the organisations’ internal factors, all the variables are significant in both types of regions with the unique exception of the training variable (train) in the case of high-income regions. This result can be explained by the more advanced education system in this type of regions, as can be seen in the indicators reported in Table 1. This could imply that the activities for improving the skills within the companies are not so relevant for innovation in the case of the high-income regions. On the contrary, in the case of lowincome regions, the deficiencies in the skills of the human resources can represent a severe constraint. Therefore, the investment in training programmes in low-income regions has a significant return in terms of innovation. It is also worth noting that the size of the standardised coefficients for planning (plan) and alertness (alert) is larger in the case of the high-income regions, suggesting that these factors play a more significant role in these regions than in the low-income ones. This could be due to the different approaches towards innovation in high- and low-income regions which have often been pointed out in the previous literature (Oughton, Landabaso and Morgan 2002; Tödtling and Trippl 2005). In high-income regions, the role of R&D and formal innovation processes targeting radical innovations is more important, requiring rigorous planning, alertness and proactivity regarding potential innovation opportunities. In the case of the low-income regions, firms focus more often on non-R&D investments, incremental innovations or activities for mere technology adoption. Those innovations could be attained through more informal processes and require more knowledge absorption framework conditions in the RIS and therefore the entrepreneurial profiles that can better foster (or hinder) innovation in different contexts may differ. This paper provides evidence of these considerations based on firm-level data for six Spanish regions, three of them high-income regions and three of them low-income ones. In this respect, the paper shows that the entrepreneurs’ personal characteristics – motivations, risk-taking, ambition and trust - influence business innovation in a different manner in high- and low-income regions (first hypothesis). In the case of the high-income regions included in the analysis, the necessity motivation of entrepreneurs limits business innovation, while trust favours it and growth ambition has no apparent effect. On the contrary, in the low-income regions analysed, the entrepreneur’s ambition for growth stimulates firm innovativeness, while necessity motivation and trust do not affect business innovation. The paper also shows that the entrepreneurs’ perception of the business environment –infrastructure, labour resources, tax, financial and administrative systemsinfluences business innovation in a different manner in the high- and low-income regions (second hypothesis).Particularly, in the high-income regions considered, the legal/administrative, financial and tax obstacles are perceived as the most severe constraints for innovation from the perspective of the managers/business owners. However, in the case of low-income regions, the workers’ qualification and infrastructures are observed to be the main constraints for innovative firms. In addition, the characteristics and perceptions of entrepreneurs influence their relationshipswith other agents in their RIS. In this respect, the paper shows that the cooperation relationships with other agents influence business innovation in a different manner in high- and low-income regions (third hypothesis). In the case of the high-income regions, cooperation with other companies in R&D activities and communication and collaboration with suppliers stimulate firm innovativeness, whereas in low-income regions the relationships with clients and with Universities and technological centres are the key channels that foster innovativeness. Thus, in low-income regions, characterised by lower levels of trust and a less entrepreneurial private sector, public support seems to have a more relevant role for innovation. Nonetheless, in high-income regions the inter-firm relationships in clusters are the main source of innovation. These results indicate that the most effective mechanisms and actions to support innovation in low-income regions can partially differ with respect to the ones that are effective in high-income regions. From this perspective, it is not possible to replicate in low-income areas the organisation and functioning of RIS in high-income areas. Policy makers should adapt their programmes to the characteristics of the RIS and the business environment in each type of region, paying particular attention to the entrepreneurial culture in the area. Innovation policy should take into account in its design and implementation that in high-income regions cooperation between companies is an effective mechanism to support innovation and, in this sense, it is better to target those entrepreneurs with high levels of generalised trust who can more frequently develop cooperative behaviours. On the contrary, innovation policy should avoid locating resources for necessity entrepreneurs, since they are observed to be less innovative. Yet, these necessity entrepreneurs should not be relegated from the perspective of public incentives in low-income regions, since they are not associated with worse results in terms of innovation in this specific context. Nevertheless, policy action in low-income regions should focus on ambitious entrepreneurs with growth aspirations (regardless of their entrepreneurship motivations) in order to achieve better results in terms of innovation. Regarding the conditions of the business environment, in high-income regions, the legal/administrative, financial and tax obstacles should be the priority aspects to be improved in order to stimulate firm innovation. On the contrary, in the case of low-income regions, the workers’ qualification and infrastructures should be attended as a priority. This paper focuses on a set of Spanish regions and therefore its results cannot be directly generalised. Yet, the considerations presented in the analysis could be relevant for other regions. In this respect, further comparative research in other regions and countries is needed in order to confirm the findings of this paper. Furthermore, the comparison between high- and low-income regions is a simplified approach that could be enriched with other criteria and considerations. The results presented in this paper indicate the existence of different patterns of insertion into value chains for companies in high- and low-income regions, which could be related to firm innovativeness. The data for this study have not allowed delving into this interesting issue that would deserve more attention in future research. Disclosure statement No potential conflict of interest was reported by the authors. Funding This work was supported by the Regional Government of Andalusia (Spain) and the European Regional Development Fund (ERDF) under the Grant number [P09-SEJ-4857]; and the Ministry of Economy and Competitiveness Spanish under the Grant number [ECO2013-42889-P] (National Plan for Scientific and Technical Research and Innovation). Notes 1. Entrepreneurs’ education was also considered as an explanatory factor, but it turned out to be non-significant in the econometric models. This does not mean entrepreneurs’ education has no effect on innovation, but its effect comes through other variables in the model, as suggested by Romero and Martínez-Román (2012). References Acs Z. J., D. B. Audretsch, E. E. Lehmann, and G. Licht. 2016. “National Systems of Entrepreneurship.” Small Business Economics 46 (4): 527–535. Acs Z. J., E. Autio, and L. Szerb. 2014. “National Systems of Entrepreneurship: Measurement Issues and Policy Implications.” Research Policy 43 (3): 476–494. Acs Z.J., P. Braunerhjelm, D.B. Audretsch, and B.Carlsson. 2009. “The knowledge spillover theory of entrepreneurship.” Small Business Economics 32: 15–30. Alvedalen, J., and R. Boschma. 2017. “A critical Review of Entrepreneurial Ecosystems Research: Towards a Future Research Agenda.” European Planning Studies 25 (6): 887– 903. Amara N., R. Landry, N. Becheikh, and M. Ouimet. 2008. Learning and Novelty of Innovation in Established Manufacturing SMEs.” Technovation 28(7): 450–463. Ar, I. M., and B.Baki. 2011. “Antecedents and performance impacts of product versus process innovation. Empirical evidence from SMEs located in Turkish science and technology parks.”European Journal of Innovation Management 14(2), 172-206. Asheim B. T., H. Lawton Smith, and C. Oughton. 2011. “Regional Innovation Systems: Theory, Empirics and Policy.” Regional Studies 45 (7): 875–891. Autio E. 1998. “Evaluation of RTD in Regional Systems of Innovation.” European Planning Studies 6 (2): 131–140. Autio E., M. Kenny, P. Mustar, D. S. Siegel, and M. Wright. 2015. “Entrepreneurial Innovation: The Importance of Context.” Research Policy 43 (7): 1097–1108. Baron R. A., and J. Tang. 2011. “The Role of Entrepreneurs in Firm-Level Innovation: Joint Effects of Positive Affect, Creativity and Environmental Dynamism.” Journal of Business Venturing 26 (1): 49–60. Beugelsdijk S. 2007. “Entrepreneurial Culture, Regional Innovativeness and Economic Growth.” Journal of Evolutionary Economics 17 (2): 187–210. Blažek J., P. Žížalová, P. Rumpel, K. Skokan, and P. Chládek. 2013. “Emerging regional innovation strategies in Central Europe: institutions and regional leadership in generating strategic outcomes.” European Urban and Regional Studies 20 (2): 275–294. Block J., and P. Sandner P. 2009. “Necessity and Opportunity Entrepreneurs and their Duration in Self-Employment: Evidence from German Micro Data.” Journal of Industry, Competition and Trade 9 (2): 117–137. Braczyk H. J., P. Cooke, and M. Heidenreich. 1996. Regional Innovation Systems. London: UCL Press. Brown R., G. Gregson, and C. Mason. 2016. “A post-mortem of regional innovation policy failure: Scotland's Intermediate Technology Initiative (ITI).” Regional Studies 50(7): 1260–1272. Clarysse B., and U. Muldur. 2001. “Regional cohesion in Europe? An analysis of how EU public RTD support influences the techno-economic regional landscape.” Research Policy 30 (2): 275–296. Cooke P., M. Gómez-Uranga, and G. Etxebarría. 1997. “Regional Systems of Innovation: Institutional and Organizational Dimensions.” Research Policy 26 (4-5): 475–491. Cooke P., and L. Leydesdorff. 2006. “Regional Development in the Knowledge-Based Economy: The Construction of Advantage.” The Journal of Technology Transfer 31 (1): 5–15. Cooke P., and D. Wills. 1999. “Small Firms, Social Capital and the Enhancement of Business Performance Through Innovation Programmes.” Small Business Economics 13 (3): 219–234. Dohse D. 2000. “Technology Policy and the Regions — the Case of the BioRegio Contest.” Research Policy 29 (9): 1111–1133. Doloreux D. 2002. “What we Should Know About Regional Systems of Innovation.” Technology in Society 24 (3): 243–263. Doloreux D., and S. Dionne. 2008. “Is regional innovation system development possible in peripheral regions? Some evidence from the case of La Pocatire, Canada.” Entrepreneurship & Regional Development 20 (3): 259–283. Doloreux D., and I. Porto-Gomez. 2017. “A Review of (Almost) 20 years of Regional Innovation Systems Research.” European Planning Studies25 (3), 371-387. Dubois A. 2016. “Transnationalising entrepreneurship in a peripheral region – The translocal embeddedness paradigm.” Journal of Rural Studies 20 (2): 1–11. Edquist C. 1997. Systems of Innovation. Technologies, Institutions and Organizations. London: Pinter Publishers. European Commission. 2009. “Regional Innovation Scoreboard –RIS. Enterprise & Industry Magazine.” Inno Europe Paper 14. European Communities. European Commission. 2017. “Regional Innovation Scoreboard –RIS. Internal Market, Industry, Entrepreneurship and SMEs.” European Union. Feldman M., J. Francis, and J. Bercovitz. 2005. “Creating a cluster while building a firm: entrepreneurs and the formation of industrial clusters.” Regional Studies 39 (1): 129–141. Fernández–Serrano J., and I. Romero. 2013. “Entrepreneurial Quality and Regional Development: Characterizing SME Sectors in Low Income Areas.” Papers in Regional Science 92 (3), 495–513. Fernández-Serrano J., and I. Romero. 2014. “About the Interactive Influence of Culture and Regulatory Barriers on Entrepreneurial Activity.” International Entrepreneurship and Management Journal 10 (4): 781–802. Forsman H. 2011. “Innovation Capacity and Innovation Development in Small Enterprises. A comparison Between the Manufacturing and Service Sectors.” Research Policy 40 (5): 739–750. Fritsch M., and R. Lukas. 2001. “Who Cooperates on R&D?” Research Policy 30 (2): 297– 312. Fritsch M., and A. Stephan. 2005. “Regionalization of Innovation Policy– Introduction to the Special Issue.” Research Policy 34 (8): 1123–1127. Gasbarro F., E. Annunziata, F. Rizzi, and M. Frey. 2017. “The interplay between sustainable entrepreneurs and public authorities: evidence from sustainable energy transitions.” Organization & Environment 30 (3): 226–252. Gatignon H, M. L.Tushman, W. Smith, and P. Anderson.2002.“A structural approach to assessing innovation construct development on innovation locus, type, and characteristics.”Management Science 48(9): 1103–1122. González-Fernández M., and C. González-Velasco. 2015. “Analysis of the shadow economy in the Spanish regions.” Journal of Policy Modeling 37 1049–1064. Gonzalez-Loureiro M., M. J. Sousa, and H. Pinto. 2017. Culture and Innovation in SMEs: The Intellectual Structure of Research for Further Inquiry.” European Planning Studies, 25 (11): 1908–1931. González-Pernía, J. L., I. Peña-Legazkue, and F. Vendrell-Herrero. 2012. “Innovation, entrepreneurial activity and competitiveness at a sub-national level.” Small Business Economics 39 (3): 561–574. Gumbau Albert M. 2017. “Entrepreneurship, Innovation and Regional Performance: Application for the Spanish Regions.” Entrepreneurship & Regional Development 29 (3- 4): 271–291. Guzmán J., and J. Santos. 2001. “The Booster Function and the Entrepreneurial Quality: An Application to the Province of Seville.” Entrepreneurship & Regional Development 13 (3): 211–228. Hauser C., G. Tappeiner, and J. Walde. 2007. “The Learning Region: The Impact of Social Capital and Weak Ties on Innovation.” Regional Studies 41 (1): 75–88. Isaksen A., and M. Trippl. 2017. “Innovation in space: the mosaic of regional innovation patterns.” Oxford Review of Economic Policy 33 (1): 122–140. Kaufmann A., and F. Tödtling. 2001. “Science–Industry Interaction in the Process of Innovation: The Importance of Boundary-Crossing Between Systems.” Research Policy 30 (5): 791–804. Kaufmann A., and F. Tödtling. 2002. “How Effective is Innovation Support for SMEs? An Analysis of the Region of Upper Austria.” Technovation 22 (3): 147–159. Koellinger P. 2008. “Why are some entrepreneurs more innovative than others?” Small Business Economics 31 (1): 21–37. Kuştepeli Y., Y. Glcan, and S. Akgngör. 2013. “The innovativeness of the Turkish textile industry within similar knowledge bases across different regional innovation systems.” European Urban and Regional Studies 20 (2): 227–242. La Porta R., and A.Shleifer. 2014. “Informality and Development.”Journal of Economic Perspectives28(3), 109–126. Leyden D. P. 2016. “Public-Sector Entrepreneurship and the Creation of a Sustainable Innovative Economy.” Small Business Economics 46 (4): 553–564. Leydesdorff L., and M. Fritsch. 2006. “Measuring the Knowledge Base of Regional Innovation Systems in Germany in Terms of a Triple Helix Dynamics.” Research Policy 35 (10): 1538–1553. Lundvall B.-. 1992. National Systems of Innovations. London: Pinter Publishers. Martin R., and J. Moodysson. 2013. “Comparing knowledge bases: on the geography and organization of knowledge sourcing in the regional innovation system of Scania, Sweden.” European Urban and Regional Studies 20 (2): 170–187. Martínez–Román J. A., J. Gamero, and J. A. Tamayo. 2011. “Analysis of Innovation in SMEs Using an Innovative Capability-Based Non-Linear Model: A Study in the Province of Seville (Spain).” Technovation 31 (9): 459–475. Martínez-Román J. A., and I. Romero. 2017. “Determinants of Innovativeness in SMEs: Disentangling Core Innovation and Technology Adoption Capabilities.” Review of Managerial Science 11 (3): 543-569.