1 Introduction

Entrepreneurship is widely recognised as key to economic development and addressing societal challenges. It is therefore not surprising that promoting entrepreneurship is high on the agenda of policymakers. Supporting opportunity recognition as the first necessary step in conducting entrepreneurial activities (Riquelme, 2013; Shane & Venkataraman, 2000; Venkataraman, 1997) is therefore of essential relevance for entrepreneurship policies. However, opportunities and their economic potential are not homogeneous, as progress in economic development is mostly the result of highly innovative entrepreneurial activities and cannot be generalised to all entrepreneurial endeavours (Autio & Yli-Renko, 1998; Shane, 2009; Szerb et al., 2019). The origin of innovative entrepreneurial activities differs as well because innovative opportunities are pursued based on different motives and require distinct conditions of contextual factors (Hayton & Cacciotti, 2013). Entrepreneurship policies aiming to stimulate entrepreneurship to enhance regional economic development therefore need fine-grained information on the regional dynamics that support regional opportunity perception, especially those that drive innovative opportunity perception.

Former research finds determinants of opportunity perception of individuals, meaning individual people, on four different levels. First, the individuals who pursue business opportunities differ in their personality traits and demographic variables from those who do not act upon opportunities (actor level) (Hayton & Cacciotti, 2013; Rinne et al., 2012; Shane, 1993; Williams & McGuire, 2010). Second, social networks of individuals matter in perceiving opportunities (network level) (Grillitsch, 2019; Ozgen & Baron, 2007; Riquelme, 2013; Spilling, 2011). Third, the economic environment of individuals impacts the perception of opportunities (environment level) (Fernandez-Serrano et al., 2019). Lastly, regional entrepreneurship culture is found to influence the opportunity recognition of individuals (culture level) (Fritsch & Kublina, 2019; Fritsch et al., 2019). All these four layers are known to influence the likelihood of perceiving opportunities. However, these determinants are often analysed either on the regional or on an individual level, which supports the call for further research to focus on an integrated perspective on entrepreneurship combining different approaches (Valliere, 2013) and the interplay of regional context and individual entrepreneurial actions (Garretsen et al., 2019; Szerb et al., 2019).

To gain better understanding of the regional dynamics that drive regional opportunity perception, it is important to comprehend entrepreneurship as a process undertaken by individuals who are embedded in a local framework with a history (Audretsch & Belitski, 2017; Autio et al., 2014; Malecki, 2018; Nayak & Maclean, 2013; Spigel, 2017). One key concept is the regional embeddedness of individuals, meaning inhabitants of a region. This can be defined as a bi-directional relationship between individuals and the economic ecosystem in which they operate. The economic evolution of the system is influenced by the agency of individuals, while the economic systems themselves, at least in part, determine the agency of individuals in undertaking entrepreneurial activities by providing specific configurations of opportunities and economic returns (Huggins & Thompson, 2020). The agency of individuals in regional economic systems relies on the subjective perception of their environment on four different levels—actor, network, economic environment, and culture level. Based on these considerations, the question arises:

  • RQ: How does the regional embeddedness of an individual relate to (innovative) regional opportunity perception?

To address this research question, a quantitative survey assessed the relation between the regional embeddedness of individuals on the four levels—actor, network, environment, and culture—and the likelihood of perceiving business opportunities by surveying inhabitants in Germany. To achieve this, two types of binary logistic regression analyses generate empirical evidence on this relationship. The first type of analyses aims to gain insights into regional opportunity perception in general, while the second type of analyses aims to gain insights into innovative regional opportunity perception specifically.

Besides extending the empirical evidence on how opportunity perception is related to the regional embeddedness of individuals, the study contributes in three additional ways. First, it delivers empirical support for the idea that entrepreneurial activities are driven by regional knowledge bases as conceptualised in the knowledge spillover theory of entrepreneurship. Second, the results indicate that regional embeddedness is important for regional opportunity perception, but its relevance is not equal for regional opportunity perception in general and innovative regional opportunity perception with the latter one being more independent of it. This has implications for the conceptualisation of entrepreneurship as a regional phenomenon. Third, the simultaneous consideration of multiple layers of regional embeddedness and their effects on opportunity perception enables the identification of the relational importance of each component. Overall, the generated insights can help to improve holistic policy approaches to stimulate entrepreneurial activities in a targeted way.

2 Conceptual framework

2.1 Business opportunities

Regions need people who recognise opportunities for innovation and development to enhance regional value generation (Fritsch & Wyrwich, 2018; Valliere, 2013). Although this is widely acknowledged, the definition of opportunities and their recognition vary in the field of entrepreneurship (Hansen et al., 2011; Riquelme, 2013). Definitions range from generating opportunities in a creative process, developing ideas into new ventures, and situations in which it becomes possible to deliver new values for customers (Hansen et al., 2011; Riquelme, 2013). However, these definitions have in common that opportunity recognition is considered fundamental to the entrepreneurial process as it represents the starting point (Riquelme, 2013; Shane & Venkataraman, 2000; Venkataraman, 1997). It is argued that opportunity recognition lays the foundation for developing new goods and services by positively influencing entrepreneurial intentions and decisions to find a new venture (Ferrero & Bessiere, 2016; Lindsay et al., 2009; Qian et al., 2013; Riquelme, 2013).

Despite varying definitions, there is one important type of conceptualisation of opportunities that individuals may recognise, which differentiates itself from other opportunities by the higher potential for economic development (Autio & Yli-Renko, 1998; Shane, 2009; Szerb et al., 2019). This type of opportunity is created by new connections between distinct fields of specialisation, which introduces structural change and disruption of market dynamics (Grillitsch, 2019). To pursue this kind of opportunity, innovation and challenging the way of doing things are key components (Cantner et al., 2017; Clydesdale, 2012; Schumpeter, 1934). Acting upon this type of opportunity is likely to result in different entrepreneurial endeavours with different impacts on regional development compared to the impact of the enactment of opportunities in general. Based on this reasoning, those innovative business opportunities and their recognition is contrasted to business opportunities in general in this study.

2.2 Regional opportunity perception

The critical first step in the entrepreneurial process, conceptualised as the identification, evaluation, and exploitation of these two types of opportunities, is to be sensitive to information about patterns in the environment (Ferrero & Bessiere, 2016; Riquelme, 2013; Shane & Venkataraman, 2000; Valliere, 2013). This alertness helps find unmet customer needs and innovative resource usage in their environment and, thus, perceive previously overlooked opportunities (Kirzner, 1979, 1997; Riquelme, 2013; Valliere, 2013).

Empirical research finds that the ability to perceive opportunities differs widely between individuals (Valliere, 2013). While some perceive or create opportunities regularly, others are not alert to these opportunities (Valliere, 2013). Thus, one of the fundamental research interests in entrepreneurship is to understand which characteristics and circumstances lead to the recognition of opportunities and ultimately to the persistence of regional entrepreneurship patterns (Riquelme, 2013; Shane & Venkataraman, 2000; Valliere, 2013; Venkataraman, 1997).

Many studies analyse entrepreneurs as isolated and homogenous entities and neglect the embeddedness of entrepreneurs in a broader context (Ferreira et al., 2017; Goktan & Gunay, 2011), even though it is widely recognised that the entrepreneurial ecosystem determines whether an individual becomes an entrepreneur and influences the type of entrepreneurial activity and the efficiency of the entrepreneurial function (Andersson & Koster, 2011; Audretsch & Belitski, 2017; Autio et al., 2014). This approach overlooks context and differences between types of entrepreneurship and opportunities by ignoring this dependence on the regional context and, thus, repeats problematic assumptions (Hayton & Cacciotti, 2013; Wheadon & Duval-Couetil, 2019). Rather than individual traits of entrepreneurs alone, their interaction with context variables shapes the entrepreneurial process (Aldrich & Cliff, 2003; Audretsch & Belitski, 2017; Fernandez-Serrano et al., 2019; Ferreira et al., 2017; Goktan & Gunay, 2011; Mason & Brown, 2014; Veciana & Urbano, 2008). Local influence is transferred through the experience of situations taking place in a specific regional context, which leads to a regionally bounded human agency (Huggins & Thompson, 2019, 2020). In this vein, entrepreneurship is interpreted as a social process that historically builds upon specific groups of people in a regional bounded context (Florida et al., 2017; Huggins & Thompson, 2020). Thus, it is highly relevant to understand entrepreneurship as a process conducted by individuals embedded into a local framework with history (Audretsch & Belitski, 2017; Autio et al., 2014; Nayak & Maclean, 2013). Based on these considerations, regional embeddedness in this study is defined as a bi-directional relation between individuals that act upon their agency in economic ecosystems and thereby form the economic evolution of the system, and the economic systems that, at least partially, determine the individual’s agency in undertaking entrepreneurial activities by providing specific configurations of opportunities and economic returns (Huggins & Thompson, 2020).

Based on this reasoning, opportunity recognition as the first critical step in the entrepreneurial process cannot be understood in isolation but requires considering the characteristics of the entrepreneur and their embeddedness in the regional context (Fernandez-Serrano et al., 2019; Huggins & Thompson, 2020; Riquelme, 2013; Sarason et al., 2010). This means that the existence of opportunities alone is not sufficient for entrepreneurial activities to take place, but environmental conditions and characteristics of the entrepreneur are of relevance (Fernandez-Serrano et al., 2019; Huggins & Thompson, 2020; Riquelme, 2013; Szerb et al., 2019).

However, people embedded in a regional context are not fully knowledgeable about the regional conditions and the structure of the system (Huggins & Thompson, 2020). Thus, not the objective evaluation of the environmental conditions, but the subjective perceptions of these are important. The perception represents the filter through which regionally embedded people see regional opportunities and the type of opportunities they perceive (Audretsch & Belitski, 2017; Fernandez-Serrano et al., 2019; Ferreira et al., 2017).

For regional opportunity perception, relevant dimensions, in which a regional inhabitant can be imbedded in, can be derived from the frameworks of regional innovation systems and entrepreneurial ecosystems. Common understandings in these strings of literature are that regional resources are the basis for the economic structure (Spigel, 2017). Both strings highlight the role of contextual factors as regional culture and institutions, social networks, and knowledge available in the region, which can be detected in the economic structure besides the characteristics of individual actors (Acs et al., 2017; Cooke et al., 1997; Spigel, 2017). Therefore, in this study, the regional opportunity perception is conceptualised as a process that relies on regional opportunities provided by the regional system. These can not be perceived by the individual agent directly, but only via its subjective perception of the regional conditions on actor, network, environment, and culture level. This conceptualisation is consistent with the notion that entrepreneurial ecosystems can be interpreted as a combination of regional and individual characteristics (Pugh et al., 2021; Stam, 2014). Figure 1 illustrates the conceptualisation of regional opportunity perception, which is applied in this study.

Fig. 1
figure 1

Conceptual framework. Source: Own illustration

Full size image

2.2.1 Regional embeddedness on actor level

The impact of personality traits on entrepreneurial activities has been studied extensively in former research. Starting with Schumpeter (1934), who describes entrepreneurs as breakers of routines and going against the odds, the prototypical innovative entrepreneur is often described as an individualist with high inner-directedness (Cantner et al., 2017; Krueger et al., 2000). Furthermore, uncertainty acceptance, individualism, and power proximity relate to innovative entrepreneurship (Hayton & Cacciotti, 2013; Rinne et al., 2012; Shane, 1993; Williams & McGuire, 2010). However, the influence of personality traits on entrepreneurial behaviour is temporarily unstable and more proximate measures tend to balance out the impact (Cantner et al., 2017; Hayton & Cacciotti, 2013; Shane, 1993; Wennekers et al., 2007).

On the regional level, there are attempts to measure the influence of aggregated personality traits on regional economic growth. Based on the geographical psychology approach (Rentfrow et al., 2008), it is found that personality traits influence local economic growth by impacting entrepreneurial activities (Garretsen et al., 2019; Obschonka et al., 2013, 2015). Reasons for geographical patterns of personality traits are argued to be local role-modelling processes and selective migration, i.e. people with similar personality traits are attracted to regions that provide a good fit (Fritsch et al., 2019; Garretsen et al., 2019; Jokela et al., 2015; Obschonka et al., 2013; Rentfrow et al., 2008). Thus, the fit of personality traits with the perceived dominant personality in the regional context is acknowledged to be relevant for entrepreneurial activities. However, there is a research gap in understanding how the good fit between individual and regional personality influences entrepreneurial behaviour (Huggins & Thompson, 2020).

If one identifies himself or herself strongly with other inhabitants of the region, it is hypothesised that the person assumes that his or her perceived shortage in products or services may be of value for others in the region as well and, thus, is more likely to perceive regional business opportunities. Therefore, the first hypothesis states:

  • Hypothesis 1: The perceived personality fit of an individual in a region relates to the regional business opportunity recognition.

2.2.2 Regional embeddedness on network level

A further widely acknowledged aspect of entrepreneurship is its interdependence with social networks. It is argued that it is unlikely that entrepreneurs develop their business in isolation, but rather the embeddedness into a social structure matters in the entrepreneurial process (Almeida & Teixeira, 2017; Rae, 2005; Riquelme, 2013). The opportunity perception of the entrepreneur is even found to be determined by the personal networks, and the maintenance and development of these personal networks are acknowledged to be a key success factor (Grillitsch, 2019; Ozgen & Baron, 2007; Riquelme, 2013; Spilling, 2011). Former research finds a positive relationship between the number of social contacts, especially the usage of weak ties, and opportunity perception (Riquelme, 2013). However, the relevance of strong and weak ties for opportunity perception varies across different societal structures as collective and individualistic societies (Ma et al., 2011).

The most relevant groups in an entrepreneur’s social network are friends, family, and colleagues (Siemens, 2010; Zhu et al., 2019). Especially in the case of inexperienced students, entrepreneurial friends and their emotions towards entrepreneurship play a crucial role in the entrepreneurial intention (Zhu et al., 2019). Thus, social contacts act as role models and provide resources that are even more relevant at the beginning of the entrepreneurial process (Siemens, 2010; Zhu et al., 2019).

The observation of role models in the social network and their attitude towards entrepreneurship can be understood as norms that establish themselves through social interactions and are largely subconscious (Andersson & Koster, 2011; Arenius & Minniti, 2005; Cantner et al., 2017; Fritsch et al., 2019; Laine, 2016; Minniti, 2005). Indeed, the presence of role models, defined as accessible, successful individuals who increase the acceptability of entrepreneurial activities, is often associated with the efficiency of entrepreneurial ecosystems (Andersson & Koster, 2011; Bosma et al., 2012; Fritsch et al., 2019; Lafuente et al., 2007; Zozimo et al., 2017). Although this mechanism seems to be relevant for entrepreneurial intentions in many cases, it is found that the relevance of these norms is moderated by group identification (Cantner et al., 2017). Thus, it can be suspected that embeddedness in a regional context is also highly relevant to evaluating regional role models’ efficiency because it governs how present region-specific role models are for an individual. As a result, the second hypothesis is formulated:

  • Hypothesis 2: The share of social contacts of an individual located in the same region relates to regional business opportunity recognition.

2.2.3 Regional embeddedness on environment level

The persistence of self-employment proportions in a region can be partly explained by the persistence of regional resources and regional industrial structure (Fritsch & Kublina, 2019; Qian et al., 2013). Especially market structures characterised by dynamic developments, e.g. a growing industry, increased demand for products, and technological development, are associated with higher opportunity perception (Riquelme, 2013). Moreover, the resources and the knowledge that spills over from established firms in the region enable entrepreneurial ventures to excel and allow for more resource and knowledge combinations (Feldman, 2014; Gonzalez-Pernia et al., 2012).

Although these findings support the hypothesis that market heterogeneity may cause opportunity perception, former research argues that market diversity is rather a symptom of high levels of new business formation than a source (Fritsch & Kublina, 2019; Riquelme, 2013). Accordingly, new ventures, especially innovative ones, can be interpreted as manifestations of regional knowledge spill-overs (Acs et al., 2009, 2013; Fritsch & Wyrwich, 2018). Thus, the subjective perception of innovative new ventures is hypothesised to relate to the recognition of underlying knowledge resources available in the region, which governs opportunity perception (Fernandez-Serrano et al., 2019). Therefore, the third hypothesis is stated as follows:

  • Hypothesis 3: The perception of the business environment, especially the perception of innovative and new ventures, relates to regional business opportunity recognition.

2.2.4 Regional embeddedness on culture level

Informal institutions defined as a framework for actions perceived as desirable or appropriate within a regional context are found to form social, economic, and political interactions (Langevang et al., 2015; Urbano & Alvarez, 2014). More specifically, informal institutions that regulate entrepreneurial activities are singled out as the main component of informal institutions related to economic growth (Beugelsdijk, 2007; Garretsen et al., 2019).

These informal institutions can be interpreted as regional entrepreneurship culture, characterised by shared perceptions, expectations, and beliefs of the regional population (Almeida & Teixeira, 2017; Hayton & Cacciotti, 2013; Laine, 2016; Ogunsade & Obembe, 2016; Spigel, 2017). By providing a general orientation of the prioritisation of values within a region, the entrepreneurship culture affects not only the number of new ventures but also their types and impact on economic development (Aparicio et al., 2016; Baumol, 1990; Bowen & De Clercq, 2008; Eunni & Manolova, 2012; Ferreira et al., 2017; Laine, 2016; Mueller & Thomas, 2000; Ogunsade & Obembe, 2016; Spigel, 2017; Urbano & Alvarez, 2014). In this vein, Spigel (2017) points out two possible cultural transmission mechanisms—cultural attitudes and histories of entrepreneurship.

The regional entrepreneurship culture is observed to change only very slowly, which explains the path-dependency of regional entrepreneurial behaviour (Feldman, 2014; Fritsch & Kublina, 2019; Fritsch & Wyrwich, 2018; Fritsch et al., 2019; North, 1994; Nunn, 2012; Williamson, 2000). Furthermore, this empirical evidence hints at a self-perpetuation of the regional entrepreneurship culture via intergenerational transmission of entrepreneurial values, which tame the uncertainty of entrepreneurial activities (Fritsch et al., 2019; Laine, 2016).

However, the effect of entrepreneurial culture on the types of entrepreneurship is mixed. While some argue that the culture channels entrepreneurial talents toward productive entrepreneurship (Baumol, 1990), others find evidence for distinct influences of specific parts of the institutional framework on different types of economic sectors (Agostino et al., 2020; Valdez & Richardson, 2013). The fourth hypothesis is formulated as follows:

  • Hypothesis 4: Regional cultures of entrepreneurship in general and innovative entrepreneurship relate to regional business opportunity recognition.

3 Data and methods

Data collection

To address the research questions, an online survey in eighteen German planning regions, containing 550 subjects, was conducted between 2nd and 15th January 2020. The participants of the study were contacted via a market research company that has access to a pool of paid respondents. To participate in the study, respondents had to fulfil two main requirements. First, all participants had to be at least 18 and at maximum 65 years old, which corresponds to the age when a full capacity to contract is given and before retirement in Germany. Second, only respondents who indicated living in one of the studied regions were eligible to participate in the study. Please see Fig. 2 for an illustration of the studied regions.

Fig. 2
figure 2

German regions with long-standing entrepreneurship patterns surveyed. Source: Own illustration (Color figure online)

Full size image

Sample construction

To gain insights into the determinants of business opportunity perception in the regional context, it is of crucial relevance to avoid biases that may be introduced by the fact that over-proportionally or under-proportionally entrepreneurially prone people answer the questionnaire. It was shown in multiple studies that gender, age, and educational status are likely to impact entrepreneurial activities undertaken by the respondents (e.g. Arenius & Minniti, 2005; Gallardo & Scammahorn, 2011). Therefore, the population statistics provided by the Federal Institute for Research on Building, Urban Affairs, and Spatial Development are taken as a reference to define the proportions of respondents regarding gender, age, and educational status for the sampling via an online questionnaire. These proportions were assured by employing a quota sampling technique. The sample description regarding the quotes is illustrated in Table 7 in the Appendix.

Data cleaning

2,250 clicks were registered and 550 full questionnaires were collected, 24% of the respondents passed the quality assurance tests and continued to the end of the questionnaire. To ensure data quality sampled by the survey, a mechanism is included in the questionnaire to exclude low-quality responses. Respondent’s answers in matrix-style questions are checked for quality by assessing the variance. If the value of the variance equals zero, the respondents are screened out. After finishing the data collection, a further data-cleaning step was applied. A respondent’s share of missing values for the research-relevant questions is calculated. In case this exceeds 15%, the respondent is excluded from the data set. Although this pattern could have multiple reasons, it could hint at low motivation and therefore lower validity of the given answers (Barge & Gehlbach, 2012). This cleaning procedure resulted in removing five respondents from the sample. Lastly, the composed measures of the personality profile suffer greatly from missing values. This is overcome by mean value imputation.

Methodological approach

Binary logistic regressions consider two dependent variables: (1) perception of regional business opportunities in general and (2) perception of innovative regional business opportunities. Four sets of control variables—individual controls, regional controls, personality traits, and interest controls—are used to ensure the robustness of the results. The validity of the results is ensured by controlling for the assumption of linearity between the log of odds and continuous independent variables. Furthermore, influential data points are identified by the value of the absolute standardised residuals. No data points with a deviance of more than three standard deviations were detected and, thus, no data points had to be removed. The variance inflation factor assessment tests for multicollinearity issues. The results of the variance inflation factor assessment are illustrated in Table 8 in the Appendix. Lastly, all models are tested against a null model containing only the constant.

The model applied in this study is the following, with X corresponding to the respective dependent variable:

$$\begin{array}{c}\text{log}\left(\frac{p\left(X\right)}{1-p\left(X\right)}\right)=Constant+\beta_1\ast\;Actor\;level\;embeddedness+\beta_2\ast\;Network\;level\;embeddedness+\beta_3\\\ast Environment\;level\;embeddedness+\beta_4\ast\;Culture\;level\;embeddedness\\\begin{array}{c}+\;Regional\;controls+Individual\;Controls+Personality\;trait\;controls\\+Interest\;controls\end{array}\end{array}$$
(1)

3.1 Dependent variables

Regional opportunity perception

Two questions asked the respondent to indicate (1) whether a business opportunity within the regional environment is recognised and (2) whether an innovative business opportunity within the regional environment is recognised. The wording of the included questions is based on the German-language questions of the GEM (Global Entrepreneurship Monitor) Adult Population Survey, as asked by the GEM Germany country team in the 2018 survey year. The question formulations were adapted by including ‘within your current designated region’ in both cases and ‘innovative’ for the assessment of the innovative opportunity recognition. When respondents were asked to answer the question on the perception of innovative business opportunities, information was shown that an innovative company is a company that offers products or services that did not exist in this form before. First, respondents indicated their opportunity perception on a 5-point Likert scale. Then, this was converted to a binary variable by summarising the upper two (opportunities are perceived) and the bottom two classes (no opportunities are perceived). Finally, all respondents that indicated the middle of the Likert scale, i.e. expressing neither opportunity perception nor no opportunity perception, were removed for analysis. This procedure causes a loss in sample size, but ordinal logistic regression analyses were not applied due to multicollinearity issues and the violation of the proportional odds assumption.

3.2 Independent variables

The embeddedness of the respondent in the regional innovation system is assessed as described in the following. Since a systemic perspective requires actor, network, environment, and culture level, those were included in the questionnaire.

Actor level

On the actor level, the identification with the regional personality is operationalised with the Big 5 personality profile, which was answered individually, and once the subject was asked to give the answers for a typical person from their designated region. The assessment of the personality traits is operationalised according to Rammstedt et al. (2013). In the case of only a slight difference between those two scales, it is interpreted as a high identification with the regional personality. This is done per observation, meaning that respondents could have evaluated the ‘typical’ person from their region differently. This is likely to be caused by different people the respondents have contact with in the region. This non-random subsample of the regional inhabitants is argued to influence the evaluation of the regional personality profile but also the perceived identification with that. Therefore, this measure estimates the subjective alignment of the own and the regional personality.

Network level

The respondents indicated how many friends, family members, and colleagues live in the same region on the network level. The restriction to these three groups of social contacts is supported by the findings of Siemens (2010) and Zhu et al., (2019) confirming the relevance of these groups. However, not so much the absolute number of contacts but their regional distribution is of high relevance. A social network that is closely located to the respondent has two advantages: (1) these people can support directly and (2) those contacts also have regional information that might help to see new business opportunities. Therefore, the respondents were asked to indicate the geographical distribution of their social contacts by answering slider questions ranging from 0 to 100%.

Environment level

The environment level, more specifically defined as the economic environment, is depicted in the perceived share of young firms, innovative companies, and innovative young firms in the region. The perception of these firms is especially relevant for the identification of business opportunities as these represent the knowledge that is regionally available and can be commercialised (Acs et al., 2009, 2013; Fritsch & Wyrwich, 2018). This information was surveyed by asking the respondent to estimate the share of these types of businesses in their designated region, ranging from 0 to 100%.

Culture level

The culture level is included with two dummy variables that describe whether a long-standing entrepreneurship pattern of high or low general entrepreneurship and high or low innovative entrepreneurship is present in the designated region of the respondent. For the first dummy variable, general entrepreneurship culture, a value of 0 indicates that the designated region of an individual shows low entrepreneurship rates consistently over a long period and a value of 1 indicates that the designated region of an individual shows high entrepreneurship rates over a long period. Analogous to the construction of the entrepreneurship culture in general, the dummy variable innovative entrepreneurship culture is created. For this, a value of 0 indicates that the designated region of an individual shows low innovative entrepreneurship rates consistently over a long period and a value of 1 indicates that the designated region of an individual shows high innovative entrepreneurship rates over a long period.

To ensure the accuracy of the cultural variables, it is necessary to limit data sampling to regions with a long-standing entrepreneurship pattern. Each region has a specific cultural configuration, leading to a culturally bounded assessment of entrepreneurial activities (Spigel, 2017). Regions with a long-standing entrepreneurship pattern are argued to have manifestations of entrepreneurial cultures that reinforce the culture. For instance, they may have the expertise and infrastructure to support specific types of entrepreneurial activities (Spigel, 2017). Comparing regions with less stable entrepreneurship patterns to those with more stable patterns may lead to a comparison of established support infrastructure and schemes for entrepreneurial activities, rather than entrepreneurial culture.

For this purpose, regions with long-standing entrepreneurship patterns are identified. German planning regions (‘Raumordnungsregionen’) are especially suitable for this because these regions build the geographical framework for the analysis of large-scale development tendencies and allow statements on disparities in infrastructure and labour market structure (Federal Institute for Research on Building, Urban Affairs, and Spatial Development, n.d.). The selection of regions was guided by data published by Fritsch and Wyrwich in 2018 and 2019, based on German planning regions (Fritsch & Wyrwich, 2018, 2019). The assignment of Fritsch & Wyrwich (2019) of each German planning region to a quintile in terms of self-employment rates in 1907 and 1925 and the average rates between 2010 and 2016 were used to identify the regions with a general long-standing entrepreneurship pattern. The regions that showed the lowest variance in the assignment into a specific quintile over these three observations were considered to inhibit a long-standing general entrepreneurial pattern. The same procedure was applied based on the data published by Fritsch & Wyrwich (2018) focusing on technology-intensive and science-based industries. Self-employment rates in 1907 and 1925 and the average of rates between 2000 and 2014 were used for this purpose. The identified regions were considered to inhibit a long-standing innovative entrepreneurship pattern. After regions with long-standing general and innovative entrepreneurship patterns are identified, regions that have long-standing patterns in both cultural variables are retrieved. Three groups of regions are identified, no regions could be detected for a long-standing pattern of high entrepreneurship rates and low proportions of innovative entrepreneurship. This procedure assures that if combinations of these two cultural elements—entrepreneurship in general and innovative entrepreneurship pattern—are influencing each other qualitatively, this is not biasing the results of the regression analyses.

Individual control variables

On the individual level, demographic variables and personal contact with entrepreneurship are surveyed. Especially the variables age, gender, educational attainment, and work experience are found to influence entrepreneurial activities by former studies (e.g. Arenius & Minniti, 2005; Fritsch & Kublina, 2019; Fritsch & Wyrwich, 2018; Gallardo & Scammahorn, 2011; Jones-Evans et al., 2011). Moreover, being self-employed or knowing entrepreneurs personally is likely to impact business opportunity perception. Lastly, because it takes time to be properly embedded into a regional system, the period of residence is included as a control variable.

Regional control

To control for broader structural differences between regions in Germany, a factor variable describing where a respondent is living is introduced. In the German context, it is widely acknowledged that there is profound variation between the northern, southern, western, and eastern parts of the country in terms of economic, innovative, and entrepreneurial activities. Thus, it is expected that this distinction captures regional effects that are important for regional opportunity perception.

Personality control variables

Because the personality profile influences entrepreneurial activities undertaken by the person (Fritsch & Wyrwich, 2017a, 2017b, 2018), the personality traits as measured by the Big 5 model are assessed and controlled for.

Interest control variables

Moreover, the personal interest of the respondents in the context of entrepreneurship is likely to influence opportunity perception. In this vein, prior research has found that media attention can increase an individual’s probability of becoming entrepreneurially active (Urbano & Alvarez, 2014). To account for awareness of business opportunities that stem from attending events about start-ups, having experience with founding ventures as well as having worked in a small or new venture, these are as well controlled for.

The construction of the variables is summarised in Table 1.

Table 1 Overview of variables
Full size table

4 Results

4.1 Descriptive results

Dependent variables — opportunity perception

Table 2 shows that the share of opportunity perception is similar in both cases—general and innovative opportunities. This finding holds when looking at unweighted and weighted data, although the weighted data is significantly lower than the unweighted results suggest.

Table 2 Descriptive statistics of the regional business opportunity perception
Full size table

Independent variables — embeddedness levels

Table 3 summarises the descriptive statistics of the variables. The most overlap between regional and individual personality can be found in the dimension of openness. However, the personality fit seems to be limited on average. The presence of social contacts in the region and the perception of an economic environment are characterised by large sample variances.

Table 3 Descriptive statistics of the regional embeddedness variables
Full size table

Figure 2 displays the assignment of German regions to the culture combinations. The selected regions distribute themselves over Germany in the pattern as illustrated. Apart from the second group of regions clustering more in the west of Germany, the distribution does not appear to be biased in a clear geographical direction.

Independent variables — controls

Table 4 summarises the descriptive statistics of the control variables.

Table 4 Descriptive statistics of the control variables
Full size table

4.2 Econometric results

To test Hypothesis 1, Hypothesis 2, Hypothesis 3, and Hypothesis 4, the driving forces of the perception of regional opportunities to find a new venture are looked at in further detail. Tables 5 and 6 display the results of the regression models with the perception of regional business opportunities as a dependent variable.

Table 5 Binary logistic regression results – business opportunities in general
Full size table
Table 6 Binary logistic regression results – innovative business opportunities
Full size table

Specifications 1 to 4 in Table 5 estimate the effects of the independent variables on the perception of business opportunities in the region in general. On the actor level, the larger the difference between the conscientiousness level of the respondent and the perceived regional stereotype, the less likely it is that general business opportunities are perceived. The same is true, as a tendency, for the personality trait openness (Model (1) p = 0.412; Model (2) p = 0.162). On the network level, there is the tendency that the more colleagues are located in the same region as the respondent, the more likely it is that general business opportunities are perceived (Model (4) p = 0.135). Furthermore, on the environmental level, the perception of opportunities is positively related to the perception of young firms, innovative companies, and, as a tendency, innovative young firms in the region (Model (4) p = 0.263). On the culture level, the presence of a culture of innovative entrepreneurship increases the probability of opportunity perception as a strong tendency (Model (3) p = 0.109; Model (4) p = 0.107).

In contrast, specifications 5 to 8 in Table 6 analyse the relation of independent variables and the perception of innovative business opportunities in the region. It is evident that, in the case of innovative business opportunities, the only consistent findings are on the environment level and network structures. On the network level, the presence of colleagues in the same region as the respondent is positively related to the recognition of innovative business opportunities. Furthermore, on the environment level, the perception of innovative companies and the perception of innovative young firms in the region increase the probability of innovative business opportunity recognition.

Based on these findings, Hypotheses 1, 2, and 4 are partly confirmed, as these can be confirmed only for one type of business opportunities. Hypothesis 3 can be confirmed in the case of both types of business opportunities.

Although the interpretability of coefficients of control variables in regression analyses is limited (Hünermund & Louw, 2020), a few observations can be made. First, it comes as a surprise that educational attainment and gender are only important for the perception of innovative business opportunities but not for recognising general opportunities and do not affect entrepreneurial activities in all forms (Arenius & Minniti, 2005; Gallardo & Scammahorn, 2011). Second, the other way around, the personality traits of the respondent are only found to matter for business opportunity perception in general and not for innovative opportunity perception, contradicting former research (Hayton & Cacciotti, 2013; Rinne et al., 2012; Shane, 1993; Williams & McGuire, 2010). This finding suggests that personality traits are not the driving factor of regional innovative opportunity perception if the regional embeddedness interaction is considered. Third, attending regional events about young firms increases the likelihood of perceiving opportunities. This finding supports that attending regional events has an impact on the entrepreneurial process. Last, watching TV shows that address entrepreneurial activities relates to a higher likelihood of recognising general business opportunities but not innovative ones. This result could suggest that TV shows focus on general business opportunities and rarely on innovative ones.Footnote 1

Overall, the results indicate that general opportunity perception is more related to the regional embeddedness of the respondents than in the case of innovative business opportunity recognition. However, the share of colleagues in the same region as the respondent and the perception of innovative companies in the region show larger coefficients in the case of innovative business opportunity recognition than in the case of general opportunity perception. Based on these findings, it can be deduced that recognising innovative business opportunities relies on fewer regional factors than the perception of general business opportunities. Whereas recognising innovative opportunities depends on the business and innovation context, the perception of general opportunities is related to a mix of personal embeddedness and regional context factors. It can be concluded that the regional innovation system does not primarily drive innovative ventures; but once established and perceived, these trigger the perception of general business opportunities in the region.

5 Discussion and conclusion

5.1 Summary

This study examines the relationship between the regional opportunity perception of regional inhabitants and the regional embeddedness of the inhabitants into the regional innovation and entrepreneurship ecosystem. Four levels of regional embeddedness are argued to be of high importance for regional opportunity perception—actor level, network level, environment level, and culture level. These four regional embeddedness levels are set in relation to regional opportunity perception in general and innovative regional opportunity perception by employing binary logistic regression analyses. These analyses are based on data primary collected by an online survey in Germany.

The analyses show three main results. First, regional embeddedness matters in recognising general and innovative business opportunities. Although there are differences between the recognition of general and innovative business opportunities and their relation to regional embeddedness, it is noteworthy that the environment is important for the perception of both. Thus, this result contributes to the empirical evidence that the perception of the regional environment of respondents matters for their ability to perceive general and innovative business opportunities (Audretsch & Belitski, 2017).

Second, innovative ventures are found to pull the recognition of opportunities in general. The observation that long-standing regional entrepreneurship cultures matter cannot be made in the case of innovative business opportunities. Rather, innovative companies present and perceived in a region have a strong signalling power and enhance the perception of more general business opportunities. Furthermore, once innovative ventures are founded, the results support the assumption that these act as a pull factor for more entrepreneurial activities (Fritsch et al., 2019). Thus, this effect is revealed to have a short-term impact if innovative companies are perceived and a long-term effect when established as a regional tradition.

Third, the perception of innovative business opportunities is less dependent on the regional context than the perception of general business opportunities. In the case of recognising general business opportunities, all levels—actor, network, environment, and culture—inhibit at least one variable with a significant relation to the recognition of general business opportunities. In the case of innovative business opportunities, only the levels of environment and networks show significant relations to innovative opportunities’ perception. This finding highlights that potential innovative entrepreneurs are less influenced by their environment when recognising opportunities. However, the results show a dependence on the social environment they are embedded in, contradicting former considerations that the social context only plays a negligible role for these potential entrepreneurs in this stage of the entrepreneurial process (Cantner et al., 2017).

5.2 Implications for entrepreneurship research

Besides the fine-grained findings regarding the relationship between regional opportunity perception and regional embeddedness, this study contributes in two more ways that concern broader conceptual frameworks.

First, the study confirms that entrepreneurial activities are a regionally contextualised phenomenon that cannot be thoroughly understood in isolation. Thus, this study contributes to the literature on agency in regional ecosystems as conceptualised by Huggins and Thompson (2020). However, the relevance of the regional context is especially high for regional opportunity perception in general, but this is less the case for innovative regional opportunity perception. This indicates that a distinction between entrepreneurial activities regarding their degree of innovation is also essential for conceptualising the relevance of regional context in frameworks such as entrepreneurial ecosystems and regional innovation systems.

Second, the results of this study support the notion that entrepreneurial activities are manifestations of regionally available knowledge. One can interpret the perception of regional young and innovative firms as the knowledgeability of the respondent about the underlying knowledge base of the region. Taken together with the positive impact of these variables on regional opportunity perception, this result goes in line with the knowledge spillover theory of entrepreneurship (KSTE) as introduced by Acs et al., (2009, 2013). In other words, the study delivers empirical evidence for the idea that opportunity perception is, at least partially, driven by commercialisation possibilities of unexploited regional knowledge. Therefore, the results support the call for more extensive consideration of the concept of learning in the entrepreneurial ecosystem literature (Pugh et al., 2021).

5.3 Implications for policymakers

To effectively stimulate entrepreneurial activities that are widely acknowledged to be of high importance for economic development and addressing societal challenges, it is of crucial relevance to gain fine-grained information about the dynamics that relate to the most essential step in the entrepreneurial process—opportunity perception (Riquelme, 2013; Shane & Venkataraman, 2000; Venkataraman, 1997). Especially the identification of determinants of innovative opportunity perception is of high importance as innovative entrepreneurial activities are predominantly held accountable for economic development (Autio & Yli-Renko, 1998; Shane, 2009; Szerb et al., 2019). The results of this study allow some policy recommendations to be made.

First, the results suggest that the perception of innovative firms in the region by inhabitants increases the likelihood of recognising regional opportunities in general. To increase the chances that regional inhabitants perceive innovative firms, it appears valuable to fund programs focusing more specifically on innovative entrepreneurial endeavours rather than following a one-size-fits-all approach, confirming former policy advice (Szerb et al., 2019). Moreover, the results of this study suggest, as a tendency, that the regional innovative entrepreneurship culture is impacting the likelihood of regional opportunity perception. Thus, support of innovative ventures has a short- and long-term impact since these enable opportunity recognition and strengthen the regional entrepreneurship tradition, which is also revealed to foster the perception of general business opportunities.

Second, further policy advice can be derived, which is also connected to the finding that the perception of innovative firms is conducive to regional opportunity perception. For this mechanism to work, it may not be sufficient to only increase the number of innovative ventures and may not even be necessary, but the existing innovative companies need to be perceived. This corresponds to the notion that the presence of innovative entrepreneurial activities alone is not per se the panacea for economic growth but needs to be complemented by other circumstances (Gonzalez-Pernia et al., 2012). Hence, it could be beneficial to invest in awareness-increasing measures to increase the perception of innovative businesses in the region, which aligns with the suggestion that successful entrepreneurship stories can be utilised in entrepreneurship campaigns (Spigel, 2017). One instrument that could help in this regard is regional events about young firms as the attendance of such events is shown to increase the likelihood of regional opportunity perception. Because the analyses show that innovative regional opportunity perception depends partly on gender and holding a university degree, it may be advisable to have a diverse representation of entrepreneurs to attract broader groups of people and spark interest in entrepreneurial careers.

Overall, a more general policy recommendation can be derived. This study highlights the importance of designing policies that support the whole entrepreneurial process and not only specific parts of it, confirming the calls for a holistic policy approach to entrepreneurship (Audretsch & Belitski, 2017; Cantner et al., 2017; Garretsen et al., 2019). Rather than stimulating the number of new ventures, policymakers are required to concentrate on the interplay of entrepreneurial activities, their quality, and their regional embeddedness, strengthening the approach of ‘place-sensitive’ policies (Garretsen et al., 2019; Iammarino et al., 2017). This policy advice aligns with the notion that policies should view entrepreneurial activities and their quality in light of the regional entrepreneurship ecosystem (Szerb et al., 2019).

5.4 Limitations

Nevertheless, the study has several limitations. First, the 5-point Likert scale transformation into a binary variable (recognition vs. no recognition of business opportunities) comes with a decrease in the sample size available for the logistic regressions. However, this loss was not to be avoided to be able to compare the qualitative difference between perceiving and non-perceiving business opportunities. Second, this study is limited in its geographical scope. Future research may also explore the factors driving opportunity recognition in regions that do not show a time-invariant pattern of entrepreneurial activities. Third, this study only covers a subset of characteristics of regional embeddedness; the reality is far more complex than can be reflected in the conducted analyses (Audretsch & Belitski, 2017). Fourth, this study assesses the specifics of innovative regional opportunity perception in comparison to regional opportunity perception in general. This approach comprises different types of innovative opportunities, for example creation-based or discovery-based opportunities.

5.5 Recommendations for future research

Apart from future research that addresses the limitations of the current study, future research may also complement the variables of this study that describe other aspects of the entrepreneurial ecosystem. A further aspect in this line is that the embeddedness levels—actor, network, environment, and culture—are likely to interact with each other in complex ways. This study neglects this complexity for the sake of identifying the main driving forces of regional opportunity perception, but future research may close this gap by applying machine learning methods that are capable of disentangling these complex interactions partially.