Social Capital and the Economic Performance of Small Businesses: Role of Entrepreneurs Experience

Kraima Mohamed Taher^*
*Correspondence: Kraima Mohamed Taher, Doctor in Methods of Finance and Accounting, University of Sfax, Tunisia, Email:

Keywords

Social capital • Relationship • Networks • Entrepreneur • Expe rience • Resource

Introduction

Currently, small firms face difficulties gaining access to resources. The problem of accessing financial resources has always posed a major threat to the development of small firms [1-3]. With the lack of technological capabilities also tending to limit their competitive success [4]. Moreover, if the business is new, it will be confronted by a lack of organizational capabilities, such as coordination and communication systems, management skills, etc. [5].

In this context, the entrepreneur's network of relationships becomes a source of strategic resources on which to build competitive advantages [6]. This is referred to as social capital, that is, the value embedded in the social relationships of individuals or groups [7]. This notion proves particularly relevant in the case of a small firm in which an entrepreneur both owns and manages the small business. In such instances, entrepreneurs, resources, their traits, even spirit, and relations, are inseparable from the firm itself. Such entrepreneurs could access technological resources through membership of professional and business associations. Their business capabilities will benefit from establishing close relationships with suppliers [8]. A rich social life involving many informal contacts (with relatives, acquaintances, and friends) emerges as a source of innovation if said contacts are with people who have new and different ideas [9]. Finally, entrepreneurs' relationships with local institutions (banks public, authorities, or the media) might provide them with access to financial resources (support, loans, or subsidies, as well as access to private investors).

Several studies address the repercussions of entrepreneurs' social capital on the performance of small firms [10-12]. Specifically, these studies explore entrepreneurs' social competence. Baron and Markman [10] and to what degree certain network characteristics impact the performance of nascent entrepreneurs [13]. Many studies do not directly measure social capital, but analyze its sources. The contribution of our study is to measure social capital as the resources embedded in the network of relationships. The basic definition of social capital states that social capital is defined as ‘networks of relationships and assets located in these networks’ [14]. It is precisely these resources that endow such networks with value and make them ‘capital’ in the sense that they may ultimately lead to future benefits in business. In this line, Lin, defines social capital as ‘resources embedded in a social structure which are accessed and/ or mobilized in purposive actions.’ However, as pointed out by Gedajlovic et al. [7] a common practice is to refer to social capital in terms of the characteristics of the relationships through which resources are expected to be derived. We assume that both the networks of relationships themselves and the resources embedded within them constitute social capital [14] and that the characteristics of these networks of relationships are the conditions required to access the embedded resources [15-17]. Regardless of how strong, close-knit, and tight the relations within a network are, the social capital of these networks lies in their ability to provide entrepreneurs with resources [18]. These resources embedded in the entrepreneur's relationship networks involve financial resources (credits or funds), technology and innovation resources (patents, technologies, etc.), commercial and business capabilities (sales, communication skills, and access to market information), human resources (motivation, qualification, etc.), resources related to quality management, and organizational capabilities .

The second contribution of our study is to conduct a comparison-oriented joint and simultaneous analysis of all the generic types of relationship networks the entrepreneur is involved in: personal, namely, associational, institutional, and professional networks. In our framework, social capital is seen as a resource located in an actor's internal ties and external ties [19]. Such that the type and content of these linkages determine access to other embedded resources [15,16,20,21]. Thus, social capital's sources lie in the social structure within which the actor is located [18]. The extent to which entrepreneurs maintain frequent links in their personal private circles (friends, family), in their professional circle (current business as well as previous businesses or jobs), with a range of agencies (volunteers, cultural or sports associations, trade unions, political parties, neighbors' associations, religious groups, etc.), or with public or private institutions will determine how much access they will have to financial, commercial, technological, and organizational resources [8,9,12]. Although prior research has explored the impact of professional and personal networks on entrepreneurs' success no works have thus far adopted a joint and comparative approach to exploring the influence of these four networks on entrepreneurs' access to resources and business performance.

From a theoretical point of view, this study's third contribution is that it provides insights regarding two boundary conditions of social capital in entrepreneurship research: entrepreneurs' experience. Although the four types of relationship networks allow entrepreneurs to access resources, we maintain that, in general, business performance is more influenced by professional and institutional networks' resources than by resources gleaned from personal and associative networks. However, each type of relationship network is not equally advantageous in all contexts. Since the resources provided by institutional and professional networks are more valuable than the resources provided by personal networks, in highly competitive contexts, all competitors will try to access institutional and professional networks. Therefore, entrepreneurs will seek inimitable networks in order to achieve competitive advantage. Thus, we propose they will be able to derive particular benefit from the resources provided by more personal and private networks in order to find new ways of improving market positioning.

In contrast, the greater the entrepreneur's experience in the industry, the more relevant the institutional and professional networks will be to the firm's economic performance, since an experienced entrepreneur will be able to access more specific and business-oriented resources provided by professional and institutional networks.

In the following section, we present the theoretical background on which the proposals are based. We first examine the resource-based view as well as the social capital approach, and we conclude with the importance of social capital as an entrepreneurial strategic resource. From the focus of the network marketing, relationship marketing, and social capital approaches, we go on to analyze those networks that generate social capital and facilitate access to valuable resources. Grounded on these theoretical foundations, and through three different sections, we successively present and argue the hypotheses, which outline the effects of social capital resources on economic performance and the effects of an industry's entrepreneurial experience. We then set out the methodological aspects and the findings of the empirical study. The work closes with comments on the principal conclusions and implications of interest to business practice.

Literature Review

The resource-based view

Resources ‘are the inputs of the productive process of a firm’ and ‘the basic unit of analysis of the theory of resources and capabilities’ Grant (1991: 118). The resource-based view theory starts from market imperfection and states that owning valuable, rare, inimitable, and non-substitutable (VRIN) business resources is a source of sustainable competitive advantage and the source of differences in the financial performance of firms competing in a similar industrial environment [22-24]. Resources are valuable when they enable firms to implement strategies that improve their effectiveness and efficiency. Rare resources are those not simultaneously possessed by many other firms. Resources are inimitable if firms that do not possess them are unable to obtain them. Finally, resources are non-substitutable if there are no strategically equivalent resources [22].

In this way, the role of resources is vindicated by the generation of value, and the existence of a positive association between managed resources and the firm's economic performance is defended. Yet, the firm should not be restricted to merely using said resources, but must also find the best way of combining them so as to create capabilities within the company [25,26]. These are the capabilities that will ultimately endow the firm with a sustainable competitive advantage.

Several works have attempted to identify their link to generating competitive advantage and these assets [27-29]. Yet, the ability to innovate and adapt to changes obviously does not depend exclusively on internal factors. Indeed, the company will develop this capability only insofar as it manages to connect with the external environment and captures these ideas and new tendencies, which might inspire them to improve their processes and products [11]. Other organizational capabilities based on intangible assets that have an impact on innovation and other economic results must, thus, be identified. Along these lines, certain studies consider the firm's social capital and relationships as strategic resources that lead to success and innovation [21,30,31].

Definition of social capital Nahapiet and Ghoshal [31] define social capital as networks of relationships that allow their members to access the different assets available in these networks. In line with Burt, literature on social capital agrees on ‘a social capital metaphor in which social structure is a kind of capital that can create a competitive advantage for certain individuals or groups when pursuing their ends,’ so that socially better-connected individuals will also be in a better position to achieve the desired results. Adler and Kwon stress that the effects of the structure and content of the actor's social relations ‘flow from the information, solidarity, and influence it makes available to the actor.’ Social capital arises, therefore, because sufficiently stable conditions exist between groups of individuals. In fact, what distinguishes social capital from other types of capital is that it resides in relationship networks and exists only if shared between network members. Since relationships between individuals are framed within networks, social capital is associated with two elements: network content (embedded resources that may be mobilized through the networks), and network characteristics.

With respect to network characteristics, the most common distinction is between ‘bonding’ and ‘bridging’ social capital. Bonding social capital refers to relationships between people in a group who know each other well. Such networks are associated with strong ties, trust, cohesiveness, and reciprocity, which allow exchange of resources between members. Bonding social capital may facilitate the pursuit of collective goals, and it is exclusive and available to the members of a group. Bridging social capital refers to horizontal ties shaping more diverse groups of people with different backgrounds, like volunteer groups or professional networks. As these networks are more diverse, they can provide members with valuable resources and explain the differential success of firms in their competitive rivalry [18]. Bridging social capital is closely related to the concepts of weak ties and structural holes. A third concept is so-called linking social capital, which refers to vertical relationships with powerful institutions and groups. Linking social capital is often characterized by weak ties. The scarcity of these types of relationships implies that linking social capital is often a powerful source of distinctive and valuable resources for individuals.

In respect to network content, the embedded resources in a network of relationships are a core concept of social capital. Lin's social resources theory proposed that access to resources embedded in social networks can lead to better socioeconomic status. Access to and use of social resources is determined by the position in the hierarchical structure and the strength of the ties. In fact, Nahapiet and Ghoshal [21] identify social capital as the sum of the actual and potential resources embedded within, available through, and derived from the network of relationships an individual has. In line with Lin, Nahapiet and Ghoshal [21] and Tsai and Ghoshal, the information and resources individuals have access to through their relationship networks are a consequence both of the type of networks and of the individual's position therein (structural social capital), as well as the ease and willingness of members to cooperate and exchange resources (relational social capital).

Many authors [20,21] describe a network's social capital as the set of current and potential resources derived from the relationships that make up a network. Granovetter [32] identify the resources embedded in relationship networks as information- type resources. Social capital has subsequently been related to resources of a different nature, ranging from the purely economic to those of a social and emotional nature.

Yet, it is important to underscore that the main aspect of social capital resources is the fact that they may be accessed and/or mobilized in purposive actions. Social capital facilitates acquisition of resources by promoting a flow of information and funds from diverse sources, and it drives the creation of intellectual capital by establishing the conditions for aiding the development of new resources. However, Payne et al. [19] find few studies that measure the effect of network connections with outcomes.

As Gedajlovic state, despite the importance of such an approach little attention has been paid to measuring the actual resources accrued from social networks. Some authors refer to social capital resources as the benefits gained from social capital (relationship networks), mainly knowledge acquisition and knowledge sharing.

Relationships networks as sources of social capital

Relationship networks constitute a nexus between social capital and the relationship marketing approach. As has been observed, social capital derives from an individual's relationship networks and from the assets located therein. For its part, the relationship marketing literature has underlined that strategic outcomes, such as relationships with channel members and customers, often become ‘market-based assets’ that add to the firm's existing resource stock. Entrepreneurs' relationship networks become a business asset since they afford a competitive advantage by providing access to, distributing, and processing more information and resources. Entrepreneurs can incorporate resources provided by these relational assets to build core competencies. Moreover, stocks of these assets can be developed, leveraged, augmented, and valued [33]. Such relational assets are primarily external to the firm and are largely intangible. Indeed, from a resource-view perspective, a firm's most important strategic assets are those based on intangible assets [34].

Brown and Greene specifically include social resources as strategic resources for growth and innovation, inasmuch as they allow access to other physical, human, financial, and organizational resources. Viedma Marti [35] indicates that social capital is an intangible resource that is primarily external and of a relational nature. Social capital, thus, becomes a source of sustainable competitive advantage, rare or scarce, as it is valuable, inimitable, and a nonsubstitutable resource. Although social capital can be imitated (individuals can have relationships with similar institutions or associations), each individual has access to separate social networks and develops different kinds of relationships within these networks and this is the inimitable aspect of social capital.

Johannisson in his analysis of business activity in the contexts of local and regional development, states the importance of business, professional, and friendship ties as well as institutional and associational links with the local community. In our research, we echo the proposals of Stone and Hughes and Johannisson and group entrepreneur networks of relationships into four categories depending on the personal, associative, professional, or institutional nature predominant in the relationships:

• Personal networks of relationships (PERS NR) with relatives, friends, and neighbors are normally symmetrical and voluntary relationships, seen among individuals sharing common characteristics and interests. Literature on social capital often considers these relationships to be related to bonding social capital.

• Associative networks of relationships ( ASSOC NR) with other members of the volunteer associations to which the entrepreneur belongs (such as business, civic, professional, political, labor, religious, cultural, social advocacy, or sports associations). They are usually formal in nature, given that on many occasions these groups are governed by explicit rules that regulate membership, commitments, and departure of members as well as how they relate to each other (internal relationships) and with other groups (external relationships) [36]. These relationships are rather in-between bonding and bridging social capital, as they can involve both weak and strong ties and both vertical and horizontal relationships in addition to mixing formal and informal mechanisms of governance.

• Professional networks of relationships (PROF NR) with partners, workers, suppliers, customers, and colleagues. Since they are related to the entrepreneur's past and present professional activities, they occur in more formal contexts than the previous ones and have been considered as a source of bridging social capital. This type of business network is usually oriented toward acquiring business-related resources.

• Institutional networks of relationships (INST NR) with representatives or members of different public and private institutions. In the case of entrepreneurs, these relationships refer to direct contacts with government officials, the media, public authorities, financial bodies, or large companies, among others. These institutional relationships are not usually voluntary in nature and are normally regulated by very specific rules. They are generally asymmetrical and their quality depends, to a large extent, on how well the institutional and legal environment in which the business activity is performed is able to function [37]. These relationships have been related to linking social capital [37,38].

Entrepreneurs' social capital and economic performance Granovetter [32], states that certain characteristics of an individual's network may shape access to new ideas that enhance an individual's ability to innovate. Moreover, a resource is more likely to generate competitive advantage if it is accessible to the enterprise, is scarce, idiosyncratic, and difficult to substitute, complements other resources of the firm, is consistent with the firm's strategies and with the characteristics of the industry or sector, and proves difficult to imitate and transfer to other companies [24,39]. Entrepreneurs' social capital merges all these characteristics and, thus, it can be concluded that the capabilities based on entrepreneurs' social capital may generate competitive advantage and, therefore, enhance performance. The positive effect of networks on business results has been highlighted in several contexts: industrial networks, industrial districts, the launch of new products, or internationalization strategies.

In our work, we contend that social capital resources contribute to improving small firms' economic performance in terms of market share, sales growth, and success in launching new products. In the case of entrepreneurs, accessing new ideas and products may not only derive from exchanging information with customers, suppliers, and members of the associations to which the entrepreneur belongs, but also from the fact that entrepreneurs' personal relationships include people from different educational backgrounds, cultures, or nationalities. The resources provided by networks help the entrepreneur achieve business success. Access to advice, technology, funding, human resources, or information may favor innovation [40], launch of new products [31,41], or entry into new markets [42]. Involvement in associations improves a community's level of social capital, thus benefitting all its members. For example, professional associations often provide entrepreneurs with advice and help when negotiating with banks and suppliers. But, nonprofessional associations are more diverse and allow local entrepreneurs to access new business opportunities. This training in negotiation proves key to securing funding and sources of future investment. Thanks to their institutional contacts, entrepreneurs may gain access to public aid programs for the commercial and technological modernization of their businesses. Personal and professional networks allow entrepreneurs to recruit reliable workers or harness new ideas for their businesses, which can lead to new products. Therefore, we propose that:

Hypothesis 1 (H1): The social capital resources of personal networks of small firms have positive effects on their economic performance.

Hypothesis 2 (H2): The social capital resources of associative networks of small firms have positive effects on their economic performance.

Hypothesis 3 (H3): The social capital resources of institutional networks of small firms have positive effects on their economic performance.

Hypothesis 4 (H4): The social capital resources of professional networks of small firms have positive effects on their economic performance.

If networks may provide resources that enhance economic performance, their effect will differ depending on the specific characteristics of each. Indeed, we can characterize each network in terms of the value of the resources provided and depending on the degree to which these networks are substitutable and imitable. These characteristics will impact business performance in different ways, as we aim to show in our next three hypotheses.

The importance of network's as resources value and economic performance

Resources are valuable when they enable firms to conceive strategies that improve their effectiveness and efficiency. Although all kinds of networks can provide valuable resources, certain networks are more likely to offer resources adapted to entrepreneurs' business needs [15]. Characterizing the resource provided by social networks is supported by the social capital literature, which links the nature of the networks to various types of social capital, that is, different types of resources. Table 1 describes networks by their embedded resources value.

As already pointed out, institutional and professional relationships are linked to bridging social capital and weak ties. Bridging social capital is characterized by connecting individuals with a wider range of agents that can provide them with a broader and, therefore, more valuable array of resources. In fact, institutional and professional networks may offer more specific (entrepreneurship-specific and industry-specific) resources and, therefore, more valuable resources since they are directly related to the entrepreneur's business or to the institutional and legal environment in which the business operates. For instance, a good relationship with suppliers may offer access to a wide range of markets in geographical terms or to new clients, which would never otherwise be possible through solely personal relationships. Relationships with professional colleagues may provide specific information concerning a particular sector (tools, technologies, forecasts, prospects, and so on) which would be difficult to secure through other means. Institutional relationships with the media, for example, would aid the task of marketing or promoting a product. A further example is the link with technology centers or universities that could supply the human resources or specialized technology a small firm would otherwise find it difficult to access through other personal or associative networks.

Personal relationships tend to be associated with so-called bonding social capital and strong ties. Although bonding social capital provides cohesion within a group, it leads to homogeneous groups. Therefore, the likelihood of it providing diverse and valuable resources is less than with bridging social capital. Personal networks offer more generic resources that are less adapted to the particular business in question. This is low embedded resource value.

The nature of associative networks places them between personal (civic, social, religious, advocacy associations, etc.) and professional or institutional (labor unions, professional colleges, political parties, etc.) networks. Thus, they can provide both non- business-related as well as business-related resources. As Teckchandani points out, professional associations and business contribute to entrepreneurial activity more than other association types. Moreover, and regardless of type, associations can be based on strong ties and provide high cohesiveness and scarce access to diverse resources; or they can be based on weak ties, with higher access to diversity. Thus, we place them in the position of low-high (medium) valuable resources.

Briefly, we propose that the resources will have greater value than networks based on strong ties and horizontal relationships. Institutional and professional and networks should, therefore, be expected to have an important effect on business performance than personal (and even associative) networks, since the resources the former provide are more valuable and are more directly linked to the firm's business activity. Thus, we propose that:

Hypothesis 5 (H5): A positive effect of social capital resources on economic performance will be greater in institutional and professional networks than in personal and associative networks.

Network substitutability over time and the role of the entrepreneur's experience

Entrepreneur's relationship networks offer a competitive advantage if the resources provided are valuable and hard to replicate. A third source of competitive advantage for an asset is that it should be difficult to substitute. Resources are non-substitutable if there are no strategically equivalent resources. As shown in Table 1, some networks can be substituted over time, whereas others are hard to substitute. Even if all the relationship networks may provide the entrepreneur with resources, preferential use of one kind of network is related with the firm's age or the entrepreneur's experience in the sector. Batjargal proposes that entrepreneurs' experience enhances the positive effects of their networks on firm performance. Sasi and Arenius [43] explain that in the early stages of a new venture, entrepreneurs rely on friends and family to obtain the information, physical and capital resources, and social support needed to turn an idea into a business reality. In other words, the entrepreneur's personal networks usually provide the initial resources needed to successfully launch a business, when it is not yet possible to develop rich enough institutional and professional networks. Entrepreneurs subsequently increase their internal and external networks with business acquaintances (suppliers, employees, partnerships, etc.) that prove more important in key market areas. They, therefore, replace resources accessed through personal networks, which are more generic and less adapted to business, with resources provided by institutional and professional networks, which are more specific and business-oriented and allow firms to grow [43]. In sum, personal and associative networks are characterized by a high degree of substitutability, whereas professional and institutional networks are hard to substitute.

Most of the works cited address the firm's age or the business life cycle as a variable engendering greater development of institutional and professional networks. Our contribution to these works is to underpin the entrepreneur's experience in the particular sector in which he/she is involved, rather than his/ her overall experience in the business world. Yet, in the case of an entrepreneur, not only should the firm's age be considered, but also the entrepreneur's whole professional career within an industry. This professional experience will enable an entrepreneur to establish institutional and professional contacts that will prove useful to his/her new venture. In this sense, we feel that over time entrepreneurs will tend to replace resources drawn from personal networks with embedded resources gained from institutional and professional networks as the latter gradually consolidate. Consequently, the longer an entrepreneur has been working in a sector, the more relevant the professional and institutional networks will prove to the firm's economic performance.

Hypothesis 6 (H6): The lower the entrepreneur's experience in the sector, the greater the positive influence of social capital resources of personal networks on economic performance.

Hypothesis 7 (H7): The greater the entrepreneur's experience in the sector, the greater the positive influence of social capital resources of professional networks on economic performance.

Hypothesis 8 (H8): The greater the entrepreneur's experience in the sector, the greater the positive influence of social capital resources of institutional networks on economic performance.

Methodology of Research

The target population of the study is small entrepreneurs in Tunisia, that is, business people who are at the same time owner and manager of a small business (50 or fewer employees). Since there is no sampling framework available for our target population. Our study drew on cooperation with Industry and Innovation Promotion Agency and Tunisian Business Directory in Tunisia.

The primary aim of these agencies is to promote economic development in the areas where they are located. Thus, they fully understand the reality of each area and can identify its key players, including local entrepreneurs. Although not strictly probabilistic in nature, this method is suitable when no sampling framework is available, as in our case. The main risk of non-probability samples is that there is no specific sampling frame that can reliably represent the population. Therefore, the sample might not prove representative. Researchers have no accurate estimates to gauge whether the sample is representative of the population or not. Despite this, in judgment-based sampling, if the experts know the population well enough, results may prove more accurate than those obtained from probabilistic sampling. Coviello and Jones indicate that judgmentbased or purposive sampling dominates in international entrepreneurship studies [30].

Data was gathered from May to December 2019, and the development agents themselves were in charge of contacting the entrepreneurs and distributing and collecting the questionnaires. Following the procedure indicated, and after eliminating some incomplete questionnaires and those of firms with more than 50 employees, a useful sample of 410 entrepreneurs was obtained. Of those surveyed, 61.3 percent of the respondents belong to rural areas and 34.7 percent to urban areas. In terms of business size, in 32.5 percent of the cases, only the entrepreneurs themselves work in the firm; in 42.8 percent of the cases, there are two to four people; in 22.9 percent of the cases, there are five to 16 workers; and in 4.3 percent of the cases, there are 17 to 48 workers. Finally, the type of businesses in the sample is quite varied vis-àvis the main activity: manufacturing (25.9%), retailing (28.2%), tourism, hotels, and restaurants (19.1%), and other services (31.1%).

Variable and Data

The widely embraced methodological proposals for measuring embedded resources in individuals' networks are the Resource Generator and the so-called Position Generator. The Position Generator has been applied successfully in social science studies. It is based on the idea that social capital can be measured by the positional characteristics of network members as a proxy variable indicating the social resource collections embedded in an individual's social network. Based on the Position Generator, Van der Gaag and Snijders developed the Resource Generator. The Resource Generator is also a survey tool for measuring individual social capital. Unlike the Position Generator, however, Resource Generator information directly refers to accessed social resources rather than occupational prestige. This proposal heralds a step forward in the attempt to measure social capital resources since it avoids using a proxy variable to gauge the resources obtained, and it focuses directly on the resources provided by the individuals involved in the network, irrespective of the position they occupy.

The Resource Generator proposed by Van der Gaaj and Snijders aims to measure, from a sociological perspective, the social capital resources of individuals as a whole, not just of entrepreneurs in particular. Thus, it includes resources that are useful for daily life. However, since our study aims to measure the impact of social capital on firms' economic performance, we focus on resources of a business nature considered to be strategic in the resourcebased view literature. To do so, based on the resources classification proposed by Rubio-Bañón and Aragón-Sánchez, we develop four formative scales to measure the social capital resources of personal, associative, institutional and professional networks. In all cases, five-point Likert scales were used, referring to the degree to which entrepreneurs consider that each type of network afforded them the chance to acquire technology, financial resources, innovation capabilities, human resources, quality management capabilities, marketing resources, and organizational capabilities. Moreover, we repeated each question addressing access to resources for each of the entrepreneur's relationship networks as suggested by Stone and Hughes. The questionnaire includes a brief description of what we understand to be personal, associative, institutional and professional networks.

Entrepreneurs' experience was measured as the number of years entrepreneurs had been working in the industry. Finally, the three items of the reflective scale of economic performance refer to two dimensions of the strategic results proposed by Walker and

Ruekert: Market results and innovation results. We performed Harman's singlefactor test to assess the possible impact of common method variance. Evidence for common method bias exists when a single factor emerges from the factor analysis or when one general factor accounts for the majority of the covariance among the measures. Exploratory factor analysis with all the indicators gave eight factors with an eigenvalue of greater than 1.0 (total variance explained was 84%), with a first factor explaining only 24% of the variance.

Control variables

To achieve a strong competitive position in a market, access to suitable resources is not enough. Firms must adopt the right strategy. Thus, as a first control variable, we identify entrepreneurs' strategies or strategic profiles into our research. The literature offers a wide range of classifications of a firm's competitive strategies. We consider that the hybrid typology, which results from integrating the proposals of Miles and Snow and Porter, proves particularly interesting due to its close relationship with the organization's marketing activities. Moreover, the hybrid classification can be used to characterize not only the firm's strategy, but also the entrepreneur's strategic profile. In line with this classification, the entrepreneur's strategic profile can be placed in one of the following categories:

Prospector strategy: This places the emphasis on the search for new business opportunities starting from the development of new products or entry into a new market. The prospector is usually associated with the pioneering launch of innovations adapted to the changing needs of the market.

Analyzer strategy: As well as working closely with customers, firms that embark on this follower strategy analyze competitors who use prospector strategies to identify their successes and failures and develop new versions of the product or service that enhance the good qualities.

Low-cost defender strategy: This strategy's principal competitive tool is price. Thus, an effort is required to reduce costs and foster economic efficiency to develop this strategy.

Differentiated defender strategy: Like the previous one, this strategy seeks to defend the firm's target and to retain present customers by offering a product that provides a greater added value or any other distinguishing feature.

A fifth strategy, the reactor strategy, although certain authors omit it since they do not believe it is a strategy in the strict sense, rather a non-strategy, given that reactor organizations do not plan their actions and display no common behavior patterns. In addition, their passive attitude is not normally the result of any deliberate intention on the part of the firm's managers.

Many works have established relationships between strategies and economic results, concluding that each kind of strategy pursues a different type of result [44,45]. Our aim, however, is not to evaluate the differential effect of each type of strategy. We confine ourselves to suggesting that firms who adopt a well-defined strategy obtain better results than those who maintain a reactor strategy.

The competitive strategy adopted by the entrepreneur was measured by means of a self-typing scale constructed around the hybrid typology proposed by Slater and Olson. For this, five descriptions of the strategic profile of the business were presented, and participants were asked to situate their enterprise in the one that best described it. This kind of self-typing scale has been used widely in previous studies [46]. Thus, we obtained five dummy variables: prospector, analyzer, low-cost, differentiated, and reactor.

Although our work focuses on small firms, the size of small firms has been considered a determinant variable of business performance. We, therefore, include it as a control variable. A firm's size was measured as the log transformation of the number of employees (logsize) rather than as a raw measure of size, as suggested in previous works. We also include the sector of activity as a control variable in order to remove possible effects on business performance. The sector was measured as four dummy variables: manufacturing, commerceretailing, tourism-restaurant, and other services). Table 2 shows the variables used in the study, their measurement indicators, and the corresponding descriptive statistics (mean and standard deviation).

Analysis and Results

In order to test hypotheses, we used moderated hierarchical regression, previously reducing the scales to a single index. As for the formative constructs, we used the partial least squares approach (PLS), an analytical technique that makes to estimate models with formative constructs and can work with nonmetric variables and data that present non-normal distributions. Specifically, Smart. PLS software was used. PLS estimation comprises estimating both the measurement and the structural models. The measurement model can involve variables measured with formative indicators and variables measured with reflective indicators. Reflective indicators are functions of the latent variable. Therefore, changes in the variable are reflected in changes in the observable indicators. Contrastingly, formative indicators are specific components of the general construct they collectively constitute. In these cases, changes in the indicators determine changes in the value of the variable.

We estimated the direct effect of resources provided by personal, professional, associative, and institutional networks on economic performance. The coefficients' level of statistical significance was calculated by means of a bootstrapping procedure with 150 subsamples randomly extracted from the original sample. Given that the scales employed to measure the social capital resources of the various networks are formative, we previously tested for the nonexistence of multi co-llinearity between the indicators that make up each scale. In Table 2, the values of the variance inflation factor (VIF) are also shown, as are the outer weights of each indicator. We observe that collinearity is not at a critical level. As for the significance of the formative indicators, Hult, Hair, Ringle, and Sarstedt explain that non-significant indicator weights should not be interpreted as indicative of poor model quality measurement. When an indicator's outer weight is no significant but its outer loading is high (above 0.50), the indicator should be interpreted as absolutely important but not as relatively important. We have included the outer loadings in Table 2, the lowest being 0.475. The absolute contribution of the indicators can, thus, be interpreted as relevant.

In order to evaluate convergent validity in formative measurement models, testing whether the formatively measured construct is highly correlated with a reflective measure of the same construct is recommended. In our work, in order to limit the length of the questionnaire, we did not include reflective scales for network resources, so we were unable to test convergent validity. Finally, discriminant validity was established since the item- to-construct correlations were higher with each other than with other construct measures.

We then multiplied the factors measuring the networks' social capital resources by the entrepreneur's experience so as to calculate the interaction variables. Independent variables were previously mean centered in order to reduce multicollinearity between the interaction terms and their constituent variables [47]. A correlation analysis was carried out prior to the regression analysis (Table 3). The highest correlation between the independent variables and the interaction terms was 0.61. Past studies suggest that correlations at this level might not pose a serious multi collinearity issue for the interaction results generated.

Our hypotheses were tested using hierarchical moderated regression. This method allows us to sequentially introduce different blocks of variables and to check their respective explanatory capacities. Four steps of regression analysis were conducted in this analysis. First, we introduced the control variables (prospector, low-cost, analyzer, differentiated, manufacturing, tourism, commerce, and logsize). Second, in order to verify H1 and H5, we included the block corresponding to the main and direct effects of the various network resources (NR): resources provided by personal, associative, institutional, and professional networks. Third, the direct effects entrepreneur experience was added. Finally, to estimate the effects suggested in, we incorporated a block with all the interaction terms among the variables in the last two blocks (of personal, associative, and institutional, professional NR with entrepreneur experience). Results are in Table 4.

The explanatory capacity of the model is limited (low R2 values), which should not concern us since our goal was not to explain entrepreneurs' economic performance, but to test the existence of the foreseen effect of social capital resources on performance. Nevertheless, Table 4 (step2) shows the positive and significant effects of the social capital resources of the PROF (β = 0.173; p < 0.001) and INST (β = 0.132; p < 0.001) NR and the non-significant effects corresponding to personal and ASSOC NR. As a result, we can accept H3 and H4, but must reject H1 and H2.

Resources obtained through entrepreneurs' professional and institutional networks significantly contribute to improving their results, while resources derived from associative and personal networks do not appear to be relevant, which seems to point in the direction indicated by H5. In order to test that the standardized beta coefficients of PROF and INST NR were significantly higher than the coefficients of personal and ASSOC NR, we performed a t-test for mean differences (Table 5). Moreover, we estimated 95 percent confidence intervals. According to Cumming and Finch's (2005) rule, two estimates can be considered as statistically significantly different from each other when the corresponding 95 percent confidence intervals overlap by no more than 50 percent. As can be seen in Table 5, the coefficient of PROF NR can be considered significantly higher than the coefficients of personal (p = 0.06) and ASSOC NR (p = 0.004). Put differently, the effect of social capital resources on economic performance is greater in the case of PROF NR than in the case of PERS and ASSOC NR. Similarly, the effect of INST NR can be considered significantly higher than the effect of ASSOC NR (p = 0.03). However, the effect of INST NR is not significantly higher than the effect of PERS NR. Hence, with this sole exception, we can accept H5.

In regard to the moderating effects of the entrepreneur's experience (H6, H7 and H8), we observe that the change in the F-statistic caused by adding the interaction effects is significant. Therefore, the interaction effects improve the explanation of economic performance. Step 4 confirms there are some significant interactions between the entrepreneur's experience and PROF (β = 0.09; p < 0.05) and INST (β = 0.011; p < 0.05) NR are significant and positive, but non-significant in the case of PERS NR. We, therefore, find support for H7 and H8.

To better understand the significant interactions, we used simple slope analysis as recommended by Aiken and West (1993). Each interaction effect was analyzed considering three conditional values of the moderator variable: the mean, one standard deviation below, and one standard deviation above the mean.

This generates three alternative β values in each case, which appear in Table 6. H6, H7 and H8 (the effect of entrepreneur experience), the influence of PROF NR on economic performance is higher when entrepreneurs have more years of experience in the industry (β = 0.237; p < 0.001) than when their experience is limited (β = 0.105; p < 0.05). Similarly, the influence of INST NR on economic performance is significant when the entrepreneur has more years of experience in the industry (β = 0.186; p < 0.001), yet is non-significant when experience is limited.

With regard to the control variables, some interesting results emerge. First, small firms' strategies impact economic performance. Although any strategy the firm actively embarks upon should be better than the reactor strategy, it seems that their impacts on performance differ. We conducted an ANOVA and a Tukey test to evaluate the different effects of strategies, with the relation between strategy and economic performance proving to be significant (F = 12.467; p < 0.001). These analyzes indicate that the prospector and analyzer strategies contribute most to improving the firm's results. Compared to the reactor strategy, the analyzer, low-cost defender, and differentiated defender strategies also improve firms' performance, although we found no differences among the effects of these three strategies.

Second, the effect of size is significant and positive, indicating that larger firms obtain better economic performance than smaller firms. As we measure firms' size as the logarithm of the number of employees, this means that performance increases with size at a declining rate.

Finally, only in the case of the manufacturing sector do we find a negative effect, showing that economic performance in the manufacturing sector is lower than in the ‘other services’ sector.

Discussion

The main theoretical implication is that it furthers the role of small entrepreneurs' social capital resources in a firm's performance. In a small business context, certain resources must be sought in entrepreneurs' relationship networks themselves. The present work bears out the relevance of so-called social capital resources vis-à-vis obtaining enhanced economic performance in terms of market and innovation results. Moreover, not all networks allow entrepreneurs to access relevant resources, with only some of the resources provided by each network actually proving valid from the business standpoint. Results from the analysis show that entrepreneurs' various relationship networks are not all equally advantageous. We find that economic performance is boosted by the resources entrepreneurs obtain via their institutional and professional networks. However, personal and associative networks do not appear to be so relevant. Yet, even though the resources afforded by personal and associative relations do not seem to impact entrepreneurial performance, this might be qualified if entrepreneurs' experience is taken into account. Entrepreneurs' business experience also helps explain the effect of the different networks' social capital on performance. As experience in the sector increases, so does the influence of professional and institutional network social capital resources on economic performance. Experience contributes to developing wider and more diverse professional and institutional networks whose influence on economic performance proves more relevant.

Previously on the social capital literature has tended to link the nature of relations (personal, associative, institutional, and professional) to various types of social capital in terms of the value of embedded resources (bonding, bridging, and linking). Our study shows that when there are no external determining factors, such a link proves to be true. As assumed, professional networks (bridging social capital) and institutional networks (linking social capital) offer entrepreneurs valuable resources. By contrast, in personal networks (bonding social capital), entrepreneurs have greater difficulty finding valuable resources. It is difficult to ascertain what kind of social capital associative networks are able to provide in terms of accessing resources. Surprisingly, in no instances do the resources afforded by such networks provide any competitive advantage. Our research shows that this link between the kind of network and the nature of the social capital can be either broken or strengthened depending on certain external factors. Specifically, entrepreneurs' experience in the sector enhances the social capital linking of institutional networks and the social capital bridging of professional networks.

Whatever the case, we must clearly bear in mind that only certain resources are significant in each type of network. Resources to which entrepreneurs have access through their personal networks (relationships with family relatives and friends) and contribute to economic performance are those related to innovation, technology, and marketing capabilities. In professional networks (relationships with partners, workers, suppliers, and customers), relevant resources for economic performance are technological, commercial (marketing), quality management, human, and organizational. Finally, the resources obtained via entrepreneurs' institutional networks (relationships with institutions or public authorities) that contribute to boosting the results of the small firm are financial, commercial (marketing), and human resources. These results are aligned with the propositions of Shipilov and Danis, who suggest that a good fit between the managerial team's type of social capital, the company's strategic profile, and environmental stability, enhances organizational performance.

Managerial Implications

In terms of the implications for small business management, integrating entrepreneurs in the relationship networks that afford them access to certain resources is a key factor in their business's future. It is, therefore, important that entrepreneurs evaluate what type of relationships they should consolidate or invest in to obtain the required resources and capabilities. Entrepreneurs can obtain financial resources through their personal networks and the associations to which they belong. They may find their technological and commercial capabilities extended if they strengthen their relationships with market agents (particularly, suppliers and customers) and may obtain organizational resources if they join associations and professional networks.

Another implication of our study is that entrepreneurs must do their utmost to maintain and strengthen their own relationship networks and to connect with and integrate into other existing ones. As Partanen et al. conclude, the importance of social capital is fundamental in the different phases of a business (marketing, innovation and sales growth). Managing and using said social capital must, therefore, remain ongoing. Yet, creating networks needs not be confined to the initiative of the entrepreneur. While large companies can create their relationship networks internally, such a task is not always feasible for entrepreneurs whose networks of contacts may initially be small. In this aspect, public authorities and, more specifically, local and regional development agencies, must play an important role when it comes to facilitating entrepreneur access to or contact with the various agents. Organizing events in which businesspeople from varying sectors participate, creating specific associations at the local level in order to bring together individuals with different capabilities, or developing activities that promote relational links among neighbors or citizens in a given area are some of the possibilities for increasing local entrepreneurs' relationship networks.

Conclusion, Limitations and Future Research

This Study analyses the personal, associative, institutional, and professional relationship networks in which the entrepreneur is involved and the resources embedded therein, and it proposes that an entrepreneur's social capital resources are determinants of his/her business' economic performance. Results show that economic performance is influenced more by institutional and professional network resources than by the other network resources. However, the entrepreneur's experience in the sector reinforces the impact of professional and institutional resources. But this study is not without its limitations and possibilities for future research. The first limitation concerns the subjective measurement of performance. Future studies should analyse the impact of networks on performance, collecting objective data on growth, sales, and benefits. In addition, the present work defines, in broad terms, the extent to which networks offer valuable resources, which are inimitable in the case of high competitive rivalry or are substitutable over time. However, research should strive to gauge entrepreneurs' perceptions of the features of the resources afforded by each network, exploring whether contextual or idiosyncratic factors in a given sector may alter the value, imitability, and substitutability of the resources embedded in personal, associative, professional, and institutional networks.

In addition, the study was carried out on a varied sample of small entrepreneurs. A differential analysis by sectors would allow us to specify the degree to which social capital affects each type of business. A more detailed description of the strategies is also needed, bearing in mind the peculiarities of each business sector, as is an analysis of the relationship between entrepreneurs' strategies and their access to resources through relationship networks. Future research should also explore the implications of firm ownership for the type of resources accessed through networks, in particular for venture capitalists.

Our study of entrepreneurs' social capital resources was conducted in Tunisia, a emerged economy. It would seem feasible to replicate the study in other similar economies, also the Euro zone countries. In a different vein, one future direction of the current research is to extend the study to other quite distinct contexts, different cultural environments depending on the role of social institutions (families, social groups, associations, etc.), or countries with different transparency and efficacy in public institutions. Only then will it be possible to evaluate the generalizability of our findings. As a first step, with our sample and the available data, we would be able to carry out a comparative analysis between the subsample belonging to rural areas and the subsample belonging to urban areas.

Finally, the study could be complemented by analyzing the various dimensions of social capital (structural, relational, and cognitive social capital) in order to shed light on which features of entrepreneurs' relationship networks (size, diversity, cohesion, relational orientation, etc.) facilitate access to useful resources. The resource-based view of competitive advantage indicates that, thanks to learning effects, many resources and most capabilities are enhanced by use. It would, therefore, prove enlightening to analyze the formation and maintenance of networks over time, in other words, the life cycle of entrepreneurs' relationship networks.