U
NIVERSITEIT VAN
A
MSTERDAM
Master Thesis
The Influence of Resource Accessibility on
Startup Performance
Name: R.D.J. van Beusekom Student number: 5947162 Supervisor: Dr. Ilir Haxhi
Second Reader: dr. Francesca Ciulli
Faculty of Economics and Business MSc. Business Administration Track: International Management Academic year: 2015 – 2016
Statement of Originality
This document is written by Rinse Douwe Johannes van Beusekom who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Abstract
Many studies have been conducted on startups and the forces these firms are subject to. In prevailing literature no consensus exists on the primary factors determining startup failure or success. This study focuses on the relation between resource accessibility and startup performance, assuming a positive relation between the two is to be found. Assuming that this relation does not stand on its own we add two possible moderators to the equation: the level of internationalization and Corporate Governance. By taking these moderators into account a more complete understanding of firm performance among startups is developed. The direct and indirect effects are tested on two levels, for all startups and for domestic startups. The results of the analyses conducted reveal no significant positive relation between the variables tested. However, the analyses indicate a negative relation between resource accessibility and startup performance. This shows that resource accessibility influences startup performance, but no evidence for the proposed moderating effects is found. This study contributes to existing literature by attributing to the ongoing debate on startups and startup performance. In addition, this study shows that existing theories put forward by business studies might not fit the profile of startup companies. On a practical level this study helps those involved in startups to better understand the driving forces of startup performance, which is helpful when building a business.
Key Words:
Startups, resource accessibility, firm performance, internationalization, corporate governance, legitimation
Table of Contents 1. Introduction ... 5 2. Literature Review ... 10 2.1 Startups ... 10 2.2 Legitimacy ... 12 2.3 Corporate Governance ... 14 2.4 Internationalization ... 15 3. Theoretical Framework ... 17
3.1 Resource Accessibility and Startup Performance ... 18
3.2 Internationalization and Startups ... 19
3.3 Corporate Governance practices and Startups ... 20
3.4 Conceptual Framework ... 23
4. Data and Method ... 23
4.1 Sample and Data Collection ... 23
4.2 Measurements ... 24 4.2.1 Dependent Variable ... 24 4.2.2 Independent Variable ... 24 4.2.3 Control Variables ... 25 4.2.3.1 Country-level Variable ... 25 4.2.3.2 Industry-level Variable ... 26 4.2.3.3 Firm-level Variable ... 26 4.2.4 Moderating Variables ... 26 4.3 Method ... 27 5. Results ... 29 5.1 Descriptive Analysis ... 29 5.2 Statistical Analysis ... 30
5.2.1 Correlation Analysis All Startups ... 30
5.2.2 Correlation Analysis Domestic Startups ... 31
5.2.4 Regression Analysis ... 33
5.2.5 Regression Analysis All Startups ... 34
5.2.5.1 Independent Variable ... 34
5.2.5.2 Moderating Variables ... 34
5.2.6 Regression Analysis Domestic Startups ... 36
5.2.5.1 Independent Variable ... 36 5.2.6.2 Moderating Variables ... 37 6. Discussion ... 38 6.1 Limitations ... 40 6.2 Future Research ... 42 7. Conclusion ... 43 8. Reference ... 46
1. Introduction
In the past decades attention on international operating startups increased (McDougall, Oviatt & Shrader, 2003). Many scholars consider startups as a significant source of economic growth and development (Oviatt & McDougall, 1994; Shrader, Oviatt & McDougall, 2000; Zahra, Ireland & Hitt, 2000), challenging the traditional theories of international business (McDougall, Oviatt & Shrader, 2003). Lumpkin and Dess (1996) argue that the aim of entrepreneurship at exploiting business opportunities correlates positively with the expansion of business, technological progress and the creation of wealth.
Furthermore, startups are considered to have unique characteristics that enable them to easily adapt to new and changing environments, which gives them a competitive edge over their larger and established competitors (Knight & Kim, 2009; Musteen & Ahsan, 2013; Zahra, Sapienza & Davidsson, 2006). Despite the advantages commonly held by startups, failure rate still remains high and no consensus on the drivers of these failures exists in literature (Buschke, 2014; Hudson & Schroeder, 1984; Huyghebaert, 2006; Lee, Kelley, Lee & Lee, 2012; Ven, Giardino, Wang & Abrahamsson, 2014; Zimmerman & Zeitz, 2002).
As compared with established competitors startups are small, experience a lack of institutional legitimacy face difficulties in attracting resources vital for their survival (Hearit, 1995; Knight & Kim, 2009; Martin & Justis, 1993; Musteen & Ahsan, 2013; Scott & Bruce, 1987). First, for stakeholders to provide a startup with the resources it needs legitimacy of the startup is required (Lee et al., 2012; Rutherford, Buller & Stebbins, 2009; Starr & MacMillan, 1990; Zimmerman & Zeitz, 2002), which means that startups are dependent on the degree of legitimacy stakeholders perceive for their survival. Second, it is important to note that the
practices of Corporate Governance (CG) underpin firm legitimacy (Coglianese, 2007). In other words, when analyzing the ability of startups to attract resources it is important to take the effects of CG practices into account since these are related to the degree of legitimacy of the firm. Interestingly, Coglianese (2007) finds that CG practices are changing with respect to the institutional arrangements the firm operates in, thus implying that differences in preferred CG practices exist between institutional environments. This brings us to the third critical aspect on which startup performance depends, the degree of internationalization.
The degree of internationalization plays an important role in relation to the survival prospects of startups. With the economy becoming more global the degree of internationalization lies at the foundation of firm competitiveness (Park, LiPuma & Prange, 2015). Findings by Lee et al. (2012) show that internationalization is associated with better survival prospects for startups. However, entering unfamiliar markets could also come with additional challenges and lead to an increase in the risk of failing (McDouall et al., 2003). When entering foreign markets firms may encounter liability of foreignness, in the sense that the firm is unfamiliar with the cultural, political, and economic landscape (Zaheer, 1995). And, as argued by Coglianese (2007), preferred CG practices are moving in the direction of the institutional environment. This would imply that when operating in different geographical areas, with different institutional environments, startups face different preferences when it comes to their CG practices.
The high failure rate of startups can be attributed to the challenges new business ventures face due to their unique characteristics. Although prior research on high failure rates among startups has been conducted, no consensus exists in literature on the main drivers of these failures. Lee et al., (2012) investigate the relation
between building alliances and the survival prospects of startups. Based on their research, they argue that startups are able to increase their likelihood of survival by forming alliances, and that this will provide them with easier access to resources. And although some scholars agree that the accessibility of resources is an important factor in whether the firm will survive (Chandler & Hanks, 1994; Oriaku, 2012), the implication of resources accessibility on firm performance is still unclear. In other words, the relation between the survival prospects of startups and the accessibility of resources remains unclear. This brings us to the main research question that will be addressed in this thesis:
(RQ1) “How does resource accessibility influence firm performance among startups?”
Since, as mentioned before, the degree of resource accessibility is related to concepts such as CG practices and internationalization, these effects will be addressed as well. The institutional environment in which a firm operates influences its perceived legitimacy (Coglianese, 2007). It is therefore valuable to study the forces that are at play for startups on an international level. Furthermore, given the relation between CG practices and firm legitimacy, both national and international, a deeper understanding of the influence of CG practices on resource accessibility is needed. Existing research on firm performance in relation to internationalization and CG practices implies a positive relation between these concepts. Han (2007) argues that internationalizing provides the firm with a competitive advantage by enhancing its ability to access resources and form alliances with established actors in these markets (Lee et al., 2012; Park et al., 2015). However, as mentioned earlier, startup companies challenge the traditional theories of international business (McDougall, Oviatt & Shrader, 2003). In existing literature no clarity exists on the workings of these concept
in relation to startups. This bring us to the second and third research question:
(RQ2): “To what extent does the degree of internationalization affect the ability to attract resources among startups?”
(RQ3):”To what extent do CG practices affect the degree of resource accessibility of startups?”
The purpose of this study is to demonstrate the relation between the degree of resource accessibility and firm performance among startups. Furthermore, this study argues that CG practices and the level of internationalization have a moderating effect on this relation. The direct effect of resource accessibility on firm performance, expressed through RQ1, occurs due to the fact that startups depend on attracting vital resources from external actors in order to survive. When the degree of resource accessibility is high, startups experience fewer difficulties in attracting vital resources, which enhances their chances of survival. The moderating effects of internationalization and CG practices on the relation between resources accessibility and firm performance are expressed through RQ2 and RQ3 respectively. Through internationalization (RQ2) startups enter new markets, which provide them with access to additional sources providing resources and a wider area for product distribution. As for RQ3, it is known that firm legitimacy positively affects the ability to attract resources (Lee et al., 2012; Rutherford, Buller & Stebbins, 2009; Starr & MacMillan, 1990; Zimmerman & Zeitz, 2002). And CG practices, being related to firm legitimacy as perceived by external stakeholders, play an important role in the firm’s ability to attract resources. In other words, CG practices impact on the legitimacy of the firm and therefore the firm’s capability to obtain resources.
With a view to answering the stated research questions, an existing sample of 347 startup companies is linked to the degree of resource accessibility and firm
performance. In this study, firm performance is measured by the Return on Equity (ROE). The ROE has been chosen as measurement because attracting external capital from external actors is one of the most critical resources startups need to acquire in order to survive. The return on equity measures the return for investors on their investment and is therefore the most suitable instrument meant for firm performance.
By answering the questions presented earlier, this study endeavors to demonstrate the impact of resource accessibility on firm performance of startups. As McDougall et al. (2003) argues, startups challenge the existing business theories. However, it is unclear what the implications of existing business concepts such as resource accessibility, internationalization and CG practices are for startups in relation to firm performance. Furthermore, much literature on startups focuses on the drivers of success, the motivations for startups to internationalize and the environmental forces influencing startup performance (Bull & Willard, 1993). The mechanisms of these concepts in relation to startup performance are, however, still undervalued. It is therefore important from a theoretical viewpoint to further investigate these mechanisms in order to develop a deeper understanding of startups in general, and more specifically, the forces that are at play in relation to startup performance.
On a practical level deeper understanding of the processes and relations within startups is vital as well. Providing managers, and other actors involved with startups, with a better awareness of the challenges startups face in order to survive, will enable them to better avoid common pitfalls and proactively engage and apply sustainable strategies. Also, an insight into these constructs provides those involved with a better perception of the mechanisms at play and might improve their chances when building a business.
accessibility and internationalization. Thereafter, the theoretical framework and proposed hypothesis will be introduced and further explanation provided. Subsequently, the research methodology and results will be explained. Finally, acknowledgements on limitations will be discussed, suggestions for further research will be offered and conclusions will be presented by answering the stated research questions.
2. Literature Review
To depict a relationship between the degree of resource accessibility and firm performance of startups, clear definitions of terms and understanding are needed. In the next section, terms will be defined and explained in order lay the foundation for the relation between them. These relations will be expanded upon within the theoretical framework.
2.1 Start-Ups
Although startups have received much attention among scholars, a clear definition has not yet been determined. Scott & Bruce (1987, p. 1-2) argue that a startup needs to comply with several requirements in order to be called a startup. Its management needs to be independent and usually holds ownership of the company, the company requires capital to facilitate its establishment, and an individual or small group holds ownership. Lastly, the company operates mainly in a local area, and owners and workers are based in one home community. However, markets need not be local.
Oviatt and McDougall (1994; 2005) describe four types of startups, based on the number of value chain activities coordinated and the number of countries in which the firm is active. The first type is referred to as import/export startups. These startups
usually operate only in a few countries and the value chain activities coordinated are limited to inbound and outbound logistics. External actors, using different structures, govern other activities in the value chain. The second type refers to multinational traders. These startups coordinate only a few activities in the value chain, but are active in a larger number of countries as compared with the first type (import/export startups). Oviatt and MacDougall (1994; 2005) also refer to these first two types as ‘New International Market Makers’.
The third and fourth types of startups are defined by Oviatt & McDougal as Geographically Focused Startup and Global Startup. Here, Geographically Focused Startups operate in a limited number of countries but coordinate many activities in the value chain. As Oviatt & McDougal describe, these firms focus on serving the specific needs of a particular region. Global Startups not only coordinate a high number of activities in the value chain, but do so in many different countries. This type of startup is seen as the most difficult to establish since it requires skills on both geographic and value chain coordination.
Furthermore, startups typically differ from larger firms in terms of organizational structure and processes. In order to compete with the established larger corporations young firms need to constantly update their capabilities and respond to changes in their environment (Musteen & Ahsan, 2013). Musteen & Ashan (2013) found that successful firms are more flexible, participative, and adaptive in comparison with unsuccessful firms.
For startups, low levels of formalization in relation to an organic structure and entrepreneurial culture within the organization are typically considered an advantage as compared with large corporations (Musteen & Ashan, 2013). However, despite these advantages characterizing startups the failure rate remains high, on account of
the challenges startups face (Buschke, 2014; Hudson & Schroeder, 1984; Huyghebaert, 2006; Lee et al., 2012; Ven et al., 2014; Zimmerman & Zeitz, 2002).
Rutherford, Buller & Stebbins (2009) argue that the main challenge for startups in order to survive is the ‘quest for legitimacy’. For startups it is crucial to be perceived as a legitimate entity with a view to obtaining the resources needed for their survival. These resources must be secured from external actors who, as stated earlier, need to recognize the firm as a legitimate entity (Lee et al., 2012; Rutherford, Buller & Stebbins, 2009; Starr & MacMillan, 1990; Zimmerman & Zeitz, 2002). The next paragraph will discuss the concept of firm legitimacy in relation to the degree of resources accessibility.
2.2 Legitimacy
The role of organizational legitimacy has received broad attention among scholars (Überbacher, 2014). Suchman (1995, p. 574) defines legitimacy as ‘a generalized
perception or assumption that the actions of an entity are desirable, proper, or appropriate within some socially constructed system of norms, values, beliefs, and definitions’. Or, as Überbacher (2014, p. 667) describes, legitimacy relates to the
judgment made by resource holding parties concerning the acceptability, desirability, or appropriateness of an organization. Since startups rely on external parties for obtaining resources (Scott & Bruce, 1987), legitimacy is seen as a critical characteristic startups need to acquire further resources necessary to their survival. Thus, stakeholders need to view the firm as a legitimate enterprise to support the startup and provide the resources needed (Lee et al., 2012; Rutherford, Buller & Stebbins, 2009; Starr & MacMillan, 1990; Zimmerman & Zeitz, 2002).
Stinchecombe (1965) argues that young firms are more likely to fail due to a lack of established relationships with external actors. Startups can build on the perceived legitimacy of the firm by initiating relationships with these actors, with a view to increasing their ability to access the resources needed (Bruton & Rubanik, 2001; Zimmerman & Zeitz, 2002). This can be related to the idea of sociopolitical
regulatory legitimacy (SRL), which is derived from regulations constructed by
external actors such as governments and other powerful players in the field (Hunt & Aldrich, 1996). However, the concept of SRL stretches beyond the way in which parties react to these regulations and related sanctions. It is also based on the concept of ‘good citizenship behavior’ of firms, which means that firms operate and act in behalf of the spirit of laws and regulations (Zimmerman & Zeitz, 2002). When firms behave as ‘good citizens’ they are perceived as more legitimate by their environment (Coglianese, 2007), enabling the firm to access the resources needed.
Another way open to firms to attain legitimacy is defined in the concept of institutional isomorphism. According to the Institutional Theory, the notion of institutional isomorphism defines the forces of the institutional context that depict the behaviors of the firm within the specific environment in which it operates (DiMaggio & Powell, 1983; Drori, Honig & Sheaffer, 2009). It explains the need for firms to comply with the institutional framework of the environment it operates in and to act accordingly. When doing so, firms could unintentionally become isomorphic with other organizations that are already established in the given environment.
Related to these underlying behaviors of ‘good citizenship’ and the concept of institutional isomorphism are the CG practices a firm adopts. These practices need to be in accordance with the desired behaviors in the environment. In the next paragraph
CG will be defined and the relation between the accessibility of resources and CG will be further discussed.
2.3 Corporate Governance
Judge, Douglas and Kutan (2008) argue that economic wealth is generated when CG practices are perceived to be legitimate, on account of the accessibility of resources (Bruton & Rubanik, 2001; Zimmerman & Zeitz, 2002). Although multiple definitions on CG exist, most of them are similar. Overall, CG is seen as the outcome of a dynamic process of interaction between stakeholders and the firm (Galanis, 2013; Aguilera & Jackson, 2003; Haxhi & Aguilera, 2015). Aoki (2000) refers to this dynamic process of interaction as ‘structure’ and defines CG as ‘the structure of
rights and responsibilities among the parties with a stake in the firm’ (2000, p. 11).
The implementation of CG is executed via Codes of Good Governance (CCG). These CCG arose as a response to corporate mismanagement and corporate scandals (Enrione, Mazza & Zerboni, 2006). The codes are a set of ‘best practices’ with which to address deficiencies between formal contracts and institutions (Haxhi & Van Ees, 2010). In doing so, CCG function as instruments of self-regulation with respect to multiple aspects within the organization, such as the board, management, supervision and the process of disclosure and auditing (Aguilera & Cuervo-Cazurra, 2004). An example of the mechanisms through which CG practices influence the workings of the organization are the ownership structures at play within the organization. Ownership structures determine the propensity of management to act in their own interest, instead of the interest of shareholders (Gillan & Starks, 2003). If the CG practices of the organization sanction management to act solely for their own benefit, the willingness of external actors such as capital providers to invest in the
organization will be compromised. In other words, CG practices provide the firm with the ability to put practices in place that lead to confidence among external investors (Aguilera, Filatotchev, Gospel & Jackson, 2008). In this manner codes ‘formalize’ CG practices, which functions as a driver for firms to either comply with these codes or explain when deviating from them (Aguilera & Cuervo-Cazurra, 2004; Haxhi & Van Ees, 2010).
It is important to note that CG practices and the legitimacy of CG vary between countries (Aguilera & Jackson, 2003; Haxhi & Van Ees, 2010; Judge et al., 2008). It is the societal construct and preferences that determine the legitimacy of CG of firms (Judge et al., 2008), and therefore the accessibility of resources for the firm (Aguilera & Jackson, 2003). With respect to the increasing number of startups operating across geographical borders (Engel & del-Pacia, 2009), it is important to consider the effects of differences in CG from one country to. Thus, the level of internationalization of startups should be taken into account when analyzing startup performance. Next, the concept of internationalization will be discussed, based on existing literature.
2.4 Internationalization
An increasing number of startup companies operate across geographical borders (Engel & Del-Pacia, 2009). These startups are commonly referred to as ‘international startups’, defined by Oviatt and McDougall (1994, p. 49) as “a business organization
that, from inception, seeks to derive significant competitive advantage from the use of resources and the sale of outputs in multiple countries”. Traditionally, the economic
view on internationalization is that it provides firms with the ability to gain a competitive advantage. Here, internationalization is commonly motivated by the
belief that the firm can benefit from profit and growth opportunities outside its domestic market (Han, 2007). This is based on the underlying assumption that these firms are well established in their domestic market prior to the development of international strategies (Bell, Crick & Young, 2004).
Startups, however, experience different challenges when internationalizing due to their young age and small size, compared to their bigger and more established competitors (McDouall et al., 2003; Zaheer, 1995). Stinchcombe (1965) identified the concept of Liability of Newness (LON), which relates to the negative effects due to a lack of experience and legitimacy as perceived by external actors of the firm (Freeman, Carroll & Hannen, 1983). These negative effects entail greater disadvantages when attracting resources, due to the fact that external actors such as capital providers must perceive the firm as legitimate in order to provide the resources needed (Lee et al., 2012; Rutherford, Buller & Stebbins, 2009; Starr & MacMillan, 1990; Zimmerman & Zeitz, 2002).
Another drawback associated with internationalizing is related to the concept of Liability of Foreignness (LOF) (Zaheer, 1995). The concept of LOF is defined as
”the costs of doing business abroad that result in a competitive disadvantage”
(Zaheer, 1995, p. 342). The LOF implies that expanding organizations have a disadvantage compared to their local competitors. According to Zaheer (1995), there are four main causes for the LOF. The first relates to the geographic distance, which leads to higher costs on account of aspects such as transportation and coordination over long distances and different time zones. Secondly, firm-specific costs could occur when entering a certain market. These can be related to the firms’s unfamiliarity with the market or the lack of a foothold in the market. Third, the expanding firm might experience costs resulting from the environment it enters, such
as economic protectionism and a lack of legitimacy compared to local competitors. And fourth, the firm might be faced with costs related to its home market environment, such as governmental implied economic restrictions.
However, expanding internationally provides opportunities as well. Competitiveness of the firm finds its base in the degree of internationalization pursuant to the globalizing economy. (Park et al., 2015). Moreover, Lee et al. (2012) argue that internationalization relates to better survival prospects for startups. They argue that internationalizing provides startups with easier access to resources due to more competence in forming alliances, as compared with startups operating within only their home country.
3. Theoretical Framework
A broad range of literature on startups exists in relation to the concepts of legitimacy and resource accessibility (Chandler & Hanks, 1994; Bruton & Rubanik, 2001; Oriaku, 2012; Zimmerman & Zeitz, 2002). It is argued that the ability of firms to attract resources depends on the firm legitimacy as perceived by external actors. Also, literature agrees on the importance to startups of obtaining these resources in order to survive (Lee et al., 2012; Rutherford, Buller & Stebbins, 2009; Starr & MacMillan, 1990; Zimmerman & Zeitz, 2002). However, little is known about either the foundation of the relationship between resource accessibility and firm performance among startups or on the variables influencing this relationship.
The proposed hypothesis will be addressed in the following section. First, the hypotheses concerning the relation between resource accessibility and firm performance of startups will be explained. Second, the hypothesized moderating effect of degree of internationalization and CG practices will be elaborated on.
3.1 Resource Accessibility and Startup Performance
The main argument of this study is that the degree of resource accessibility has a direct effect on firm performance of startups. It is known that attracting capital is one of the main challenges for startups, which often leads to failure when startups fail do so (Hearit, 1995; Huyghebaert, 2006; Zimmerman & Zeit, 2002). One mechanism through which attracting resources, such as capital, positively influences survival prospects of startups is that it buys the firm time to learn and overcome obstacles (Cooper, Gimeno-Gascon & Woo, 1994). So it is clear that attracting resources is of vital importance for startups to survive. Cooper et al. (1994) argue that access to resources positively influences the ability of firms to solve problems. For example, employees’ level of education or experiences might enhance the startup’s ability to contact the right networks and provide itself with problem-solving knowledge concerning the challenges it faces (Cooper et al., 1994).
In current literature there is a consensus on the importance of resource accessibility in relation to the survival prospects of firms (Chandler & Hanks, 1995; Oriaku, 2012). This study argues that the degree of resource accessibility directly influences firm performance of startups. Obstacles commonly faced by startup firms such as financial constraints, little experience and a lack of an existing network are root causes for the high failure rate among these firms (Huyghebaert, 2006; Lee et al., 2012; Ven et al., 2014; Zimmerman & Zeit, 2002).
This study argues that the accessibility of resources is therefore a main indicator of the survival prospects of the firm. If the firm has easy access to the resources required its chances of survival will improve. The first hypothesis thus states:
H1: The degree of resource accessibility has a positive impact on startup performance.
Research by Pe’er & Keil (2013) supports the proposed relation expressed through H1. They found that, when based in ‘clusters’, startups with inferior resources compared to their competitors benefit most from the ‘cluster’. These startups are able to maximize the use of accessible resources such as skilled labor and equipment in order to compensate for the lack of internal resources (Pe’er & Keil, 2013). In other words, startups are able to maximize their performance when the resources needed are more easily accessible.
3.2 Internationalization and Startups
Internationalization is commonly believed to be motivated by the expectation of the firm that it will benefit from profit and growth opportunities outside the domestic market (Han, 2007). However, this belief is based on the assumption that the firm is well established in its domestic market (Bell, Crick & Young, 2004). In comparison with their established and larger competitors, startups experience a different set of opportunities and challenges when internationalizing (McDouall et al., 2003; Zaheer, 1995). Internationalizing startups often gain a competitive advantage through international expansion but they might also experience negative consequences, such as the liability of newness and the liability of foreignness (Freeman et al., 1983; Stinchcombe; 1965; Zaheer, 1995).
One advantage that is commonly associated with internationalization is the ability for firms to form strategic alliances (Han, 2006; Oviatt & McDougall, 1994). As Lee et al. (2012) argue, internationalizing enhances the survival prospects for startups, due to the increasing ability to form new alliances as opposed to firms
operating solely in their domestic market. When entering into new markets startups provide themselves with better access to resources. This leads to the following hypothesis:
H2: A higher degree of internationalization of startups has a positive effect on firm performance.
Research by Han (2006) shows that internationalization provides startups with the capability to form ties and networks, which are indispensable to build a competitive business as they provide startups with increased access to resources. Furthermore, being part of a network strengthens the legitimacy of the firm as perceived by external actors, which enhances the possibilities to attract resources from these actors. In sum, internationalization provides firms with the ability to access resources through alliances, which positively affects firm performance (Han, 2006; Oviatt & McDougall, 1994).
3.3 Corporate Governance practices and Startups
It is clear that attracting the resources required for startup survival is one of the main challenges startups face. According to literature, the inability to attract enough resources at an early stage is mostly caused by to the lack of legitimacy of startups as perceived by external actors and the non-existence of relationships with these actors (Bruton & Rubanik, 2001; Lee et al., 2012; Rutherford, Buller & Stebbins, 2009; Starr & MacMillan, 1990; Zimmerman & Zeitz, 2002). When perceived as a legitimate entity, external actors are willing to provide the firm with the resources needed. Also, legitimacy deflects questions on, and provides the firm with, ‘the right of existence’, enabling the firm to function within its environment (Baum & Oliver, 1996; Freeman, Carroll & Hannen, 1983; Hearit, 1995; Lee et al., 2012; Rutherford,
Buller & Stebbins, 2009). The practices of CG lay the foundation of firm legitimacy (Coglianese, 2007). These practices of CG enable the firm to gain confidence of the environment and therefore of external investors (Aguilera et al., 2008). Judge et al., (2008) argue that the constructs and preferences of the environment determine firm legitimacy through the perceived legitimacy of the firms CG practices. Another important aspect related to the legitimacy of a firm is the ability of firms to form relationships with external actors. If the firm conforms to the norms and expectations of the environment through CG practices, it is able to build relationships with important actors, which enhances the survival prospects of the firm (Baum & Oliver, 1996).
As stated earlier, firm legitimacy as perceived by external actors is of great importance for startups in order to survive. And CG practices, laying at the base of the perceived legitimacy of the firm, play a vital role in determining whether the startup is identified as a legitimate enterprise. Also, when the firm is able to incorporate GC practices with respect to the expectations of its environment, it creates the opportunity to form relationships with external actors (Bruton & Rubanik, 2001; Coglianese, 2007; Lee et al., 2012; Zimmerman & Zeitz, 2002).
Baum & Oliver (1996) state that an increase in perceived legitimacy of the firm enhances resources accessibility. To increase firm legitimacy, startups need to incorporate GC practices that are in line with expectations from the environment. In other words, startups need to act as ‘good citizens’ in order to enhance their ability to access resources and form relationships (Zimmerman & Zeitz, 2002).
Another function contributing to firm legitimacy as perceived by external actors is the protection of investors (Porta, Lopez-de-Silanes, Shleifer & Vishny, 2000). When investors’ rights are protected it increases the firms’ legitimacy and the
willingness of investors to provide the firm with the resources needed. When control of the firm is centralized within the firm it reduces the valuation of corporate assets (Porta et al., 2000), which has a negative effect on the ability of the firm to attract resources. This leads to the following hypothesis:
H3a: Decentralized company ownership positively influences startup performance.
Gillan & Starks (2003) argue that a high level of inside ownership reduces the ability of outside to influence and control the organization, which results in a decrease in value of the company. Consequently, this could lead to difficulties for the firm when attracting resources such as capital. This leads to the following hypothesis:
H3b: A greater dispersion of company ownership has a positive effect on startup performance.
The proposed relationship between CG practices and firm performance, expressed through H3a and H3b, is based on existing research on CG practices and firm legitimacy. As argued by Gillan & Starks (2003), ownership structures in the firm influences the ability of management to act in their own interest instead of the interest of other stakeholders. If CG practices do not exist in the firm or allow management to ignore the interest of shareholders it will undermine the willingness of external actors to invest in the organization (Aguilera, Filatotchev, Gospel & Jackson, 2008). Furthermore CG Literature shows that CG practices influence the firms’ ability to form ties with its environment (Bruton & Rubanik, 2001; Coglianese, 2007; Lee et al., 2012) and that firm legitimacy is affected by the CG practices the firm incorporates (Baum & Oliver 1996; Zimmerman & Zeitz, 2002). In other words, through CG practices startups are able to form relationships with their environment and increase their legitimacy, which provides the startups with greater access to resources and therefore better firm performances.
3. 4 Conceptual Framework
The proposed hypotheses are summarized in Figure 1 below.
4. Data & Method
The following section describes the methodological aspects of this study. First, the data collection will be addressed, as well as the sources used to gather the data. Secondly, information on variables used will be provided and elaborated on. Third and last the data conducted for this study and the regression models will be explained.
4.1 Sample and Data Collection
For the purpose of this study data has been retrieved from the Orbis database. The Orbis database contains information on over 200 million companies worldwide, with an emphasis on private company information. In order to create a suitable sample for this study the following criteria were used: companies with an active status, companies incorporated in or after 2011, companies with at least one shareholder and known number of total shareholders, companies with a maximum of 25 employees (categorized as small), and companies with a known number of subsidiaries abroad. These criteria reduced the sample to a total of 347 companies. However, a number of
companies needed to be excluded at a later stage due to missing data, which resulted in a total sample of 227 companies.
4.2 Measurements
4.2.1Dependent Variables
The model tested in this study consists of one dependent variable, being firm performance of startups. In existing literature, studies on firm performance of startups is commonly based on the survival prospects of firms. It is found that the ability of the firm to attract resources is vital in order to survive (Huyghebaert, 2006; Lee et al., 2012; Oriaku, 2012; Ven et al., 2014; Zimmerman & Zeitz, 2002). This study will therefore build on these findings by measuring firm performance of startups in relation to the accessibility of resources.
In this study firm performance is measured by Return on Equity (ROE). Attracting capital from external actors is seen as the main challenge for startups in order to survive (Cooper et al., 1994; Hearit, 1995; Huyghebaert, 2006; Zimmerman & Zeitz, 2002). The ROE measures a firm’s profit generation by using money invested by shareholders and is therefore an appropriate measuring instrument, since it identifies to the direct relation between the startup and its capital providers. Data on company ROE is provided by and collected from the Orbis database.
4.2.2 Independent Variables
The independent variable used in this study, resource accessibility, is measured by the amount of shareholder funds provided to the firm and the number of shareholders of the firm. Data on shareholder funds and number of shareholders is retrieved from the Orbis database.
Total shareholder funds and number of shareholder is an appropriate measurement of resource accessibility, since capital is a vital resource startups need to attract in order to survive (Cooper et al., 1994; Hearit, 1995; Huyghebaert, 2006; Scott & Bruce, 1987; Zimmerman & Zeit, 2002). Furthermore, multiple shareholders providing funds to the firm are indicative of the ability of the startup to form relations with external actors, which in itself is an important resource for startup companies (Huyghebaert, 2006; Lee et al., 2012; Ven et al., 2014). Building relationships and alliances with external actors provides the firm with the potentiality to attract resources and build legitimacy as perceived by the firm’s environment (Baum & Oliver, 1996; Bruton & Rubanik, 2001; Coglianese, 2007; Zimmerman & Zeitz, 2002).
4.2.3 Control Variables
For this study control variables are required at three levels: country level, industry level, and firm level. The variables used to measure these levels will be elaborated on in the next section.
4.2.3.1 Country level
The control variable used on country level is based on the home-country of the company. Data on country of origin is provided by the Orbis database. Countries have been dummy coded as 1 if the country of origin is located in Western-Europe and 0 if the country of origin is located outside of Western-Europe. The geographical area of Western-Europe is used as reference group as most firms originated in this specific area.
4.2.3.2 Industry level
At industry level another control variable is put in place. Controlling for industries is important since the type of industry influences the amount of capital that is invested, or needs to be invested, and this amount differs from one type of industry to another. The data has been retrieved from the Orbis database and is classified according to NACE (the European industry standard classification system), consisting of four digits. Firms are coded 1 if they belong to the industry and 0 if they do not. The following four most prominent industries are selected for this study.
- Manufacturing
- Wholesale and retail trade
- Financial and insurance activities
- Professional, scientific and technical activities
Companies are coded 0 if they do not belong to one of the industries presented above, and companies who do are coded as 1.
4.2.3.3 Firm level
On firm level the control variable is based on the number of employees of the firm in order to control for firm size, since the size of the firm might influence the amount of resources needed to operate. Furthermore, it is suggested that firm size influences the performance of the firm (Kogut & Sing, 1988).
4.2.4 Moderators
The focus of this study lies on the effect of resource accessibility on firm performance of startups. It is expected however, that this relationship is influenced by at least two
moderating variables. These moderators are (1) the level of internationalization and (2) the CG practices of the startup.
The level of internationalization of the startups is measured by the number of foreign subsidiaries of the startup. The Orbis database provides data on the total number of subsidiaries and the countries in which these subsidiaries are located. Countries in which the startup owns more than one subsidiary are counted as a single foreign subsidiary.
The CG practices are regrettably difficult to measure with the available data. Due to the fact this study focus on startup firms, little is known on specific CG practices that are incorporated in the firms. However, the Orbis database does provide information about the number of shareholders and the BvD independence indicator. The BvD Independence Indicator shows the degree of independence of the company with regard to its shareholders. The indicator is divided in four levels, A, B, C and D. The levels relate to the maximum percentage of shareholding by one shareholder. Level A indicates that no shareholder holds over 25 percent of shareholding. Level B indicates that at least one shareholder holds over 25 percent but no more than 50 percent of shareholdings. Level C & D indicate that one shareholder owns over 50 percent of the total shares.
4.3 Method
The proposed hypotheses are tested using a regression analysis. The regression analysis is used when one or more independent variables are expected to influence one dependent variable. The regression analysis is expressed through the following equation:
Here, β0 and β1, 2, 3, n represent the regression coefficients, where β0 stands for the intercept and β1, 2, 3, n indicate the slope of the relation to X1i, 2i, 3i, ni. The slope can be interpreted as the change in units by the dependent variable, as a result of the change by one unit of the independent variable. And last, the ε is the error describing the difference between the estimated Xli and the actual Xli.
To test the theoretical framework presented linear regression is conducted, an overview of the tested models is displayed in table 1.1. First, in model 1, the relation between the IV and DV is tested without controlling for any variables. Then, in model two, the control variables are tested in relation to the DV (Return on Equity). In model three, the relation between the ID and DV is tested, including the control variables presented. As mentioned, the ID and DV are respectively the degree of resource accessibility (measured through shareholder funds) and firm performance (measured through the Return on Equity). The fourth, fifth and sixth models test the expected moderating relations by adding one of the proposed moderating variables at a time, internationalization, BvD, and number of shareholders respectively. Here, the number of countries in which the startups has subsidiaries represent the level of Internationalization, and the BvD score and the number of shareholders represent the CG practices. Last, in model seven, all variables are included.
Table 1.1: Stepwise regression on performance of all startups
Model Control Variable IV Moderating Variable
Country Industry Firm Funds Internationalization BvD #Shareholders 1 X 2 X X X 3 X X X X 4 X X X X 5 X X X X 6 X X X X 7 X X X X X X X
In order to test the regression analysis for only domestic startups the same models are used, displayed in Table 1.2, with the exclusion of the level of internationalization, since this is equal to zero.
Table 1.2: Stepwise regression on performance of domestic startups
Model Control Variable IV Moderating Variable
Country Industry Firm Funds Internationalization BvD #Shareholders 1 X 2 X X X 3 X X X X 4 X X X 5 X X X X 6 X X X X 7 X X X X X X 5. Results
In the following section the results of the study are presented. The first part covers the descriptive statistics of the variables discussed. In the second part the statistical analysis will be elaborated on.
5.1 Descriptive Analysis
Analysis of the descriptive statistics, displayed in Table 2.1, delivers the mean, standard deviation and correlation or each variable presented in the model of all startups. In total 227 startup companies are analyzed in this study, spread over 24 countries throughout the OECD. Of the total number of companies 119 are domestic, meaning that these startups have no subsidiaries abroad. The descriptive statistics for domestic startups are shown in Table 2.2. The remaining 108 startups are defined as international, meaning the company has at least one subsidiary located in a foreign country.
The average ROE of the startups is calculated over the last year available in Orbis (being the year 2015). For all startups this leads to an average ROE of 26.363 US dollar, with a standard deviation (SD) of 50.617 US dollar, meaning that there is a notable variance in ROE between the companies. The same applies for the amount of shareholder funds held by the companies, with a mean of 3.538,925 US dollar and a SD of 6.507,377 US dollar. This results in a wide spread of shareholder funds held by the companies tested, which means great variance of resource accessibility within the sample.
For domestic startups the average ROE is 28.343 US dollar, with a SD of 50.159 US dollar. For the amount of shareholder funds the mean is 2.043 with a SD of 4949,159, meaning that for both ROE and shareholder funds notable variance applies for this sample.
Furthermore, no indication of multicollinearity appears in either matrices, since no correlation over 0.80 are shown.
5.2 Statistical Analysis
5.2.1 Correlation Analysis – All startups
The correlations between the DV and IV’s are summarized in table 2.1. The correlation matrix presented indicates several significant correlations, of which the notable ones will be discussed further in this section. First, significant correlations in relation to firm performance, measured through ROE, will be discussed. Surprisingly, a negative correlation is found between shareholder funds and startup performance (r=-0.246, p < 0.01). This means, that an increase in shareholder funds is related to a lower performance of the startup. Furthermore, a significant correlation is found between the number of shareholders and firm performance (r- -0.154, p < 0.05).
Which tells us that an increase in the number of shareholders has a negative effect on startup performance.
A positive correlation is found between internationalization and shareholder funds (r = 0.368, p < 0.01). Meaning that a higher degree of internationalization is related to a higher amount of shareholder funds. Furthermore, the matrix shows a negative correlation between BvD and shareholder funds (r = -0,172, p < 0,01). This indicates that the amount of shareholder funds is expected to be lower when shareholdings are more dispersed. And last, as could be expected, the number of shareholders seems positively related to more dispersed shareholdings (r = 0,297, p < 0,01). No correlation values exceed 0.80, indicating that no multicollinearity exists (Field, 203). It is important to note however, that the correlations found are relatively low overall.
Table 2.1: Descriptive statistics and correlations of all startups
Mean SD 1 2 3 4 5 6 7 RoE 26363 50622 Country 12.37 6106 0.208** Industry 2.08 1521 0.036 -‐0.076 #Employee 5.62 5360 0.098 0.333** -‐0.016 Funds 3538.923 6507.377 0.246** 0.185** 0.102 0.044 Internationalization 1.76 1333 -‐0.049 -‐0.015 0.197** 0.110* 0.368** BvD 1.93 0.631 -‐0.098 0.007 -‐0.101 -‐0.073 0.172** 0.182 #Shareholders 3.74 5618 -‐0.154* 0.087 -‐0.021 -‐0.035 0.060 0.039 0.297** *. Correlation is significant at the 0.05 level (2-‐tailed).
**. Correlation is significant at the 0.01 level (2-‐tailed). 5.2.1 Correlation Analysis – Domestic startups
The correlation analysis conducted for the sample of domestic startups, Table 2.2, shows a single significant correlation with respect to the stated hypotheses. Here, a negative correlation between shareholder funds and ROE is shown (r – 0.186, p <
0.05), indicating that an increase in shareholder funds negatively affects the ROE. However, no significant correlation concerning the other hypotheses are shown in the results, meaning that the proposed moderating effect of internationalization and CG practices does not seem to be associated with better startup performance.
Table 2.2: Descriptive statistics and correlations of domestic startups
Mean SD 1 2 3 4 5 6 RoE 28343 50.159 Country 12.50 5.655 -‐0.171 Industry 1.866 1.461 -‐0.071 -‐0.037 #Employee 4.97 5.159 0.164 0.350** -‐ -‐0.086 Funds 2.043 4949.159 -‐0.186* 0.076 0.142 0.084 BvD 2.05 0.467 -‐0.126 0.038 -‐0.040 -‐0.003 0.019 #Shareholders 3.30 2.353 -‐0.074 -‐0.119 -‐0.092 0.178 0.037 0.310** *. Correlation is significant at the 0.05 level (2-‐tailed)
**. Correlation is significant at the 0.01 level (2-‐tailed) 5.2.2 Multicollinearity
Although the results shown in the correlation matrix do not indicate the presence of multicollinearity, it is important to test the data to make sure there is none. In this section the test on multicollinearity will be explained and elaborated on.
It is important to test on multicollinearity to ensure that the findings on the explanation of variance are reliable. When multicollinearity emerges, it might seem like the variance is explained, but actually this is partially caused due to interdependencies between independent variables. As explained by Farrar & Glauber (1967, p. 93), multicollinearity constitutes a threat to both the specification and effective estimation of the relationship that is sought through the regression analysis.
When analyzing whether multicollinearity exists, one has to evaluate the variance inflation factor (VIF) and the tolerance (Field, 2013). The VIF is an indicator
of the possible linear relationship between variables. When the VIF values are greater than 10, there is a substantial change that multicollinearity exists. The tolerance statistics generally indicates a potential problem when under 0.2 and a serious problem when below 0.1. As shown in table 3.1 and 3.2 the data used for this study are not subject to problems with multicollinearity. The highest VIF values presented in the tables for respectively all startups and domestic startups are 1.240 and 1.197. Therefore, the test on multicollinearity does not lead to the exclusion of variables presented in the model.
Table 3.1: Collinearity all startups Table 3.2: Collinearity domestic startups Collinearity Statistics Collinearity Statistics
Tolerance VIF Tolerance VIF Country 0.834 1.2 Country 0.851 1.175 Industry 0.945 1.059 Industry 0.954 1.048 #Employees 0.866 1.155 #Employees 0.835 1.197 Funds 0.806 1.240 Funds 0.953 1.049 Internationalization 0.814 1.228 BvD 0.897 1.115 BvD 0.853 1.173 #Shareholders 0.860 1.162 #Shareholders 0.890 1.123 5.2.3 Regression Analysis
In this section the results of the hierarchical regression analyses are presented and further discussed, starting with the analysis for all startups followed by the analysis for domestic startups. The regression analyses are conducted based on the models presented in the method section. Outcomes of the analyses are considered significant when p < 0.05, p < 0,01, and p < 0.001.
The proportion of variance in the DV that is predictable by the IV is interpreted through R2 (R squared). The R2 measures the explained variance of the
total variance in the DV, caused by the IV. In this study, it is the variance in startup performance, caused by the degree of resource accessibility of the startup.
5.2.4 Regression analysis - All Startups
In total seven models are tested, according to the sequence as presented in table 1.1. The first model tests the relationship between the IV and DV without the control variables, which appears to be significant, with R2= 0.056 and p < 0.01. The second model tests the relation between the control variables and the DV. This relation also appears to be significant with R2 = 0.0312 and p < 0.01.
5.2.4.1 Independent Variable
The first hypothesis is tested in model 3, which covers the direct relationship between resource accessibility (shareholder funds) and firm performance (RoE). Here, the model is controlled for type of industry, firm size (measured by the number of employees) and country of origin. The outcome of this analysis shows an adjusted R2 of 0.077 with a significant level of p < 0.001. However, the data indicates a negative relationship between funds and RoE, meaning that the first hypothesis is rejected.
5.2.4.2 Moderating Variables
This section covers the proposed moderating role of internationalization and CG practices. Here, internationalization is measured by the number of countries in which the startup owns a subsidiary, and the CG practices are measured by ownership using the independency indicator (BvD) and number of actual shareholders. Before testing the moderating effects a number of steps have been conducted in order to be able to enter the moderating variables in the regression analysis. First, the moderating
variables and the IV are centered around zero by subtracting the mean. Second, by multiplying the centered variables with the centered IV the moderators are created.
The regression analyses are tested following the sequence presented in table 1, firs the level of internationalization, followed by BvD and the number of shareholders. If the explained variance of the DV significantly increases, the proposed moderating variables have indeed a moderating effect on the relation between the IV and DV.
To begin with, the effect of internationalization is tested in model four. It measures the impact of internationalization on the relation between resource accessibility on startup performance. The results show a significant moderating effect for the model as a whole, with R2 = 0.070 and p < 0.001. However, results on the effect of internationalization are proven to be non-significant with R2 = 0.042 and p > 0.05. Based on these results, the H2 is rejected.
The third hypothesis is tested in models five and six. Both models five and six appear to be significant as a whole, with respectively (R2 = 0.088, p < 0.001) and (R2 = 0.087, p < 0.001). However, the proposed moderating effects in the models seem to be non-significant. Model five tests the effect of BvD on the relationship between resource accessibility and firm performance. This relationship appears to be non-significant, with R2 = -0.0678 and p > 0.05. The sixth model tests the effect of the number of shareholders on the relationship between resource accessibility and firm performance, which also leads to non-significant outcomes with an R2 = 0.704 and p > 0.05. Based on these results H3 is rejected as proven to be non-significant.
Last, taking all variables together, model seven is tested. The results indicate the model to be significant as a whole, with R2 = 0.069 and p <0.01. Furthermore, the IV shows to be significant as suggested in model three, with R2 = -3.460 and p < 0.01.
Table 4.1: Regression analysis all startups Model 1 2 3 4 5 6 7 N 227 227 227 227 227 227 227 F-stat significance 14.457*** 3.491* 5.712*** 3.823*** 4.621*** 4.586*** 3.391** R2 0.06 0.045 0.093 0.094 0.112 0.111 0.098 Adj. R2 0.056 0.032 0.077 0.070 0.088 0.087 0.069 Constant 33.127 43.173 40.529 43.173 35.462 33.215 41.417 Control Variables Country -1.623** -0.142** -0.142* -0.146* -0.133 -0.150* Industry 0.716 0.049 0.049 0.039 0.050 0.051 Firm 0.315 0.062 0.064 0.053 0.061 0.062 Independent Variable Funds -0.246*** -0.227*** -3.460*** Moderator Internationalization Funds*Nrcountries 0.042 0.405 Moderator CG-Practices Funds*BvD -0.678 -0.907 Fund*Nrshareholders 0.704 0.428 Dependent variable: ROE * Significance level at 0.05 ** Significance level at 0.01 *** Significance level at 0.001
5.2.5 Regression Analysis – Domestic startups
In order to conduct the regression analysis on domestic startups only six models are tested, due to the exclusion of internationalization. The first model tests the relationship between the IV and DV excluding the control variables, which appears to be significant with R2 = 0.026 and p < 0.01. The second model tests the relation between the control variables and the DV. Here, results show R2 = 0.046 and p > 0.05 for the model as a whole, indicating a non-significant relation.
5.2.5.1 Independent Variable
The first hypothesis is tested in the third model, addressing the direct relationship between resource accessibility and firm performance. The analysis shows R2 = 0.078
with p > 0.05, meaning that this is a non-significant relationship, implying that resource accessibility does not influence startup performance.
5.2.5.2 Moderating Variables
In this section the proposed moderating role of CG practices for domestic firms is covered. These relations are tested through model four and five through BvD and the number of shareholders respectively. Both tests show no significant results, meaning that both the level of BvD and the number of shareholders do not significantly influence startup performance.
Last, model six is tested taking all the variables together. The results show no significant relation when testing for all variables, with R2 = 0.087 and p > 0.05.
Table 4.2: Regression analysis domestic startups
Model 1 2 3 4 5 6 N 119 119 119 119 119 119 F-stat significance 4.178** 1.847 2.416 1.763 1.380 1.773 R2 0.034 0.046 0.078 0.058 0.046 0.087 Adj. R2 0.026 0.021 0.046 0.025 0.013 0.038 Constant 32.190 42.156 40.102 44.165 42.153 41.417 Control Variables Country -0.135 -0.110 -0.149 -0.135 -0.124 Industry -0.066 -0.037 -0.070 -0.066 -0.041 Firm 0.111 0.138 0.111 0.110 0.135 Independent Variable Funds -0.186** -0.184* -0.172 Moderator CG-Practices Funds*BvD -0.112 -0.095 Fund*Nrshareholders -0.015 -0.013
Dependent variable: ROE
* Significance level at 0.05
** Significance level at 0.01
6. Discussion
The aim of this study is to provide a better understanding of the mechanisms influencing firm performance of startups. In order to do so data is tested based on the hypothesis proposed. The previous section presented the outcomes of the data analyses conducted. In this section the findings of the data analysis will be discussed, in the light of the existing literature as presented earlier.
First, as stated in hypothesis H1, the relationship between resource accessibility and firm performance is tested. Here, resource accessibility is measured through the amount of shareholder funds and firm performance through the RoE. Surprisingly, in contrast with what existing literature implies, the findings indicate a negative relationship between the amount of shareholder funds and the RoE of the firm. This negative relationship applies for both the test on all startups and the test on domestic startups. Therefore, the first hypothesis is not supported and is rejected based on the results.
H1: The degree of resource accessibility has a positive impact on startup performance.
A possible explanation for the rejection of H1 might be that startups often depend on heavy investment in expanding activities, meaning that the firm might not be profitable at all in the first years (Cooper et al., 1994). As cooper et al. (1994) argue, attracting resources such as capital buys the firm time to learn and overcome the challenges faced in the startup period.
For the second hypothesis, the moderating effect of internationalization is tested. In existing literature there is no consensus on whether internationalization has a positive effect on firm performance of startups. This study proposes that there is a relation between degree of internalization and firm performance. This is based on
arguments made by Park et al. (2015) and Lee et al. (2012), who state that international expansion provides the firm with an competitive edge over others and that international expansion leads to easier access to resources and larger markets. However, the analysis conducted on the influence of internationalization on firm performance shows no significant results. Therefore, the second hypothesis is also rejected.
H2: A higher degree of internationalization of startups has a positive effect on firm performance
Despite the possible advantages startups experience when operating internationally, a number of drawbacks are discussed in literature. For example, Stinchcombe (1965) developed the concept of Liability of Newness (LON). The concept of LON is related to negative consequences for startups when entering foreign markets, caused by their lack of experience and knowledge of that specific market. Furthermore, as argued by Stinchcombe (1965), existing actors operating in the new market might perceive the ‘newcomer’ as an inexperienced or illegitimate firm, which has a negative effect on the startup’s performance and its ability to benefit from international activities (Freeman, Carroll & Hannen, 1983).
Zaheer (1995), identified another concept related to potential drawbacks for startups when operating abroad, the concept of Liability of Foreignness (LOF). The LOF implies that internationalizing results in a competitive disadvantage on account of the costs that are involved in international activities.
Hypotheses H3a and H3b bear upon the relation between CG practices and firm performance. Based on existing literature, it is expected that CG practices affect the willingness of external actors to support to the firm. For this study, CG practices are defined by ownership of the firm. Two measures of ownership are considered
