Does mimicking pay off? Trait and frequency imitation of internationalization actions and their effect on substantive performance under changing conditions of uncertainty

(1)

Does mimicking pay off? Trait and frequency imitation of

internationalization actions and their effect on substantive performance

under changing conditions of uncertainty

MSc Business Administration Thesis Strategy track

Student: Marleen ten Damme /Student № 11861193

University of Amsterdam, Faculty of Economics and Business Supervisor: Dr. Marten Stienstra

University of Amsterdam, Amsterdam Business School Version: Final

(2)

Statement of Originality

This document is written by Marleen Jeanice ten Damme who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document are 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.

(3)

Abstract

Staying ahead of competition is a core topic of strategic management. One of the difficulties of gaining and remaining a competitive advantage is the interorganizational imitation of firms. The homogeneity amongst firms that is derived from this imitation has long been, and still is, a core topic for explanations of performance differences between organizations. Research regarding the effects of imitation on substantive performance remains inconclusive and contradicting. This study tries to add to the current body of literature by studying the relationship between trait and frequency imitation on substantive performance under changing conditions of uncertainty. It tests the hypotheses of the individual relationships between trait imitation and substantive performance, frequency imitation and substantive performance and the moderation effect of the degree of uncertainty. It expects that the relationship between trait imitation or frequency imitation of internationalisation actions and substantive performance is moderated by uncertainty such that this relationship is stronger when uncertainty is high than if uncertainty is low. By studying a sample derived from the Claessens & van Horen (2015) database, 347 commercial banks from 52 Emerging Market or OECD Market countries are analysed between 1970-2013. A database with the studied variables is created from multiple databases for example COMPUSTAT, HISTORIC ORBIS database and annual reports. Outcomes of this study suggest that trait imitation of internationalisation actions has a negative relationship with performance and for frequency imitation, the relationship with substantive performance does not occur. This means that imitating firms based on traits (in terms of size of a firm) leads to a decrease in substantive performance whereas frequency imitation does not matter for substantive performance. Outcomes suggest that it does not pay off to imitate firms based on size and that being a first or second mover does is not a path towards increased substantive performance.

Key words: new institutional theory; trait imitation; frequency imitation; uncertainty; substantive performance; legitimacy; isomorphism; internationalisation actions; industry leader; distance;

(4)

1. Introduction

This first chapter introduces the topic of this study. It follows with a description of the research question, the research gap and the contributions of this study and finalizes with a description of the outline of this document.

1.1. Research background

"Now, here, you see, it takes all the running you can do, to keep in the same place. If you want to get somewhere else, you must run at least twice as fast" (Carroll, 1946, pp. 178-179). Lewis Carroll’s novel ‘Through the Looking Glass’ refers to the Red Queen effect which has important implications for competitive business positions. For businesses nowadays, it is difficult to ‘run faster’. Even though businesses focus on renewal and innovation, it remains difficult to stay ahead of competition. One of the difficulties of staying ahead of competition is imitation, the central topic of this study. Imitation implies that firms copy parts of the strategic behaviour of other firms that are perceived legitimate. The homogeneity in an industry derived from imitation of organizations strategic renewal actions, makes it difficult for firms to stay ahead of competition.

The reasons for performance differences amongst organizations is a relevant and core topic of strategic management. Due to, for example, developments in information technology, changes in the competitive environment, evolving customer preferences and frequent innovations, companies struggle to sustain their competitive advantage. These developments on the one hand require stability of short-term competitive forces and direct firms towards exploiting assets, avoiding unproven concepts and high-risk operations (Volberda, Baden-Fuller, & Van den Bosch, 2001a). On the other hand, these development direct organizations towards exploring opportunities in order to disrupt their advantage and those of their competitors (D'Aveni, 1994).

It is apparent that many organizations have a long history of imitating strategic behaviour from companies in similar or divergent industries. A recent article in the Financial Times describes the imitation behaviour of key successful and large organizations in global industries: “Amazon and Facebook grabbed headlines with ground-breaking moves last week: segueing into bank accounts and music, respectively. Ground-breaking, that is, for the West – such moves are business as normal in China. (. . .) Silicon Valley is increasingly looking to

(7)

the Chinese playbook, according to analysts. If you want to know what Amazon might do in the next three years, look at what Alibaba has done over the past two years (. . .). But while China certainly offers clues, it is not a pathway every company can follow. (. . . ). Unfortunately for consumers, investors and China’s own tech titans, it can be a tough to replicate their services in more developed markets without the advantages they have enjoyed domestically” (Lucas, 2018, p. Paragraph 1).

I derive the following key questions from these empirical findings: is this type of behaviour successful? Does imitation behaviour lead to additional substantive performance for firms?

The academic literature on imitation behaviour and its effect on substantive performance is extensive. The meta-analyses by Heugens & Lander (2009) defines a positive effect of this behaviour on both symbolic and substantive performance (Heugens & Lander, 2009). Symbolic performance is the extent to which organizations have legitimacy, status and reputation (Deephouse & Suchman, 2008), whereas substantive performance is defined as the extent to which an organization is able to generate accounting based profits or Return On Assets (ROA) (Heugens & Lander, 2009). Multiple scholars debate that isomorphism requires the selection of organizational strategies and structures to enhance substantive performance (Heugens & Lander, 2009); (Deephouse, 1999); (Chen & Hambrick, 1995). Key reasons for a positive relationship of isomorphism with substantive performance are: 1) Firms that adopt isomorphic behaviour can attract higher quality resources at better terms & conditions (Baum & Oliver, 1991). 2) Templates are often led by early adopters because of their focus on improving efficiency and quality (Westphal, Gulati, & Shortell, 1997). However, this does not suggest that late conformers do not benefit from these templates substantively (Kennedy & Fiss, 2009). 3) The demonstration of legitimacy that derives from performing isomorphic strategies gives leeway to competitive differentiation. This legitimacy improves substantive performance (Deephouse, 1999). Concluding, researchers expect isomorphism to improve substantive performance of organizations (Heugens & Lander, 2009).

However, debates in the literature on imitation behaviour and its effect on substantive performance still exist. Several studies conclude on the negative relationship between isomorphism and substantive performance (Meyer & Rowan, 1977). Three key reasons for this negative relationship are defined 1) Formal structures and processes could lead to dissatisfaction within informal work organizations (Heugens & Lander, 2009). 2) Investments

(8)

costs. Isomorphism will result in lower performance when the resources that are required, having a higher investment value somewhere else (Barreto & Baden-Fuller, 2006). 3) Sustained competitive advantage could be endangered because isomorphism lowers the firms ability to differentiate itself from its competitors (Deephouse, 1996).

1.2. Research questions, gap and contributions

New institutional theory describes three modes of imitation behaviour: outcome, trait and frequency imitation (Haunschild & Miner, 1997). The central topics for this study are trait and frequency imitation. Trait imitation is the imitation of practices of an organization that stands out in terms of specific traits e.g. success or size. Frequency imitation is imitation of practices that have been adopted frequently by several organizations (Haunschild & Miner, 1997).

Even though Heugens & Lander (2009) describe the positive effect of isomorphism on substantive performance, the specific forms of imitation behaviour as defined by Haunschild and Miner (1997) are not analysed in this study. Fernhaber & Li (2010) do describe the specific effects of trait imitation and frequency imitation on substantive performance and conclude on a positive relationship. However, the study by Fernhaber & Li (2010) focuses on new venture internationalisation actions and not on incumbent organizations. Besides, to the best of my knowledge, there is no unified understanding about the individual effects of trait imitation and frequency imitation on substantive performance moderated by uncertainty. The shortcomings of the existing studies, combined with the contradictions in the literature on the effect of imitation on substantive performance, is the synthesized gap that defines the necessity for this study.

Strategic renewal is defined as: “the activities a firm undertakes to alter its path dependence” (Volberda, Baden-Fuller, & Van den Bosch, 2001a, p. 44). The specific activities that are imitated and form the basis of this research require definition and reasoning. A variety of strategic actions were examined in earlier research on this topic for example: corporate governance decisions (Fligstein, 1985), corporate acquisition strategies (Haunschild, 1993), new market entry (Haveman, 1993), market positions (Greve, 1998), preferred investment banker (Haunschild & Miner, 1997) and, more recently, decisions in merger & acquisition strategies (Yang & Hyland, 2012). The core topic of this study is imitation of

(9)

“Globalisation has been driven by multinationals capital and technology and produced a structural change in business networks”. “(. . .) the global competitive landscape of innovation and imitation have significantly changed the relative position of many Nation-States”. (Brondoni, 2013, p. 12). Fernhaber & Li (2010) as well describe the central importance of internationalisation for this topic. They find that the international entry of a new venture is partly an imitative response to the internationalization of other firms that have the same home country industry. They also find that imitation behaviour moderates the relationship between new venture international entry and its profitability (Fernhaber & Li, 2010). Several researchers have applied their knowledge about organizational imitation to the better understanding of internationalization actions by incumbent firms (Gimeno et al., 2005); (Henisz & Delios, 2001). Research describes the central importance of imitation of the internationalization strategy of industry firms, therefore, the strategic renewal actions involved in this study, are imitation of strategic internationalisation actions.

This paper contributes to existing literature by extending the current research on imitation of incumbents’ internationalisation actions. It contributes to the findings on imitation behaviour of incumbents on substantive performance and studies the specific and individual relationship of trait and frequency imitation of internationalisation actions on substantive performance moderated by uncertainty. This moderating effect is interesting for research as it can give further insight into the effects of uncertainty on organizational imitation and substantive performance. The organizational environment provides a context in which individual efforts of dealing with uncertainty and constraints rationally, can lead to homogeneity (DiMaggio & Powell, 1983). Haunschild & Miner (1997) suggest that trait and frequency imitation are more likely under changing conditions of uncertainty (Haunschild & Miner, 1997). By including uncertainty as a moderator, companies that perceive more uncertainty than others can be analysed. To conclude, this study adds to the current theoretic field as the perspectives of trait imitation and frequency imitation of internationalisation actions and their relationship with substantive performance moderated by uncertainty have not been combined yet.

This study gives answer to the central research questions: What is the effect of trait imitation

and frequency imitation of internationalisation actions on substantive performance of incumbent firms? How is this moderated by uncertainty?

(10)

The practical relevance of this study is proven by its contribution to business by giving insight in the topic of imitation and its effect on substantive performance. Businesses can use this study to become informed on making strategic decisions about if it pays off to imitate based on traits or frequency and can determine a business strategy accordingly.

1.3. Thesis outline

This first chapter introduced part of the previous literature on trait and frequency imitation, substantive performance and uncertainty to define the research gap and contributions of this study. The second chapter of this deductive study gives a full review on the current state of literature on these relevant topics. The third chapter explains theoretical framework by presenting the hypotheses derived from the literature review and gives a visual representation of the studied hypotheses in the conceptual model. The next chapter describes the data collection method and statistical procedures that are performed. The results are explained in chapter four. Subsequently, chapter five discusses the results of the descriptive, bivariate and regression analysis. Finally, chapter six describes the discussion, conclusions, limitations and implications for future research.

(11)

2. Literature review

This chapter describes the relevant findings of current literature regarding strategic renewal, new institutional theory, isomorphism, strategic imitation behaviour, imitation of internationalization actions and uncertainty. In a first attempt to answer the stated research questions posed in the first chapter of this study, the existing theories in the literature are examined. The context of this paper is new institutional theory (Scott, 2001); (Tolbert & Zucker, 1996); (Wicks, 2001). As both trait imitation and frequency imitation depend on their social environment, this is the most suitable context (Fernhaber & Li, 2010). This chapter is structured by using a funnel approach, this means that first the broader concepts of this study are described and each following paragraph zooms into the core topic of this study.

2.1. Strategic renewal

Organizations in today’s increasingly dynamic environments are characterized by significant and unpredictable developments in the field of technology, politics and economics. Strategic renewal is a key consideration in understanding the long-term survival and success of firms as firms need to transform themselves. The central importance of strategic renewal for long term success of organizations has received little attention compared to the topic of strategic change in general (Agarwal & Helfat, 2009). This topic is evidently critical for further research as strategic renewal develops both opportunities and challenges for firms, firm groups and entire industries. There are several definitions for strategic renewal in the literature. Various researchers describe strategic renewal as a process of change (Huff & Huff, 1992); (Floyd & Lane, 2000) or as “the process, content and outcome of the refreshment and replacement of attributes of an organization that has the potential to affect its long-term prospects” (Agarwal & Helfat, 2009, p. 283). This study defines strategic renewal in line with the definition by Volberda et al. (2001a). It refers to strategic renewal as the activities that organizations undertake to transform their path dependence (Volberda, Baden-Fuller, & Van den Bosch, 2001a). These activities include managerial behaviour, exploration versus exploitation and intra-organisation learning (Volberda, Baden-Fuller, & Van den Bosch, 2001a). Strategic renewal research benefits from using multiple lenses and literature.

(12)

Managerial theories of strategic renewal as described by Volberda et al (2001a) focus on both selection and adaptation perspectives. The selection perspective suggests that renewal is limited by industry norms, scarcity of resources and structural inertia. This perspective describes that firms are similar and are limited by their core competences (Volberda, Baden-Fuller, & Van den Bosch, 2001a). On the other hand, the adaptation perspective suggests that firms change and can overcome their rigidities. It describes that firms can learn to alter their behaviour and explore new capabilities (Volberda, Baden-Fuller, & Van den Bosch, 2001a).

New institutional theory is a key contribution in the field of selection perspectives of strategic renewal (Volberda, Baden-Fuller, & Van den Bosch, 2001a); (DiMaggio & Powell, 1983). This strategic renewal is achieved by creating alignment with changing industry norms and common logics (DiMaggio & Powell, 1983); (Greenwood & Hinings, 1996). Volberda et al (2001a) describe four ideal kinds of strategic renewal journeys that firms can adopt in order to cope with increasing environmental demands and pressures.

The first journey is the Emerging Renewal journey. The selection perspective of strategic renewal drives this Emerging Renewal journey (Volberda, Baden-Fuller, & Van den Bosch, 2001a). This journey is characterized by its ‘follow the market’ principle. It is rooted in the assumption that a company should be outwardly oriented and passive, it assumes that their role is to strengthen market signals to the benefit of the organization. This reactive market approach has historically been used by several financial service firms and these firms have experienced the positive effect of adopting this emergent journey (Volberda, Baden-Fuller, & Van den Bosch, 2001a).

The second strategic renewal journey is the Directed Renewal journey. This journey describes its ‘top management should be in control’ principles. The journey assumes that they have control over the environment, that renewal is driven by managerial intentions and that the organization is adaptive (Volberda, Baden-Fuller, & Van den Bosch, 2001a). The third journey is Facilitated Renewal, with its principle of ‘increasing variety of renewal initiatives’, lower level management are active in the choices for renewal. There is much more room for learning compared to the Emerging Renewal journey, but both are passive towards the environment. The fourth journey, Transformational Renewal, entails the ‘company-wide renewal process’. In this journey top management believes that it can influence the

(13)

environment and to achieve this, working together with lower levels in the organization is key (Volberda, Baden-Fuller, & Van den Bosch, 2001a).

The Directed Renewal and the Emerging Renewal journeys are typical for traditional financial service firms, where following the market is key (Volberda, Baden-Fuller, & Van den Bosch, 2001a). The other two journeys, are less common in mature markets. These journeys represent situations where following the market is not enough or top management is not in control (Volberda, Baden-Fuller, & Van den Bosch, 2001a).

2.2. New institutional theory

New institutional theory is concerned with the influence of the institutional context on organizational behaviour and the reasons for homogeneity of organizational forms and practices (Scott, 2001); (Tolbert & Zucker, 1996); (Wicks, 2001). Whereas for example Woodward (1965) and Child & Kieser (1981) aim to explain the differences between firms, opposing studies are concerned with reasons for homogeneity amongst organizations (DiMaggio & Powell, 1983). According to this opposing research, once a field is established, there is an inescapable push towards homogenization (DiMaggio & Powell, 1983). The majority of the new institutional theorists determine that an organization requires “more than material resources and technical information if they are to survive and thrive in their social environments” (Scott et al., 2000, pp. 641-643). Substantial change occurs as a result of making firms more similar whilst not making them necessarily more efficient. This structural change seems less driven by efficiency or competition. Firms are rational actors and organizational similarity is increased whilst changing them (DiMaggio & Powell, 1983).

The similarity of processes or structures between organizations can be the result of imitation or development under similar constraints. The result of this conformity comes from similar isomorphic pressures (DiMaggio & Powell, 1983). The concept isomorphism is described by Hawley (1968) as a process that forces one unit in a population to resemble other units that face similar environmental conditions (Hawley, 1968). Meyer (1979) and Fennell (1980) define two types of isomorphism: competitive and institutional (Meyer J. W., 1979); (Fennell, 1980). Competitive isomorphism does not present a fully adequate picture of the modern world of organizations, as it focuses on firms competing for resources and customers but misses the importance of political power and institutional legitimacy (Carroll & Delacroix,

(14)

& Lander, 2009) and the perspective taken in this study. New institutional theory is a key contribution to the field of strategic renewal selection perspectives, as renewal journeys result from coercive, normative and mimetic isomorphism (Volberda, Baden-Fuller, & Van den Bosch, 2001a); (DiMaggio & Powell, 1983).

2.3. Isomorphism

DiMaggio and Powell (1983) define three types of drivers for isomorphism: normative, coercive and mimetic behaviour leading to similarity between firms. Coercive isomorphism describes pressures by other organizations on a firm on which it depends for critical resources or institutions upholding the cultural expectation of the society in which it functions. This includes political influences and pressures (DiMaggio & Powell, 1983). The second form of isomorphism, normative pressures, is associated with professionalism. These are pressures on the organization to adhere to norms set by other firms in the organizational field that define conditions and methods of work (DiMaggio & Powell, 1983). The third form of isomorphism, mimetic isomorphism, derives from an organization’s perception of other organizations’ successful actions. Responses to environmental uncertainty, for example a lack of technological understanding, unclear goals or symbolic uncertainty created by the environment, are determined as mimetic isomorphism (DiMaggio & Powell, 1983). Firms have the tendency to become similar to other organizations in their field that they perceive as legitimate or successful. According to DiMaggio & Powell (1983) “the ubiquity of certain kinds of structural arrangements can more likely be credited to the universality of mimetic processes than to any concrete evidence that the adopted models enhance efficiency” (DiMaggio & Powell, 1983, p. 148). Mimetic isomorphism occurs when organizations respond to change by imitating the behaviour of others in a standardized way (Barreto & Baden-Fuller, 2006). Legitimacy is crucial for organizations survival, when managers face unclear situations, imitation is more likely to happen then (Dacin, 1997); (Deephouse, 1996); (Meyer & Rowan, 1977); (Scott, 1987).

My question based upon this part of the literature review is if this type of isomorphism is the solution to gaining additional substantive performance? Does it pay off to imitate?

(15)

2.4. Types of imitation behaviour

Drawing on new institutional theory as described, Haunschild and Miner (1997) distinguish three modes of selective imitation behaviour.

The first is frequency imitation. This is imitation of practices that have already been adopted by large numbers of other organizations (Haunschild & Miner, 1997). There are several reasons for frequency imitation to occur. One of the reasons for organizations to imitate practices of other firms is the increasing legitimacy of that practice (Tolbert & Zucker, 1983); (DiMaggio & Powell, 1983). The desire for legitimacy leads firms to adopt legitimate practices (Meyer & Rowan, 1977). Legitimacy can also occur unconsciously, when practices linked to frequency imitation are ‘taken for granted’ and therefore adjusted to in an unconscious manner (Zucker, 1977); (March, 1981). There are various studies that support frequency imitation. For example, research by Fligstein (1985) found that an organizations tendency to adopt an organizational form, is related to the number of organizations that adopted this form as well (Fligstein, 1985). Li (1995) supports this notion by finding evidence for the statement that the probability of a firms presence in a given country is higher when there are other organizations active in this country (Li, 1995).

The second mode of imitation behaviour is trait imitation. This is imitation of practices of organizations that stands out in terms of characteristics like successfulness or size (Haunschild & Miner, 1997). Trait imitation may also occur to gain status by imitating higher status organizations (Fombrun & Shanley, 1990). Often, these organizations are large and successful (Haunschild & Miner, 1997).

The third mode is outcome imitation, this is imitation of practices that have previously produced positive outcomes for others (Haunschild & Miner, 1997). It focuses on the practices or structures that produce positive outcomes for others. Unlike trait and frequency imitation, this type of imitation does not focus on features of other organizations but on the perceived consequences of certain practices.

Research by Haunschild & Miner (1997) confirmed that these three modes of imitation function independently and do not influence each other (Haunschild & Miner, 1997). Trait and frequency imitation are within the context of new institutional theory, which is the

(16)

context of this study. These types of imitation derive from the pursuit of legitimacy or from using taken-for-granted practices.

2.5. Imitation behaviour of internationalisation actions

The previous paragraph described the literature on imitation behaviour. This paragraph describes the extensive literature on imitation behaviour of internationalisation actions.

Pereira (2015) describes the central importance of internationalisation for this topic: “The rationale for a move into a new market may be one of opportunity. Overseas markets may offer exciting new prospects, high growth potential or the chance to better serve existing clients. Alternatively, it may be one of risk, entrepreneurs could be responding to a domestic market that is saturated, has slow growth or is too competitive” (Pereira, 2015, p. Paragraph 1). “Globalisation has been driven by multinationals capital and technology and produced a structural change in business networks”. “… the global competitive landscape of innovation and imitation have significantly changed the relative position of many Nation-States”. (Brondoni, 2013, p. 12).

Various studies on internationalization of incumbent firms suggest that firms display imitation behaviour by analysing other firms in similar positions in order to make a decision (Henisz & Delios, 2001); (Yiu & Makino, 2002). Internationalisation actions are a core topic in studies about isomorphism (Yiu & Makino, 2002); (Fernhaber & Li, 2010) and imitation behaviour of internationalisation actions is the strategic renewal topic of this study. Although conventional perspectives on internationalisation actions direct attention towards economic motives, the leading perspective in this study is new institutional theory. From a new institutional theory perspective, internationalisation actions of organizations are significantly influenced by isomorphism (Yiu & Makino, 2002). Research by for example Gimeno & Hoskission (1997) debate that organizations considering internationalisation could follow the actions of other successful similar firms (Gimeno & Hoskisson, 1997). Besides this, international acting firms may dictate expectations regarding efficiency by observing the performance of other firms and implement the observations to their own business (Roberts & Greenwood, 1997). To conclude about the importance of the topic of internationalisation in mimetic isomorphism, internationally operating firms actively seek for imitation in their steps towards further internationalisation (Yiu & Makino, 2002).

(17)

The importance of internationalisation actions is underlined by the drivers of cross border integration: technological improvements and policy changes. These mechanisms of integration have engaged more countries in the world economy (Ghemawat, 2007). Firms that succeed in becoming isomorphic with their environments can maximize their legitimacy and increase their resources and survival capabilities (Meyer & Rowan, 1977).

International expansion can be explained as a process of knowledge development and commitment (Johanson & Vahlne, 1977). They argue that a lack of knowledge, as a result of differences for example language or culture, is an important obstacle to international decision-making. The necessary knowledge can only be achieved through experience abroad and this knowledge can then lead to effective strategic decisions (Johanson & Vahlne, 1977). Thus, the firm’s foreign strategy is a critical decision for the future success of the organization.

2.6. Uncertainty

In interorganizational research, uncertainty shows to affect important organizational actions and it has a strong role in imitation practices (Haunschild & Miner, 1997). Firms perceive uncertainty about efficiency and structure (March & Olsen, 1976). Haunshild and Miner (1997) debate that the perceived high level of uncertainty is a reason for imitating firms (Haunschild & Miner, 1997). Uncertainty strengthens the importance of the social

considerations such that when uncertainty increases, social comparisons are a basis decision making (Festinger, 1954). Prior research suggests that frequency and trait represent social factors and that frequency imitation and trait imitation is more likely under conditions of uncertainty (Haunschild & Miner, 1997). This means, both for trait and frequency imitation, that even though firms are likely to imitate other organizations, they are more likely to do so under conditions of uncertainty.

(18)

3. Theoretical framework

The chapter presents the theoretical framework of this study. The chapter is structured as follows: first the hypotheses derived from the literature are described, second the conceptual model, a visual representation of the stated hypotheses is presented. Third, the boundary conditions and assumptions of the model are described.

3.1. Hypotheses

Imitation and substantive performance

The concept of the effect of institutional norms on organizational performance and the effect of isomorphic templates on performance has been researched extensively (Heugens & Lander, 2009); (Barreto & Baden-Fuller, 2006). There is consensus that isomorphism is good for organizations as it “avoids confusion, makes them intelligible, makes them legitimate, gives them funding and avoids coercive state sanctions” (Donaldson, 1995, p. 201). “Isomorphism has a positive effect on symbolic performance: the extent to which they generate positive social evaluations” (Deephouse & Suchman, 2008, p. 69). Besides the effects on symbolic performance, multiple scholars debate that isomorphism requires the task of selecting organizational structures and strategies that improve substantive performance (Deephouse, 1999); (Chen & Hambrick, 1995). The meta-analysis by Heugens & Lander (2009) concludes on the positive effect of isomorphic behaviour on substantive performance (Heugens & Lander, 2009). The adoption to isomorphic templates enhances substantive performance and defined reasons are: 1) Firms that are isomorphic can attract resources of better quality at preferred terms and conditions (Baum & Oliver, 1991) 2) “Early adopters often pioneer templates because they improve efficiency and secure quality” (Westphal, Gulati, & Shortell, 1997, p. 380). “This by no means implies that late conformers stop benefiting substantively” (Kennedy & Fiss, 2009, p. 911). 3) “Isomorphic strategies do allow for competitive differentiation” (Deephouse, 1999, p. 148). Concluding, isomorphism is expected to improve substantive performance of organizations (Heugens & Lander, 2009).

However, the meta-analysis by Heugens & Lander (2009) concludes on this positive effect, theorists still debate the negative effect of isomorphism on organizations substantive performance (Meyer & Rowan, 1977). From the literature, there are three reasons defined for this negative relationship 1) Investments that are made to increase isomorphism may have positive opportunity costs: if the required resources have a higher value for investment

(19)

somewhere else, isomorphism will result in a decrease in substantive performance (Barreto & Baden-Fuller, 2006) 2) Set structures and work processes could lead to conflicts within the informal organization of work (Heugens & Lander, 2009) 3) Isomorphism lowers the firm’s ability to differentiate itself from competition, this can lead to a decrease in sustained competitive advantage (Deephouse, 1996).

Imitation of internationalization action and substantive performance

Mitchell, Shaver and Yeung (1992) studied the effect of international operations on substantive performance of the parent firm. Outcomes suggest that both an increase and a decrease in international presence has a negative effect on the survival of the parent firm (Mitchell, Shaver, & Yeung, 1992). The focus of this study is however on the substantive performance of the imitating firm.

The effects of trait imitation on substantive performance are explained by new institutional theory. Argued is that large firms in an industry serve as a reference group that other ventures look at to make decisions. Trait imitation of large industry competitors can lead to improved performance because of 1) A firm can benefit from the existing infrastructure and knowledge. Firms that imitate can benefit from the informal knowledge spill over and qualified workers that large firms possess (Birkinshaw & Hood, 2000). 2) Firms that internationalize can create an increase in demand, other firms can benefit from this increase (Fernhaber & Li, 2010). 3) By observing large existing organizations, other firms could obtain knowledge to determine if it is an efficient decision to internationalize or not and to define and research the effects on their substantive performance (Anand, Glick, & Manz, 2002).

Fernhaber & Li (2010) describe the effects of the modes of mimetic isomorphism as described by Haunschild & Miner (1997) on substantive performance. They find support for the hypothesis that the level of internationalization by the largest firm within a home country (trait imitation) has a positive relationship with new venture international entry and performance. Their findings suggest that international entry is more positively related to substantive performance in industries in which the three largest firms have a high level of internationalization (Fernhaber & Li, 2010).

(20)

Even though the research by Fernhaber & Li (2010) focus on new venture establishment instead of on incumbent firms, the meta-analysis of Heugens & Lander (2009) combined with the literature reviewed in this paragraph leads to the implication that trait imitation has a positive effect on substantive performance. It is therefore hypothesized that:

Hypothesis 1a: Trait imitation of internationalization actions has a positive relationship with substantive performance

However, as highlighted in the literature review, multiple scholars debate the negative effects of imitation on substantive performance. The negative effects of isomorphism on substantive performance are debated by Meyer & Rowan (1977) as conformity may conflict with efficiency (Meyer & Rowan, 1977) and it may create tensions with internal routines or decrease the potential to differentiate from the competition (Volberda et al., 2011). It is therefore hypothesized that:

Hypothesis 1b: Trait imitation of internationalization actions has a negative relationship with substantive performance

New institutional theory explains the social processes that lead ventures to base their international entry decision on the frequency of internationalization actions of other industry firms. This means, imitation behaviour occurs by conforming to internationalization of other firms in the industry despite their effects on profitability or efficiency. Firms imitate others in order to increase their level of legitimacy (Meyer & Rowan, 1977). Although increased performance is not a requisite for this type of imitation behaviour, it is of importance to this study. A review of the literature on the effect frequency imitation on substantive performance leads to contradicting outcomes. Honig & Karlsson (2004) find that frequency imitation is not significantly related to substantive performance, whereas it did on symbolic performance (Honig & Karlsson, 2004). However, other theorists debate that firms from Emerging Markets improve their substantive performance when they imitate other organizations with the same home country. As Emerging Market countries are generally resource poor, they benefit from large firms, resulting in an increase in substantive performance (Brouthers, O'Donnell, & Hanjimarcou, 2005). Fernhaber & Li (2010) suggest that new ventures are likely to achieve higher levels of performance based on frequency imitation as they can make use of the

(21)

available knowledge of industry firms. According to Sapienza et al (2006) knowledge of foreign markets will increase the probability of venture growth (Sapienza et al., 2006).

Fernhaber & Li (2010) confirm that, in line with the previous findings, new ventures are most likely to imitate actions that are frequently performed (Fernhaber & Li, 2010). Furthermore, they found evidence that international entry based on imitation of frequently performed actions improve substantive performance. When there is a high level of internationalization in an industry, there is a stronger positive relationship between international entry and substantive performance (Fernhaber & Li, 2010).

This study does not focus on new ventures like described in the research by Fernhaber & Li (2010). However, the literature review in this paragraph combined with the meta-analysis by Heugens & Lander (2009) leads to substantial evidence to posit the following hypothesis:

Hypothesis 2: Frequency imitation of internationalization actions has a positive relationship with substantive performance

Research by Li (1995) describes the international actions of firms as a dynamic process (Li, 1995). She describes the positive side of this uncertainty: “The results show that firms benefit from learning and experience in foreign operations, which improves the chances of success for subsequent foreign investments” (Li, 1995, pp. 333-334). Gimeno & Hoskin (1997) debate that organizations with foreign entry follow actions of other successful similar ventures because of environmental uncertainty. Imitating companies in an industry however does not guarantee efficiency as firms operate under modes of uncertainty (Gimeno & Hoskisson, 1997).

Various strategic management theorists debate that organizational fit with the environment is an important driver of firm performance (Donaldson, 2001). Research by Volberda et al (2011) describes the concept of contingency and institutional fit on firm performance (Volberda et al., 2011). Whereas contingency theory describes increasing performance as a result of the fit between environmental contingencies and the firm (Donaldson, 2001), Institutional theory explains firms’ performance as an effect of external legitimacy and support (DiMaggio & Powell, 1983). Both theories are concerned with the organizations

(22)

Volberda et al (2011) use the degree of environmental uncertainty to represent the contingency variable. Environmental uncertainty is operationalized as the level of dynamism within the environment and the degree to which changes are unpredictable. From this study, important conclusions are derived as to the effect of uncertainty on firm performance. Volberda et al (2011) strongly support the notion that higher degrees of environmental uncertainty require higher levels of responsiveness. They posit that contingency fit of a firm’s technology, structure and culture (organizational variables) with the level of environmental uncertainty, is positively related to firm performance. A better fit of the organizations variables with the level of uncertainty leads to higher performance. According to Donaldson (1987), higher performance is a consequence of fit between organizational variables and environmental factors (Donaldson, 1987). This implies a positive relation between uncertainty and performance. It is therefore hypothesized that:

Hypothesis 3: There is a positive relationship between uncertainty and substantive performance

New institutional theory considers interorganizational imitation as a standard response to uncertainty (DiMaggio & Powell, 1983). According to Haunschild & Miner (1997) uncertainty has a positive influence on imitation behaviour (Haunschild & Miner, 1997).

Mimetic isomorphism includes the process of organizational change and imitation in uncertain situations in order to gain legitimacy (DiMaggio & Powell, 1983). Suchman (1995) defines organizational 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” (Suchman, 1995, p. 573). Mimetic isomorphism provides legitimacy and is more likely to happen when managers face unclear situations (Cyert and March, 1963), as legitimacy becomes crucial for organizational survival (Dacin, 1997); (Deephouse, 1996); (Meyer & Rowan, 1977); (Westphal, Gulati, & Shortell, 1997). Barreto & Baden-Fuller (2006) confirm the existence of imitation on strategic choices and the importance of legitimacy for the dynamics of imitation (Barreto & Baden-Fuller, 2006). As organizations tend to decrease uncertainty (Cyert & March, 1963), imitation of behaviour could provide legitimization (Meyer & Rowan, 1977); (DiMaggio & Powell, 1983).

(23)

Legitimacy-based imitation enables organizations to increase their probability of survival (DiMaggio & Powell, 1983); (Meyer & Rowan, 1977).

McGrath & Macmillan (1995) posit that firms that internationalize exist in a high state of uncertainty. To improve their chances of success, they imitate what they assume as legitimate strategies undertaken by other firms in the industry (McGrath & MacMillan, 1995). Henisz & Delios (2001) studied the effect of imitation under conditions of uncertainty and conclude that when uncertainty is high, imitation provides legitimization and information to a decision in a market affected by uncertainty (Henisz & Delios, 2001).

The earlier mentioned studies agree that imitation is more likely when uncertainty arises (Henisz & Delios, 2001); (DiMaggio & Powell, 1983); (Haunschild & Miner, 1997). Firms may gain legitimacy from imitating common behaviour in an industry. According to Henisz & Delios (2001) a higher degree of uncertainty, leads to a positive relationship between imitation behaviour of internationalisation actions and substantive performance. Consequently, when uncertainty is lower, the relationship between imitation behaviour of internationalisation actions and substantive performance is weakened (Henisz & Delios, 2001).

The notion of fit as described by Volberda et al (2011) which implies that a better fit between the organizations variables and the level of uncertainty leads to higher substantive performance combined with the academic research which describes that the probability of imitation is higher under conditions of uncertainty and this positive influence on substantive performance leads to the following hypotheses:

Hypothesis 4: The degree of uncertainty moderates the relationship between trait imitation of internationalization actions and substantive performance such that this relationship is stronger when uncertainty is high than if uncertainty is low

Hypothesis 5: The degree of uncertainty moderates the relationship between frequency imitation of internationalization actions and substantive performance such that this relationship is stronger when uncertainty is high than if uncertainty is low

(24)

3.2. Conceptual model

The conceptual model illustrates the hypotheses that are tested in this study.

Figure 1: Conceptual model

3.3. Boundary conditions and assumptions

To test the hypotheses, several boundary conditions apply. This study assumes that all internationalisation actions are visible and observable to organizations. I also assume that all organizations have the resources to imitate other firms and that, when organization do not imitate, this is a conscious decision.

(25)

4. Data and method

The existing literature about the relationship between imitation of internationalisation actions of incumbent firms on substantive performance needs further elaboration in order to answer the main research question. This chapter explains the data and methods used in order to perform empirical research. The framework that examines the relationship between imitation behaviour of firms and substantive performance is based on deductive quantitative analysis. Quantitative research is applicable for objective research on the relationship between variables, to test theories and hypotheses and is therefore applicable. To test the hypotheses, this study performs secondary data analysis in the form of database research as this is the basis for the methodology of research that is built upon for example in studies by Barreto & Baden-Fuller (2006) and Fernhaber & Li (2010).

4.1. Industry

While OECD Market banks reduced their international presence (however still controlling 89% of all foreign bank assets), non-OECD Market banks more than doubled their presence internationally over recent years (Claessens & van Horen, 2015). After the global financial crisis, global banking is going through a structural transformation with a great variety of players and a more regional focus. The impact of the crisis on global financial institutions has focused almost entirely on European and U.S. banks as these were the banks that integrated globally before the crisis and thus, were affected the most by it. However, banks from Emerging Market and Developing Market countries form an important part of the global financial banking landscape as well (Van Horen, 2011). Developments in banking systems globally does not necessarily represent development regionally. Although some host countries experienced a decline in foreign bank presence between 2007 and 2013, other host countries experienced an increase (Claessens & van Horen, 2015).

Financial service institutions are specifically interesting for this research because of the extensive amount of research available in this industry, its high level of uncertainty, conflicts between exploration and exploitation and the extensive presence of imitation behaviour in the banking industry (Barreto & Baden-Fuller, 2006); (Volberda, Baden-Fuller, & Van den Bosch, 2001a); (Volberda et al., 2001b). Besides, international entry represents a visible choice for banks such that it enables this research to define international entry in a host

(26)

country as a central issue in strategy. This research is applied to new international entry of financial service institutions and provides practical relevance of the studied topic.

4.2. Sample

For this study, commercial banks that have internationalized between 1970 – 2013 are taken into account for the sample. In order to create a comprehensive database with a sample that is feasible within the restricted time frame of this study, a restriction on bank home country and host country is made. Home country is defined as the country where the majority of the stocks are held. The host country is the country where the bank internationally expanded to in the form of opening a branch in this country. Banks that do not originate from a country that signed the Convention on the Organisation for Economic Cooperation and Development (OECD) or belong to Emerging Markets as described in the Standard and Poor’s Emerging Market and Frontier Markets indexes by the World Bank (2000) are excluded from the sample. Claessens & Van Horen (2015) support the assertion that OECD and Emerging Markets are the most interesting home country markets for research regarding international expansion of commercial banks as they increased their international presence (Claessens & van Horen, 2015). Host countries include OECD Markets, Emerging Markets and Developing Markets as described by the World Bank (2000). The aforementioned restrictions create a sample of 347 commercial banks in 52 countries that internationalized between 1970 and 2013. For this study, the sample that is used, is equal to the population of the study.

4.3. Data collection

The data used in this study is retrieved from multiple secondary data sources. The sample is derived from the database on bank ownership from De Nederlandsche Bank (Claessens & van Horen, 2015). This database contains full ownership information over the period 1960-2013 for 5,498 commercial, savings and cooperative banks active in 139 countries. It contains for each bank the name of the bank, year of establishment, a description of its ownership (foreign or domestic), country of ownership and geographic region. Foreign ownership is determined at >50% of the shares being domestically owned. Only banks with more than 1 locations that have reported to Bankscope/Bureau Van Dijk are included in the database and 1% of the data is missing (Claessens & van Horen, 2015).

(27)

The original database of Claessens & Van Horen (2015) included some double counts of banks and a lack of uniformity in names of banks that belong together (for example: ING bank and ING banks). In the preparation phase of the data collection, this study unified the names of the banks that belong together. In order to create a comprehensive database that can be used for statistical analyses, information from various secondary data sources are combined into one. After having determined the sample based on the database of Claessens & Van Horen (2015), additional data is derived from the HISTORIC ORBIS database, COMPUSTAT and annual reports for each individual bank.

4.4. Measures

This chapter describes the operationalization of the variables to test the hypotheses of this study.

4.4.1. Dependent variable

This paragraph describes the dependent variable, substantive performance, which measures the effect of the independent variables.

Substantive performance

A suitable measure of substantive performance is the extent to which the company is able to generate accounting-based profits or increase their overall Return on assets (ROA) (DiMaggio & Powell, 1983). “ROA is one of the most commonly used measures of performance in the strategy and organizational literature” (Barreto & Baden-Fuller, 2006, p. 1568). Also, ROA is the indicator that is most closely analysed by bank analysts and bankers themselves, allowing comparisons between firms of various sizes (Barreto & Baden-Fuller, 2006). A commonly used time lag for substantive performance to be visible is one year. Therefore, the ROA one year after expansion is determined as measure for substantive performance. ROA is derived from the annual reports of each individual bank.

4.4.2. Independent variables

According to Haunschild & Miner (1997) a there are three types of imitation (Haunschild & Miner, 1997). Trait and frequency imitation are the two independent variables in this study. Outcome imitation is not taken into account as it is not part of new institutional theory.

(28)

Trait imitation of internationalisation actions

As Haunschild (1993) proposes, a strong test of imitation has to satisfy three conditions. These conditions are 1). Certain behaviour by the ‘industry leader’ at time period t, 2). The possibility and ability of other actors to observe the behaviour of the ‘industry leader’ and 3). Imitation behaviour of the behaviour performed by the ‘industry leader’ at a given time period t + x (Haunschild P. , 1993). Whereas ‘t’ is the moment of entry by the industry leader in a certain country and ‘x’ is the time period that other organizations follow this entry.

The independent variable trait imitation is measured by defining an ‘industry leader’ which stands out in terms of large size of the bank. Out of the created sample, an industry leader per home country is determined by its trait of being the largest in terms of size. Size is operationalized as total asset size of the bank and data is retrieved from COMPUSTAT. In total for 34 countries, industry leading banks are determined.

Trait imitation of internationalisation actions is measured when a bank with the same home country as the industry leader expands its branch to the same host country as the industry leader. This is called branch expansion. Fernhaber & Li (2010) debate that “players within the same home country are most visible to a firm and serve as a potential reference set” (Fernhaber & Li, 2010, p. 2).

Trait imitation is calculated in number of years after the industry leader that this bank establishes a branch in a foreign country. Having determined the industry leader with data from COMPUSTAT, the database of Claessens & Van Horen (2015) indicates the year of entry of the other banks. Comparing the year of entry of the industry leader to the year of entry of the other banks in the sample, trait imitation in number of years is determined. Barreto and Baden-Fuller (2006) describe that entry in a foreign country is easy, according to them it only took a few months to plan and open a branch (Barreto & Baden-Fuller, 2006). This means that imitation behaviour could take place speedily, hence branch expansion one month after the industry leader and longer is determined as trait imitation. A measure closer to zero indicates a higher indication of trait imitation occurring. Evidently, a measure further from zero indicates a less strong indication of trait imitation occurring.

This introduces the concept of distance and has important implications for the interpretation of the results. The concept of distance in this study indicates that, the higher the number of

(29)

evidence of trait imitation there is. Evidently, banks that established in a country before the industry leader and industry leaders itself were excluded from measurement of this variable.

Frequency imitation of internationalisation actions

Frequency imitation is the second independent variable and includes the imitation of actions that have been taken by a larger number of other organisations (Haunschild & Miner, 1997). There are two forms in which this can occur, consciously firms may adopt legitimate practices of other firms (Meyer & Rowan, 1977). The other form occurs when frequent visible practices become accepted by other firms unconsciously (Zucker, 1977); (March, 1981). This type of imitation is measured to analyse its relationship with substantive performance and to create possible comparisons between trait and frequency imitation.

Frequency imitation is measured as an ordinal variable. Data of all banks in the sample are taken and ranked according to entry in a host country coming from a specific home country. This variable includes all banks in the sample and the numbers are ranked based on first to entry from a certain home country to certain host country, second, third etc. The information of entry in a host country is derived from the database on bank ownership from De Nederlandsche bank (Claessens & van Horen, 2015).

4.4.3. Moderator

The moderator variable affects the direction and/or the strength of the relationship between an independent variable and a dependent variable. For this study, the moderator is expected to give additional insights and contribution to the literature that is already known about the effects of imitation behaviour on substantive performance.

Uncertainty

Uncertainty is referred to as environmental variability or volatility (Child, 1972) and is considered a dimension of environmental uncertainty (Scott, 1992). It is described as the perceived frequency of change of turnover of market instability over time and the uncertainty caused by interconnectedness between organizations (McArthur & Nystrom, 1991). Measures of uncertainty using secondary data frequently include the volatility of net sales in an industry and the volatility of operating income (Keats & Hitt, 1988). This assumes that volatility is a proxy for unexpected change and that past levels of volatility will continue in the same

(30)

relative levels for each industry (Harrington, 2001). This construct is measured as a coefficient of variation to determine the volatility of net sales. It is measured as the average net sales at moment of entry to a host country and the two years before entry. Over these three years, the standard deviation and average are determined to measure volatility. Net sales are derived from the HISTORIC ORBIS database which includes key company financials in a standardized format.

4.4.4. Control variables

As in other internationalization studies, control variables are used to avoid biased results. Previous research on the effect of imitation of internationalisation actions have used control variables for age, size and previous international experience (Fernhaber & Li, 2010) to increase internal validity. Additionally, the control variable home country is created as OECD Market countries may have different implications for this study compared to Emerging Market countries. Past performance is a control variable that is only used in the robustness check of this study to increase validity.

Age

Older firms typically tend to have a larger amount of resources and a larger network that they can build upon (Fernhaber & Li, 2010). These advantages can as well be used when internationalizing. Age can influence a company’s propensity to internationalize and grow. This control variable is operationalized by comparing the year of first existence of the bank to the year of entry of this same bank in the host country. Year of first existence of the bank is derived from the HISTORIC ORBIS database. Year of entry of the bank in a host country is derived from the Claessens & Van Horen database (2015).

Size

Whereas age leads to a higher amount of resources, large sized firms have more resources available that could influence their decision to internationalize (Bloodgood, Sapienza, & Almeida, 1996). Size is measured by taking the LOG of net sales at the moment of entry of the bank in the host country in USD (*1000). The net sales are derived from the HISTORIC ORBIS database.

(31)

Previous international experience

A company with previous international experience has a higher probability of new venture internationalization (Bloodgood, Sapienza, & Almeida, 1996). Previous international experience is a dummy variable that is operationalized by determining if the company had a branch outside of the home country before the year of entry that is included in the sample. For the dummy variable, 1 means that there is previous international experience and 0 means that there is no previous international experience. This information is derived from the Claessens & Van Horen database (2015).

Home country

As the sample home country is restricted to two country groups: OECD Market and Emerging Market, two dummy variables are created for the home country variable. A home country belongs to the OECD Market group when it is listed as an OECD country (OECD, 2018). OECD Market and Emerging Market countries are banks located in home countries that are included in the Standard and Poor’s Emerging Market and Frontier Markets indexes as determined by the World Bank classification (World Bank, 2000). For the control variable home country OECD, a 1 confirms that the variable is an OECD Market country. Consequently, a 0 disconfirms that the variable is an OECD Market country and therefore is an Emerging Market country. For the control variable home country Emerging Market, a 1 confirms and a 0 disconfirms. Information regarding home country is derived from the Claessens & Van Horen (2015) database and World Bank data (2000).

Past performance

Assumed is, that previous substantive performance might increase future performance. Therefore, the control variable past performance is included in this study. Past performance is measured as the ROA one year before entry in the host country. Data is derived from the Annual Reports of the individual banks.

(32)

4.4.5. Overview of measures

Table 1 gives an overview of the measures used in this study.

(33)

4.5. Reliability & validity

This chapter explains the reliability and validity of the studied outcomes.

Reliability

The data used in this study are all derived from secondary data sources that are critically evaluated. The Claessens & Van Horen (2015) and HISTORIC ORBIS databases prove their reliability as the databases are used for scientific research frequently. The Claessens & Van Horen (2015) database and accompanying article is used in multiple studies for example by Ogena et al (2016), Buch & Goldberg (2017) and Faia & Weder (2016). Independent audits of the Annual Reports used in this study, confirm the reliability of the reports.

Construct validity

Construct validity explains the degree to which a test measures what it claims to measure. For this study, the constructs are determined based on previous scientific literature. The literature used for the operationalization of the constructs is described in the previous chapter. The fact that constructs are operationalized based on existing academic literature strengthens the construct validity of this study.

External validity

In this study, the sample used is equal to the population. There are no missing values in the sample. Outcomes of this study cannot be generalized to other countries that are not taken into account in the sample but are present in the database by Claessens & Van Horen (2015). Outcomes of this study can as well not be generalized to other industries besides the commercial banking industry.

Internal validity

In order to increase the internal validity, control variables are included and an additional control variable for past performance in included to perform a robustness check. This strengthens the internal validity of this study as there is more certainty in measuring what is intended to be measured. A limitation in terms of internal validity is however that no certainty can be given that imitation behaviour is indeed imitation behaviour. This study focuses on

(34)

quantitative data and a decisive factual analysis on the question if imitation behaviour is indeed imitation behaviour cannot be given.

Robustness check

As described, to increase the validity of this study further, a robustness check of past performance is conducted. The effect of past performance as a control variable is added to see if similar or divergent outcomes derive. For the robustness check, the regression analyses for trait imitation and frequency imitation are run again with the control variable past performance. Past performance could abate a large portion of the explained variance and is therefore a valuable measure for the robustness check.

Please refer to the results section in chapter 5 and appendix 8.1 for the outcomes of the robustness check of past performance for this study.

4.6. Statistical procedure

Statistical analyses are performed by using the Statistical software Package for Social Sciences version 25 (SPSS). The composed database is uploaded in SPSS and several preparation and statistical procedures were performed. There is no missing data in the database, so missing data were not defined or deleted. For the control variables previous international experience, home country OECD Market and home country Emerging Markets, dummy variables are created.

After the preparation of the data, the dependent and independent variables were checked for outliers by means of a boxplot. Subsequently, a normality check for the dependent variable (substantive performance) is performed. Normality is a requisite in order to perform regression analysis in a later stage of the statistical procedure. After checking normality, univariate analyses in the form of descriptive statistics are performed. The descriptive statistics give the first insight in the data and possible errors. After the univariate analysis, the bivariate analysis is performed which includes the correlation matrix. Finally, to test the hypotheses of this study, two Ordinary Least Squares hierarchical regression analysis (OLS) are performed. The first OLS regression analysis involves trait imitation and the second involves frequency imitation. In order to determine the interaction effect, the centered mean of the independent variables (trait imitation and frequency imitation) and the moderator (uncertainty) have been computed in SPSS. In the first regression analysis, the interaction

Does mimicking pay off? Trait and frequency imitation of internationalization actions and their effect on substantive performance under changing conditions of uncertainty