The Rise of Regional Strategies:
Performance Implications and the Moderating
Impact of International Experience
Longitudinal Analysis of the Aerospace and Defense Industry
Master Thesis
MSc. Business Administration - International Management Supervisor: Dr Johan Lindeque
Second reader: Dr Niccolò Pisani Student: Philipp Garlichs Student ID: 10731253
Date: 26.01.2015
1
Statement of originality
This document is written by Student Philipp Garlichs 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.
2
Acknowledgements
I would like to thank my family and friends for supporting me throughout the thesis process. Equally grateful I am for the interesting ideas and the support by my supervisor Dr Johan Lindeque and my second reader Dr Niccolò Pisani.
3
Table of contents
1. Introduction ... 6
2. Literature review ... 9
2.1. The DOI-performance relationship ... 9
2.2. Firm- and Country-specific Advantages ... 11
2.3. Liability of Foreignness ... 12
2.4. The role of international experience ... 13
2.5. The regionalization theory ... 15
2.6. The DOI-performance debate in a regional context ... 17
3. Conceptual framework ... 21
4. Methodology ... 24
4.1. Sample and data collection ... 24
4.2. Classification of regional orientations ... 25
4.3. Variables and measures ... 26
4.4. Model specification ... 30
5. Results ... 32
6. Discussion ... 40
6.1. The support for the hypotheses ... 40
6.2. The industry’s shift towards a global orientation ... 44
6.3. The moderating impact of international experience ... 46
7. Conclusion ... 50
7.1. Summary and implications for theory and practice ... 50
7.2. Limitations and directions for future research ... 53
8. References ... 55
4
Index of Figures and Tables
Figure 1: The three-stage model of multinationality and performance ... 10
Figure 2: Conceptual model ... 21
Figure 3: The A&D companies’ geographic orientation compared between 1999 & 2013... 44
Figure 4: The moderating impact of high and low levels of international experience ... 47
Table 1: Regional orientations defined by the proportions of sales by region ... 15
Table 2: Classification of regional orientations ... 26
Table 3: Overview of the variables and their measurements ... 29
Table 4: Summary statistics and correlation matrix of variables ... 33
Table 5: Panel data analysis results for Profit Margin ... 35
Table 6: Panel data analysis results for Return on Invested Capital ... 37
Table 7: Panel data analysis results for Return on Assets ... 39
Table 8: Overview about the support for the hypotheses ... 40
Table 9: Correlation matrix of variables (Sub-sample 1999-2005) ... 65
5
Abstract
The major objective of this study is to investigate whether performance can be better predicted when a regionalization approach is taken to the degree of internationalization (DOI). This transition is extended by the incorporation of international experience as a moderating factor due to its importance in the DOI-performance debate. Against the regionalization argument and based on the concepts of firm- and country-specific advantages and the liability of foreignness, it is argued that the internationalization of the aerospace and defense (A&D) industry indicates incentives to expand beyond the home region, which is positively moderated by the company’s accumulated international experience. Nevertheless, considering the recent acceleration of regionalization research, regional orientation is expected to provide an increasingly important explanation of performance. The hypotheses are tested against longitudinal data from a panel of the largest 20 A&D companies whose geographic dispersion of sales was traced from 1999 to 2013. The results reveal that the A&D companies have adopted higher degrees of global orientation that are positively related to performance but not moderated by their international experience. Contrary to the identified expansion pattern, the comparison of two sub-samples shows that regional orientation is becoming relatively more important in accounting for positive performance over time. International experience, instead, draws a different explanation by indicating a compensation effect for highly globalized companies, whereas there is a less clear picture on the regional level. On the basis of the results of this research, it can be concluded that regional strategies become increasingly important for driving performance and that rethinking of the internationalizing A&D companies is required.
Keywords: Regionalization theory, performance, aerospace and defense industry, international experience, global orientation
6
1. Introduction
The recent acceleration of research in the regionalization theory of international business (IB) has led to new insights and is increasingly calling for specific application and longitudinal data to examine the importance of multi-national enterprises’ (MNE) regional strategies (Ghobadian et al., 2014; Oh and Rugman, 2014). Consequently, this study tries to understand whether performance can be better predicted if we take a regionalization approach to the degree of internationalization (DOI).
The theoretical rationale to extend the DOI-performance debate derives from Rugman and Verbeke (2004). Based on dispersion of sales, the authors identify that MNE operations are primarily located in the extended Triad regions, namely NAFTA, EU, and the Asia Pacific region with a strong focus on their respective home region (Rugman and Verbeke, 2004). Thus, most MNEs are regionally and not globally oriented. In line with a majority of regionalization studies, regional orientation refers to the home region orientation (Gilbert and Heinecke, 2014). The regional agglomeration is largely attributed to the concept of liability of foreignness (LoF), which describes the additional costs a company has to bear in a market abroad due to its foreignness (Zaheer, 1995; Rugman and Verbeke, 2004). According to Rugman and Verbeke (2008) and Qian et al. (2013), regional integration indicates the lower degree of LoF within than between regions, which is supported by certain patterns of trade and foreign direct investment (FDI) (Rugman and Oh, 2013; Gilbert and Heinecke, 2014). Zaheer (1995) argues that the spatially and psychically closer a company is to a foreign market, the lower the LoF and therefore the costs of expanding the operations abroad. Consequently, regional characteristics, such as low distance, common currency, or free trade agreements, can be considered as main reasons for regionalization patterns (Zaheer, 1995; Rugman and Verbeke, 2004).
Whereas the LoF describes the additional costs a foreign company has to bear, the benefits of doing business abroad are conceptualized as firm-specific advantages (FSAs) and country-specific advantages (CSAs) by Rugman (1981). These unique capabilities attributed to the firm or the location can be leveraged, transferred, or recombined within the MNE network in order to create competitive advantages (Rugman and Verbeke, 2008). Thus, the trade-off between advantages and disadvantages of internationalization a company is facing is described by FSAs and CSAs on the one hand and the LoF on the other hand (Almodóvar and Rugman, 2014).
Even though these concepts are widely accepted and their regional characteristics are increasingly integrated in IB research, it is still highly controversial whether and to which
7 degree internationalization and, in this context, regional orientation can explain MNE performance (Sukpanich and Rugman, 2007; Kudina, 2012; Qian et al., 2013; Gilbert and Heinecke, 2014; Oh et al., 2014). Research has shown the importance of regional strategies in various industries ranging from the world’s food and beverage MNEs (Filippaios and Rama, 2008) to the world’s automotive sector (Rugman and Collinson, 2004). However, these investigations primarily focus on the performance implications of different regional strategies (Rugman and Collinson, 2004; Collinson and Rugman, 2008; Filippaios and Rama, 2008; Rugman and Verbeke, 2008). More recently, researchers are increasingly interested in how regional orientation has changed over time and how performance is affected (Oh and Rugman, 2014; Oh et al., 2014).
A frequently applied and adapted model in order to show the process of internationalization with regard to regionalization theory is the horizontal S-curve model by Contractor et al. (2003), which illustrates the basic relationship between international expansion and performance (Almodóvar and Rugman, 2014; Oh and Contractor, 2014; Oh et al., 2014). One of the central moderators of this relationship is a company’s accumulated international experience, which importance in the course of internationalization is highlighted by the widely accepted process model by Johanson and Vahlne (1977) that argues for expansion in dependence on a company’s experience. Kudina (2012) emphasizes the dual function of international experience in reducing the LoF a company is facing on the one hand and developing experienced-based FSAs on the other hand. Taking this into account, one would intuitively assume the positively moderating impact of international experience on the relationship between DOI and performance, however, empirical evidence is limited and the transition to a regional level of analysis is insufficiently studied (Luo and Peng, 1999; Ruigrok and Wagner, 2003; Mohr et al., 2013).
The dynamics in the beneficial character of global or regional orientations and the corresponding moderating impact through international experience are particularly important for the recent environmental changes in the global aerospace and defense (A&D) industry. The primary reason for that is a shift in demand conditions that affects regional orientations while these companies are strongly embedded in their regional or national environments due to contractual agreements with governments (Deloitte, 2014). Whereas US A&D companies are primarily home region or even home country oriented, European A&D companies are rather bi-regional oriented, because the US is by far the biggest market for A&D products (Rugman, 2005). In 2013, the US defense spending exceeded the amount spent by the next 13
8 countries combined (Washington Post, 2013, January 7). This imbalance shows how the US market was shaping the geographic orientation of A&D companies.
Patterns of global defense spending are however changing. The US government has enacted the “US Budget Control Act” in 2011, which plans to reduce the defense budget from 645.7 billion USD in 2012 annually by 54.7 billion USD from 2013 up to 2021 in order to reduce the fiscal deficit (Center of Budget and Policy Priorities, 2012). In contrast, emerging markets such as China, India, or Brazil show annual growth rates in defense spending ranging from 6-10% between 1992 and 2012 (McKinsey & Company, 2014). This change in demand conditions clearly illustrates the pressure on A&D companies to re-invent their business model through geographic and/or product diversification in order to survive in the increasingly competitive environment. Region-based sales data shows that the A&D companies have anticipated the change in demand conditions and have increased their degree of global orientation (Deloitte, 2014). It remains open whether their adopted geographic orientations drive their corporate performance and how their accumulated international experience moderates this relationship. Based on previous findings, it is generally expected that the higher the degree of global orientation, the more the A&D companies can make use of scale and scope economies and drive performance, which is positively moderated by their accumulated international experience (Yang et al., 2013; Mohr et al. 2013). The term degree of global orientation is used to represent the DOI within the context of regionalization theory and to delineate the regionalization theoretical paradigm from the DOI-performance debate.
In order to investigate the internationalization of the A&D industry and its performance implications with regard to global and regional strategies, an industry-specific panel data analysis is conducted. This contribution to regionalization literature is in line with recent suggestions by Asmussen and Goerzen (2013) for future research to underpin the relationship between geographic dispersion and performance in a regional context. Another contribution arises through the sector specific longitudinal data regarding regional strategies, encouraged by Oh and Rugman (2014) in the sense of a further investigation of the development of regional orientation over time. In a more refined approach, the beneficial character of global or regional strategies and the associated impact of international experience is investigated by comparing two sub-samples, thus, allowing for the identification of trends in the relative importance of global or regional orientation for driving performance. Besides the theoretical contributions, there are also important managerial implications for A&D companies indicating the necessity of rethinking with regard to their prevalent modus operandi.
9
2. Literature review
2.1. The DOI-performance relationship
The relationship between the degree of internationalization and firm performance is a central theme in international business research, which has only received limited attention in the regionalization literature (Sukpanich and Rugman, 2007, Kudina, 2012; Oh and Rugam, 2010; Oh and Contractor, 2014). The regionalization argument (Rugman & Verbeke, 2004) is a special case of the degree of internationalization and these literatures are integrated below to provide the foundation for assessing how regionalization strategies affect MNE performance. First, the basic relationship between DOI and performance will be discussed. In a second step, the driving and restricting forces of internationalization are explained. Therefore, the conception of FSAs and CSAs on the one hand and the LoF on the other hand are introduced. Further, international experience as a key moderator is introduced. Finally, the DOI-performance evidence is embedded and specified within the context of regionalization theory. Over 100 studies have focused on a general relationship between DOI and performance highlighting the controversy in IB research over the relationship (Hennart, 2007). Whereas earlier literature has emphasized the purpose of risk reduction, more recent work is increasingly identifying profitability as a company’s motive for geographic diversification (Hennart, 2007). In this sense, it is generally assumed that a higher DOI leads to higher performance due to economies of scale and scope, learning, and resource access (Qian et al., 2008). The underlying concept is the internalization of transactions due to imperfect markets as stressed by Rugman (1979). Rugman (1981) further refines his theory through the conception of FSAs and CSAs, which has become increasingly elaborated and largely accepted as driver of international expansion in IB research. These driving forces of internationalization are contrasted by the additional costs an internationalizing company has to bear that are embodied in the concept of LoF (Zaheer, 1995). The conceptions of FSAs/CSAs and the LoF will be discussed in-depth later in the literature review.
Besides these two main forces of internationalization, researchers have provided various analyses investigating the DOI-performance relationship with regard to a diverse set of impact factors (Vermeulen and Barkema, 2002; Chang and Rhee, 2011; Chao and Kumar, 2010). From a process point of view, Vermeulen and Barkema (2002) provide evidence that due to a firm’s limited capability to absorb foreign subsidiaries, the pace, rhythm, as well as product and geographic scope of international expansion negatively moderate MNE performance. In contrast, Chang and Rhee (2011) argue that a rapid expansion can be beneficial for performance when an industry is highly internationalized and when firms
10 possess competitive advantages due to resources and capabilities. Chao and Kumar (2010) take a contextual approach and argue that successful internationalization is largely defined by institutional settings, such as social networks. Others, such as Bausch and Krist (2007), provide evidence that a company’s capabilities and intangible assets, e.g. in form of R&D, positively influence the DOI-performance relation.
Importantly, all these approaches presented above rely heavily on organizational learning and experience, as acknowledged by Hutzschenreuter et al. (2007) who emphasize that path-dependent experience is a common feature of internationalization theories and a central moderator of the DOI-performance relationship. This path-dependent experience will be analyzed in-depth later in the discussion.
At this point it can be summarized that even though driving and restraining forces of internationalization are broadly analyzed in IB research, how DOI and performance exactly relate to each other is still one of the central debates in the field. Researchers have identified positive and negative linear, U-shaped, as well as inverted U-shaped DOI-performance relationships (Brewer, 1981; Grant, 1987; Gomes and Ramaswamy, 1999; Ruigrok and Wagner, 2003). More recently, a three-stage model hypothesizing a horizontal S-shaped relationship between DOI and performance, see figure 1, was introduced by Contractor et al. (2003) and verified and further refined by Lu and Beamish (2004).
Figure 1: The three-stage model of multinationality and performance Source: Contractor et al., 2003
11 Following Contractor et al. (2003), an internationalizing company is first faced with LoF and learning costs, which lead to lower performance. In the second stage, a company’s FSA/CSA combinations, which are partly also result from international experience, are expected to lead to competitive advantages, such as scale or scope economies. These advantages are supposed to lead to a superior level of performance while the further experience gained reduces the effects of any LoF. At a certain degree of internationalization it is expected that the performance starts to decrease with further expansion, primarily driven by a saturation effect of profitable markets and the exponentially increasing coordination costs due to the complexity of the MNE network (Contractor et al., 2003). Due to the strong home country orientation of A&D companies it is expected that the majority could still be at the first stage of this process. Only a few companies, such as Airbus Group, Boeing, or General Electric, represent clearly second stage progress and generating positive returns at a high DOI.
In order to provide a solid foundation to transfer the DOI-performance debate to a regionalization level, the following sections are focused on analyzing driving and constraining forces of internalization as well as international experience as moderator more deeply.
2.2. Firm- and Country-specific Advantages
According to Rugman and Verbeke (1992), the driving forces of internationalization are rooted within a company’s FSAs and a home-/host-country’s CSAs. FSAs encompass proprietary assets and transactional advantages, such as routines and capabilities (Rugman and Verbeke, 1992). Intangible assets, frequently measured by R&D intensity, are largely used as proxies for a company’s FSAs (Rugman and Sukpanich, 2006). The difficulties to imitate or substitute intangible assets when compared to tangibles are the rationale why they are expected to create a higher value and thereby a specific advantage (Barney, 2002). FSAs can be further distinguished as location bound and non-location bound, with the former being exploited only in a particular location and hard to transfer across borders, the latter being exploited on a global basis and more easily transferred across borders within the MNE network (Rugman and Verbeke, 1992). Furthermore, Rugman and Verbeke (2004) identify upstream FSAs (related to measures of assets) to be more transferable than downstream FSAs (related to measures of sales) due to the specific local demands in sales activities. In contrast, CSAs are a wide range of location-bound benefits for companies, such as factor endowment or taxation (Rugman and Verbeke, 1992). CSAs can either be used in a static way in a host country environment or in a leveraged way to generate new FSAs through a recombination process of host country CSAs and MNE FSAs (Rugman and Verbeke, 1992; Luo, 2002).
12 As an MNE is operating in more than one country, subsidiaries are in the position to leverage their specific CSAs to generate or to recombine transferable FSAs with host country CSAs to create competitive advantages (Rugman and Verbeke, 1992). Exemplified by the A&D industry, FSAs can be expected to be strongly related to technological know-how, whereas basic CSAs are market attractiveness and advanced factor endowments. For defense focused companies the major market pull factor is in the end the defense budget of a certain country (McKinsey & Company, 2014). Considering the controversial character of the industry, political support for the sector can be seen as a CSA, whereas a company’s lobbying capabilities can be seen as a corresponding FSA illustrating the broadness and industry-specificity of the FSA/CSA conception. All in all, competitive advantages through FSAs, leveraged CSAs or recombination are expected to ultimately lead to superior performance (Verbeke, 2009; Rugman et al., 2012).
2.3. Liability of Foreignness
Having considered the driving force of international expansion, this section presents the restricting force, namely the LoF, in order to explain why going abroad implies not only additional revenues but also additional costs.
Rugman and Verbeke (2003) argue that the main limiting force for expanding business operations across borders is the LoF. The term was introduced to IB research by Hymer (1976) who describes LoF as the additional “costs of doing business abroad” that a company has to bear due to its nonlocal status. Zaheer (1995) specifies the LoF concept in a more detailed way by categorizing the associated costs. According to Zaheer (1995), LoF occurs due to spatial distance, unfamiliarity, lack of roots in the host country environment, lack of legitimacy and economic nationalism, and/or home country restrictions (Zaheer, 1995). For European A&D companies, for example, additional costs of entering the US market might be relatively high as they might experience difficulties in competing with large US prime contractors that are preferred by the US government. Similar, different legal restrictions regarding the import and export of defense-related products impact the LoF that A&D companies are facing. Exemplifying the additional costs of doing business abroad shows that especially the institutional environment shapes the LoF for A&D companies. The resulting disadvantage for foreign companies can be compensated through their FSAs (Hymer, 1976). FSAs that create sufficient value to outweigh the additional costs arising from LoF lead to competitive advantages in foreign markets, e.g. due to technological leadership (Zaheer, 1995).
13 The LoF concept can be delineated from other approaches to decompose distance, such as Hofstede’s (1983) national cultures or Ghemawat’s (2001) distance concept, as the financial manifestation of the distance consideration mentioned above and serves thereby as a foundation to explain the internationalization of MNEs (Rugman and Verbeke, 2004; Asmussen, 2009).
2.4. The role of international experience
Internationalization is fundamentally driven by weighing the advantages derived from FSAs and CSAs and the disadvantages arising from a LoF. The factor that might tip the scale is a company’s accumulated international experience (Goerzen and Asmussen, 2007, Kudina, 2012). International experience can thereby be seen as the learning process of internationalization comprising experiences in foreign markets that gradually build a company’s knowledge position (Eriksson et al., 2000). Learning and experience are in this context a very broad conception to acquire tacit knowledge in order to decrease uncertainty encompassing personal experience, imitation, or cooperation (Forsgren, 2002, p. 261). Consequently, experience can be considered an intangible FSA that appears in a variety of forms, e.g. internal processes or human and relational capital (Forsgren, 2002). Because every experience influences in some way a subsequent decision, the firm is acting in a path-dependent way (Eriksson et al., 2000). The role of path-dependency can best be captured by the wording of Eriksson et al. (2000),
“Path-dependence is the incremental process where the pattern of behavior by firms is contingent upon and a function of its past international experience. The knowledge accumulated in the past forms the trajectory for the future internationalization behavior of firms.”
(Eriksson et al., 2000, p 308)
There are two underlying concepts that need to be mentioned when talking about path-dependency as they crucially shape the process of organizational learning, namely “bounded rationality” and “bounded reliability” (Verbeke and Greidanus, 2009). Bounded rationality considers the ability of the human mind to process limited information in a limited timeframe (Gigerenzer and Goldstein, 1996). Thus, bounded rationality affects the exploitation of FSAs (Verbeke and Kenworthy, 2008). Bounded reliability, in contrast, accounts for the limitation of individuals to realize a desired outcome (Verbeke and Greidanus, 2009). Consequently, knowledge accumulation and application, such as transferring and exploiting non-location bound FSAs, is limited by the ability of managers to realize intended outcomes or prevent
14 other parties from not living up to their commitments (Verbeke and Greidanus, 2009). Importantly the concept of bounded reliability does not, as in the case of opportunism assume business actors will seek to actively take advantage of partners when it is rational to do so, but rather recognizes that preferences may change (Rugman and Verbeke 2005; Gilbert and Heinecke, 2014).
Ruigrok and Wagner (2003) embed the path-dependent experience defined as organizational learning as key moderator in the DOI-performance relationship and provide evidence that MNEs undergo a learning process that leads to superior performance. Bowen (2007) provides further support for this argument by identifying international experience as a main source of heterogeneity among internationalizing firms and their performance outcomes. More precisely, Hsu and Pereira (2008) operationalize organizational learning into social, market, and technological learning. Whereas technological learning has no significant impact, social and market learning are identified to positively moderate the DOI-performance relationship. The importance of social and market learning illustrates the significant influence of cultural and economic distance (Hsu and Pereira, 2008).
From a theoretical point of view, the effect of organizational learning is captured by the internationalization process model by Johanson and Vahlne (1977). Different to mainstream economic theory by Hirsch (1976), which is based on discrete rational choices, these authors explain FDI patterns as a learning-by-doing process in which a company’s resource commitment and geographic scope increases over time as it gains more experience (Johanson and Vahlne, 1977).
Whereas these studies indicate the importance of international experience in the DOI-performance debate, Luo and Peng (1999) and Ruigrok and Wagner (2003) provide evidence showing that the amount of experience gained depends crucially on the number of years a company is operating abroad. To which degree a company can make use of its international experience, however, depends largely on the company’s DOI (Luo and Peng, 1999). The underlying reason is that the applicability of experience in creating FSAs and reducing LoF increases with geographic dispersion, illustrating the moderating impact of international experience on the relationship between the DOI and performance (Luo and Peng, 1999).
In conclusion, the DOI-performance debate has produced vast knowledge about how DOI and performance can be related to each other and which factors pose major influence. But in order to analyze corporate performance in the contemporary IB landscape the dimension of DOI needs to be adjusted, because most MNEs have been observed to have regional strategies of different natures (Rugman and Verbeke, 2004).
15
2.5. The regionalization theory
Traditionally, the growing IB research has led to the mainstream globalization theory that implies companies are becoming more global due to increasing heterogeneity among markets, demand, and regulation (Kim and Hwang, 1992).
More recently, Rugman and Verbeke (2004) have shown that the global business landscape is regionalized, which is supported by Ghemawat’s (2003) work using macro level data, and calls this degree of incomplete global market integration semi-globalization. According to Rugman and Verbeke (2004) the world’s economic activity is dominated by the Triad regions, namely NAFTA, EU, and Asia Pacific. This regional classification extends the so-called Triad power concept by Ohmae (1985) who initially incorporated only Japan instead of the Asia Pacific region. The definition of the Triad regions derives from the fact that those regions are home of the most MNEs and the primary geographic origin of business innovations (Ohmae, 1985; Rugman and Verbeke, 2004). The region-based agglomeration is the result of institutional, cultural, administrative, economic and geographic characteristics (Rugman and Verbeke, 2004). From a theoretical perspective, the process model of Johanson and Vahlne (1977) is arguably reflected in the identified regionalization patterns as stages of international expansion. Rugman and Verbeke (2004) assess the sales proportions of companies operating in these regions and characterize them as being either home region oriented, host region oriented, bi-regional oriented or global oriented. This conception in detail is shown in table 1 as a set of criteria based on the distribution of MNE activities through measures such as sales, assets or employees (Rugman and Verbeke, 2004).
Classification Home region 1. Host region 2. Host region
Home region oriented ≥ 50 % - -
Host region oriented - ≥ 50 % -
Bi-regional oriented 20% ≤ x < 50% 20% ≤ x < 50% -
Global oriented 20% ≤ x < 50% 20% ≤ x < 50% 20% ≤ x < 50%
Table 1: Regional orientations defined by the proportions of sales by region Source: Rugman and Verbeke, 2004
The data show that most of the Fortune 500 companies are home region oriented, while just a small percentage of the overall sample represents purely global companies (Rugman and Verbeke, 2004). Asmussen (2009) further provides measures of intra- and inter-regionalization identifying that most home region oriented companies are in fact home country oriented. This is supported and further specified by the findings of Hejazi (2007) who
16 shows that the economic activities of US MNEs within the NAFTA region are primarily driven by their US operations. Taking this home country effect into account is particularly important for analyzing the A&D industry as it is strongly rooted in the U.S. economy. These further findings resulted from criticism raised about the classification and methodology applied by Rugman and Verbeke (2004). Osegowitsch and Sammartino’s (2008) data review shows that there are far more bi-regional and global firms than a decade ago, thereby casting doubt on Rugman and Verbeke’s (2004) arguments, but more recently, Oh and Rugman (2014) refute significant changes in regional orientations through a longitudinal study of the
Fortune 500.
Even though Rugman and Verbeke’s (2004) approach is largely accepted in IB research, it breaks up the continuous character of the DOI by introducing different types of regional orientations that neglect this continuum. This can be illustrated briefly by the comparison between a home-region and host-region orientated company. The company in the host-region could concentrate its operations just within one foreign country, whereas the home-region oriented company could have its operations within all countries of its home region. Although one company is operating in another region, it is less internationalized than the one operating in its home region. Consequently, in contrast to previous research, such as Johanson and Vahlne (1977), the classification by Rugman and Verbeke (2004) is not based on an underlying continuous scale of DOI. As this analysis aims to embed the DOI-performance debate within a regionalization context, it is crucial to have a common basis for the two theoretical paradigms, therefore, Rugman and Verbeke’s (2004) approach is less suitable and requires adaptation to the focal issue of MNE’s degree of global orientation and performance.
A different approach for the classification of regional orientations, which is based on breadth and depth of international operations and allows for an increasing degree of global orientation throughout its regional orientations, is introduced by Aggarwal et al. (2011). MNEs are characterized in dependence on distribution of sales and investments and can be recognized as purely domestic, home-regional, trans-regional, or global (Aggarwal et al., 2011). The classification introduces a transition from the DOI way of thinking to the regionalization theory as it assumes regional orientations based on an increasing geographic scope. Another crucial benefit of this classification is that it takes into account that home region orientation is often based on home country orientation as shown by Hejazi (2007), which is largely neglected by the approach of Rugman and Verbeke (2004). As A&D companies are strongly embedded in their home countries due to governmental contracts,
17 Aggarwal et al.’s (2011) classification is deemed more appropriate for this study. This issue will be further elaborated in the methodology section by introducing an integration of the two approaches in order to account for the specific focus of this study while acknowledging the prevailing concept in regionalization research (Rugman and Oh, 2010; Oh et al., 2014).
In general, the different regional strategies pursued by the MNEs call for an adaptation of the DOI-performance debate, because the basic concepts to explain internationalization, namely the FSA/CSA conception and the LoF, are fundamentally influenced by the regional agglomeration of business (Kudina, 2012).
2.6. The DOI-performance debate in a regional context
This section is concerned with specifying the DOI-performance debate through the regionalization perspective. The importance of the regionalization theory is illustrated by various studies analyzing the performance implications of DOI in a regional context (Oh and Rugman, 2006; Collinson and Rugman, 2008; Oh et al., 2014). Rugman and Oh (2010) highlight the differences to prior research by showing that traditional measurements of DOI and performance are insufficiently applicable to contemporary IB and emphasize the incorporation of a regional level of analysis. According to Rugman and Verbeke (2004), the same applies to the major forces of internationalization discussed above, which need to be specified with the regionalization context.
The character of the FSA/CSA conception as an operationalization of the source of competitive advantages in foreign markets allows its application to explaining regionalization patterns (Rugman and Verbeke, 2004). In this regard, certain CSAs could also be summarized as region-specific advantages (RSAs), such as for example promotional programs by the EU (Lee and Rugman, 2012). Similarly, location-bound FSAs can be bound to a particular region, such as human resource management practices of Japanese firms (Collinson and Rugman, 2008). This regional scope of an FSA arises because adaptation can be relatively easy due to common characteristics among countries within that region, transferring those FSAs to different regions requires significantly more adaptation for profitable exploitation (Collinson and Rugman, 2008). In this sense, Rugman and Verbeke (2004) have proposed that the regional orientation is partly due to the limited transferability of FSAs outside the home region, which is empirically supported by Rugman and Sukpanich’s (2006) findings that most of their tested FSAs can be more profitably exploited within the home region.
The transferability of FSAs across regions is strongly related to the character of LoF within the regionalization theory (Rugman and Verbeke, 2007). The authors identify that
18 there is not only intra-regional, but also inter-regional LoF that affects the transferability of FSAs and thereby the internationalization in different ways. This phenomenon is referred to as the non-linearity of LoF, unlike prior research, no linear relationship between LoF and compounded distance is expected (Rugman and Verbeke, 2007). An alternative operationalization of the LoF, which essentially represents the same idea, but highlights the increasing importance of LoF within the context of research in regionalization theory, derives from Asmussen and Goerzen (2013) who distinguish between cultural, inter-institutional, and inter-regional LoF.
Intra-regional LoF encompasses the costs that result from spatial distance as well as from structural/relational and institutional differences among countries in a region (Qian et al., 2013). Inter-regional LoF explains regional differences and includes additional costs through complexity and diversities of operations, the existence of trade unions, and the creation of fit with the environment in matters of institutions and production factors (Qian et al., 2013). Qian et al. (2013) emphasize that both dimensions of LoF coexist and pose an influence jointly or separately on the internationalizing MNE.
Qian et al. (2013) further specify that intra-regional LoF depends on the differences between two countries and a firm’s experience and capabilities, whereas inter-regional LoF is dependent on regional characteristics. Consequently, a firm can make use of its region-specific know-how among all countries within a region, but a firm’s experience and the relative distance between countries are moderating the additional costs resulting from LoF. This illustrates that home region oriented companies are primarily faced with intra-regional LoF, which allows a higher degree of transferability of FSAs. In contrast, by entering a new region a company has to deal with inter- and intra-regional LoF that can have a joint effect on corporate performance (Qian et al., 2013).
Important in the context of this analysis is that international experience is predominantly moderating the intra-regional LoF. Even though there’s space to decrease inter-regional LoF through learning and experience it is limited through organizational complexity (Qian et al., 2013). The important role of path-dependent international experience is also highlighted as the company’s expansion patterns identified in regionalization theory can be explained by the Uppsala model by Johanson and Vahlne (1977) (Oh et al., 2014). Further support derives from Mohr et al. (2013) who stress the importance of international experience as a moderating factor between internationalization and performance. In particular, against their expectations, the authors find evidence suggesting that international experience is not region-bound, therefore, concluding that the moderating effect of experience is not
19 limited to intra-regional LoF (Mohr et al., 2013). Conversely, this supports the conclusion that international experience is also reducing inter-regional LoF, which is up to a certain degree in line with Qian et al. (2013). Relating this evidence to the findings of Luo and Peng (1999) and Mohr et al. (2013) allows the inference that the longer a company is operating abroad, the more international experience the company has accumulated, and therefore the lower the intra- and inter-regional LoF. Consequently, international experience is expected to be beneficial for corporate performance within but also outside of the home region, but its moderating impact is expected to be stronger in an intra-regional setting due to the better applicability as suggested by Qian et al., (2013).
Taking the similarities with the DOI-performance debate into account, Oh and Contractor (2014) have refined the horizontal S-curve model by Contractor et al. (2003) under consideration of regionalization theory. In this context, the first two stages encompass the intra-regional expansion while the third stage represents the inter-regional expansion (Oh and Contractor, 2014). There are two facts that call the model strongly into question. First, it is purely based on a sample of US companies. And second, as shown by Hejazi (2007) the NAFTA region is very sensitive to the home country effect. Therefore, the expansion in focus illustrates higher degrees of inter-regional LoF due to a lack of international experience, because experience derives primarily from the US operations. In contrast, European companies have to deal with a greater diversity and are able to gain more international experience (Edfelt, 2009). Based on this reasoning, the analysis here will stick to the initial version of the model by Contractor et al. (2003) in order to show the expected performance.
In a different approach to capture the regional characteristics of the DOI-performance relationship in the horizontal S-Curve model by Contractor et al. (2003), Oh et al. (2014) identify that the model is only applicable to intra-regional expansion, inter-regional expansion, instead, is found to be captured by an S-curve fit. The slightly different curve is caused by the relative degree of LoF and network complexity. The limited complexity of intra-regional networks allows companies to increase performance through intra-regional diversification, without being strongly confronted with overwhelming coordination costs (Sethi and Judge, 2009). Moreover, those firms are only faced with the intra-regional LoF. However, there are incentives for inter-regional expansion due to limited growth opportunities in the home region, but companies are then faced with intra- and inter-regional LoF and the effects of complexity become more significant (Oh et al., 2014). The baseline is that the way of reasoning stays the same; only the strength of LoF and network complexity varies. Therefore, it can generally be stated that internationalization up to a certain maximum, which
20 reflects the overwhelming coordination costs, leads to higher performance, but is moderated first by intra-regional and later by intra-and inter-regional LoF (Oh et al., 2014).
Nevertheless, in IB research there are various analyses providing evidence that expansions within the home region are leading to higher performances than those across or within other non-home regions (Collinson and Rugman, 2008, Kudina, 2012). Others, such as Qian et al. (2010) and Oh et al. (2014), emphasize the benefits of inter-regional expansions and are more in favor of an inverted U-shaped relationship, which promotes the idea of overwhelming costs due to complexity at a certain internationalization level. Similar to the DOI-performance debate, it is still highly debatable how regional orientation and performance are related to each other and how this is moderated by a company’s international experience.
In order to contribute to IB research, this A&D industry-specific study will investigate the DOI-performance debate in a regional context. The analysis will shed further light on the debate whether a global or a regional strategy is beneficial for corporate performance. Furthermore, the emergence and acceleration of research in regionalization theory in recent years seem to indicate a trend from a global towards a regional level of analysis (Sukpanich and Rugman, 2007, Kudina, 2012; Rugman and Oh, 2010; Oh and Contractor, 2014). In order to account for the dynamics in IB research, this study will also investigate whether there is a shift in terms of the relative importance of global or regional orientation and their impact on performance. It is generally expected that the regional orientation is becoming stronger in driving performance due to the increasing degree of regional integration in the extended Triad regions and the consequent impact on intra- and inter-regional LoF the companies are facing (Lucas, 2007; Qian et al., 2013).
A second dimension within this study is the incorporation of international experience as a moderating factor of the relationship between the degree of global orientation and performance because of its importance in IB research. Thus, this study will elaborate on the controversy that arises when international experience is transferred in the context of regionalization theory. As experience-based FSAs are expected to increase with geographic dispersion, it remains unclear to which degree a regional orientation is appropriate to capture this effect.
A third issue this study aims at is the longitudinal analysis of regional orientations as proposed by Oh and Rugman (2014). Whereas Oh and Rugman (2014) indicate that there are no substantial changes in regional orientations over time, it is expected that the A&D companies in focus increase the degree of global orientation due to the substantial changes in demand conditions.
21
3. Conceptual framework
Having set the theoretical foundation, this section is concerned with developing the hypotheses arising when the DOI-performance debate is transferred in a regional context and embedding them in a conceptual framework. First, it will be investigated whether the degree of global orientation drives performance. Second, international experience is introduced as a moderating factor. Third, it is investigated whether there is a change over time in the relationship in focus by repeating the analysis within the two sub-samples. In figure 2, the conceptual framework is visualized in order to clarify the dimensions of analysis in focus.
Figure 2: Conceptual model
The investigation of the relationship between DOI and performance in a regional context has shown mixed results (Collinson and Rugman, 2008; Qian et al., 2010; Kudina, 2012). The evidence by Oh et al. (2014) for the discussed broader consistency between intra- and inter-regional diversification and the horizontal S-curve model infers that the basic relationship is positive. This inference assumes that companies are aware of the additional costs a higher degree of global orientation implies and that the decision is only made when superior performance in the long-run is expected. The limitation of the relationship on a very high degree of global orientation is neglected in the first place as the A&D industry is less internationalized and in accordance to Rugman and Verbeke (2004) there are just a few companies that can be considered as global and might encounter those problems. Consequently, it can be stated that the baseline relationship between the degree of global orientation and performance is expected to be positive. As the analysis transfers the DOI-performance debate into a regional context, the various studies having shown this positive relationship in the DOI-performance debate, such as Delois and Beamish (2001), Goerzen and Beamish (2003), and recently Yang et al. (2013) support the claim that:
22 In a second step, the international experience a company has gained will be incorporated in the sense of the process model by Johanson and Vahlne (1977). The overall applicability of the Uppsala model within the context of regionalization theory as well as the role of organizational learning allows assuming that a firm’s accumulated knowledge impacts the relationship between the degree of global orientation and performance. Thereby, the accumulation of knowledge is generating knowledge-based FSAs on the one hand and is decreasing the degree of LoF through organizational learning on the other hand (Kudina, 2012). In line with Luo and Peng (1999) and Ruigrok and Wagner (2003), it is assumed that the higher the number of years a company is operating abroad, the more international experience it has gained. Because companies with a small geographic scope can make less use of its international experience, it can be argued that the moderating effect of international experience is expected to be beneficial for highly geographically diversified companies (Mohr et al., 2013). Those companies are in the position to exploit the benefits of being global, such as scale and scope economies, while decreasing the additional costs through international experience by lowering the degree of LoF and creating intangible FSAs (Luo and Peng, 1999; Kudina, 2012). Consequently, it can be stated that:
Hypothesis 2: Higher degrees of international experience will positively moderate the relationship between the degree of global orientation and performance outcome (H1).
Having established the basic relationships, the focus is now turned towards the relative changes in the nature of the relationship over time. There are only a few studies examining the geographic orientation with consideration of regionalization theory over time, such as Osegowitch and Sammartino (2008), Rugman and Oh (2013), and Oh and Rugman (2014). Whereas Osegowitch and Sammartino (2008) find that large companies are becoming more global, Oh and Rugman (2014) refuse this by showing that only a small proportion of firms have changed their geographic scope over time. As the sales-related home country orientation of A&D companies in focus has decreased by an average of 9% and the home region orientation by an average of 8% over the years 1999-2013, it would be reasonable to assume that it has become more beneficial for corporate performance to increase the degree of global orientation. Nevertheless, the various recent studies having shown the beneficial character of regional strategies, especially in relation to the home region, such as Collinson and Rugman, (2008), Kudina, (2012), Oh and Rugman (2012), and Oh and Rugman (2014), support the idea of an increasing importance of a regional geographic scope. Consequently, even though it is assumed that the general relationship between degree of global orientation and performance is
23 positive, it is expected that the relative explanatory power of the regional orientation for corporate performance increases in place of global orientation over time:
Hypothesis 3a: The positive impact of the degree of global orientation on corporate performance will decline over time.
As it is expected that the relative importance of higher degrees of global orientation in driving performance decreases, companies either are less able to make use of their position in the form of achieving scale and/or scope economies or face overwhelming costs due to coordination and complexity at the more extreme levels of the degree of global orientation as suggested by the horizontal S-curve model (Contractor et al., 2003; Lu and Beamish, 2004). In both cases, the experience of being international as intangible FSA is crucial in order to maintain or improve corporate performance at higher degrees of global orientation. This can for example be achieved by either use organizational learning in the form of best-practice to reduce coordination costs or apply the accumulated knowledge to improve scale economies, a concept which is essentially related to learning curve effects (Barney, 2002; Kudina, 2012). Thus, the relative importance of the moderating effect of international experience is expected to increase over time. The continuous accumulation of knowledge is expected to compensate for the relatively lower importance of the degree of global orientation in driving performance:
Hypothesis 3b: The degree to which international experience positively moderates the relationship between the degree of global orientation and performance increases over time.
After having set the theoretical foundation for the analysis, the next section is dealing with the methodological formulation of the research design in order to capture the phenomenon in focus in the contemporary business environment.
24
4. Methodology
4.1. Sample and data collection
The A&D industry with its recent development serves as a suitable environment in order to test the hypotheses formulated above. A&D companies generally have very clear geographic orientations due to long-term governmental contracts. However, a recent trend in changing demand conditions has given rise to increasing internationalization (Deloitte, 2014). Following the reasoning above would suggest that internationalization is a way to improve performance. The degree to which this applies in a regional context and how international experience moderates this relationship remains unclear. This practical framework for providing further insights in the regional DOI-performance research is operationalized in the following part before the statistical test adopted is explained.
The sample consists of the 20 largest companies operating in the aerospace and defense (A&D) industry in 2013. Size is measured by A&D-related sales, which were assessed by Deloitte (2014). In a first attempt, the sample was created by following Rugman and Verbeke (2004) and using the A&D companies listed in the Fortune 500. Initial investigations have shown that the majority of companies within the sample are strongly rooted in the US economy. In order to reduce the home country bias identified by Hejazi (2007), the sample was extended to 25 companies as this scope allows an increasing breadth and depth of geographic orientations. Later, five companies were dropped out due to a lack of data availability.
As the development of geographic orientation over time is investigated, it is crucial to add a time dimension to the cross-sectional dimension. Therefore, the 20 companies are observed over 15 years, namely 1999-2013, in order to capture the rise of defense budgets up to recent events that tremendously changed the demand conditions, such as the “US Budget Control Act” (Center of Budget and Policy Priorities, 2012; Mc Kinsey and Company, 2014). This combination of cross-section and time-series leads to a panel data research design. As panel data combines the two dimensions of analysis, it allows investigating the dynamics of change while accounting for the individual heterogeneity within the sample (Baltagi, 2008). Besides this, panel data provides more variability and degrees of freedom than cross-section or time-series do, thus, leading to more efficient estimators (Baltagi, 2008). A detailed discussion of the advantages and disadvantages of using a panel data research design can be found in Hsiao (1986).
The analysis is repeated by using two unequal sub-samples, the first ranging from 1999-2005 and the second ranging from 2006-2013. The decision to have a larger second
sub-25 sample is based on the reasoning to reduce the impact of the financial crisis in 2007-2008. The comparison of the two sub-samples allows for another dimension of analysis that is concerned with the relative change in the effects of the independent variables. Thus, it can be investigated whether there is a trend regarding the explanatory power of global or regional orientation for corporate performance.
The data needed to explain the regional orientation of the focal A&D companies are the distributions of sales on a regional level. In addition, to account for the home country effect, sales data for the home country was also collected. This methodological approach and the definition of the extended Triad regions derive from Rugman and Verbeke (2004). The most consistent way of gathering the data is using the annual reports of the companies as secondary data sources, which is in line with longitudinal data collections of regional orientations, such as Rugman and Collinson (2004) or Oh and Rugman (2014). On a product level, a diversification ratio was incorporated in order to specify a company’s focus on aerospace or defense product lines. Because the content and structure of annual reports differ over time, the availability of data varies by year and firm. As reported by Oh and Rugman (2014), recent reports tend to be more informative regarding sales and asset distributions and this is reflected in the data collection and noted where appropriate for understanding findings and their discussion.
4.2. Classification of regional orientations
This analysis is strongly related to the process model by Johanson and Vahlne (1977), therefore, the classification of regional orientations by Rugman and Verbeke (2004) is less suitable for two reasons. First, the classification cannot be related to a hierarchy of increasing degree of global orientation and thereby impact of international experience. For example, host-region orientation can rely solely on the operations within a single country in a foreign region indicating a lower degree of international experience than operating in the entire home region. Second, the classification fails to consider the importance of the home country, which is very significant for companies operating in the NAFTA region as shown by Hejazi (2007). The approach presented here will relate the measurements applied by Rugman and Verbeke (2004) to the classification by Aggarwal et al. (2011). Consequently, regional orientations can be characterized as domestic, home-regional, trans-regional, and global. In more recent publications, Oh and Rugman (2014) have also applied a domestic dimension of regional orientation, which supports the reasoning above. Table 2 provides an overview about the schematic definition of regional orientations used within this analysis. Regarding the order of
26 the regions it is assumed that the expansion path of A&D companies depends on market attractiveness measured by defense budgets. Consequently, the following prioritization of regions is assumed: NAFTA, EU, Asia Pacific, and Rest of the World (ROW).
Orientation Home country Home region Host region 1 Host region 2/ROW1
Domestic ≥ 90% - - -
Home region < 90% ≥ 50% - -
Trans-regional - 20% ≤ x < 50% ≥ 20% -
Global - 20% ≤ x < 50% ≥ 20% ≥ 20%
Table 2: Classification of regional orientations
Source: Adapted from Rugman and Verbeke (2004) and Aggarwal et al. (2011)
The definition of sales proportions is essentially related to Rugman and Verbeke’s (2004) approach. The threshold of 10% to account for the strong home country orientation within the A&D industry is increasingly used in IB research (Qian et al., 2010; Oh et al., 2014). Furthermore, as the host region orientation is neglected in order to create an increasing degree of global orientation, as discussed above, the 50% limitation for host regions is removed. Different to Rugman and Verbeke (2004), the ROW is incorporated as an alternative second host region1. Even though, there are no regional characteristics in the ROW, it is justified by the importance of international experience in this analysis as well as the increasing share of the BRIC-countries in the global economy (Douglas and Craig, 2011). Overall, this classification assumes a hierarchical construct of increasing DOI and the associated impact through international experience embedded within a regional context.
4.3. Variables and measures
In order to establish a model, it is first necessary to discuss the variables necessary to include for capturing the phenomenon in focus. An overview about the different variables used and their operationalization is presented in table 3after the discussion of the applied metrics.
Dependent variables
The performance of a firm is measured by using three different indicators having found to be frequently applied in IB research by Richard et al. (2009). First, Profit Margin (PM), also referred to as Return on Sales, is very common in regionalization literature in order to account for the market-based profitability (Barney, 2002; Rugman and Oh, 2010; Almodóvar and Rugman, 2013). In order to measure the operational profitability while avoiding the bias
27 through a firm’s financial leverage, which is naturally neglected by the Return on Equity, the
Return on Invested Capital (ROIC) is used as a second performance measurement (Barney, 2002; Chang and Rhee, 2011). Moreover, as the capital structure is neglected, the ROIC allows an underlying measurement of the risk that is associated with the company’s operations (Brigham and Houston, 2011). Third, Return on Assets (ROA) is an equally popular rather accounting related indicator that controls for the effectiveness with which a company is using its assets (Lu and Beamish, 2004; Kudina, 2012; Qian et al., 2013). All of these performance measurements are lagged by one year (t+1) as it is a common method in IB research in order to allow for causality (Lu and Beamish, 2004). This combination of lagged performance measurements makes it possible to capture the effect of geographic expansion.
Independent variables
The independent variables used in this analysis to describe a company’s geographic dispersion are widely used in regionalization research (Vermeulen and Barkema, 2002; Qian et al., 2013; Rugman and Oh, 2013; Oh and Rugman, 2014). In this context, two variables are introduced to account for the scale and one for the scope of geographic dispersion. The scale measurements are foreign-to-total sales (F/T sales) and home-region-to-total sales (HR/T
sales), which capture the percentages of sales inside and outside of the home country and the home region (Rugman and Li, 2007; Kudina, 2012; Oh and Rugman, 2014). The scope measurement is the entropy index for inter-regional dispersion (Entropy Inter-region) by Hitt et al. (2006). According to Aggarwal et al. (2011) and Oh et al. (2014), the entropy measure allows the consideration of breadth and depth of regional diversification, as it accounts for the number of regions and their relative importance. In this context, the higher the number of the inter-regional entropy index, the more equally geographically diversified are the companies among regions (Qian et al., 2013; Oh et al., 2014). In turn, a low entropy index would be associated with a company that is highly home region oriented.
Moderator variable
The accumulated international experience a company has gained is commonly described by the number of years a company is operating abroad (Luo and Peng, 1999; Mohr et al., 2013). Due to a lack of data availability, which is a frequent problem in panel data research designs as described by Oh and Rugman (2014), the reporting of foreign income taxes of a company was used as a proxy. It is not a widely adopted measure, but it clearly signals for how long
28 firms have paid taxes; therefore, it is a fairly accurate measure of how long they have operated in foreign countries. The data were gathered by using COMPUSTAT.
Control variables
There is a broad bandwidth other than the independent and moderating variables mentioned above, that can cause higher performance (Lu and Beamish, 2004). Therefore, including a variety of control variables is necessary in order to account for their impact and establish a clear focus on the relationship of interest. In this context, the control variables are firm size, age, leverage, R&D intensity, defense ratio of sales, and foreign-to-total defense budget (Lu and Beamish, 2004; Elsayed and Paton, 2005; Goerzen and Asmussen, 2007).
According to Elsayed and Paton (2005) the size of a firm might affect the performance due to the possibility of scale economies, whereas the age of a firm accounts for maturation effects as it is assumed that companies become more efficient and effective over time (Delaney and Huselid, 1996). Another performance affecting factor is the leverage of a
company, which is important to take into account as it can be used to change capital costs, thereby, distorting the actual performance of a firm (Elsayed and Paton, 2005). As A&D companies are strongly related to technological progress, a company’s R&D intensity can also be considered as a driver of performance. In this context, R&D intensity is used as a proxy for a company’s intangible assets. According to Almodóvar and Rugman (2013), theory suggests that intangible assets lead to competitive advantages and ultimately to superior performance.
Focusing more on industry-specificity, two factors that might also have a substantial impact on A&D corporate performance are framed as the defense sales ratio and the
foreign-to-total defense budget. The former is a measurement for the production diversification. As the shrinking defense budgets are primarily affecting the defense product lines, an A&D company might increase the proportion of aerospace products in order to increase performance. The latter is also a fundamental driver of performance in the A&D industry, as it is expected that a strongly home country or region oriented company, as it is regularly the case in the A&D industry, benefits significantly from an increase of the defense budget of the particular country.
This combination of control variables acknowledges the commonly used variables in the regionalization theoretical stream in IB research and is extended by incorporating the industry-specific factors that might pose an impact on performance (Lu and Beamish, 2004; Elsayed and Paton, 2005).
29
Variable Operationalization Sources
Dependent variables Profit Margin (PM)
Net income before interest and taxes / total sales Barney, 2002; Richard et al., 2009; Rugman & Oh,
2010
Return on Invested Capital (ROIC)
[Net operating profits - adjusted taxed] / invested capital Barney, 2002; Chang & Rhee, 2011
Return on Assets (ROA)
Net earnings / average total assets Lu & Beamish, 2004; Kudina, 2012; Qian et al., 2013
Independent variables Foreign-to-total sales (F/T sales)
Foreign sales / total sales Rugman and Li, 2007; Kudina, 2012; Oh & Rugman,
2014
Home region-to-total sales (HR/T sales)
Sales within the home region / total sales Rugman and Li, 2007; Kudina, 2012; Oh & Rugman,
2014
Entropy Index Inter-region (Entropy Inter-region)
∑ "#ln ()(
*)
,
#-( with N accounting for the number of regions and Ri for the
proportion of sales within each region
Hitt et al., 2006; Aggarwal et al., 2011; Qian et al., 2013; Oh et al., 2014
Moderator variable
International Experience Number of years a company is paying foreign income taxes Luo and Peng, 1999; Mohr et al., 2013
Control variables
Firm size Natural logarithm of total assets Elsayed & Paton, 2005
Firm age Natural logarithm of years a company is operating since its foundation Delaney & Huselid, 1996
Firm leverage Total assets / total debt Elsayed & Paton, 2005
R&D intensity R&D expenditures / total sales Lu & Beamish, 2004; Almodóvar & Rugman, 2013
Defense ratio sales Defense related sales / total sales -
Foreign-to-total defense budget Totalized foreign defense budgets / globally totalized defense budgets -
