The global logistics sector : integrating regional or global distribution networks?

(1)

1

The global logistics sector: integrating

regional or global distribution networks?

Master Thesis

Author: Sjirk Prins

Supervisor: Dr. Johan Lindeque

Second Reader: Dr. Niccolò Pisani

Date: 31 January 2014

University of Amsterdam: Amsterdam Business School

MSc Business Studies – International Management

Word count: 21.190 words (excluding figures and tables)

(2)

2

Abstract

The traditional idea that the largest MNEs in the world are truly global has been challenged by scholars (Rugman, 2005; Rugman and Verbeke, 2004) as it is argued that even the largest MNEs are not globally orientated but rather regional. This study examines the regional orientation of MNEs by conducting a qualitative multiple case study on the regional or global nature of the world’s largest logistic service providers (LSPs). These firms are often recognized for their role in global trade as these firms ship products between regions across the world. This study first seeks to determine the global or regional nature of these firms and then aims for an explanation. A longitudinal element is necessary to be able to explain why LSPs are regional or global orientated and the data analysis therefore covers a seven-year period, ranging from 2006 – 2012. The qualitative data in this study exists of articles from selected newspapers covering the seven-year and the Financial Times is used as the key source, supplemented by other newspapers where necessary. The data analysis resulted in support for the proposition that formal institutional barriers pose strong constraints on LSPs’ international strategies and weak support for the propositions with regard to the transferability of brand FSAs and service quality FSAs. This does not indicate that the working propositions for which evidence is weak are incorrect, as there is some evidence for their relevance. Further research is however necessary to fully clarify the influence of these FSAs on the internationalization of LSPs. Key words: regionalization, international strategy, logistics, postal services, internationalization, institutions

(3)

3

Acknowledgement

I sincerely thank everybody who supported me during the process of writing this thesis. In particular I would like to thank Johan Lindque for supervising this thesis, for providing me with valuable insights and for offering guidance were necessary. I am very happy with the way this thesis developed and the process would have been much tougher without the constructive comments I received and the inspiring discussions we had.

I would furthermore like to thank my fellow students with whom I studied the topic of regionalization. Though we all went our own way pretty quick after the initial meetings, the discussions I had with some of you were often very clarifying and this thesis benefited a lot from it.

Last but not least, I would like to thank my family and friends for all the support I received. Writing this thesis was a long and sometimes difficult process and I’m thankful you listened to me when it was needed.

(4)

4

Figures and tables

Tables

Table 1: Classifications of strategic orientations and distribution of 14 the Global Fortune 500 MNEs in 2001 and 2012

Table 2: Relationship between political economy and the SOEs degree of 22 internationalization

Table 3: Overview of selected LSPs in this study 31

Table 4: Sources and number of newspaper articles per company 33 Table 5: Overview of primary codes including description and the WP to 34

which they relate

Table 6: Within-case Analysis FedEx, 2006-2012 36

Table 7: Within-case Analysis United Parcel Service (UPS), 2006-2012 39 Table 8: Within-case Analysis Deutsche Post DHL, 2006-2012 42

Table 9: Within-case Analysis State-owned LSPs, 2006-2012 47

Table 10: Cross-case Analysis 51

Table 11: Results for the working propositions in this study 56

Figures

Figure 1: Development of the private LSPs’ host region presence between 57 2000 and 2012

(7)

7

Introduction

The world of international business and trade changed significantly in the past decades as a result of what is conceptualized as globalisation (Garrett, 2000). Globalization is a widely acknowledged phenomenon (Garreth, 2000) and although the existence of the phenomenon is clear there remains much discussion around its definition, as a result of which it remains “a poorly understood phenomenon” (Rugman and Verbeke, 2004, p. 3). The definition of globalisation which will be used in this paper is the one which was formulated by Rugman and Verbeke (2004). These authors describe globalisation as the process of “increased economic interdependence among nations” (2004, p. 3). Key drivers behind this process are multinational enterprises (MNEs) and it is in particular the world’s 500 largest MNEs which are responsible for a large part of world trade (Rugman and Verbeke, 2004). MNEs create the supposed globalisation through their efforts to internationalize their businesses. Because MNEs internationalize into different parts of the world it is often assumed that these firms are global firms, doing business all around the world and pursuing global strategies. This global nature is often emphasized by the MNEs themselves. A nice illustration of this is Deutsche Post DHL which states in its 2011 annual report that “about 470.000 employees in more than 220 countries and territories form a global network focused on service, quality and sustainability” (Deutsche Post DHL, 2011, p. I) However, while the scientific literature often speaks about these global firms as well (see for example Yip, 2002) there is empirical evidence that many of the world’s largest firms’ activities are regionally based rather than globally (see for example Rugman and Verbeke, 2004; Rugman and Girod, 2003; Fillipaios and Rama, 2008; Oh, 2009).

This thesis addresses this disagreement with regard to the global or regional character of MNEs by examining the internationalization of the largest players in a specific international market. It will do so by executing an in-depth multiple case study in which the degree of internationalisation of a number of MNEs is analysed as well as the factors influencing their internationalization. Both firm specific factors as well as institutional aspects which might have an influence on the regional or global nature of the MNEs in the focal sector are analysed. The focal sector in this study is the international postal services and logistics sector, which includes the private MNEs FedEx, UPS and Deutsche Post DHL and the state-owned firms China Post, Japan Post Holdings, La Poste, Poste Italiane and US Postal,. These MNEs provide a wide range of services including postal services, express services and parcel services and these services are provided to both individual consumers and to other businesses (B2B- relationships). The performance of these services rapidly changed as a result of technological progress in the early 1990s of which especially the rise of the internet and improved transportation methods were very important (Garrett, 2000; Berglund et. al., 1999). MNEs had to respond to these changes and they did so by engaging in networks which added the flexibility needed for the performance of its services. This flexibility was needed because customers demanded a global reach and no single company could offer this, unless they internationalized and engaged in networks with

(8)

8 other companies (Castells, 2010). This engagement in networks with other firms required the MNE to revise its internal structure as well: “to be able to internalize the benefits of network flexibility the corporation had to become a network itself and dynamize each element of its internal structure: this is in essence the meaning and the purpose of the ‘horizontal corporation’ model” (Castells, 2010, p. 176). Postal and logistic firms, thus, started to form inter-organizational relationships and to engage in networks with other firms in order to create the global reach which their customers were asking for. These customers, however, were customers in the home market. When MNEs acted upon these requests they were guided by a client-following strategy (Majkgard and Deo Sharma, 1998; Bowen and Leinbach, 2004). In the later stages of the market development and foreign market entry client-following strategies were complemented with or adapted into market seeking strategies as MNEs started to seek business opportunities in the markets they were so far only serving for home market clients (Morgan et. al., 2006; Bowen and Leinbach, 2004). Market seeking strategies became more and more dominant as important postal and logistics markets in North America and Europe matured and the logistic service providers (LSPs) from these regions sought market opportunities in host regions (Rushton and Walker, 2007). These motivations in the LSP´s international strategies mainly apply to private LSPs as the state-owned LSPs are mostly home country focused. It is furthermore expected that the drivers behind their internationalization differ from the drivers of the private LSPs internationalization.

There are several reasons for why studying the international postal and logistics sector in the context of regionalisation makes sense. First, it is a sector which is strongly associated with globalisation. Just as airlines, shipping companies and other forms of trade logistics firms transport products across the globe. An increased interconnectedness between different parts of the world includes the exchange of products. The international logistics sector, thus, offers services for transportation across the world and one would assume at first sight a high degree of international market integration and globalised firms: “By designing, managing, and monitoring the world’s flows of goods and information, logistics is justifiably regarded as the backbone of the globalisation phenomenon” (Straube et.al., 2008, p. 1). This paper examines the truth of this statement by investigating the internationalisation process of the key players in the market. A second interesting aspect about the logistics sector is that it is a service sector and not a manufacturing sector. It is both in the IB literature in general and in the regionalization literature in specific that service MNEs are understudied as most scholars pay attention to manufacturing firms (Hitt et. al., 2006; Lommelen et. al., 2002). Studying a service sector may therefore add new insights in how global service MNEs are and why this is the case. An interesting additional aspect of studying the international logistics sector is that it hasn’t been studied in this context before. A final interesting aspect in studying this sector is that it includes both state-owned firms and private firms. This mixture may add additional value to the study as it is interesting if and how patterns of international activities differ among these different

(9)

9 types of firms. The unique character of this sector is expected to add valuable new information to the regionalisation debate.

The analysis is longitudinal allowing this study to examine changes over time and to give insights in which regions these MNEs invest. The time period in which the degree of internationalization of the MNEs in this sector will be studied is 2006 – 2012. By examining the degree and nature of the internationalisation of these MNEs this thesis tries to add more clarity to the scientific literature with regard to the global or regional nature of the world’s largest MNEs. Furthermore, it also tries to create new insights for managers. Daily managerial activities include strategy formulation and execution and doing this on a regional or a global scale can create significant differences in performance.

How have institutional factors and firm specific advantages shaped the global presence of the major logistics companies in the Fortune Global 500?

In order to formulate an answer to the research question, this paper starts by giving an overview of the state of the art in the literature on regionalisation and the internationalization of business activities. This overview includes both conceptual discussions and empirical discussions and it will conclude with the formulation of a number of working propositions which reflect the theoretical dimensions relevant to the international postal and logistics sector. After the theoretical foundations of this research have been developed the methodology in this study will be discussed and this chapter includes a discussion of the research philosophy as well as a discussion on case selection and data analysis. The consequent within-case analysis and the cross-case analysis provide the evidence for the working propositions in this thesis. The analysis shows strong support for the influence of formal institutional barriers in this sector. These barriers exist in varying forms in almost all regions across the world and these barriers indicate the importance many governments attach to this sector in particular in emerging markets where the postal and logistics sector is often underdeveloped. This aspect is further discussed in the discussion chapter which will furthermore contain a paragraph on the development of this industry. The largest (private) firms in this sector appear to slowly but continuously increase their host region presence which has no consequences for their regional orientation yet as all MNEs in this sector are home-region orientated. However, when these trends will continue this sector contains some good candidates for future bi-regional focused MNEs.

(10)

10

Literature Review

The literature review will start with an overview of the literature on the core topic of this thesis, which is the regionalization of business activities. Recent studies have shown that a large majority of the world’s largest MNEs operate mainly within the home region (Rugman and Verbeke, 2004). The first section of the first part of the literature review discusses both conceptual and empirical studies. The second section will discuss the phenomenon of regionalization through the lens of the most important approaches in IB, which are the RBV and the IBV. The third section discusses the key themes in regionalization. The second part of this literature review discusses the nature of LSPs and this chapter finishes with the formulation of a number of working propositions.

Regional nature of international business activities

Globalization, as already defined in the introduction (Rugman and Verbeke, 2004), thus refers to the “international integration of markets in goods, services, and capital” (Garrett, 2000, p. 942). The supposed causes of the increasing international market integration are technological innovations, increased international economic activity and liberalism as the dominant political ideology in many countries’ foreign policy (Garrett, 2000). While the debate focuses in particular on the existence of the phenomenon, the causes and the consequences of the phenomenon and the newness of the phenomenon, less attention has been paid to the scale and scope of the assumed international market integration. It can well be argued that this is not the case and that countries have a preference for cross-border integration with particular countries, especially countries which are geographically close. This is confirmed by Franklin and Rose (2002), who furthermore show that the trade between countries which belong to the same trading block is three times larger than trade with countries outside the trading block. These authors thereby suggest that “cross-border integration may be stronger within such regions than between them” (Asmussen, 2008, p. 1192). Additionally, Ghemawat (2003) argues that the levels of international market integration differ across the world, while in some parts of the world the levels of international market integration are very high, the levels are very low in other parts. Ghemawat (2003) therefore concludes that what we witness today is not globalization but rather globalization: “This condition of incomplete cross-border integration, referred to here as semi-globalization, is more complex than the extremes of total insulation and total integration because it involves situations in which the barriers to market integration at borders are high, but not high enough to insulate countries completely from each other” (Ghemawat, 2003, p. 139).

Where globalization on a macro-level does not seem to exist, scholars recently started to doubt the global nature of the world’s largest MNEs as well (Rugman and Verbeke, 2004; Rugman, 2005). As already argued in the introduction, many MNEs and scholars emphasize the global nature of these MNEs. A closer analysis of the operations of the world’s largest MNEs, however, reveals that the

(11)

11 supposed global nature can only be observed at a very small number of MNEs (Rugman and Verbeke, 2004). Most MNEs, in contrast, operate primarily within the home region (see table 1).

The idea that firms internationalize into their home region rather than around the globe was first conceptualised by Ohmae in 1985 (in Rugman and Verbeke, 2004, p. 4): “setting up an extensive distribution capability for the product ex ante, throughout the triad, may entail high, irreversible, fixed costs, and therefore high risks, if the superproduct somehow does not deliver on its sales expectations. Conversely, if the superproduct is first marketed at home, rival companies in other legs of the triad are expected to rapidly create an equivalent product, capture their home triad region market, and dominate distribution in that market”. The triad legs in Ohmae’s words are the US, the EU and Japan. Ohmae’s work is used as a foundation of Rugman and Verbeke’s (2004) study. These authors extended the triad regions to the NAFTA region, the European Union and Asia. Based on a sample of the 500 largest companies in the world (based on sales) these authors found that 365 of the 500 largest companies in 2001 had a presence outside their home region.

Rugman and Verbeke (2004) come up with five explanations for why MNEs logically internationalize into their home region and not into other parts of the triad. First, the firm’s products or services are either not equally accessible or equally attractive around the world. In other words, the firm does not succeed well enough to adapt its products to local demand. Second, a firm’s global presence can be limited due to the limited transferability of its FSAs. If a firm’s FSAs are primarily location bound (LB) the firm will not succeed to transfer these FSAs to the other parts of the triad. Additionally, a third reason for why MNEs have difficulties to build up a global presence is that the transferred non-location bound (NLB) FSAs need to be combined with LB FSAs and an inability to access and develop these LB FSAs results in a limited performance in terms of sales. Fourth, a lack of globalness of a firm’s business might appear a result of limited local adaption: “if the MNE's market position is very different in the various regions of the world this indicates the need for very different competitive strategies” (Rugman and Verbeke, 2004, p. 6). Finally, a firm’s relative lack of success in other triad regions might be explained by different environmental circumstances. As a result of these dynamics in the functioning of the MNE, Rugman and Verbeke (2004) found very few companies to be truly global. Most MNEs fell into one of the other categories which Rugman and Verbeke (2004) developed (table 1).

An important concept in both the literature on international strategy and the literature on regionalization is the liability of foreignness (LOF) (Zaheer, 1995). The LOF is defined as “the costs of doing business abroad that result in a competitive disadvantage for an MNE subunit” (Zaheer, 1995, p. 342). These costs have different sources: 1) they can result from the spatial distance to the host country, 2) they can result from unfamiliarity with the host-country’s environment, 3) they can result from economic nationalism and/or lack of legitimacy of foreign firms in the host country and 4) they

(12)

12 can result from the home country environment. The LOF will result in an unequal competition between domestic and foreign firms as the profitability of the foreign firms will be lower because of the LOF.

The LOF was traditionally a concept which was taken into consideration primarily on a country-level. However, since Rugman and Verbeke (2004) observed the regional preference of the majority of the world’s largest MNEs the LOF appears to have a different effect within a region compared to outside-region investments. Rugman and Verbeke (2007) argue that this is because “the additional costs of doing business abroad are often much higher when venturing into other regions of the world than when expanding intra-regionally, in the home triad region” (Rugman and Verbeke, 2007, p. 204). These authors then distinguish between intra-regional LOF and interregional LOF. This can be interpreted in two ways. LOF is either absent within the home region and only plays a role outside the home region or it is significantly higher outside the home region than within the home region (Asmussen, 2008). In either case, however, MNEs will have a preference for intra-regional expansion.

The operations of the largest MNEs in the world, thus, appear to be less global as assumed by themselves and by scholars. MNEs rather tend to organize their operations in their home region or in some cases in their home region and in one additional region. It is important to explain this phenomenon and there are two schools in IB which offer appropriate explanations. These are the resource-based view (RBV) and the institutions-based view (IBV).

Resource-based explanations of regionalization

The RBV stream in the international strategy literature focuses on internal factors influencing strategic decision making. The core argument is that strategy is driven by firm specific differences (Peng et. al., 2007). Successful strategy and superior performance then result from the possession of valuable, rare, inimitable and non-substitutable resources and capabilities (Barney, 1991), also called firm-specific advantages (FSAs). When firms internationalize their businesses they transfer these FSAs to the host country. However, the degree to which these FSAs can be transferred globally is limited as these FSAs always interact with country-specific advantages (CSAs) (Rugman, 2005).

The exploitability of a MNEs FSAs in a host region primarily depends on the nature of the specific FSAs. The general distinction is being made between FSA which are location bound (LB) and FSAs which are non-location bound (NLB). When expanding to another country a firm transfers its NLB FSAs to the host region. These NLB FSAs form the basis for performance and competitive advantage in the host region. However, these FSAs are successful in the home region because of the presence of complementary LB FSAs. In order for the MNE to be successful in host regions, it needs to develop LB FSAs in the host region as well. Global firms succeed in this mission, but most firms do

(13)

13

Focus of the MNE Criteria Distribution of Global

Fortune 500 firms (2001)

Distribution of Global Fortune 500 firms (2012)

Largest MNEs per category 2012

Home-region orientated Firms with at least 50% of their sales in the home region

320 (64%) 215 (43%) 1. Toyota

2. Volkswagen 3. E.ON Bi-regionally orientated Firms with at least 20% of their

sales in each of two regions, but less than 50% in any one region

25 (5%) 39 (7,8%) 1. Banco Santander

2. Int. Bus. Machines 3. Citigroup

Host-region orientated Firms with more than 50% of their sales in one region other

than their home region

11 (2,2%) 13 (2,6%) 1. HSBC Holdings

2. Noble Group 3. Rio Tinto Group Global orientated Firms with at least 20% of sales in

all triad regions, but less

than 50% in one of these regions

9 (1,8%) 26 (5,2%) 1. Novartis

2. Nokia 3. Roche Group Insufficient or missing data Firms for which data is insufficient

or for which data is not available

135 (27%) 207 (41,4%) -

Table 1: Classifications of strategic orientations and distribution of the Global Fortune 500 MNEs in 2001 and 2012 Source: Rugman and Verbeke (2004); own research

(14)

14 not and.Rugman and Verbeke (2007) argue that a majority of the MNEs stay regionally focused because they are unable to develop the required LB FSAs in host regions. This inability to develop LB FSAs results from unfamiliarity with the host country’s environment, which is identified by Zaheer (1995) as one of the sources of the LOF. A second reason that these authors give is that NLB FSAs cannot be transferred universally around the globe. Though it is theoretically assumed that these FSAs can be transferred globally, the degree to which this is possible decreases with the distance between the home country and the host country (Rugman and Verbeke, 2007; Cuervo-Cazurra et. al., 2007). Thus, in practice “it is very difficult for firms to transfer their successful regional FSAs into global FSAs” (Rugman, 2005, p. 39) and as a result many firms are home-region focused despite potential global locational advantages. This is also confirmed by Collinson and Rugman (2008). These authors concluded in their empirical study of the regional nature of Japanese MNEs that these “are innately tied to the regional and country specific factors, the political, economic, social context and business infrastructures of Japan;; and have evolved FSAs specifically to compete in this environment” (Collinson and Rugman, 2008, p. 13). As a result, most Japanese MNEs are home-region focused and do not succeed to internationalize their business into other regions. This limited transferability makes MNE again suffering from the LOF, which in this case can best be explained by the distance related costs which include amongst others transportation and coordination costs (Zaheer, 1995).

It appears, thus, to be difficult for MNEs to operate in host regions. However, it is not only as a result of this that MNEs tend to expand within the home region. Regional expansions are to an important degree also very attractive for MNEs. This has to do with the reach of LB FSAs. As noted before, when the MNE expands to other regions in the world it transfers its NLB FSAs and has to develop new LB FSAs to complement its NLB FSAs. However, when the MNE expands within the region it is much easier and more important, less costly, to adapt existing LB FSAs and transfer these to the within-region host country: “location-bound FSAs can often easily be augmented so as to become deploy-able in the entire home region” (Rugman and Verbeke, 2007, p. 201). LB FSAs become in this way bound to the home region and not only to the home country. This process, of increasing the reach of LB FSAs, is also indirectly fueled by processes of regional integration. Through regional integration, countries within the region remove trading barriers amongst each other and depending on the degree of integration start to align certain policies (Rugman, 2005). As market characteristics in within-region countries converge as a result of these integration efforts, it becomes easier and less costly for MNEs to expand within the region. What then emerges is the intra-regional LOF as opposed to the interregional LOF (Rugman and Verbeke, 2007). MNEs suffer less from the LOF when they expand within their home region when compared to the LOF in other regions. The abovementioned reasons are the resource-based explanations for this higher interregional LOF.

(15)

15

Institution-based explanations of regionalization

Institutional approaches emphasize the influence of informal and formal dimensions of the institutional context on the MNE’s strategy. The informal dimensions relate to customs, traditions and other culture-related aspects of the context and these aspects generally change very slowly. The formal dimensions include laws and regulations and change quicker and more easily than the informal dimensions of a country’s institutional context (North, 1991;; Williamson, 2000). Both pose constraints on business strategy and raise the costs of doing business in a host country. However, institutions also reduce uncertainty in the business environment. This section discusses how regionalization of businesses can be explained from an institutional perspective.

There exist two dominant perspectives within the IBV stream in the IB literature: the sociological approach as developed by Scott (1995) and the more economic approach which is developed by North (1991). The sociological approach argues that organizations face coercive, normative and cognitive pressures in host countries to which they have to respond in order to obtain legitimacy. North (1991) approaches institutions from a perspective of transaction costs. Institutions in a host country determine the transaction costs for firms if they aspire to expand into that country’s market and the transaction costs can be so high that firms simply decide to not enter that market. Though the sociological approach is appropriate for analysing differences between an MNE’s subsidiaries, especially when the MNEs structure can best be conceptualised as an inter-organizational network (Ghoshal and Bartlett, 1990; Eden and Miller, 2004), it is less appropriate for explaining why or why not an MNE expands to markets in specific geographic areas. Since the aim of this thesis is to explain precisely this phenomenon, North’s (1991) economic approach is better suited.

Institutions have historically emerged to reduce the uncertainty which is always in involved in any exchange or transaction. The institutions became necessary as exchanges and trade evolved from personal transactions with hardly any uncertainty to impersonal transactions characterized by high levels of uncertainty (North, 1987; North, 1991). When there is a high level of uncertainty, the transaction costs involved in the specific exchange will be high as well. These transaction costs arise from “problems of being able to measure the attributes of what is being exchanged and problems of enforcement of the terms of exchange” (North, 1987, p. 420). Transaction costs can be lowered by using more formal contracts. However, formal contracts alone won’t decrease uncertainty. To avoid cheating and opportunism, “elaborate institutional structures must be devised that constrain the participants and so minimize the costly aspects I have just described. As a result, in modern Western societies we have devised formal contracts, bonding of participants, guarantees, brand names, elaborate monitoring systems and effective enforcement mechanisms” (North, 1987, p. 421).

It is at the core of the IBV-approach that institutions, also called the ‘rules of the game’ (North, 1991), affect a firm’s strategy and performance: “an institution-based view of strategy focuses

(16)

16 on the dynamic interaction between institutions and organizations, and considers strategic choices as the outcome of such an interaction” (Peng et. al., 2008, p. 922). MNEs obviously have a preference for expanding to countries with strong formal institutions in which their property rights are ensured and in which they do not face additional costs of cheating or renegotiation. However, strong formal institutions are not necessary. As long as the MNE expects the gains from FDI to outweigh the transaction costs, entrance into the specific country can still be attractive. In these cases, it is necessary for the MNE to think about ways in which the transaction costs resulting from weak institutions should be dealt with (Meyer, 2001).

What is also frequently perceived is that governments design institutions which deliberately raise the transaction costs for foreign firms who wish to enter that market: “foreign firms are often discriminated against by the formal rules of the game in many countries” (Peng et. al., 2008, p. 925). These so-called entry barriers are raised by governments for several reasons, but mainly to protect domestic industries. This is an important source of the LOF which MNEs face in host countries (Zaheer, 1995). It is a general and long-held belief that welfare will be maximized under the conditions of free trade. It is also generally believed that entry barriers destroy welfare. Almost all countries therefore try to reduce these entry barriers. However, “economic welfare can only be expected to increase at the national level, if norms of reciprocity prevail with the other countries involved in the internationalization process” (Rugman and Verbeke, 2005, p. 7). Such reciprocity can be achieved in several ways such as bilateral agreements, multilateral agreements or even sometimes regional integration efforts. The latter is a key theme in the regionalization debate and will therefore be discussed separately, later in this chapter. However, it is also the key IBV explanation for the regional nature of IB. Regional solutions to entry barriers are more efficient and effective than multilateral agreements which can have a global coverage, like for example the WTO: “the demise of global trade and investment liberalization leads to second best arrangements in the form of regional, triad-based government policies and corporate strategies” (Rugman and Verbeke, 2003a, p. 2). It is for this reason that MNEs will suffer less from the LOF in cases of intra-regional expansions as compared to inter-regional expansions.

In the absence of reciprocity, however, MNEs will suffer from the LOF in host countries but also from potential restrictions from the home country institutions. This fourth and final source of the LOF (Zaheer, 1995) can occur when the MNE wants to transfer specific technological or managerial knowledge to countries with which the home country has hostile relationships. This source of the LOF cannot directly be linked to a geographic preference because such restrictions differ per firm, per sector and per home and host country if they occur. This barrier is furthermore less common than the previous three sources of LOF (Zaheer, 1995) and applies only to specific cases in which technological or managerial knowledge is transferred to countries of which the home government believes it hurts its own strategic position in the global arena.

(17)

17

Key themes in regionalization

This final section of the first part of this theoretical framework discusses some key themes in the regionalization debate. The themes which will be discussed are the home-country effect, sector and asset specificity, regional integration processes and the effect of ownership type.

Home-country effect

The empirical finding of Rugman and Verbeke´s (2004) study, that over 80% of the top 500 MNEs in the world is home-region oriented is theoretically extended by Hejazi (2007). This author tested the degree to which US MNEs internationalized and concluded that the dispersion of MNE activity is rather national than global: “It is likely the case that as companies are born and as they expand domestically and become successful, it is in response to local preferences and local needs. As the company moves into a globalization phase, it may still rely most heavily on its local market, which is more in line with the demands the company is best able to satisfy. Furthermore, it is likely that any global expansion will not be uniform across the world to the extent demands and needs for its products are not” (Hejazi, 2007, p. 6-7). The importance of the home market to the MNE is called the home-country effect and it received considerable attention in the regionalization literature. Several empirical studies showed the importance of the home market to the firm. Criticism on the regionalization thesis, however, argues that the evidence for regionalization is the result of this strong home-country effect. MNEs might seem to be regionally organised but when analysed in more detail they appear to be mostly national based, critics argue (Hejazi, 2007).

Sector specificity

Another important element in the literature on regionalization is sector specificity. Several empirical studies showed that the degree of global or regional integration depends very much on the sector involved. Rugman and Girod (2003), for example, analysed the extent to which the international retail sector is globalised. Based on a sample of 49 retail MNEs the authors conclude that there is no evidence that would support the notion of globalisation in this sector. The authors therefore argue that firms should not focus on developing global strategies but instead should come up with what the authors call a regional solution to strategy. A second sector-specific study that confirms the regional focus of MNEs is Filippaios and Rama’s (2008) study of the degree of internationalisation of MNEs in the food & beverage sector. In a sample of 81 MNEs the authors found that a large majority had a regional orientation and that only nine of these MNEs met the criteria for being global. The authors therefore conclude that “though F&B MNEs seem at ﬁrst sight good candidates to become global companies, only a few of them pursue a truly global strategy” (Filippaios and Rama, 2008, p. 68). To make clear the argument that regionalization strategies are not a stepping stone towards global strategy the authors state that “no evidence was found of a linear sequence in international expansion - home-region focused strategy, bi-home-regional strategy and finally global strategy” (Filippaios and Rama, 2008, p. 67). Another sectoral study on the regionalization of MNEs is the study of Oh and Rugman (2006).

(18)

18 These authors examined the degree of internationalisation in the cosmetics and toiletries industry. Based on a sample of 100 MNEs within this sector the authors confirm the regional scope: “Cosmetics MNEs are home region oriented both in the geographic scope of upstream FSAs and of downstream FSAs” (Oh and Rugman, 2006, p. 170).

Asset specificity

Patterns of global presence do not only differ significantly across sectors, but can also differ within a single MNE. MNEs can possess different upstream and downstream FSAs which each have their own potential to penetrate foreign markets. These FSAs are captured in different business units and the potential to ‘become global’ can as a result differ per business unit (Rugman and Verbeke, 2008a). Upstream FSAs are related to production and quite a number of MNEs possess upstream FSAs in all triad regions “to take advantage of market imperfections in markets for raw materials, labor, components and other intermediate good, and even to source final goods” (Rugman and Verbeke, 2008a, p. 308). Downstream FSAs, in contrast represent the geographic scope of sales, the measure which is used in many studies to determine the global presence of an MNE and which is used in this study as well. MNEs often have much more difficulties to extend the scope of these FSAs to countries outside the home region (Rugman and Verbeke, 2008a). The MNEs in the global postal and logistic service sector perform their service in many occasions as part of another firm’s supply chain as logistics is often outsourced by manufacturing firms. In contrast to manufacturing firms, service firms cannot separate production and sales as these are not separable for services (Rugman and Verbeke, 2008a). Production and sales, thus, take place in the same geographic region. However, most firms in the focal industry perform different services which as well may have different potential to penetrate foreign markets.

Regional integration

Regional trade agreements (RTAs) are according to Banalieva et. al. (2012) one of the reasons for the observed regional nature of international business. The importance of these agreements cannot be underestimated as all WTO members except Mongolia are involved in some form of a regional integration process (WTO, 2011). The reason for countries to engage in regional agreements or regional integration is that it stimulates economic growth for countries within the region. This growth originates in the removal of inter-region trade barriers. Though the benefits would be even greater if trade barriers across the whole world would be removed, this is “so challenging to accomplish that regional deals emerge as realistic alternatives” (Peng, 2012, p. 230). The ability of regional trade agreements to affect the MNEs strategy is also acknowledged by Hejazi (2007, p. 25): “trade and investment policy are themselves important determinants of MNE location, and as such, should be viewed as one of the benefits that is likely to accompany such treaties”. Such policies reduce the uncertainty of investing within the region and as a result also lower the transaction costs of doing business within the home-region: “developed economic institutions help firms avoid costly

(19)

19 transactions, such as contract renegotiation and dispute resolution, thus reducing transaction costs” (Banalieva et. al., 2010, p. 801). This is a powerful incentive for firms to invest within the region as compared to inter-region investments.

Important in this respect is the degree of regional integration. An RTAs is just one form of regional integration in which policies are to a very low degree integrated if they are integrated at all. The effect of regional integration depends on the degree of integration: the more integrated RTAs are, the stronger its function as an institution and the greater its influence on firm-level strategy (Banalieva et. al., 2012). However, the effects of regional integration on firm strategy do not come only from the degree of integration. As Banalieva et. al. (2010) explain, the RTA is a multidimensional concept. These authors identified RTA experience, diversity and market size as important dimensions of RTAs and they empirically tested the effects of these dimensions on the geographic focus of MNEs. After examining the relationship between the different RTA dimensions and the global presence of EM MNEs Banalieva et. al. (2010) conclude that there is a negative relationship between home region market size and global orientation and a positive relationship between RTA diversity and experience and global orientation. RTA diversity hereby refers to the institutional variations within the region and RTA experience refers to length of experience the home country has with the RTA. These results are, however, slightly different for Triad MNEs. The authors only find a significant negative relationship between RTA experience and global orientation, suggesting that the longer the experience Triad MNEs have with the RTA the more home region focused they are.

Ownership

A final important topic in the regionalization debate with regard to the postal and logistics sector is the effect of ownership on the internationalization and regionalization of postal and logistic firms. Among the largest postal and logistics firms in the world are both state-owned firms as well as private firms and these different types of ownership could have an important influence on the internationalization and global focus of these firms. While the literature on why and how private firms internationalize is abundant, less has been written about the internationalization of state-owned enterprises (SOEs). The few articles that analyzed this topic were mainly about the internationalization of Chinese SOEs and to move beyond these Chinese SOEs Estrin et. al. (2012) performed an analysis of SOEs from 47 countries. These SOEs were all listed on the stock market and the government was in all instances the majority shareholder.

The key difference between private enterprises and SOEs is that private enterprises are entirely profit driven and that SOEs, of which the government is the majority shareholder, have other objectives than mere profit as well. These objectives are determined by national political processes and these processes differ across countries. Important differences in these processes with regard to SOEs are the strength of democratic institutions and the historical political legacy of a country, which

(20)

20 determine the SOEs objectives and its function in society (Estrin et. al., 2012). Table 2 illustrates that the economic system influences the objectives of the SOE and its degree of internationalization. In contrast to SOEs in capitalist systems, SOEs in social democratic and state-capitalist systems also have social objectives. This decreases the international ambitions of the SOE: “to the extent that SOEs aim to satisfy social objectives, for example by creating employment, these will be attained by operating primarily within the home country” (Estrin et. al., 2012, p. 13).

Table 2: Relationship between political economy and the SOEs degree of internationalization Source: Estrin et.al. (2012)

The type of ownership has also important consequences for the resources and capabilities a firm can develop. Majority state-ownership often results in weak corporate governance, weak monitoring of managers, limited capital market constraints and soft budget constraints (Estrin et. al., 2012). What results from these consequences is that SOEs “will be less competitive than privately owned counterparts and therefore less able to develop capabilities or ‘ownership advantages’ (Dunning, 1993) that would be transferable to be exploited abroad, and which are a key driver for international growth” (Estrin et. al., 2012, p. 17). SOEs will thus be limited in their international ambitions because they are less able to develop FSAs in their home market which will enable them to compete in competitive host country markets. Private firms in contrast operate by definition in competitive markets and these firms necessarily have to develop FSAs in order to survive. Because these firms possess the FSAs to potentially compete in foreign markets, these firms are expected to internationalize to a larger extent than SOEs. The limited ability and need of SOEs to develop FSAs, is empirically confirmed by Rugman and Li (2007). These authors examined the potential of Chinese MNEs to become global or bi-regional firms. As many of the largest Chinese firms are state-owned the authors had the necessity to distinguish between both types of firms. Important to note is that the SOEs Rugman and Li (2007) talk about operate in protected industries and these SOEs “are poor candidates for internationalization because they have been slow to engage in the realities of market driven efficiency. As they go abroad their domestic monopoly protection, with its resulting inefficiency, will serve them badly in competitive foreign markets” (Rugman and Li, 2007, p. 341).

Political Economy Objectives of SOEs Degree of internationalization of SOEs

Capitalist Profit SOEs emulate private firms, incl.

internationalization Social-democratic Profit and social goals

like employment

Low, social goals are attained domestically

State-capitalist Social goals and national development

Low, SOEs operate to inefficient to compete in int. markets

(21)

21

Logistic service providers as business networks

Logistic service providers (LSPs) are part of an often complex network of linkages with both competitors and clients (Lemoine and Dagnaes, 2003). It is necessary in the context of this research to distinguish between both because the former deals with the distribution and delivery once a client relationship is established, the latter deals with sales. Since the focus of this research is on the geographic distribution of sales, the relationships between the focal MNEs and clients needs to be studied.

The LSPs sell their services mostly to other businesses, which can range from large manufacturing MNEs to small e-businesses (Lemoine and Dagnaes, 2003). As already explained in the introduction, the relationships with these firms were an important driver of the internationalization of these MNEs (Majkgard and Deo Sharma, 1998; Bowen and Leinbach, 2004; Farrell and Wood, 1996). All these relationships between the MNE and its clients constitute the specific part of the MNE’s business network which is analyzed in this study. Relationships with these clients have to be built and building a successful relationship with a client may require a long period of time and a lot of resources to be invested. It is furthermore a process which is essentially informal (Johanson and Vahlne, 2009). However, once a stable relationship in which both sides trust each other is build it “is an important [intangible] firm resource” (Johanson and Vahlne, 2009, p. 1414) and for service firms it even is “the most important asset” (Teece, 2003, p. 902). What the firm actually does when building these relationships is building a reputation for being a reliable partner to its client (Hitt et. al., 2006). Reputation is so important for service firms because it is the most effective way for these firms to sell their service: “a firm's reputation and the associated trust in specific partners by clients are likely to be more important resources for service firms than their specialty expertise (Cooper et. al., 2000; Grosse, 2000), especially in generating demand for the firms' services from customers in foreign countries” (Hitt et. al., 2006, p. 1140).

Though the literature on the internationalization of service firms discusses the concept of reputation, this study takes a broader approach by identifying brand image as the crucial concept. Brand image is defined as “a concept of a brand that is held by the consumer” which is “largely a subjective and perceptual phenomenon that is formed through consumer interpretation, whether reasoned or emotional” (Dobni and Zinkhan, 1990, p. 118). It is the transferability of these downstream brand-related FSAs which is crucial for the establishment of new client relationships in host country markets. However, “the larger the psychic distance, other things being equal, the more difficult it is to build new relationships. This is the effect of the liability of foreignness” (Johanson and Vahlne, 2009, p. 1414). This implies that brand image is an intangible asset which, although it is transferred within the network, will also be restricted to this network with the home market partner if it is to be transferred to distant countries. Because the psychic distance exists in particular between

(22)

22 regions and less within regions (Rugman and Verbeke, 2007; Rugman and Oh, 2012) intangible brand FSAs are expected to be more difficult to transfer inter-regionally. The first working proposition therefore is:

WP1: The degree of international transferability of intangible brand FSAs to the host region impedes the capability to build new client networks.

Institutional distance has also very important consequences for the quality of the provided service. Because in contrast to the quality of products, which “can usually also be made homogeneous in terms of quality, because quality control routines can be deployed during production” (Rugman and Verbeke, 2008, p. 399) services “may be associated with a higher degree of heterogeneity, to the extent that the human element, not captured fully in routines, remains important” (Rugman and Verbeke, 2008, p. 399). Service quality is therefore an intangible FSA which, in contrast to the quality of most products, cannot be transferred universally because it is largely embedded in human routines: “the problem with expanding to high distance, host environments (including cultural, economic and institutional distance), is that it becomes more difficult to guarantee service quality and to correct quality deficiencies” (Rugman and Verbeke, 2008, p. 399). Important to emphasize is that service quality in this context does not refer to consumer perception but to operational difficulties which the LSP faces. Where brand image is a downstream FSA, service quality is thus an upstream FSA of which it is expected that it is difficult to transfer to distant host markets. High distance exists in particular between regions and less within regions (Rugman and Verbeke, 2007) and service quality therefore is more likely to decrease when the MNE expands inter-regionally:

WP2a: The degree of international transferability of operational service quality FSAs to host region countries limits the global expansions of LSP MNEs.

The cause of the limited cross-border transferability of the upstream service quality FSAs is to be found in the entry modes which LSPs use. The entry mode choice results from the nature of the specific services in this sector. One important characteristic of these is the limited possibility of decoupling production and consumption (Rugman and Verbeke, 2008b). This means that LSPs do not have the opportunity to learn before they engage in FDI, as for example manufacturing firms have through exporting. A second important aspect in this respect is the non-storability of services: “a disconnect may occur between supply and demand, without the option to “ship” the excess supply of services in one location to high demand occurring elsewhere” (Rugman and Verbeke, 2008b, p. 399). Given the limited possibility to decouple production and consumption and the non-storability of services (Rugman and Verbeke, 2008b) investments in host region markets in which demand is uncertain are very risky. When investments are risky, MNEs will be reluctant to make these investments. A strategy which LSP MNEs use to deal with uncertain demand is the acquisition of local

(23)

23

LSPs. These LSPs possess the very important domestic client network which is a necessary condition to do business in this industry (Johanson and Valhne, 2009). However, the costs which are involved in these acquisitions are usually high compared to other entry modes and it is also likely that additional costs will occur in the post-acquisition phase as the newly acquires firm needs to be integrated in the existing corporate structures (Datta and Puia, 1995). This in particularly applies to situations in which institutional distance between acquirer en acquired is high and in which the acquired firm needs to be “tightly integrated into the acquirer” (Slangen, 2006, p. 161), as it is the case in the focal sector where newly acquired firms need to be integrated in the worldwide networks of the LSPs. Research furthermore shows that in these situations foreign expansion performance will be low due to integration problems (Slangen, 2006). Integration problems are thus expected to be the source of perceived service quality decrease which, as argued in working proposition 2, occurs in particular in high distance situations. Because distance exist in particular on a regional level and not so much on a country level (Rugman and Verbeke, 2007) it is expected that these problems will occur in inter-regional expansions. Working proposition 2b therefore is:

WP2b: LSP MNEs face challenges in inter-regional expansions which result from high entry costs and costs related to the integration of the acquired local LSPs in the own organizational structures.

The ownership distinction is already discusses earlier this chapter and it became clear that publicly traded companies have a much larger need to develop FSAs to survive compared to their state-owned counterparts (Estrin et. al, 2012; Rugman and Li, 2007). Competition with SOEs is therefore very attractive for privately owned enterprises, because they will most probably be capable to win the competition with the SOEs given their stronger resources and capabilities. This opportunity for privately owned companies is recognized by host country governments and foreign firms aiming at expansion are in many markets confronted with a wide range of entry barriers and privileges for the domestic postal service. Privileges for domestic companies are an important institutional barrier for foreign MNEs who seek to enter a specific market (Slangen and Beugelsdijk, 2010) and these barriers include institutional differences with the home market and deliberately raised entry barriers to protect domestic markets. The most extreme entry barrier is the granting of a monopoly to a national postal service provider. However, as many postal markets around the world are liberalized in the past decades (Panzar, 2004) most markets are open to foreign enterprises and thus open to competition. It appears, however, that despite many postal markets are open to competition these markets lack a level-playing field because the domestic postal service enjoys privileges in many instances. The most important reason for host governments to protect their domestic postal service against foreign competitors is the large importance of these firms for the domestic employment: “the postal sector is one of the largest employers in Europe. Privately employed postal workers typically work longer hours and for less pay than publicly employed postal workers. Political opposition to reform [market

(24)

24 liberliazation] was therefore likely to be massive” (Knudsen, 2000, p. 37). To illustrate, the United States Postal Service and China Post are ranked sixth and seventh in the list of the largest employers worldwide in 2012 in terms of employees (Fortune Global 500, 2012; see table 3). Though these protective measures can have different forms, they are expected to apply especially to firms from other regions as intra-regional trade barriers in the postal sector are removed in many regions in the world (Panzar, 2004). The third working proposition therefore is:

WP3: Market-seeking LSP MNEs face challenges in their inter-regional expansions due to protective entry barriers which are deliberately raised by governments in host-region countries.

The abovementioned drivers and barriers to LSP MNE’s internationalization efforts are expected to be particularly relevant to listed companies and less to SOEs. Differences in geographic presence between publicly traded companies and SOEs which result from different international strategies were already noticed by Mascarhenas (1989), who argued that SOEs were “focused on their domestic market with a narrow product line and stable consumer base” (Mascarhenas, 1989, p. 582) and concluded that “ownership may be a theoretical based, parsimonious approach to parsing the international strategic space” (Mascarhenas, 1989, p. 582). This argument is also made by Estrin et. al. (2012) who concluded that “state ownership reduces a firm’s degree of internationalization” (Estrin et. al., 2012, p. 2). Differences between SOEs in their degree of internationalization are then explained in terms of the home country institutional environment (Estrin et. al., 2012). However, there could also be an alternative explanation. Because one thing Estrin et. al. (2012) did not account for in their analysis is the difference in political salience between public interests in different sectors. A public policy issue is politically salient if it has “widespread interest among a large segment of likely voters” (Bonardi and Keim, 2005, p. 555). If the public interest which the SOE needs to serve is highly salient, government authorities will have a strong incentive to serve this interest (Bonardi and Keim, 2005). Though the relationship between the salience of a specific social issue and the degree of internationalization of an SOE appears to be understudied, it can be argued that if the public interest can only be served domestically the strategic focus of the SOE becomes more and more domestic as the salience of the relevant social issue increases. In the focal sector, the public interest which is most salient is employment (Lovelock, 1999; Knudsen, 2000, see also table 3). LSPs in all regions in the world, regardless the institutional environment, employ many people and labor costs can make up 80% of the total costs (USPS, 2012). The salience of the public interest relevant to the SOE is therefore expected to be an important barrier to the internationalization of the SOE. The final working proposition is:

WP4: State-owned LSPs remain home country orientated because the postal sector is politically salient and political objectives have to be achieved domestically.

(25)

25

Methodology

The research question in this thesis will be answered by using a multiple-case study design (Yin, 1994). This chapter will start with a discussion of the ontological and epistemological foundations on which this research design rests (Brannick and Coghlan, 2007). These more abstract assumptions are about the research philosophy of the researcher and they logically justify the methodology adopted in this thesis. When these foundations are clear, the multiple-case design will be discussed and particularly important with regard to the design are quality criteria and case selection of the design (Yin, 1994). This section furthermore includes a description of the cases in this research and it will conclude with a discussion of the types of data which are used in this study, the sources of these data and the analytical strategy used to analyse the data.

Ontology

The research ontology describes the assumptions about what the world is, epistemology describes what we can know about this world and the research methodology describes the very specific way in which it is possible to gather that knowledge of the world (Fleetwood, 2005). There are two broad ontological stances. These are the objectivist and the subjectivist positions. The objectivist ontology argues that the world exists independent of the human mind: “social and natural reality have an independent existence before human cognition” (Brannick and Coghlan, 2007, p. 62). The subjectivist ontology, in contrast, argues that “what is taken as reality is an output of human cognitive process” (Brannick and Coghlan, 2007, p. 62). The latter is particularly relevant when social actors with possible different perceptions of reality are involved in the research. The ontology of this research, however, is an objectivist ontology as this paradigm is the most applied paradigm in management studies and it takes the existence of organisations for granted and they are seen as “objective material entities” (Reed, 2005, p. 1622).

Epistemology

How we can know the world and the nature of knowledge is articulated by epistmological positions and there are four major epistmological paradigms: positivism, post-positivism, critical theory and constructivism (Brannick and Coghlan, 2007). What we can know of the world logically follows from our belief of the nature of the world. When we adhere to the objectivist ontological position and believe that the world exists independent from the human mind and therefore out of objective entities, then it is most logical to assume that the nature of knowledge is also independent of the human mind. In other words, what we can know is objective knowledge which is free from interpretations and different perceptions (Brannick and Coghlan, 2007). This epistemological belief is also called the post-positivist paradigm (Hyde, 2000) and is appropriate for answering the main research question.

(26)

26

Qualitative Multiple-Case Study Design

Iin this study a multiple-case study is adopted as the study contains multiple cases of the same phenomenon. The case study method in general is defined as an “in-depth study of a particular instance, or a small number of instances, of a phenomenon” (Hyde, 2000, p. 83). This method is appropriate to use if the aim of the study is to examine a “contemporary phenomenon within its real-life context” (Yin, 1994, p. 13) and the research question is a ‘how’ or a ‘why’ question (Yin, 1994; Hyde, 2000). The approaches of Yin (1994) and Eisenhardt (1989) conform to the objectivist ontology and post-positivist epistemology as argued above.

There have traditionally been a number of objections against the use of case study research. Lack of rigor has been a major argument against case study research: “too many times, the case study investigator has been sloppy and has allowed equivocal evidence or biased views to influence the direction of the findings and conclusions” (Yin, 1994, p. 9). Other objections are the limited potential to generalize the results to broader contexts and the enormous length of many case study reports (Yin, 1994). These objections refer to general quality criteria for research which include the construct validity, the internal and external validity and the reliability for more positivist inclined studies (Yin, 1994). An important way to increase the rigor of case studies is the use of triangulation (Hussein, 2009; Yin, 1994) which “refers to a process by which a researcher wants to verify a ﬁnding by showing that independent measures of it agree with or, at least, do not contradict it” (Meijer et. al., 2002, p. 146). Different types of triangulation exist and the relevant types will be discussed with regard to the quality criteria of case study research (Hussein, 2009; Yin, 1994).

Construct validity is concerned with the measurement of the concepts being studied (Yin, 1994). In the context of this research, the overall phenomenon which is studied is the regionalization of business activities and patterns of regionalization can have resource-based explanations, institutions-based explanations or network-based explanations. Construct validity deals first with the correctness of the concepts to explain the regionalization phenomenon and second with the way in which these concepts are measured (Yin, 1994) and it is theoretically driven. The RBV and IBV are the main perspectives in international strategy and these schools are also appropriate to explain a regional focus of international strategy (see for a brief discussion Rugman and Verbeke, 2004). Furthermore, the potential influence of network outsidership is conceptualized as an important influence on international business strategy (Johanson and Valhne, 2009). These concepts are measured by using multiple sources of data, which is also called methodological triangulation (Hussein, 2009), see later, and the use of multiple sources of evidence is an important way to increase construct validity (Yin, 1994).

The internal validity of the research mainly relates to the correctness of the observed relationships. Though it is mainly concerned with causal relationships, it is also relevant to case

The global logistics sector : integrating regional or global distribution networks?