Master Thesis
How International Expansion Influences the Product Diversity of US Retailers: Exploring the Moderating Role of Institutional Distance and Internationalization Speed
Final Submission Date: June 22, 2018
Name: Queeny Cheung Student Number: 10971718
First Supervisor: Dr. Vittoria Scalera
Word Count: 18.291 (excluding front page, references & appendix)
MSc Business Administration - International Management Track University of Amsterdam
Statement of Originality
This document is written by Student Queeny Cheung who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Abstract
Research on geographic and product diversification has proven its importance in the International Business (IB) literature. However, many of these studies looked at the consequences of diversification strategies on firm performance and not its interrelationship. Some studies have analyzed the effect of product diversity on geographic expansion strategies in order to research the concept of standardization or localization. However, little is known about the opposite, i.e. the effect of geographic dispersion on product diversity. Therefore, this study sheds light on this relationship, while taking the possible moderating effect of institutional distance and internationalization speed into account. It is important to take the macro environment including cultural, political, economic and legal factors into consideration as these factors may influence the retailer’s decision-making process in how, where and when to expand to foreign markets. Furthermore, internationalization speed is an under researched concept in the IB literature and deserves more attention, as this may influence the ability of retailers to diversify their products when expanding to foreign markets. Since large retailers can represent a sizeable and important sub‐population of the entire retail industry and 70 percent of the top 10 in the 250 Global Powers of Retailing 2018 list (Deloitte, 2018) is dominated by US retailers, the data consists of large US retailers operating between 2 and 106 years internationally and is extracted from Orbis. To conduct this study, I use hierarchical OLS regressions to test the proposed relationships. The results show that there is a linear relationship between geographic dispersion and product diversity. Furthermore, statistical evidence supports a positive moderating effect of internationalization speed. This study contributes to the literature by showing that geographic dispersion has a positive effect on product diversity, and that internationalization speed positively moderates this effect. There is also almost support found for a negative moderating effect of institutional distance on the relationship between geographic dispersion and product diversity. In addition, it contributes to the literature by showing the effect of the least researched facets of firms’ internationalization process, and the importance of acknowledging the influence of internationalization speed in relation to diversification strategies.
Key words: Retail; Geographic Dispersion; Product Diversity; Institutional Distance; Internationalization Speed.
Table of Contents
1 Introduction 6
2 Literature Review 11
Internationalization of Retail firms 11
Entry Modes 13
Diversification within Retail firms 15
Geographic Diversification 17 Product Diversification 19 Institutional Distance 20 Internationalization Speed 21 Research Gap 23 3 Hypothesis Development 25 4 Methodology 32
Data Collection and Sample 32
Description of Variables 34 Independent Variable 34 Dependent Variable 34 Moderating Variables 35 Control Variables 36 Empirical Analysis 38
5 Data Analysis and Results 40
Normalcy, Independence, Homoscedacity and Linearity 40
Descriptive Statistics and Correlations 41
Results 43
6 Discussion 48
Theoretical Contributions 51
Practical and Managerial Recommendations 52
Limitations and Future Research 53
7 Conclusion 55
Acknowledgement 57
Appendices
Appendix A: Normality Testing 67
Appendix B: Normal Distribution of Errors & Homoscedacity 67
Appendix C: Multicollinearity Testing 69
Appendix D: Regression Output H1 69
Appendix E: Regression Output H2 71
Appendix F: Regression Output H3 72
List of Figures
Figure 1. Ansoff Product-Market Matrix 16
Figure 2. Conceptual Model 31
Figure 3. Moderation Model Path Diagram 38
Figure 4. Multiple Regression Scatter Plot 44
Figure 5. Interaction Plot 46
List of Tables
Table 1. Names, Descriptions, Operationalization and Outcomes of Variables 37 Table 2. Bivariate Correlation Matrix and Descriptive Statistics 41
1 Introduction
Apparel retailer UNIQLO expects their international revenue to overtake that of UNIQLO Japan for the first time in 2018, which is interesting as retailers are often more focused on their domestic market than foreign regions (Oh, Sohl and Rugman, 2015). Seventeen years ago, they opened their first overseas store in London and rapidly became an important global apparel player (UNIQLO, 2017). However, firms face challenges as they grow and so has UNIQLO and other retailers over the years (Penrose, 1959). More rapid growth leads to more challenges, and more managerial capacity is needed to overcome these challenges. As an increasing number of firms expand into global markets, the impact of international diversification has generated an important area of scholarly research (e.g., Capar & Kotabe, 2003; Contractor, Kundu, & Hsu, 2003; Hitt, Hoskinsson, & Kim, 1997; Lu & Beamish, 2004). However, current literature draws inconclusive results.
Since global diversification involves dimensions of geographic as well as product markets (Kim, Hwang, & Burgers, 1989), many firms have also diversified their products. In the example of UNIQLO, the retailer offers a large range of warm-weather products that suit the tropical climate year-round, in order to ensure that people that need every day wear visit their store first (UNIQLO, 2017). However, the current literature does not provide one comprehensive explanation on the relationship between geographic diversification, product diversification and internationalization (Hitt, Hoskisson & Ireland, 1994; Tallman & Li, 1996; Delios & Beamish, 1999; Geringer, Tallman & Olsen, 2000; Hashai & Delios, 2012). For example, drawing on the resource-based approach, Davies, Rondi and Sembenelli (2001) argue that for diversified products, firms can utilize internationalization and product diversity to maximize its proprietary assets. Likewise, Delios and Beamisch (1999) argue that the two growth strategies complement each other, because expanding the firm’s geographic scope can generate or acquire new assets that can be used to enter non-core business lines. On the opposite, Tallman and Li (1996) argue that expanding to foreign markets improves the performance of low product-diversified firms by accommodating risk diversification and intensifies their competence to exploit economies of scope.
Most studies in the existing literature examining this relationship use manufacturing firms as their sample, while research shows that contextuality between industries is salient, meaning not all empirical results are feasible across different sectors (Fleming & de Oliveira
Cabral, 2016; Grant, Jammine & Thomas, 1988; Kirca, Roth, Hulst & Cavusgil, 2012; Sukpanich & Rugman, 2007). Retail firms need to invest in the development of a store network to sell products in foreign markets, while manufacturing firms can internationalize sales by exporting products to foreign markets (Oh et al., 2015). Furthermore, building a network of retail stores involves many factors, including selecting and managing local real estate, suppliers, human resources and logistics channels (Corstjens & Lal, 2012). Therefore, I expect the internationalization process of manufacturing firms to be different from the internationalization path of retailers, as it is also important to acknowledge that the success of retail MNEs is determined by effectively managing an international store network and understanding foreign customers further than just product-related preferences (Oh et al., 2015). There may be some overlap, like the need to understand the foreign market and its operations, but given the reasons above, it is likely that the internationalization process of manufacturing firms and retailers differs in certain phases.
Since retailers are required to respond to the culture of their customers in order to achieve success (Dawson, 1994; 2007), owning the capability to understand and adapt to the different dimensions of consumer culture is a managerial capacity that is fundamental and is transferred to the host country by the international retailer (Currah & Wrigley, 2004; Wrigley & Currah, 2003). However, missing this capability either in the firm internally or in the transfers to a particular host country can lead to ‘failed’ international retailing (Dawson, 2007). Marks & Spencer is an example of missing this capability internally, as they lacked a “clear retail positioning and design” (Burt, Mellahi, Jackson & Sparks, 2002, p. 213). Moreover, they did not have knowledge about the needs of their customer on a global level, which implied little synergy and cooperation effects between the (foreign) subsidiaries. An example of missing this capability in transferring knowledge and other resources to a particular host country, is IKEA’s original entry to Japan. In 1974, IKEA decided to enter the Japanese market with a local partner but withdrew in 1986. This was mainly due to the lack of understanding the needs of the Japanese consumer and thinking that the “assemble it yourself” philosophy with a standardized product range could work in every country. Another example is Wal-Mart’s entry in Germany, as they tried to bring their successful US formula in an unmodified manner to the German market. As a result, they did not have sufficient knowledge about the market structure, as well as key cultural and political issues. This shows that the retailer needs to consider and match the firm’s systems and operations
to the economy and culture of the host country. More importantly, the retailer has to transfer into the host market the capability to adapt to local culture (responsive capability), as well as the
capability to change local markets (proactive capability) (Dawson, 2007).
Having said this, lack of knowledge about foreign markets and foreign operations “is an important obstacle to the development of international operations” (Johanson & Vahlne, 1977: p.23). Consequently, learning and knowledge are crucial to the internationalization process of the firm (Hutzschenreuter & Matt, 2017). Vermeulen and Barkema (2002) argue that the rapid expansion of subsidiaries to foreign markets can negatively affect firm performance, because managerial decision-making is imperfect and takes time due to “bounded rationality and limited cognitive scope” (p. 640). However, there are benefits to rapid international diversification, such as the potential to obtain first-mover advantages (Mohr & Batsakis, 2017). I argue that instead of looking only at the effect on firm performance, although it is the main objective of firms (Dawson, 2007), it could be interesting to look at internationalization speed, as this is also an under researched concept.
Thus, I investigate the effect of geographic dispersion on the product diversity of retailers with the following research question: ‘How does the level of geographic dispersion influence the
product portfolio diversification of a retail firm and to what extent does the institutional distance and the level of internationalization speed moderate this relationship?’. It is important to take
the macro environment including cultural, political, economic and legal factors into consideration as these factors may influence the retailer’s decision-making process in how, where and when to expand to foreign markets. Furthermore, internationalization speed deserves more attention, as this may influence the ability of retailers to diversify their products when expanding to foreign markets. I measure diversity with the entropy measure and analyze the relationship between the two diversification strategies. Thereafter, I examine how the institutions of the host country and the pace of expansion play a role in the relationship between geographic dispersion and product diversity. All the data for this study is extracted from Orbis. The sample consists of very large US retailers operating between 2 and 106 years internationally. Subsequently, I use hierarchical OLS regressions to examine the relationships.
The results show that there is a positive linear relationship between geographic dispersion and product diversity. A negative moderating effect of institutional distance was almost supported, but no statistical support was found. However, significant statistical
evidence indicate that internationalization speed positively moderates the relationship between geographic dispersion and product diversity. More specifically, the results show that higher internationalization speed has a bigger positive moderation effect on product diversity than low internationalization speed. This is a surprising outcome, since existing theory on internationalization speed argues that retailers take a path-dependent approach to the internationalization process (Mohr & Batsakis, 2017; Wiersema & Bowen, 2008). These interesting results can be explained by the different expansion strategies of retailers. The retailers that choose organic growth will internationalize slowly, replicating their retail formats to a foreign market. Hence, product diversity also increases slowly. On the other hand, retailers that choose to rapidly expand internationally and try to obtain first-mover advantages (Mohr & Batsakis, 2017) also faster increase their product diversity.
This study contributes to the current IB literature by adding to the theory of internationalization and diversification. First, when looking at the retail internationalization process, Alexander and Myers (2000) argue that more research is needed to understand the different processes that exist in retail firms. That is, the effort to firm specific resources and skills across national boundaries and the internationalization process occurring within the firm itself. With regards to that, this study looks further into the effect of internationalization on the retailer’s product diversity. As a lot of previous research is focused on manufacturing firms and not on service firms, the retail industry is very interesting to look at, since retailers cannot serve foreign markets through exports (Batsakis & Mohr, 2017). Second, in congruence with former studies by Casillas and Moreno-Menéndez (2014) and Batsakis and Mohr (2017), internationalization is an under researched concept. Hence, this study sheds more light on the importance of internationalization speed. The results contribute to the literature by showing that internationalization speed moderates the relationship between geographic dispersion and product diversity, and that future studies comparing internationalization with regards to the product portfolio of the retailer should take this factor into account.
This study also provides practical contributions for retail management. For example, managers of retail firms can use the generated knowledge of the study to rethink their internationalization strategies regarding to expansion speed and taking into account if they need to develop and introduce new products to foreign markets. The results of this study may change the view of seeing rapid internationalization as a threat and a big risk, while this study shows that
this has a positive moderating effect on the relationship between geographic dispersion and product diversity. Moreover, as there is limited support for a negative moderating effect of institutional distance on the relationship between geographic dispersion and product diversity, managers should keep in mind that restrictions in foreign institutional environments may lead to difficulties in diversifying their product assortment. In this case, standardization should be considered.
The thesis is structured as follows: the first section discusses the theoretical foundation of this study by framing the literature to identify a clear research and concise research question. In the second section, the hypotheses development and theoretical framework are described. The third section discusses the research design and methodology that are to conduct this study, which is followed by the description of the results. Lastly, the implications, limitations and future research recommendations are described in the discussion and concluding section.
2 Literature Review
The literature review starts by exploring and explaining the internationalization process of retail firms. Then, related entry mode strategies are discussed. Thereafter, the next section looks into different diversification strategies. Specifically, geographic and product diversification are highlighted. The third section addresses research on the host country’s institutions internationalization speed as moderating effects. Lastly, the research gap summarizes the findings of the literature review and points out the gap for this research, resulting in a concise research question.
Internationalization of Retail Firms
International diversification is increasingly seen by firms as a strategic option to seek sustained competitive advantage (Nachum & Zaheer, 2005). In the past this growth strategy was seen as a way to diversify risks, while current literature perceives international diversification as risk increasing, due to higher exposure to uncertain environments (Alkpinar, 2009). Moreover, recent studies suggest international diversification as a means of market power. However, general literature on internationalization is primarily focused on manufacturing firms (Grant et al., 1988; Sukpanich & Rugman, 2007). It is therefore important to explore which theories also apply to retail firms. The definition of internationalization is widely explained, but in this study, it is considered “a strategy through which a firm expands the sales of its goods or services across the borders of global regions and countries into different geographic locations or markets” (Hitt, Ireland, & Hoskisson, 2007: p.251). What is important to note is that terms like internationalization or international expansion, geographic diversification and multi-nationality tend to refer to the same strategic management construct in the literature, while in reality they might measure different things. However, to follow the majority, it will be simultaneously used in this study to explain how firms expand to foreign markets.
According to Zentes, Morschett and Schramm-Klein (2017), the retailing process is based on two elements: sourcing and selling. Modern retail as we know it today has developed after the industrial revolution and has shifted from being a goods producing process to a demand satisfying process. Retail in 2018 is not just about selling a product, it is rather a service provider to the consumer. This can be explained by the Transaction Cost Economics theory (TCE): due to the power shift from the producer to the consumer, it is now no longer a matter of minimizing
the costs within the supply chain from the producer's point of view but minimizing the find-or search issues that the consumer of today encounters. Retail trade therefore only arises if the transaction costs associated with the direct supply of the producer to the consumer are higher than the sum of the costs involved in delivering products to retailers on the one hand and the retailer supplying the consumer on the other (Quix, 2016). Moreover, the possibility of modifying the value chain in order to gain control over the channels is salient for retail firms, both at the domestic as international level. In this case, Porter’s concept of the value chain (1985) could be the source of competitive advantage, like ZARA, H&M and IKEA convincingly show.
This study mainly focuses on cross-border retailing, meaning the expansion of store operations across the borders of the home country, since it has accelerated greatly over the last two decades (Howard, 2004). Therefore, international retail operations can be defined as the operation of shops or other forms of retail distribution, in more than one country. However, relatively speaking this is still a limited activity for most retailers. Rugman and Girod (2003) argue that retail firms mainly focus on their domestic market due to high capital intensity and location specificity in the retail sector (Campbell & Verbeke, 1994; Rugman & Verbeke 2008). Furthermore, building a network of retail stores involves many factors, including selecting and managing local real estate, suppliers, human resources and logistics channels (Corstjens & Lal, 2012). Therefore, it is important to acknowledge that the success of retail MNEs is determined by effectively managing an international store network and understanding foreign customers further than just product-related preferences (Oh et al., 2015). An example is the world-leading fast fashion company Inditex, being present with 7,292 stores different chains (ZARA, Pull & Bear, Massimo Dutti, Bershka, Stradivarius, Oysho, ZARA HOME and Uterqüe) in 93 markets (Inditex, 2017). The Inditex business model is strongly based on efficient vertical integration and shows that managing the links in the value chain efficiently is one of the factors that can lead to success. More specifically, Zara is able to shorten the design process and distribution to store to two weeks, while the industry average is 5-6 months, leading to more flexibility to respond to trends and thus competitive advantage (Moreno & Carrasco, 2016).
When determining the firm’s performance at different levels of international diversification, researchers have found several non-linear relationships, such as U-shaped (Capar & Cotabe, 2003; Lu & Beamish, 2001), S-shaped (Contractor et al., 2003; Lu & Beamisch, 2004) and M-shaped relationships (Almodovar & Rugman, 2014). However, as the empirical
findings provide inconclusive results, it is acknowledged by recent studies that contextuality is salient, and a generic shape of the curve is not feasible across different sectors (Fleming & de Oliveira Cabral, 2016; Kirca et al., 2012). Contractor et al. (2003) found a U-shaped curve for capital-intensive firms like retailers, which means that international diversification initially creates negative performance, but after a certain point will improve and increase performance, thus supporting a U-shaped relationship. This is also supported by the research of Capar and Kotabe (2003). Furthermore, because retail MNEs mainly focus on their domestic market or home-region market (Rugman & Girod, 2003), this is highly generalizable to the specific industry.
Entry Modes
Consequently, when retail firms internationalize, entry and operation strategies need to be considered. However, this depends on the strategic approach of the firm at home, its market position, the market conditions in foreign countries, the amount of resources available for expansion into foreign markets, the acceptable risk level and the desired level of control. Moreover, “the selected entry method indicates the level of control that the retailer seeks to exert over their foreign operations, the degree of flexibility required in order to effectively respond to market conditions that their foreign enterprise may face” (Moore & Fernie, 2005, p.16). Gielens and Dekimpe (2001) describe two international entry waves from the late 1970s to 1998. The first wave shows that retail firms expand into proximate countries and adopt equity investments. During the second wave, there is development to more distant areas, greenfield expansions, and joint ventures. Moreover, Pederzoli and Kuppelwieser (2015) argue that international retailers seem to follow a twofold approach. On the one hand, they use organic growth to control their international operations and their brands in foreign markets (high level control). Entry modes they use in this case are often direct exports (for example through e-tailing), wholly owned subsidiaries through e.g. outlet multiplication or acquisitions (Zentes et al., 2017). On the other hand, they tend to use franchising as a way to develop internationally, because it is a low-cost strategy (McGoldrick, 1995) and reduces the risk of international investments in developing countries far from the domestic market. This is in line with the work of Alexander and Quinn (2002), who highlight that retailers that do not franchise in the domestic market employ franchising only as an internationalization strategy. Another low level of control and risk
approach is a joint venture, where two or more retailers establish a joint firm in order to enter a new market. “Joint ventures provide the incoming retailer with an opportunity to learn about operations in a new market, while at the same time giving indigenous retailers the opportunity to learn from an international player” (Alexander & Doherty, 2009, p. 258). An example of a retailer that often enters foreign markets this way is Carrefour (Gollnhofer & Turkina, 2015).
As retail formats and assortments may need to be adapted to the entry mode, or respectively the other way around, a major challenge in international retailing is standardization versus adaptation (Zentes et al., 2017). When retailers adopt a standardized strategy, they sell global products that do not need adaptation. However, most products need some changes in the product or promotion strategy to fit the targeted market (Sternquist, 2007). This is because in retailing, the product is the retail business. There are four basic options to approach standardization versus adaptation in international retailing (Zentes, Swoboda & Schramm-Klein, 2013, p. 49–65):
• Domestic market orientation • Global orientation
• Multinational orientation
• Glocal orientation (“think global, act local”)
Domestic orientation relates to retail activities only in the country of origin. Global retailers (as distinct from multinational retailers) tend to operate via standardization, which is the replication of concepts abroad as if targeted markets are homogeneous, thereby ignoring national or regional differences (Salmon & Tordjman, 1989). Moreover, standardization facilitates global retailers’ rapid international expansion (Salmon & Tordjman, 1989; Sternquist, 1997). Glocalization is explained as incorporating “the local context to establish a strategy at the local level, while logalization uses local information as a resource to develop a standardized strategy at the global level” (Kim, Lee & Stoel, 2017).
This market-oriented analysis is in line with the integration/responsiveness framework (I/R framework) by Bartlett and Ghoshal (1989) in the context of international management (Morschett, Schramm-Klein & Zentes, 2015, p. 32–35). Furthermore, these four strategic options in international retailing largely determine decisions over market selection, timing, mode of entry and operations in foreign countries and, the marketing approach itself (e.g. store location, assortment, pricing and the communication mix) (Zentes et al., 2017). On the other hand,
Gripsrud and Benito (2005) advocate that “while manufacturing companies typically enter markets to exploit location advantages” international retail activity “is more likely to be motivated by demand factors” of the market (p. 1672).
Swoboda, Elsner and Morschett (2014) show that retailers use all four strategies (international, global, multinational and transnational) of the I/R framework, but their preferences for these strategies vary across retail sectors (food and non-food). They denote that transnational and multinational strategies are successfully used in the food sector and transnational and global strategies in the non-food sector. This shows that retail internationalization is diverse, and scholars have argued that food retailers, such as Tesco, show a rather multinational behavior (Palmer, 2005), whereas fashion retailers, such as Escada and Esprit, exhibit a rather global behavior (Goldman, 2001; Pederzoli, 2006). Furthermore, other scholars have argued that non-food retailers, such as IKEA, show adapted and integrated behavior (Jonsson & Foss, 2011). To deepen the literature review, diversification within retail firms is discussed in the next section.
Diversification within Retail
Diversification for retail firms is associated with value creation or risk minimization at the customers end, which means that customer characteristics are important determinants for the type of diversification (Oh et al., 2015). Furthermore, it is argued in the literature that increasing levels of diversification can have positive effects on performance, due to economies of scope and scale, market power effects, risk reduction and learning effects (Geringer et al., 2000). This is because the greater the international operations of a firm, the greater the opportunities to leverage strategic resources. Here, the Ansoff Product-Market Matrix presented in figure 1can be used to explain growth strategies and their risks. Market penetration in the lower left quadrant means the enlargement of present market share and has the lowest risks of the four options, since the firm already knows that the product works in the existing market. Market development in the upper left quadrant relates to putting an existing product into a foreign market. Firms can do this for example by entering new geographic markets by moving from local to regional to national and eventually international or global. This may require the business to acquire new capabilities e.g. exporting and understanding different cultures. Third, product development in the lower right quadrant means introducing a new product in the existing market of the firm. Here, the firm
focuses on the current customer and seeks to understand their underlying needs better to find opportunities for new suitable products development.
Figure 1. Ansoff Product-Market Matrix
Diversification in the upper right quadrant involves “offering new products to new markets” and includes three main strategies (Zentes et al., 2017). The first strategy is horizontal diversification, which means that the firm diversifies into a related business field at the same value chain level. The retailer thus opens or acquires stores dedicated to new product categories. However, the distinction between diversification and product development becomes therefore blurred, as adopting new products or retail formats often attracts new consumer segments. Second, vertical diversification involves forward integration (to customer) and backward integration (to supplier). The latter diversification option is more common in retail where retailers often operate production facilities or have full control over the production process, and originally sell to customers directly. Firms that made use of backward integration are e.g. Gap, IKEA, H&M and HEMA. More specifically, in 2015, IKEA bought a forest in Romania in order to have better control over their supply chain. It is also beneficial for quality control and environmental concerns. Forward integration is relatively less common, but still exists. In this case, the producer moves up in the supply chain and start selling its product to consumers. Examples here are: Benetton, Levi’s, Rituals and ZARA (Quix, 2016). Instead of relying on outside partners, ZARA manages almost fully the design, production, warehousing, distribution and logistics of the products, which is different from traditional retailers. The last diversification strategy is conglomerate diversification and involves offering new products or new services to
new markets that are unrelated to the firm’s core activities. However, researchers in IB argue that diversification into unrelated domains can often lead to low performance, as the firms do not have any knowledge or experience related to these new domains (Geringer et al., 2000; Oh et al., 2015).
IB literature shows that both product and international diversification play important roles in determining the strategic behavior of firms (e.g. Hitt et al., 1994; Tallman & Li, 1996). To explore the benefits and costs of geographic and product diversification strategy, the resource-based view and the transaction costs theory are applied in most studies, for instance like Grant et al. (1988), Tallman and Li (1996) and Geringer et al. (2000). Therefore, in the next two subsections, geographic and product diversification are further explored as they are considered the most dominant growth strategies for firms (Hashai & Delios, 2012; Mudambi & Mudambi, 2002).
Geographic Diversification
Most of the studies analyzing geographic diversification focus on a global context (e.g. Lu & Beamisch, 2004), but more recent studies have acknowledged the importance of a regional approach, after the study of Rugman and Verbeke (2004) concluded that the world’s 500 largest firms are mainly operating within their home regions of the triad of North America, the European Union, and Asia. Drawing on the concept of liability of Regional Foreignness (LORF), which are the costs and managerial constraints incurred by MNEs expanding into foreign regions, sources of disadvantages are for example different customer taste, lack of connection within the regional production network and local competition. This results in most of the MNEs confining their cross-border activities within their respective home regions and experiencing negative performance in the initial phase of cross-regional expansion (Kim & Aguilera, 2015). This is also in line with the arguments of Rugman and Girod (2003) for retailers. As customer characteristics are important to geographic diversification (Oh et al., 2015), this logically supports the nature of home-region orientation.
In order to reflect the different demand-based characteristics, geographic diversification can be divided by intra-regional and inter-regional diversification (Oh et al., 2015). Intra-regional diversification considers a retailer’s diversification to other countries within its home triad region, whereas inter-regional diversification means expansion to geographic markets
across triad regions outside the home region. This study focuses on both geographic diversification strategies, even though retail firms mainly focus on their domestic market due to high capital intensity and location specificity in the retail sector (Campbell & Verbeke, 1994; Rugman & Verbeke, 2008), because in retailing, geographic diversification primarily happens by establishing new outlets either in their domestic market or in a foreign market (Zentes et al., 2017).
The literature provides several main internationalization approaches, e.g. the Monopolistic Advantage of Hymer (1976), Dunnings (1988) Eclectic Paradigm/OLI theory and the Uppsala Model (Johanson & Vahlne, 1977). When focusing on retail MNEs, the latter is more applicable, as expansion patterns where firms change or increase their international activities gradually are typical (e.g. Metro Group) (Zentes et al., 2017). Drawing on the concept of psychic distance, retailers often venture first into familiar countries that are psychologically near. Next, they cautiously approach countries which are less familiar or considered psychologically remote. However, Childs and Byoungho (2014) argue that this method is only partially supported, since they find that fashion retailers after initially moving to culturally close countries, moved to countries with close cultural proximity to each other rather than close to home market.
When retailers expand their business to foreign markets, they can develop and gain capabilities through consecutive foreign entries and businesses where they have a competitive advantage, enabling diversification into non-core businesses. Drawing on the resource-based view this perception argues that resources are firm-specific advantages (FSAs), critical sources and outcomes of international expansion, which explains the importance and widespread use of the resource-based view when analyzing internationalization strategy (Delgado-Gomez, Ramirez-Aleson & Espitia-Escuer, 2004; Kotabe, Srinivasan & Aulakh, 2002). In a similar vein, Zentes et al. (2017) assert that only retailers who achieve success in their home market will have the necessary FSAs for geographic expansion to foreign markets, as the capability to adapt is part of the knowledge base of all retailers (Dawson, 2007).
Product Diversification
When assessing geographic expansion, retailers also need to (re)consider their product assortment, which is part of their retail marketing strategy. Therefore, retailers planning to internationalize their operations must define their marketing mixes in domestic and foreign markets based on the four basic types of international retailing within the I/R framework. These retail marketing strategies can be divided into two elements: format and assortment. For the assortment, a distinction can be made between the structural aspect and the operational aspect. The structural aspects refer to e.g. the basic categories and the quantity of SKUs sold, the level of product quality, the price level and the share of store brands. The specific products a retailer offers construct part of the operational aspect (Zentes et al., 2017). The format refers to the type of retail formula and Keep, Hollander and Calantone (1996) claim that product diversification indicated a retail firm's diversification across its retail formats. For example, food stores can have different retail formats, such as hypermarkets, supermarkets, convenience stores and superstores (Oh et al., 2015). Preceding studies made reference to a retail format based on its specific assortment of product and service, store design, and location, targeted to match specific consumer segments (González-Benito, Munoz-Gallego & Kopalle, 2005). Moreover, leading retailers have started to diversify across retail formats, indicating that retailers market an increasingly diversified portfolio of product lines (Deloitte, 2018).
From different theoretical perspectives, related product diversification is found to be essentially different from unrelated product diversification (e.g., Jones & Hill, 1988; Kim et al., 1989; Palepu, 1985; Rumelt, 1974). Building on the theories related to this type of diversification, Chang and Wang (2007) identify five key factors through which the product diversification strategy has important impacts on the performance of international diversification. These five effects are inter-division knowledge learning, synergy, internal control mechanism, adjustment of internal settings with external environments and governance costs. Chang and Wang (2007) argue that although internationally diversified firms may be affected by product diversification, the influences on the above five factors may not be the same for different product diversification strategies.
Rumelt (1974) asserts that related product diversification provides performance advantages, since different product areas can lead to increase of knowledge, while unrelated diversification adds administrative burdens without economies of scope in developing
competencies. The relationship between product diversification mode and performance is well established and can be divided in two directions: the type of diversification and degree of diversification. Subsequently, studies following Rumelt’s (1974) measure have generally found that related diversification leads to higher performance levels than unrelated diversification. However, exogenous factors such as industry effects and firm-level variables tend to absorb much of the effect of diversification type. The ability to develop and expand products to multiple countries can increase sales and reduce operating costs due to economies of scope and scale, given a firm does not over-expand (Geringer et al., 2000).
Institutional Distance
However, retailers are required to respond to the culture of their customers if they want to be successful (Dawson, 1994; 2007). Therefore, if they want to enter a foreign market, the formal and informal institutions of the host country need to be considered (Huang & Sternquist, 2007). Institutions are “rules of the game” and can influence firm behavior (North, 1990). In a general view, Kostova and Zaheer (1999) define institutional distance as the differences or similarities between the firm’s home and host country institutions. Wan and Hoskisson (2003) emphasize the importance of the country environmental context in influencing firm strategies and found that beneficial environment factors in the home country can act as a moderator of the relationship between international diversification and performance. Their results suggest that firms in more stable home country environments enjoy increasing performance levels when they diversify internationally, whereas those in less stable environments do not achieve substantial performance benefits. However, not much research has been conducted on the impact of the host country institutions.
The environmental differences retail firms experience across foreign countries are often not considered in strategy research. Past literature in corporate diversification has repeatedly excluded in what way the national environmental context is related to firm-level strategies. The studies that did consider the home and host country endowments mostly oriented their analysis specifically toward the entry mode decision rather than international diversification in general (Werner, 2002). However, Wan (2005) argues drawing on institutional economics (North, 1990), that country environmental contexts can be perceived as a resourceful environment for conducting transactional activities. For instance, retailers adapted to the evolving environment in
China after its acceptance of the WTO but have also adopted country-specific actions to deal with China’s institutional environment (Dawson, 2007). According to North (1990), “Institutions are the rules of the game in a society or, more formally, are the humanly devised constraints that shape human interaction” (p. 3). Therefore, retailers may need to adapt to these soft rules in order to gain success in the host market.
Even though retailers entering foreign institutional environments may suffer from liability of foreignness (LOF) (Zaheer, 1995), lack of legitimancy (Kostova & Zaheer) and psychic distance that often constraints knowledge transfer when doing business abroad (Johanson & Vahlne, 1977), there are certain motivations for firms to still enter host countries with high risks, e.g. market potential or (knowledge) specific resources (Kochhar & Hitt, 1995). However, Hitt, Tihanyi, Miller and Connelly (2006) argue that the potential “fit” between home and host institutions, capabilities, and resource endowments are not sufficiently analyzed, and a better understanding is needed with further exploration.
Furthermore, Hitt et al. (2006) suggest that the host country resource endowment is an important moderator in the firm’s choice of markets for diversification. This view is supported by Alexander, Rhodes and Meyers (2011), who argue that GDP in the home market creates organizational development and competition, creating a push factor to encourage retailers to enter international markets. Also, retail spend per capita in the host market creates a pull factor to attract retailers into a specific market. This means that managerial decisions are constrained by structural market conditions. However, the role of institutional environments of the host country in decision-making to enter particular international markets is not fully explored yet. More importantly, it has never been considered a possible moderator in the relationship between geographic dispersion and product diversification.
Internationalization Speed of Retailers
Understanding and adapting to a new market may take some time. However, since the mid 1990s the pace of the internationalization process of retailers appears to have quickened (Dawson, 2007). Internationalization speed is a time-based indicator of “how many foreign expansions a firm undertakes in a certain period of time” (Vermeulen & Barkema, 2002, p. 643). Three different approaches of speed are identified in the IB literature. The first conceptualization is that of time elapsing between a firm’s foundation and its first international activity (Zahra &
George, 2002), which is mainly focused on pre-internationalization. The second approach is a more comprehensive observation (Mathews & Zander, 2007) using for example, the average number of foreign markets per year (Vermeulen & Barkema, 2002). The third approach is focused on gaining a profound understanding of how internationalization processes develop: the time elapsing between two consecutive activities in different stages within the internationalization process (Casillas & Acedo, 2013).
Fast internationalization and a rapid exploration of foreign markets generally provide performance and growth advantages for firms (Autio, Sapienza & Almeida, 2000; Oviatt & McDougall, 1994; Powell, 2014). On the other hand, Vermeulen and Barkema (2002) found that the rapid foundation of subsidiaries in foreign markets can negatively affect firm performance, because managerial decision-making is imperfect and takes time due to “bounded rationality and limited cognitive scope”. However, there are benefits to rapid international diversification, such as the potential to obtain first-mover advantages (Mohr & Batsakis, 2017). In the context of retail diversification, these advantages can relate for example to securing desired retail locations or resources such as a well-developed customer base, before the late-movers have a chance (Gielens & Dekimpe, 2001; 2007; Fuentelsaz, Gomez & Polo, 2002). However, the extent to which they realize such potential, depends on certain product market contingencies and the actions of competitors (Kerin, Varadarajan & Peterson, 1992). This means that rapid diversification does not mean that first mover advantages are guaranteed.
Another benefit of rapid internationalization is that it allows firms to quickly develop their strategic resources to improve their competitive position and exploit their valuable resources more quickly (Mohr & Batsakis, 2017). For example, H&M has developed a strong global brand name, which allows them to grow faster than brands that are lesser known. However, when firms increase their internationalization speed, complexity and coordination costs will increase as well, leading to potential time compression diseconomies. In the context of retailers, as these types of firms are often market-seeking, costs effect due to greater international diversification is less pronounced in comparison to firms that are efficiency or strategic asset seeking (Mohr & Batsakis, 2017). In a similar vein, Mohr and Batsakis (2017) found that the negative effects of geographic scope will not outweigh the positive effects, due to the inseparability of production and consumption in the retailing sector.
Schu, Morschett and Swoboda (2016) find that the internationalization speed depends on the next entered country and that the product portfolio also changes. Furthermore, even though internationalization is an important trend, evidence show differences between traditional and online retailers, as the latter internationalizes much faster (Dawson & Mukoyama, 2014). In summary, internationalization speed is an under researched concept, and requires further investigation (Casillas & Moreno-Menéndez, 2014).
Research Gap
A main body of research into internationalization and product diversification is established, but the current literature does not look into its interrelationship. Most of the studies only relate internationalization and product diversification to firm performance, as the main objective of firms is to increase sales (Dawson, 2007). Moreover, most research is conducted with manufacturing firms as their sample, making it interesting to focus on retail firms to analyze to what extend the same empirical results apply. It is argued that global retailers tend to operate via standardization, thereby ignoring national or regional differences (Salmon & Tordjman, 1989). Additional to this, standardization facilitates global retailers’ rapid international expansion (Salmon & Tordjman, 1989; Sternquist, 1997). This means that increasing internationalization to foreign markets lead to the standardization of products in a retailer’s firm portfolio. However, anecdotal evidence shows that retailers actually diversify their product assortment because they need to tailor their products to a different consumer, like UNIQLO does with their all year-round warm-weather clothing in tropical climates. Retailers are also required to respond to the culture of their customers if they want to be successful (Dawson, 1994; 2007). This, in turn, might be affected by the institutions of the host country, since retailers need to adapt to the differences between their home and host country, for example in infrastructure and regulations. However, past literature in international diversification has repeatedly excluded in what way the national environmental context is related to firm-level strategies (Wan, 2005).
Furthermore, due to the high capital intensity of the retail sector, it is argued that retailers often first internationalize within their home-region, as this leads to lower liability of foreignness and the characteristics of their new consumer correspond more with existing customers in the home country. However, some retailers like Inditex show that rapid internationalization and an efficient vertical integration can lead to optimal performance. Research also shows that fast
internationalization and a rapid exploration of foreign markets generally provide performance and growth advantages for firms (Autio et al., 2000; Oviatt & McDougall, 1994; Powell, 2014). However, rather than focusing on firm performance, this study looks at the effect of geographic dispersion on the product portfolio of a retailer. Moreover, it adds the organizational learning process to the effect, which has not been done before. Hence, I examine the moderating effects of the institutional distance between the home and the host country and internationalization speed on the relationship between geographic dispersion and product diversification. Consequently, the following research question is formulated:
How does the level of geographic dispersion influence the product portfolio diversification of a retail firm and to what extent does the institutional distance and the level of internationalization
3 Hypotheses Development
In the following chapter, I build upon the literature review by critically reviewing and working towards clear directions of relationship, resulting in the hypotheses. The framework in figure 2 illustrates the relationship between geographic dispersion and product diversity, as well as the moderating variables: institutional distance and the internationalization speed of the retailer, which will be tested individually.
Hypothesis 1
In the literature it is argued that only firms who achieve domestic success will have the necessary FSAs for internationalization (Zentes et al., 2017). Drawing on the resource-based view, when firms diversify within the scope of their resources and capabilities, they obtain economies of scale and scope by leveraging and maximizing their current strategic resources across business units in utilizing their assets and capabilities e.g. production facilities, distribution channels, or even brand names (Meyer, Wright & Pruthi, 2009; Pennings, Barkema & Douma, 1994). Hence, firms exploit these FSAs with the aim of generating rent and strive to develop them efficiently (Tsang, 2000). In retail perspective, this means that establishing a brand name and a strong customer base in the home market can help firms achieve international success.
However, the question remains, although extensive research has been conducted on either the separate or joint impact of geographic and product diversification on firm performance, what the interrelationship is between geographic dispersion and product diversity (Hashai & Delios, 2012). The studies that did analyze the interrelationship yield different results. For example, Hashai and Delios (2012) show that under-diversification in one dimension and over-diversification in the other will lead to an expansion of the former at the expense of the latter, thus leading to a negative correlation between geographic and product diversification. Davies et al. (2001) found a positive linear relationship, which indicates complementarity of the two growth strategies. Hence, they argue that for diversified products, firms can utilize internationalization and product diversity to maximize its proprietary assets. Likewise, Delios and Beamisch (1999) argue that the two growth strategies complement each other, because expanding the firm’s geographic scope can generate or acquire new assets that can be used to enter non-core business lines. Different from this, Kumar (2009) found a negative linear
relationship, possibly because of limitations in replicating and transferring tacit resources, indicating a substitution effect between geographic and product diversification. Moreover, Kumar (2009) asserts that international and product diversification strategy decisions are made endogenously and simultaneously after taking in consideration the availability of specific resources, while past research assumed this process to be independent from the two options.
As mentioned throughout the literature review, the ability of firms to leverage their unique resources can lead to higher competitive advantage, economies of scale and scope as well as knowledge inflow and outflows. Palich, Cardinal and Miller (2000) claim that geographic diversification reduces the advantages of related product diversification, due to difficulties to synergy formation in marketing, production, and technology internationally. However, I argue that greater geographic scope leads to reduced risks as well as increase returns from product diversification, due to new market opportunities (Kim, Hwang & Burgers, 1993). The resource-based view suggests that firms develop unique resources that they can exploit in foreign markets or use foreign markets as a source for acquiring or developing new resource-based advantages. Thus, based on this theoretical perspective, I assume that greater geographic scope leads to the absorption of tacit knowledge about global opportunities and the capability to leverage such knowledge in a way not matched by competitors. This means that retailers will have the capabilities to reduce risks when expanding to foreign markets. Furthermore, this can also have a positive effect on related product diversification, as new market opportunities can be leveraged by the firm.
Meanwhile, according to Hitt et al. (1994), the combination of increasing geographic and product diversification creates synergies that enable firms to differentiate their products while incurring lower costs than non-diversified firms and increase the stability of their returns. This is more difficult when the firm is only operating in one market. Retailers that for example set up a strong international store network can achieve more sales volume and thus economies of scale and scope. Moreover, resource sharing can be exploited by the retailer, as well as core competences and interdependencies across business units produced by geographic dispersion.
Thus, altogether, I argue that higher geographic dispersion of retail firms makes it easier for them to diversify their products. However, this is only the case when they achieved domestic success and have acquired the necessary FSAs for internationalization. This means that I assume that as geographic dispersion increases, product diversification will also increase. However, this
will be only up to a certain point, after which, as geographic dispersion keeps increasing, product diversification decreases. Thus, I assume that there is a curvilinear relationship between geographic dispersion and product diversification. If product diversification would increase infinitely, the retail firm would not be able to keep an aligned brand image and lose its vision. For instance, Jones (2003) assert that while there are significant benefits of high product development rates, rising development rates also can lead to escalating diseconomies, such as scope limitations, cannibalization and increased portfolio complexities. After a certain point, these costs will negatively influence the increasing benefits of product development, and thus diversity. Hence, I predict that the relationship will show an inverted U-shaped curve. The corresponding hypothesis is then formulated as:
H1: Geographical dispersion has an inverted U-shaped relationship with product diversification.
Hypothesis 2
The international expansion process of retailers leads to the transfer of retail management technology or the establishment of international trading relationships across regulatory, economic, social, and cultural boundaries (Alexander, 1993). From this perspective, the macro environment including cultural, legal, political, and economic factors affect the retailers’ decision-making process as to where, when, and how to expand into foreign markets (Huang & Sternquist, 2007; Vida & Fairhurst, 1998). Furthermore, the literature on global expansion strategy assumes for the most part that strategic decisions or management in the retail sector are driven by economic and political variables (Egelhoff, 1988; Kobrin, 1982; Sternquist & Jin, 1988). Therefore, the institutional theory focuses on the role of the institutional environment in affecting in organizational behavior. In order to gain legitimacy, firms are expected to follow the institutional rules (DiMaggio & Powell, 1983).
When a retailer expands its operations to a foreign market, it will serve new customers whose characteristics are fundamentally different from the characteristics of their existing customers in the home country (Oh et al., 2015). Gollnhofer and Turkina (2015) show that the cultural institutions play an important role in strategic decisions of retailers. Hence, in order to gain international success and legitimacy, it is important for retail firms to integrate the cultural
component into their overall strategy, as the firm will face a substantially different set of competitors and may suffer from latecomer disadvantages. To be able to serve these different markets, the retailer may need to utilize very different types of resources, as well as store networks, then at home due to variations in shopping behaviors and infrastructures (Gereffi, 1999). Furthermore, high location specificity and government regulations generally make it challenging to transfer firm-specific resources to foreign markets (Rugman & Verbeke, 2008). Therefore, to gain access to resources in a foreign market so the retailer can operate effectively, firms are often expected to rely on local partners (Anand & Delios, 1997; Quer, Claver & Rienda, 2007). This explains why retailers adopt low risk approaches like joint ventures. In a similar vein, Padmanabhan and Cho (1996) and Li, Poppo and Zhou (2010) show that joint venture partners can offer this tacit local knowledge and experience, which also decrease cultural distance (Li et al., 2010)
However, there is no empirical evidence on the effect of the institutions in the host country on the relationship between internationalization and product diversity, while in the retail context this is important to consider due to the isomorphic pressures both in the external as internal environment. Isomorphism is considered a constraining process that forces one unit in a population to resemble other units that face the same set of environmental conditions (Dimaggio & Powell, 1983). However, this depends on: selection (number of firms in the field and capacity to carry contextual change) and adjustment (diversity of organizational forms). As firms compete not just for resources and customers, but for political power, institutional legitimacy, social and economic fitness as well, it is important to consider how institutional isomorphic change occurs. Furthermore, Chan, Finnegan and Sternquist (2011) assert that retailers' sales growth is negatively related to the number of retail formats and number of countries of operation. When entering a foreign country, the retailer is confronted with certain “rules of the game” and literature shows that institutions play an essential role in this process (North, 1991). Hence, if the institutional environment has strong advantages and incentives for the retailer to leverage, the retailer may experience less difficulty in expanding with new retail formats and products.
In the regulatory pillar of the institutional framework, Chan et al. (2011) argue that the host country's governance infrastructure reduces retailers’ uncertainty about what legal protection they can expect from the legal system. This means that it is also predictable that when the host country’s governance infrastructure gets stronger, retailers tend to enter the country
earlier and commit with more resources. However, the downside of this is when the governance gets restrictive, retailers may postpone their entry or enter with low resource modes. For instance, India banned foreign retail direct investment, preventing foreign retailers such as Wal-Mart, Carrefour, and Tesco from investing in and operating their own stores (Chan et al., 2011). As a solution, retailers entered India through franchise agreements (Wonacott, 2006). On the other hand, the host country can provide strong inducements such as incentives to attract firms (Hollander, 1970; Meyer & Scott, 1992). For example, in China, there are special economic zones (SEZs) along the coast that provide tax incentives and lower foreign exchange restrictions for overseas investors, attracting foreign firms who are motivated to acquire the benefits of investment in China (Grewal & Dharwadkar, 2002; Ma & Delios, 2007).
Altogether, I propose that the level of institutional distance between the home and the host country can have a negative moderating effect on the relationship between geographic dispersion and product diversification. Expanding to a foreign market means that the retailer will cope with an uncertain environment that is also further complicated by the simultaneously adapting behavior of other retailers or firms (March, 2006). According to Haans, Pieters and He (2015), (inverted) U-shaped relationships can be moderated by a moderator by shifting the turning point of the curve left- or rightward, and the curve can become more flat or steep. In the case of institutional distance, I argue that this moderator will flatten the inverted U-shaped relationship, without the turning point shifting, due to challenges in overcoming institutional differences and its effect on the retailer’s overall expansion strategy. Therefore, the following hypothesis is developed:
H2: Institutional distance between the home and the host country negatively moderates the relationship between geographic dispersion and product diversification.
Hypothesis 3
With regards to new product development and introduction, internationalization speed is an under researched concept, which actually deserves more attention (Lee, Smith, Grimm & Schomburg, 2000; Mohr & Batsakis, 2014; Wong, 2002). Time has been regarded as a central competitive dimension in retail research, but “explanation of conditions under which the speed of international expansion increases or decreases” remains elusive (Luo, Hongxin
Zhao & Du, 2005, p. 755). Anecdotal evidence has also underlined the importance of time-based competition in retailing and how this is reflected in the, often rapid, internationalization of fashion retailers, such as ZARA (Ghemawat & Nueno, 2006; Quinn & Falley, 2010), and hypermarket chains, like Tesco and Carrefour (Coe & Hess, 2005; Coe & Wrigley, 2007; Lowe & Wrigley, 2010). However, in the literature it is argued that retailers take a path-dependent approach to the internationalization process. Therefore, I argue that the concept of internationalization speed needs to be further explored.
Mohr and Batsakis (2017) argue that not correctly judging local demand for the retailer’s products due to managers lacking time for adequate market research can lead to inefficiencies and opportunity costs for overestimation. When demand is underestimated, it can lead to loss of reputation which will be costly to rebuild. The faster firms enter new markets, the more likely such errors may occur. This is in line with the “bounded rationality” concept of Vermeulen and Barkema (2002). As managerial decision-making is imperfect and takes time, rapid internationalization speed will have a negative effect on gaining the adequate information about the foreign market and therefore influence the success of the retailer. This is in line with Wiersema and Bowen (2008), who argue that in the short run, increased coordination and control costs impose limitations to the firm’s fixed bundle of resources and assets. This means that retailers need time to transfer and process tacit knowledge, or otherwise they will likely face time compression diseconomies. Retailers that do not spend sufficient time gathering information about the relevant regulations and specific consumer preferences of the host countries are more likely to make mistakes that lead to additional costs (Mohr & Batsakis, 2017). This is interesting because this may be one of the reasons why retailers are often focused on organic growth and therefore replicating their retail formats to a foreign market. This also means that there is a focus on market development and not on product assortment.
Hence, I expect that internationalization speed negatively moderates the effect between the relationship of geographic dispersion and product diversification, thus flattening the inverted U-shaped relationship without shifting the turning point, due to the constrains of retailers absorbing tacit knowledge and learning in a short period of time. Accordingly, the following hypothesis is formulated:
H3: Internationalization speed negatively moderates the relationship between geographical dispersion and product diversification
In conclusion, figure 2 gives a comprehensive overview of the proposed hypotheses.
