|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 HOW DOES INTERNATIONAL AND FORMAT DIVERSIFICATION AFFECT THE FINANCIAL PERFORMANCE OF RETAILERS? By JEREMY MIANXIN LIM A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT O F THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2011
2 2011 Jeremy Mianxin Lim
3 To my family, especially my parents, who have supported and encouraged me through all of my endeavors, for which I am eternally grateful I also dedicate this dissertation to my advisor, Dr. Barton Alan Weitz, whose insights, patience and kindness I am forever indebted to. Last but not least, I dedicate this dissertation to members of my committee, the marketing faculty at The University of Florida and my fellow PhD students, whose advice, encouragement and patience I treasure.
4 ACKNOWLEDGMENTS Even prior to me joining the doctoral program in August of 2006, I have looked to my advisor Dr. Barton Alan Weitz for advice and encouragement I remembered first meeting him during my campus interview at Florida; his first words were: What would you like to know about my PhD program in a cheerful and genuinely welcoming tone. Ever since, he has been a source of inspiration both in my academic career and in my personal life. He is one of the most determined successful and benevolent persons I have ever met and I wished I had spent more time during my doctoral program learning from him. His encouragement and unfaltering support is t he reason I am able to complete this dissertation, for which I will remain eternally grateful. Next, I would like to acknowledge the support of the acclaimed marketing faculty at Florida, in particular: Dr. Lyle Brenner, Dr. Steve Shugan, Dr. Joe Alba and Dr. Alan Cooke, who have provided me with substantial guidance and encouragement over the course of my studies. I am also grateful to my colleagues, especially Dr. Melissa Minor, Dr. Xiaoqing Jing, Dr. Yuying Shi, Dr. Steven Sweldens Dr. Hyunjoo Oh and Dr Fang Wang, for the experiences we have shared. Last but definitely not least, I would like to acknowledge the incredible amount of emotional and physical support from my parents that helped me make it through the enduring hardships of a top doctoral program.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF TABLES ............................................................................................................ 8 LIST OF FIGURES .......................................................................................................... 9 LIST OF ABBREVIATIONS ........................................................................................... 10 ABSTRACT ................................................................................................................... 11 CHAPTER 1 INTRODUCTION .................................................................................................... 13 Retail Strategy and Retailers Product Market Portfolio Defined ............................. 13 Evolution in the Retail Climate and the Importance of this Research ..................... 13 International market diversification ................................................................... 14 Format diversification ....................................................................................... 15 Interactive synergies between market and format diversification ..................... 15 Unique aspects of retailers not addressed by other research ........................... 16 Timing ............................................................................................................... 17 Intended Contributions ............................................................................................ 17 2 LITERATURE REVIEW .......................................................................................... 20 Un derlying Concepts Influencing Firms Diversification Decisions .......................... 20 Economies of Scale and Scope ........................................................................ 20 Firm Specific Assets ......................................................................................... 21 Order of Entry and the Learning Curve ............................................................ 22 Risk Reduction ................................................................................................. 23 Market Powe r ................................................................................................... 24 Other Factors Influencing Diversification Decisions ......................................... 24 Nature of Diversification Performance Relationship ............................................... 25 Linear Impact of Firms Diversification on Their Financial Performance ........... 25 Positive linear impact of diversification on firm performance ............................ 26 Negative linear impact of diversification on firm performance .......................... 26 Non Linear Impact of Firms Diversification on Their Financial Per formance ... 27 U shaped impact of diversification on firm performance ................................... 28 Inverted U shaped impact of diversification on firm perfor mance ..................... 28 S shaped impact of diversification on firm performance ................................... 29 Cross Country Characteristics and their Impact on Successful Diversification ....... 30 3 CONCEPTUAL FRAMEWORK AND HYPOTHESES ............................................. 32
6 Conceptual Considerations and Assumptions of the Model .................................... 32 Diversification Implications for Retailers and Manufacturers ................................... 33 Benefits and Costs of Diversification ...................................................................... 35 Diversification benefits ...................................................................................... 35 Diversification cost ........................................................................................... 36 International Market and Format Diversificati on and Retailer Financial Performance ........................................................................................................ 37 International Market Diversification Intensity .................................................... 37 Format Diversification Intensity. ....................................................................... 38 International Market Format Diversification Interaction .................................... 39 Effects of the Characteristics of the Portfolio of International Mark et and Format on Retailer Financial Performance ...................................................................... 40 Cultural Dissimilarity ......................................................................................... 40 Economic Dissimilarity ..................................................................................... 41 Format Dissimilarity .......................................................................................... 42 4 METHOD ................................................................................................................ 44 Sample .................................................................................................................... 44 Measures ................................................................................................................ 44 Financial Market Performance .......................................................................... 44 Diversification Intensity ..................................................................................... 45 Format Dissimilarity .......................................................................................... 46 Cultural Dissimilarity ......................................................................................... 47 Economic Dissimilarity ..................................................................................... 48 Summary Statistics ........................................................................................... 49 Estimation ............................................................................................................... 49 Endogeniety Issue ............................................................................................ 49 Method ............................................................................................................. 50 Substantive Issue ............................................................................................. 52 First Stage Estimation ...................................................................................... 53 SecondStage Estimation ................................................................................. 55 5 RESULTS ............................................................................................................... 59 First Stage Results ................................................................................................. 59 Test of Hypotheses (SecondStage Results) .......................................................... 61 International Market and Formats Intensity of Financial Market Performance .. 61 Cultural, Economic, and Format Diversity and Financial Performance ............. 63 6 DISCUSSION ......................................................................................................... 69 Summary of Results ................................................................................................ 69 Directions for Future Research ............................................................................... 71
7 APPENDIX A DATA APPENDIX ................................................................................................... 74 B SAMPLE OF RETAILERS ...................................................................................... 80 C RESULTS FROM FACTOR ANALYSIS OF HOST COUNTRY ECONOMIC VARIABLES ............................................................................................................ 88 D FORMAT DISSIMILARITY SURVEY ...................................................................... 90 LIST OF REFERENCES ............................................................................................... 93 BIOGRAPHICAL SKETCH .......................................................................................... 102
8 LIST OF TABLES Table p age 4 1 Summary statistics for main variables ................................................................ 58 4 2 Correlation matrix for main variables .................................................................. 58 5 1 Estimated results: instruments affecting diversification decisions ....................... 64 5 2 Effect of country and format portfolio constituents on Tobins Q ....................... 65
9 LIST OF FIGURES Figure page 3 1 Conceptual framework ........................................................................................ 43 4 1 Scree plot from factor analysis of host country economic variables ................... 57 5 1 International diversification and retailers Tobins Q ........................................... 66 5 2 Format diversification and retailers Tobins Q .................................................... 66 5 3 Impact of varying levels of market dissimilarity on Tobins Q. ............................ 67 5 4 Impact of varying levels of format dissimilarity on Tobins Q. ............................. 68
10 LIST OF ABBREVIATIONS Constituents Geographical and/or operational format members in the retailers portfolio D P Link Link between the effects of firm diversification and corresponding performance GPR Global Powers of Retailing Hofstede Scores Cultural constructs developed by Dr. Geert Hofstede Retailers Portfolio The set of markets and formats retailers operate in WBI World Bank Development Indicators
11 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy HOW DOES INTERNATIONAL AND FORMAT DIVERSIFICATION AFFECT THE FINANCIAL PERFORMANCE OF RETAILERS? By Jeremy Mianxin Lim December 2011 Chair: Barton Alan Weitz Major: Business Administration A conceptual derivative of the literature on strategic growth management across firms product/market portfolio (i.e. Ansoffs, Boston Consulting Groups and McKenzies produc t/mar ket matrix) this research strives to provide similar empirically based guidance to retailers. Specifically, this research examines the impact of retailer s portfolio of countries and formats on their financial performance, while accounting for dissimilarities across the elements of the retailers portfolio. I nitially drawing on the methodology and constructs used in studies on market entry and firm performance in formulating dissimilarity constructs this study subsequently departs from the micro aspects of market entry and product launch sequence in an effort to provide a holistic overview on how retailers should strategically manage their product market portfolio to achieve optimal financial performance, while accounting for both the effect of heterogeneous geographical and product markets and any corresponding higher order interactive effects (i.e. synergies in particular product market combinations). Based on a sample of the largest global retailers over a fiveyear period, the results indicate that the relationship between both the number of counties and formats
12 and financial performance is U shaped. However, due to the negative interaction between the number of countries and formats, the overall effect on financial performance is more complex within the range of the observations. Dissimilarities in retail formats and the cultural and economic characteristics of the countries (in which the retailer operates in) have significant negative effect on the retailers financial performance, but the impact of t hese dissimilarities in the retailers portfolio is less than the impact from the number of countries and formats in the retailers portfolio. My research shows that retailers operating a limited number of formats in many countries or many formats in a few countries have the best financial performance.
13 CHAPTER 1 INTRODUCTION Retail Strategy and Retailers Product Market Portfolio Defined A fundamental strategic marketing decision facing firms is: how to determine their product market portfolio the set of markets in which a particular firm chooses to operate in and the product(s) directed toward these respective markets (e.g. Aaker 2005, pp. 3033; Kerin and Pederson 2007, pp. 710) A retailers product market portfolio is the result of the retailer pursuing two growth strategies: (1) market diversification adding new markets to their operating portfolio and (2) format d iversification launching new formats (Gielens & Dekimpe 2001) Examples of the former include: McDonalds expanding into China and Wal Mart expanding into Argentina, while examples of the latter include: Wal Mart expanding its operations from discount stores to include hypermarkets and warehouse clubs and British retailer Tesco expanding its operations to include convenience stores (i.e. Tesco Express). Market diversification alters the number and nature of markets in which retailers operate in (demandside diversification) while format diversification affects the different operational resources utilized by retailers (supply side diversification) Evolution in the Retail Climate and the Importance of this Research Faced with maturing markets and increasing competition, retailers like most other firms are faced with two particularly pressing strategic issues: 1) how to adapt to challenges arising from both globalization and technological advances and 2) how to ascertain the extent to which retailers should tradeoff their core operations [that they have proven core competencies in] to capitalize on the resulting growth opportunities that have become ev er more so viable.
14 In their search for new growth opportunities, retailers face critical decisions concerning the degree to which they should engage in market and format diversification, along with the characteristics of the resulting operational portfolio of international markets and formats that the retailer operates in (Gielens & Dekimpe 2001; Levy & Weitz 2009) International market diversification Many academic and industry observers extol the need for retailers to pursue international market diver sification. For instance, Gielens and Dekimpe (2001) argue that: to avoid a pure market share game in increasingly saturated domestic markets, retailers are increasingly forced to also look for new geographical markets and Retail Forward, in its outline of the evolving retail landscape, emphasizes that: Global Scope [in 2015] will be a necessity, not an option to grow the top line and bolster the bottom line. Global expansion will be a key avenue for retailers in developed markets to generate new sources of revenue to offset slower sales growth at home . They [American retailers] will be late to the game, especially compared with large European retailers . (Pollack, 2007, p. 11). Indeed, the evidence suggest that retailers are taking the above sentiments to heart and devoting proportionally more resources to cross border operations. For instance, Higgins ( 1997) and M ulhern (1997) noted that the worlds 100 largest retailers are growing twice as fast abroad as they are domestically, with the top35 largest retailers averaging one new market entry per year. However, the returns are unclear given the lack of empirical research on entry and/or post entry performance (Feeser & Willard 1990; Gielens & Dekimpe 2001; Sharma & Kesner 1996) Anecdotal evidence suggest that retailers realize a
15 substantially lower [or even negative] return on investment (ROI) on foreign compared to domestic operations. For ins tance, Carrefour and Wal Mart not only maintain uniformly lower average ROIs on foreign versus domestic operations but have also sustain losses in many markets (The Economist, 1999).1Format diversification Gielens and Dekimpe (2001) asserts that few international retailers realize comparable margins or break even volumes in foreign markets and also suggest that there is a longrun optimal level of in ternational market diversity for retailers. The other growth opportunity pursued by retailers is format diversification. Format diversification is the variety of formats that retailers utilize to offer goods or services to their tar get market(s). Format diversity in the retail industry is conceptually similar to industrial and product diversity constructs used in studies of diversification strategies engaged in by manufacturing firms. Format diversification, like product and industr ial diversification, focuses on the variety of unique resources and capabilities utilized by firms. For example, expertise in fashions and merchandise budget planning are key performance drivers for retailers operating apparel specialty store; while exper tise in supply chain management and cost control are performance drivers for hypermarket retailers. Interactive synergies between market and format diversification Gielens and Dekimpe (2001) noted that format diversification evolved as a more aggressive c ompetitive response from retailers simply offering new and broader store assortments, while RetailWire ( 2009) suggest that new formats [particularly, the more 1 Losses were so heavy in certain markets and necessitated retreat.
16 successful formats] are the result of retailers evolving to meet changing consumer habits and current economic conditions. Implicit in the latter report is the emphasis on retailers focus on customer orientation and delivering form ats with a high value proposition to consumers (i.e. demanddriven product extension) as opposed to the former justification (i.e. competition/resource [supply] driven product extension). Unique aspects of retailers not addressed by other research T he benefits of international market and format diversification on firm performance are unclear. Research examining the affects of international market diversification and financial performance is mixed. Research has found positive and negative linear relationsh ips as well as nonlinear relationships (e.g. Annavarjula and Beldona 2000; Hitt, Tihanyi, Miller and Connelly 2006; Li 2007). Similarly, research on the effects of product/industry diversification and financial performance has produced inconsistent results with some research suggesting that product diversification moderates the international market diversificationperformance relationship (Chang and Wang 2007). But almost all of the extant research has examined the impact of diversification on the performance of manufacturing firms. For example, Morgan and Rego (2009) examined the effect of brand portfolios (the number of brands and market segments) on the financial performance of brand manufactures. Very few studies have focused on the impact of market and format diversification strategies for firms in service industries and even fewer for the retail industry. This lack of research on diversification in service and retail industries is problematic for two reasons. First, service industries are faster growing than manufacturing industries worldwide and dominate the business environment in most developed countries. Second, there are significant differences in the operating
17 characteristics of retailers and other service firms compared to manufacturing firms differences than can affect the diversificationperformance relationships (Dawson 1994, 2007 and Wrigley, Coe and Currah 2005). Due to the lack on research on the diversificationperformance relationship and the unique aspects of retail and other service industries, several scholars have questioned the generalizability of the extant manufacturingdominated research on diversification and have called for studies on the effects of diversification in service industries (c. f. Agarwal & Ramaswami 1992; Errami lli & Rao 1993; Ekeledo & Sivakumar 1998). Timing Heigtening globalization in the retail industry is intensifying crossborder competition. This coupled with the compounding effect of the recession on consumer spending in domestic markets, is forcing many retailers to rethink and restructure themselves so as to preserve their competitive position and preserve [or even increase] their value proposition to consumers ( Gielens and Dekimpe, 2001; RetailWire 2009; Sage, 2009). Indeed, Sage (2009) noted that the shrinking domestic prospects, along with the market growth in emerging economies, is prompting many retailers to implement not only market but also format diversification plans at the onset of available cash flow. Wal Marts plan to open its first cashandcarry store in India is a very characteristic example that underscores the relevance of this research. Intended Contributions Given the highly contagious globalization fever affecting retailers, the recent recessionary impact on the retail industry and the lack of empirical support and
18 guidance on market diversification strategies, specific to retailers, this research is especially timely and relevant (Gielens and Dekimpe, 2001). Specifically, this study considers a very specific subset of firms with uni que characteristics, namely service type firms that have largely been underrepresented by the extant market diversification literature. In addition, this study provides additional contributions by questioning whether there exist a longrun optimal level of market diversification for top performing retailers, while taking into account dissimilarities in cultural and economic conditions across geographic markets and examining the interactive impact from pursuing two separate diversification strategies. It als o addresses some inconsistencies in prior research by examining the linear, nonlinear, and interactive effects of market format portfolios; correcting for endogeniety in the estimated effects of diversification intensity on performance; and using a financi al market measure (Tobins Q) rather than accounting measures of diversification portfolio performance. Another unique aspect of this study is its focus on the operational dissimilarities across different retail formats and its corresponding impact on retailers financial performance. To the best of my knowledge, this is one of two studies to account for this important yet under researched dimension format familiarity as termed by Gielens and Dekimpe (2001) in the link between diversification and f inancial performance. Furthermore, t his research utilizes a global sample and takes a holistic and long run outlook when examining the effects of both international market and format diversification specific to the retail context.
19 The questions addressed in this research are: How does international market and format diversification affect the financial performance of retailers? Are there combinations of countries and formats that yield higher financial performance than others? What portfolio characterist ics should retailers seek to exploit market and operational synergies and improve financial performance? Does format diversification improve or degrade the financial performance arising from international diversification? In the following section, I out line my conceptual framework and hypotheses concerning the relationship between characteristics of retailers diversification portfolios and their financial performance. Then I describe my sample and methodology for estimating the relationship between the characteristics of retailers diversification portfolios and their financial market performance. After discussing the results of my estimation, I review this studys limitations and provide directions for future research.
20 CHAPTER 2 LITERATURE REVIEW This chapter begins with an interdisciplinary exposition of the relevant concepts influencing market expansion decisions. Next, the discussion (in the second section) turns to past research on aggregate expansionary strategies and firm performance, while highlighting their inconsistencies and shortcomings. With this background knowledge, I then proceed to review studies examining differences between cross country characteristics and their resulting implications in the third section of this chapter The fou rth section repeats this process for cross format characteristics, drawing on the findings from the product diversification literature. The fifth section highlights important characteristic that are unique to servicetype firms, especially retailers and ex pose shortcomings and the lack of generalizability of the extant diversification literature. Finally, I summarize this review with a synthesis of the literature reviewed and provide support for the relevance and need for this research in the sixth section. Underlying Concepts Influencing Firms Diversification Decisions There are several commonly discussed factors motivating firms to diversify, they include: potential economies of scale and scope, increased market power, the ability to leverage firm specif ic assets and the potential for firms to lower risk by hedging across multiple markets and product lines. Economies of Scale and Scope International and format diversification enable retailers to draw on scale economies to reduce costs. By expanding operations to new markets and using new formats, a firms fixed costs are spread across the increased revenue opportunities
21 arising from diversification (Caves, 1996). However, cost savings may not always be realized due to the increased complexity of managing diversified firms. As firms organizational hierarchy expands, they may encounter greater difficulties in disseminating information to business units resulting in transmission and coordination inefficiencies, such as the loss or distortion of information (Hoskisson & Hitt1988; Williamson 1967). Furthermore, the larger hierarchical structure of diversified firms is conducive to employee shirking; thereby further incurring cost associated with either heightened monitoring or decreased worker productivity (Calvo & Wellisz 1978). Hence, t he increased governance costs of operating a diversified firm may even exceed the benefits of diversification (Hitt et al. 1997; Tallman & Li, 1996). Firm Specific Assets These scale and scope opportunities are enhanced for fir ms possessing valuable firm specific assets, such as well known and highly regarded brand names, unique systems and processes, customer loyalty and managerial skills. These intangible, firm specific assets cannot be readily transacted due to market imperfe ctions but can not only afford firms higher than normal returns in new markets (Markides, 1992) but also enhance their re source allocation efficiency (Froot et al., 1994; Lang et al., 1995; Markides, 1992; Palich et al., 2000). In addition, researchers pr opose that, in contrast to nondiversified firms, diversified firms have greater access to both internal and external sources of financing (Lang & Stulz, 1994) and this expanded access provides diversified firms with privileged access to less costly financ ing (Froot et al., 1994; Lang et al., 1995). However, the transfer of the desired firm specific asset(s) may be met with a range of obstacles including hostility and excessive restrictions in the host market, currency
22 fluctuations and last but not least deteriorating home market performance (Williams, 2001). The latter point is especially relevant to service based firms (in particular to retailers) due to the vast amount of resources needed to establish a retail and distribution network and retailers may not have sufficient resources to do so (Levy & Weitz, 2006). Furthermore, this extensive investment has to be undertaken with great uncertainty since retailers do not have the same luxury of test marketing or throttling productions that is available to manufacturing firms. Indeed, the extensive upfront investment to establish a retail brand and presence will have to be undertaken long before retailers can ascertain whether their retail format will ever to take off (Golder & Tellis 1997) Alexander (1990) and Gielens and Dekimpe (2001) also pointed out that the transfer of firm specific assets (especially in the case of retailers) may not be readily accepted by foreign customers and retailers management and operational culture as well as their retail concept may not be readily implemented in the host culture. Implementation is especially pertinent for service industries especially retailin g due to the large amount of interpersonal interactions involved. Order of Entry and the Learning Curve In addition, diversification can lead to inefficiencies arising from external challenges of operating in a new environment (Hymer 1976, Zaheer 1995). Some of these disadvantages, related to diversification, are due to inefficiencies resulting from the lack of market reputation (Barkema, Bell & Pennings 1996); limited knowledge about the new environment, which leads to lower efficiencies than native fir ms (Hyman 1976); and increased uncertainty from having to operate in a complex and unfamiliar
23 environment, characterized by a different set of economic, political and legal factors (Sambharya 1996). However, studies examining firm entry outcomes have shown that there is a learning effect at play. Specifically, incumbent entrants can mitigate the level of unfamiliarity by entering markets that are similar to their existing markets. Indeed, successfully diversified firms typically enter markets that are simil ar to their home markets or markets that they have experience in (Mitra & Golder 2001). The above studies show that entry decisions in general are not unilateral but rather encompass elements of self selection (with regard to size, market potential and sim ilarities and also the portability and acceptance of the firms product and operating concept in the host country). Firms that are successful at market and format diversification can then enjoy the benefits of scope and can match the appropriate format(s ) with the needs of the target market. Risk R eduction In a related vein, diversification is also argued to reduce a firms risk by reducing the effects of nonsystematic fluctuations over more business units (Berger & Ofek 1995; Kim et al. 1993; Lewellen1971; Markham 1973). This risk reduction spawns a feedback effect, affording firms with reduced risks to realize better external financing terms (Duffie & Singleton 1997) and improved borrowing ability (Shleifer & Vishny 1992). Though valid for the manufacturing industry, this particular rationale may not apply to retailers and may even be detrimental to retailers performance due to the extensive resources needed to establish a retail network in just one additional market. In fact, Gielens and Dekimpe (2001) and Williams (2001) both suggest that retailers who enter
24 new markets (especially markets that are highly competitive) may do so at the expense of their home market. Market Power The ability to cross subsidize internal resources provides yet another benef it to diversified firms: market power. Larger diversified firms can negotiate better terms with suppliers and can even demand exclusive relationships, engage in predatory pricing against competitors (Berger & Ofek 1995; Markham 1973) and may even deter ent ry by threatening to respond with a pricing war (Salone, 1987). Indeed, Gielens and Dekimpe (2001) found evidence that the retail chains that have the best long run performance are those who enter early, with substantial scale while offering a store format that is both new to the host market and familiar to the parent firm. Other Factors Influencing Diversification Decisions As noted in an earlier subsection, diversified firms may be afforded privileged financing and are further able to improve thei r performance by exploiting tax advantages associated with increased borrowing (Melicher and Rush, 1973; Shleifer and Vishny, 1992). Williamson (1964) and Jensen (1986) discuss the agency problems that arise from hierarchical versus market governance. Specifically, managers have an incentive to over diversify since their compensation is often tied to the revenues generated (Murphy 1985). In addition, revenue growth creates new positions for which these managers may be promoted to (Barker 1986). The tendency to over diversify is further exacerbated in mature firms with substantial free cash flow (Jensen 1986; Mueller 1972).
25 This observation further strengthens the case for this research; in particular, this research can serve as a guiding principle as to what level of diversification is optimal resulting in better decisions that are guided by empirical evidence as opposed to managerial incentives. Nature of DiversificationPerformance Relationship The benefits of international market and format diversifi cation on firm performance are unclear. Research examining the affects of international market diversification and financial performance is mixed. Research has found positive and negative linear relationships as well as nonlinear relationships (e.g. Annavarjula and Beldona 2000; Hitt, Tihanyi, Miller and Connelly 2006; Li 2007, Palich et al. 2000). Specifically, Palich et al. (2000) in their synthesis of three decades of research on diversification and firm performance conclude that this area of inquiry falls far short of consensus. Similarly, research on the effects of product/industry diversification and financial performance has produced inconsistent results with some research suggesting that product diversification moderates the international mar ket diversificationperformance relationship (Chang and Wang 2007). Linear Impact of Firms Diversification on Their Financial Performance Earlier research focused on the benefits or costs of diversification and proposed either a positive or negative linear relationship between international and product/industry diversification and performance. C onsistent with basic marketing guiding principles that stress core competencies and market orientation, a survey of the literature revealed that proportionally mor e studies found positive linear as opposed to negative linear relationships between diversification and performance (Hitt et al., 2006).
26 Positive linear impact of diversification on firm performance Support for this model stems largely from theories of mar ket power and internal market efficiencies (Gort, 1962; Grant, 1998; McCutcheon, 1991; Scherer, 1980). Specifically, diversified firms are able to use their asymmetric financial strength (i.e. crosssubsidization across other business units) to engage in predatory pricing driving their more focus rivals out of the market and thereafter reaping gains from higher pricing strategies (Saloner, 1987). Market power can also yield a positive linear relationship for diversified firms as they are able to secure favorable buying and selling arrangements from other firms that are both suppliers and customers of the diversified firm (Schrer, 1980; Sobel 1984). Implicit in these justifications is that a certain degree of market efficiency exists. However, these theories may not be suitable in global and mature markets where competitive intensity is high or in retail markets where margins are sufficiently low Negative linear impact of diversification on firm performance While the agency justification provided in the proceeding section can also lead to spurious diversification decisions and negative firm performance, the extant literature focuses largely on the governance cost justification. Specifically, diversified firms incur increase governance cost, which detracts fr om their performance (i.e. excess returns). Vermeulen and Barkema (2002) in their study found that the speed of internationalization, spread of the geographic and product markets and irregularity in expansion patterns can further negatively moderate the di versification performance relationship.
27 Another study by Siegel, Omer, Rigsby & Theerathorn (1995) found that increased international diversification is also associated with increased total risk to the firm, resulting in a negative relationship between diversification and performance. Non Linear Impact of Firms Diversification on Their Financial Performance More recent research, considering both the costs and benefits of diversification, has proposed and empirically investigated complex, nonlinear relationships such as inverted U shaped, U shaped, or S shaped (Hitt et al 2006). These models posit that moderate levels of diversification is better than none (Palich et al., 2000) and the curvilinear models stem from an array of moderators, such as: geogr aphy (Li and Qian, 2005, Nachum, 2004, Vermeulenn and Barkema, 2002), munificence in the home country (Wan and Hoskisson, 2003), speed of internationalization, spread of product markets entered (Hitt, Hoskisson and Kim, 1997, Vermeulenn and Barkema, 2002) competitive intensity (Martin, Swaminathan and Mitchell, 1998, Sakar, Cavugsil and Aulakh, 1999), level of intangible assets (Lu and Beamish, 2004), firm size (Dragun, 2002, (Qian, Yang and Wang, 2003), R&D intensity (Kotabe, Srinivasan and Aulakh, 2002, Qian, Yang and Wang, 2003), marketing intensity (Kotabe, Srinivasan and Aulakh, 2002), level of international experience (Bloodgood, Sapienza and Almedia, 1996, Daily, Certo and Dalton, 2000, Ramaswamy, 1995), novelty and differentiability of the firms product offerings over those found in the host countries (Bloodgood, Sapienza and Almedia, 1996), commonalities between markets entered (Riahi Belkaoui, 1996), mode of entry and investment intensity upon entry (Gielens and Dekimpe, 2001).
28 U shaped impact of diversification on firm performance One rationale for a U shaped relationship considers both governance costs and learning effects. Initially the governance costs reduce firm performance, but over time, firms learn how to operate new formats and in new m arkets, reducing the governance costs, and performance begins to increase. However, as diversification continue to increase the governance costs increase to a point at which they can not be effectively dealt with and performance decreases. Another rationa le for this relationship stems from the use of predatory pricing by the incumbent entrant. Initial performance may be negative due to deliberate pricing strategies and intense competition from established firms in the host country. Subsequent performance i mproves as the entering firm mitigates losses by increasing prices and gains market share (Berger & Ofek 1995; Markham 1973) Firm performance may also subsequently improve yielding an inverted U shaped diversification performance relationship as: their product/concept gains acceptance in the host market(s) (Bloodgood, Sapienza and Almedia, 1996, Gielens and Dekimpe, 2001), they increase their market presence (i.e. retail network) or improve their image in the host country ( Gielens and Dekimpe, Kotabe, Sr inivasan and Aulakh, 2002, Sakar, Cavugsil and Aulakh, 1999) and when they improve their global market position and image (Barkema, Bell & Pennings 1996; Bloodgood, Sapienza and Almedia, 1996). Inverted U shaped impact of diversification on firm performance The theoretical rationale for the inverted U shaped relationship between diversification and performance stresses the initial positive effects of diversification up to a point at which the governance and agency costs arising from further diversification out weight the benefits.
29 Several studies have found support for this particular form diversification performance proposition (Hitt, Hoskisson and Kim, 1997; Gomes and Ramaswamy, 1999). Specifically, Hitt et al., (1997) finds that while firms can enjoy substantial growth opportunities in new markets (Markides, 1992, Hitt et al. 2006) the cost of coordination increases with increased operating scope. In addition, selective entry of markets based on market potential can also result in an inverted U shaped diversificationperformance relationship as firm experience diminishing excess returns and increased competition with each subsequent entry (Gielens and Dekimpe, 2007). In particular, Ramaswamy (1995) finds that diversificationperformance is positively r elated to the level of control and coordination within the firm and the ability to exercise a high level of control is decreasing with firm size ( Hoskisson & Hitt1988; Williamson 1967). The case for an inverted U shape diversificationperformance relations hip is further strengthened by combining the latter theory with the selective entry hypothesis as coordination and control become increasingly difficult in markets that are increasingly unrelated ( Mitra and Golder, 2001, Riahi Belkaoui, 1996, Sambharya 1996). Yet another rationale supporting an inverted U shaped diversificationperformance relationship is the diminishing excess returns that firms receive from new markets over time (Markides, 1992) S shaped impact of diversification on firm performance Fin ally, the rationale for the S shaped relationship suggests that some level of diversity is needed before learning effects take hold, but when they do, performance
30 increases to a point at which there are diminishing returns to additional learning and divers ification. CrossCountry Characteristics and their Impact on Successful Diversification A survey of the entry literature provides support for the selective entry hypothesis (Gielens and Dekimpe, 2007, Mitra and Golder, 2001). In particular, these studies suggest that firms stage their entry decisions to target enter markets that are most similar to their home market and/or markets they are familiar with ( RiahiBelkaoui, 1996) and subsequently use the knowledge acquired from existing markets to guide futur e entry decision (Mitra and Golder, 2001). Despite this background, Palich et al. (2000) noted that few studies spanning strategic management, economics and finance (with the exception of : Bettis, 1981; Christensen and Montgomery, 1981, Markides and Willi amson, 1994, Nayyar, 1992, Rumelt, 1974, 1982, Galai and Masulis, 1976, Higgins and Schall, 1975, Levy and Sarnat 1970, Lewellen 1971) have considered the moderating impact of related versus unrelated diversification on firm performance. Even fewer have t aken into account the level and type of diversification on the diversificationperformance relationship resulting in no clear consensus on how commonalities across diversified operations affect the diversificationperformance relationship (Paclich et al. 2000). On the other hand, several authors argue that differences in markets can provide firms with access to substantial growth opportunities in foreign countries (Hitt et al. 2006, Markides 1992) and can also accentuate their existing core competencies and gain unique knowledge (Hitt et al. 2006)
31 Of the studies that do account for similarities across countries, the findings are mixed. In particular, Riahi Belkaoui (1996) in his survey of 31 French multinational organizations finds that performance gains are greater with unrelated
32 CHAPTER 3 CONCEPTUAL FRAMEWORK AND HYPOTHESES My conceptual framework for examining the effects of international market and format diversification on retailer financial performance is illustrated in Figure 3 1. This framework suggests that the financial performance of retailers is affected by two characteristics of their international market portfolios the intensity of international market diversification and the cultural and economic dissimilarity (lack of sy nergy) of the international markets in which the retailer operates. Similarly, financial performance is also affected by the intensity of format diversification and the dissimilarity of the formats operated by the retailer. Finally, I propose an interac tion between international market and format diversification intensity. The dashed feedback paths in the framework represent the endogenous confounds in the diversification decision making. Specifically, the retailers financial performance affects its decisions to diversify as well as diversification affecting financial performance. In the following section, I discuss the requisite assumptions as well as several unique characteristics of this framework Next, I argue that the effects of diversification on financial performance are different for retailers and manufacturers. Then I review the costs and benefits of diversification as a background for the rationales I used to support the hypotheses illustrated in my framework. Conceptual Considerations and Assumptions of the Model As previously mentioned, this research is distinct from studies that focus on entry and performance as it abstracts from micro market considerations based on the assumption of aggregate rationality prevailing over stochastic vari ations in the long run and across the cross section. Specifically, this model assumes that retailers make
33 rational choices to maximize their expected return of investment when evaluating subsequent growth opportunities and their calculus involves a joint o ptimi zation across expected market and expected format potential ,[(format expertise, international expert ise, control variables)]itE and the corresponding risk based on their unique constraints denoted by ,(.)itC and information set at time t (denoted by t ). Specifically, I assume that retailers enter markets that are the most lucrative, given their existing market format portfolio and will continue to do so in descending order of the expected marginal products and cross products.1 ,1,t ,1Max [(format expertise, international ex pertise, control variables)|] s.t. (.) and (.)itit it itE CC Specifically, ret ailers are faced with the following constrained optimization: (3.1) Based on this assumption, the aggregate count variables, especially the interaction between the country and format counts are highly endogeneous while the diss imilarity measures can be considered predetermined conditional on the information set (Davidson & MacKinnon 1993) Diversification Implications for Retailers and Manufacturers Two critical differences between ret ailers and manufacturers that can affect the portfolio diversificationperformance relationship are: (1) the local nature of retailing and (2) the greater complexity of retailing compared to manufacturing operations (Dawson 1 This assumption is supported by research examining McDonalds expansion strategy; see Lafontaine & Leibsohn (2004)
34 1994 2007; Finn and Quinley 2010; Wrigley et al 2005). Even though retailers may have extensive global operations, their offering to consumer s is predominantly made through their stores and the market for these stores is local. Most consumers patronized store in close geographic proximity to where they work and live. Retailers, and other consumer service providers, need to satisfy the needs of local markets and compete against other retailers in these local markets. Thus, as they diversify internationally, retailers typically need to have operations (stores) in many more locations and be more sensitive to local cultural differences compared to manufacturers. For example, there is significant worldwide heterogeneity in preferences for food offered by supermarkets and restaurants, but little heterogeneity in the use of microprocessors. Because a key benefit offered by retailers is an assortment of products, retailers typically interact with more suppliers than manufacturing firms (Capar & K otabe 2003; Quinn & Fally 2010) Even a relatively small retail chain will operate stores in over 500 locations, deal with a thousand suppliers, and hundreds of thousands of customers (Alexander & Myers 2000) The complexity of managing this extensive network of customers, suppliers, and locations are dramatically heightened when retailers engage in international or format div ersification. In addition to the challenges of managing employees in a large number of remote locations, there are significant supply chain management challenges in getting the right products to the right places at the right time. Indeed, the difference in delivery requirements for manufacturing firms versus service firms (espec ially retail firms) gives rise to greater uncertainty for service firms ex ante, which may lead to different post entry outcomes (Lafontaine & Leibsohn 2004)
35 Benefits and Costs of Diversification The intensity of international market and format diversification, the extent to which a retailer operates in multiple markets with multiple formats, affects the retailers revenues, costs, and subsequent financial performance. In this section I briefly review the costs and benefits of diversification. Diversification benefits Some of the benefits of diversification are: economies of scale and scope, more effective asset management, opportunity to use market power, and risk reduction. International market and format diversification enable retailers to draw on scale economies to reduce costs. By expandi ng operations to new markets and using new formats, a retailers fixed costs are spread across the increased revenue opportunities arising from diversification (Caves, 1996). These scale and scope opportunities are enhanced for retailers possessing valuabl e firm specific assets, such as well known and highly regarded brand names, unique systems and processes, and managerial skills. Diversified retailers, in contrast to nondiversified retailers, can have greater access to both internal and external sources of financing (Lang & Stulz, 1994) and this expanded access provides diversified firms with privileged access to less costly financing (Froot et al., 1994; Lang et al., 1995). Diversification also reduces a retailers risk by spreading the effects of nons ystematic fluctuations over more business units (i.e. Berger & Ofek 1995; Kim et al. 1993;). This risk reduction produces a feedback effect affording retailers with reduced risks to realize better external financing terms (Duffie & Singleton 1997) and im proved borrowing ability (Shleifer & Vishny 1992).
36 The ability to cross subsidize internal resources provides yet another benefit to diversified retailers: market power. Larger diversified retailers can negotiate better terms with suppliers and can even demand exclusive relationships, engage in predatory pricing against competitors and may even deter entry by threatening to respond with a pricing war (Berger & Ofek 1995; Markham 1973). Diversification cost Some of the costs of diversification are due to increased management complexity, agency issues, and familiarity. Perhaps the primary cost related to diversification is the increased complexity of managing diversified firms. As retailers diversify, their management hierarchy expands and they encounter greater difficulties in disseminating information to business units, resulting in transmission and coordination inefficiencies and the loss or distortion of information (i.e. Hoskisson and Hitt 1988; Williamson 1964). Furthermore, the larger hierarchical st ructure of diversified retailers is conducive to employee shirking; thereby further incurring cost associated with either heightened monitoring or decreased labor productivity (Calvo and Wellisz 1978). Williamson (1964) and Jensen (1986) discuss the agency problems that arise from hierarchical versus market governance. Specifically, managers have an incentive to over diversify since their compensation is often tied to the revenues they generate (Murphy 1986). In addition, revenue growth creates new positions to which these managers may also be promoted. This tendency to over diversify is further exacerbated in mature firms with substantial free cash flow (Jensen 1986; Mueller 1972). In addition to these increased internal governance costs, diversification can lead to inefficiencies arising from external challenges of operating a new format or being in a new international market (i.e. Hymer 1976, Zaheer 1995). Some of these
37 disadvantages, related to diversification, are due to inefficiencies resulting from the lack of market reputation (Barkema, Bell and Pennings 1996); limited knowledge about the new environment, which leads to lower efficiencies than native firms (Hyman 1976); and increased uncertainty from having to operate in a complex and unfamiliar environment, characterized by a different set of economic, political and legal factors (Sambharya 1996). International Market and Format Diversification and Retailer Financial Performance In this section, I discuss how these benefits and costs associated wi th diversification affect the financial performance of retailers. International Market Diversification I ntensity Early research on international market diversification focused primarily on either the benefits or costs and proposed either positive or negative linear relationships between international market diversification and the performance of manufacturers. Earlier research focused on the benefits or costs of diversification and proposed either a positive or negative linear relationship between international and product/industry diversification and performance. More recent research, considering both the costs and benefits of international market diversification, has proposed and empirically investigated complex, nonlinear relationships such as inverted U shaped, U shaped, or S shaped (Hitt et al 2006). The theoretical rationale for the inverted U shaped relationship between international market diversification and financial p erformance stresses the initial positive effects of diversification up to a point at which the governance and agency costs arising from further diversification out weight the benefits. The rationale for a U shaped
38 relationship considers both governance costs and learning effects. Initially the governance costs reduce firm perfor mance, but over time, firms learn how to operate new formats and in new markets, reducing the governance costs, and performance begins to increase. Finally, the rationale for the S shaped relationship suggests that some level of diversity is needed before learning effects take hold, but when they do, performance increases to a point at which there are diminishing returns to additional learning and diversification. As discussed previously, most of the prior empirical research on international diversificat ion is based on samples of manufacturing firms, but there are significantly greater governance costs (managing many remote locations and suppliers) associated with the international diversification of retailers compared to manufacturers as discussed previously. Thus, I propose that retailer with a modest number of international markets in their portfolio will have the highest governance costs as a percent of sales and the lowest financial performance. Retailers operating in a small number of international markets will have lower relative costs and higher performance because they have sufficient slack to manage the governance issues in a limited set of markets. The performances of retailers with many international markets in their portfolios have lower relative costs and higher performance because they have learned to effectively manage many remote locations. Thus, I propose that: H1 : The relationship between the intensity of international market diversification and retailer financial performance is U shaped. Format Diversification I ntensity. Using a similar logic, due to the complexity of retail operations, the governance costs incurred by launching a new format are substantial. However, as retailers
39 develop a larger portfolio of formats, they learn how to manage the multiple formats for efficiently and extract the economies of scale and scope from operating many formats. In addition, with many format, retailers are better able to match the format with the needs of different markets. Thus, H2 : The relati onship between the intensity of format diversification and retailer financial performance is U shaped. International Market F or mat Diversification Interaction Many manufacturing firms, as they diversify internationally, also diversify their product portfolios as well. Prior research suggests that product diversification (the manufacturing equivalent of format diversification for retailers) moderates the international diversificationperformance relationship (Hitt, Hoskinsson, & Kim, 1997; Kim, Hwang.& Bur gers1989). However, the empirical evidence on the moderating effects between product and international diversification is mixed (Chang and Wang 2007). Some research has reported a positive moderating effect while other studies have found a negative moderat ing effect. The rationale for the positive moderating effect is that the experience gained from dealing with a diversified product portfolio enables manufacturers to better deal with the complexities associated with international market diversification. In addition, the access to diversified portfolio of products enables firms to better tailor their offering to the unique needs of different countries. The rationale for the negative effect is that product diversification, when coupled with international div ersification, dramatically increases the complexity of the management tasks and reduces efficiency, and further decreases performance. In light of the heightened demand and supply side complexity of retailing compared with manufacturing, I propose that:
40 H 3 : The intensity of international market and forma t diversification interact with each other producing a negative effect on retailer financial performance. Effects of the Characteristics of the Portfolio of International Market and Format on Retailer Financial Performance The extant international diversification research discussed in the previous sections simply examines the effects of the intensity, or degree, of international market and format diversifciation on performance but does not consider the char acteristics of the diversified portfolio. However, a number of researchers have suggested that the degree to which the products [or in the case of retailers, the formats] and international markets are related affects the diversificationperformance relationship (i.e. Cavusgil 1983) For example, Rumelt (1982), in his classic diversification research study, found differences in financial performance between firms pursuing related and unrelated product diversification strategies. Rumelt speculated that the performance premium afforded to firms with related product diversity stems from their ability to exploit core competencies and create unique and defensible product market positions, a view echoed by marketing scholars (Davison 1983, Jaworski & Kohli, 1993; Hunt & Morgan, 1995). In this research, I examine the following three factors affecting the degree to which the international markets and formats in a retailers diversification portfolio are unrelated: (1) the cultural dissimilarity of the countries in the retailers portfolio, (2) the economic dissimilarity of the countries in the retailers portfolio, and (3) the operational dissimilarity of the formats in the retailers portfolio. Cultural D issimilarity Increased challenges arise when a retailers portfolio consists of international markets with dissimilar cultures because the retailer must develop an understanding of
41 a number of different norms and values for both its customers and employees. To effectively satisfy the needs of customers and employees, the retailer has to develop a wide variety of merchandise assortment, formats, and human resource policies and procedures. Managers in culturally diverse countries are also less able to take advantage of economies of scale and scope involving joint purchasing, advertising, brand building programs, distribution, and systems for workload planning, visual merchandising, and employee performance appraisal. Thus cultural dissimilarity in a retailers portfolio of countries reduces the retailers ability to develop scale economies, exploit synergies, and realize the benefits of increased revenues from diversification (Palich & Gomez Mejia, 1999, Gomez Mejia & Palich 1997). Thus, H4 : The degree of cultural dissimilarity within a retailers international ma rket por tfolio decreases the retailers financial market performance of. Economic D issimilarity Increased operational costs also arise when a retailer has a portfolio of countries at different stages of economic development. For example, a key competitive asset of Wal Marts is its systems and skills in supply chain management. However, this capability was developed in develop in the United States with its well developed infrastructure. Wal Mart may not be able to exploit this asset when operating in less developed countries with limited infrastructure. Prior research supports this contention that economic dissimilarity can impact on firms performance (Baffoe Bonnie & Khayum 2003; Beckerman 1956; Davidson 1983; Dunning 1973) H5 : The degree of dissimilarity in the economic development of the countries in diversification the financial market performance of retailers.
4 2 Format D issimilarity Different skills and resources are required to effectively operate different formats. For example, brand building, merchandise select ion, pricing, and store design and visual merchandise skills, skills related to generating demand, are needed to effectively operate apparel specialty stores that typically target fashionoriented customers. On the other hand, cost control factors such as supply chain management are more important in the effective management of discount store that typically target valueconscious customers. By operating different formats, retailers can provide an attractive offering to different market segments but their abilities to exploit operational efficiencies and scale economies decreases. However, the benefit of operating a number of different formats is moderated by the degree to which the formats require similar operating skills. Thus, one would expect that a ret ailer operating discount stores and warehouse clubs, two formats focusing on cost control skills would have better performance than a retailer operating apparel specialty stores and warehouse clubs. H6 : The dissimilarity in the operational characteristics of different formats within a portfolio reduces the financial market performance of retailers.
43 Figure 31. Conceptual f ramework Financial Performance of RetailerIntensity of International Market Diversity in Retailer s Portfolio Intensity of Format Diversity in Retailer s Portfolio Dissimilarity of Cultural and Economic Market Conditions in Retailer s Portfolio Dissimilarity of Formats in Retailer s Portfolio
44 CHAPTER 4 METHOD In this section, I describe the sample, measures for the constructs, sources of the d ata, and econometric model used to estimate the effects of the degree of international and format diversity and portfolio characteristics on financial performance. Sample The sample of retailers used in this study was the worlds largest global retaile rs as identified in the annual Deloitte Touche Tohmatsus Global Powers of Retailing reports published from 2002 to 2007 (Deloitte 2002, 2003, 2004, 2005, 2006, 2007). These reports, also published as addendums to the January issues of Stores magazine, provided data on the 200 largest retailers from 2002 to 2004 and on the 250 largest retailers from 2005 to 2007. However, some retailers in these reports were not be included because financial market performance measures of retail performance were not available (i.e. they were not publically traded or were part of larger diversified/holding firms), In addition, I eliminated retailers that were either automobile, food, and hotel franchisors or franchisees. These restrictions, along with firms entering and exit ing the largest retail list, resulted in data set composed of 762 annual observations for 170 retail firms. Measures Financial Market P erformance As noted by Hoskisson, Hitt, Johnson and Moesel (1993), market based financial measures of firm performance are superior to accounting measures because they are more closely related to the firms objective of maximizing stockholder and less
45 sus ceptible to managerial discretion (Barney 1997), making them more suitable for studying diversificationperformance linkages (Palich et al., 2000). In this study, I used Tobins Q as the measure of financial market performance. Tobins Q is the ratio of a firms market value to the cost of replicating the firms assets. It is a proxy for the average return on a firm's c apital anticipated by the market. Hence, it is a preferred measure of firm performance because it is forward looking and provides an unbiased comparison across firms (Wernerfeldt and Montgomery 1988, Srinivasan and Hanssens 2009). Particularly, under the assumption of financial market efficiency, changes in future earnings and costs will be capitalized into share prices. Thus, if diversification is beneficial, past expansion should be related to higher share value (i.e. high Q), no t high current returns to shareholders. Further, because investors require a return related to the riskiness of an asset, comparing returns without adjusting for riskiness yields biased results. In contrast to stock return comparisons or accounting perfor mance measures, no risk adjustment is required to compare Q across firms. Tobins Q was operationalized as the retailers market capitalization at the end of the fiscal year divided by firms total assets. Diversification I ntensity In prior research, the most commonly used measures of international diversification are the firms percentage of overall sales, profit, assets, or employees in nondomestic markets. However, these measures have been criticized for not capturing the actual nature of heterogenei ty in international diversification (Vachani 1991). Firms can have the same level of diversity if their percent of international sales are the same but they operate operating in few or many countries.
46 In a similar manner, product diversity, the manufact uring industrys parallel to retail format diversity, is typically operationalized as the number of SIC categories in which the firm operates or an entropy measure based on the percentage of sales or profit generated in each SIC category. The SIC categories are particularly poor at characterizing the operating characteristics of services firms, especially the operating characteristics of the various operating formats. I used a simply count of the number of countries that retailer have operated in to measur e international diversity and a simple count of the number of formats to measure format diversity 1Format D issimilarity These counts were based on data provided in the Global Powers in Retailing reports (Deloitte 2002 2007). These count measures provide a richer portrayal of the level of diversification than a simple percent of international/product category sales or profits. I used expert judges to assess the similarity of the retail formats in each retailers portfolio. Ten experts (academic scholars senior retail executives and consultant to the retail industry) were asked to indicate the degree to which each retail format pair were similar in terms of their operations using a sevenpoint Likert scale [very dissimilar = 1 and very similar = 7]. The experts made these judgments for the 78pairs of formats, derived from the following 13 retail formats: apparel specialty, catalogue or electronic retailer, 1 Another commonly used measure of diversification is an entropy y measure because it considers both the number of countries in which a firm operates in and the proportion of total sales or profits in each country (see Palepu 1985). The entropy measure is calculated by summing the percentage of sales in each countries weighed by the natural log of one divided by the percentage of sales in the country. I was not able to use this measure of diversification because firm sales or profit by country and by retail format are not readily available using secondary sources.
47 convenience store, departmental store, drug store, electronic specialty, hard goods discounter, hypermarket, home improvement, nonapparel/other specialty, soft goods discounter, supermarket and warehouse club. Eight experts responded and the reliability of their responses was acceptable [Alpha = 0.81]. The average of the mean expert rating for each pair in a retailers portfolio was used to create a format dissimilarity index. A copy of the survey questionnaire is included in Appendix D. Cultural D issimilarity Our measure of cultural dissimilarity of a retailers portfolio of countries was based on t he four cultural dimensions developed by Hofstede (1984) : individualism masculinity, power distance, and uncertainty avoidance. The cross cultural scores, available for 79 countries, have been widely used in studies of consumer innovation (Steenkamp, ter Hofstede & Wedel 1999) consumer taste (Craig, Greene & Douglas 2005) decision marking (Roth 1995) ethics (Franke & Nadler 2008; Waldman, de Luque, Washburn, House, Adetoun, Barrasa, Bobina, Bodur, Chen, Debbarma, Dorfman, Dzuvichu, E vcimen, Fu, Grachev, Duarte, Gupta, Den Hartog, de Hoogh, Howell, Jone, Kabasakal, Konrad, Koopman, Lang, Lin, Liu, Martinez, Munley, Papalexandris, Peng, Prieto, Quigley, Rajasekar, Rodriguez, Steyrer, Tanure, Thierry, Thomas, van den Berg & Wilderom 2006) managerial functions (Costigan, Insinga, Berman, Ilter, Kranas & Kureshov 2006; Smith, Dugan & Trompenaars 1996; van Oudenhoven, Mechelse & de Dreu 1998) negotiations (Campbell, Graham, Jolibert & Meissner 1988; Graham 1985) and retail image (Bianchi & Ostale 2006; O'Grady & Lane 1997) To calculate cultural dissi milarity, I created a composite index that is a variant of the index used by Kogut and Singh (1988) I first divided the country rating for each of
48 the four Hoefstede dimensions by its standard deviation. Then I calculated the average dimension dissimilarity between all countries in the retailers portfolio. Next, I took the average of the squared average dimension dissimilarity across the four dimensions. In the few cases, where cultural distances scores were not available, I either made the appropriate substitutions or dropped the observation point altogether. For instance, cultural distances for former and present territories or colonies were substituted with scores of the governing country. Thus, this index is: 22 ,,Cultural dissimilarity = ((()/(1)/2!)/)/ 4ikij i kj iHHnn where: ijH denotes the value of the ith Hofstede dimens ion for the jth country. i = 1 .. 4 denotes the ith original Hosftede dimension; and, k, j = 1 .. n index the country pairs for which the firm has operations in; n denotes the number of countries in the firms portfolio for that particular year. Eco nomic D issimilarity This index was developed using a similar approach to the development of the cultural dissimilarity index using factors assessing the economic climate and technological development of the countries rather than Hofstedes culture scores. As noted earlier, variations in economic orientation across countries has been shown to affect the internationalization decisions of firms and is also likely to affect a firms performance. To account for this effect, I obtained various economic measur es that would later comprise the economic distance construct and instrumental variables. A total of 86 relevant World Bank Indicators [WBI] time series for the dynamic variables [for the years 1997 to 2006]
49 and 13 relevant Central Intelligence Agencys World Factbook 2007 entries for the static [or reasonably static over the analysis period] variables were obtained for the relevant countries and their corresponding descriptions can be found in Appendix D The factors scores for four dimensions (development status and infrastructure technological advancement and efficiency, international orientation and market potential) were based on an exploratory factor analysis of 62 economic indicators from the World Bank (WDI 2007). Factor analysis results are provided in Appendix D and the Scree plot is shown below. Summary Statistics The means, standard deviations, and correlation matrix for the variables are shown in Table 11 and 12 respectively The sources of the data used in the study are in the Appendix A Estim ation Endogeniety I ssue A critical issue that has not been addressed in most diversificationperformance studies is the endogenous link between a firms decision to diversify its portfolio of countries and formats and its performance. While diversification can improve firm performance, improved performance can also generate more resources that motivate and enable firms to expand into new markets and launch new formats. Li (2007) emphasizes that the causality of the international diversificationperformance relationship remains ambiguous. He refers to Tallman and Li (1996, p.181) who say evidence suggests that firms with significant performance advantages tend to be multinationals but that the direction of the causal relationship may well run from high le vels of firmspecific capabilities to higher performance to international diversification,
50 rather than from capabilities to multi nationality to higher performance. Li (2007) then goes on to say that there are virtually no studies that have examined the two way relationship between international diversification and firm performance simultaneously. An assumption necessary for regular estimation is that ,0 E T i,titxu This condition, referred to as the sequential conditional exogeneity condition, is likely to have been violated based on theories on how firms sequentially choose to enter markets (Mitra & Golder 2002; Montgomery & Wernerfelt 1988; Wernerfelt & Montgomery 1988) survivor bias (Denrell 2003) and other heterogeneous firm characteristics (e.g. size, structure, industry) affecting diversification decisions (Campa & Kedia 2002) Method Simultaneous equations is an econometric technique that is used to estimate multiple interdependent variables of interest in a series of equations involving the interdependent variables of interest and other exogenous variables and includes as examples: substitution into reduced form equations, IV estimation, 2SLS, 3SLS and seemingly unrelated regression models (Wikipedia 2006) The latter is a technique for enhancing the efficiency of regression systems with correlated error terms (Wikipedia 2008b) and the remainder, except for substitution into reduced form, are manifestations of IV type techniques. In particular, 3SLS is 2SLS with the addition of an extra step that enhances efficiency thru estimation of the error components using the residuals from the second stage (Sola 2004) The Tobit model is used for dealing with latent variables that are prone to inconsist ent estimates when using traditional estimation techniques due to issues that arise from censoring (Wikipedia 2008c) Another model, the Heckman selection model,
51 does not appropriately deal with the issue at hand. Specifically, the Heckman selection model is used to estimate parameters when the sample is prone to self selection (i.e. estimating wage relationships when wages are only avail able for working individuals) and corrects for this self selection by the use of a Probit model to estimate the probability of being employed and subsequently using this result to weight the ensuing regression/ estimation (Hopkins 2005; Wikipedia 2008a) Hen ce, while this method corrects for the sample selection problem, it does not correct for the inherent endogeneity between performance and level of diversification (the endogeneous relationship between the dependent and independent variables). It may be beneficial to implement the Heckman correction with another technique that corrects for the endogenous relationship, thereby accounting for 1) the self selecting sample of companies that diversify and 2) the endogeneous relationship amongst those who diversif y. However, the first issue may very well be equally addressed by the use of dummy variables to account for heterogeneous sample characteristics and estimating separate equations for the different sample groups. Perhaps the most well known techniques used in econometrics for the correction of endogeneous relationships between the dependent and independent variables are those of the IV type, specifically: 2SLS and 3SLS. The use of such techniques will help in isolating the exact performance premium attributed solely to diversification levels and NOT to a mix of diversification levels and other unobserved correlated constructs (e.g. size, managerial efficiency, TMT diversity, etc.) that seems to plague many of the existing diversification studies. It is perhaps for this reason that researchers have not yet reached a consensus on the exact diversification relationship that seems to change
52 with each passing sample or the varying relationships across different firm characteristics, such as size (see: Dragun 2002). A review of the literature revealed that a handful of the many existing studies used the following endogeneity correction method(s): 2SLS (Campa & Kedia 2002) and Heckman Correction (Campa & Kedia 2002; Graham, Lemmon & Wolf 2002; Hyland & Diltz 2002; Lang & Stulz 1994) However, a subset of these studies look at a totally different issue: the self selection into diversification rather the level of diversification (Campa & Kedia 2002; Hyland & Diltz 2002) Hence, while it is appropriate for those studies to correct for endogeneity using a Heckman approach, it is not appropriate in our case to solely rely on that method. Substantive Issue The key issue is to derive nonendogeneous country and format count variables for use in predicting the relationship between level of diversification and performance for a sample of retailers. Focusing on the first stage analysis, the analysis in which we use various instruments to tease out the bidirectional effect from the respective count variables, we require that our instruments be correlated with the problematic variable ( the count variables) but not correlated with the error term (the residual after regressing performance on level of diversification, distance moderators and their interaction terms). This implies that the instruments need to be uncorrelated not only with the current periods firm performance construct but also that of previous lags. This is due to the possibility that firms make entry/ expansion decisions in periods of good performance that manifest in future periods but firms can also abort such plans inter mittently if performance starts to head south (e.g. Toyota delays plans to open additional plants in the US).
53 The following is a suggested list of instruments types, listed in an order that depicts decreasing correlation with the error terms (i.e. ranging from better to worse instruments). Historic variables (e.g. firm founding year, continent firm was originally established in, etc.) Relatively static home country specific characteristic variables (e.g. uncertainty index, social index, etc.) Relatively dyn amic home country variables (e.g. infrastructure expenditure, tourism rates, etc) Industry specific variables (e.g. competition index for a certain industry) Firm specific variables (e.g. firm size, alternative performance index, etc.) First Stage E stimati on The orthogonality conditions needed for least squares to produce unbiased estimates was potentially violated for a variety of reasons. For instance, correlations could arise due to heterogeneous firm characteristics (e.g. size, structure, industry) of h ighly versus lowly diversified retailers (Campa & Kedia 2002) sequential entry decisions (Mitra & Golder 2002; Montgomery & Wernerfelt 1988; Wernerfelt & Montgomery 1988) and survivor bias (Denrell 2003) To correct for the potential orthogonality violation and endogeneity bias, I used a two stage estimation model with panel autoregressive disturbance. However, standar d two stage estimation techniques were not suitable for my model since both my first stage dependent variables were count variables. Standard Instrumental Variables method utilize normal distribution estimation techniques for the first stage and is hence
54 n ot appropriate for count variables that follow either a Poisson or Negative Binomial distribution. My modified estimation technique involved first regressing two sets of instrumental variables (IVs) on each of the two decision variables: retailers country count and format count. Intuitively, this step can be thought of as a screen for the endogenous confounds, that drops out with the other unexplainable confounds in the error term from the first stage. The first stage regression is aimed at teasing out t he bi directional causality between firm performance and the explanatory variables. I assume that endogeneous effects lie predominantly in the aggregate count variables for countries and formats based on conjectures and some evidence that firms diversify i n a strategic manner (Mitra & Golder 2002; Montgomery & Wernerfelt 1988; Wernerfelt & Montgomery 1988) Appropriate instruments should correlate well with the count variables but not with performance related constructs (Davidson & MacKinnon 1993; Wooldridge 2002) ; hence, we chose to use 9 firm specific variable and 11 threeyear lagged country WBI specific variables since evidence suggest that it takes approximately that amount of lag for international responses (Flowers 1976) and in implementing managerial actions (Williams, Hoffman & Lamont 1995) Theoretical and empirical rationales drove the selection of IVs. T he IVs used in the first stage were variables that depicted the characteristics of the retailers home country influencing diversification interest such as infrastructure, tourism, industry competition, and continent and characteristics of the retailer such as founding date, sales, and number of employees.
55 These IVs used in the first stage to estimate the number of countries and formats in a retailers portfolio were lagged by three years to reflect the retailers planning horizon. Research suggests that managers often make involved entry/expansion decisions for the mid to distant future (Alexander & Myers 2000; Dawson 2001; Rugman & Girod 2003) Justification for my use of a threeyear lag can be found in cases involving decisions on f oreign direct investment (Flowers 1976), by examining the substantially different firm characteristics (R&D to Asset ratio) evident threeto four years preceding firms diversification attempt(s) (Hyland and Diltz 2002) and even in micro managerial issues such as the persistence of managerial employee relational conflict (Klass and Denisi 1989). The sources of the data for the IVs is described in Appendix A Next, I transformed and recoded the data when necessary, and windsorized the data symmetrically at the 5 % end points. I then estimated the first stage models with two negative binomial regression s. SecondStage E stimation The predicted values for the decision variables retailers country counts and format counts from the first stage model w ere then transformed in accordance with main, interactive, and nonlinear terms of my model used to estimate the factors I hypothesize to explain retailers financial performance (Tobins Q). Tobins Qi,t = ( 4 1)
56 + economic dissimilarity I estimated the model described by Equation 41 using Stochastic Moderated Regression (SMR), a method proposed by Gatignon and Vosgerau (2006) for handling collinear interaction terms. SMR allows for random moderating relationships, thereby enabling other variables [other than the hypothesized moderators] to influence the stochastic relationship. This is particularly useful when the exact nature of moderating relationship is: (1) random, (2) not fully understood or (3) misspecified. Additional benefits of SMR include: achieving more efficient estimates in t he presence of multicolinearity, (2) determining the direction of the moderating causality and (3) not detecting a spurious moderating effect when none exists particularly when the main effects are moderately high to highly correlated. The latter is relevant sinc e correlation between the format distance variable and format count variable is 0.747 and the correlation between physical economic and cultural distance and the country count variable are 0.5881, 0.5869 respectively. Using SMR, estimation of Equation 4 1 is done by estimation of Equation 4 2 Equation 4 4 jointly. However, there are still two issues remaining before we can proceed further. ,,0,1 ,,2 ,,PERF FORMAT COUNTCOUNTRY COUNTitiit itit ititu ( 4 2) ,,1,, ,,2 ,,AVG FORMAT DISTCOUNTRY COUNTtii iti itit ( 4 3)
57 ,,2,,,1 ,,,2 ,3 ,,2 AVG CULTURALAVG ECONOMIC FORMAT COUNTtiit itit it i it ( 4 4) First, while cr oss sectional time series data provides for a richer and more in depth analysis (Gujarati 2003) it also requires additional assumptions about the error structure. Specifically, firms have varying resources and characteristics, which will likely result in heterogeneous outcomes and error terms. In addition, these firms are also likely to demonstrate temporal dependence and variability. Since subsequent tests indicated no presence of the latter effect, we decided to use a heteroskedastic and robust covariance matrix to mitigate the former concern and to attain more efficient estimators. Thus, I specified a heteroskedastic and robust covariance estimator. Figure 41. Scree plot from factor analysis of host country economic variables
58 Table 41. Summary statistics for main variables MEAN S.D. MIN MAX TOBIN'S Q RATIO 1.01 0.85 0.15 3.33 COUNTRY COUNT 5.16 6.49 1 23 FO RMAT COUNT 2.36 1.64 1 6 AVG. CULTURAL DISSIMILARITY 0.71 0.94 0 5.93 AVG. ECONOMIC DISSIMILARITY 5.29 21.99 0 378.58 AVG. PHYSICAL DISTANCE 2468.11 2905 0 11426.71 AVG. FORMAT DISSIMILARITY 2.57 2.42 0 6.14 Table 42. Correlation matrix for main var iables TOBIN'S Q RATIO COUNTRY COUNT FORMAT COUNT AVG. CULTURAL DISSIMILARITY AVG. ECONOMIC DISSIMILARITY AVG. PHYSICAL DISTANCE AVG. FORMAT DISSIMILARITY TOBIN'S Q RATIO 1 COUNTRY COUNT 0.01 1 FORMAT COUNT 0.22 0.2 1 A VG. CULTURAL DISSIMILARITY 0.09 0.44 0.26 1 AVG. ECONOMIC DISSIMILARITY 0.13 0.08 0.16 0.43 1 AVG. PHYSICAL DISTANCE 0.06 0.59 0.13 0.53 0.13 1 AVG. FORMAT DISSIMILARITY 0.27 0.14 0.75 0.2 0.12 0.06 1
59 CHAPTER 5 RESULTS First Stage Results The estimated coefficients for the first stage are shown in Table 51 The DurbinWu Hausman test was conducted to determine the potential impact of endogeneity in a regression estimated via instrumental variables (IV). The null hypothesis that an o rdinary least squares (OLS) estimator of the same equation would yield consistent estimates was rejected for the the country and format equations ( 2 1Pr (13.03)0.0003 ob and 2 1Pr (3.33)0.0679 ob respectively. Thus ignoring endogeneity among the regr essors would have had deleterious effects on OLS estimates. While the estimated coefficients for the IVs were used to correct for the endogeneity bias, there some interesting results supporting my a priori hypotheses about the effects of the IVs. As expected, the size of the retail firm as measured in annual sales had a significant positive effect on both the number of countries in a retailers portfolio as well as the number of format utilized. As retailers grew larger they sought growth opportunities outside their home country and from operating different formats. In addition, a number of retailers home country characteristics affected their interest in diversifying their portfolio of countries and formats. The ratio of tourist arrivals to departure was a surrogate measure of the brand image of the countries. Retailers with headquarters in countries with a strong brand image were prone to exploit this advantage by entering more countries.
60 Retailers from home countries with high GDP per capita and popu lation were in a better position to expand internationally and operate different formats since they were more likely to have greater access to the resources needed to drive such expansion. On the other hand, these retailers also benefited from being situated in countries with large domestic markets and thus were more likely to have significant domestic growth opportunities. my results suggest that retailers in home countries with high GDP per capita and populations focused on their growth opportunities in their home countries. The prevalence of computers in a retailers home country was a surrogate measure of home country technology and infrastructure development. The significant positive coefficient indicated that technology and infrastructure depth in t he retailers home country were enablers of international expansion and format diversification. Retailers that were headquartered in the European or North American continents (compared to those from other continents) and operated specialty stores formats o perates diversified into more countries. Perhaps the intensity of retail competition and scale economies due to increasing consolidation resulted in North American and European retailers having a greater tendency to diversify internationally. In addition, the greater international diversity of European retailers may be due to the limited size of their domestic markets, greater regulation in their home countries, and/or their proximity to other countries for potential expansion. Specialty store retailers were more prone to expand internationally while supermarket retailers were less prone. These results may be due to the substantial heterogeneity of food taste globally and relative homogeneity for merchandise sold in by specialty store retailers.
61 The per iod from 1961 to 1985 is characterized as the era of internationalization (Nelson & Wright 1992; O'Rourke, Taylor & Williamson 1996) Thus the business climate might have stimulated retailers operating prior to 1961 that had developed the operational skills to internationalize. Test of Hypotheses (SecondStage R esults) The second stage results outlining the effects of international and format diversification and portfolio characteristics on financial market performance are shown in Table 52 International Market and Formats I n tensity of Financial Market P erforma nce The number of countries and formats in a retailers portfolio had a significant effect on the retailers financial performance (Tobins Q). The estimated coefficient for the number of countries in a retailers portfolio was positive and significant ( p<.01), while the estimated coefficient for number of formats in a retailers portfolio was negative and .745, p<.01). These simple main effects suggested that retailers operating in more countries and retailer s operating fewer formats have significant and positive. The significance of these nonlinear terms and signs of the coefficients indicates that the relationship between the number of countries and formats in a retailers portfolio and its financial market performance is U shaped supporting H1 and H2. In addition, there was a signi 0.731, p<.01) for the interactions of number of formats and number of countries supporting H3. This interaction coupled with significant squared terms suggested a complex relationship between retailer financial performance and the international market and format
62 diversifications of their portfolios. The nature of these relationships is illustrated in Figures 5 1 and 52 In Figure 51 the estimated Tobins Q as a function of international market was plotted for r etailers with the average levels of cultural and economic dissimilarity at several levels of format diversification. At low levels of format diversification (number of formats less than four), international diversification had a positive relationships wi t h financial market performance (i.e. Tobins Q ) But at higher levels of format diversification (number of format great that 4), the relationship between international diversification and financial market performance was just the opposite. The estimated Tobins Q decreased as the retailer operates in more countries. At modest levels of format diversification (num ber of countries) was weaker. Note that the relationship between international market diversification was basically U shaped, but only a portion the U is shown with the range of the data 1 to 30 countries because the number of formats shifted the U horizontally. The more countries in which a retailer operated, the higher was its estimated Figure 5 2 illustrates the effects on retailers Tobins Q as a function of format diversification. The number of formats in the retailer operates was plotted for retailers with the average levels of cultural and economic dissimilarity at several levels of international market diversification. In this cas e, the relationship between Tobins Q and the number of formats operated was consistently U shaped with the range of data, however, the minimum level of Tobins Q increased as the number of countries in which the retail operates increased. Note that when a retailer operates one format, its financial performance basically increased as the retailers contains more countries. On the other
63 hand, when a retailer operates eight formats, its financial performance decreased when its portfolio included more countri es. Cultural, Economic, and Format D iver sity and Financial Performance The results in Table 52 also provided support of propositions H4 and H5. Both the 0.03302, p<.01) dissimilarity of the countries in a retailers portfolio had a significant effect on the retailers Tobins Q. Figure 51 illustrates the effect of these market dissimilarities on the estimated Tobins Q. In Figure 53 the estimated Tobins Q is plotted as a function of the number of countries for three levels of cultural and economic dissimilarity retailers with high market dissimilarity (75th percentile for bother cultural and economic dissimilarity), median levels of market dissimilarity (50th percentile), and low levels of market di ssimilarity (25th percentile). Other variables affecting Tobins Q (format dissimilarity and number of formats) were set at the mean levels for the sample. Figure 5 3 illustrates that the estimated Tobins is higher for retailers that operate a portfolio involving similar versus dissimilar markets. Finally, my analysis indicates that format dissimilarities in a retailer portfolio had a 0.107, p<.01) support in g proposition H6. In Figure 54 the estimated Tobins Q was plotted as a function of the number of format for three levels of format dissimilarity retailers with high format dissimilarity (75th percentile), median levels of format dissimilarity (50th percentile), and low lev els of format dissimilarity (25 percentile). Other variables affecting Tobins Q (cultural and economic dissimilarity and number of countries) were set at the mean levels for the sample. Figure 5 5 illustrates that the estimated Tobins
64 was high er for retailers that operated a portfolio involving similar versus dissimilar formats. Table 51 Estimated results: instruments affecting diversification decisions COUNTRY COUNT FORMAT COUNT LN (SALES) .1853 *** .1565 *** PERCENTAGE OF POPULATION LIVING IN URBAN SETTINGS .0091** .00245 POPULATION DENSITY (PPL / SQ KM) .00042 *** .00018 *** LN (POPULATION) .2593 *** LN (GDP PER CAPITA) .4945 *** .4724 *** COMPUTER PREVELANCE (PER 100 PEOPLE) .0112 *** .00594 ** TOURIST ARRIVAL DEPARTURE RATION .0 2622 .01256 FOUNDED PRIOR TO 1961? (F=0, T=1) .4344 *** .1078 ** EUROPEAN? (F=0, T=1) 1.994 *** .578 *** N. AMERICAN? (F=0, T=1) .308 ** .1286 SPECIALTY TYPE RETAILER? (F=0, T=1) .5368 *** SUPERMARKET TYPE RETAILER? (F=0, T=1) .3115 *** CONSTANT 3.225 *** 1.572 ** LN(ALPHA) CONSTANT .675 *** N 822 822 LN LIKELIHOOD 2060 1368 CHI2 960 499.6 AIC 4149 2756 BIC 4215 2803 p < .1, ** p < .05, *** p < .01
65 Table 52 Effect of country and format portfolio constituents on Tobins Q Tobin s Q Estimated Coefficient Standardized Coefficient Number of Countries 0.0746*** 0.397*** Number of Formats 0.768*** 0.745*** Number of Countries Squared 0.00205*** 0.366*** Number of Formats Squared 0.125*** 0.761*** Number of Countries x Number of Formats 0.0307*** 0.731*** Cultural Dissimilarity 0.0116* 0.0107* Economic Dissimilarity 0.00159*** 0.0332*** Format Dissimilarity 0.044*** 0.107*** Constant 2.089*** N 737 Chi Square (8) 2270.98 PseudoR2 0 .127 p< .10, ** p<.05, *** p<.01
66 Figure 51. International diversification and retailers Tobins Q Figure 5 2. Format diversification and retailers Tobins Q
67 A B Figure 53 Impact of varying levels of market dissimilarity on Tobins Q; A) Across multiple countries.; B) Across multiple formats.
68 A B Figure 54 Impact of varying levels of format dissimilarity on Tobins Q; A) Across multiple countries.; B) Across multiple formats.
69 CHAPTER 6 DISCUSSION As discussed previously, the prior research on international market and product diversif ication for manufacturing firms has reported a wide variety of relationships negative, positive, inverted U shaped, U shaped, and S shaped --between diversification intensity and financial performance. Some of these differences may be due to differenc es in methodology such as failure to consider potential endogeniety bias or the different measures used to assess performance and diversification. However, these differences might also be due to the industries sampled. Summary of Results My results support an inverted U shaped relationship consistent within my premise that retail operations are more complex than manufacturing operations and thus the scale and scope economies are realized after retail firms have considerable learning from diversification experiences. The inverted U shaped relationship in my results for retail firms is consist with Porters (1985) business strategy dictum that getting stuck in the middle is a prescription for below average financial performance. In the context of int ernational market and format diversification, stuck in the middle is pursing both forms of diversification with middling intensity. My results suggest the highest levels of financial performance are achieved when retailers either operate one format in m any international markets or multiple formats in one market. While the estimated coefficients indicate an inverted U shaped relationship, my research findings suggest that international and format diversification have significant interactive and nonlinear effects on performance. Due to these interactive and
70 nonlinear effects, the nature of the functional relationships between international diversification varies for different levels of format diversification. Thus, the relationship between diversificati on and performance can appears to take various functional forms with in the range of the data. The managerial implication of this research is that are retailers seek growth through diversification, the two approaches that lead to strong financial performance are to operate a large number of formats in a few countries or use a small number of formats in many countries. The first approach exploits the retailers assets related to operations, while the second approach exploits the retailers assets related t o home market. The lowest level of financial performance arises when a retailer has a portfolio that involves operating a median level of formats in a median level of countries. For retailers with this type of portfolio the cost of diversification outwei gh the benefits. When looking at diversification within the retail sector, it appears that diversification intensity, the number of countries in which a retailer operates and the number the formats utilized, has a greater impact on financial performance t han the similarities between of countries and format within a retailers portfolio. However, weaker estimated effects of format and country dissimilarity on financial market performance may be due to the quality of the dissimilarity measures. Assessing t he number of countries in which a retailer operates and the number of formats used in straight forward. However, there is probably more error in assessing portfolio dissimilarities, particularly cultural and economic dissimilarities. Both of these measur es require transformations of data to form an index. In addition, even though the measures of cultural characteristics used in this research are widely used to assess
71 culture, they are based on dated, small, convenience sample of people living in the countries. Directions for Future Research As i ndicated, this research is one of the few to examine the effects for diversification for service business, specifically retailers. While my research findings do differ somewhat from previous diversification research, these differences might be due to different methods and measures rather than the differences in industries. Thus a direction for future research would be to examine the potential differences between manufacturing and services industry using the same methodologies with samples of firms in both industry sectors. Another critical direction for future diversification research is to examine the moderating effects of managerial actions on the diversificationperformance relationship. For example, the effects of operating portfolios with many international market and/or formats might be mitigated or exasperated by the activities the retailers choose centralize versus decentralize and the ownership structure of the operations in the retailers portfolio. Ahold chooses to decentralize most decision making and even uses differ names for chains in the same and different countries. It also uses a variety of on ownership ranging from cooperatives to direct foreign investment other hand, Wal Mart takes a more centralized decision making approach and use direct foreign investment. Which operating and ownership approaches are more effective or does the effectiveness of the approach depend on the number of countries and formats in the retailers portfolio? Due to data limitations I was unable to study the moderating effect of networ k size or the effect of selective cross country format implementation on performance.
72 However, past research has shown that retail investment intensity during entry can impact retailer s long run post entry performance (Gielens and Dekimpe 2001), presumably due to information asymmetries, pioneer advantage and market power Hence, future research can include the moderating effect of retail network size on retailers financial performanc e. Along those lines, one could also examine whether there exist any systematic impact on retailers performance based on how aggressively and uniformly retailers distribute their existing formats over each geographic retail network cluster. G iven the present hype about opportunities in emerging markets, one can extend this research to study the differential impact of retailers who predominantly maintain either a portfolio of developed or developing geographic market constituents. Such an endeavor may reveal interesting findings regarding the transferability and implementation of specific retail concepts and expertise under varied market conditions while taking into consideration the size of the retail network. For instance, while emerging markets may affor d retailers with higher market potential and growth, these markets may also prove more volatile to retailers trying to market and implement their retail concept and this risk is often not priced in the decision making process ex ante. Accounting for the intangible cost and volatility of operating in emerging markets may lead retailers to rethink their diversification portfolio strategy. Several authors posit that retailers can engage in selective entry and utilize knowledge gained from increasingly diverse markets to guide sequential expansion decisions Given this learning justification, a final recommendation for future research is to split the sample into four groups: 1) retailers from developed home markets who are
73 primarily confined to developed host m arkets [compact developed markets portfolio], 2) retailers from developed home markets who are primarily confined to emerging host markets [varied emerging market portfolio], 3) retailers from emerging home markets who are primarily confined to developed host markets and [varied developed markters portfolio] 4) retailers from emerging host markets who are primarily confined to emerging host markets [compact emerging markets portfolio] A comprehensive examination across these subgroups could reveal interest ing findings on the differential efficacy of the learning process and overall portfolio performance.
74 APPENDIX A DATA APPENDIX Construct Classification Variable Transformation Source First Stage Company Variables Performance Retai l Sales NA Global Powers of Retailing 2002 07 Descriptive Information Founding Year NA Hoovers (http://premium.hoovers.com); Google Finance (http://finance.google.com) Country List NA Global Powers of Retailing 2002 07 Format List NA Global Powers o f Retailing 2002 07 First Stage Country Variables Country Metrics Urban Population (% of total) (Instrumented/Fact or Rotated) World Bank Development Indicators, 2007 (http://publications.worldbank.org/WDI/indicators) Population (Instrumented/ Fact or Rotated) World Bank Development Indicators, 2007 (http://publications.worldbank.org/WDI/indicators) Population Density (Instrumented/Fact or Rotated) World Bank Development Indicators, 2007 (http://publications.worldbank.org/WDI/indicators) GDP ( per capita) (Instrumented/Fact or Rotated) World Bank Development Indicators, 2007 (http://publications.worldbank.org/WDI/indicators) Personal Computers (per 1000 individuals) (Instrumented/Fact or Rotated) World Bank Development Indicators, 2007 (http://p ublications.worldbank.org/WDI/indicators) International Tourist Arrivals/Departures International Tourist Arrival/International Tourist Departure (Instrumented/Fact or Rotated) World Bank Development Indicators, 2007 (http://publications.worldbank.org/WDI /indicators) Second Stage Performance Tobin's Q (Share Price x Shares Outstanding)/Total Assets Wharton Research Data Services Compustat (http://wrds.wharton.upenn.edu/) Share Price (Adjusted) NA Wharton Research Data Services Compustat (ht tp://wrds.wharton.upenn.edu/); Google Finance (http://finance.google.com) Shares Outstanding NA Wharton Research Data Services Compustat (http://wrds.wharton.upenn.edu/); Google Finance (http://finance.google.com) Total Assets NA Wharton Research Dat a Services Compustat (http://wrds.wharton.upenn.edu/); Google Finance (http://finance.google.com) Diversification level Country Count NA Global Powers of Retailing 2002 07 Format Count NA Global Powers of Retailing 2002 07 Cultural Dissimilarity Po wer Distance Index NA Hofstede Scores (http://www.geert hofstede.com/hofstede_dimensions.php) Individualism NA Hofstede Scores (http://www.geert hofstede.com/hofstede_dimensions.php) Masculinity NA Hofstede Scores (http://www.geert hofstede.com/hofsted e_dimensions.php) Uncertainty Avoidance Index NA Hofstede Scores (http://www.geert hofstede.com/hofstede_dimensions.php) Economic Dissimilarity Economic Dissimilarity Factor Scores Weighted Average See enclosed factor analysis results in Appendix.
75 Sci entific and Technical Articles (per 1000 individuals) (Instrumented/Fact or Rotated) World Bank Development Indicators, 2007 (http://publications.worldbank.org/WDI/indicators) Merchandize Exports (Instrumented/Fact or Rotated) World Bank Development Indica tors, 2007 (http://publications.worldbank.org/WDI/indicators) Working Age Population (ages 15 64) (Instrumented/Fact or Rotated) World Bank Development Indicators, 2007 (http://publications.worldbank.org/WDI/indicators) Dependents to Working Age Populat ion Ratio (Instrumented/Fact or Rotated) World Bank Development Indicators, 2007 (http://publications.worldbank.org/WDI/indicators) Information and Telecommunication Expenditure (Instrumented/Fact or Rotated) World Bank Development Indicators, 2007 (h ttp://publications.worldbank.org/WDI/indicators) Real Interest Rate (Instrumented/Fact or Rotated) World Bank Development Indicators, 2007 (http://publications.worldbank.org/WDI/indicators) Domestic Credit Provided by Banking Sector (Instrumented/Fact or Rotated) World Bank Development Indicators, 2007 (http://publications.worldbank.org/WDI/indicators) Gross Domestic Savings (% of GDP) (Instrumented/Fact or Rotated) World Bank Development Indicators, 2007 (http://publications.worldbank.org/WDI/indicators ) Physical Distance Geographical Coordinates NA World Factbook, 2007 (https://www.cia.gov/library/publications/the world factbook) Format Dissimilarity Format Dissimilarity Survey Items Weighted Average See enclosed survey questions in Appendix. Mi sc. Variables Performance Revenue NA Wharton Research Data Services Compustat (http://wrds.wharton.upenn.edu/); Thompson One Analytics (http://www.thomsononeim.com); Hoovers (http://premium.hoovers.com); Google Finance (http://finance.google.com) N et Income NA Wharton Research Data Services Compustat (http://wrds.wharton.upenn.edu/); Thompson One Analytics (http://www.thomsononeim.com); Hoovers (http://premium.hoovers.com); Google Finance (http://finance.google.com) Descriptive Information Cost o f Goods Sold NA Wharton Research Data Services Compustat (http://wrds.wharton.upenn.edu/); Thompson One Analytics (http://www.thomsononeim.com); Hoovers (http://premium.hoovers.com); Google Finance (http://finance.google.com) Average Inventory (Inventory[t] + Inventory[t 1])/2 Wharton Research Data Services Compustat (http://wrds.wharton.upenn.edu/); Thompson One Analytics (http://www.thomsononeim.com); Hoovers (http://premium.hoovers.com); Google Finance (http://finance.google.com) Data may not be exclusive to the listed category.
76 Home Country Economic Variables GINI index Income share held by highest 20% Poverty headcount ratio at national poverty line (% of population) Age dependency ratio (dependents to working ag e population) Birth rate, crude (per 1,000 people) Mortality rate, under 5 (per 1,000) Net migration Population ages 15 64 (% of total) Population density (people per sq. km) Population growth (annual %) Population in urban agglomerations > 1 million (% of total population) Population, female (% of total) Population, total Survival to age 65, female (% of cohort) Survival to age 65, male (% of cohort) Urban population Urban population (% of total) Urban population growth (annual %) Foreign direct investment, net (BoP, current US$) Foreign direct investment, net inflows (% of GDP) Foreign direct investment, net outflows (% of GDP) Exports of goods and services (BoP, current US$) Goods exports (BoP, current US$) Goods im ports (BoP, current US$) Imports of goods and services (BoP, current US$) Net income (BoP, current US$) Service exports (BoP, current US$) Service imports (BoP, current US$) Trade in services (% of GDP) Average number of times firms spent in me etings with tax officials Average time to clear exports through customs (days) Broadband subscribers (per 100 people) Business disclosure index (0=less disclosure to 10=more disclosure) Business entry rate (new registrations as % of total) Contai ner port traffic (TEU: 20 foot equivalent units) Cost of business start up procedures (% of GNI per capita) CPIA building human resources rating (1=low to 6=high) CPIA business regulatory environment rating (1=low to 6=high) CPIA debt policy rating (1=low to 6=high) CPIA economic management cluster average (1=low to 6=high) CPIA efficiency of revenue mobilization rating (1=low to 6=high) CPIA equity of public resource use rating (1=low to 6=high) CPIA financial sector rating (1=low to 6=high ) CPIA fiscal policy rating (1=low to 6=high) CPIA gender equality rating (1=low to 6=high) CPIA macroeconomic management rating (1=low to 6=high) CPIA policies for social inclusion/equity cluster average (1=low to 6=high) CPIA policy and institu tions for environmental sustainability rating (1=low to 6=high) CPIA property rights and rule based governance rating (1=low to 6=high)
77 CPIA public sector management and institutions cluster average (1=low to 6=high) CPIA quality of budgetary and fin ancial management rating (1=low to 6=high) CPIA quality of public administration rating (1=low to 6=high) CPIA social protection rating (1=low to 6=high) CPIA structural policies cluster average (1=low to 6=high) CPIA trade rating (1=low to 6=high) CPIA transparency, accountability, and corruption in the public sector rating (1=low to 6=high) Credit information availability index (0=less info to 6=more info) Daily newspapers (per 1,000 people) Ease of doing business index (1=most business fr iendly regulations) Firms offering formal training (% of firms) Firms that do not report all sales for tax purposes (% of firms) Firms using banks to finance investment (% of firms) Firms with female participation in ownership (% of firms) Fixed line and mobile phone subscribers (per 100 people) High technology exports (% of manufactured exports) High technology exports (current US$) Households with television (%) IDA resource allocation index (1=low to 6=high) Information and communicat ion technology expenditure (% of GDP) Information and communication technology expenditure (current US$) Information and communication technology expenditure per capita (US$) International Internet bandwidth (bits per person) International Internet bandwidth (Mbps) International tourism, expenditures (% of total imports) International tourism, number of arrivals International tourism, number of departures International voice traffic (minutes per person) International voice traffic (out and in, minutes) Internet users (per 100 people) ISO certification ownership (% of firms) Legal rights of borrowers and lenders index (0=less credit access to 10=more access) Losses due to theft, robbery, vandalism, and arson (% sales) Management ti me dealing with officials (% of management time) Micro, small and medium enterprises (per 1,000 people) Mobile phone subscribers (per 100 people) New businesses registered (number) Passenger cars (per 1,000 people) Patent applications, nonresiden ts Patent applications, residents Personal computers (per 100 people) Price basket for Internet (US$ per month) Price basket for mobile (US$ per month) Price basket for residential fixed line (US$ per month) Private credit bureau coverage (% of adults) Procedures to build a warehouse (number) Procedures to enforce a contract (number) Procedures to register property (number) Public credit registry coverage (% of adults) Pump price for diesel fuel (US$ per liter) Rail lines (total rout e km) Railways, goods transported (million tonkm)
78 Researchers in R&D (per million people) Rigidity of employment index (0=less rigid to 100=more rigid) Roads, goods transported (million ton km) Scientific and technical journal articles Secure Internet servers (per 1 million people) Start up procedures to register a business (number) Tax payments (number) Technicians in R&D (per million people) Telecommunications investment (% of revenue) Telecommunications revenue (% GDP) Telephone average cost of call to US (US$ per three minutes) Telephone faults (per 100 mainlines) Telephone mainlines (per 100 people) Time required to build a warehouse (days) Time required to enforce a contract (days) Time required to obtain an operating license (days) Time required to register property (days) Time required to start a business (days) Time to prepare and pay taxes (hours) Time to resolve insolvency (years) Total businesses registered (number) Total tax rate (% of profit) Trad emarks, nonresidents Trademarks, residents Unofficial payments to public officials (% of firms) Value lost due to electrical outages (% of sales) Vehicles (per 1,000 people) Vehicles (per km of road) Consumer price index (2000 = 100) GDP defl ator (base year varies by country) Inflation, consumer prices (annual %) PPP conversion factor, GDP (LCU per international $) Real effective exchange rate index (2000 = 100) Wholesale price index (2000 = 100) Bank capital to assets ratio (%) Ba nk nonperfoming loans to total gross loans (%) Domestic credit to private sector (% of GDP) Lending interest rate (%) Listed domestic companies, total Net foreign assets (current LCU) Risk premium on lending (%) Highest marginal tax rate, corpo rate rate (%) Highest marginal tax rate, individual rate (%) Social contributions (% of revenue) Taxes on exports (% of tax revenue) Taxes on goods and services (% of revenue) Taxes on international trade (% of revenue) Exports as a capacity to import (constant LCU) Exports of goods and services (% of GDP) GDP per capita (constant 2000 US$) GDP per capita growth (annual %) Gross savings (% of GDP)
79 Household final consumption expenditure (constant 2000 US$) Imports of goods and servic es (annual % growth) Net income from abroad (constant LCU) Trade (% of GDP) Death rate, crude (per 1,000 people) Employment in agriculture (% of total employment) Employment in industry (% of total employment) Employment in services (% of total employment) Employment to population ratio, ages 15 24, total (%) GDP per person employed, index (1980 = 100) Labor force participation rate, total (% of total population ages 15 64) Labor force with primary education (% of total) Labor force wi th secondary education (% of total) Labor force with tertiary education (% of total) Labor force, female (% of total labor force) Labor force, total Life expectancy at birth, total (years) Literacy rate, youth total (% of people ages 1524) Lon g term unemployment (% of total unemployment) Unemployment, male (% of male labor force) Unemployment, total (% of total labor force) Unemployment, youth total (% of total labor force ages 15 24) Agricultural raw materials exports (% of merchandise exports) Agricultural raw materials imports (% of merchandise imports) Computer, communications and other services (% of commercial service exports) Computer, communications and other services (% of commercial service imports) Export quantum/quant ity index (2000 = 100) Export value index (2000 = 100) Import quantum/quantity index (2000 = 100) Import value index (2000 = 100) Manufactures exports (% of merchandise exports) Manufactures imports (% of merchandise imports) Merchandise export s (current US$) Merchandise imports (current US$) Merchandise trade (% of GDP) Travel services (% of commercial service exports) Travel services (% of commercial service imports)
80 APPENDIX B SAMPLE OF RETAILERS RETAILER COUNTRY OF ORIGIN GPR OBSER VATION YEARS FINANCIAL DATA AVAILABLE? 7 ELEVEN (SOUTHLAND) USA 2002 2003 YES 84 LUMBER USA 2005 NO A&P USA 2002 2003 YES ABERCROMBIE USA 2007 NO ADVANCE AUTO USA 2003 2004 2005 2006 2007 NO AEON JAPAN 2002 2003 2004 2005 2006 2007 YES ALBERTSONS USA 2002 2003 2004 2005 2006 2007 YES ALDI GMBH GERMANY 2002 2003 2004 2005 2006 2007 NO ALIMENTATION COUCHE_TARD CANADA 2005 2006 2007 NO ALTICOR INC USA 2002 2003 2004 2005 2006 2007 NO AMAZON.COM USA 2002 2003 2004 2005 2006 2007 YES AMES USA 2002 2003 YES AMWAY USA 2002 NO ARCADIA GROUP UK 2002 2003 2004 2005 2006 2007 YES ARMY & AIR FORCE EXCHANGE USA 2002 2003 2004 2005 2006 2007 NO AS WATSON/ HUTCHINGSON WHAMPOA HONGKONG 2004 2005 2006 2007 NO ASBURY AUTOMOTIV E USA 2003 2004 2005 NO AUTOGRILL ITALY 2005 NO AUTONATION USA 2002 2003 2004 2005 YES AUTOZONE USA 2002 2003 2004 2005 2006 2007 YES AVON USA 2002 2003 2004 2005 2006 2007 YES AXFOOD AB SWEDEN 2002 2003 2005 2006 2007 YES BARNES & NOBLE USA 2002 2003 2004 2005 2006 2007 YES BAUGUR GROUP ICELAND 2007 NO BAUHAUS GERMANY 2006 2007 NO BED BATH AND BEYOND USA 2002 2003 2004 2005 2006 2007 YES BEIJING GOME HOME APPLIANCE CHINA 2005 NO BELK, INC. USA 2003 2005 2006 2007 NO B ERKSHIRE HATHAWAY USA 2005 2006 2007 NO BERTELSMANN AG GERMANY 2002 2003 2004 2005 2006 2007 YES BEST BUY USA 2002 2003 2004 2005 2006 2007 YES BEST DENKI CO. JAPAN 2002 2004 2005 2006 2007 YES BIC CAMERA JAPAN 2005 2006 2007 NO BIG LOTS USA 2002 2003 2004 2005 2006 2007 YES BILL HEARD ENTERPRISES USA 2003 NO BJ'S WHOLESALE USA 2002 2003 2004 2005 2006 2007 YES BL LO HOLDINGS USA 2007 NO BLOCKBUSTER USA 2006 2007 NO BOOTS GOUP UK 2002 2003 2004 2005 2006 2007 YES BORDERS GROU P USA 2002 2003 2004 2005 2006 2007 YES BRINKER INTERNATIONAL USA 2005 NO
81 BURLINGTON COAT USA 2003 2004 2005 2006 2007 NO C&A BELGIUM 2002 2003 2004 2005 2006 2007 NO CAINZ HOME JAPAN 2006 2007 NO CANADIAN TIRE CANADA 2002 2003 2004 2005 2006 2007 YES CAPRABO, SA SPAIN 2005 2006 2007 NO CARMAX USA 2005 NO CARREFOUR FRANCE 2002 2003 2004 2005 2006 2007 YES CASA BAHIA BRAZIL 2006 2007 NO CASEY'S GENERAL USA 2005 2006 2007 NO CASINO GUICHARD FRANCE 2002 2003 2004 2005 2006 2 007 YES CBRL GROUP USA 2005 NO CCA GLOBAL USA 2005 NO CELESIO AG GERMANY 2005 2006 2007 NO CENCOSUD S.A. CHILE 2007 NO CENTRES DISTRIBUTEURS E LECLERC FRANCE 2002 2003 2004 2005 2006 2007 NO CHARMING SHOPPES USA 2004 2005 2006 200 7 NO CIRCUIT CITY USA 2002 2003 2004 2005 2006 2007 YES COLES MYER AUSTRALIA 2002 2003 2004 2005 2006 2007 YES COMCAST/QVC USA 2002 2003 2004 YES COMPANHIA BRASILEIRA DE DISTRIBUICAO SA GRUPO PAO DE ACUCAR BRAZIL 2002 2003 2004 2005 2006 YES COMPA SS UK 2002 2003 2004 2005 YES COMPUSA USA 2003 2004 2005 2006 2007 NO CONAD CONSORZIO ITALY 2002 2003 2004 2005 2006 2007 NO CONTROLADORA COMMERICAL MEXICANA MEXICO 2002 2003 2004 2005 2006 2007 YES COOP ITALIA ITALY 2002 2003 2004 2005 2006 2007 NO COOP KOBE JAPAN 2002 2005 2006 NO COOP NORDEN AB SWEDEN 2004 2005 2006 2007 NO COOP NORWAY NORWAY 2002 2003 NO COOP SWITZERLAND SWITZERLAND 2002 2003 2004 2005 2006 2007 YES COOPERATIVE GROUP UK 2002 2003 2004 2005 2006 2007 NO CORA FRANCE 2002 2003 NO COSTCO USA 2002 2003 2004 2005 2006 2007 YES CVS USA 2002 2003 2004 2005 2006 2007 YES DAIEI JAPAN 2002 2003 2004 2005 2006 2007 YES DAIRY FARM HONGKONG 2002 2003 2004 2005 2006 2007 YES DAISO SANGYO JAPAN 2005 2006 2007 NO DALIAN DASHANG GROUP CO. LTD. CHINA 2006 NO DANSK SUPERMARKED DENMARK 2002 2003 2004 2005 2006 2007 YES DARDEN RESTAURANTS USA 2002 2003 2004 2005 YES DEBENHAMS PLC UK 2004 2005 2006 2007 NO DECATHLON GROUP FRANCE 2006 2007 NO DEFENSE COMM USA 2004 2005 2006 2007 NO DELHAIZE GROUP BELGIUM 2002 2003 2004 2005 2006 2007 YES DELHAZIE AMERICA (FOODLION) USA 2002 YES DELL USA 2003 2004 2005 2006 2007 NO DICK'S SPORTING USA 2007 NO
82 DILLARDS USA 2002 2003 2004 2005 2006 2007 YES DIRK ROOSSMANN GERMANY 2007 NO DISTRIBUCION Y SERVICIO CHILE 2006 2007 NO DM DROGERIE GERMANY 2004 2005 2006 2007 NO DOHLE HANDELSGRUPPE GERMANY 2004 2005 2006 2007 NO DOLLAR GENERAL USA 2002 2003 2004 2005 2006 2007 YES DOLLAR TREE STORES US A 2005 2006 2007 NO DOUGLAS HOLDING GERMANY 2005 2006 2007 NO DSG INTERNATIONAL UK 2002 2003 2004 2005 2006 2007 YES EAST JAPAN RAILWAY JAPAN 2006 2007 NO EDEKA ZENTRALE AG GERMANY 2002 2003 2004 2005 2006 2007 NO EDGARS CONSOLIDATED SAFRICA 2007 NO EDION JAPAN 2005 2006 2007 NO EL CORTE INGLES SPAIN 2002 2003 2004 2005 2006 2007 NO EL PUERTO DE LIVERPOOL MEXICO 2007 NO ESSELUNGA ITALY 2002 2003 2004 2005 2006 2007 NO ETS FRANZ COLRUYT BELGIUM 2004 2005 2006 2007 NO EUROM ADI SPAIN 2002 NO FA ANTON SCHLECKER GERMANY 2002 2003 2004 2005 2006 2007 NO FAMILY DOLLAR USA 2002 2003 2004 2005 2006 2007 YES FAMILYMART CO., LTD. JAPAN 2006 NO FAST RETAILING JAPAN 2003 2004 2005 2006 2007 NO FDB DENMARK 2002 2003 NO FEDERATED DEPARTMENT STORES USA 2002 2003 2004 2005 2006 2007 YES FEMSA COMERCIO MEXICO 2007 NO FINIPER S.P.A. ITALY 2006 2007 NO FLEMING USA 2002 YES FOCUS WIKES GROUP LTD. UK 2005 2006 NO FOODLAND ASSOCIATED LTD. AUSTRALIA 20 04 2005 2006 NO FOOT LOCKER USA 2002 2003 2004 2005 2006 2007 YES FOOTSTAR USA 2003 NO FUJI CO. JAPAN 2005 2006 2007 NO GAMESTOP USA 2007 NO GAP, INC USA 2002 2003 2004 2005 2006 2007 YES GATEWAY USA 2003 NO GIANT EAGLE USA 2002 2 003 2004 2005 2006 2007 NO GIB BELGIUM 2002 YES GIGAS K'S DENKI JAPAN 2006 2007 NO GLOBUS HOLDING GERMANY 2002 2003 2004 2005 2006 2007 NO GREAT UNIVERSAL STORES UK 2002 2003 2004 2005 2006 2007 YES GROUP 1 AUTOMOTIVE USA 2002 2003 2004 2005 YES GROUPE AUCHAN FRANCE 2002 2003 2004 2005 2006 2007 NO GROUPE GALERIES LAFAYETTE FRANCE 2002 2003 2004 2005 2006 2007 YES GRUPO CARSO MEXICO 2002 YES GRUPO EROSKI SPAIN 2002 2003 2004 2005 2006 2007 NO GRUPO GIGANTE MEXICO 2002 2003 2004 200 5 2006 2007 YES
83 GRUPPO PAM S.P.A., GECOS S.P.A. ITALY 2005 2006 2007 NO GS RETAIL CO. SKOREA 2006 2007 NO H&M SWEDEN 2002 2003 2004 2005 2006 2007 YES HACHETTE FRANCE 2006 2007 NO HANKYU DEPARTMENT JAPAN 2002 2003 2004 2005 2006 2007 YES H E BUTT GROCERY USA 2002 2003 2004 2005 2006 2007 NO HEIWADO CO. JAPAN 2002 2003 2004 2005 2006 2007 YES HENDRICK AUTOMOTIVE GROUP USA 2003 2005 NO HMV GROUP PLC UK 2005 2006 2007 NO HOLMAN ENTERPRISES USA 2003 NO HOME DEPOT USA 2002 2003 2 004 2005 2006 2007 YES HORNBACH GERMANY 2006 2007 NO HUDSON'S BAY CANADA 2002 2003 2004 2005 2006 2007 YES HY VEE, INC USA 2002 2003 2004 2005 2006 2007 NO IAC/INTERACTIVE USA 2005 2006 2007 NO ICA AB SWEDEN 2006 2007 NO IFA SPAIN 2002 NO IKEA SWEDEN 2002 2003 2004 2005 2006 2007 NO INDITEX SPAIN 2003 2004 2005 2006 2007 NO INSIEME ITALY 2002 NO INTERDIS ITALY 2002 NO INTERMARCHE FRANCE 2002 2003 2004 2005 NO INTERMEDIA 90 ITALY 2002 NO INTIMATE BRANDS USA 2002 2003 YES ISETAN JAPAN 2002 2003 2004 2005 2006 2007 YES ITM DEVELOPPEMENT INT FRANCE 2006 2007 NO ITO YODADO CO. LTD. JAPAN 2002 2003 2004 2005 2006 YES IZUMI CO. LTD. JAPAN 2002 2003 2004 2005 2006 YES IZUMIYA CO. JAPAN 2002 2003 2004 2005 2006 2007 YES J SAINSBURY UK 2002 2003 2004 2005 2006 2007 YES JC PENNY USA 2002 2003 2004 2005 2006 2007 YES JEAN CROTU GROUP CANADA 2004 2005 2006 2007 NO JERONIMO MARTINS PORTUGAL 2005 2006 2007 NO JIM PATTISON GROUP CANADA 2005 2006 2007 NO JOHN LEWIS UK 2002 2003 2004 2005 2006 2007 YES JOSHIN DENKI JAPAN 2006 2007 NO KARSTADTQUELLE GERMANY 2002 2003 2004 2005 2006 2007 YES KATZ GROUP CANADA 2006 2007 NO KESA ELECTRICALS UK 2005 2006 2007 NO KESKO FINLAND 2002 2003 2004 2 005 2006 2007 YES KINGFISHER UK 2002 2003 2004 2005 2006 2007 YES KINTETSU DEPARTMENT JAPAN 2003 2004 2005 2006 2007 NO KMART HOLDING CORP. USA 2002 2003 2004 2005 2006 NO KOHL'S CORP USA 2002 2003 2004 2005 2006 2007 YES KOJIMA CO. JAPAN 2002 2003 2004 2005 2006 2007 YES KONINKLIJKE NETHERLANDS 2002 2003 2004 2005 2006 2007 YES KOOPERATIVE FORBUNDET GROUP SWEDEN 2002 NO
84 KOTOBUKIYA JAPAN 2002 YES KROGER USA 2002 2003 2004 2005 2006 2007 YES LAURUS N.V. NETHERLANDS 2002 2003 2006 2007 YES LAWSON, INC. JAPAN 2006 NO LEROY MERLIN FRANCE 2002 2003 2004 2005 2006 2007 NO LIBERTY MEDIA USA 2005 2006 2007 NO LIFE CORPORATION JAPAN 2002 2003 2004 2005 2006 2007 YES LIMITED BRANDS USA 2002 2003 2004 2005 2006 2007 YES LINENS 'N T HINGS USA 2005 2006 2007 NO LITHIA MOTORS USA 2005 NO LITTLEWOODS SHOP UK 2002 2003 2004 2005 2006 2007 NO LOBLAW CANADA 2002 2003 2004 2005 2006 2007 YES LONGS DRUG USA 2002 2003 2004 2005 2006 2007 YES LOTTE SHOPPING SKOREA 2003 2004 2005 2 006 2007 NO LOUIS DELHAIZE BELGIUM 2004 2005 2006 2007 NO LOWE'S USA 2002 2003 2004 2005 2006 2007 YES LUXOTTICA GROUP ITALY 2005 2006 2007 NO LVMH FRANCE 2002 2003 2004 2005 2006 2007 YES MANOR SWITZERLAND 2005 NO MARKS & SPENCER UK 2002 2 003 2004 2005 2006 2007 YES MARUI CO. JAPAN 2002 2003 2004 2005 2006 2007 YES MASSMART HOLDINGS SAFRICA 2005 2006 2007 NO MATSUMOTO KIYOSHI JAPAN 2005 2006 2007 NO MATSUZAKAYA CO. JAPAN 2002 2003 2004 2005 2006 2007 YES MAUS FRERES SWITZERLAND 2 002 2003 NO MAXEDA/ROYAL VENDEX KBB NETHERLANDS 2002 2003 2004 2005 2006 2007 YES MCDONALDS USA 2002 2003 2004 2005 YES MEIJER USA 2002 2003 2004 2005 2006 2007 NO MENARDS USA 2002 2003 2004 2005 2006 2007 NO MERCADONA SPAIN 2002 2003 2004 2005 2006 2007 NO MERVYN'S, LLC USA 2006 2007 NO METCASH GROUP AUSTRALIA 2007 NO METCASH TRADING SAFRICA 2004 2005 2006 2007 NO METRO AG GERMANY 2002 2003 2004 2005 2006 2007 YES METRO RICHELIEU INC CANADA 2002 2003 2004 2005 2006 2007 YES MICH AELS STORES USA 2003 2004 2005 2006 2007 NO MIGROS GENOSSENSCHAFTS BUND SWITZERLAND 2002 2003 2004 2005 2006 2007 YES MILLENNIUM RETAILING JAPAN 2005 2006 2007 NO MITCHELLS&BUTLERS UK 2005 NO MITSUKOSHI JAPAN 2002 2003 2004 2005 2006 2007 YES MUSGRAVE GP IRELAND 2005 2006 2007 NO MYCAL JAPAN 2002 YES NEIMAN MARCUS USA 2002 2003 2004 2005 2006 2007 YES NEXT PLC UK 2003 2004 2005 2006 2007 NO NORDSTROM USA 2002 2003 2004 2005 2006 2007 YES NORGESGRUPPEN NORWAY 2002 2003 2004 2005 NO
85 NORMA LEBENSMITTELFILIALBETRIEB, GMBH & CO. KG GERMANY 2005 2006 2007 NO ODAKYU ELECTRIC RAILWAY CO. LTD. JAPAN 2002 2005 2006 YES OFFICE DEPOT USA 2002 2003 2004 2005 2006 2007 YES OFFICEMAX USA 2002 2003 2004 2005 2006 2007 YES ORGANIZACIO N SORIANA S.A. DE C.V. MEXICO 2002 2003 2004 2005 2006 2007 YES OTTO GROUP GERMANY 2002 2003 2004 2005 2006 2007 YES OUTBACK STEAKHOUSE USA 2005 NO PATHMARK STORES USA 2002 2003 2004 2005 2006 2007 YES PAYLESS SHOESOURCE USA 2002 2003 2004 2005 20 06 2007 YES PEPBOYS USA 2002 YES PEPKOR SAFRICA 2002 2003 YES PETSMART,INC. USA 2003 2004 2005 2006 2007 NO PHONES4U UK 2006 2007 NO PICK N PAY RETAILERS SAFRICA 2004 2005 2006 2007 NO PPR GROUP FRANCE 2002 2003 2004 2005 2006 2007 YE S PRAKTIKERBAUUND GERMANY 2007 NO PRESIDENT CHAIN STORE TAIWAN 2006 2007 NO PUBLIX SUPERMARKETS USA 2002 2003 2004 2005 2006 2007 YES QUIKTRIP CORP. USA 2006 2007 NO RACETRAC PETROLEUM USA 2006 2007 NO RADIOSHACK USA 2002 2003 2004 2 005 2006 2007 YES RALEY'S INC. USA 2002 2003 2004 2005 2006 2007 NO REITAN HANDEL NORWAY 2003 2004 2005 2006 2007 NO RETAIL VENTURES USA 2005 2006 2007 NO REWE ZENTRAL AG GERMANY 2002 2003 2004 2005 2006 2007 NO RINASCENTE ITALY 2002 YES RIT E AID USA 2002 2003 2004 2005 2006 2007 YES ROSS STORES USA 2002 2003 2004 2005 2006 2007 YES ROUNDY'S, INC. USA 2005 2006 2007 NO RUDDICK/ HARRIS TEETER USA 2003 2005 2006 2007 NO S GROUP FINLAND 2002 2003 2004 2005 2006 2007 NO S.A.C.I. FALABEL LA CHILE 2006 2007 NO SAFEWAY UK 2002 2003 2004 2005 YES SAFEWAY USA 2002 2003 2004 2005 2006 2007 YES SAKS USA 2002 2003 2004 2005 2006 2007 YES SAVE MART SUPERMARKETS USA 2006 2007 NO SCHNUCK MARKETS USA 2005 NO SCHWARZ UNTERNEHMENS GERMANY 2002 2003 2004 2005 2006 2007 YES SEARS USA 2002 2003 2004 2005 2006 2007 YES SEARS CANADA CANADA 2002 2003 YES SEIBU DEPARTMENT JAPAN 2002 2003 2004 YES SELEX ITALY 2002 NO SEVEN & I HOLDINGS JAPAN 2007 NO SHANGHAI FRIENDSHI P CHINA 2005 NO SHEETZ, INC. USA 2006 2007 NO SHERWIN WILLIAMS USA 2002 2003 2004 2005 2006 2007 YES
86 SHIMAMURA JAPAN 2005 2006 2007 NO SHINSEGAE SKOREA 2003 2004 2005 2006 2007 NO SHOPKO STORES USA 2002 2003 2004 2005 2006 2007 YES SHOPPERS DRUG CANADA 2004 2005 2006 2007 NO SHOPRITE HOLDINGS SAFRICA 2002 2004 2005 2006 2007 YES SHV MAKRO NETHERLANDS 2004 2005 2006 2007 NO SIGNET GROUP UK 2006 2007 NO SIRIO ITALY 2002 NO SKYLARK JAPAN 2002 2003 2004 2005 YES SOBEYS C ANADA 2002 2003 2004 2005 2006 2007 YES SOMERFIELD GROUP UK 2002 2003 2004 2005 2006 2007 YES SONAE/MODELO CONTINENTE PORTUGAL 2002 2003 2004 2005 2006 2007 YES SONIC AUTOMOTIVE USA 2002 2003 2004 2005 YES SPAR JAPAN JAPAN 2002 NO SPAR OESTERREICHISCHE WARENHANDELS AUSTRIA 2002 NO SPAR OSTERREICHISCHE AUSTRIA 2003 2004 2005 2006 2007 NO SPIEGEL USA 2002 2003 2004 YES STAPLES USA 2002 2003 2004 2005 2006 2007 YES STARBUCKS USA 2003 2004 2005 NO STATER BROS. USA 2002 2003 2004 200 5 2006 2007 YES SUPERVALU USA 2002 2003 2004 2005 2006 2007 YES SYSTEME U FRANCE 2002 2003 2004 2005 2006 2007 NO TAKASHIMAYA JAPAN 2002 2003 2004 2005 2006 2007 YES TARGET USA 2002 2003 2004 2005 2006 2007 YES TCHIBO HOLDING GERMANY 2002 2003 NO TENGELMANN VERWALTUNGSUND BETEILIGUNGS GERMANY 2002 2003 2004 2005 2006 2007 NO TESCO PLC UK 2002 2003 2004 2005 2006 2007 YES THE BIG FOOD GROUP (ICELAND) UK 2002 2003 2004 2005 2006 YES THE CARPHONE UK 2006 2007 NO THE DAIMARU JAPAN 2002 2003 2 004 2005 2006 2007 YES THE GOLUB CORP. USA 2004 2005 2006 2007 NO THE MARUETSU, INC. JAPAN 2002 2004 2005 2006 2007 YES THE MAY DEPARTMENT STORES CO. USA 2002 2003 2004 2005 2006 YES THE PANTRY USA 2002 2003 2004 2005 2006 2007 YES THE SEIYU LTD. JAPAN 2002 2003 2004 2005 2006 YES THE SPORTS AUTHORITY USA 2006 2007 NO TJX USA 2002 2003 2004 2005 2006 2007 YES TOKYU DEPARTMENT STORE JAPAN 2002 2003 2004 2005 2006 2007 YES TOKYU STORE CHAINS JAPAN 2005 NO TOY R US USA 2002 2003 2004 20 05 2006 2007 YES TRICON RESTAURANTS USA 2002 NO UNITED AUTO GROUP USA 2002 2003 2004 2005 YES UNY JAPAN 2002 2003 2004 2005 2006 2007 YES V.T.INC USA 2003 2004 2005 NO
87 VALUE CITY USA 2004 NO WALGREEN USA 2002 2003 2004 2005 2006 2007 Y ES WALMART USA 2002 2003 2004 2005 2006 2007 YES WALMART MEXICO MEXICO 2002 2003 YES WAWA, INC. USA 2007 NO WEGMANS FOOD USA 2002 2003 2004 2005 2006 2007 NO WENDYS USA 2005 NO WESTFARMERS AUSTRALIA 2005 2006 2007 NO WH SMITH UK 200 2 2003 2005 2006 2007 YES WHOLE FOODS USA 2004 2005 2006 2007 NO WILLIAMS SONOMA USA 2005 2006 2007 NO WINN DIXIE USA 2002 2003 2004 2005 2006 2007 YES WM MORRISON SUPERMARKETS UK 2002 2003 2004 2005 2006 2007 YES WOOLWORTHS AUSTRALIA 2002 2003 2004 2005 2006 2007 YES WOOLWORTHS UK 2004 2005 2006 2007 NO YAMADA DENKI JAPAN 2002 2003 2004 2005 2006 2007 YES YODOBASHI CAMERA CO. LTD. JAPAN 2002 2003 2004 2005 2006 2007 NO YORK BENIMARU JAPAN 2006 2007 NO YUM! BRANDS (TRICON) USA 2003 20 04 2005 NO ZALE USA 2005 NO TOTAL 200 200 200 250 250 250
88 APPENDIX C RESULTS FROM FACTOR ANALYSIS OF HOST COUNTRY ECONOMIC VARIAB LES Component Matrix(a) Component 1 2 3 4 URBAN INDEX .806 9.644E 02 .444 .268 SCIENCE INDEX .804 4.136E 02 .522 .142 MERCHANDISE EXPENDITURE INDEX .794 7.353E 02 .442 .168 SAVINGS INDEX .637 .136 .219 .323 URBAN RATIO INDEX .550 .256 .496 .233 TOURISM INDEX .547 .328 .349 .145 DOMESTIC CREDIT INDEX .397 .820 .188 .1 04 ADULT GROWTH RATIO 7.857E 02 .749 .176 .386
89 ADULT RATIO .393 .626 .341 .389 INTEREST RATE INDEX .342 .196 .568 .174 INFRASTRUCTURE INDEX .140 .165 .455 .754 Extraction Method: Principal Component Analysis. a 4 components extracted. Rota ted Component Matrix(a) Component 1 2 3 4 URBAN INDEX .949 .131 7.634E 02 6.877E 02 SCIENCE INDEX .939 .106 .194 9.418E 02 MERCHANDISE EXPENDITURE INDEX .649 6.372E 02 .264 .232 SAVINGS INDEX .401 .804 6.926E 02 .217 URBAN RATIO INDEX .137 .723 .241 .263 TOURISM INDEX .141 .691 7.327E 02 7.076E 02 DOMESTIC CREDIT INDEX .316 .595 .207 .283 ADULT GROWTH RATIO .283 .201 .860 .124 ADULT RATIO .135 .478 .723 .209 INTEREST RATE INDEX .263 .328 .610 .444 INFRASTRUCTURE INDEX .17 7 .130 7.574E 02 .876 Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. a Rotation converged in 8 iterations.
90 APPENDIX D FORMAT DISSIMILARITY SURVEY
93 LIST OF REFERENCES Aaker, David A. (2005). Strategic Market Management, 7th Edition. Hoboken, NJ:Wiley. Agarwal, Sanjeev and Sridhar N. Ramaswami (1992), "Choice of ForeignMarket Entry Mode Impact of Ownership, Location and I nternalization Factors," Journal of International Business Studies, 23 (1), 127. Alexander, N (1990), "Retailers and international markets: motives for expansion," International Marketing Review, 7 (4), 7585. ___ and Hayley Myers (2000), "The Retail Internationalisation Process," International Marketing Review, 17 (45), 33453. Alvarez, L. H. R. (1996), "Demand uncertainty and the value of supply opportunities," Journal of Economics Zeitschrift Fur Nationalokonomie, 64 (2), 16375. Annavarjula, Madan and Sam Beldona (2000), "'Multinationality Performance Relationship: A Review and Reconceptualization," International Journal of Organizational Analysis, 8 (1): 4867. Ansoff, H. Igor (1957), Strategies for Diversification, Harvard Business Review, 35(September October), 113124. BaffoeBonnie, John and Mohammed Khayum (2003), Contemporary Economic Issues in Developing Countries, Westport, Connecticut: Praeger. Barkema, Harry G., John H. J. Bell, and Johannes M. Pennings (1996), "Foreign Entry, Cultural Barr iers, and Learning," Strategic Management Journal, 17 (2), 15166. Barney, Jay B. (1997. Gaining and Sustaining Competitive Advantage. Reading, MA: Addison Wesley Berger, Philip G. and Eli Ofek (1995), "Diversifications Effect on Firm Value," Journal of Finance, 50 (3), 953953. Beckerman, W. (1956), "Distance and the Pattern of IntraEuropean Trade," Review of Economics and Statistics, 38 (1), 3140. Berger, P. and E. Ofek (1995), "Diversifications Effect on Firm Value," Journal of Finance, 50 (3), 95353. Bianchi, Constanza C. and Enrique Ostale (2006), "Lessons Learned from Unsuccessful Internationalization Attempts: Examples of Multinational Retailers in Chile," Journal of Business Research, 59 (1), 140147. Campa, Jose M. and Simi Kedia (2002), "Explai ning the Diversification Discount," Journal of Finance, 57 (4), 17311762.
94 Campbell, N. C. G., J. L. Graham, A. Jolibert, and H. G. Meissner (1988), "Marketing Negotiations in France, Germany, the UnitedKingdom, and the UnitedStates," Journal of Marketing, 52 (2), 4962. Capar, N and M Kotabe (2003), "The Relationship between International Diversification and Performance in Service Firms.," Journal of International Business Studies, 34 (4), 34556. Calvo, Guillermo A. and Stanislaw Wellisz (1978), "Superv ision, Loss of Control, and Optimum Size of Firm," Journal of Political Economy, 86 (5), 94352. Cavusgil, S. Tamer (1983), Market Similarity and Market Selection: Implications for International Marketing, Journal of Business Research, 11 (4), 439 56. Ch ang, ShaoChi.,and Chi Feng Wang (2007), "The Effect Of Product Diversification Strategies on the Relationship between International Diversification and Firm Performance," Journal of World Business, 42 (1), 6179. Costigan, R. D., R. C. Insinga, J. J. Berman, S. S. Ilter, G. Kranas, and V. A. Kureshov (2006), "A cross cultural study of supervisory trust," International Journal of Manpower, 27 (78), 76487. Craig, C. Samuel, William H. Greene, and Susan P. Douglas (2005), "Culture Matters: Consumer Acceptance of US Films in Foreign Markets," Journal of International Marketing, 13 (4), 80103. Davidson, Russell and James G. MacKinnon (1993), Estimation and Inference in Econometrics, New York: Oxford University Press. Davidson, William H. (1983), "Market S imilarity and Market Selection: Implications for International Marketing Strategy," Journal of Business Research, 11 (4), 439 56. Dawson, John A. (1994), "Internationalization of Retailing Operations," Journal of Marketing Management, 10, 267282 ___ (2007), "Scoping and Conceptualizing Retailer Internationalization," Journal of Economic Geography, 7 (4), 373397. Deloitte (2002), Global Powers of Retailing, New York: Deloitte Touche Tohmatsu. ___ (2003), Global Powers of Retailing, New York: Deloitte Touche Tohmatsu. ___ (2004), Global Powers of Retailing, New York: Deloitte Touche Tohmatsu. ___ (2005), Global Powers of Retailing, New York: Deloitte Touche Tohmatsu. ___ (2006), Global Powers of Retailing, New York: Deloitte Touche Tohmatsu. ___ (2007), Global Powers of Retailing, New York: Deloitte Touche Tohmatsu.
95 ___ (2008), Global Powers of Retailing, New York: Deloitte Touche Tohmatsu. Denrell, Jerker (2003), "Vicarious Learning, Undersampling of Failure, and the Myths of Management," Organization Scienc e, 14 (3), 227243. Duffie, Darre.. and Kenneth J. Singleton (1997), "An Econometric Model of the Term Structure of InterestRate Swap Yields," Journal of Finance, 52 (4), 1287321. Dunning, J H (1973), "The Determinants of International Production," Oxfor d Economic Papers, 25 (3), 289336. Dragun, D (2002), "Challenging the rhetoric: Internationalisation, size and financial performance," European Retail Digest, 36, 2533. Economist, The (1999), "Shopping all over the world," in The Economist. Ekeledo, Ikec hi and K. Sivakumar (1998), "Foreign Market Entry Mode Choice of Service Firms: A Contingency Perspective," Journal of the Academy of Marketing Science, 26 (4), 27492. Erramilli, M. Krishna and C. P. Rao (1993), "Service Firms' International Entry Mode Ch oice: A Modified TransactionCost Analysis Approach," Journal of Marketing, 57 (3), 19 38. Feeser, H.R. and G.E. Willard (1990), "Founding strategy and performance: A comparison of high and low growth high tech firms," Strategic Management Journal, 11 (2), 8798. Flowers, Edward Brown (1976), "Oligopolistic Reactions in European and Canadian Direct Investment in the United States," Journal of International Business Studies, 7, 4355. Franke, George R. and S. Scott Nadler (2008), "Culture, Economic Development, and National Ethical Attitudes," Journal of Business Research, 61 (3), 254264. Froot, Kenneth A., David S. Scharfstein, and Jeremy C. Stein (1994), A Framework for Risk Management, Harvard Business Review, 72(November/December), 91102. Gatignon, H and J Vosgerau (2006), "Moderating effects: The myth of mean centering," INSEAD Working Paper, version April 2006. Gielens, K. and M.G. Dekimpe (2001), "Do international entry decisions of retail chains matter in the long run?," International Journal of Research in Marketing, 18 (3), 23559. ___ and MG Dekimpe (2007), "The Entry Strategy of Retail Firms into Transition Economies," Journal of Marketing, 71 (2), 196212.
96 Golder, P. N. and G. J. Tellis (1997), "Will it ever fly? Modeling the takeoff of really new consumer durables," Marketing Science, 16 (3), 25670. Gomez Mejia, Luis R. and Leskue E. Palich (1997), "Cultural Diversity and the Performance of Multinational Firms," Journal of International Business Studies, 28 (2), 309335. Graham, JL (1985), "C ross cultural marketing negotiations: a laboratory experiment," Marketing Science, 4 (2), 13046. Graham, JR, ML Lemmon, and JG Wolf (2002), "Does Corporate Diversification Destroy Value?," The Journal of Finance, 57 (2), 695720. Grant, RM (1988), "On dom inant logic, relatedness and the link between diversity and performance," Strategic Management Journal, 9 (6), 639 42. Gujarati, Damodar N. (2003), Basic Econometrics (4th ed.), Boston: McGraw Hill. Higgins (1997), "The internationalization of food retaili ng," CIES, Food Business News, 8. Hitt, Michael A., Robert E. Hoskisson & Hicheon Kim (1997), "International Diversification and Firm Performance in Product Diversified Firms," Academy of Management Journal, 40(4), 767798 ___, Laszio Tihanyi, Toyah Miller and Brian Connelly (2006), "International Diversification: Antecedents, Outcomes, and Moderators," Journal of Management, 32(6), 831867. Hofstede, Geert H. (1984), Culture's Consequences: International Differences in Work Related Values (Abridged ed.), Beverly Hills: Sage Publications. ___ and Michael A. Hitt (1988), "Strategic Control Systems and Relative R and D Investment in Large Multiproduct Firms," Strategic Management Journal, 9 (6), 605621. Hopkins, Dan (2005), "Graduate Methods Masters Class," Vol. 2008. Hoskisson, RE and MA Hitt (1988), "Strategic control systems and relative R&D investment in large multiproduct firms," Strategic Management Journal, 9 (6). Hunt, Shelby D. and Robert M. Morgan (1995), "The Comparative Advantage Theory of Competi tion," Journal of Marketing, 59, 115. Hyland, David C. and J. David Diltz (2002), "Why Firms Diversify: An Empirical Examination," Financial Management, 31 (1), 5181.
97 Hymer, Stephen H. (1976), The International Operations of National Firms, Cambridge, Ma ssachusetts: MIT press. Jaworski, Bernard J. and Ajay K. Kohli (1993), "Market Orientation: Antecedents and Consequences," Journal of Marketing, 57, 5370. Jensen, Michael C. (1986), "Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers," Ameri can Economic Review, 76 (2), 323329. Kerin, Roger A. and Robert A. Peterson (2007). Strategic Marketing Problems, 11th Edition. Upper Saddle River, NJ: Pearson Kim, W. Chan, Peter Hwang and William P. Burgers (1989), "Global Diversification Strategy and C orporate Profit Performance," Strategic Management Journal, 10 (1), 45 57. Kogut, Bruce and Harbir Singh (1988), "The Effect of National Culture on the Choice of Entry Mode," Journal of International Business Studies, 19 (3), 411432 Lafontaine, F and D Leibsohn (2004), "Beyond Entry: Examining McDonald's Expansion in International Markets," Mimeo, University of Michigan. Lang, Larry H. P. and Rene M. Stulz (1994), "Tobins Q, Corporate Diversification, and Firm Performance," Journal of Political Economy, 102 (6), 12481280. ___, Annette Poulsen, and Rene Stulz (1995), "Asset Sales, Firm Performance, and the Agency Costs of Managerial Discretion," Journal of Financial Economics, 37 (1), 3 37. Levy, Michael and Barton Weitz (2009), Retail Management, 7th Editi on. New York: McGraw Hill. Lewellen, W. G. (1971), "Pure Financial Rationale for Conglomerate Merger," Journal of Finance, 26 (2), 52137. Li, Lei (2007), "Multinationality and Performance: A Synthetic Review and Research Agenda," International Journal of Management Reviews, 9 (2), 117139. Li, S. X. and R. Greenwood (2004), "The effect of withinindustry diversification on firm performance: Synergy creation, multi market contact and market structuration," Strategic Management Journal, 25 (12), 113153. Lu, J. W. and P. W. Beamish (2004), "International diversification and firm performance: The S CURVE hypothesis," Academy of Management Journal, 47 (4), 598609.
98 Markham, Jesse W. (1973), Conglomerate Enterprise and Public Policy, Boston, Massachusetts: Harv ard University Press. Markides, C. C. (1992), "Consequences of Corporate Refocusing Ex Ante Evidence," Academy of Management Journal, 35 (2), 398412. Martin, JD and A Sayrak (2003), "Corporate Diversification and Shareholder Value: A Survey of Recent Li terature," Journal of Corporate Finance, 9 (1), 3757. Melicher, R. W. and D. F. Rush (1973), "Performance of Conglomerate Firms Recent Risk and Return Experience," Journal of Finance, 28 (2), 38188. Mitra, Debanjan and Peter N. Golder (2002), "Whose Culture Matters? Near Market Knowledge and Its Impact on Foreign Market Entry Timing," Journal of Marketing Research, 39 (3), 350 365. Montgomery, C. A. and B. Wernerfelt (1988), "Diversification, Ricardian Rents, and TobinQ," Rand Journal of Economics, 19 (4), 62332. Morgan, Neil and Lopo Rego (2009), Brand Portfolio Strategy and Firm Performance, Journal of Marketing, 73(January 2009) 5974. Mueller, Dennis C. (1972), "A Life Cycle Theory of the Firm," Journal of Industrial Economics, 20 (3), 199219. M ulhern, F.J. (1997), "Retail marketing: from distribution to integration," International Journal of Research in Marketing, 14 (2), 10324. Murphy, KJ (1986), "Incentives, learning, and compensation: A theoretical and empirical investigation of managerial l abor contracts," The RAND Journal of Economics, 5976. Nelson, Richard R. and Gavin Wright (1992), "The Rise and Fall of American Technological Leadership: The Postwar Era in Historical Perspective," Journal of Economic Literature, 30 (4), 19311964. O'Gra dy, Shawna and Henry W. Lane (1997), "Culture: An Unnoticed Barrier to Canadian Retail Performance in the USA," Journal of Retailing and Consumer Services, 4 (3), 159170. O'Rourke, KH, AL Taylor, and JG Williamson (1996), "Factor Price Convergence in the Late Nineteenth Century," International Economic Review, 37, 499530. Palepu. Krishna (1985), "Diversification Strategy, Profit Performance, and the Entropy Measure of Diversification," Strategic Management Journal, 6 (3), 239255.
99 Palich, Leslie E., & Gom ez Mejia, Luis R. (1999),. "A Theory of Global Strategy and Firm Efficiencies: Considering the Effects of Cultural Diversity," Journal of Management, 25 (4), 587606. ___, Laura B. Cardinal, and C. Chet Miller (2000), "Curvilinearity in the DiversificationPerformance Linkage: An Examination of over Three Decades of Research," Strategic Management Journal, 21 (2), 15574. Pollack, Elaine (2007), Retailing 2015: New Frontiers. Columbus, OH: TNS Retail Forward Porter, Michael E. (1985). Competitive Advantage: Creating and Sustaining Superior Performance. New York: Free Press Quinn, Thomas F and JeanMichel Fally (2010), "Retail Global Expansion: The Journey Starts at Home," Deloitte Development. RetailWire and Dechert Hampe & (2009), "Retail Formats in Transit ion," in The Retail Next Studies. Roth, Martin S. (1995), "The Effects of Culture and Socioeconomics on the Performance of Global Brand Image Strategies," Journal of Marketing Research, 32 (2), 163 175. Rugman, Alan and Stephane Girod (2003), "Retail Multi nationals and Globalization: The Evidence is Regional," European Management Journal, 21 (1), 2437. Rumelt, Richard P. (1982), "Diversification Strategy and Profitability'," Strategic Management Journal. 3 (4), 359369. Sage, Alexandria (2009), "U.S. Retai l Lull Means Prep Time for International Expansion," in Reuters. Sambharya, Rakesh B. (1996), "Foreign Experience of Top Management Teams and International Diversification Strategies of US Multinational Corporations," Strategic Management Journal, 17 (9), 739 46. Shleifer, A. and R. W. Vishny (1992), "Liquidation Values and Debt Capacity a Market Equilibrium Approach," Journal of Finance, 47 (4), 134366. Sharma, A. and I.F. Kesner (1996), "Diversifying entry: Some ex ante explanations for postentry survi val and growth," Academy of Management Journal, 63577. Smith, P. B., S. Dugan, and F. Trompenaars (1996), "National culture and the values of organizational employees A dimensional analysis across 43 nations," Journal of Cross Cultural Psychology, 27 (2), 23164. Sola, Martin (2004), "Three Stage Least Squares and FIML," Vol. 2008.
100 Steenkamp, JanBenefict E. M., Frenkel ter Hofstede, and Michel Wedel (1999), "A Cross National Investigation into the Individual and National Cultural Antecedents of Consumer Innovativeness," Journal of Marketing, 63 (2), 5569. Srinivasan, Shuba and Dominique Hanssens (2009), Marketing and Firm Value: Metrics, Methods, Findings, and Future Directions. Journal of Marketing Research, 46(June), 293312. Tallman, Stephen and Jiatao Li (1996), "Effects of International Diversity and Product Diversity on the Performance of Multinational Firms," Academy of Management Journal, 39 (1), 17996. Vachani, Sushil (1991)," Distinguishing between Related and Unrelated International Geogr aphic Diversification: A Comprehensive Measure of Global Diversification," Journal of International Business Studies, 22 (2), 307322. van Oudenhoven, J. P., L. Mechelse, and C. K. W. de Dreu (1998), "Managerial conflict management in five European countr ies: The importance of power distance, uncertainty avoidance, and masculinity," Applied Psychology an International ReviewPsychologie AppliqueeRevue Internationale, 47 (3), 439 55. Vermeulen, F. and H. Barkema (2002), "Pace, rhythm, and scope: Process dependence in building a profitable multinational corporation," Strategic Management Journal, 23 (7), 637 53. Waldman, D. A., M. S. de Luque, N. Washburn, R. J. House, B. Adetoun, A. Barrasa, M. Bobina, M. Bodur, Y. J. Chen, S. Debbarma, P. Dorfman, R. R. Dz uvichu, I. Evcimen, P. P. Fu, M. Grachev, R. G. Duarte, V. Gupta, D. N. Den Hartog, A. H. B. de Hoogh, J. Howell, K. Y. Jone, H. Kabasakal, E. Konrad, P. L. Koopman, R. Lang, C. C. Lin, J. Liu, B. Martinez, A. E. Munley, N. Papalexandris, T. K. Peng, L. Pr ieto, N. Quigley, J. Rajasekar, F. G. Rodriguez, J. Steyrer, B. Tanure, H. Thierry, V. M. Thomas, P. T. van den Berg, and C. P. M. Wilderom (2006), "Cultural and leadership predictors of corporate social responsibility values of top management: a GLOBE stu dy of 15 countries," Journal of International Business Studies, 37 (6), 82337. Wernerfelt, Birger and Cynthia A. Montgomery (1988), "TobinQ and the Importance of Focus in Firm Performance," American Economic Review, 78 (1), 246250. Williams, DE (1992), "Motives for retailer internationalization: their impact, structure, and implications," Journal of Marketing Management, 8 (8/9), 824. Williams, RJ, JJ Hoffman, and BT Lamont (1995), "The Influence of Top Management Team Characteristics on M Form Implementation Time.," Journal of Managerial Issues, 7 (4). Williamson, Oliver E. (1964), The Economics of Discretionary Behavior: Managerial Objectives in a Theory of the Firm, Englewood Cliffs, New Jersey: PrenticeHall.
101 Wooldridge, Jeffrey M. (2002), Econometric analysis of cross section and panel data. Cambridge, Mass.: MIT Press. World Bank (2007), World Development Indicators, http://www.worldbank.org/data/. Wrigley, Neil, Neil M. Coe and Andrew Currah (2005), "Globali zing Retail: Conceptualizing the DistributionBased Transnational Corporation (TNC)," Progress in Human Geography 29 (4), 437457 Zaheer, Srilata (1995), Overcoming the Liabilities of Foreigness," Academy of Management Journal, 38 (2), 34363.
102 BIOGRAPHICAL SKETCH Born in 1981 to parents Tow Seng Lim and Hui Phing Low, Jeremy Mianxin Lim attended The University of Florida from 2006 to 2011 and received his PhD in Business Administration in the fall of 2011. Prior educational experienc e include: a Master of Arts in Economics from Queens University in 2005 and a Bachelor of Science from The University of Michigan Ann Arbor in 2002. Jeremy Mianxin Lim now resides in Oakland Township, Michigan.