international business
queen24
internationalmarketingarticle.pdf
David M. Szymanski, Sundar G. Bharadwaj, & P. Rajan Varadarajan
Standardization versus Adaptation of Internationai Mariceting Strategy:
An Empiricai Investigation An issue debated frequently in the international marketing literature centers on whether a business should pursue a strategy that is standardized across national markets or adapted to individual national markets. Of the two as- pects relating to standardization of marketing strategy across national markets—(1) standardization of the pattern of resource allocation across marketing mix variables integral to a business's marketing strategy and (2) standard- ization of the strategy content with respect to individual marketing mix variables—the latter has been the sub ĵect of numerous conceptual articles. However, there is a relative dearth of empirical studies on both issues. To partially fill this void, this study addresses empirically the question of the standardization of the pattern of resource alloca- tion among marketing mix variables across national markets. The question is addressed by examining whether com- petitive strategy and industry structure variables affect market share and business profits similarly or dissimilarly across Western markets, that is, the U.S., U.K., Canada, and Western Europe. The results reveal that with few exceptions, the effects of competitive strategy and market structure variables generalize across these markets. The study findings provide insights into both the merits of standardizing the strategic resource mix across Western mar- kets and the competitive strategy and market structure variables that are major explanators of business perfor- mance across Western markets.
I N the international marketing literature, the desirability of pur-suing a strategy of standardization of marketing mix and other competitive strategy vadables across national markets versus adaptation to individual national markets has been de- bated extensively (e.g., Ghoshal 1987; Levitt 1983; Walters 1986; Wills, Samli, and Jacobs 1991; Wind 1986; Yip 1989). In recent years, however, the debate centering on the pros and cons of pursuing a strategy of total standardization across national markets versus complete adaptation to individual markets has given way to a more fruitful dia- logue focusing on the (1) desired degree of standardization (or adaptation) with respect to various competitive strategy variables such as branding, advertising, sales promotion, and pricing (Riesenbeck and Freeling 1991) and (2) moder- ating effects of organizational and environmental contin- gencies on the desired degree of standardization (or adapta- tion) with respect to these variables (Quelch and Hoff 1986). However, two related issues that merit further explo- ration in an international context are the degree to which (1) the nature of the underlying relationships between competi- tive strategy variables and business performance are similar across national markets and (2) certain competitive strategy
David M. Szymanski is Associate Professor of Marketing, Department of Marketing, College of Business Administration and Graduate School of Busi- ness, Texas A&M University. Sundar G. Bharadwaj is Assistant Professor of Marketing, Emory Business School, Emory University. P. Rajan \ferada- rajan is the Foley's Professor of Retailing and Marketing, Department of Marketing, College of Business Administration and Graduate School of Busi- ness, Texas A&M University. The authors gratefully acknowledge the many helpful comments of the anonymous JM reviewers and the editor and the research support provided by the Strategic Planning Institute.
variables are relatively more important determinants of per- formance across national markets than other variables.
Realistically, a business competing in or contemplating competing in multiple national markets should first develop an understanding of the nature of the underlying relation- ship between competitive strategy variables (e.g., advertis- ing, personal selling effort, etc.) and performance (e.g., mar- ket share and profitability) to determine whether the pat- terns of these relationships are similar across national mar- kets. Specifically and as explicated in Figure 1, the mar- keting strategy formulation process in multinational firms can be viewed as comprising a series of decisions pertain- ing to the business's (1) strategic orientation (standardiza- tion vs. adaptation), (2) desired degree of standardization of the strategic resource mix (i.e., pattern of resource alloca- tion among advertising, promotion, personal selling, and other marketing mix variables), and (3) the desired degree of standardization of the strategy content (i.e., decisions on product positioning, brand name, appropriate media, con- tent of advertisements, etc.). Making effective decisions re- lating to strategic orientation, resource allocation across stra- tegic variables, and strategy content with respect to individ- ual marketing mix elements, in turn, can be viewed as being contingent on managers' understanding (derived from sources internal and external to the business) of relationships among strategic levers (i.e., marketing mix and other competitive strategy variables under the control of the business that impact performance and also are im- pacted by performance), industry drivers (i.e., industry struc- ture variables that impact performance and in turn can be af- fected at times by a business's performance),yirw drivers (i.e., the characteristics of the firm and business unit, such
Journal of Marketing Vol. 57 (October 1993), 1-17 Standardization Versus Adaptation / 1
as firm-specific and business-specific skills and resources that can impact the strategic levers), and business perfor- mance. Unfortunately, there is a relative dearth of empirical information from sources external to the business on the determinants of business performance across foreign and do- mestic markets (exceptions include Ayal and Hirsch 1982; Craig, Douglas, and Reddy 1987; Douglas and Craig 1983; Kotabe 1990; Kotabe and Omura 1989; Kotabe and Murray 1990; Lecraw 1983; Roth, Schweiger, and Morrison 1991; Samiee and Roth 1992; Samli 1977; Takada and Jain 1991). The lack of sufficient information from external sources persists in many respects in spite of the potential value of infonnation on the determinants of business perfor- mance across multiple national markets, particularly to managers of multinational firms and managers whose firms are entering international markets for the first time.
In an attempt to compensate for this void in the interna- tional marketing literature, we focus on two key questions facing intemational businesses: "Does an opportunity exist for standardizing the strategic resource mix across national markets?" and "On which strategic variables should busi- nesses place relatively greater emphasis across national mar- kets?" To provide insights into tiiese issues, we examine a number of marketing mix variables for their effects on busi- ness profits in Western markets, i.e., the U.S., U.K., Can- ada, and Western Europe. Specifically, we build on the ear- lier cross-sectional work by Douglas and Craig (1983) and Craig, Douglas, and Reddy (1987), which examined the im- pact of marketing mix and marketplace factors on business performance across the U.S. and European markets. We build on these efforts by (1) expanding the set of marketing mix variables to include service quality, product line breadth, vertical integration, advertising expenditures, and shared marketing programs across businesses from the same parent company; (2) controlling for industry drivers and additional competitive strategy variables (i.e., research and development [R&D] expenditures, shared customers, and shared facilities) whose omission from the empirical model might otherwise lead to specification errors; (3) study- ing more than twice as many businesses as those examined by Douglas and Craig (n=1556 vs. n=734), which can pro- duce more accurate estimates of relationship form; and (4) studying businesses that represent a greater array of na- tional markets (i.e., the U.K. and Canada in addition to the U.S. and Western Europe) to gain additional insights into whether marketing mix-business performance relationships generalize across national markets. Furthermore, we ex- amine one slice of the conceptual model in Figure 1 (paths 1 and 2) as the initial study in a programmatic stream of re- search directed at developing a better understanding of the strategic orientation, resource mix, and content that would enable a business to achieve superior performance in the global marketplace.
Generalizability of Performance Relationships Across
Western iVIarkets One way of conceptualizing the relationships among compet- itive strategy, industry structure, market share, and profita- bility is to follow the path model presented in Figure 2. The model represents strategic levers and industry drivers affect-
ing both market share and profits and market share as di- rectly impacting business profits. The face validity of this representation is suggested by numerous empirical studies that over the years have examined the effect of competitive strategy and market structure variables on market share and profits, as well as the numerous empirical studies modeling market share as an antecedent of business profits (see Capon, Farley, and Hoenig [1990]; Szymanski, Bharadwaj, and Varadarajan [1993] for reviews of this literature).
Before delving into the rationale behind the relation- ships depicted in Figure 2, a discussion of whether the de- terminants of business performance in general would be sim- ilar among firms serving the U.S., U.K., Canadian, and West- ern European markets is in order. Though empirical evi- dence focusing directly on this issue is scant, theory sug- gests that business strategies and their effects on firm perfor- mance should generalize across national markets that are similar economically, politically, and culturally (cf. Jain 1989; Levitt 1983; and Takeuchi and Porter 1986). Market similarities on these dimensions hold constant the factors that could moderate the relationship between strategic lev- ers/industry drivers and firm performance. If this perspec- tive is correct, then comparing the economic, political, and cultural characteristics of the U.S., U.K., Canadian, and Western European markets could provide preliminary in- sights into whether we could expect the determinants of busi- ness performance to be similar across these four markets.
Overall, the evidence suggests that the U.S., U.K., Cana- dian, and Western European markets are more similar eco- nomically, politically, and culturally than they are different. The markets exhibit similarly high levels of per capita in- come and gross national product, which Terpstra (1988) sug- gests are important indicators of the demand for consumer and industrial goods, respectively; they display similar lev- els of political stability and government intervention into business practices (Cateora and Keaveney 1987); and many barriers to trade between certain pairs of Western markets have already been or are being eliminated through agree- ments such as the U.S.-Canada Free Trade Agreement, the Single European Act, and other acts of coalition (Dudley 1990; Tillier 1992).
In addition, the four markets can be characterized as dis- playing cultural similarities that add support to the supposi- tion that competitive strategy/market structure relationships with business performance would generalize across West- em markets. For one, the nuclear/am/Z>' (parents and chil- dren) is the rule in Western markets (Terpstra 1988), and the family is the primary agent of socialization among pre- adolescent children (Ward et al. 1986). We might therefore expect to find many similar values and family decision- making strategies across buyers from different Westem mar- kets as found by Douglas (1976, 1978) in her comparisons of consumers in France versus the U.S. Second, consumers in these Westem markets on average are highly and simi- larly educated, which suggests that they may hold similar perceptions and respond similarly to strategic variables (Cateora and Keaveney 1987). The latter possibility is sup- ported in cross-national studies that compared buyers' atti- tudes toward advertising (Anderson and Engledow [1977] compared buyers in the U.S. versus Germany; Barksdale et al. [1982] made comparisons among buyers in the U.S., U.K., Canada, and Norway), buyers' price consciousness
2 / Journal of Marketing, October 1993
FIGURE 1 Marketing Strategy Formulation in a Multinational Firm
Information Bank on Business Performance Relationships by National Markets^-"
Firm Drivers • Financial Resources • Other Firm-Specific Skills and Resources
n n Industry Drivers • Industry Structure • Marketplace Characteristics
Business Performance
T Strategic Levers • Competitive Strategy Variables • Synergistic Relationships
Global Competitive Strategy': Decisions and Outcomes IS
J l l
strategic Orientation
•Total standardization of marketing mix elements across national markets
' Total adaptation of marketing mix elements by national markets
Strategic Resource Mix
Pattern of resource allocation among marketing nnix elements
• Total standardization across national markets
' Total adaptation to individual national markets
(2)
IT BusinessPerformance Strategic Content
Degree of standardization of the content of marketing mix elements across national markets
Element 1 Element 2
High Low High Low
Element N High Low
^ The performance relationships presented as bi-directionai reflect the fact that over time, firm performance couid impact certain firm drivers (e.g., successful firms are able to attract better managers and skilled workers). Industry drivers (e.g., industry structure), and strategic levers (e.g., experience and scale effects that provide an opportunity for the firm to lower prices); and firm drivers (e.g., financial assets) could impact on Industry drivers (e.g., industry structure), and both firm and Industry drivers could impact on strategic levers (e.g., possible pricing points, quality of R&D efforts, and advertising strategies employed by the firm). •> The factors listed under the respective headings are illustrative rather than exhaustive of firm drivers, industry drivers, and strategic levers. For additional insights into industry drivers and strategic levers, see Yip (1989). <= The focus here is limited to marketing strategy Variables.
Standardization Versus Adaptation / 3
FIGURE 2 Path Model Representing the Relationship Among Strateglo Levers,
Industry Drivers, and Business Performanoe
Relative Market Share
strategic Levers: '^I = Quality of customer service ^2 = Product quality '̂ 3 = Product line breadth M = New products ^̂ 5 = R&D expenditures ^6 = Backward vertical integration ^7 = Forward vertical integration ^8 = Advertising expenditures ^9 = Promotion expenditures
= Sales force expenditures = Product price = Shared customers = Shared marketing programs = Shared facilities
Industry Drivers: = Marketing growth rate = Industry concentration
(Anderson and Engledow [1977]; Hester and Yuen [1987], who compared U.S. with Canadian consumers; and Lehmann and O'Shaughnessy [1974], who contrasted U.S. buyers with U.K. buyers), and buyers' valence for product/ service quality (Barksdale et al. [1982]; Colvin, Heeler, and Thorpe [1980], who studied buyers in Germany, the U.K., and Sweden; Lehmann and O'Shaughnessy [1974]; and Lewis [1991], who compared U.S. consumers to those con- sumers residing in the U.K.).
Evidence at the aggregate level also suggests that West- em markets are similar economically, politically, culturally, and in other ways. Huszagh, Fox, and Day (1985) clustered 21 countries (U.S., U.K., Canada, Western European coun- tries, and Japan) on the basis of nine economic and social welfare variables (life expectancy, average work week, per- centage employed in service, consumer price index, unem- ployment, govemment spending per capita, manufacturing
percentage of GNP, urbanization, and private spending percentage of GNP) and found that the clustering algorithm grouped together the U.S., U.K., Canada, and the Westem European countries of Belgium, France, Luxembourg, Neth- erlands, and the former West Germany. Another study by Sethi (1971) clustered 91 countries on the basis of 29 polit- ical, socioeconomic, trade, transportation, communication, biological, and personal consumption variables, and found that the clustering algorithm placed the U.S., U.K., Canada, and most European countries (with the exception of Greece, Italy, and Spain) in the same cluster. Finally, Goodnow and Hansz (1972) sorted 100 countries into low-, moderate-, and high-risk markets to enter using 59 country descriptors (the U.S. was not included in the study), in which a low-risk market was defined as politically stable, noninterventionist vis-a-vis business, not legally restrictive of business, culturally harmonious, and economically devel-
4 / Journal of Marketing, October 1993
oped (Gatignon and Anderson 1988). They found that the U.K., Canada, and the European countries of Austria, Bel- gium-Luxembourg, Denmark, France, Italy, Netherlands, Norway, Sweden, Switzerland, and the former West Ger- many were clustered together as low-risk markets in which to conduct business.
In summary, theory suggests that strategic relationships should generalize across national markets that are similar economically, politically, and culturally; and the available empirical evidence suggests that the U.S., U.K., Canadian, and Western European markets are relatively similar along these three dimensions. We therefore hypothesize that, on av- erage, the effects of strategic levers and industry drivers on business performance would generalize across the four West- em markets. Though it is plausible that, like Douglas and Craig (1983), we may find that certain relationships gener- alize across markets and others do not, recent advances in communication that have led to greater exchange of informa- tion across national markets suggest that Western markets are more homogeneous than before (Archer 1990; Douglas and Craig 1991; Ohmae 1987; Smith 1990), and hence more similarities than differences should characterize the un- derlying relationships between strategic levers/industry driv- ers and business performance across the four Western markets.
The next section provides a rationale for the path rela- tionships represented in Figure 2. Consistent with the hy- pothesis that performance relationships will be the same across the four markets, this discussion is generic (i.e., with- out reference to the four markets).
Strategic Levers, Industry Drivers, and Business Performance
strategic Levers
Product/service quality. Though several studies have uncov- ered a positive relationship between product quality and mar- ket share (Buzzell and Wiersema 1981a, 1981b), differ- ences in how quality is defined (e.g., conformance versus su- perior performance; see Garvin 1988) could influence the re- lationship between quality and market share. When supe- rior quality is defined as greater conformance to product specifications, higher costs and any accompanying higher prices that could reduce sales are not necessarily impera- tive. However, when superior quality is defined as superior or more expensive features, the accompanying higher costs passed on to buyers in the form of higher prices could lead to reduced sales.
The relationship between quality and profit is also equiv- ocal. When increased costs from the use of more expensive components, less standardized procedures, greater emphasis on product innovation to sustain a high-quality position, and higher promotional expenditures to convey a position of superior quality to customers (Farris and Reibstein 1979; Phillips, Chang, and Buzzell 1983) cannot be passed on to customers, profit margins would be squeezed, and quality and profit would be inversely related. On the other hand, consumers may be willing to pay a premium for better qual- ity goods and services (Buzzell and Gale 1987), and supe- rior product quality could also reduce rework and service costs (e.g., warranty costs [Garvin 1988]) to increase mar-
gins as well as protect the business from outside forces that reduce them (e.g., bargaining power of buyers).
Product line breadth. The net effect of offering a broad line of products on market share is expected to be positive for several reasons. A broad product line can reduce the chances of market share erosion from consumers' variety- seeking behavior (i.e., variety seeking can occur within the business's offerings), and/or when consumers are deal prone (i.e., by ensuring that one of the business's offerings is on deal). A broad product line can also minimize market share erosion by acting as an entry barrier (i.e., closing al- ternative product forms, sizes, etc., as entry points; and preempting shelf-space [Rao and Rutenberg 1979]), and it may be preferred by buyers whose needs change with time by lowering their purchase risks and transaction costs be- cause they would be familiar with the general characteris- tics of the business and its offerings.
The impact of product line breadth on the level of busi- ness profits, however, appears to be equivocal. On one hand, the modularization benefits of producing a broad line could allow the business to achieve product differentiation while using common parts across products in the line to help reduce costs (Hayes, Wheelwright, and Clark 1988). Conversely, the impact on profits could be negative be- cause of the increased costs that could result from more com- plex operations (Johnson and Kaplan 1987; Lubben 1988), increased material handling and inventories (Kekre 1987; Lubben 1988; McClain and Thomas 1980), more diverse process fiows (Banker, Datar, and Kekre 1988; Karmarkar 1987), more monitoring of operations (Miller and Vollman 1985), and more resources required for scheduling the shorter production runs commonly associated with produc- ing a broad line of goods (e.g., Abegglen and Stalic 1985; Johnson and Kaplan 1987).
New products. New products can have a favorable ef- fect on market share to the extent that they satisfy customer needs better than the existing goods (Davidson 1976) and prevent competitors from taking away a business's custom- ers with their own new products (Hayes and Abernathy 1980). Though new products can cannibalize the sales of ex- isting products and consume the marketing resources that would otherwise go to them (Hambrick and Schecter 1983), continuous and constant innovation are competitive imper- atives for maintaining and/or building share. Hence, in the aggregate, market share at the strategic business unit (SBU) level can be expected to vary positively with the rate of new product introductions.
The relationship between new product introductions and profits, however, is equivocal. There is evidence to sug- gest that new product introductions could hurt business prof- its in the short run. Developing and introducing new prod- ucts can require large investments in R&D and plant and equipment (Woo 1987) and in advertising and promotion to build consumer awareness. Yet, evidence from studies on ' 'excellent'' companies suggests that new products and prof- its are positively related in the long term (Peters and Water- man 1982; Maidique and Hayes 1984); other evidence sug- gests that new products and profits are positively related when consumers are willing to pay a price premium for su- perior new products that cannot be readily duplicated by
Standardization Versus Adaptation / 5
competitors (Booz, Allen, and Hamilton 1981; Cooper 1986).
Research and development expenditures. R&D directed at both product innovation (e.g., introducing innovative new products) and process innovation (e.g., improving qual- ity and/or lowering production costs) are likely to endow a business with enduring competitive advantages. All else equal, R&D expenditures can be expected to be positively related to market share because innovative new products that match buyer needs more closely are likely to command a larger market share. Innovative new products and improve- ments in existing products resulting from investments in R&D also are less likely to be vulnerable to price wars and more likely to command a price premium. However, since investments in R&D tend to have long gestation periods (Ravenscraft and Scherer 1982), the positive relationship be- tween R&D expenditures and profitability often are more pronounced in lagged R&D-profit relationships than contem- poraneous R&D-profit relationships.
Vertical integration. It seems plausible to expect that vertical integration would be positively related to both mar- ket share and profits. Vertic^ly integrating operations can lead to supply assurances (Amihud and Lev 1981), more consistent product quality (Scherer 1980), shorter turna- round in producing the product and making it available to consumers (Chandler 1977), and higher entry barriers (Buzzell 1983). These factors could translate into more sales and a greater share of the served market.
Vertically integrating operations could also have a posi- tive impact on profits because (1) business profits that would otherwise go to other members of the distribution chain are internalized (Aaker and Jacobson 1987) and (2) the reduced transaction costs often offset the increased costs associated with being more integrated (e.g., coordina- tion costs [Buzzell 1983; Ravenscraft 1983; Williamson 1975]). However, riecent studies focusing on alternatives to vertical integration that offer the benefits of vertical integra- tion without the accompanying costs and disadvantages sug- gest a weak relationship between vertical integration and profitability. Nonintegration strategies—such as maintain- ing close, long-term working relationships with suppliers and long-term contracts—and quasi-integration strategies— such as minority equity investment, strategic alliances, joint ventures, loan guarantees, and prepurchase credits—have been proposed as alternatives to full integration (Hanigan 1984). In fact, a conceptualization of vertical integration as involving operations that are 1(X)% owned and physically in- terconnected to supply 100% of a business's needs by the business itself is viewed as outmoded by some scholars (Har- rigan 1985).
Expenditures on communication elements. All else equal, expenditures on advertising, sales promotion (Assmus, Farley, and Lehmann 1984; TfeUis 1988), and the sales force (Anderson 1988) are conducive to increased sales and market share. Expenditures on these communica- tion elements can increase customer awareness, facilitate the formation of positive attitudes and behavioral intentions among prospective consumers, and ultimately increase sales for the business.
The impact of increased marketing communication ex- penditures on profit, however, is less clear. The sign of the
effect would depend on whether the change in sales more than offsets, exactly offsets, or less than offsets the change in expenditures. All else equal, the lower the expenditure in- curred in realizing a dollar of sales, the higher a business's profits would be. However, marketing communication ex- penditures and profits would not demonstrate a positive re- lationship with profits when the increased sales from spend- ing more on communication elements just offsets or fails to offset the reduction in profit margins that occur when the business spends more. Tlie nature of the underlying relation- ship is further confounded by the degree to which the scale economies associated with increased expenditures on market- ing communications and the scope economies that result from sharing marketing costs with other businesses in the firm's portfolio are captured by the respective business unit.
Product price. An inverse relationship between price and market share is commonly thought to exist. However, the possibility of there being a positive relationship be- tween price and sales when higher prices refiect greater pres- tige of ownership and/or superior quality cannot be com- pletely overlooked (Michael and Becker 1973; Monroe and Krishnan 1984). Hence, establishing the sign of the relation- ship between price and market share is best left as an empir- ical issue, and the same holds true for establishing the sign of the relationship between price and profits. The sign of the relationship between price and profit would depend on the shape of the demand curve (monotonically decreasing, vertical, or monotonically increasing [as in the case with gif- fen goods]). When the demand curve is monotonically decreasing, the relationship between price and profit would also depend on the point along the demand curve at which the business is operating. In the inelastic portion of a mon- otonically decreasing demand curve, total revenues decline with increases in price, whereas in the elastic portion of the demand curve, total revenues more than offset the lost sales from charging higher prices up to the point where the elas- ticity of demand equals one (Henderson and Quandt 1980).
Shared customers, marketing programs, and facilities. All else equal, marketing synergies can be expected to dem- onstrate a positive relationship with market share. For exam- ple, selling products to shared consumers allows buyers to draw on their information bank to form associative evalua- tions of the business's newest offerings (Nayyar 1990). Cus- tomers who have favorable impressions of some of the firm's offerings would be more favorably disposed toward purchasing the offerings of other businesses in the firm's portfolio. Synergies may also allow a business to offer pack- age deals to customers, initiate joint promotions (e.g., mul- tiple coupons in one mailing), and cross-sell to customers of other SBUs in the firm's portfolio.
Businesses could also achieve greater efficiency in cur- rent operations to increase profit margins when they share customers, facilities, and/or marketing programs witfi other business units in the firm's portfolio of businesses (Lecraw 1984; Lemelin 1982; Nayyar 1990; Rumelt 1982; Wells 1984). These synergistic relationships can lead to cost sav- ings through the sharing of knowledge, marketing skills, and marketing programs and resources such as the sales force, warehouses, etc. Woo and Cooper (1985) found the gains accruing to high market share businesses to be greater
6 / Journal of Marketing, October 1993
than the gains accruing to low market share businesses. How- ever, establishing synergies is not costless. There may have to be compromises between formulating optimal marketing programs for individual business units versus developing a common marketing program for a number of businesses; and the search for synergies can increase the complexity of a business's management system because of the need to coordinate competing schedules and maintain an atmos- phere of cooperation (Ghoshal 1987). In the aggregate, how- ever, one would expect the benefits of sharing customers, marketing programs, and/or facilities to outweigh the costs of administering and coordinating these shared "resources."
Industry Drivers
Market growth rate. All else equal, the higher profit po- tential of high-growth markets should make these markets more attractive to nonincumbent businesses. We would therefore expect to witness more businesses leaving low- or no-growth markets to enter high-growth markets (assuming there are no significant barriers to entry). The correspond- ing effect would be that the combined market share of com- peting businesses in the high-growth markets would be dis- persed over a larger number of competitors. Consequently, we would expect to find an inverse relationship between market growth rate and market share.
Whereas operating in high-growth markets may be in- consistent with having high market share, high-growth mar- kets and high profitability can be very consistent. High- growth markets are often characterized by relatively higher gross margins, marketing costs, and prices (that rise at a rate below costs), rising productivity, increased investment to keep pace with growth, low or negative cash flow, more new customers, higher levels of demand per customer, and a higher probability that the product will meet some custom- ers' needs. The net effect of these cost reductions, cost in- creases, rising profit margins, and rising sales that often char- acterize high-growth markets seems to be increased profits (Buzzell and Gale 1987).
Industry concentration. Though by definition, industry concentration and market share are positively related, the positive relationship often exhibited between industry concentration and profit (e.g., Domowitz, Hubbard, and Pe- tersen 1986; Weiss 1974) could result from several factors. The structure-conduct/strategy-perfonnance (SCP/SSP) para- digm, for example, posits that industry structure variables (e.g., concentration) influence a business's conduct (strat- egy) and eventually impact its financial performance (Bain 1951; 1956). Embedded in this SCP/SSP perspective is the view that businesses attempt to control the output in the mar- ket through the exercise of monopoly power or colluding with other businesses to drive up prices and profit. Though the question of whether the association between industry concentration and profitability is attributable to the exercise of market power by incumbent businesses or efficiency dif- ferences among businesses (see Demetz 1973; Gale and Branch 1982; Ravenscraft 1983; Smirlock, Gilligan, and Marshall 1984) is a point of controversy, both viewpoints imply a positive relationship between industry concentra- tion and profits.
Market Share/Profit Relationship
Finally, the relationship between market share and profita- bility is expected to be positive. Efficiency theory predicts that businesses with large market shares are cost-efficient be- cause of experience curve and scale effects that ultimately lead to greater profitability (Day and Montgomery 1983; Buzzell and GaJe 1987). Market power theory posits that businesses with large market shares have the power to ob- tain inputs at lower costs, extract concessions from channel members, and set prices rather than be price takers to in- crease their profits (Martin 1988; Schroeter 1988). The pos- itive relationship between market share and profitability also has been supported in numerous empirical studies (see Capon, Farley, and Hoenig 1990; Szymanski, Bharadwaj, and Varadarajan 1993).
Methodology Among secondary data sources, there seems to be growing recognition that the PIMS (Profit Impact of Market Strat- egy) data base has value for providing initial insights into business phenomena across national markets (Buzzell and Gale 1987; Craig, Douglas, and Reddy 1987; Douglas and Craig 1983; Douglas and Rhee 1989; Manu 1992; Yip 1991). The PIMS data base contains competitive strategy, market structure, and business performance data on thou- sands of business units; and the recent expansion of the in- ternational portion of the data base now makes it even more attractive for conducting cross-national market compari- sons. Moreover, empirical evidence suggests that PIMS- based findings are both reliable and valid. Phillips, Chang, and Buzzell (1983) found that independent observations taken over time from the same respondents measured the same latent variable; Marshall and Buzzell (1990) showed that the PIMS data base and the Federal Trade Communica- tion line of business data produced similar profit relation- ships; and Hagerty, Carman, and Russell (1988) found that marketing mix elasticities generated from the PIMS data base corresponded with the published literature and had pre- dictable relationships with market characteristics. In addi- tion, Rumelt (1991) provides evidence that business unit fac- tors (the level at which the PIMS program gathers data) ac- count for most of the differences in financial performance. Industry structure variables and corporate level factors ac- counted for relatively less of the explained variance in performance.
These attractive features of the PIMS data base factored into our decision to use it for examining the generalizability of performance relationships across Western markets. At the same time, we recognize that the PIMS data base has lim- itations (cf Anderson and Paine 1978; Kerin, Mahajan, and Varadarajan 1990). Overall, PIMS member companies are Fortune 500 U.S. companies and British, Canadian, and Western European companies of similar size. Neither the number of SBUs from individual member companies in- cluded in the data base nor the extent to which various com- panies are already global players is known. Though there is considerable dispersion in the size, strategy, performance, and other characteristics of SBUs in the data base, research- ers have expressed concerns about the appropriateness of generalizing the findings of studies based on PIMS busi- nesses to the larger population of businesses (i.e., SBUs of
Standardization Versus Adaptation / 7
large, medium, and small firms and of non-PIMS compa- nies). In addition, the considerable leeway PIMS member companies have in defining their business units (i.e., a divi- sion, product line, or profit center that produces/markets a well-defined set of related goods that serve a clearly de- fined set of customers, and/or competes with a well-defined set of competitors [Buzzell and Gale 1987, p. 32]) could lead to considerable variance in the level of aggregation at which they provide information on industry structure, busi- ness strategy, and performance variables. Another limita- tion of the PIMS data base is that it pools cross-sectional data from a sample of businesses across different products and markets, which can bring into question the appropriate- ness of analyzing a pooled sample of heterogeneous busi- nesses (Bass, Cattin, and Wittink 1978). Unfortunately, con- fidentiality constraints associated with the PIMS data base preclude testing the appropriateness of all pooling procedures.
Though these and other limitations of the PIMS data base should not be ignored, the similarities in the magni- tude of the estimated relationships often found among PIMS-based and non-PIMS-based studies (Hagerty, Car- man, and Russell 1988; Marshall and Buzzell 1990) sug- gest that, overall, the data base could be viewed as a useful source of information for generating initial insights into whether strategic levers and industry drivers impact busi- ness performance similarly across national markets.
Specifically, we used the SPI4 data base (i.e., four-year averaged data), which has the added feature of controlling better than single-year data for abnormal data fiuctuations because of economic and other conditions that could affect the estimated path coefficients. The sampling frame was fur- ther restricted to include only those business units having 90% or more of their served market located in one Western market; i.e., either the U.S. (n=1168 businesses), U.K. (n=189), Canadian (n=86), or Western European (n=113) market (i.e., Austria, Belgium, Denmark, Finland, France, Germany, Greece, Italy, Netherlands, Norway, Portugal, Spain, Sweden, and Switzerland). Businesses classified in the PIMS program as serving "other countries" or more than one national market were excluded from the sample.
Estimation Procedures
Both the correlation- and regression-based approaches to moderator analysis advocated by Arnold (1982, 1984) were used to examine whether the predictor-criterion variable re- lationships specified in Figure 2 differed by the location of the served (national) market (LOSM). Path analysis in turn was used to estimate the magnitude of the causal relation- ships represented in Figure 2. The value of path analysis over regression per se is that path analysis can provide in- sights into the magnitude of direct effects (i.e., the effect of a predictor variable on the adjoining criterion variable), in- direct effects (i.e., the effect of a predictor on the criterion variable through an intervening variable; for example, the impact of product quality on profits through market share), spurious effects (i.e., the portion of the correlation between two criterion variables caused by the sharing of a common causal variable) and the effect coefficients (i.e., the total ef- fect equal to the direct plus indirect effects [Pedhazur 1982]). The path methodology therefore permits us to exam- ine (1) whether an indirect effect embellishes, diminishes.
or negates an associated direct effect and (2) the degree to which the market share-profit relationship is spurious.
Expressing the Variabies in the Modei
Profit and market share, the two endogenous variables in the performance model, are measured as return on invest- ment (ROI) and relative market share, respectively. ROI is used rather than return on sales because ROI more fully re- lates profits to the assets used in achieving them; and rela- tive market share (a business's market share relative to the combined market share of its three major competitors) is used instead of absolute market share because the sum con- straint (the predicted market shares of individual businesses should add up to 100) and the bound constraint (the pre- dicted market shares of individual businesses should be be- tween zero and 100), two conditions that must be satisfied when using absolute measures of market share, cannot be sat- isfied when the data are pooled across industries as they are here (Varadarajan and Dillon 1982). Relative market share is thought to also capture more comprehensively the scale and bargaining effects associated with a business's relative size in its served market (Buzzell and Gale 1987). The op- erational definitions for the endogenous and exogenous var- iables can be found in Buzzell and Gale (1987) and the PIMS data base manual.
Results The question of whether the relationships among strategic levers, industry drivers, and business performance general- ize across businesses serving the U.S., U.K., Canadian, and Western European markets was examined by viewing the lo- cation of the served market as a potential moderator varia- ble that could affect the strength (i.e., correlation) and/or form (i.e., the magnitude of the estimated regression coeffi- cient) of the hypothesized relationships.
Strength of the Reiationship
To test for the degree to which the LOSM moderates the re- lationships between strategic levers/industry drivers and business performance, the z-transformed correlations be- tween each Xj and y were compared across the four Western markets with a chi-square test with three degrees of free- dom (Arnold 1982; Cohen and Cohen 1983). The respec- tive z-transformed correlations and chi-square test results are reported in Table 1. The findings reported in Table 1 in- dicate that (1) the strength of the variable relationships dif- fered significantly (p < .05) across the respective served markets for only five of the 17 predictor variables (29.4%) when profitability was the performance measure of interest and (2) the strength of the relationships across markets did not differ for any of the predictor variables when market share was the associated factor of interest. Overall, these findings indicate that the LOSM does not moderate the strength of the relationship between marketing strategies/ marketplace conditions and the business's market share performance, and only on an infrequent occassion does the LOSM moderate the strength of the business's respective financial performance relationships.
Form of the Reiationship
Moderated regression analysis (MRA) was used to test whether the LOSM moderates the form of respective ex-
8 / Journal of Marketing, October 1993
TABLE 1 z-Transformed Correlations Between the Respective Exogenous and Endogenous
Variables by Location of the Served iVlarket
Exogenous Variabie
Strategic Levers: Quality of customer service
Product quality
Product line breadth
New products
R&D expenditures
Backward vertical integration
Forward vertical integration
Advertising expenditures
Promotion expenditures
Sales force expenditures
Product price
Shared customers
Shared marketing programs
Shared facilities
industry Drivers: Market growth rate
Industry concentration
iVlari(et Share: Relative market share
Endogenous Variabie
ROI
MS ROI
MS ROI
MS ROI
MS ROI
MS ROI
MS ROI
MS ROI
MS ROI
MS ROI
MS ROI
MS ROI
MS ROI
MS ROI
ROI
MS ROI
ROI
The U.S. Mari(et
.30
.15
.40
.26
.38
.16
-.02 -.07
.08 -.11
.10
.06
.08
.20
.24
.14
.22
.13
.09
.04
.17
.05
.01
.05
-.01 .01
.03 -.07
.05
.07
.02
.38
The U.K. iVlarl<et
.29
.13
.55
.27
.45
.06
-.04 -.08
.01 -.11
.06 -.18
.12 -.07
.31
.23
.21
.19
.16
.01
.27
.14
.05
.10
.03
.08
.01
.01
.14
.37
.20
.41
The Canadian
iUlari(et
.31
.05
.38
.00
.35 -.09
.01
.35
.11
.06
.21
.01
.39
.12
.26 -.03
.19 -.04
.09 -.08
.16
.08
.03
.27
-.01 -.01
-.01 -.05
-.12
.16 -.01
.19
The Western European
iVlarket
.38
.08
.49
.32
.56
.10
-.13 .05
.29 -.10
.05 -.10
.11
.06
.41
.15
.28
.10
-.02 .06
.30
.19
.08
.02
.00
.03
-.09 .05
.13
.18
.33
.34
.84 1.21
3.98 5.73
4.06 6.03
1.36 8.11"
1.93 2.24
3.41 10.58"
7.33 12.30"
4.27 3.91
.40 1.61
2.20 1.27
2.75 2.97
.68 4.00
1.11 .83
1.52 2.18
4.49
13.95" 13.91"
16.10"
3 MS is the relative markel share variable. ROI is the return on investment variable. ^ The z-transformed correlations differed significantly across the four markets at alpha ^ .05
ogenous-endogenous variable relationships specified in Fig- ure 2 (Arnold 1982; Shanna, Durand, and Gur-Arie 1981; Stone 1988). MRA is hierarchical regression with three equa- tions and contingency effects modeled as interaction terms. In a bivariate model with one moderator variable (m), the three equations would be
(t) y = a + b,x; (2) y = a + b|X + (3) y = a + b,x +
; and
By examining the change in explained variance between var- ious pairs of the three models, one can infer whether m is a
moderator variable, and if it is, what type. The change or ab- sence of change in explained variance can be interpreted as follows (see Arnold 1982; Sharma, Durand, and Gur-Arie 1981):
• Step I. If the change in the R^ for equation 3 over equation 2 is statistically significant, conclude that the factor moder- ates the form of the relationship and proceed to Step 2. If the change in explained variance is not significant, go to Step 3.
• Step 2. Detennine whether m is related to the criterion fac- tor. If it is not related to the criterion variable, m is a pure moderator (i.e., interacts with the predictor but is not related to either the predictor or the criterion variable); otherwise m
Standardization Versus Adaptation / 9
is a quasi moderator variable (interacts with predictor and is related to the criterion or predictor variable).
• Step 3. Determine whether m is related to the criterion vari- able. If m is related to the criterion variable and the change in R̂ for equation 2 compared to equation 1 is statistically significant, then m is not a moderator variable. Instead, m is a predictor variable and equation 2 is the appropriate model. On the other hand, if m does not interact with the predictor variable and m is not related to the criterion variable, equa- tion 1 is the appropriate model.
A summary of the MRA findings when market share was the criterion variable and then when profitability was the criterion variable is reported in Table 2. The MRA re- sults strongly suggest that the LOSM does not moderate the predictor-criterion relationships either when market share or profitability is the criterion variable of interest. In both cases, specifying the LOSM as an interaction term did not increase significantly the explanatory power of the respec- tive model. The change in R^ for equation 3 over equation 2 was only 3%, both when market share was the en- dogenous variable and profit was the endogenous variable. Both changes in explanatory power are nonsignificant (p > .50). The MRA was also carried out with only the U.K., Ca- nadian, and Western European subset of businesses to deter- mine whether the disproportionate number of U.S. firms in the sample could be masking the moderating effects of the LOSM. The findings when k=3 were the same as when k=4. The LOSM (k=3) neither moderated nor acted as a pre- dictor of the business's market share or financial perfor- mance. (Further MRA revealed that business type—i.e., in- dustrial versus consumer—neither moderated nor predicted performance effects.)
The failure to uncover significant moderator effects sug- gested that we proceed to Step 3 of the MRA procedure to examine whether the LOSM was related to either market share or profitability. The analysis was carried out first by calculating the change in explained variance for equation 2 versus equation 1 for the respective market share and prof- itability models as reported in Table 2. The change in ex- plained variance when market share and profitability were the criterion factors, respectively, was near zero and nonsig- nificant. Second, the categorical LOSM factor was modeled as a series of dummy variables (n=3 dummy variables) and regressed on market share and subsequently on profitability to ascertain whether the LOSM was related to either market share or profitability. The LOSM was found to explain just .8% (nonsignificant) of the variance in market share and just .2% (nonsignificant) of the variance in profitability.
Because the findings indicate that the LOSM is related to neither market share nor profitability and does not mod- erate the direct effects of strategic levers/industry drivers on business performance, pooling the data across the U.S., U.K., Canadian, and European subsamples was considered appropriate (i.e., the equivalent to equation 1). The decision to pool the business data across markets was also reinforced by the relative similarity in the values of the indirect effects between any variable i and profitability across the four sub- models (see Table 3). Examining the magnitude of the indi- rect effects estimated for each of the four subsamples of busi- nesses reveals that the differences across subsamples are rel- atively small (cf. Slomczynski, Miller, and Kohn 1981).
Pooled Results
The results on the basis of analysis of the pooled sample of businesses reported in Tables 3 and 4 provide insights into the major explanators of business performance when serv- ing the four Western markets. Regarding the market share submodel, the findings (Table 4) revealed that 11 of the 15 paths (73.3%) between the respective exogenous variables and market share were statistically or marginally significant (p < .10). The four nonsignificant paths had backward and forward vertical integration, shared marketing programs, and market growth rate as the predictor variable, respec- tively. Two factors having a significant coefficient (i.e., new products and sales force expenditures) were associated with a negative sign, contrary to a priori expectations.
When profitability was the performance variable of inter- est (Table 3), we found that nine of the 17 path coefficients were statistically or marginally significant. However, five of the variables' signs were opposite to the direction pre- dicted. Forward and backward vertical integration, R&D, ex- tent of sharing of facilities, and market growth rate showed a negative relationship to profitability. In contrast, the find- ings revealed that market share had a positive effect on profit as expected, which was relatively large in magnitude and virtually nonspurious (88.6% of the association was nonspurious).
Findings from an analysis of the pooled data further re- vealed that the direct effect of product quality on profit was reinforced by the indirect effect of this exogenous variable on business performance. They also suggested that the neg- ative and direct effects of R&D and shared operating facili- ties were tempered by the positive and indirect effects that these two factors exerted on profits. Finally, market share and product quality had the greatest effect on business prof- its among the businesses in the pooled sample. The effect co- efficients for market share and product quality were .31 and .20, respectively, whereas the absolute value of the next larg- est effect coefficient was (-).IO for R&D.
Discussion Our findings have implications for two key questions faced by managers of businesses competing in multinational mar- kets: "Does a significant opportunity exist for standardiz- ing the strategic resource mix across national markets?" and "On which competitive strategy variables should inter- national businesses place greater emphasis in their strategic programs?"
Standardization Versus Adaptation of the Strategic Resource Mix
In reference to strategy content, it has been argued that adopting a standardized approach instead of a multidomes- tic approach to serving multinational markets is desirable be- cause sales can be increased by developing a consistent image of the product across national markets; and costs can be lowered by pooling production activities across coun- tries, moving production to low-cost locations without rede- fining the production process, and capturing the economies associated with formulating and implementing a single mar- keting plan (Walters 1986; Yip 1989). Advocates of multido- mesdc strategies, in contrast, point out that because few mar-
10 / Journal of Marketing, October 1993
TABLE 2 Moderated Regression Results for the Market Share and Profit Modeis
Descriptive Statistics
iModel R2 p level AR2 p level Market Share Modei: (1) Market
structure.and competitive strategy^ .28 40.53 .0001
(2) Market structure, competitive strategy, and LOSM"
(3) Market structure, competitive strategy, LOSM, and their interactions
Profit Modei: (1) Market
structure, competitive strategy, and market share*^
(2) Market structure, competitive strategy, market share, and LOSM
.29 34.59 .0001 .01 1.17 >.5O
.32 10.87 .0001 .03 1.02 >.5O
.17 18.80 .0001
.18 16.21 .0001 .01 .89 >.5O (3) Market
structure, competitive strategy, market share, LOSM, and their interactions .21 5.53 .0001 .03 .76 >.5O
^Interpretation: Market structure and competitive strategy variables were regressed as main effects on market share. "LOSM is the "location of the served market" variable. ''Interpretation: Market structure, competitive strategy, and market share variables were regressed as main effects on profitability.
kets are exactly alike, some adaptadon of markedng pro- grams is necessary to ensure that buyer needs are satisfied effecdvely and sales maximized (Quelch and Hoff 1986; Wills, Samli, and Jacobs 1991).
Our findings address the extent to which it would be de- sirable for businesses to standardize their strategic resource mix when serving muldple Westem markets. Failing to find that strategic resource mix effects generalize across markets would suggest that a standardized approach is not likely to be effecdve for serving these markets, whereas finding that performance reladonships generalize across markets sug- gests that a standardized approach could be conducive to su- perior performance. The findings suggest that a multina- donal business that employs a similar pattem of resource al- locadon among markedng mix variables when serving the U.S., U.K., Canadian, and Western European markets would find that, on average, the standardized approach evokes similar performance responses in Westem markets.
Specifically, our findings show that the strength and form of the relationships between the various marketing mix, other competitive strategy, market structure, and business performance factors are relatively similar across the four markets. Hence, businesses may be better off standardizing their strategic resource mix to capture the benefits pur- ported to be associated with a standardized approach to serv- ing muldple nadonal markets. However, further research di- rected at the nature of the underlying reladonship between a multinational business's strategic orientation and its mar- ket share/financial performance is required to illuminate fully this reladonship.
Reiative Emphasis on Competitive Strategy Variabies
In addition to offering insights into the opportunity for em- ploying a standardized resource mix with respect to market-
Standardization Versus Adaptation / 1 1
TABLE 3 Decomposition of Exogenous Variable Effects on Profitabiiity by Location of Served Market and Pooled Sample
The U.S. Market The U.K. Market
independent Variables
Hypothesized Effect on
Profitabiiity Direct Indirect Effect Totai Effect Effect Coefficient r
Direct indirect Effect Totai Effect Effect Coefficient r
Strategic Levers: Quality of customer service Product quality Product line breadth New products R&D expenditures Backward vertical integration Fon/vard vertical integration Advertising expenditures Promotion expenditures Sales force expenditures Product price Shared customers Shared marketing programs Shared facilities
Industry Drivers: Market growth rate Industry concentration
Market Share: Relative market share R 2 ( R 2 adj.) F (p value) MaxVIF= df.
.01
.138
.00
.068
-.01 .03 .04 .02
-.03 -.04"
.078
.01 -.09
.068
.00
.318
.18 (.17) 15.35 (.0001) 1.94
1150
.03
.07
.08 -.02
.03
.01
.00
.03
.02 -.01
.02
.01 -.01
.02
.01
.03
.04
.20
.08 -.08 -.11
.00
.03
.07
.04 -.04 -.02
.08
.00 -.07
.07
.03
.31
.158
.258
.168 -.078 -.118
.068
.02
.148
.138
.04
.05
.05
.00 -.088
.05"
.02
.368
.06
.06 -.06 -.07 -.09 -.188 -.07
.07
.12" -.09
.04 -.01
.09
.00
.138
.06
.328
.05
.10
.10 -.02
.01 -.03
.01
.02
.03
.00
.01
.00
.01
.05
.00
.07
_
.28 (.20) 3.64 (.0001) 2.05
161
.11
.16
.04 -.09 -.08 -.21 -.06
.09
.15 -.09
.05 -.01
.10
.05
.13
.13
.32
.13"
.268
.06 -.08 -.11 -.188 -.07
.238
.198
.01
.14"
.10
.08
.01
.14
.208
.398
The Canadian Market The Western European Market
Hypothesized Independent Effect on Variabies Profitability
Strategic Levers: Quality of customer service + / - Product quality +/- Product line breadth + / - New products + / - R&D expenditures + Backward vertical integration + Fonward vertical integration + Advertising expenditures +/- Promotion expenditures + / - Sales force expenditures +/- Product price + / - Shared customers + Shared marketing programs + Shared facilities +
Industry Drivers: Market growth rate + Industry concentration +
Market Share: Relative market share +
R2 ( R 2 adj.) F (p value) Max VIF<= df.
Direct Effect
.18 -.06 -.11 -.398
.03 -.04 -.04
.01
.11 -.16
.16
.358 -.15 -.05
-.12 .00
.16
.31 (.13) 1.76 (.05) 2.35
68
indirect Effect
.02
.04
.03 -.02
.01
.01
.06
.03 -.02 -.02 -.01
.01 -.02
.01
.00
.04
Effect Coefficient
.20 -.02 -.08 -.41
.04 -.03
.02
.04
.09 -.18
.15
.36 -.17 -.04
-.12 .04
.16
Totai r
.05
.00 -.098 -.348
.06
.01
.12 -.03 -.04 -.08
.08
.278
.00 -.05
-.12 -.01
.19"
Direct Effect
.02
.22" -.04
.01 -.208 -.148 -.14" -.01 -.05 -.10
.07 -.03
.01 -.10
.02
.238
.288
indirect Effect
.03
.04
.09 -.03
.05
.02 -.01
.03
.06 -.06
.05
.00
.01 -.03
.01
.06
.28 (.16) 2.20 (.008) 2.24
95
Effect Coefficient
.05
.26
.05 -.02 -.15 -.12 -.15
.02
.01
.04
.12 -.03
.02 -.13
.03
.298
.28
Totai r
.18"
.318
.10 -.05 -.10 -.10 -.06
.15
.10
.06
.198
.02
.03
.05
.13
.328
.33
12 / Journal of Marketing, October 1993
TABLE 3 Continued
Pooled Sample
Independent Variables
Strategic Levers: Quality of customer service Product quaiity Product iine breadth New products R&D expenditures Backward verticai integration Forward vertical integration Advertising expenditures Promotion expenditures Saies force expenditures Product price Shared customers Shared marketing programs Shared facilities
Industry Drivers: Market growth rate Industry concentration
Market Share: Reiative market share
R2 ( R 2 adj.) F (p value) Max ViF<: df.
Hypothesized Effect on Direct
Profltabiiity Effect
+/- .02 + / - .13a + / - -.03 +/- -.06a + -.13a + -.04'' + -.04''
+/- .03 +/- .03 + / - -.03 + / - -.02 + .07a + .02 + -.07a
+ .07a + .01
+ .3ia
.17 (.16]
Indirect Effect
.04
.07
.08 -.02
.03
.01
.01
.03
.02 -.02
.02
.01 -.01
.02
.01
.05
-
1 18.80 (.0001) 1.83
1525
Effect Coefficient
.06
.20
.05 -.08 -.10 -.03 -.03
.06
.05 -.05
.00
.08
.01 -.05
.08
.06
.31
Total r
.14a
.24a
.13a -:07a -.10a
.01
.01
.15a !i3a .03 .07a .06 .02
-.05a
.05a
.07a
.35a
a Significant at aipha ^ .05 '' Significant at aipha > .10 "= Max VIF (i.e., the maximum variance infiation vaiue associated with any one of the predictor variables in the respective model) provides a formal method for detecting the presence of multicoiiinearity. A Max ViF in excess of 10 is often taken as an indication that multicoilinearity may be unduly influencing the ieast squares estimates (Neter, Wasserman, and Kutner 1990).
ing mix variables for serving Westem markets, our findings provide managers with insights into the factors that multina- tional businesses should emphasize in their marketing strat- egies. Specifically, offering a broad product line and selling high-quality products with high levels of customer service seem to be especially conducive to superior market share per- formance. Selling high-quality goods, integrating vertically, focusing on new products, following a strategy of increas- ing market share, sharing customers among business units, and competing in high-growth markets appear to be condu- cive to superior financial performance. The sign of the rela- tionship between R&D expenditures and profitability re- ported in Table 3 is contrary to a priori expectations, and this finding should be weighed against the fact that invest- ments in R&D are often characterized by relatively long ges- tation periods and investment-profitability lags (Raven- scraft and Scherer 1982) that could not be captured in this study.
The findings also lend support for efficiency and mar- ket power theories (Day and Montgomery 1983; Buzzell and Gale 1987; Martin 1988; Schroeter 1988; Staten, Um- beck and Dunkelberg 1988) by demonstrating that the mar- ket share-profit relationship is positive and statistically sig- nificant and nonspurious. Though, in our model, we were un- able to control for all factors that may be related to market share and profitability or capture the underlying relation- ship between these two factors (e.g., focusing on more ma- ture businesses), the findings suggest that on average pursu- ing a strategy of building market share within various West-
em markets can be conducive to achieving superior finan- cial perfonnance. The results further suggest that marketing communications are crucial to a business's market share per- formance in particular, capturing perhaps the reduced price sensitivity (cf. Comanor and Wilson 1967; Nelson 1974, 1975; Klein and Leffier 1981), stronger brand preferences, and brand loyalty (Schultz and Vanbonacker 1978) among buyers often associated with products that are intensely pro- moted. Finally, the findings indicate that the breadth of the product line impacts profits not directly, but indirectly by market share. The latter finding accents the critical nature of fully examining and taking into account both the direct and indirect factors impacting business performance.
Limitations of the Study and Directions for Further Research
The limitations of our study should also be considered when interpreting our findings. Moreover, some of the study's limitations are suggestive of directions for fiirther re- search efforts. For example, future studies should extend the analysis to include additional national markets as well as closely examine the appropriateness of treating Westem Europe as a single segment as we and others have done (Ayal 1981; Anderson and Coughlan 1987; Boyd and Ray 1971; Lillien and Weinstein 1984). Future studies might also (1) adopt a longitudinal approach; (2) expand the set of predictor variables to include other strategic levers such as
Standardization Versus Adaptation / 1 3
TABLE 4 Exogenous Variable Effects on Market Share by Location of the Served ii/larket
Hypothesized Exogeneous Effect on
Variable Mari«tShai
Strategic Levers: Quaiity of customer service +/- Product quality +/- Product line breadth + New product +/o R&D expenditures + Backward verticai integration + Fonward verticai integration + Advertising expenditures + Promotion expenditures + Sales force expenditures + Product price +/- Shared customers + Shared mari<eting programs -t- Shared facilities -t-
industry Driver: Mari<et growth rate +
R^ (R^ adj.) F (p ievel) Max ViF df.
U.S.Maricet
Beta
.12
.22
.24 -.08 .10 .02 .01 .09 .05
-.05 .05 .03
-.01 .07
.04
.26 26.71
1.94 1152
t
3.85a 7.01a 8.72a
-3.06a 3.70a
.72
.38 2.79a 1.55
-1.62 1.98" 1.15 -.49 2.40a
1.35
(.25) (.0001)
U.K.Martet
Beta
.14
.36
.33 -.09 .04
-.08 .05 .09 .06 .04 .04
-.04 .10 .16
.05
t
2.22a 5.13a 4.92a
-1.52 .57
-1.24 .72
1.08 .77 .53 .62
-.52 1.26 2.19a
.72
0.42 (.37) 7.91 (.0001) 2.03
163
Canadian iUlarket
Beta
.13
.28
.16 -.07 .03 .02 .38 .18
-.11 -.06 -.06 .10
-.06 .05
-.01
t
.94 1.94" 1.36 -.67 .24 .18
3.22a 1.24 -.75 -.50 -.52 .84
-.51 .38
-.11
.35 (.21) 2.53 (.005) 2.34
70
European Market
Beta
.12
.19
.31 -.12 .16 .05
-.05 .14 .18
-.20 .13
-.01 .02
-.03
.08
t
1.16 1.78" 3.40a
-1.50 2.13a
.55 -.60 1.55 1.87"
-2.25a 1.32 -.11 .27
-.26
.99
.49 (.42) 6.25 (.0001) 2.15
97
Pooled Sampie
Beta
.12
.22
.25 -.09 .09 .01 .03 .12 .05
-.05 .06 .05
-.02 .05
.03
t
4.60a 8.29a
10.69a -3.97a 4.03a
.53 1.36 4.32a 1.84"
-2.25a 2.50a 1.85" -.62 2.23a
1.21
.28 (.28) 40.53 (.0001)
1.83 1527
asignificant at p ^ 0.05 "Significant at p < 0.10
distribution intensity and firm-specific assets (e.g., manage- ment skills, corporate reputation/image, and corporate/ brand equity) and industry and firm drivers of business per- formance that are not specified in our model; and (3) exam- ine whether the findings based on an analysis of the PIMS businesses generalize to the broader spectrum of non-PIMS businesses.
Also, additional studies might examine more fully the de- sired degree of standardization of individual marketing mix elements across national markets. These implementation is- sues include the content of advertisements and promotions (Cavusgil and Nevin 1981; Gilly 1988), governance struc- ture of channels of distribution (e.g., Anderson and Cough- lan 1987), administration of shared programs, personal sell- ing strategies appropriate within each foreign market, and so on. Though we offer evidence to suggest that the pattern of allocation of resources among marketing mix variables could be standardized across Westem markets, the extent to which the content of marketing strategy and individual mar- keting mix elements should be standardized across national markets or adapted to individual national markets is not ad-
dressed in this study. To varying degrees, adapting the con- tent of marketing mix elements to match the needs of na- tional markets being served can be important to both the short- and long-run success of multinational businesses. For instance, an important macroenvironmental dimension with respect to which differences exist across the four markets ex- amined (but were not examined because of data constraints associated with the PIMS data base) are those regarding the regulations governing the content of advertising, media through which goods and services can be advertised, and types of sales promotion programs that can be employed. Colvin, Heeler, and Thorpe's (1980, p. 73) discussion on im- plementation issues in terms of pattern standardization is particularly pertinent in reference to the latter issue. Here, a strategy is designed from the outset to be susceptible to ex- tensive modification to suit local conditions while maintain- ing sufficient common elements to nunimize the drain on re- sources and management. Examining these and other issues would add significantly to the limited body of empirical ev- idence available that addresses determinants of business per- formance across national markets.
REFERENCES Aaker, David A. and Robert Jacobson (1987), "The Role of Risk
in Explaining Differences in Profitability," Academy of Man- agement Journal, 30 (June), 277-96.
Abegglen, James C. and George Stalk, Jr. (1985), KAISHA, The Japanese Corporation. New York: Basic Books.
Amihud, Yakov and Baruch Lev (1981), "Risk Reduction as a Managerial Motive for Conglomerate Mergers," The Belt Jour- nal of Economics, 12 (Autumn), 605-17.
Anderson, Carl R. and Frank T Paine (1978), "PIMS: A Reexam- ination," Academy of Management Review, 4 (July), 602-12.
Anderson, Erin (1988), "Strategic Implications of Darwinian Ec- onomics for Selling Efficiency and Choice of Integrated or In- dependent SalesfoTces,"Management Science, 34 (May), 599- 618.
and Anne T. Coughlan (1987), "International Market Entry and Expansion via Independent or Integrated Channels of Distribution," Joumat of Marketing, 51 (January), 71-82.
Anderson, Ronald and Jack Engledow (1977), " A Factor Analytic Comparison of U.S. and German Information Seekers," Jour- nal of Consumer Research, 3 (March), 185-96.
14 / Journal of Marketing, October 1993
Archer, Margaret (1990), "Theory, Culture and Post-Industrial So- ciety," in Global Culture: Nationalism, Globalization and Mod- ernity, Mike Featherstone, ed. Newbury Park, CA: Sage Publi- cations, Inc.
Arnold, Harvey J. (1982), "Moderator Variables: A Clarification of Conceptual, Analytic, and Psychometric Issues," Organiza- tional Behavior and Human Performance, 29 (April), 143-74.
(1984), "Testing Moderator Variable Hypotheses: A Reply to Stone and Hollenbeck," Organizational Behavior and Human Performance, 34, 214-24.
Assmus, Gert, John V. Farley, and Donald R. Lehmann (1984), "How Advertising Affects Sales: Meta-Analysis of Economet- ric Results," Joumai of Marketing Research, 21 (February), 65-74.
Ayal, Igal (1981), "International Product Life Cycle: A Reas- sessment and Product Policy Implications," Joumai of Market- ing, 45 (Fall), 91-6.
and Seev Hirsch (1982), "Marketing Factors in Small Country Manufactured Exports: Are Market Share and Market Growth Rate Really Important?," Joumai of International Busi- ness Studies, 13 (Fall), 73-84.
Bain, Joe S. (1951), "Relation of Profit Rate to Industry Concen- tration: American Manufacturing 1936-1940," Quarterly Jour- nal of Economics, 65 (August), 293-324.
(1956), Barriers to Competition. Cambridge, MA: Har- vard University Press.
Banker, Rajiv D., Srikant M. Datar, and Sunder Kekre (1988), "Relevant Costs Congestion and Stochasticity in Production En- vironments," Journal of Accounting and Economics, 10 (April), 171-98.
Barksdale, Hiram C , William D. Perreault Jr., Johan Amdt, J. A. Bamhill, Warren A. French, Michael Halliday, and Jehiel Zif (1982), " A Cross-National Survey of Consumer Attitudes To- wards Marketing Practices, Consumerism and Government Reg- ulations," Columbia Joumai of World Business, 17 (Summer), 71-86.
Bass, Frank M., Phillippe J. Cattin, and Dick R. Wittink (1978), "Firm Effects and Industry Effects in the Analysis of Market Structure and Profitability," Joumai of Marketing Research, 15 (February), 3-10.
Booz, Allen and Hamilton, Inc. (1982), New Product Management for the 1980s. New York: Booz, Allen, and Hamilton, Inc.
Boyd, Harper W. and Michael L. Ray (1971), "What Big Agency Men in Europe Think of Copy Testing Methods," Joumai of Marketing Research, 8 (May), 219-23.
Buzzell, Robert D. (1983), "Is Vertical Integration Profitable?" Harvard Business Review, 61 (January-February), 92-102.
and Bradley T Gale (1987), The PIMS Principles: Link- ing Strategy to Performance. New York: The Free Press.
and Frederik D. Wiersema (1981a), "Successful Share- Building Strategies," Harvard Business Review, 59 (Januaiy- February), 135-44.
and (1981b), "Modelling Changes in Market Share: A Cross-Sectional Analysis," Strategic Management Joumai, 2 (January-March), 27-42.
Capon, Noel, John U. Farley, and Scott Hoenig (1990), "Deter- minants of Financial Performance: A Meta-Analysis," Manage- ment Science, 36 (October), 1143-59.
Cateora, Philip R. and Susan Keaveney (1987), Marketing: An In- ternational Perspective. Homewood, IL: Richard D. Irwin, Inc.
Cavusgil, Tamer S. and John R. Nevin (1981), "The State of the Art in International Marketing: An Assessment," in Review of Marketing, Ben Enis and Ken Roering, eds. Chicago: Ameri- can Marketing Association.
Chandler, A. (1977), The Visible Hand: The Managerial Revo- lution in American Business. Cambridge, MA: Harvard Univer- sity Press.
Cohen, Jacob and Patricia Cohen (1983), Applied Multiple Re-
gression/Correlation Analysis for the Behavioral Sciences. Hillsdale, NJ: Lawrence Erlbaum Associates.
Colvin, Michael, Roger Heeler, and Jim Thorpe (1980), "Devel- oping International Advertising Strategy," Joumai of Market- ing, 44 (Fall), 73-9.
Comanor, William S. and Thomas A. Wilson (1967), "Adver- tising, Market Structure and Performance," Review of Econom- ics and Statistics, 49 (November), 423-40.
Cooper, Robert G. (1986), Winning at New Products. Reading, MA: Addison-Wesley Publishing Company, Inc.
Craig, C. Samuel, Susan P. Douglas, and Srinivas K. Reddy (1987), "Market Structure, Performance and Strategy: A Com- parison of U.S. and European Markets," in Advances in Inter- national Marketing, Vol. 2, S. Tamer Cavusgil, ed. Greenwich, CT: JAI Press, Inc., 1-21.
Davidson, Hugh J. (1976),'Why Most New Consumer Brands Fail," Harvard Business Review, 54 (March- April), 117-20.
Day, George S. and David B. Montgomery (1983), "Diagnosing the Experience Curve," Joumai of Marketing, 47 (Spring), 4 4 - 58.
Demetz, Harold (1973), "Industry Structure, Market Rivalry, and Public Policy," Joumai of Law and Economics, 16 (April), 1- 19.
Domowitz, Ian, Glenn R. Hubbard and Bruce C. Petersen (1986), ' 'Business Cycles and the Relationship Between Concentration and Price-Cost Margins," The Rand Joumai of Economics, 17 (Spring), 1-17.
Douglas, Susan P. (1976), "Cross-National Comparisons and Con- sumer Stereotypes: A Case Study of Working and Non-work- ing Wives in the U.S. and France," Joumai of Consumer Re- search, 3 (June), 12-20.
(1978), " A Cross-National Exploration of Husband- Wife Involvement in Selected Household Activities," in Ad- vances in Consumer Research, Vol. 6, William L. Wilkie, ed. Provo, UT: Association for Consumer Research.
and C. Samuel Craig (1983), "Examining Performance of U.S. Multinationals in Foreign Markets," Joumai of Inter- national Business Studies, 14 (Winter), 51-62.
and (1991), "Spatial Dimensions of Inter- national Markets," in Research in Marketing, Suppl. 5, Avijit Ghosh and Charles A. Ingene, eds. Greenwich, CT: JAI Press, Inc.
and Dong Kee Rhee (1989), "Examining Generic Com- petitive Strategy Types in U.S. and European Markets," Jour- nal of International Business Studies, 20 (Fall), 437-63.
and Christine D. Urban (1977), "Life-Style Analysis to Profile Women in International Markets," Journal of Mar- keting, 41 (July), 46-54.
Dudley, James W. (1990), 1992 Understanding the New European Market. Dearborn, MI: Financial Publishing Inc.
Farris, Paul W. and David J. Reibstein (1979), "How Prices, Ad Expenditures, and Profits Are Linked', Harvard Business Re- view, 57 (November-December), 173-84.
Gale, Bradley and Ben Branch (1982), "Concentration Versus Mar- ket Share: What Determines Performance and Why Does It MaX.- t&Tl" Antitrust Bulletin, 27 (Spring), 83-106.
Garvin, David A. (1988), Managing Quality. New York: The Free Press.
Gatignon, Hubert and Erin Anderson (1987), "The Multinational Corporation's Degree of Control Over Foreign Subsidiaries: An Empirical Tfest of the Transaction Cost Explanation," Jour- nal oflMw, Economics, and Organization, 4 (Fall), 305-36.
Ghoshal, Sumantra (1987), "Global Strategy: An Organizing Framework," Strategic Management Joumai, 8 (September- October), 4 2 5 ^ 0 .
Gilly, Mary C. (1988), "Sex Roles in Advertising: A Comparison of Television Advertisements in Australia, Mexico, and the United States," Joumai of Marketing, 52 (April), 75-85.
Standardization Versus Adaptation / 1 5
Goodnow, James D. and J. Hansz (1972), "Environmental De- terminants of Overseas Market Entry Strategies," Joumat of In- temational Business Studies, 2 (Spring), 33-60.
Hagerty, Michael R., James M. Carman, and Gary J. Russell (1988), "Estimating Elasticities with PIMS Data: Methodolog- ical Issues and Substantive Implications," Joumat of Market- ing Research, 25 (February), 1-9.
Hambrick, Donald C. and Steven M. Schecter (1983), ""nima- round Strategies for Mature Industrial-Product Business Units," Academy of Management Joumat, 26 (June), 231-48.
Harrigan, Kathryn R. (1984), "Formulating Vertical Integration Strategies," Academy of Management Review, 9 (October), 638-52.
(1985), Strategic Flexibitity: A Management Guide for Changing Times. Lexington, MA: Lexington Books.
Hayes, Robert H. and William J. Abemathy (1980), "Managing Our Way To Economic Decline," Harvard Business Review, 58 (July-August), 67-77.
, Steven Wheelwright, and Kim B. Clark (1988), Dy- namic Manufacturing. New York: The Free Press.
Henderson, James M. and Richard E. Quandt (1980), Microecon- omic Theory: A Mathematical Approach. New York: McGraw- Hill, Inc.
Hester, Susan B. and Mary Yuen (1987),' 'The Influence of Coun- try of Origin on Consumer Attitude and Buying Behavior in the United States and Canada," in Advances in Consumer Re- search, Melanie Wallendorf and Paul Anderson, eds. Provo, UT: Association of Consumer Research, 538-42.
Huszagh, Sandra M., Richard J. Fox, and Ellen Day (1986), "Global Marketing: An Empirical Investigation," Columbia Journal of World Business, 21 (Twentieth Anniversary Issue), 31-44.
Jain, Subhash C. (1989), "Standardization of Intemational Mar- keting Strategy: Some Research Hypotheses," Journal of Mar- keting, 53 (January), 70-9.
Johnson, Thomas H. and Robert. S. Kaplan (1987), Relevance Lost: The Rise and Fall of Management Accounting. Cambridge, MA: Harvard Business School Press.
Karmarkar, Uday S. (1987), "Lotsizing, Manufacturing Lead Times and Utilization," Management Science, 13 (March), 419-23.
Kekre, Sunder (1987), "Perfonnance of Manufacturing Cell with Increased Product Mix," HE Transactions, 19 (September), 329-39.
Kerin, Roger A., Vijay Mahajan and P. Rajan Varadarajan (1990), Contemporary Perspectives on Strategic Market Planning. Bos- ton: AUyn and Bacon.
Klein, Benjamin and Keith B. Leffler (1981), "The Role of Mar- ket Forces in Assuring Contractual Performance," Journal of Political Economy, 89 (August), 615-41.
Kotabe, Masaaki (1990), "Corporate Policy and Innovative Behav- ior of European and Japanese Multinationals: An Empirical Investigation," Journal of Marketing, 54 (April), 19-33.
and Janet Y. Murray (1990), "Linking Product and Pro- cess Innovations and Modes of International Sourcing in Global Competition: A Case of Foreign Multinational Firms," Journal of Intemational Business Studies, 21 (Third Quarter), 383-408.
-and Glenn S. Omura (1989), "Sourcing Strategies of Eu- ropean and Japanese Multinationals: A Comparison," Joumal of Intemational Business Studies, 20 (Spring), 113-30.
Lecraw, Donald J. (1983), "Performance of Transnational Cor- porations in Less Developed Countries," Journal of Intema- tional Business Studies, 14 (Spring/Summer), 15-33.
(1984), "Diversification Strategy and Performance," The Industrial Journal of Economics, 33 (Spring/Summer), 179-98.
Lehmann, Donald R. and John O'Shaughnessy (1974), "Differ-
ences in Attribute Importance for Different Industrial Prod- ucts," Joumal of Marketing, 38 (April), 36-42.
Lemelin, A. (1982), "Relatedness in the Patterns of Intra-Industry Diversification," The Review of Economics and Statistics, 64 (November), 646-57.
Levitt, Theodore (1983), "The Globalization of Markets," Har- vard Business Review, 61 (May-June), 92-102.
Lewis, Barbara R. (1991), "Service Quality: An Intemational Com- parison of Bank Customers' Expectations and Perceptions," Joumal of Marketing Management, 7 (January), 47-62.
Lillien, Gary L. and David Weinstein (1984), "An Intemational Comparison of the Determinants of Industrial Marketing Expen- ditures," Joumal of Marketing, 48 (Winter), 46-53.
Lubben, Richard T. (1988), Just-in-Time Manufacturing. New York: McGraw-Hill, Inc.
Maidique, Modesto A. Robert H. Hayes (1984), "The Art of High- Technology Management," Sloan Management Review, 26 (Winter), 1-31.
Manu, Franklyn A. (1992), "Innovation Orientation, Environment and Performance: A Comparison of U.S. and European Mar- kets," Joumal of Intemational Business Studies, 23 (Second Quarter), 333-59.
Marshall, Cheri T. and Robert D. Buzzell (1990), "PIMS and the FTC Line-Of-Business Data: A Comparison," Strategic Man- agement Joumal, 11 (May-June), 269-82.
Martin, Stephen (1988), "Market Power and/or Efficiency," Re- view of Economics and Statistics, 70 (August), 331-5.
McClain, John O. and L. Joseph Thomas (1980), Operations Man- agement. Englewood Cliffs, NJ: Prentice-Hall, Inc.
Michael, Robert T. and Gary S. Becker (1973), "On the New The- ory of Consumer Behavior," Swedish Journal of Economics, 75 (December), 378-95.
Miller, Jeffrey G. and Thomas E. Vollman (1985), "The Hidden Factory,'' Harvard Business Review, 63 (September-October), 142-50.
Monroe, Kent B. and R. Krishnan (1984), "Effects of Price on Sub- jective Product Evaluations," in Perceived Quality: How Con- sumers View Merchandise in Stores, J. Jacoby and J. Olson, eds. Lexington, MA: Lexington Books.
Nayyar, Praveen R. (1990), "Information Asymmetries: A Source of Competitive Advantage for Diversified Service Firms," Stra- tegic Management Joumal, 11 (November-December): 513-9.
Nelson, Philip (1974), "Information and Consumer Behavior," Joumal of Political Economy, 81 (July-August), 729-54.
(1975), "The Economic Consequences of Advertising," Joumal of Business, 48 (April), 213-41.
Neter, John, William Wasserman, and Michael H. Kutner (1990), Applied Linear Regression Models. Homewood, IL: Richard D. Irwin, Inc.
Ohmae, Kenichi (1987), "The Triad Worid View," Journal of Business Strategy, 7 (Spring), 8-19.
Pedhazur, E.J. (1982), Multiple Regression in Behavioral Re- search. Fort Worth, TX: Holt, Rinehart and Winston, Inc..
Peters, Thomas J. and Robert H. Waterman, Jr. (1982), In Search of Excellence: Lessons From America's Best Run Companies. New York: Harper & Row.
Phillips, Lynn W , Dae R. Chang, and Robert D. Buzzell (1983), "Product Quality, Cost Position and Business Performance: A Test of Some Key Hypotheses," Journal of Marketing, 47 (Spring), 26-43.
Quelch, John A. and E. J. Hoff (1986), "Customizing Global Mar- keting," Harvard Business Review, 64 (May-June), 59-68.
Rao, Ram C. and David P. Rutenberg (1979), "Preempting an Alert Rival: Strategic Timing of the First Plant by Analysis of Sophisticated Rivalry," The Bell Joumal of Economics, 10 (Au- tumn), 412-28.
Ravenscraft, David J. (1983), "Structure-Profit Relationships at the Line of Business and Industry Level," Review of Econom-
16 / Journal of Marketing, October 1993
ics and Statistics, 65 (February), 22-3 L and Frederic M. Scherer (1982), "The Lag Structure of
Returns to Research and Development," Applied Economics, 14 (December), 602-20.
Riesenbeck, Hajo and Anthony Freeling (1991), "How Global are Global Brands?" The McKinsey Quarterly, 4, 3-18.
Roth, Kendall, David M. Schweiger, and Allen J. Morrison (1991), "Global Strategy Implementation at the Business Unit Level: Operational Capabilities and Administrative Mecha- nisms," Journal of International Business Studies, 22 (Third Quarter), 369-402.
Rumelt, Richard (1982), "Diversification Strategy and Profita- bility," Strategic Management Journal, 3 (October-Decem- ber), 359-69.
(1991), "How Much Does Industry Matter?" 5/ra- tegic Management Journal, 12 (March), 167-85.
Samiee, Saeed and Kendall Roth (1992), " T h e Influence of Global Marketing Standardization on Performance," Journal of Marketing, 56 (April), 1-17.
Samli, Coskun A. (1977), "An Approach for Estimating Mar- ket Potential in East Europe," Journal of International Busi- ness Studies, 8 (Fall-Winter), 49-54.
Scherer, Frederic M. (1980), Industrial Market Structure and Economic Performance. Chicago: Rand McNally & Co.
Schroeter, John R. (1988), "Estimating the Degree of Market Power in the Beef Packing Industry," Review of Economics and Statistics, 70 (February), 158-62.
Schultz, Randall R. and W. G. Vanbonacker (1978), " A Study of Promotion and Price Elasticity," Paper #657, Institute for Re- search in the Behavioral, Economic and Management Sci- ences, Purdue University.
Sethi, Prakash S. (1971), "Comparative Cluster Analysis for World Markets," Journal of Marketing Research, 8 (August), 348-54.
Sharma, Subash, Richard M. Durand, and Oded Gur-Arie (1981), "Identification and Analysis of Moderator Variables," Journal of Marketing Research, 18 (August), 291-300.
Slomczynski, Kazimierz M., Joanne Miller, and Melvin L. Kohn (1981), "Stratification, Work and Values: A Polish- United States Comparison," American Sociological Review, 46 (December), 7 2 0 - ^ .
Smirlock, Michael, Thomas Gilligan, and William Marshall (1984), "Tobin's q and the Structure-Performance Relation- ship," American Economic Review, 74 (December), 1051-60.
Smith, Anthony D. (1990), "Towards a Global Culture?" in Global Culture: Nationalism, Globalization and Modernity, Mike Featherstone, ed. Newbury Park, CA: Sage Publications Inc.
Staten, Michael, John Umbeck, and William Dunkelberg (1988), "Market Share/Market Power Revisited," Journal of Health Economics, 1 (March), 73-83.
Stone, Eugene F. (1988), "Moderator Variables in Research: A Review and Analysis of Conceptual and Methodological Is- sues," in Research in Personnel and Human Resources Man- agement, Vol 6, G.R. Ferris and K.M. Rowland, eds. Greenwich, CT: Jai Press, Inc.
Szymanski, David M., Sundar G. Bharadwaj, and P. Rajan Var- adarajan (1993), "An Analysis of the Market Share-Profitabil- ity Relationship," Journal of Marketing, 57 (July), 1-18.
Takada, Hirokazu and Dipak Jain (1991), "Cross-National Anal- ysis of Diffusion of Consumer Durable Goods in Pacific Rim Countries," Journal of Marketing, 55 (April), 48-54.
Takeuchi, Hirotaka and Michael E. Porter (1986), "Three Roles of International Marketing in Global Strategy," in Com- petition in Global Industries, Michael E. Porter, ed. Boston: Harvard Business School Press, 111-46.
Tellis, Gerard J. (1988), "The Price Elasticity of Selective De- mand: A Meta-Analysis of Econometric Models of Sales," Journal of Marketing Research, 25 (November), 331-41.
Terpstra, Vem (1988), International Dimensions of Marketing. Bos- ton: PWS-Kent Publishing.
Tillier, Alan (1992), Doing Business in Today's Westem Europe. Lincolnwood, IL: NTC Business Books.
Varadarajan, P. Rajan and William R. Dillon (1982), "Intensive Growth Strategies," Journal of Business Research, 9 (De- cember), 49-64.
Walters, Peter G. (1986), "International Marketing PoUcy: A Dis- cussion of the Standardization Construct and Its Relevance for Corporate Policy," Journal of International Business Stud- ies," 17 (Summer), 55-69.
Ward, Scott, Thomas S. Robertson, Donna M. Klees, and Hubert Gatigon (1986), "Children's Purchase Requests and Parental Yielding: A Cross-National Study," in Advances in Consumer Research, Vol 13, Richard J. Lutz, ed. Provo, UT: Association for Consumer Research, 629-32.
Weiss, L. W. (1974), "The Concentration-Profits Relationship and Antitrust," in Industrial Concentration: The New Learn- ing, H.J. Goldschmid et al., eds. Boston: Little Brown.
Wells, John (1984), In Search of Synergy. Unpublished doctoral dis- sertation. Harvard University.
Williamson, Oliver (1975), Markets and Hierarchies. New York: The Free Press.
Wills, James, A. Coskun Samli, and Lawrence Jacobs (1991), "De- veloping Global Products and Marketing Strategies: A Con- struct and Research Agenda," Journal of Academy of Market- ing Science, 19 (Winter), 1-10.
Wind, Yoram (1986), "The Myth of Globalization," Journal of Consumer Marketing, 3 (Spring), 23-6.
Woo, Carolyn (1987), "Path Analysis of the Relationship Be- tween Market Share, Business-Level Conduct and Risk," Stra- tegic Management Journal, 8 (September), 149-68.
and Arnold C. Cooper (1985), "Corporate Settings of Ef- fective Low-Share Businesses," in Strategic Marketing and Management, H. Thomas and D. Gardner, eds. New York: John Wily and Sons, 465-81.
Yip, George S. (1989), "Global Strategy: In a World of Na- tions?," Sloan Management Review, 31 (Fall), 29-41.
(1991), " A Performance Comparison of Continental and National Businesses in Europe," International Marketing Review, 8 (July), 31-9.
Standardization Versus Adaptation / 1 7
