DETERMINANTS OF THE SUCCESS OF GLOBAL AND LOCAL BRANDS IN LATIN AMERICA

The purpose of this paper is to address the issue of the implementation of global and local brands in Latin America by drawing on contingency theory to develop and test hypotheses relating to how product category characteristics affect the success of global and local brands in the region. Hypotheses are tested using data obtained from top brands rankings reported in five Latin American markets (Argentina, Brazil, the Caribbean and Central America, Chile and Mexico). The study design considers estimating a logistic regression on a binomial dependent variable measuring whether 475 top brands are global or local brands, with product category characteristics as independent variables. Results reveal that product categories related to subscriptions, local tastes, high-tech, and global citizenship do have an impact on the success of global and local brands in Latin America.


INTRODUCTION
In the 1990s, with increased globalization and the opening of bilateral trade agreements with other countries, the Latin American markets were enhanced with a huge number of products and brands available for sale.Currently, markets in the region are rapidly integrating across borders, and for many product categories, consumers are able to choose between global and local alternatives.
Additionally, consumers in emerging countries are experiencing improved standards of living and lifestyle and exhibiting a desire to consume global brands (Bhardwaj, Park, & Kim, 2011).It has been more than thirty years since Theodore Levitt (1983) wrote about global brands as the future of firm success.Global brands were supposed to sweep away national differences.However, they have not reached the promise that Levitt envisioned (Pitta & Franzak, 2008).Although the automotive, computer, and electronics product categories, among others, are well-known for their strong global brands, many product categories are still characterized by their local brands in Latin America.Indeed, in most top brand rankings published by research agencies at the national level, top positions feature a mixture of global and local brands.
However, consumers may prefer either local or global brands, depending on contextual factors other than brand ownership, such as the particular product category.Whereas global brands may be preferred for some product categories, local brands may be favored for others.In fact, Winit et al. (2014)  but also provides managerial implications by emphasizing that marketers must consider product category characteristics before selecting between global and local brands.

CONCEPTUAL FRAMEWORK
Figure 1 shows the conceptual framework for this study.The sources of global brand value and the processes through which global brands contribute to firm value differ systematically across product categories and brands (Bengtsson, Bardhi, & Venkatraman, 2010;Steenkamp, 2014).Global and local associations operate in a manner very similar to brand associations in the transference of perceptions (Iversen & Hem, 2011).A brand's perceived globalness had an important effect on consumer evaluation (Punyatoya, 2013).However, a wide range of contextual factors could exert influence on consumption choices and contribute to within-country heterogeneity (Douglas & Craig, 2011).Consumers may prefer either local or global brands, depending on contextual factors such as the particular product-category.This section reviews the literature to derive hypotheses about the effects of product category characteristics on the success of global and local brands.
Firms in different industries show different levels of motivation for internationalization (Zeng, Zeng, Xie, Tam, & Wan, 2012).It is possible that the success of global and local brands varies significantly when the type of product is considered.Schuling and Kapferer (2004) note that the push toward development of global brands has been driven more by supply-driven considerations linked to costs than by market considerations.In the 1980s and 1990s, Japanese firms (e.g., Toyota, Panasonic) applied production quality controls and scale efficiencies to market standardized products across the world.American and European firms (e.g., Procter & Gamble, Unilever, L'Oréal, Nestlé) transferred profit and loss responsibility from individual country managers to worldwide strategic business units, becoming truly global firms in the process (Quelch, 2003).When many firms went global, they took the products that were successful in their domestic markets and began selling them in foreign markets.Some were successful and quite literally created the product category in the host countries (Pitta & Franzak, 2008).A key advantage of globalization is firms' opportunity to benefit from strong economies of scale.Standardized brands can generate significant cost reductions in all areas of the business system, including research and development, manufacturing, and logistics (Craig & Douglas, 2000;Levitt, 1983).However, many product categories such as hotels, electricity distribution, gas distribution, fast food restaurants, department stores, and supermarkets demand a strong local presence (e.g., buildings, production facilities, stores for operations).Consequently, significant opportunities could exist for local (vs global) brands in product categories that require a local presence, a local production.Hence:  When switching costs are substantial or the switching processes especially painful, dissatisfied customers are likely to maintain business relationships with existing service providers and resist the dissolution of the relationship.Schuling and Kapferer (2004) provide information on the usage of local and global brands.
They show higher ratings for local brands (42.9%) than for global brands (37.4%).However, the usage intention indicates a different pattern: ratings are slightly higher for global brands (47.5%) than for local brands (46%).This might indicate that consumers are attracted to global brands but that, in reality, they buy local brands.This difference between usage and usage intention could be higher in product categories related to subscriptions (e.g., pension fund management, retail banking, and telecommunications services).
The subscription is a business model where the customer must pay a subscription price to have access to the product/service.Generally speaking, these are product categories with which the customer, once a provider has been chosen, does not switch brands unless a specific action, which entails costs for both the firm (marketing effort) and the customer (e.g., psychological, financial costs), merits the switch (Hidalgo, Manzur, Olavarrieta, & Farías, 2007;Venkatesan & Kumar, 2004).Hence: H2: Product categories related to subscriptions are positively associated with local brand success.
Research in social psychology has identified cues of trustworthiness in face-to-face situations (e.g.eye contact, lateral flexion of head).Customers in several product categories tend to build a face-to-face relationship with the brand employees.
In many product categories (e.g., airlines, health care services) local brands compete on the basis of well-established face-toface relationships with their customers (Dawar & Frost, 1999).In several product categories, local brands represent many years of face-to-face relationships and customers have developed close relationships with these brands over the years.In these product categories, local brands hold important advantages, such as long-standing emotional links that represent a critical proximity to customer values (i.e., being a brand people grew up with) and high awareness levels (Kapferer, 2002;Wolfe, 1991).Hence: The signaling theory holds that by manipulating signals, firms use brands to inform consumers about product positions.Thus, local brands "signal" respect for and unique fit into cultural traditions, and pride in representing the local economy.Local brands are frequently seen as more authentic and down-to-earth, thus offering a more intimate basis for nurturing consumer-brand relationships.Local brands may be perceived as local icons, associated with local culture, tradition, and country (Winit et al., 2014).Local brands are not without their strengths, most notably, in their responsiveness to local needs, including those needs perhaps not precisely met by global brands (Schuling & Kapferer, 2004), and by virtue of their close connection to the local cultural narrative (Steenkamp, Batra, & Alden, 2003).
Culture, it seems, is harder to standardize than economic organization and technology.Consumers attribute personalities to brands that fit their own cultural values, not the values of the producer of the brand (Anand & Delios, 2002;Becker-Olsen, Taylor, Hill, & Yalcinkaya, 2011;Farías, 2007;Hofstede, 2001;Manzur, Uribe, Hidalgo, Olavarrieta, & Farías, 2012;Mooij & Hofstede, 2010).Several product categories as food (breakfast, meal solutions, frozen and fresh foods, snacks, cookies and candy), beer, media, and mineral water, could be local maybe because of a better response to local needs, flavors and tastes (e.g., advertisements, ingredients).The complexity of food, in particular, is noteworthy.Food preferences vary dramatically over the globe because of basic differences in local tastes, customs, and access to fresh, local products.It would be strange if curry powder, which is part of the Malay and Indian culture, were to use a "Western" brand name and image to make it more commercial.
Generally, a product that is closely related to its culture must use local branding strategy (Douglas et al., 2001;Harun, Kassim, Igau, Tahajuddin, & Al-Swidi, 2010;Kapferer, 2002;Wolfe, 1991).Hence: The importance of the trust construct has already been demonstrated in sustaining buyer and seller relations.Trust is an attitudinal or mindset construct.It is the consumer's belief about whether or not the brand keeps its promises.Trust is defined as a willingness to rely on an exchange partner in whom one has confidence.Consequently, trust is a major determinant of relationship commitment (Moorman, Deshpandé, & Zaltman, 1993).It represents the confidence that the relational party in an exchange will not exploit the other's vulnerability.Accordingly, to trust a brand implicitly means that there is a high probability or expectancy that the brand will result in positive outcomes for the consumer.A trustworthy brand is one that consistently keeps its promise of value to consumers through the way the product is developed, produced, sold, serviced and advertised.In Europe, Schuling and Kapferer (2004) documented that the image of trust is significantly stronger for local brands than for global brands (22.1% vs. 17.9%), and local brands (22.1%) benefit more from a significantly stronger image of reliability than do global brands (17.9%).Similarly, Pinar, Girard, and Eser (2012) show that brand image is significantly higher for local banks than for global banks.Trust reflects cumulative effects over time on loyalty in high-involvement, high-service product markets.Consequently, significant opportunities could exist for local brands in product categories that require image of trust (e.g., financial services, retail banking, nonprofit organizations, medicines, health care services).Therefore: H5: Product categories related to trust are positively associated with local brand success.
In China, consumers perceived global brands as high quality products, durable, innovative, and customer-oriented product design.Chinese consumers perceived local brands as low quality products, low trustworthiness issue, and poor product design (Ar & Kara, 2012;Jap, 2010;Laforet & Chen, 2012).Firms from emerging markets are not known for having the most innovative technology, superior human capital, or worldwide-recognized brands.They may also experience shortages of financial capital, which attenuates the implementation of marketing strategies (Magnusson, Westjohn, & Zdravkovic, 2011;Wright, Filatotchev, Hoskisson, & Peng, 2005).In Latin American countries, several studies have demonstrated that the brand is seen by customers as a sign of product quality, assisting them to make their purchasing decisions (e.g., Hidalgo & Farías, 2006;Hidalgo et al., 2007;Lopes, Garcia, Santos, & Schiavo, 2013;Manzur, Hidalgo, Olavarrieta, & Farías, 2005, 2011;Merino & Gonzalez, 2008;Milberg & Sinn, 2008;Olavarrieta, Hidalgo, Manzur, & Farías, 2006, 2012;Olavarrieta, Manzur, Hidalgo, & Farías, 2008;Rojas-Méndez, Erenchun-Podlech, & Silva-Olave, 2004;Torres, Hidalgo, & Farias, 2007;Zanette, Lourenço, & Brito, 2013).The electronics, computer, and automotive product categories have global brands, and these products appear to be more universal in terms of how they are used and how they are viewed.Computers and electronics, in particular, were also in earlier product life cycle stages when they became available around the world (Pitta & Franzak, 2008).Jaffe and Martinez (1995) found that Mexicans rate American and Japanese electronic brands much more positively than Mexican electronic brands.Consequently, in Latin America small opportunities could exist for local (vs global) brands in high-tech product categories.Hence: H6: Product categories related to high-tech are negatively associated with local brand success.
In the 1980s, technology opened borders digitally, if not physically, initially with satellite communication, then with wired and wireless transmissions.Media (e.g., television, newspapers, internet) showed lifestyles, reflecting a quality of life, inspiring a demand for something better (Pitta & Franzak, 2008), and increasing similarity in lifestyles across the world (Alden, Steenkamp, & Batra, 1999).A global identity means that consumers feel they belong to the global community and identify with a global lifestyle.Global brands carry an "exporting" culture, attract the "cosmopolitan" elite, and induce the aspiration to live in a global consumer community.Hall (1991) describes a growing global mass culture as driven by Western societies in terms of technology, capital, techniques, and advanced labor.With images and stories emanating from the West, this global mass culture "is centered in the West and it always speaks English" (Hall, 1991, p. 28).Hall fears that minority cultures or cultures in developing countries are likely to follow Western culture in the process of globalization (Maynard & Tian, 2004).However, scholars have suggested that globalization is not necessarily a one-way process flowing from the West to the rest (Hamilton, 1994;Hofstede, 2001;Inglehart & Baker, 2000).They argue that the development of a global economy does not lead to an increasing level of uniformity of Western culture across the world; instead, it contributes to the continual development of cultural diversity.
In developing countries, individuals associate global brands with having symbolic meanings, which enhances the emotional reward such as a sense of pleasure and happiness upon using these brands (Batra, Ramaswamy, Alden, Steenkamp, & Ramachander, 2000;Bhat & Reddy, 1998;Kinra, 2006).
Findings also suggest that consumers in less developed countries associate global brands with an aspiration toward the lifestyles of more advanced economies (Alden et al., 1999).Consequently, in developing countries, a brand's country of origin not only serves as a "quality halo" or summary of product quality (Han, 1989), but also possesses a dimension of non-localness that, among some consumers and for some product categories, contributes to attitudinal liking for status-enhancing reasons (Batra et al., 2000).Hence:

METHODOLOGY
In Latin America, there were five studies that reported rankings of brands published over the past years.Such reports were used in our study -an attempt to obtain more recent ones was unsuccessful.Consequently, the present study includes the 100 top brands reported in Argentina (Centro de Estudios de Opinión Pública, 2009), the Caribbean and Central America (Fernández et al., 2009), Chile (TheLab Y&R, 2010) and Mexico (Millward Brown, 2009), and the 75 top brands reported in Brazil (Sart Dreamaker, 2009), i.e., the analysis is set in the five largest Latin American markets.Spanish is the main language in the majority of Latin American countries.Portuguese is spoken primarily in Brazil, where it is both the official and the national language.
Consequently, for each top brand reported, three codersgraduate students -fluent in Spanish and Portuguese, were instructed to code for the following items: global or local brand and product category (See Exhibit 1).A total of 39 product categories were used.These were: alcoholic beverages, airlines, beer, breakfast (excluding cereals), cars, cereals, credit cards, clothing, computers, conglomerates, department stores, electronics, energy distribution, fast food restaurants, financial services, frozen and fresh foods, health care services, home appliances, home care, hotels, internet websites, jewelry and accessories, meal solutions, media (television, radio, newspapers, magazines), medicines, mineral water, nonprofit organizations, pension fund management, personal care, soccer teams, soft drinks, snacks, cookies and candy, retail banking, sportswear, supermarkets, telecommunications services, tobacco, and other brick-and-mortar retailers.All disagreements in coding (overall agreement > 95%) were discussed among the three coders, and a final consensus was reached (Szymanski & Henard, 2001).
In order to examine the source of variations in the success of local and global brands, a detailed measurement of specific product category characteristics (local production, subscriptions, face-to-face relationships, local tastes, trust, high-tech, global citizenship) would be needed.These measures were obtained through three new coders -graduate students -as no secondary data sources comprehensively cover the large array of product categories included in this study (See Exhibit 1).For each product category, the coders were shown the name of the product category and a picture of a typical product.All disagreements in coding (overall agreement > 80%) were discussed among the three coders, and a final consensus was reached (Szymanski & Henard, 2001).Therefore, all variables used in this study were categorical measurements with two categorical levels.Table 1

RESULTS
Discriminant function analysis is very similar to logistic regression, and both can be used to answer the same research questions.
Logistic regression does not have as many assumptions and restrictions as discriminant analysis.Logistic regression is the common choice nowadays, since the assumptions of discriminant analysis are rarely met (Cohen, Cohen, West, & Aiken, 2003).
Consequently, in order to test the hypotheses, the study design considers estimating a logistic regression on a binomial dependent variable measuring whether top brands are global or local brands with product category characteristics as independent variables.
Specifically, standardized regression estimates were obtained by regressing local brand success on the independent variables -local production, subscriptions, face-to-face relationships, local tastes, trust, high-tech, global citizenship, and the markets (Argentina, Brazil, Chile, Mexico) as control variables.A positive regression coefficient means that an increase in its associated variable will increase (decrease) the probability of local (global) brand success.On the other hand, a negative regression coefficient means that an increase in its associated variable will increase (decrease) the probability of global (local) brand success.
The parameters are estimated using maximum likelihood, employing the binomial logistic regression procedure of IBM SPSS Statistics 21.These are reported in Table 2. To begin, the variance inflation factors for each regression coefficient ranged from a low of 1.092 to a high of 3.012, suggesting that the variance inflation factors were at acceptable levels (Hair, Black, Anderson, & Tatham, 2006).
Since no particularly strong multicollinearity among the independent variables is found, all of them are included in the final model.
As can be seen, the overall regression is significant The hypothesized model has a classification hit rate (percentage of correct classification) of 79.4% (See Table 3).
Press's Q Statistic is used to compare the number of correct classifications with the total sample size and the number of groups.
As the value (163.88)exceeds the critical value (6.63), prediction accuracy exceeds the value assigned by chance.Consequently, the prediction accuracy is greater than that expected by chance.
In the light of these test results, the hypothesized model appears to have satisfactory explanatory and predictive abilities.
The hypothesized model was compared to alternative models (considering interactions between product category characteristics and markets).The hypothesized model demonstrated a better model fit than the alternative models because results show no significant interaction effects between product category characteristics and markets, and chi-square and Nagelkerke's R 2 in the alternative models are slightly larger than that of the hypothesized model (Zheng, Yang, & McLean, 2010).
If there are no significant interaction effects, the hypothesized model could be valid for more than just Latin American countries.

Managerial implications
The

Future research directions
The findings of this study must be viewed in the light of its limitations.First, since this study involved only top brands, the findings may not be applicable to unknown brands.Future research should include more brands, and similar studies should be carriedout in other Latin American countries and/or other regions (e.g., Asia, Europe) to expand the applicability of the findings.
Second, studies that consider the success of global and local brands across different populations and consumer segments in Latin America (and other emerging economies) in order to check for demographic effects (e.g., age, gender, household income) might be another important avenue of research.Additionally, this study failed to identify any cultural factors that would moderate the effect of global and local brands.All consumers within a country are not identical.Indeed, there is substantial variation within a country, and also considerable overlap among different countries.Consumer differences should be incorporated into recommend that future research should consider an extended range of product categories in terms of their characteristics.Whereas much research examines the relationship between brand origin perceptions and attitudes, few studies have focused on the roles played by product category characteristics on global/local brand evaluations and choice.Understanding variations in the success of global and local brands is important to academics focused on identifying their sources, and to managers whose branding strategies may depend on the extent of success of global and local brands within the product categories they manage.The purpose of this paper is to address the issue of the implementation of global and local brands by drawing on contingency theory to develop and test hypotheses relating to how product category characteristics affect the success of global and local brands.The findings of this study are of use to marketing managers in Latin America, whose branding strategies may partially depend on an understanding of the determinants of the success of global and local brands in the region.This research represents a novel attempt to incorporate product category characteristics as key factors affecting global versus local brand success.This research not only contributes to the extant literature,

H1:
Product categories that require a local production are positively associated with local brand success.

Figure
Figure 1.Conceptualization of the study

H4:
Product categories related to local tastes are positively associated with local brand success.
H7: Product categories related to global citizenship are negatively associated with local brand success.
(Chisquare (d.f.= 11) = 220.25,p < 0.01).Product categories related to subscriptions and local tastes have a positive effect on local brand success (p < 0.05).Therefore, H2 and H4 are supported.Product categories related to local production, face-to-face relationships, and trust have no effect on local brand success (p > 0.05).Thus, H1, H3 and H5 are not supported.Product categories related to global citizenship and high-tech are negatively associated with local brand success (p < 0.01).Consequently, H6 and H7 are supported.These results reveal that while four product characteristics (subscriptions, local tastes, global citizenship and high-tech) do have an impact on the success of global and local brands, three product characteristics (local production, face-to-face relationships, and trust) have no effect on the success of global and local brands.Although not the point of interest in this research, and considering the significance of significant values associated with Brazil and Chile (p < 0.01), perhaps an additional comment would be worthwhile about the meaning of these values.The results show that in Brazil and Chile there are significantly more local brands than in the Caribbean and Central America, Argentina, and Mexico.It seems that Mexico and the Caribbean and Central America have more global brands because of its proximity to the United States.Argentina has less local brands maybe because of the relative weakness of the local economy.In contrast, Brazil and Chile had more local brands, maybe because of the strength of their economies, which, in a way, provides for the creation and development of strong local brands.

Table 1 .
Data summary

Table 3 .
Classification accuracy findings of this research are of use to managers in the To date, there is a gap in the literature as to how global and local brands compete in Latin America.This study raises a number of important issues for marketing managers who are reluctant to expand abroad and compete with local or global brands, or who are struggling with their current globalization process.Diagnosing levels of local brand success and differences across product categories is important for managers who need to coordinate brands in a competitive multi-category environment.As such, this research addresses an important concern among researchers and managers regarding how to achieve the success of local and global brands in Latin America.The findings indicate that the success of local and global brands in Latin America is product category-dependent.It follows that a firm's branding strategy should also be product categorydependent in the region.This work indicates that in product categories related to subscriptions and local tastes, significant opportunities may exist for local brands.In contrast, local brand success is significantly reduced in product categories related to high-tech and global citizenship.These results suggest that managers should not develop local brands in product categories closely related to high-tech and global citizenship.The findings of this study provide the basis for several recommendations for international marketing managers in terms of the selection of product categories.Managers can use this knowledge by selecting product categories scoring high on hightech and global citizenship, and product categories scoring low on subscriptions and local tastes.These constructs, though related, exert independent influences.To identify appropriate product categories, each characteristic should be measured and assessed.
not only contributes to the extant literature, but also provides managerial implications by emphasizing that marketers must consider product category characteristics before selecting between global and local brands.Marketers should avoid assuming that the motives underlying global brand success are universal.