Ethnocentrism at the Coffee Shop Industry: A Study of Starbucks in Developing Countries

Numerous studies have identified that ethnocentric consumers prefer buying products produced nationally rather than products from overseas. Nevertheless, this study has been focused on big economies, that’s why its application in developing countries is still scarce, especially if we want to know their relationship with a global and foreign brand that operates in these developing countries. On the other hand, this study has been centralized in products, and that is the reason why the studies in hedonic services are also scarce. The objective of this study is to identify the antecedents and consequences of ethnocentrism in hedonic services by applying the variable to a developing country such as Colombia. Considering the cultural and economic importance of Colombian coffee and, regarding a global brand, foreign and hedonic services like Starbucks. This allows the presentation of a theoretical model that forms the basis of an empirical study with a sample of 305 consumers of Starbucks in Bogotá. Via data analysis with Smart PLS 3.0, it was observed that although the Colombian consumer is ethnocentric, they do possess loyalty towards the cited global brand. These results provide us with implications for the management of international businesses as well as with future research pathways.


Introduction
As has been documented in some pieces of research, consumers very often favor the acquisition of nationally produced products above foreign products. In some cases, this is due to the influence that is wielded by the origin of those products on the customer's intention to buy Baughn et al. [1]. Ethnocentrism is one of the variables studied to better understand this behavioral pattern. In the sphere of consumption, ethnocentrism allows us to identify the level of influence that the group of origin exerts on the local consumer in relation to their buying habits and also in their perception of the quality of foreign, Shimp et al. and Sharma et al. [2,3].
Nonetheless, according to de Ruyter et al., Shankarmahesh, and Thelen et al. [4][5][6], literature is scant when seeking information about the implications of the consumer's ethnocentrism as applied to the services sector (SET). According to Vivek et al. [7], ethnocentric tendencies are more at the fore when choosing a service provider (SET) in comparison to tendencies in relation to the selection of tangible products. Contrary to the fact that the marks or signals denoting the country of origin are disappearing from goods, the origin of service providers is indeed one of the factors that influence the intention to purchase.
According to Speece et al. [8], the nationality of the staff the customer interacts with during the presentation of the service and the origin country of that service also impact the consumer's perception and intention of buying, this is the reason why it is necessary to research ethnocentrism in the scope of services.
On the other hand, for Clark [9], international services embrace a diverse range of activities, making it very complex to generalize them. For this reason, through the study of ethnocentrism in services, many psychological factors could be analyzed to know the possible attachment with a particular service, depending on the brand or country of origin of that service, Salciuviene et al. [10].
These international services are accompanied on many occasions by a global brand that allows the identification of the services provider in the world. This global brand sets standard communication aspects through its logo, image, positioning, and final consumer type, Akaka et al. [11]. This allows global brands to be accepted and have more desirability from the local consumer, Özsomer et al. [12] and in this way, they are seen by the consumer as a high-quality brand with higher prestige and more functional and symbolic benefits, the reason why the global character from a brand directly impacts the product or service that comes with it, Dimofte et al., Özsomer, and Xie et al. [13][14][15].
This may become more evident in the hedonic services sector given that these types of services are centered on a consumer experience in which the consumer seeks to obtain a personalized experience that satisfies their need for pleasure, emotion, and entertainment whilst utilitarian services, in contrast, possess a purely functional character, Bigné et al. [16]. According to Pérez et al. [17], catering services are hedonic services that are focused on pleasing consumers and affective responses may overwhelm cognitive responses. The relevance of analyzing this concept in terms of developing countries springs from this fact.
It is also important to point out that in the services context exists a higher perception of hedonic services as those services with a foreign name and for that reason, a preference for those brands, Salciuviene et al. [10].
For this reason, a deeper study of ethnocentrism in hedonic services may provide an academic perspective to be used by multinational service companies wishing to develop their business in various parts of the world. Keeping in mind that companies that commence selling not only products but also services, increase their profitability between 8 to 8.5%, Crozet et al. [18], the growing internationalization of this sector can only serve to increase interest in this study.
Nonetheless, the study of consumer ethnocentrism has concentrated on developed countries in which the importation of products is seen as something negative (especially when the country of origin is a developing country) and for this reason, ethnocentric consumers prefer to purchase national products. However, the purchase of imported products in developing countries can generate a type of symbolic consumption, playing the role of a status symbol and representing a means to stand out from the crowd, Wang et al. [19].
Emerging markets are being presented with high growth of the middle-class consumer, who has greater acquisition power. For this reason, international enterprises have increased their presence in these countries through different commercial activities. Considering this, it is important that marketing specialists provide better attention to particular characteristics from these developing countries and their consumers since the literature has centralized in the study of big economies and not in developing countries. On the other hand, taking into account that in a majority of cases, global brands come from developed countries, it is crucial to understand the differences between the consumption patterns in emerging markets, compared with consumers in developed countries. So, diverse strategies could be generated for these global brands to be successful in emerging countries, Wang et al. [20].
For this reason, the overvaluation from a usually western foreign culture and the corresponding underestimation from the national culture, is a common topic in studies from emerging market's economics, Montero [21]. For Van et al. [22] the not ethnocentric segments with a big size could be attractive market niches for overseas companies, where marketing strategies can be made in the international component of the brand. While for local enterprises, the strategy of emphasizing the internal origin of their products could end up being negative. Meaning that it is necessary for international services enterprises to apply different strategies, depending on the consumer's ethnocentrism in the country where they operate.
With this in mind, the principal objective of this current study is to produce a proposal for SET analysis that enables comprehension of factors that explain consumer ethnocentrism in regard to international companies offering services of a hedonic character in developing countries with a global brand. To this end, we have focused on Starbucks in Colombia. It is worth noting that Starbucks began operations in Colombia in 2014. This North American coffee shop chain offers a brand-named experience that engages its customers in a multi-sensorial and hedonic manner via experiential marketing, Ding et al. [23]. For the purposes of this investigation, Colombia is considered a developing country, owing to the fact that in 2018, it was placed last in social inequality; it is a country in which it takes 11 generations for a family to attain the average income, OECD [24]. On the other hand, the per capita GDP of Colombia, according to the World Bank, is 6667 US dollars and according to the United Nations Development Program (UNDP) of 2017, the Human Development Index (HDI) of Colombia was 0.747.
According to The International Coffee Organization [25], by May 2019, Colombia occupied the third place as a world coffee exporter, where its total production for the harvesting year 2018/2019 is estimated at 8.2 million sacks, 2.7% more than the previous period 2017/2018. It is also important to note that the USA is the main receptor of this Colombian coffee, with 46.5% of all exportations from this country, this is the reason why coffee is a very important product in the Colombian economy.
Hwang et al. [26] consider that consumer loyalty allows for a preference towards one particular service provider; this generates commitment to buy from them again, social and emotional bonds, and leads to less sensitivity towards the price of the service, Ogba and Rowley [27,28]. Ethnocentric consumers tend to develop a marked loyalty towards national brands, Zeithaml et al. [29].
For Burgess et al. [30] the knowledge of marketing phenomena drifts almost exclusively from researches made in industrialized countries with a high income, for this reason, there is a need to generate new marketing inputs that come from emerging economies to contribute to the academic literature. In this way, it could be said that emerging markets are radically different than they are in traditional, capitalist, and industrialized societies. That is why the boom of these markets offers new investigation opportunities in marketing, Sheth [31].
For Wang et al. [20] more investigations are needed to know how the image of a global brand develops in an emerging market and also to know if these brands are able to adapt to that market through the addition of local cultural values. In the same way, to know if concepts such as national identity, patriotism, and ethnocentrism affect, in a very important manner, the development of these brands in those markets.
For this reason, it is needed to enlarge the research of global brands in these developing countries, even though the global brands show a solid image that provides them a competitive advantage compared to local brands, these brands should also face the resistance of consumers with a high level of ethnocentrism and patriotism, Wang and Chen; Wang and He, and He et al. [19,20,32], besides global brands in the context of developing countries, should also face different challenges in terms of service delivery, of communication, specific cultural meanings, and local tension regarding the rivalry between global and local brands; and where some keys for the development of the global brand in a developing country should have the skills to work out a direct dialogue with the local culture, meaning its ability for local adaptation, without losing its globality.
On the other hand, also it is expected from these brands to be humble, instead of being haughty despite their global importance, Suarez et al. [33]. This way, it is needed to go deeper into the concept of global brands in developing countries, with the ethnocentrism of the consumer in hedonic services, the reason why this study presents an academic input to the short amount of existing literature that embraces these topics as a whole.
With these facts in mind, this work provides a deeper understanding of how the level of ethnocentrism in the selection of services influences loyalty in the realm of hedonic services and considers antecedents such as patriotism, collectivism, and individualism in consumers in a developing country (which is also an exporter and regular consumer of one of the best coffees in the world); this approach involves looking at a foreign company from a developed country offering a service.
That is why the research presented has the following structure: after presenting the introduction, a literature check is conducted with SET as the main study variable over the variables patriotism, individualism, and collectivism; while loyalty is used as a consequence of the SET. On the other hand, image is used as an antecedent of loyalty. Therefore, it could be proposed that these relationships construct and present a theoretical model with its respective hypothesis. Next, the methodology used will be detailed to prove the previous hypothesis, which allows us to analyze the investigative work results. Finally, conclusions, implications, limitations, and future research lines are presented.

Consumer Service Ethnocentrism SET
Working from a sociological perspective, Sumner; Camacho et al. [34,35] defined the concept of ethnocentrism as the perception that individuals hold of the group they belong to as well as their evaluation of this group compared to other groups. It is for this reason that the perception held of one's own group becomes a point of reference.
According to Lee et al. [36] ethnocentrism, which is a lack of acceptance of cultural diversity and intolerance for outgroups, has a tendency to lead to negative prejudices towards other cultural groups, for this reason, ethnocentrism is involved in xenophobia.
This ethnocentrism produces a phenomenon known as consumer ethnocentrism in the North American consumer that is characterized by the level of appreciation demonstrated towards products from their own country of origin. This appreciation of the place of origin is grounded in the fear of possible losses to economic interests in their own country resulting from high imports. This leads to the intention of not buying any foreign products, Shimp et al. [2].
In relation to this topic, Luque-Martínez et al. [37] consider that ethnocentric consumers have a high tendency towards expressing biased and over-valued judgments of local products. This tendency allows them to emphasize the positive aspects of these products and undervalue or minimize the virtues of imported products. It follows that ethnocentric consumers will assume the values of their ethnic group or country as symbols of pride and as representative of national unity. With these concepts in mind, Sharma [38] concludes that consumer ethnocentrism represents a favorable tendency towards anything related to the group to which one belongs (the in-group) and a bias against anything related to groups to which one does not belong (the out-group). This results in an attitude with three distinct dimensions: affective reaction, cognitive reaction, and behavior. This behavior is reflected in the beliefs that consumers hold about purchasing services offered by foreign companies.
For Vivek et al. [7], the ethnocentrism of consumers seeking service products is a form of bias on the part of the consumer that generates a preference for service providers that are of the same ethnic group as themselves, thus avoiding the possible risk to their own group that the selection of foreign service providers represents.
Likewise, Sharma [39] considers that high levels of SET have a negative effect on how the quality of services is perceived and on the level of satisfaction felt towards services provided by foreign operators.
In summary, as can be observed by a review of the literature listed in Table 1, the limited number of studies of service ethnocentrism (SET) does not enable conclusions to be made about the nature of the concept nor about the variables that explain it. The conclusion is that ethnocentrism in the selection of services makes reference to the consumer bias that manifests as a preference to contract services provided by national companies above services provided by foreign companies Vivek et al. [7].

Author(s) and Year
Definition (Shimp & Sharma, 1987, p. 280) "Represents the beliefs held by American consumers about the appropriateness, indeed morality, of purchasing foreign-made products." (Sharma, Shimp & Shin, 1994, p. 27) "Specifically, consumer ethnocentricity has the following characteristics: first, it results from the love and concern for one's own country and the fear of losing control of one's economic interests as the result of the harmful effects that imports may bring to oneself and countrymen. Second, it contains the intention or willingness not to purchase foreign products. For highly ethnocentric consumers, buying foreign products is not only an economic issue but also a moral problem."

Patriotism
For Sharma et al. [3] patriotism is the dedication of a person, via feelings of love and devotion, to their country of origin. For Mihalyi [40], it also functions as a defense mechanism when dealing with third parties. Balabanis et al. [41] define patriotism as a strong feeling of attachment and loyalty towards one's own country but which does not generate hostility towards other countries. Durvasula et al. [42] consider that patriotism is intricately linked with the pride one feels when regarding one's own country and that it is reflected in favoritism towards the people, modes of action, and values of one's own country. Patriotism, then, is entwined with the perceptions that are held by individuals regarding their country of origin and all that is associated with it. For Blank [43], there is a distinction between patriotism and nationalism. The first tends to attribute positive values such as the level of commitment people feel towards their country of origin, while nationalism has negative implications as it heightens the glorification of the home nation above all others. As such, it should be treated differently.
Han [44] demonstrated that consumer patriotism has significant effects on the intention to buy national products compared to the purchase of foreign products. Following this, when high levels of patriotism are encountered, the people, cultures, and products from origins distinct to those of the consumer will probably be rejected. Highly patriotic consumers support national producers, considering it a duty and an expression of loyalty towards their home country. For Puncheva-Michelotti et al. [45], patriotism ends up awarding a competitive advantage to local products and strengthening sensations of national identity felt by individuals via the purchase of these products. Other studies confirm the positive relationship that exists between patriotism and the ethnocentrism of consumers, Sharma et al. [3], even in the services sector, De Ruyter et al. [4]. With these conclusions in mind, we set up the first hypothesis.

Individualism and Collectivism
For Triandis [46], values related to individualism and collectivism exist in every person. For the majority of people, tendencies towards collectivism are the first to develop within a primary group such as the family. Later, other more individualistic tendencies are developed.
For Triandis et al. [47], people who develop more individualistic values present firm characteristics of self-sufficiency and separation in regard to their own group whilst people with collectivist tendencies present firm characteristics of family integration and interdependence in regard to their own group. In view of these facts, in collectivist cultures, both the behavior of their members as individuals and the behavior of the group is guided by the group; on the other hand, in individualistic cultures, the behavior of individuals within the group is driven by the individuals themselves. It is for this reason that when conflicts arise in individualistic cultures between the objectives of the group and the objectives of the individual, the differences are resolved in favor of the individual whilst in collectivist cultures these differences are resolved in favor of the group, Kapoor et al. [48]. In collectivist cultures, it is possible to identify how the members of the group tend to subordinate their own personal goals to those of their particular group; in contrast, it is possible to observe how the members of individualistic cultures use the group only as a means to achieve their own personal goals, Hui et al. [49].
Hofstede [50] defines culture as a collective phenomenon in which people live in the same social setting and where culture is learned and not inherited; the author adopts five dimensions as descriptors: power distance, individualism versus collectivism, masculinity versus femininity, uncertainty avoidance, and long-term orientation. Referring to the individualism versus collectivism dimension, Colombia presents one of the lowest incidences of individualism in Latin America. As such it can be concluded that it is a collectivist society in which strong relationships are maintained between group members and where reliance on the group is considered important for the achievement of its objectives, Robles et al. [51]. Nonetheless, Colombian society is highly focused on the achievement of success via competition. Although it is a collective society, the implication of this focus is that rivalry with members of other groups or even with other social classes is generated and also that it confers status, Hofstede [52]. With the previous studies by Sharma et al., Ruyter et al., and Van Birgelen et al.; [3,4,53] in mind, we consider it probable that people from collectivist societies demonstrate strong ethnocentric tendencies whilst those from individualist societies do not. From this the following hypotheses were put forward: Hypothesis 2 (H2). Collectivism positively affects ethnocentric tendencies in the hedonic services.

Loyalty
Customer loyalty can be defined as a perception of the superiority of the product that a consumer buys, generating a social bond and an emotional attachment, Ogba et al. [27]. It is for this reason that loyalty is demonstrated both by repeat purchases and in the attitude of consumers towards the product, thus determining the level of commitment to the brand, Beerli [54]. For Rowley [28], this is the means by which loyalty generates less sensitivity to fluctuating prices and represents less cost to obtain new clients and as a result, greater profitability. For Bowen et al. [55] however, the benefits of loyalty are demonstrated by a more favorable attitude towards the company, commitment to buying the product, and the making of recommendations to other people. East [56] considers that in the services sector, loyalty is a deeply-rooted commitment on the part of the consumer, generating repeated future purchases of a service in a consistent manner, despite various influences designed to change the behavior.
For Hwang et al. [26], client loyalty generates positive direct marketing via word of mouth, where the consumer easily identifies their preferences towards a given service provider. For their part, Schlesinger et al. [57] identify the benefits of loyalty as a tendency that lasts beyond the time when the consumer has ceased to use the service. It is for this reason that loyalty is referred to as both the period in which the consumer is actively using the service and the period after the use of the service, thus contributing to improving and promoting the image and reputation of the service they received.
As noted by Zeithaml et al. [29] ethnocentric consumers generally reward their loyalty to national brands, thus reflecting their intention to remain within their own group, negatively impacting any loyalty towards a foreign product Abosag et al. [58]. Makanyeza [59] affirms that loyalty generates repeat purchases, the result of the consumer's commitment to the preferred brand. The greater the ethnocentrism of the consumer, the greater the negative impact on loyalty to a foreign brand. With this in mind, the following hypothesis was formulated: Hypothesis 4 (H4). Ethnocentrism in the hedonic services negatively affects loyalty to a foreign global brand.

Image as an Antecedent to Loyalty
Brand image is the associations (attributes, benefits, and attitudes) that a consumer holds in their memory about a particular brand, Keller [60] whilst for Schlesinger et al. [57], image is the sum of beliefs, ideas, and impressions that a person has of an object. For Palacios-Florencio et al. [61] (p. 1275), "the corporate image is based on the impressions of the firm held by both internal and external stakeholders. This image is determined by the organization's actions, such as how it treats its employees, customers, and society".
With this in mind, and owing to high levels of competition, companies are obliged to make great efforts to manage their brand image, Abosag et al. [58]. Also, consumers generally tend to link their perceptions of a product with the image of the country in which it is manufactured and for this reason, the country of origin of a brand will influence the image of the product, Nebenzahl et al. [62]. Significant evidence exists that the country of origin is a potent signal that shapes the perceptions and evaluations of the brand on the part of the consumer, Heslop et al. [63]. In marketing literature, sufficient evidence exists of the significant impact that global brands have on the consumer's perceptions and judgments about a particular product and on their loyalty, Leclerc et al. [64].
According to Pérez et al. [17] (p. 4), the catering services "are value-expressive, which means that consumers will generally make their decisions to use the service based on the image that is associated to the company and the association that will be created between the company's image and his/her image".
For Steenkamp et al. [65], the positive image that a consumer has of a foreign country is also transferred to the products manufactured in that country and, as such, it is possible-in some countries-to detect the perception that global brands possess greater quality and prestige. Ostrowski et al. [66] found that a significant relationship exists between the image of a company that provides services in a hedonic market and the loyalty of the clients. Brunner et al. [67] for their part, consider that a positive link also exists between perceived image and consumer loyalty towards train service companies: in the foodservice sector, Abosag [58] found that image increases consumer loyalty. With these studies in mind, the following hypothesis was formulated: Hypothesis 5 (H5). The corporate image of hedonic services positively affects consumer loyalty.

Data Collection and Sampling
A quantitative study based on a structured questionnaire was used to verify the hypotheses proposed by the model. The data collection process took place in the second half of 2019 in the city of Bogotá via a personal survey targeted at Colombian customers of Starbucks. Convenience sampling was used, and 305 valid questionnaires were obtained. Table 2 shows the principal characteristics of the study. With regards to the demographic profile of the survey participants, Table  3 shows the distribution of the sample according to the classification variables: 51.8% were men and 48.2% were women. The age range of survey participants was 56.4% between 18 and 25 years of age, 24.6% between 26 and 36 years of age, 12.8% between 36 and 45 years of age, 4.6% between 46 and 55 years of age with only 1.6% of survey participants older than 55 years of age. In relation to education, 41% were secondary school educated, 44.9% had university studies, and 12.1% postgraduate studies. In relation to their employment situation, 38.4% were students and 43.3% were currently working. Regarding earnings, 35.1% earned between two and three times the minimum legal wage.

Data Collection and Sampling
A quantitative study based on a structured questionnaire was used to verify the hypotheses proposed by the model. The data collection process took place in the second half of 2019 in the city of Bogotá via a personal survey targeted at Colombian customers of Starbucks. Convenience sampling was used, and 305 valid questionnaires were obtained. Table 2 shows the principal characteristics of the study. With regards to the demographic profile of the survey participants, Table 3 shows the distribution of the sample according to the classification variables: 51.8% were men and 48.2% were women. The age range of survey participants was 56.4% between 18 and 25 years of age, 24.6% between 26 and 36 years of age, 12.8% between 36 and 45 years of age, 4.6% between 46 and 55 years of age with only 1.6% of survey participants older than 55 years of age. In relation to education, 41% were secondary school educated, 44.9% had university studies, and 12.1% postgraduate studies. In relation to their employment situation, 38.4% were students and 43.3% were currently working. Regarding earnings, 35.1% earned between two and three times the minimum legal wage.

Measurement of the Variables
The various items included within each studied variable were adapted to the scope of the study, based on the different measurement scales used in previous studies. For this study, seven-point Likert-type scales were used (with 1 representing "totally disagree" and 7 representing "totally agree"). To measure the SET, the Sharma [38] scale, made up of 24 items and three dimensions was used: affective reaction, cognitive reaction, and behavior. In terms of individualism, the eight items proposed by Triandis et al. [68] were used, grouped into two dimensions: horizontal individualism and vertical individualism. To measure collectivism, eight items were used, also grouped into two dimensions: vertical collectivism and horizontal collectivism, based on the Triandis et al. [68] scale. The nine items used to measure patriotism were adapted from the proposed scale of Levinson [69]. The items used to measure loyalty were obtained from Kim et al. [70]. Finally, seven items that corresponded with identity were adapted from the scale proposed by Palacios-Florencio et al. [61].

Analysis and Discussion of the Results
A descriptive summary of the results obtained in the study (see Table 4) shows that the sampled participants possess an elevated ethnocentric character. All the measurements from this item are above 3.5 on a scale of 1 to 7 and 18 of the 24 items measured are above 4. Similarly, levels of patriotism are high given that five of the seven items are above 3.5. On the other hand, the sample also reflects an elevated collectivist character as seven of the eight items rated above 5. Nonetheless, a medium to high level of individualism is also observed given that four of the eight items are above 5. This result reflects the observations made by Robles et al. [51] that, despite the collectivist character of Colombian society, it is also highly oriented towards success achieved in a competitive market. The resulting rivalry generated with other groups and even with other social classes increases status Hofstede [52]. In contrast, loyalty towards Starbucks is relatively moderate whilst their image is highly rated. This can be understood as a result of the hedonic nature of their service. The collected data were analyzed in two stages. The first stage consisted of validating the measuring instrument and in the second stage the structural model was assessed, Barclay et al. [71]. Partial least squares regression was applied using Smart PLS version 3.0.
This technique was chosen due to the fact that the use of PLS continues gaining traction as it is employed in diverse marketing studies and at an international level. It allows the inclusion of both formative and reflective constructs, Diamantopoulos [72]. In our case, the image construct was considered formative based on a study carried out previously by Palacios-Florencio et al. [61] who considered that the construction of image includes formative indicators. In the analysis of these constructs, the weights factor must be analyzed instead of the factor loadings, Chin [73]. The authors acknowledged the possible multicollinearity of formative indicators, Roldán [74] On the other hand, modeling via PLS sought to predict the dependent variables, allowing the maximum expression of the R 2 ratio of the explained variance of the dependent variables. This allowed estimations of the parameters to be based on the capacity to minimize residual endogenous variance. In this case, PLS permits a better adaption to predictive studies when compared with other tools, Barroso [75]. The decision to use PLS is also justified due to the predictive nature of this study, plus the fact that in this study, the image construct is considered a construct with formative indicators.
The evaluation of the measurement model required analysis of the internal consistency of the scale, individual item reliability, and convergent and discriminant validity, Barclay et al.; . To analyze the reliability of reflective constructs, the Cronbach α coefficient of composite reliability and the extracted average variance (AVE) were used. All values obtained via the Cronbach α values such as composite reliability showed scores above 0.7 although the AVE of patriotism is the only variable with a value below 0.5. In contrast, two of the items of the patriotism variable (PT8 and PT9) have loadings of less than 0.6 while the SET variable of the item (SET12) has a loading of less than 0.5. With these results in mind, and after carrying out the respective eliminations, when PT8 and PT9 are eliminated it is observed that the AVE of the patriotism variable is above 0.5. The item SET12 was also eliminated due to its loading below 0.5.
From the results presented in Table 4, we can affirm the reliability and convergent validity of the scales used for measuring the different variables included in our model. Since the image variable is defined as a formative construct, its evaluation is made at the indicator level and via the evaluation of possible multicollinearity using the Variance Inflation Factor (VIF) and determination of the significance of weights. With respect to the FIV, all the items have a loading of less than 0.3 while the weight of four items (IMAG1, IMAG2, IMAG4, and IMAG6) is not significant. These indicators are eliminated from the results shown in Table 5 due to their non-compliance with the previously outlined criteria. Finally, to confirm the discriminant validity, an evaluation was made using the criteria put forward by Fornell [76] and the Heterotrait-Monotrait ratio (HTMT), Henseler et al. [77]. In relation to the criteria of Fornell [76], the square root of the AVE of each variable should be greater than the correlations that this has with the other variables of the model. As can be observed in Table 6, all the square roots of the AVE of each construct are greater than the correlation with any other construct in the model, Fornell [76]. With respect to the (HTMT) ratio, authors such as Henseler et al. [77] consider 0.90 as an appropriate maximum cut-off value. As can be observed in Table 7, all the values of the (HTMT) ratio are below 0.9.  Once the psychometric properties of the measurement instrument were assessed, the analysis of the structural model via PLS proceeded based on the same criteria as was employed to calculate the significance of the parameters (a process of bootstrapping, of 5000 sub-samples of the size of the original sample was carried out). The R 2 obtained via bootstrapping was the first stage in the evaluation of the predictive capacity of the structural model. This indicated the degree of variance in a construct explained by the predictor variables of the construct in the model.
In relation to the predictive power of the model, the criteria proposed by Falk et al. [78] were followed. They considered that the minimum R 2 value is 0.1; this value indicates that the model presents a minimum level of significance. Table 8 shows that the R 2 value of all the dependent factors is above 0.1 (as per the critical level). Next, the predictability of the model was evaluated by applying the Stone-Geisser (Q 2 ) test to each dependent construct via the blindfolding procedure. As can be observed in Table 8, none of the Q 2 values had a score of zero or below zero which indicates the predictive relevance of the endogenous variables. After using the R 2 calculation to check the predictive power of the model-a measurement that determines the amount of variance of the endogenous variables, explained by the constructs they predict-levels of between 0 and 1 should result. As can be observed in Table 8, all the values are above the established limit and as such, the previously formulated hypotheses possess an adequate predictive level that permits the assessment of the significance of the previously stated relationships.  The positive link that exists between patriotism and SET is evidenced by the results obtained. As such, the Hypothesis H1 is confirmed. Due to this and according to the significance of the relationship previously stated, it can be confirmed that this construct wields a positive and significant effect on SET (β = 0.312; p < 0.000). In the same way, collectivism also has a positive and significant influence on SET. This leads to the confirmation of Hypothesis H3 (β = 0.254; p < 0.000). Apart from this, a negative relationship is observed between individualism and SET, allowing us to confirm Hypothesis H2. Considering the significance of this result, we can conclude that individualism has a negative and statistically significant effect on SET (β = −0.100; p < 0.036). According to the results obtained, a positive and statistically significant relationship exists between SET and loyalty towards foreign brands (β = 0.148; p < 0.001), thus rejecting Hypothesis 4, given that this was proposed as a negative relationship between the two constructs. One possible explanation for this result is the hedonic nature of the service that Starbucks provides, as previously mentioned. The sample, however, is made up of a largely youthful public, 81% of those interviewed were aged between 18 and 35 and the majority had tertiary studies. For this segment of the population, foreign services generate a status that allows them to rise above the rest, socially speaking. This may explain their loyalty, in a positive sense, towards the brand. Finally, based on the information obtained, a positive and significant link exists between brand image and consumer loyalty. With this finding, Hypothesis H5 is also confirmed (β = 0.612; p < 0.000).

Discussion and Theoretical Implications
This study intends to contribute to the research into the concept of service ethnocentrism (SET) related to hedonic services. Similar studies have, to date, focused on developed countries.
Sharing the vision of Sheth et al. [79], the boom of emerging markets is not just inevitable but it is also having an important impact in the practice and theory of marketing; where it should change from a colonial mentality to a global mentality, that is why the investigation of emerging markets is increasingly becoming a need. It implies to invite future academics, to perform more research in these markets every day. Under this perspective, this job has contributed to the scarce literature that exists about ethnocentrism in hedonic services and also about the global brand management that is developed in the context of an emerging market like Colombia, generating contributions that allow global and foreign brand managers to perform strategies for excellent development in these emerging markets.
The results obtained serve to confirm, firstly, a positive and significant relationship between patriotic sentiment and collectivism with SET, as was demonstrated in previous studies, Sharma et al.; De Ruyter et al. [3,4] the understanding that these constructs elevate levels of SET is supported by this study. Secondly, this study confirms the existence of a significant and negative relationship between individualism and SET, having observed how a greater level of individualism lowers the level of SET, Sharma et al. and de Ruyter et al. [3,4]. Thirdly, this study examined the relationship between SET and loyalty, with the result that Hypothesis H4 was not supported. These results suggest that the relationship between the two variables is positive and significant whilst Hypothesis H4 proposed a negative and significant relationship, in agreement with the work by Abosag et al. and Makanyeza [58,59]. On the other hand, this study serves to confirm the positive and significant relationship between image and brand loyalty. In this case, as occurs with authors such and Abosag et al. and Brunner et al. [58][59][60][61][62][63][64][65][66][67], our results confirm the great influence that image exerts on Colombian consumer loyalty towards the Starbucks brand.

Managerial Implications
Various implications for the management of international service companies can be derived from this study. In this study it is concluded that SET does not influence brand loyalty in the hedonic services and, as such, enhancing all the experiential and hedonic aspects of the consumption of the service is a relevant means of decreasing the effects of the ethnocentrism of a country. In any case, we recommend a prior analysis of the ethnocentric character of the citizens of a country as a key variable when formulating strategies of internationalization in multinational companies.
It is worth highlighting that global brands end up doing a counterbalance to ethnocentrism, in this manner, we consider the global character from the hedonic service brands allows to generate an important position that generates acceptance and desirability from an ethnocentric consumer.
About this same topic, He et al. [32] determined that global brands may have a higher success abroad, when they add local elements from this local country, meaning that compatibility could exist between a global brand and the cultural items from that country, this improves the perceived image as the purchase intention towards the global brand. Considering this, for this study we consider that Starbucks decided to get into the Colombian market in 2014, with this type of strategy from a global brand with cultural items from Colombia. As stated before, Colombia is one of the three highest coffee producers in the world and thus their population is very proud of this product. For this reason, Starbucks stores in Colombia utilize decoration, pictures of Colombian coffee landscapes, the Colombian coffee growers, the harvesting process, Colombian coffee farms, and some other representative elements, besides mentioning that they sell 100% Colombian coffee (even though it is possible to buy grains of coffee from different countries).
On the other hand, for Diallo et al. [80], on his study about the brands in Colombia, it can be concluded that the Colombian consumer values the image of a brand more than his previous consumption experience. So, it can be evidenced that global brands as Starbucks have such a solid image that gives them a competitive advantage that allows them to overcome the resistance from consumers who have a high level of ethnocentrism and patriotism, Wang and Chen; Wang and He, and He et al. [19,20,32]. Then, we can conclude that Starbucks with its global brand has had the capability to generate a direct dialogue with this emerging market, through the adaptation of the local Colombian culture without losing its globality.
In the past years, the amount of open innovation dynamics has shown great growth, for this reason, many worldwide companies such as Starbucks have started to create open innovation communities so that the users take part in the innovation activities and development of their own products. This way, Starbucks is able to reduce the doubt in the decision-making process, by generating continuous interaction with local customers that allows them to offer products according to the needs of those consumers, Wu et al. [81]. In the case of Colombia, this is another reason that has allowed the images of global brands to adapt to the local market, overcoming the ethnocentrism of the Colombian consumer.
On the other hand, supported on the previous studies from Martins et al. [82] about business model innovation from companies like Starbucks, let us consider for this study the fact that this brand has successfully gotten into the Colombian market in a positive way, through four punctual processes: first, getting to convince the Colombian consumer of the concept of "coffee bars", where some elements from a cafeteria and some services and environmental elements from a bar are mixed, allowing them to develop innovation as much in the service delivery as in their stores, through a combination of both concepts. Second, comparing their business models in different markets and finding substantial differences between them. Third, considering the previous information, modifying their business schemes according to each market, this let them highlight, even more, some predominant values through what they call "slot filling" according to each kind of market. In the case of Colombia, this can be evidenced in the decoration of their shops, in their background music, and in the hedonic products and services offered to the local consumer as stated before. Fourth, their unit business managers are capable of identifying those differences and are able to offer solutions to fill those gaps or "slot filling".
This combination of these four factors allowed them to have a better comprehension of the customer needs, to be able to compete in an emerging market like Colombia. Considering this, we consider that open business models play a very important role to be able to develop global and foreign brands in the emerging markets, overcoming the local consumer's ethnocentrism.

Limitations and Future Research
To conclude, we consider worth mentioning that some limitations of this study suggest future research paths. The first arises from the convenient character of the sample. On the other hand, the limited geographic area in which the study was applied suggests that a comparison with results from other Latin American countries may be interesting. For example, studies about previous positive experiences from brands in developing countries made by Diallo et al. [80], where they compare Colombia and Brazil, demonstrated that homogeneous conclusions cannot be made within these Latin-American countries.
Other possible future research paths could contemplate the analysis of new relationships such as the mediating effect of image on SET and loyalty as well as the study of new variables that would allow us to understand and explain why an ethnocentric consumer would feel loyalty towards a foreign brand. One new variable could be xenocentrism, defined as the people who like a society different from their society and who evaluate everything using that different society as a reference, and it entails the tendency to reject their own products, which deteriorates local cultural values, Camacho et al.; Lee et al. [35,36].
In Latin America, this xenocentrism is demonstrated by the over-estimation of foreign cultures (generally Western cultures) due to local economic inequality; it is especially notable when these markets are dependent on an economy that also exerts forces of cultural imperialism, Gaur [83].
For Balabanis et al. [84], the global and foreign brands may use signals of local identity to adapt better in so-called emerging markets and they may attract the ethnocentric consumers, and at the same time highlighting their global or foreign origins, especially if they sell hedonic products or services. They will be also able to attract xenocentric consumers as well, who have a clear preference for foreign brands. Another limitation of this work was that its methodology was developed applying quantitative techniques that were processed through PLS, it is suggested for new investigations to develop diverse qualitative techniques to improve even more the measurement of included variables and their adaptation to different industries.
Conspicuous consumption is another variable that we propose for further research. Piron [85] made reference to the desire of consumers to demonstrate their capacity to pay for luxury goods that, with the intention of impressing others via their consumption, also seek to project the buyer in society, meaning that their utility is more social than functional, Mason [86]. These variables may exert a moderating function on the relationships analyzed by this study. Finally, the service selected-Starbucks-possesses a clearly hedonic character that can be strongly associated with the theoretical framework of the experience economy, Pine [87]. Our interest is to continue working in this field, leaving open the possibility that future research paths may consider other providers in the utilitarian service sector.
For McIvor et al. [88], COVID-19 has allowed us to evidence how in some political situations, historical and negative legacies from extreme nationalism have been enhanced and for that reason, a new perspective of global survival and resistance has been cast based on this pandemic. On the other hand, about this new reality, McQueen et al. [89], considers that the reactionary ideas from the philosopher Nietzsche about the topic of the herd and his cultural nationalism relapse over the empathy attitudes towards the society's weakest members.
This appears as an answer to the negative attitudes from some leaders and their managements regarding COVID-19, where some of these leaders have shown a dislike to the wellness of more vulnerable social groups and their social responsibility towards them. The crisis generated from COVID-19 has allowed it to have two political mainstreams. On one side some leaders are feeding populism and nationalism through patriotism, affecting in a negative way the international solidarity. On the other hand, some other leaders have been using international cooperation between countries, so all interested parties assume a shared responsibility and where it is forecasted that developing countries may take longer to recover Leal et al. [90]. This way, as a future line of investigation, the patriotism variable should be researched to see how it is affected by COVID-19, the political mindsets taken by every country, and how it affects SET.
Finally, Supphellen et al. [91] determined that the consumer's ethnocentrism did not have any effect on foreign brands, while these brands were considered superior; however, it could happen that highly ethnocentric consumers would not evaluate the products or services in the same way they do with brands, so researches based on particular hedonic services can be performed, without considering its brand.