International analysis of the countries where Brazilian franchise chains operate

R E S U M E N R E F E R E N C E S


INTRODUCTION
In Brazil, the franchising segment has gained prominence in the last decade due to its increased representation in the Brazilian economy. There are approximately 2,426 franchise chains operating in Brazil. Comprising approximately 104.543 franchised units, it has reported a 12% growth in comparison with the previous year. Sector revenues stand at US$ 43 billion per year, with a growth rate of 14% compared with 2011 (ABF, 2013).
This growth is beneficial institutionally for the entity representing the franchising sector, Associação Brasileira de Franchising (ABF -Brazilian Franchising Association), which has been increasingly representing a sector of major relevance for the Brazilian economy. Thus, topics such as internationalization of franchises are now part of the agenda of activities at ABF, which is promoting the internationalization process of Brazilian franchises. For that, partnerships have been established at the federal level with Agência Brasileira de Promoção de Exportação e Investimentos (APEX -Export Promotion Agency) aiming to increase the disclosure of Brazilian franchises in the international market and, consequently, the development of these operations abroad (Macedo, 2011;Simões, 2011).
Even though only 4.7% of Brazilian franchise chains have international operations, the process of internationalization of Brazilian franchise chains reported a growth of 300% in the last decade (Macedo, 2011;Simões, 2011). Currently, approximately 110 Brazilian franchise chains operate in all continents (Basílio, 2013). Nevertheless, little is known about the commitment of international Brazilian franchise chains as well as their operations in foreign countries. In order to fill this gap, this paper aims to demonstrate which external environment factors are involved in the international commitment of Brazilian franchise chains. More specifically, our objectives herein are (i) to understand which external country characteristics lead to international franchising operations and (ii) to ascertain the influence of such characteristics in the commitment of franchise chains in each country they operate.
By analyzing the external factors of countries in which Brazilian franchise chains operate, we expect to find different sets of countries with similar institutional characteristics. Therefore, once the group formed by countries sharing similar characteristics is set, our next step is to verify the existence of significant differences among the international commitment of the franchise chains belonging to each one of these groups. Thus, it is intended herein to comprehend the differences in the international commitment of franchise chains, taking into consideration the external environment characteristics of the countries in which the franchises operate.
Our research contributes to the field of international business and more specifically to the understanding of internationalization of franchises in three ways. First, by adding to the literature since only a small number of papers on the internationalization of franchises can be found in the literature nowadays (Melo & Andreassi, 2010).Then, by filling a theoretical gap within the internationalization of Brazilian franchises where we present the formulation and analyses of environmental characteristics of the clusters formed by the countries in which Brazilian franchises operate. Up to now, the most relevant papers published in academic journals and conference abstracts on Brazilian franchise chains can be classified into two different categories: (i) exploratory and quantitative researches on internationalization drivers (Marques, Merlo, & Nagano, 2009;Aguiar, Consoni, & Bernardes, 2013), as well as on the comparative organizational characteristics among internationalized and domestic chains (Melo, Parente, Oliveira Jr., & Borini, 2012), and (ii) case studies which analyze the management of franchises abroad (Ribeiro & Melo, 2007;Borini, Souza, Biskamp, Coelho, & Sadzinski, 2013).
Finally, regarding research involving the internationalization of Brazilian franchise chains, there are only five exploratory researches conducted but they all have a sampling limitation, covering 14 chains (Rocha, Borini, Spers, Khauaja, & Camargo, 2012); 15 chains (Fundação Dom Cabral, 2011;Marques, 2006); 16 chains (Fundação Dom Cabral, 2013) and 20 chains (Borini, Rocha & Spears, 2012). The research that involved the largest number of internationalized Brazilian franchises corresponded to 33 chains (Barbosa & Mariotto, 2012). It is therefore intended to present herein a study with a larger sample size, involving 54 Brazilian franchise chains and 157 units operating abroad in order to comprehend the international commitment of a large number of Brazilian chains operating abroad.
In sum, our paper makes significant contributions to franchise studies by creating a group of relevant environmental variables that enable the understanding of the international commitment of franchise chains in countries sharing similar institutional characteristics. We pointed out, however, that this study relates only to the countries in which Brazilian franchises have some international experience. The contributions made herein are based on the variables "market opportunity" and "business efficiency -trust and ease in doing business", both of which directly support the researches accomplished by Aliouche and Schlentrich, 2009e 2011, and Alon and Shoham, 2010 Moreover, the managerial contributions of this paper lie in helping the selection of destination countries according to the characteristics of the external environment of these markets, through a framework that considers the demographic, economic, political and government, corporate and legal factors. The grouping of countries into clusters can contribute to franchisors in the management of markets with similar institutional characteristics, assisting in the definition of policies of regional business.
In addition, this analysis may help Brazilian franchise chains not yet internationalized to identify international opportunities based on the experience of Brazilian franchise chains that already have international experience. It also contributes to the internationalized Brazilian franchise chains, providing the analysis by international markets yet unexplored.
The managerial contribution of this paper is implied in the assumption that once the two proposed vectors are found, it becomes possible for managers to understand how the international commitment in a country can be more or less related to market opportunities and to trust and ease in doing business. This understanding implies in knowing beforehand the degree of intensity of the international expansion in each country and also in determining if the barriers or advantages from this expansion derive from the market opportunity or/and from the business efficiency.
The rest of the paper is organized as follows: In the following section the theoretical framework will be presented, where studies on franchise internationalization will be analyzed, as well as studies which analyze the external environment of the chains' international strategy. Then the methodology section presents the indicators related to external environment factors and to the international commitment of franchises, whose information were obtained from secondary data. Next, the results present the outcomes from the exploratory factor analyses containing external environment indicators; the clusters result from the obtained factors conjunction and the relationship between the group of countries in which the franchises operate and their international commitment. Finally, by discussing the results and implications found in this research, some explanations and suppositions will be presented along with the final conclusion and considerations.

LITERATURE REVIEW AND THEORETICAL FRAMEWORK
The scientific publication in international journals regarding international commitment of franchise chains dates from the 1990s. This literature has focused on understanding how the processes of internationalization of franchise chains happen and the reasons why these chains start operating abroad (Altinay & Miles, 2006;Aliouche & Schlentrich, 2011).
In the 2000s, however, many scholars studying franchising started to focus on the devastating consequences that international expansion plans caused to the enterprises, which did not take into account the international environment of the countries of destination (Han & Diekman, 2001;Aliouche & Schlentrich, 2009). Before this, many academic papers that started to focus on countries where franchise chains installed operational units, gained more academic relevance (Aliouche & Schlentrich, 2009, 2011Alon & Shoham, 2010).
The first empirical study about environmental factors in internationalizing franchises involved the franchisee's investment risk analysis when choosing a country to operate in. Aliouche and Schlentrich's (2009) proposal is focused on the analysis of European countries aiming to identify and assess risks in these countries. Aliouche and Schlentrich (2009) aimed to develop a model to assist franchisors by selecting countries to invest in. The analysis and data categories include political and economic risk (Country Risk Index), legal and regulatory risk (Ease of Doing Business), market potential (population and gross domestic product [GDP] per capita) and cultural distance (Hofstede). The main purpose of the study was to develop an international index of countries for investment regarding American franchise chains. The authors focused on analyzing the profile of the countries of the European Union (EU), given the political risk, economic strength and culture proximity among EU countries. The main contribution of these authors was, therefore, to create an instrument for choosing international markets by franchise chains, which aimed to expand to all 25 European countries. Despite the discussion on the relevance of internal resources of franchise chains and the entry procedures as variables, which could complement the choice made by countries to expand internationally, these approaches were not empirically tested in this model. Aliouche and Schlentrich (2011) continued to improve the model elaborated in 2009 and created a global index involving the analysis of 143 countries for investment. The prerogative of the authors is that the improved model is the most comprehensive instrument towards the choice of countries for investment involving franchise chains. The environmental variables and the data sources involve political and economic risks (Country Risk Index); legal and regulatory risks (Ease of Doing Business); cultural and geographic distance (Hofstede); and market opportunity (population and GDP per capita). The authors emphasize the relevance of this study by pointing out that even large franchise chains have human resources, financial and logistic limitations and, therefore, they must properly select countries for investment. This selection of markets precedes the entry procedure decision that franchise chains must take

INTERNATIONAL ANALYSIS OF THE COUNTRIES WHERE BRAZILIAN FRANCHISE CHAINS OPERATE
into consideration before analyzing the characteristics of the external environment (Aliouche & Schlentrich, 2011).
Finally, Alon and Shoham (2010) argue that the creation of country clusters is useful because it brings up facilitators to manage franchise chains. The franchisors can therefore select countries that have similar environmental characteristics to the countries of their own, which facilitate the choice of countries for expansion. The findings of this study indicated eight clusters, which involved 56 countries, whereby the economic, demographic, cultural and political environment were analyzed. These analyses were based on the following indicators: Gini, service percentage of the economy, population, Ease of Doing Business Index, percentage uppermiddle class, urbanization, economic freedom, political risk and culture distance.
Based on the indicators of the external environment that were used in the studies above mentioned, we propose the following framework to be empirically tested (see Figure 1).
The indicators, presented in our theoretical framework, were divided into two groups a priori: one group is related to the market opportunity, which involves indicators regarding population size and purchasing power of consumers i.e., demographic and economic factors; the second one is associated to the business efficiency -trust and ease in doing business -in a determined country, which involves indicators regarding political and legal stability, as well as the administrative conditions necessary for doing business such as less bureaucracy and corruption i.e., political, legal and administrative factors.

Data collection
The selection of the sample relied on an official database exclusively released by the Brazilian Franchising Association, being all the information gathered from April 2012. The list obtained from the database includes 62 franchise chains with international operations and, after a statistic treatment which excluded the outliers and chains with considerable number of incomplete information, 54 Brazilian franchise chains with international operations in 26 countries were selected, which implies in 157 franchises operating abroad; this is, therefore, the absolute number of the sample.
To all 157 franchises selected there are neither missing data related to their external environment nor to their international commitment. The 157 franchise units from the 54 franchise chains vary considerably: some chains have only one unit abroad while others have more than 16. The number of units showed in Table 1 consists of the international commitment indicator, which is the dependent variable of the model presented in Figure 1. It is important to emphasize that the aim of this research is to know if the international commitment differs according to characteristics of the external environment of the group of countries in which Brazilian franchises operate. Table 2 shows that these numbers reflect the data obtained from ABF, demonstrating the internationalization of Brazilian franchises prominently to countries like Portugal, United States, Paraguay, Argentina, Angola and Spain.  Regarding external environment factors, the independent variables form a group divided into (i) market opportunity and (ii) business efficiency -trust and ease to doing business. Note that "market opportunity" is divided into demographic and economic factors. With regard to demographic factors, one needs to highlight the importance of analyzing the population of a country as well as its level of urbanization. Choosing countries with high rates of urbanization implies the presence of shopping centers and shopping malls, a retail environment favorable to franchises (Yavas, 1988;Arthur Andersen, 1996;Aliouche & Schlentrich, 2009, 2011Alon & Shoham, 2010). These data were obtained from the World Urbanization Prospects [WUP] (UN) and were measured in absolute number of people belonging to a certain country and living in urban areas. The economic factors involve the analysis of the consumption power of the population of a foreign market and, for this purpose, they must consider the local income per capita (Yavas, 1988;Arthur Andersen, 1996;Alon, 2006;Alon & Shoham, 2010). These data are a continuous variable measured in millions of dollars and were obtained from the "International Comparison Program [ICP]" (World Bank).

INTERNATIONAL ANALYSIS OF THE COUNTRIES WHERE BRAZILIAN FRANCHISE CHAINS OPERATE
In turn, factors related to business efficiency -trust and ease in doing business (Eroglu, 1992;Alon and McKee, 1999;Aliouche & Schlentrich, 2009, 2011Alon and Shoham, 2010) -are indicators of the political and legal system, especially regarding the trust in the governmental system and in the legal and regulatory systems of the country. These data were obtained from the International Country Risk [ICR] (PRS Group) and vary from 0 to 1, given that the higher the score, the higher the level of trust in the governmental, regulatory and legal system. The indicators related to bureaucratic matters serve as  The data related to the ease of doing business in the country were obtained from the ranking Doing Business, given that the lower the rate, the easier it is to do business. Since these data are inverse to the other ones, it is expected that the factor analysis indicate opposite results in comparison with the other business efficiency variables. The corruption data vary from 0 to 10: the higher the rate, the lower the level of corruption in the country. Figure 2 summarizes the variables involved in the analysis of the Brazilian franchise chains with operations abroad.
Given the fact that all variables, descriptive statistics and correlations have been presented, the following session aims to indicate through an exploratory factor analysis in how many factors the external environment indicators can be reduced. After obtaining the factors, a cluster analysis will be conducted aiming to determine the different groups of countries in which the franchises operate. Finally, an analysis of variance will be performed in order to verify if there is any difference among the average number of units abroad in each of the groups found in the cluster analysis. Table 3 shows the independent variables, their means and standard deviation and the existing correlation. There are no missing data in the selected sample. Two observations have to be nevertheless made: the first one is that the correlation of the variable "ease to doing business" is negative, since the higher the ease to doing business, the lower the numerator; the other variables have the opposite behavior (the higher the indicator, the higher the numerator). The second observation is that it is possible to verify correlations among the variables, which can be considered medium (between 0.5 and 0.7) and strong

Figure 2: Composition of the Variables
(above 0.8) in both groups: "market opportunity" (population of the country, urban population and purchasing power parity) and "business efficiency -trust and ease to doing business" (governmental and regulatory risks, corruption and ease in doing business). The one exception is the variable "purchasing power parity", which has a strong correlation to the variable "corruption perception". This outcome may indicate an association between purchasing power parity and a determined business efficiency group. Factor analysis was used to check whether the variables composed two expected or more representative factor of the external environment aspect. The factor analysis is exploratory and it was carried out with eigenvalues greater than 1. The rotation method used was the "varimax" (Hair, Black, Babin, Anderson, & Tatham, 2006).
In the case of factor analysis, the KMO statistics showed results at 0.734, signaling the appropriateness of the construct (Corrar, 2007). In turn, all Bartlett's Test of Sphericity (BTS) indicate p<0.050 (sig.), confirming the required assumption for the analysis (Hair et al., 2006;Corrar, 2007). In Table 4, it is possible to observe a high level of commonalities, since the results are superior to 0.800 (Fávero, Belfiore, Silva, & Chan, 2009). In this table, the factor loadings and the total variance explained are also represented.
In Table 4, it is also possible to observe that two factors were retained: factor 1, representing "business efficiency -trust and ease in doing business"; and factor 2, represented by "market opportunity". Note that the purchasing power parity showed a low loading regarding the other factors; it remained however in the factor 2 "market opportunity".
Given both retained factors, the second part of the statistical test consists of the grouping of countries in which Brazilian franchises operate. Firstly, a hierarchical cluster analysis was performed using the method between groups and the farthest neighbor, aiming to find an approximate number of clusters into which the sample could be divided. This number was three clusters, according to the standard of the coefficients of homogeneity (between groups) and the Pearson correlation (farthest neighbor). Assuming a solution with three clusters, we performed a Kmeans Cluster analysis, coming to a combination of three clusters as the most suitable to explain the sample (p <0.01). Figure 3 shows the components of each cluster and Table 5 shows the results of the clusters based on the characteristics of the external environment.
The results show that cluster 3, containing 20 franchises, which have the United States as a host country, is the one that presents the better market opportunities (p<0.05, check Table  6 ANOVA). On the other hand, regarding business efficiency -

INTERNATIONAL ANALYSIS OF THE COUNTRIES WHERE BRAZILIAN FRANCHISE CHAINS OPERATE
trust and ease in doing business -cluster 3 is the second most attractive one (p<0.05). Cluster 2, represented by 72 franchises, has the lowest market opportunity (p<0.05) due to the low population density. On the other hand, cluster 2 is the one that presents the best conditions of business efficiency -trust and ease in doing business (p<0.05). Finally, cluster 1 is the one that presents the worst business efficiency conditions [trust and ease in doing business (p<0.05)]. In terms of market opportunity, however, it is significantly superior to cluster 2 (p<0.05). Table 7, shows the test used to verify the main objective of this paper i.e., if there is any difference in the international commitment of the franchises, which operate in each one of these clusters. It is important to emphasize that some franchises are placed in one cluster whilst other ones are placed in more than one -some franchises, like CCAA and Carmen Steffens, are placed in the three clusters. This sort of placement doesn't interfere in the results because what one can observe in Table 7 is the mean of the international commitment of each franchise. Therefore, if the mean of a certain franchise is expressed by "n", then "n" will have the same weight to all clusters the franchise is inserted to.
The result shows that the average international commitment of the chains, which are inserted in cluster 1, is approximately seven international franchises, which doesn't differ significantly from the nine international franchises in cluster 2. There is however a significant difference (p<0.05) between the franchises of cluster 2 (eight franchises) and cluster 3 (four franchises). This difference indicates that in the mean the international commitment of the franchises operating in the United States (cluster 1) is smaller than the one of the franchises that predominantly operate in developed countries and in small Latin American countries (cluster 2). Figure 4 shows the dispersion of these clusters.

DISCUSSION AND IMPLICATIONS
Our findings suggest that in cluster 1, formed by underdeveloped countries in Latin America and Africa, the Brazilian franchises seek market opportunity, even though operating in countries with low business efficiency indicators (trust and ease to doing business). The explanation for this behavior lies in a higher ability developed by companies arising from emerging countries, like Brazilian franchise, in dealing with the external difficulties in their countries of origin. These difficulties involve contractual limitations, underdeveloped market regulations, inefficient judicial system, and high bureaucracy, among others. The ability to manage these scenarios can be reproduced in other emerging countries through similar problems. Thus, the disadvantage of operating in home countries with weak external aspects becomes a competitive advantage when operating in other emerging countries (Khanna & Palepu, 1997;Ghemawat & Khanna, 1998;Cuervo-Cazurra, 2006;Cuervo-Cazurra & Genc, 2008). Furthermore for Latin American and Portuguese language countries, this proximity is beneficial to companies arising from developing markets, making it easier to transfer resources and coordinate these markets (Cuervo-Cazurra & Genc, 2008). This gets even more evident when comparing to the franchises of cluster 3, which operate in the United States and are also seeking for market opportunity. However, it is possible to notice in the mean that the international commitment of the franchises belonging to this cluster is smaller than the other ones.
On one hand, the North American market is shaped by superior conditions in comparison to cluster 1, both in terms of market opportunity and in terms of business efficiency (trust and ease in doing business); on the other hand, Brazilian franchises operate in a highly competitive market in which it is harder to establish/conquer an immediate competitive position.
In cluster 2, consisted of developed countries and small Latin American countries, the business efficiency (trust and ease in doing business) conditions are the main attractiveness to open up a franchise, since in terms of market opportunity this cluster is the worst among the three analyzed ones.
As there is a significant difference in the international commitment in clusters 2 and 3, the business efficiency (trust and

Figure 4: Clusters Dispersion
ease in doing business) together with market opportunity creates an extremely competitive environment, which is attractive but also demands certain caution for committing internationally. It is therefore possible to infer that the international commitment is mainly driven by business efficiency (trust and ease in doing business) instead of market opportunity.

CONCLUSION
This paper has demonstrated that the market opportunity and the business efficiency (trust and ease in doing business) were the two factors related to the external environment, which can explain the international commitment of Brazilian franchises. The paper shows that the franchise chains operating in the United States (cluster 3) have an inferior international commitment in comparison to the franchises, which operate in developed countries and in small Latin American countries (cluster 2). It is also possible to notice a large number of franchises that operate in countries (cluster 1) with worse business efficiency due to the advantage of learning how to operate in a country that could have some similarities with Brazil.
This paper presents a research involving the largest number of Brazilian franchise chains operating abroad. Unless regarding to external factors and international commitment, this is the largest representation ever studied. This is the first paper in Brazil that establishes and determines how external factors explain the different strategic behavior of the international commitment of Brazilian franchises. Finally, the result of the three distinct clusters show how the market opportunity and the business efficiency (trust and ease in doing business) work as drivers to the international operation of Brazilian franchises.
In the end, this paper contributes to the existing literature by presenting the external environment relevant factors of the countries where the Brazilian franchises chains operate. Other studies had already been conducted with a focus of analysis on the international environment for North American franchising chains, through the creation of indexes for choosing the markets (Aliouche & Schlentrich, 2009, 2011. As well as the creation of clusters involving countries included in the Globe Study (Alon & Shoham, 2010). However, the studies were not exclusively focused on the behavior of franchise chains coming from emerging countries, specifically Brazil.
In terms of managerial implications, the agglutination into clusters of a series of countries in operation through the Brazilian franchise chains can facilitate the choice of franchise executives for international markets through systemic analyses involving two groups of factors in the external environment. It is primarily a matter for each franchise chain to analyze which attributes of the international environment considered in detail herein are the most relevant ones to its international operations, given their internal needs and competencies. Another contribution to the executive environment refers to the assistance in the management of these markets, aiming at establishing regional business policies based on markets with similar characteristics. Thus, this research is beneficial for both Brazilian franchise chains wishing to start operations abroad, as Brazilian franchise chains that already have international operations.
In turn, it is important to highlight the limitations of the study that can be set up as opportunity for future research, in particular, those related to the analysis of the external environment, which refer to the selection of indicators. Even though we chose indicators edited by credible international institutions, the formation of the external environment factors were limited to the variables selected. The deepening of the variables can express more accurately the behavior of these factors.
Even though the indicator "cultural distance" shows up in some studies analyzed in the theoretical framework (Alon & Soham 2010;Aliouche & Schlentrich, 2011), this indicator will not be used in this paper. The first reason for this choice is that the authors above mentioned use Hofstede's index to measure distance; if the same index were used herein, the sample would have been severely reduced due to missing data. The other reason is that, even though the GLOBE indicator could have been used, the cultural indicator only measures cultural distance, whereas all the other indicators measure country's absolute data.
In addition, this research only included countries where Brazilian franchise chains have some kind of international operations. Thus, it did not include other potential markets for Brazilian franchise chains. Other researchers can deepen this research by demonstrating the impact of environmental characteristics and the success of franchise chains that operate in each one of the three clusters. Other relevant research topics that could be explored relate to the following questions: in which cluster are the Brazilian franchise chains more successful and which environmental aspects influence such success? These research topics can contribute to the continuity and improvement of the empirical knowledge of the theoretical framework proposed in this paper.

International analysis of the countries where Brazilian franchise chains operate
This paper aims to demonstrate which external environment factors are involved in the international commitment of Brazilian franchise chains. Our objectives herein are to understand which external country characteristics lead to international franchising operations and to ascertain the influence of such characteristics in the commitment of franchise chains in each country they operate in. The database has 54 Brazilian franchise chains with international operations in 26 countries, which implies in 157 franchises operating abroad. Regarding external environment factors, the independent variables form a group divided into market opportunity and business efficiency -trust and ease to doing business. The result of the three distinct clusters show how the market opportunity and the business efficiency (trust and ease in doing business) work as drivers to the international operation of Brazilian franchises. The paper shows that the franchise chains operating in the USA (cluster 3) have an inferior international commitment in comparison with the franchises which operate in developed countries and in small Latin American countries (cluster 2). It is also possible to notice a large number of franchises that operate in underdeveloped countries from Latin America and Africa (cluster 1) with worse business efficiency due to the advantage of learning how to operate in a country that could have some similarities with Brazil Keywords: franchise chains, franchising, Brazilian franchise chains, international business, entrepreneurship.