Shareholder internationality and importing activities of emerging market firms

In light of the increasing efforts made by emerging market firms to engage in international business through importing activities, identifying the characteristics that motivate importing business and contribute to its success is practically and theoretically meaningful. Drawing upon a knowledge-based view (KBV), we examine how the shareholder internationality affects a firm’s importing activities. We hypothesize that the shareholder internationality can facilitate a firm’s import initiation and contribute to the diversity of importing countries of origin. Moreover, the divergence of shareholders’ nationality backgrounds may hinder import initiation, but motivate importing from dispersed countries of origin during the import development process. A longitudinal analysis of Colombian firms supports our hypotheses.


Introduction
Integrating into the global economy by participating importing activities has become increasingly important for firms as a result of reduced trade restrictions, improved international trade infrastructure, and global integration (Aykol et al. 2012). Especially in most emerging markets, importing has a considerable value not only in financial terms but also in diversity, technology diffusion, and employment (Bianchi and Saleh 2011;Goldberg et al. 2010). However, the international business literature on firm internationalization has mostly focused on foreign direct investment (FDI) and exporting (e.g., Iqbal et al. 2019), leaving aside the study of importing as a key international business activity (Bianchi and Saleh 2011). Although insightful studies have been conducted by previous researchers to investigate topics such as importer-exporter relationships (e.g., Aykol and Leonidou 2018;Bianchi and Saleh 2011;Silva et al. 2019), a basic and critical issue, namely importer profile, has received limited research attention (Aykol et al. 2013;Liang and Parkhe 1997;Quintens et al. 2006). Questions such as "what are the differences between importers and non-importers" and "what are the behaviors of importers in searching for sources of supply" have not been fully explored. Since identifying importer characteristics helps to offer a better understanding of the design and implementation of firm importing strategies, scholars have called for diverse perspectives to find valuable antecedents of firm importing behaviors (Aykol et al. 2013). Apart from firm characteristics (e.g., age) and product characteristics (e.g., design stability), shareholding characteristics that may capture the knowledge flow and influence firm strategic decisions are worth exploring (Jiang et al. 2018;Singla et al. 2017).
In emerging economies, the role of shareholders is predominantly influential in a firm's decision making due to one notable feature of corporate governance, namely high ownership concentration (Bhaumik et al. 2010;Peng et al. 2008). Ownership concentration describes the distribution of stock ownership (Hill and Snell 1989). In emerging markets where the institutional environments are weak, ownership concentration tends to be high (Vega Salas and Deng 2017). Young et al. (2008) posit two reasons to explain the phenomenon. First, due to the weak institutions, emerging market firms do not trust professional managers and outside investors, making the ownership decentraliazation difficult. Second, since boards of directors are less likely to play a strong monitoring and controlling role due to the lack of formal and informal institutional support, emerging market firms have to depend on dominant ownership to keep potential managerial opportunism in check. Therefore, major shareholders own controlling power over the firms, possess the ability to constrain managers, and influence the decision (Hill and Snell 1989;Liu et al. 2011). A practically and theoretically meaningful perspective to investigate the role of shareholders is the shareholder internationality which describes the mixture of nationalities among shareholders. On the one hand, with the globalization of investment, foreign shareholders in firms have brought substantial changes to corporate governance (Aguilera et al. 2019;Desender et al. 2016). On the other hand, the impact of shareholder internationality on firm internationalization has received broad attention from international business scholars (e.g., Filatotchev et al. 2008;Singla et al. 2017). While the stream of research mainly focuses on the exporting or FDI activities (e.g., Bhaumik et al. 2010), few studies have noticed that international shareholders, as sources of international knowledge, may also influence importing, which is a way of entering international markets and may facilitate exporting and subsequent internationalization activities (Aykol et al. 2012).
Drawing upon a knowledge-based view (KBV), this study analyzes how international shareholders and the divergence of their nationality backgrounds influence a firm's import initiation and import development strategies. First, we posit that the presence of international shareholders in the firm (i.e., shareholder internationality) has positive relationships with import propensity and the diversity of importing countries of origin. These two hypotheses emphasize the complementary effects among foreign and local shareholders. Second, we focus on the shareholder heterogeneity and hypothesize that the divergence of shareholders' nationality backgrounds is negatively related to import propensity but positively related to the diversity of importing countries of origin. To test these hypotheses, we employ panel data of Colombian firms from 2009 to 2016, and the empirical results support the hypotheses.
This study makes three theoretical contributions. First, we extend the international business literature by investigating the antecedents of importing activities in emerging markets which is largely unexplored in previous studies. Second, we underscore the value of studying how shareholders' characteristics drive international business in emerging markets (Cuervo-Cazurra 2012; Singla et al. 2017) from the importing perspective. Third, this study enriches the KBV literature because it identifies different types of knowledge, and analyzes both advantages and disadvantages of knowledge heterogeneity in international business. In addition, it advances the KBV literature by identifying shareholder internationality as an important source of knowledge to facilitate importing activities.

Theoretical background
Importing as an internationalization process The process of business internationalization includes outward activities, such as exporting and overseas investment, and inward activities, such as importing (Karlsen et al. 2003). With more firms moving toward global strategic sourcing, importing activities have become increasingly important (Aykol et al. 2012). A firm that seeks expansion may need to import to gain access to diverse materials, advanced technology, and global quality improvement (Hanafee 1984). Such a firm can also build relationships with foreign firms (Karlsen et al. 2003), learn from importing intermediates to motivate local innovation (Augier et al. 2013;Mody and Yilmaz 2002), enlarge product scope (Goldberg et al. 2010), and gain other important benefits that can give the firm a competitive advantage in current market and reduce risks.
The import development process is suggested to be a valuable topic for further exploration (Aykol et al. 2013). Import initiation needs to overcome hurdles. One barrier inhibiting import involvement is the extra costs arising from trade, such as exchange rate fluctuations and transportation delays (Katsikeas and Dalgic 1995). The other barrier is the uncertainty of foreign markets and institutions. A firm that lacks information on global supply sources and foreign business practices may stick to local purchasing (Ford 1984). Such barriers may affect the risk orientation of the decision-makers of a firm. Therefore, a firm's belief in its capacity to overcome barriers may facilitate import initiation and further development.
However, once the importing process begins, importers start to appreciate the benefits in cost reduction, quality improvement, technology spillovers (Augier et al. 2013), and other opportunities, and then, importing becomes a desirable strategy (Liang and Parkhe 1997). At this stage, firms need to determine their import source strategies, such as source diversification versus concentration (Aykol et al. 2012;Aykol et al. 2013). Rather than relying on a single supplier, source diversification denotes that a firm allocates its import sources across different suppliers (Muhammad 2012). While taking geographic factors into account, the diversification of importing countries of origin refers to a firm that has a broad scope of import sources across the world rather than only focusing on suppliers from a few countries (Patel et al. 2018). The reasons for expanding the set of importing countries of origin include (1) searching for suppliers with the best quality or price, (2) controlling country-specific risks on price and supply chain, (3) leveraging sunk costs and using capabilities already built, (4) gaining flexibility from the ease of switching, and (5) learning knowledge from diverse sources for further application (Jaimovich 2012;Wolak and Kolstad 1991). All these factors may contribute to firms' competitiveness.

Shareholder internationality and firm internationalization
The impact of shareholder internationality on firm internationalization has been studied in prior research with mixed findings. Some studies suggest that it affects a firm internationalization (e.g., Erramilli 1996;Hennart and Larimo 1998), whereas others find no significant effect (e.g., Kim and Daniels 1991;Lien et al. 2005). Those findings have encouraged open discussion in the literature. For the specific measurement of internationalization on this issue, some studies focus on FDI. For instance, a study on the internationalization of Indian firms confirms the value of international shareholders and examines how their good interaction increases a firm's outward FDI (Singla et al. 2017). Other studies on exporting, such as Filatotchev et al. 2008, have found that foreign shareholders' control of strategic decisions has a positive relationship with export propensity. However, there is still a gap in import literature concerning how foreign participation in a firm's ownership affects its import decisions, such as whether to import and where to import. Since importing is an important mode of entering international market and may facilitate exporting and subsequent internationalization activities (Aykol et al. 2012), international ownership may influence the effectiveness and the efficiency of both import initiation and development.
In addition, studies have focused on the advantages of multinational firms to expand in emerging markets through local partnerships (Meyer 2001;Meyer et al. 2009;Tong et al. 2008), but this is not enough. The selection of partners is both difficult and critical for success in internationalization (Hitt et al. 1997b). Apart from the assessment of complementary assets that are relatively visible, shareholders' motivations and capabilities can be diverse, and may significantly influence the collaboration efficiency when implementing strategies such as internationalization (Hitt et al. 2000;Singla et al. 2017). Thus, an in-depth examination is needed to discuss how the inner structure of international shareholders affects their interaction and consequently, the internationalization decisions. Heterogeneous owners provide diverse resources that enhance a firm's capacity for internationalization (Singla et al. 2017). However, the differences among them cause different behaviors, methods, and goals. These obstacles may hinder effective internal communication, create conflicts, and lead to different motivations in internationalization decisions (Tihanyi et al. 2003;Wu and Sekiguchi 2019).

A KBV on shareholder heterogeneity
This study draws upon a KBV framework, which regards knowledge as the core strategic resource and the source of sustainable competitive advantage (Conner 1991;Grant 1996a, b;Spender 1994). Having been applied to diverse areas, such as organizational studies, strategic management, and marketing, KBV studies can be divided into two streams from the perspective of knowledge. One stream is somewhat static and focuses on the knowledge base of a firm consisting of two types of knowledge, namely information and know-how (Kogut and Zander 1992). The other is relatively dynamic, which emphasizes the exploitation, creation, combination, integration, and application of the knowledge base (Gibson 2001;Nonaka 1994;Nonaka et al. 2006). Moreover, KBV research has been conducted at different levels, including intrafirm levels. Intra-firm knowledge plays a critical role in successful internationalization (Inkpen and Dinur 1998;Minbaeva et al. 2003;Prashantham 2005). KBV is extensively applied in international business research (e.g., Dau 2018;Meng et al. 2019). First, two types of objective informational knowledge are necessary for conducting international business, namely (1) market knowledge about customers, suppliers, competitors, and other resources in target markets (Morgan et al. 2003), and (2) institutional knowledge about government, laws, regulations, and norms in foreign markets (Eriksson et al. 1997). Second, know-how or experiential knowledge involves (1) experiential market knowledge, (2) experiential institutional knowledge, and (3) internationalization knowledge concerning the understanding of the compatibility of a firm's present capability and the capability needed in international business (Eriksson et al. 1997). Such knowledge facilitates a firm's better understanding of foreign markets and institutions, enabling it to discover opportunities and reduce risks in internationalization.
To ensure effective knowledge processes, the management of knowledge heterogeneity is indispensable. Knowledge heterogeneity refers to "the state of a collective knowledge structure of an organizing unit in which the members configure knowledge of various attributes (e.g., tacit vs. explicit) from different disciplinary, managerial or technical areas by utilizing different processing methods" (Tsai 2016(Tsai : p. 1165). Heterogeneous knowledge can be beneficial since it offers a possibility for further knowledge creation and use. However, it may also hinder effective knowledge communication and integration (Kearney et al. 2009).
Although knowledge itself is abstract, it is held by individuals and is exhibited at multiple levels, such as groups, organizations, and networks (Kogut and Zander 1992). Prior studies generally regard individuals or firms as carriers of knowledge, and utilize the diversity of backgrounds and experiences as proxies for heterogeneous knowledge (e.g., Dau 2016;Elango and Pattnaik 2013). For example, the diversity in educational or functional backgrounds denotes the existence of knowledge heterogeneity (e.g., Meng et al. 2019;Smith et al. 2005). In corporate governance, shareholders can be the proper carriers for knowledge at an intra-firm level. Different types of owners, such as family owners and foreign owners, tend to own diverse knowledge (Liesch and Knight 1999;Singla et al. 2017).

Shareholder internationality and import propensity
The presence of foreign shareholders in a local firm helps to reduce the level of risk aversion of local shareholders and encourages the firm to import. In emerging markets, local shareholders often have a risk aversion tendency when considering importing activities. First, they are generally driven by the motivation to improve their competitiveness in domestic market, so they tend to buy from local suppliers to ensure the stability of resource supply. Local shareholders and suppliers usually share the same culture, business routine, and a common understanding of the sources of purchasing (Liang and Parkhe 1997;Niffenegger et al. 1982). Even if they need materials or components that are scarce in the home market, they still opt for local sources first even at a higher cost. Second, importing is a risky strategy as a firm needs to pay in advance and wait for the goods, which may adversely affect the stability of supply and endanger existing business lines (Singla et al. 2017). Third, once a firm decides to import, a series of factors must be considered, such as the social, legal, economic, and political frameworks in the countries of origin (Madhok 1997). However, the lack of specialized human capital makes the firm unable to obtain specific knowledge about foreign countries, which is necessary to deal with different conditions in the external market (Gomez-Mejia 1988).
Foreign shareholders are not only sources of capital but also sources of knowledge, which can reduce the uncertainty and add positive anticipation of import, and thereby, reduce the risk aversion of local shareholders. First, the participation of foreign shareholders in a firm may compensate for the absence of specialized human capital in emerging markets. They transfer knowledge of foreign markets and institutions (Beamish and Banks 1987), enabling local shareholders to anticipate and compare the benefits and costs between local and foreign sources. Moreover, the knowledge of foreign shareholders can enable the absorption of technology and enhance the capability needed for global sourcing (Cuervo-Cazurra 2012; Lane et al. 2001), which can help the local shareholders see more advantages than risks in import (Beck 1999). Second, since foreign firms gain the present ownership in the focal firm by outward FDI, they gain experiential knowledge, and thus, can understand and manage the risks involved in internationalization (Singla et al. 2017); this may also reduce the risk aversion of local shareholders. Therefore, the complementarity of local and foreign shareholders can remove the knowledge barriers in importing, and reveal its advantages by reducing the uncertainty of doing business in different markets and adding positive anticipation of importing, which contributes to importing initiation.
Hypothesis 1a: Shareholder internationality has a positive relationship with import propensity.

Shareholder internationality and the diversity of importing countries of origin
A firm undergoing importing activities with the presence of foreign shareholders tends to move toward a geographically dispersed distribution of importing efforts. First, the presence of foreign shareholders motivates the firm to compare suppliers worldwide and to optimize its import supplier arrangements, thereby, contributing to import diversification. Since products within the same category may have different prices and quality across countries of origin with different market environments, fruitful knowledge brought by foreign shareholders enables a firm to assess suppliers comprehensively by evaluating product prices and quality as well as import costs, and choose the most suitable suppliers on a global scale (Muhammad 2012). Therefore, the locations of the firm's suppliers are likely to be dispersed.
Second, a firm with foreign shareholders tends to have strong capabilities and confidence to get involved in importing on a greater scale and utilize diversification strategy to control the increasing risks. Along with the increasing scale of imports, the impact of risks on the firm may grow. To reduce risks, such as supply chain disruption caused by country-specific shocks, importing from diverse countries of origin can be helpful (Jaimovich 2012). Compared with firms with only local shareholders, a firm with foreign shareholders is more likely to pursue the diversification strategy due to stronger  (Kogut and Zander 1992). Local shareholders have informational and experiential knowledge concerning local market and present business (Williamson and Wan 2018), while foreign shareholders bring about informational and experiential knowledge about different markets, institutions, and international operations. The combination of such knowledge held by foreign and local shareholders can create new knowledge in the firm, strengthening its capabilities and encouraging it to become better embedded in its local and international business (Cohen and Levinthal 1990; Kogut and Zander 1992).
Hypothesis 1b: Shareholder internationality has a positive relationship with the diversity of importing countries of origin.

Divergence of shareholders' nationality backgrounds and importing
Divergence of shareholders' nationality backgrounds and import propensity The divergence of shareholders' nationality backgrounds describes the variance of heterogeneous ownership. It may cause difficulty in reaching an agreement on whether to import; therefore, shareholders are less likely to initiate importing. First, the present state and behavior of an individual are the result of past experiences, conditions, and constraints (Simsek et al. 2015). Accordingly, different nationality backgrounds imply that each shareholder has a specific set of values, beliefs, and preferences, which may cause conflict and hamper effective interaction (Watson et al. 1993). Second, based on KBV literature, knowledge heterogeneity in shareholders with divergent nationality backgrounds may cause difficulties for internal knowledge processes (Tsai 2016), such as knowledge exchange, integration, and utilization. As a result, mutual understanding and coordination among shareholders are hindered. Therefore, it will be difficult to reach a consensus on whether to adopt risky strategies such as importing.
Hypothesis 2a: Divergence of shareholders' nationality backgrounds has a negative relationship with import propensity.

Divergence of shareholders' nationality backgrounds and the diversity of importing countries of origin
Shareholders with different nationality backgrounds have heterogeneous knowledge about markets, institutions, and international operations (Tihanyi et al. 2003). In a firm with shareholders more diverse in nationality backgrounds, a higher degree of knowledge heterogeneity is present (Tsai 2016). On the one hand, such broad knowledge offers a firm more information about the distribution of product price and quality across countries and about the environments of different markets. On the other hand, the exchange and combination of such diverse knowledge can create new knowledge (Cohen and Levinthal 1990). Initial informational and experiential knowledge, as well as new knowledge, may deepen the understanding of different markets, institutions, and how to do international business. Therefore, in the import development process, as the divergence of shareholders' nationality backgrounds increases, a firm may command information about a larger number of countries, enabling them to search for suitable suppliers extensively (Aykol et al. 2012). As a result, the distribution of the final portfolio of importing countries of origin is likely to be geographically dispersed (Fig. 1).
Hypothesis 2b: Divergence of shareholders' nationality backgrounds has a positive relationship with the diversity of importing countries of origin.

Data and sample
This study uses Colombian firms from 2009 to 2016 as the research setting. The reasons we focus on Colombian firms are as follow. First, there is a call to develop research in understudied emerging economies (Hoskisson et al. 2000). Colombia is a typical Latin American economy with a liberal economy, relatively small population, underdeveloped human capital, and low growth rate (Fastoso and Whitelock 2011). Second, consumption expenditure in Colombia represents more than 80% of the total demand, and imports are higher than exports in recent years (United Nations 2016); hence, it offers a suitable environment to analyze importing. Finally, there are internal social conflicts, and the institutional environments are weak in Colombia (World Bank 2017). Therefore, the ownership concentration is high, making Colombia a proper context to investigate shareholder characteristics Young et al. 2008). This study combines several sources to compile a panel dataset. The first data source is the BvD Osiris database. It provides firm-level information about the shareholders' nationality and helps to measure shareholder internationality and the divergence of shareholders' nationality backgrounds. The second data source is the Colombian Customs. This source provides information about the import amount in U.S. dollars per year, firm names, countries of origin, and tariffs. The third source is the Informa Colombia S.A. which provides information about a firm's inception time of activities to measure the firm's age and experience. The fourth source is the Heritage Foundation which provides information about the institutional index per country and per year, which helps to measure the divergence of shareholders' nationality backgrounds. The last source is the open data from the Colombian government used to measure the firm size variable. The research has access to more than 30,000 firms' information about importing from the Colombian customs organization and the Colombian government's open data. In addition, 261 Colombian firms' information about Colombian shareholders is obtained using the Osiris database. However, the overall information about shareholders' nationality is scarce, which limits the total size of our research sample to 115 Colombian firms.

Dependent variables
Import propensity This variable reflects a firm's motivation for importing, showing its avoidance or predilection for internationalization. The propensity variable is widely used to capture the international behavior of a firm (Fernandez and Nieto 2006;Lu et al. 2009). We create a dummy variable to indicate whether a firm imports each year.
Diversity of importing countries of origin This variable shows the degree of internationalization. Prior studies use several variables to measure the degree of internationalization, such as the number of countries that a firm operates in (George et al. 2005;Reuer and Leiblein 2000;Vermeulen and Barkema 2002). However, this measure of diversification does not reflect the risks involved because firms may import products as samples or in very small amounts. In this study, we use the entropy measure of diversification which considers the total imports and imports from per origin country. One of the advantages of this measure is that it gives more weight to the import amount than the import activity. It has been widely used to measure diversification (Deng et al. 2018;Hitt et al. 1997a). Therefore, the diversity of importing countries of origin is measured by ∑ i [P i × ln(1/ P i )], where P i is the proportion of imports from overseas market i to total imports from all markets in U.S. dollars.

Independent variables
Shareholder internationality It is represented by a dummy variable and takes the value of 1 if a firm has at least one foreign shareholder and 0 if the firm only has local shareholders.

Divergence of shareholders' nationality backgrounds
This variable shows the variation of shareholders' backgrounds by analyzing the absolute deviation of the shareholders' nationality backgrounds. We use the Heritage Foundation's Index of Economic Freedom, which provides 10-dimension indices of country institutions and an overall index per year. Each index is on a 100-point scale, with 100 points denoting the highest degree of economic freedom. We use the overall dimension index to measure each shareholder's nationality background since we regard individual shareholders as a reflection of a set of complex economic environments in their nation, and the overall dimension index can incorporate diverse aspects of the economic situations, such as business freedom, trade freedom, investment freedom, and property rights. The divergence of shareholders' nationality backgrounds is measured by (Σ i │a i -ā│) / n, where a i is the overall dimension index of a shareholder's background according to their nationality i, ā is the average overall dimension index of all the shareholders, and n is the number of shareholders in the firm.

Control variables
Experience Considering that the accumulation of knowledge may influence the firm's importing decisions, we subtract the date of a firm's founding year from the current evaluation year to measure the firm's experience as a control variable following Chang et al.'s work (2013).
Value-added tax (VAT) We employ the average VAT that each firm pays for importing products per year to control for non-ownership behaviors. If a firm does not register imports in a period, the average VAT paid by firms in the same industry is generated to represent its VAT. Previous studies support the effects of taxes on importing (Fung et al. 2011).
Tariff Government policies can influence demand and a firm's profitability through tariffs. For example, falling domestic tariffs are associated with a firm's declining profits (Baggs and Brander 2006). Therefore, tariffs may influence imports. We control the average tariffs paid by a firm for imports per year, or the average tariffs of a firm's industry if it does not import.
Size This variable is related to the level of internationalization because it reflects organizational complexity (Singla et al. 2017). The number of employees is used in this study to represent the variable.
Year effects and industry effects are included in all models.

Model design
This study uses two models for the regressions in Stata 13.0. First, since our first dependent variable, import propensity, is a dichotomous outcome variable, we employ a panel data Probit model to evaluate the relationships in H1a and H2a.
Âdivergence of shareholders'nationality backgrounds where X denotes a vector of control variables as specified earlier.
Second, a linear regression model is used to examine the relationships in H1b and H2b, as the second dependent variable, the diversity of importing countries of origin, is constructed based on the entropy measure. This model is run only if a firm imports in a given year (i.e., the variable import propensity equals 1).
H1b : diversity of importing countries of origin H2b : diversity of importing countries of origin where X denotes a vector of control variables as specified earlier.
To decide between fixed and random effects, we use the Hausman test (Greene 2008), and confirm that a random effect model is more appropriate than a fixed effect model. We lag all explanatory and control variables by 1 year to help clarify causal relationships.

Baseline results
Table 1 provides descriptive information and the correlation coefficients among variables. Almost all correlations among independent and control variables are lower than 0.3. The average variance inflation factor is 1.19, which is well below 10, suggesting no concern of multicollinearity. The independent variable shareholder internationality or the presence of international shareholders in the firm is on average 0.635. The divergence of shareholders' nationality backgrounds is on average 1.813, representing a low value compared with the median, as the values range from 0 to 9.950. Table 2 provides regression results. Model 1 shows a positive relationship (p = 0.002) between import propensity and shareholder internationality, which confirms H1a. Foreign shareholders bring informational and experiential knowledge about foreign markets, institutions, and international operations, reducing the uncertainty of doing business in different markets, and revealing the advantages of importing, thereby, reducing the risk aversion of local shareholders. The result in Model 2 shows a significantly positive relationship (p = 0.025) between the diversity of importing countries of origin and shareholder internationality, which supports H1b. The combination of the local and foreign shareholders' knowledge enables a firm to buy from global suppliers and reduce risks, leading to a dispersed distribution of importing countries of origin.
Model 3 reports the regression results of the relationship between import propensity and the divergence of shareholders' nationality backgrounds, which is significantly negative (p = 0.003), confirming H2a, since knowledge heterogeneity hinders knowledge understanding, exchange, and integration processes and thus it is difficult to reach a consensus on whether to adopt risky strategies such as importing.
Model 4 shows a significantly positive relationship (p = 0.065) between the diversity of importing countries of origin and the divergence of shareholders' nationality backgrounds, which supports H2b. This result shows that once an importing decision is taken, the diversity of shareholders offers a wide range of knowledge about a large number of sourcing countries. Hence, the importing countries of origin may be diverse.

Robustness tests
We conduct three sets of tests to ensure the robustness of the findings. First, we use the panel data Logit models to replace the Probit models to evaluate import propensity. Logit regression is also commonly used when the dependent variable is binary (e.g.,  Estrin et al. 2009). The results in Models 1 and 3 of Table 3 support H1a and H2a. Second, we employ panel data Tobit models instead of linear regression models to test H1b and H2b. Tobit regression is used to avoid estimation bias when the dependent variable is censored (Wooldridge 2002). The results in Models 2 and 4 of Table 3 also support H1b and H2b. Finally, we use the Herfindahl index to measure the diversity of the importing countries of origin instead of entropy. The Herfindahl index is calculated by 1 -∑S j 2 where S j is the proportion of a firm's imports from country j, which has been used to measure diversification in firm internationalization (Vega Salas and Deng 2017). The results provide further support for H1b and H2b (results are omitted for brevity, but are available upon request).

Discussion
Recent years have witnessed an increase in the emerging market firms' participation in importing activities (Aykol and Leonidou 2018). However, several important issues, such as the relationship between shareholder characteristics and importing, remain unanswered. This research takes shareholder internationality as a potential antecedent of importing and examines the relationship between the shareholder internationality and the internationalization of the business. This study illustrates whether firms with certain governance characteristics are more likely to import. The arguments of this research are based on the view that foreign shareholders own informational and experiential knowledge about foreign markets, institutions, and internationalization processes, and shareholders from different backgrounds may have heterogeneous knowledge. We analyze the interaction including both complementarity and conflict among international shareholders caused by their divergent knowledge at the intra-firm level. Based on a KBV framework, this study has specified the relationship between shareholder internationality and importing. We first examine shareholder internationality from the role of foreign shareholders and then, discuss the divergence of international shareholders' nationality backgrounds. We consider a firm's importing from two dimensions. One is the import propensity concerning whether a firm will import or not. The other is the diversity of importing countries of origin, which measures whether a firm will import from suppliers in different countries or from suppliers within the same country. Our findings are largely in line with those in related studies. First, this study confirms that the complementary resources and understanding of different institutions are indispensable for managing the uncertainty of doing business abroad (Anderson and Gatignon 1986) and conducting international business (Madhok 1997). It is consistent with the literature that argues that shareholder internationality affects a firm's internationalization process (Erramilli 1996;Filatotchev et al. 2008;Hennart and Larimo 1998;Singla et al. 2017). Second, the findings on the dual effects of shareholder heterogeneity on internationalization are consistent with previous studies. Heterogeneous owners provide heterogeneous resources that can enhance emerging market firms' capacity for doing international business (Singla et al. 2017). However, the differences among shareholders involve different behaviors, methods, and goals, creating conflicts in making internationalization decisions (Tihanyi et al. 2003). For instance, the national diversity of foreign directors on the board has a positive relationship with international market operations (Estelyi and Nisar 2016). However, personal differences on the board may lead to conflict in the workplace because this diversity may misguide the common interests of the group (Farrell and Hersch 2005;Hilscher and Sisli-Ciamarra 2013).
Although based on the Colombian dataset, the findings of this study may shed light on other emerging markets, such as China. At present, China is the second-largest importing country in the world. Firms in China significantly depend on importing to acquire materials and products at a lower cost, better quality, and greater novelty. Although a large number of studies have explored the ownership-internationalization relationships in China (e.g., Boyd and Solarino 2016;Lu et al. 2009;Vega Salas and Deng 2017), few studies investigate the importing activities regardless of the critical role of importing for Chinese firms. Moreover, the ownership concentration is high in Chinese firms (Liu et al. 2011;Lu et al. 2009), which shares similarities with Colombian firms. This study suggests that Chinese firms, similar to Colombian firms that require the knowledge owned by international shareholders, should pay attention to the design of ownership structure to conduct importing activities effectively and efficiently. The similarities among emerging markets suggest the generalizability of our findings. However, we also acknowledge that each emerging market has unique characteristics, which offers opportunities for open discussion on the ownership-importing relationships in other emerging markets to explore the boundary conditions of our findings.
By analyzing shareholders' characteristics and their interactions, this research makes several contributions to the studies on shareholder-internationalization relationships in emerging markets. First, this study enriches the international business literature (e.g., Deng et al. 2018;Meyer et al. 2009;Wan 2005) by acknowledging importing as an important means of internationalization, and investigating its potential antecedents. International exchange activity, as a fundamental form of internationalization, has two sides, exporting and importing; one cannot exist without the other (Liang and Parkhe 1997). However, systematic research on this topic extensively focuses on export (e.g., Filatotchev et al. 2008;Gomez-Mejia 1988), leaving importing studies insufficient (Katsikeas and Dalgic 1995). This study utilizes an import development perspective, and incorporates the import propensity and the diversity of importing countries of origin in one study, enriching the import behavior literature (e.g., Aykol et al. 2013). It measures a firm's importing decisions from both the motivation and involvement perspectives, which opens a broad avenue for further research. Moreover, prior work on this issue is mainly concerned whether a firm will import from multiple suppliers (e.g., Whitney et al. 2014) while neglecting the geographical distribution characteristics of these suppliers. This study considers geographical distribution related to a firm's importing diversification by investigating the diversity of importing countries of origin. Additionally, this research identifies the corporate governance antecedents of importing behaviors and sourcing strategies, which have received limited attention in prior import studies (Aykol et al. 2013).
Second, the present study answers the call for incorporating owners' attitudes and decision making into studies in the contexts of emerging markets (Cuervo-Cazurra 2012; Singla et al. 2017). Most of the extant studies on shareholder internationality have focused on developed market firms (e.g., Ahmadjian and Robbins 2005;David et al. 2006;Desender et al. 2016). This research is conducted in Colombia, whose research findings can be generalized to many other neighboring countries in Latin America due to the high level of homogeneity (Aguilera et al. 2017), as well as other emerging markets such as China and India. In addition, extant literature primarily examines family and institutional owners (Boyd and Solarino 2016;Fernandez and Nieto 2006;Thomsen and Pedersen 2000), whereas we consider local and foreign shareholders or international shareholders, thus enriching the literature with more ownership types. To the best of our knowledge, this research is the first attempt to explore the mechanism through which shareholder internationality characteristics act as antecedents of importing behaviors.
Third, by employing the importing context, this study extends the KBV literature. As KBV is extensively used in IB research (e.g., Dau 2016), the micro-foundation of knowledge processes is not completely understood (Foss and Pedersen 2004). The theoretical framework of this study is integrated, identifying different kinds of knowledge in the intra-firm level. Moreover, the findings suggest there are both advantages and disadvantages of shareholders' knowledge heterogeneity in a global sourcing context. Although several previous studies have examined knowledge heterogeneity in firms' top manager teams (e.g., Smith et al. 2005), the research objectives here focus on the shareholders. Furthermore, this research contributes to the KBV literature by emphasizing import studies that have been ignored before. Previous studies have examined knowledge management in the multinational and international joint venture contexts (e.g., Gupta and Govindarajan 2000;Inkpen and Dinur 1998) and the export context (e.g., Morgan et al. 2003). However, theoretical or empirical studies need to be conducted to explore various ways of knowledge management in international business, such as importing. Two managerial implications emerge from this study. First, international shareholders can serve as the sources of worldwide knowledge about markets, institutions, and international operations, which help a firm to optimize its source strategy and take part in importing activities. This study highlights the importance of the complementarity of foreign shareholders and local shareholders in reducing the risks of doing international business. Second, attention needs to be given to the ownership structure to ensure the efficiency of understanding, interaction, and communication among shareholders in emerging market firms where high ownership concentration exists. It is critical to recognize the present stage of a firm when deciding the shareholder structure. In the decision-taken stage, the convergence of the shareholders' backgrounds is necessary. If their backgrounds have a high level of heterogeneity, it may cause conflict in the agreement to take risky decisions, and make further communication, cooperation, and work difficult. However, if a firm is undertaking an import business then divergent shareholders will be beneficial. Deciding on the suitable shareholder structure is important, or if the structure is settled down, enhancing the communication efficiency among the shareholders is recommended.
Admittedly, this study has several limitations that offer future research opportunities. First, the lack of available data hinders the identification of the degree of each shareholder's ownership participation. One of the most notable characteristics of emerging economy firms, such as Latin American and Chinese firms, is ownership concentration (Lu et al. 2009;Vega Salas and Deng 2017). However, due to data constraints, we cannot analyze the impact of the largest owners' behaviors and risk tolerance on a firm's international decisions. Second, one may dig deeper into the analysis of importing countries of origin in the future. For instance, one may wonder if a firm based in an emerging market will prefer importing from other emerging countries rather than developed countries because local shareholders have a better understanding of emerging markets' characteristics, and foreign shareholders have successful experiences in emerging countries. Moreover, foreign shareholders possess knowledge and experience about international markets, especially about their national or regional markets. Thus, the preference of shareholders to import from their own country can be investigated in a future analysis. Third, Colombia is a country with relatively simple trade modes, which means that Colombian firms import mostly for consuming in domestic markets rather than exporting. Therefore, this study offers the baseline findings regarding the relationships among shareholder internationality and importing behaviors. There are potential research opportunities to examine the relationships in other emerging markets with more complex and comprehensive trade modes, such as China, which can shed light on the boundary conditions of our findings.

Conclusion
This study explores an important, but largely unexplored question in emerging markets, namely the antecedents of importing behaviors. Based on a KBV framework, we identify shareholder internationality as a critical source of firm knowledge flow that may influence importing decisions. The empirical results support our hypotheses. The presence of international shareholders in the firm (i.e., shareholder internationality) has positive relationships with import propensity and the diversity of importing countries of origin. We also find that the divergence of the shareholders' nationality backgrounds Wang and Vega Salas Frontiers of Business Research in China (2020)  is negatively related to import propensity but positively related to the diversity of importing countries of origin. The findings contribute to the research on shareholderinternationalization relationships in emerging market contexts by acknowledging importing behaviors as a topic that merits further investigation and encouraging studies on the role of shareholders in firm strategic decisions.