The Moderating Effects of Host Country Governance and Trade Openness on the Relationship between Cultural Distance and Financial Performance of Foreign Subsidiaries in Latin America

: Cultural distance (CD) is an important driver of foreign expansion strategy at the ﬁrm level. However, its effects can be more or less signiﬁcant depending on the contextual characteristics of the host country, such as the quality of formal institutions and the openness to international trade. Therefore, it is argued that strong formal institutions in the host country can effectively reduce the adverse impact of CD. Additionally, due to the more frequent interactions with foreign cultures, countries open to foreign trade can positively accommodate the effects of CD. The study tests these assumptions using data from the Orbis database and the World Bank and ﬁnds a reduction in the adverse impact of CD on the ﬁnancial performance of foreign subsidiary ﬁrms with robust formal institutions in the host country. Moreover, the negative effects of CD increase with higher degrees of trade openness. Thus, the results indicate that foreign subsidiary ﬁrms operating in host countries that are more open to foreign trade will have to conform to the higher expectations from the local culture.


Introduction
The international business literature identifies cultural distance (CD) as an important driver of globalization strategy, for example, choice of location, degree of ownership, and entry and establishment goals (Kirkman et al. 2006;Beugelsdijk et al. 2018;Wang and Larimo 2020). However, there are concerns regarding the theoretical arguments, methodological procedures, and empirical results found in CD studies (Shenkar 2001;Dow 2017;Verbeke et al. 2017). According to Shenkar et al. (2020), to advance the knowledge of international business, there is a need to identify the conditions that allow the cultural traits to shape the internationalization outcomes. The study contributes to this debate by focusing on how formal institutions and the degree of trade openness in the host country moderate the effects of CD on the financial performance of foreign subsidiary firms.
Formal constraints are rules that human beings devise, while informal constraints include conventions and codes of behavior (North 1990). The institutions can be classified into "three pillars": regulatory, cognitive, and normative (Scott 1995) to understand how they impact firm behavior. The regulatory pillar includes laws and regulations that provide stability and order in societies. The cognitive pillar represents "world views" (i.e., how individuals in society relate to things). The normative pillar includes social values, norms, and culture. Kostova (1996) adopts this classification and defines institutional distance as the difference between two countries regarding their regulatory, cognitive, and normative institutions. According to Kostova et al. (2020, p. 470), "when companies do business across borders, they face a challenge not only to learn new ways of conducting certain functions noted by Whetten (1989, p. 492). "temporal and contextual factors set the boundaries of generalizability, and as such constitute the range of the theory." Results regarding the negative effects of CD on the financial performance of foreign subsidiaries in Latin America are in line with previous research (Singh et al. 2017;Beugelsdijk et al. 2018). However, findings reveal that the adverse effects of CD are moderated negatively by formal institutions in the host country. Therefore, strong formal institutions provide an effective mechanism to curb the adverse effects of CD on the financial performance of foreign subsidiary firms. Furthermore, contrary to the hypothesis, we find that the negative impacts of CD increase in countries that are more open to foreign trade. The increased negative effects of CD in countries that are more open to foreign trade are likely to be associated with the higher requirements on foreign subsidiary firms to conform to the expectations and requirements of the local culture.
Following the introduction (Section 1), the remaining parts of this study are organized into five major sections. Section 2 presents the theoretical background and hypotheses. The data and method are presented in Section 3. Section 4 discusses the results, and Section 5 provides the conclusions, limitations, and directions for future research.

Literature Review, Theoretical Background and Hypothesis
Institutions can be classified according to formality (North 1990) into three pillars (Scott 1995). According to Peng et al. (2009), formal institutions incorporate the regulatory framework and include formal rules, regulations, laws, and enforcement mechanisms and characteristics. Informal institutions encompass the normative and cognitive pillars and include norms, culture, and ethics. Moreover, culture is at the core of informal institutions ).
Formal institutions refer to "the rules of the game in a society or, more formally, are the humanly devised constraints that shape human interaction. They structure incentives in human exchange, whether political, social, or economic" (North 1990, p. 3). A hierarchy exists within the different levels of formality in the institutional framework, such that formal institutions can be "enacted to modify, revise, or replace informal constraints" (North 1990, p. 47). Therefore, it is crucial to consider the interactions with formal constraints when investigating the effects of informal institutions.

Cultural Distance and Performance
Studies have shown the relationship between cultural characteristics and economic development (Hofstede et al. 2010;Correa da Cunha et al. 2022a). Moreover, studies have shown that cultural traits can contribute to agency costs (Orlova 2020). Additionally, foreign firms are at a disadvantage (Zaheer 1995) due to the cultural distance between home and host countries (Kogut and Singh 1988). Cultural distance represents the "extent to which the shared norms [ideas, beliefs] and values in one country differ from those in another" (Drogendijk and Slangen 2006, p. 362). Most studies emphasize their negative effects (Beugelsdijk et al. 2018) because the distance represents the liability of foreignness (Zaheer 1995) as it creates friction (Shenkar et al. 2008;Shenkar 2012) and increases the costs of doing business abroad (Cuervo-Cazurra and Genc 2011). We test the following hypothesis to assess the effects of CD on the financial performance of foreign subsidiary firms: H1. CD lowers the financial performance of foreign subsidiary firms in Latin America.

The Moderating Effects of Host Country Formal Institutions
Formal institutions are strong "if they support the voluntary exchange underpinning an effective market mechanism" and weak "if they fail to ensure effective markets or even undermine markets" (Meyer et al. 2009, p. 63). Studies show that more developed formal institutions in the host country provide better conditions for firms to operate as it lowers transaction costs and increases financial performance (Cuervo-Cazurra and Genc 2011;Maseland 2013;Zaheer et al. 2012;Hernández and Nieto 2015;Konara and Shirodkar 2018).
According to North (1990), formal and informal institutions can interact in a complementary or suppressive manner. Institutional complementarity refers to a condition in which "the presence (or efficiency) of one [institution] increases the returns from (or efficiency of) the other" (Hall and Soskice 2001, p. 17). Alternatively, the suppressive relationship refers to a condition in which the formal rules may be "enacted to modify, revise, or replace informal constraints" (North 1990, p. 47). According to Pejovich (1999, p. 170), "formal institutions suppress, but fail to change informal institutions." Furthermore, "formal rules change, but the informal constraints do not" (North 1990, p. 91). Thus, we argue that, due to the higher hierarchy and priority of formal institutions, the strength of formal institutions and governance characteristics in the host country conditions the effects of CD. When governance is stronger, it lowers the effects of CD and vice-versa. To test these assumptions, we propose the following hypothesis: H2. The stronger the governance characteristics in the host country, the lower the negative effects of CD on the financial performance of foreign subsidiaries.

The Moderating Effects of Host Country Trade Openness
Trade openness represents the country's orientation towards foreign trade. It measures the ratio of imports and exports compared to the size of the economy. Reyes et al. (2019) show that trade openness affects inward foreign direct investment in Latin America and, thus, creates incentives for foreign firms to invest in the region. Countries that are more open to foreign trade interact more frequently with other cultures. For instance, Lee and Wen (2020, p. 140) identified that "the government-imposed trade liberalization significantly increases the entry threat from foreign competitors, offering a specific setting in which to capture the dynamic nature of changing competition over time." Coyne and Williamson (2009, p. 4) state that "the more open a country is to the trade, the more likely it is to possess culture conducive to increased social and economic interactions." Thus, we argue that firms operating in countries with different degrees of trade openness face the varying impact of CD. The more frequent and intense interaction with foreign cultures might explain why the effect of CD in more open host countries is lesser than those in closed economies. Hypothesis H3 tests this argument: H3. The higher the degree of trade openness, the lower the negative effects of CD on the financial performance of foreign subsidiaries.

Methodology
This study uses a quantitative approach using the econometric technique of panel data to test the hypotheses. Due to fast-paced and highly volatile conditions in Latin America, panel data can check the consistency in patterns and relationships for the variables included in this study. The data on foreign subsidiaries is from the Orbis database. However, not all firms provide complete data or participate in the survey every year, leading to missing data and unbalanced data panels. The final sample for this study is selected considering the tradeoff between a larger number of individual subsidiary firms over a shorter period and a larger number of individual firms over an extended period. The final sample comprises foreign subsidiaries operating in the ten largest economies in Latin America: Argentina, Brazil, Colombia, Chile, Ecuador, Mexico, Panama, Peru, Uruguay, and Venezuela, over three consecutive years from 2013 to 2015, totaling over 4200 firm-year observations. Due to this vast diversity, Latin America provides a relevant context for this study as "societal, cultural, and economic characteristics that make the region an ideal 'natural laboratory' to build and test management theories" (Aguinis et al. 2020, p. 615). Observing the same individual firms over three consecutive years provides more robust estimates because of the rapidly evolving domestic economic and political conditions in Latin America. Furthermore, bootstrap analysis improves the stability of the parameter estimates given the reduced number of individual subsidiary firms (Efron and Gong 1983). When comparing bootstrap estimates with the original dataset, the significance of the parameter estimates improves in the bootstrap sample. Also, the results in both samples are consistent for all empirical tests.

Dependent Variable
We use profit margins, a widely cited measure in literature, to measure the financial performance of the subsidiaries (Venkatraman and Ramanujam 1986;Hitt et al. 1997;Correa da Cunha et al. 2020). Furthermore, the profit margin is less susceptible to different asset valuations that result from the time of investment or depreciation (Michael Geringer et al. 1989, Contractor et al. 2003. When comparing firms in different industries, profit margin provides a more equitable alternative to measure firm performance as firms in various sectors use assets differently. In turbulent contexts, such as in emerging markets, sustaining the company's profit margins become even more challenging and reflects management's effectiveness at investing in projects that add value (Chopra and Mier 2017). The profit margin data is obtained from the Obis database.

Control Variables
Several controls known to affect the financial performance of foreign subsidiary firms are included at the subsidiary, home, and host country-level: size of the economy, trade openness, country-level governance, industry sector (i.e., industrial vs. service firms), industry sector annual growth, subsidiary annual sales growth, subsidiary size, and subsidiary market share (Capon et al. 1990;Dikova 2009;Cuervo-Cazurra and Genc 2011;Hernández and Nieto 2015;Konara and Shirodkar 2018). Figure 1 illustrates the general framework of the study. Int. J. Financial Stud. 2022, 10, x FOR PEER REVIEW 6 of 16 subsidiary market share (Capon et al. 1990;Dikova 2009;Cuervo-Cazurra and Genc 2011;Hernández and Nieto 2015;Konara and Shirodkar 2018). Figure 1 illustrates the general framework of the study.

The Moderation Tests
The moderation tests are performed by analyzing how changes in the moderator variable affect the coefficients. We follow the procedure described by Hayes (2013) to test hypotheses H2 and H3-it involves evaluating the change in the slope of the regression line that results from adding and subtracting one standard deviation from the moderator variable.

Results
We use the random effects (REs) estimation method to "control for unobserved heterogeneity because some of the independent variables are time-invariant" (Salomon and Wu (2012, p. 359). Therefore, the REs allows verifying the effects of CD, which remain constant over time. Additionally, White's test (Andrews 1991;MacKinnon and White 1985;White 1980) indicates no heteroskedasticity issues. Moreover, Lu and White (2014, p. 178) indicate that "a now common exercise in empirical studies is a 'robustness check,' where the researcher examines how certain 'core' regression coefficient estimates behave when the regression specification is modified in some way, typically by adding or removing regressors." The consistency of the results is verified as the size and signal for the coefficients and the characteristics of the models remain unchanged when new variables are introduced during the moderation tests. Values lower than 2 for the variance inflation factor (VIF) test for all variables show no multicollinearity issues (Rogerson 2001). Table 1 shows the test results for hypotheses H1 and H2.

The Moderation Tests
The moderation tests are performed by analyzing how changes in the moderator variable affect the coefficients. We follow the procedure described by Hayes (2013) to test hypotheses H2 and H3-it involves evaluating the change in the slope of the regression line that results from adding and subtracting one standard deviation from the moderator variable.

Results
We use the random effects (REs) estimation method to "control for unobserved heterogeneity because some of the independent variables are time-invariant" (Salomon and Wu (2012, p. 359). Therefore, the REs allows verifying the effects of CD, which remain constant over time. Additionally, White's test (Andrews 1991;MacKinnon and White 1985;White 1980) indicates no heteroskedasticity issues. Moreover, Lu and White (2014, p. 178) indicate that "a now common exercise in empirical studies is a 'robustness check,' where the researcher examines how certain 'core' regression coefficient estimates behave when the regression specification is modified in some way, typically by adding or removing regressors." The consistency of the results is verified as the size and signal for the coefficients and the characteristics of the models remain unchanged when new variables are introduced during the moderation tests. Values lower than 2 for the variance inflation factor (VIF) test for all variables show no multicollinearity issues (Rogerson 2001). Table 1 shows the test results for hypotheses H1 and H2.
The negative and statistically significant effect of cultural distance (CD_KS) for all empirical arrangements show that the cultural distance negatively affects the financial performance of firms in Latin America (Table 1), supporting hypothesis H1. Furthermore, results reveal that host country governance has a positive and significant direct effect on the financial performance of foreign subsidiary firms. Moreover, the results support hypothesis H2 as the strength of governance in the host country negatively moderates the effects of CD. The decrease in the coefficient of the main relation (between CD_KS and financial performance) verifies the negative moderation. It changes from statistically significant −1.033 (p-value < 0.01) at low governance levels (−1 std. deviation) to −0.806 (p-value < 0.01) at moderate levels and the effects become even lower (−0.578 and p-value < 0.05) when host country governance becomes stronger (+1 std. deviation). Figure 2 shows the change in the slope of the regression line that represents the effects of CD on the profit margin of foreign subsidiaries firms when the quality of governance in the host country improves (solid line, +1 std. deviation) and worsens (dotted line, −1 std. deviation).  Table 2 displays the changes in the average of the six WGI variables in the countries included in this study, ranging from 1996 (the first year the indicators were available) to 2020 in intervals of five years (when available). In countries such as Argentina, Brazil, Chile, Ecuador, Mexico, Panama and Venezuela, the deterioration in the quality of formal institutions is associated with an increase in the negative effects of CD experienced by foreign subsidiary firms operating in the region. On the other hand, a few exceptions, such as Colombia, Peru, and Uruguay, show some improvements in the quality of formal institutions. In these three countries, the effects of CD in 2020 are expected to be lower compared to 1996. financial performance) verifies the negative moderation. It changes from statistically significant −1.033 (p-value < 0.01) at low governance levels (−1 std. deviation) to −0.806 (pvalue < 0.01) at moderate levels and the effects become even lower (−0.578 and p-value < 0.05) when host country governance becomes stronger (+1 std. deviation). Figure 2 shows the change in the slope of the regression line that represents the effects of CD on the profit margin of foreign subsidiaries firms when the quality of governance in the host country improves (solid line, +1 std. deviation) and worsens (dotted line, −1 std. deviation).  Table 2 displays the changes in the average of the six WGI variables in the countries included in this study, ranging from 1996 (the first year the indicators were available) to 2020 in intervals of five years (when available). In countries such as Argentina, Brazil, Chile, Ecuador, Mexico, Panama and Venezuela, the deterioration in the quality of formal   Table 3 presents the test results for hypothesis H3, which relates to the moderating effects of trade openness on the relationship between CD and the financial performance of foreign subsidiary firms.  Concerning hypothesis H3, the test results in Table 3 reveal that contrary to expectations, trade openness positively moderates the effects of CD on the financial performance of foreign subsidiaries in Latin America. This can be verified by the increase in the coefficient of the cultural distance (CD_KS) variable, which changes from −1.738 (p-value < 0.01) at low degrees of trade openness (−1 std. deviation) to −2.173 (p-value < 0.01) at intermediate values and −2.609 (p-value < 0.01) at high degrees of trade openness (+1 std. deviation). Figure 3 displays the positive moderating effects of trade openness on the relationship between CD and the financial performance of foreign subsidiary firms in Latin America. The change in the slope of the regression shows that the negative effects of CD on the financial performance of foreign subsidiaries increase when trade openness in the host country increase by 1 std. deviation (solid line) and decreases as the host country becomes less open to foreign trade (dotted line, −1 std. deviation).
at intermediate values and −2.609 (p-value < 0.01) at high degrees of trade openness (+1 std. deviation). Figure 3 displays the positive moderating effects of trade openness on the relationship between CD and the financial performance of foreign subsidiary firms in Latin America. The change in the slope of the regression shows that the negative effects of CD on the financial performance of foreign subsidiaries increase when trade openness in the host country increase by 1 std. deviation (solid line) and decreases as the host country becomes less open to foreign trade (dotted line, −1 std. deviation). These results provide a more nuanced view of the conditions under which the effects of CD can be higher or lower. By looking at the moderating effects of trade openness, these findings complement the literature regarding the asymmetric effects of CD. Magnani et al. (2018) show that exports from Brazil, a relatively closed economy with a trade openness index (exports + imports as % of GDP) lower than 30%, face more significant challenges associated with CD when doing business in Italy (a much more open economy with a trade openness index greater than 60%) than Italians adjusting to the Brazilin culture. Similarly, Selmer et al. (2007) verified that Germans find it easier to adapt to the local culture in the U.S. (trade openness index of approximately 25%), in contrast to American expatriates in Germany, which is a much more open economy to international trade with a trade openness index of over 90%. The moderating effects show that the different and asymmetric effects of CD experienced by the different groups of expatriates might result from the degree of trade openness in the host country. Table 4 shows the changes in the trade openness index in the countries over forty years, starting in 1980 in intervals of five years. Trade openness increases in most countries in the region-the average trade openness index for Latin America and the Caribbean in 2020 is 46.6, 50% higher than 30.3 in 1980. Therefore, according to our findings, as countries in the region became more open to foreign trade over the past 4 decades, the negative effects of CD experienced by foreign subsidiary firms operating in the region might have increased during the period.  These results provide a more nuanced view of the conditions under which the effects of CD can be higher or lower. By looking at the moderating effects of trade openness, these findings complement the literature regarding the asymmetric effects of CD. Magnani et al. (2018) show that exports from Brazil, a relatively closed economy with a trade openness index (exports + imports as % of GDP) lower than 30%, face more significant challenges associated with CD when doing business in Italy (a much more open economy with a trade openness index greater than 60%) than Italians adjusting to the Brazilin culture. Similarly, Selmer et al. (2007) verified that Germans find it easier to adapt to the local culture in the U.S. (trade openness index of approximately 25%), in contrast to American expatriates in Germany, which is a much more open economy to international trade with a trade openness index of over 90%. The moderating effects show that the different and asymmetric effects of CD experienced by the different groups of expatriates might result from the degree of trade openness in the host country. Table 4 shows the changes in the trade openness index in the countries over forty years, starting in 1980 in intervals of five years. Trade openness increases in most countries in the region-the average trade openness index for Latin America and the Caribbean in 2020 is 46.6, 50% higher than 30.3 in 1980. Therefore, according to our findings, as countries in the region became more open to foreign trade over the past 4 decades, the negative effects of CD experienced by foreign subsidiary firms operating in the region might have increased during the period.

Discussion
We contribute to the knowledge of cultural distance by showing that its effects can be more or less significant depending on the characteristics of the host country. In line with previous research (Beugelsdijk et al. 2018), our findings reveal that CD represents a liability of foreignness (Zaheer 1995) to foreign subsidiary firms operating in Latin America as it affects the financial performance of these firms negatively. Furthermore, studies show that the effects of CD might be asymmetric (Shenkar 2001;Selmer et al. 2007;Zaheer et al. 2012;Magnani et al. 2018) as the same distance can have significantly different effects depending on the standpoint of the observer. This study contributes to this debate by providing a more nuanced view of the conditions under which the impact of CD can become higher or lower.
First, by analyzing how formal and informal institutions interact (North 1990;Pejovich 1999;Hall and Soskice 2001;Correa da Cunha 2019), results reveal that the effects of CD are suppressed by the strength of formal institutions in the host country. In that sense, findings show that formal institutions prioritize more than informal ones (e.g., culture). Therefore, the adverse effects of CD tend to be lower in host countries with more supportive formal institutions and stronger governance mechanisms. These results can help foreign firms understand what can be expected regarding the effects of CD by considering the characteristics of the host country and the tradeoff between entering a more culturally distant host country versus a more developed country in terms of formal institutions. Moreover, countries can become more competitive in attracting foreign investment when they provide more supportive governance structures that reduce the negative effects of cultural differences, which increases the profitability of foreign subsidiary firms.
Second, contrary to expectation, although trade openness in the host country positively affects the financial performance of foreign subsidiary firms, it positively moderates (i.e., increases) the adverse effects of CD. Thus, a higher trade openness index of a host country leads to an increase in the negative effect of CD on the financial performance of foreign subsidiary firms. These results suggest that the more open the economy to trade (i.e., imports and exports), the higher the pressure for foreign subsidiary firms to conform to the demands of the local culture, i.e., the more open the economy becomes to foreign trade, the higher the competition and the bargaining power of customers. There are more options to replace foreign subsidiary firms that fail to accommodate the needs and requirements of the local culture. Thus, countries that are more open to foreign trade might have more frequent and sometimes more diversified experiences doing business with foreign cultures. However, it does not necessarily translate to a more tolerant environment when accommodating the cultural differences between the foreign subsidiary firm and the local environment. Host countries that are more open to foreign trade have higher competition from foreign firms (Lee and Wen 2020). Therefore, foreign subsidiary firms will face higher demands from local stakeholders to conform to the expectations of the local culture, which explains the significantly higher negative effects of CD.
This study has several relevant implications for policymakers and practitioners. These insights can assist governments to improve the attractiveness of countries in the region by implementing more supportive formal institutions. It has been shown that strong formal institutions offer positive and significant business advantages, but they can also moderate the adverse effects of CD. Thus, by improving the governance in countries in Latin America, policy formulation in this area can improve the conditions for foreign subsidiary firms to operate in the region. Furthermore, as countries in Latin America become more open to foreign trade, foreign subsidiary firms must increase their awareness regarding the adverse effects of the CD. As a result, foreign subsidiaries operating in the region can identify alternatives to adjust more favorably to the local culture.

Conclusions
This study shows that the effects of CD are highly dependent on other contextual characteristics of the host country. Investigating how formal institutions and trade openness in the host country moderates the impact of CD provides a more nuanced view of the conditions under which the effects can be higher or lower. It can be concluded that although CD between countries remains relatively stable over time, changes in formal institutions and policies that affect foreign trade can modify the way CD affects the financial performance of foreign subsidiary firms in a much shorter period. Being aware of such implications can help companies decide how and where to invest in increasing overall firm performance. Firms can evaluate the tradeoff between entering a host country that is more distant in terms of CD but has more developed formal institutions than a similar country in terms of national culture with less supportive formal institutions. Countries can develop better governance and policies that provide better conditions for foreign subsidiary firms to operate.
This study has limitations, which provide great opportunities for future research. Future research could strengthen our findings and contributions by adopting a more nuanced approach, such as the one proposed by Ionascu et al. (2004) to investigate how contextual characteristics of the host country affect the relationship between cognitive, normative, and regulatory distances. Moreover, they can focus on different contexts and specific measures of CD to account not only for the size but also for the direction of the distance. This can help advance our understanding of how different contextual elements in the host country affect the financial performance of foreign subsidiary firms.