Cultural differences and cross-border investment project performance: an analysis of the Polish banking sector

Abstract In recent decades, the Polish banking sector has experienced a large number of mergers and acquisitions (M&As) as well as the establishment of several new banks. The success of such investment projects can be influenced by numerous factors, including the cultural differences between the country of the bank initiating the transaction and Poland. The objective of this article is to assess the influence of these cultural differences on the performance of cross-border investment projects carried out in the Polish banking sector from 1994 to 2018. The results of this study confirm that cultural differences influence bank performance: the culturally closer the countries are, the better the banks perform. Specifically, the dimensions of power distance, individualism, uncertainty avoidance and masculinity are the most relevant to bank performance. Future research should be focused on determining the role that cultural differences exert on the cross-border consolidation of the European banking sector.


Introduction
According to the literature (Kolaric & Schiereck, 2014;Ferretti et al., 2010;among others), investment projectsmergers and acquisitions (M&As)mostly benefit the shareholders of the acquired banks, and the effect on the shareholders of the bank initiating or conducting the transaction is neutral at best. Despite this evidence, in the last twenty years, there has been an unprecedented increase in such operations. Although cross-border investment projects are not the main objective of banking operations, they help achieve more rapid development and a stronger competitive position in the market. For banking institutions, such projects are a path to achieving objectives, and they provide an important alternative for organic growth. Additionally, the skilful use of such projects can be a decisive factor in increasing strategic shareholder value. However, the success of such projects is influenced by several factors, including the cultural differences between the country of the bank initiating the transaction and the country in which the investment is made (Petrenko & Stolyarov, 2019).
Although some research has studied the influence of cultural differences on the success of these investment projects, the results are not definitive. Datta and Puia (1995) suggest that acquisitions characterized by a high cultural distance are accompanied by lower wealth effects on acquiring firm shareholders, whereas Morosini et al. (1998) find a positive association between national cultural distance and crossborder performance. More recently, Reus and Lamont (2009) conclude that cultural distance has a double impact.
The aim of this paper is to assess the influence of cultural differences on the performance of cross-border investment projects, including M&As and the founding of new banks. For this purpose, this paper studies the relationship between cultural distance and performance of all investment projects carried out in the Polish banking sector in the 1994-2010 period. Furthermore, this study analyses the influences of each cultural dimension on bank performance. To analyse the effects of such projects in the long term, complete historical data since the entrance of foreign capital up to 2018 are considered; hence, the sample constitutes an unbalanced panel covering a 25-year period.
The Polish banking sector is an especially relevant focus for this study because of the large number of M&As that have occurred and the large number of new banks that have been established in the country. In this respect, at the end of 2019, foreign investors controlled over 17 commercial banks and all foreign bank branches. Controlling interests were owned by investors from 18 countries, with a major role played by investors from Spain, Germany, France and Holland (KNF-Polish Financial Supervision Authority, 2020).
The privatization of the banking sector and the principles of foreign capital participation in the takeovers of national commercial banks have been some of the most controversial and politically charged issues in the Polish banking sector. In 2004, Poland had the largest banking sector in the European Union (Korzeb & Samaniego-Medina, 2019). The main issue discussed in connection with Poland's accession to the European Union was concern about a further increase in the share of foreign shareholders in the structure of the banking sector. In fact, it was not until June 2017 that the ownership structure of the sector changed because of the takeover of Bank Pekao SA by Polish capital. At that point, Polish investors' share of the assets of the sector outpaced that of foreign investors for the first time since 1999.
To achieve the aim of this study, this paper uses fixed effects models with panel data and robust standard errors (Wooldridge, 2010), which control for unobservable time-invariant individual-specific characteristics. The v2 test statistics suggest that a unit-specific error component exists for each regression equation, and the Hausman test verifies the utilization of the individual fixed effects.
The results confirm that cultural differences influence bank performance. The culturally closer the countries are, the better the banks perform. Specifically, bank performance is significantly influenced by the dimensions of power distance, individualism, uncertainty avoidance, and masculinity.
This article contributes to the existing literature by providing a more extensive view of the relationship between shareholder performance and differences in national culture. The findings of this study can be applied to the sector as a tool for supporting decision making in cases of new foreign investment and for evaluating existing transactions.
The remainder of this article is structured as follows. Section 2 reviews the most significant literature and establishes the hypotheses. Section 3 describes the data and methodology employed in the empirical research. Section 4 presents and discusses the results obtained. Section 5 summarizes and presents the main conclusions reached.

Research on cultural diversity
Intercultural management did not emerge as a scientific discipline until the mid-20th century. The pioneering work in this area of research was performed by ethnologist Hall (1960) and Hall and Hall (1959), who described the cultural differences between countries. However, the seminal work on national cultural differences in international business was authored by Hofstede (1980). Hofstede defined culture as 'the collective programming of the mind which distinguishes the member of one group or category of people from another'. His research on the IBM corporation led him to identify four dimensions of national culture. i) The first is the power distance dimension, which indicates the extent to which the less powerful members of institutions and organizations within a country expect and accept that power is distributed unequally.
ii) The second dimension is the individualism versus collectivism factor, where individualism applies to societies in which the ties between individuals are loose and everyone is expected to look after him-or herself and his or her immediate family, while collectivism, as the opposite of individualism, refers to societies in which people from birth onward are integrated into strong, cohesive in-groups that continue to protect them throughout their lifetime in exchange for unquestioning loyalty. iii) Uncertainty avoidance is the third dimension. It indicates the extent to which the members of a culture feel threatened by ambiguous or unknown situations. iv) The fourth dimension is the masculinity versus femininity factor, where masculinity is defined as a societal preference for achievement, heroism, assertiveness and material rewards for success, while femininity represents a preference for cooperation, modesty, caring for the weak and quality of life. Furthermore, Hofstede (1980) emphasized that the characteristics of national cultures are shaped over long periods of time in the context of historical, geographical and economic conditions.
In more recent research, Hofstede has identified fifth (Hofstede & Bond, 1988) and sixth dimensions (Hofstede et al., 2010): v) long-term versus short-term orientation, which is related to the choice of focus for people's efforts, the future or the present and the past; and vi) indulgence versus restraint, which is related to gratification versus control of basic human desires connected to the enjoyment of life.
Following this line of research, Trompenaars and Hampden-Turner (1997) indicate that every culture distinguishes itself from others by the specific solutions that it chooses for certain problems that reveal themselves as dilemmas. It is convenient to look at these problems under three headings: i) those that arise from relationships with other people; ii) those that come from the passage of time; and iii) those that relate to the environment. In addition, House et al. (2004) surveyed 17,300 middle managers from 61 countries to develop the GLOBE project. GLOBE empirically establishes nine cultural dimensions that make it possible to capture the similarities and/or differences in norms, values, beliefs, and practices among societies (power distance, uncertainty avoidance, humane orientation, collectivism I (institutional), collectivism II (in-group), assertiveness, gender egalitarianism, future orientation, and performance orientation).
Finally, Schwartz (2006Schwartz ( , 1994Schwartz ( , 1992 introduces a new concept in relation to the cultural dimension. According to this author, human values are structured as a universal, motivational and circular continuum. Schwartz's main thesis concerns the structure (the circular continuum) and the contents of values (their motivational contents). Schwartz (1992) and Schwartz et al. (2012) characterize this continuum of values as circular, which means that the principle of the similarity of neighbouring values is supplemented by the principle of the opposition of the values located on the opposite sides of the circle.

Research on the effect of cultural differences on M&as
Numerous studies have attempted to explain the relationship between cultural distance and M&A performance. In this respect, several studies have found a negative relationship between the two concepts. Datta and Puia (1995) suggest that cross-border acquisitions, on average, do not create value for acquiring firm shareholders. The authors emphasize that acquisitions characterized by high cultural distance are accompanied by lower wealth effects for acquiring firm shareholders. The negative relationship between cultural distance and acquirer performance is caused by the high degree of integration challenges in distant cultures, which may result in high potential costs. Similar results are obtained by Beugelsdijk et al. (2018), who find that firms are less likely to expand to culturally distant locations. According to these authors, cultural distance has a strong negative effect on subsidiary performance but no effect on the performance of the whole multinational company. In addition, the authors find that the effects of cultural distance are not sensitive to time but are sensitive to the cultural framework used and the home country of the company.
However, other studies have obtained opposite results. For instance, Morosini et al. (1998) find a positive association between national cultural distance and crossborder performance using a multidimensional measure of national cultural distance. In the authors' opinion, the cross-border acquisitions that tend to perform better are those in which the routines and repertoires of the target's country of origin are, on average, more distant than those of the acquirer's country. Similarly, Chakrabarti et al. (2009) argue that cultural distance produces positive effects through its potential to present acquirers with access to cultures and practices with which they are not familiar. The authors find that cross-border acquisitions perform better in the long run if the acquirer and the target come from countries that are more culturally disparate.
Finally, there are studies that provide mixed findings. Reus and Lamont (2009) conclude that cultural distance has two-sided impacts. These authors indicate that national cultural distance impedes clarity with regard to the key capabilities that need to be transferred and that it constrains communication between acquirers and their acquired units, thus having a negative indirect effect on acquisition performance. In contrast, Sarala (2010) suggests that organizational cultural differences and organizational cultural preservation increase conflict and that partner attractiveness decreases conflict, while national cultural differences have no influence on the level of conflict. Ahammad et al. (2016) indicate that national cultural distance shows no significant effect on knowledge transfer or cross-border acquisition success; however, organizational cultural differences have a significant impact on knowledge transfer and a strong effect on cross-border acquisition success. The authors' results support the view of Weber et al. (2011Weber et al. ( , 2009): national cultural distance and organizational cultural differences are dissimilar constructs that differently affect M&A success.
The aim of this study is to assess the influence of cultural differences on the performance of cross-border investment projects. As the nature of these relationships is not clear because previous studies present mixed findings, two opposing hypotheses are formulated: Hypothesis 1a: The culturally closer the countries are, the better the banks perform.
Hypothesis 1b: The culturally further apart the countries are, the better the banks perform.

Research on the effect of cultural dimensions on the banking sector
Managing cultural differences across geographically dispersed locations is one of the central challenges for international research and practice in the banking sector. Previous studies (Boubakri et al., 2017;Bussoli, 2017;Ashraf et al., 2016;Kanagaretnam et al., 2011Kanagaretnam et al., , 2014 find that uncertainty avoidance and individualism are the cultural dimensions that affect risk-taking banks. In this respect, Kanagaretnam et al. (2011) find, for a sample of international banks, that uncertainty avoidance is negatively related to bank risk taking, whereas individualism is positively associated with it. Bussoli (2017) confirms these findings using a sample of European banks. Ashraf et al. (2016) obtain a similar result; however, they find that the power distance dimension is likewise relevant to determine bank risk taking: the lower the power distance is, the higher the risk taking.
These cultural differences become more relevant in times of crisis. In this regard, Kanagaretnam et al. (2014) suggest that banks with low individualism and high uncertainty avoidance are less likely to fail or experience financial trouble during a crisis period. These findings are confirmed by Boubakri et al. (2017).
Finally, in a line of research similar to that of this study, Ashraf and Arshad (2017) find that the national culture of a parent bank's home country has a greater impact on the risk-taking behaviour of the foreign affiliates of multinational banks than the national culture of their host country. Specifically, foreign affiliates engage in more risk taking if the parent banks' home country's cultural values include a low power distance, low uncertainty avoidance and high individualism.
Accordingly, the empirical evidence shows that the dimensions of power distance, individualism and uncertainty avoidance have a greater influence on bank risk taking. Therefore, these dimensions are considered to establish the following hypotheses in relation to bank performance: Hypothesis 2: There is a relationship between the power distance dimension of the country of the main shareholder and Polish bank performance.
Hypothesis 3: There is a relationship between the individualism versus collectivism dimension of the country of the main shareholder and Polish bank performance.
Hypothesis 4: There is a relationship between the uncertainty avoidance dimension of the country of the main shareholder and Polish bank performance.
However, with respect to the other dimensions identified by Hofstede et al. (2010), there is no empirical evidence of their relevance to determining bank risk taking. Accordingly, the following hypotheses are formulated: Hypothesis 5: There is no relationship between the masculinity versus femininity dimension of the country of the main shareholder and Polish bank performance.
Hypothesis 6: There is no relationship between the long-term versus short-term orientation dimension of the country of the main shareholder and Polish bank performance.
Hypothesis 7: There is no relationship between the indulgence versus restraint dimension of the country of the main shareholder and Polish bank performance.

Sample
The sample considers all the investment projects that took place in the Polish banking sector during the 1994-2010 period, including 83 projects, 55 M&As and 28 cases of the creation of new banks. These operations were conducted by 28 financial entities from 17 countries, comprising a sample of 780 observations. Table 1 shows the number of investment projects analysed by investor and by country.
To analyse the effects of the projects in the long term, the sample considers complete historical data since the entrance of foreign capital up to 2018, constituting an unbalanced panel covering a 25-year period . The data were mainly obtained from bank reports (annual financial reports, corporate financial statements, the National Court Register, publications in the 'Rzeczpospolita' daily and 'Gazeta Prawna', and announcements by the Polish Press Agency). Financial and accounting information on these firms was obtained from the Thomson Reuters Eikon database. Table 2 shows the total number of observations by country and by period.

Variables
This study selects variables to measure bank performance, cultural dimensions, cultural distance, and bank characteristics. As a dependent variable, a well-established measure of a company's profitability is used: return on equity (ROE) (Molyneux & Thornton, 1992;Bourke, 1989). It is calculated as the company's net income to shareholder equity and determines a company's capacity to generate value for its investors in comparison to its cost of capital. With respect to the cultural dimension variables, following Hofstede (1980), Hofstede and Bond (1988) and Hofstede et al. (2010), six dimensions of national culture are used to analyse the role of cultural differences: power distance (PDI), individualism vs. collectivism (IDV), uncertainty avoidance (UAI), masculinity vs. femininity (MAS), long-term orientation vs. short-term orientation (LTO), and indulgence vs. restraint (IND) 1 . Hofstede's dimensions are a widely used tool for measuring cultural differences (Breuer et al., 2018;Kirkman et al., 2006), and they are especially relevant in the banking sector (Ahunov & Van Hove, 2020;Yunanda et al., 2019;Zhang et al., 2018;Ashraf et al., 2016;Baptista & Oliveira, 2015;Carretta et al., 2015).
PDI expresses the inequality tolerance of a society regarding power distribution. A high value in this dimension means that individuals accept their role in society regardless of whether it is justified. IDV refers to the role that people play in groups and how they integrate into society; a high value indicates unstable or unstructured societies, while a low value means the opposite: stable societies. UAI expresses the degree to which the members of a society feel uncomfortable when they are in situations of uncertainty. A high value expresses intolerance regarding beliefs and greater comfort with regulatory limits; a low value indicates open-minded societies in which people easily trust one another. Regarding MAS, a high value indicates a more aggressive and competitive society than a feminine society (low value), where cooperation and consensus are the most important values. With respect to LTO, a high LTO represents societies with a pragmatic focus that values education and saving; in contrast, low-LTO societies prioritize immediate results and adopt a normative focus. Finally, IND indicates the degrees of tolerance and liberty. A high score means indulgence regarding human impulses and freedom of expression, while a low value expresses more restrictive cultures.
To measure the cultural distance in Hofstede's dimensions between the country of the main shareholder and Polish banks, this study calculates two measures: the cultural distance index (CDI) and Pearson's correlation (PEARSON).  Country 1994Country -1998Country 1999Country -2003Country 2004Country -2008Country 2009Country -2013Country 2014Country -2018 Total   Austria  15  10  5  5  5  40  Belgium  0  10  10  8  1  29  Belgium-Fr.  0  0  4  3  0  7  Denmark  1  7  5  0  0  1 3  France  2  11  28  16  3  60  Germany  30  59  43  16  5  153  International shareholders  31  7  3  14  8  63  Ireland  3  7  5  5  5  2 5  Italy  2  10  13  10  8  43  Japan  0  7  10  5  0  22  Netherlands  12  15  8  5  5  Following Kogut and Singh (1988), this study calculates the CDI between Poland and the j th country based on the following equation: (1) where I ij is the index for the i th cultural dimension and j th country, I iPL denotes Poland's score on the i th cultural dimension, and V i represents the variance in the index of the i th dimension. The CDI denotes the composite of Kogut and Singh's cultural distance index (Kogut & Singh, 1998) based on the deviation from the Polish ranking along each of Hofstede's cultural value dimensions for each country. The lowest (highest) CDI values indicate the smallest (greatest) cultural distance between the country of the acquiring bank and Poland.
Although the first natural measure of differences between countries is the CDI (Kogut & Singh, 1998), Pearson's correlation is used as an alternative method to determine the interdependence of the cultural dimensions between both countries (Abraham, 2019;Buszko, 2018;Bernardi, 2006). The closer the correlation is to 1, the culturally closer the countries are.
Finally, this study uses 5 accounting variables as bank characteristics for i) capital structure, ii) credit risk, iii) debt quality, iv) long-term solvency and v) firm size. To analyse the effect of capital structure on companies' performance, the debt-to-equity (DE) ratio is used. This measure is a leverage ratio that compares a company's total liabilities with its total shareholders' equity (Ghosh & Moon, 2010;Hurdle, 1974). It is widely used in the literature regarding the determinants of banks' profitability (Trujillo-Ponce et al., 2013). With respect to credit risk, this study selects the Z score measure (ZSCORE), which describes a bank's capital level in relation to volatility in its return on assets (ROA). This accounting-based variable explains the capability of a bank to absorb variability in its ROA (Li et al., 2017;Khan et al., 2013), and it is generally used as a proxy of individual risk for the banking sector (Khan et al., 2017;Baselga-Pascual et al., 2015;Chiaramonte et al., 2015;Laeven & Levine, 2009; among others). The Z score is defined as follows: where ROA is return on assets. Concisely, the Z score measures the distance to default in terms of the number of standard deviations of the ROA (using a three-year rolling window). High Z score values signal a lower probability of default (Shim, 2017), and vice versa.
In relation to debt quality, the ratio of total deposits to total debt (T_DEPOSD) is considered (Valipour & Moradbeygi, 2011;Garc ıa-Teruel & Mart ınez-Solano, 2007). Total deposits represent the sum of non-interest-bearing deposits, interest-bearing deposits and other deposits (amounts due to financial institutions, customers and the public sector). Long-term solvency is measured through the debt-to-total assets ratio (DTA), that is, the book value of total debt divided by the book value of total assets (Valipour & Moradbeygi, 2011;Ghosh & Moon, 2010;Cai et al., 2008).
Finally, the Neperian logarithm of bank total assets (Ln_TA) is calculated to linearize the effect of size on firm performance (di Pietro et al., 2019; Gir alez-Puig & Berenguer, 2018; Gerhardt & Vennet, 2017). Table 3 presents the variables and their definitions.

Methodology
This paper estimates two models to study both the relationship between cultural distance and bank performance and the influence of the individual cultural dimensions on bank performance. Model 1 measures the effect of the CDI and cultural dimension variables on ROE. Therefore, the model is as follows: where ROE is the measure of banks' performance and CDI is the cultural distance index. PDI, IDV, UAI, MAS, LTIO and IND correspond to each of the measures of the cultural dimensions. X it includes bank characteristics (capital structure, credit risk, debt quality, long-term solvency and firm size) lagged by one year to lessen the endogeneity issues potentially arising from simultaneity concerns. The subscript i represents firms (i ¼ 1, … , n), and the subscript t represents time in years (t ¼ 1, … , t). yeardummy is a set of year dummy variables; l i is the firm's effect; and e it is the error term.
The second model reinforces the results of the first model, measuring the effect of Pearson's correlation and the cultural dimension variables on ROE. Thus, model 2 is as follows: where Pearson is Pearson's correlation and the other variables are the same as those in equation (3).
With the aim of testing the relevance of the differences between every individual dimension for the main shareholder and the performance of Polish banks, this study employs the distances between the cultural dimensions of both countries, in absolute differences, as variables in models 3 to 8. Thus, the models are as follows: where Cultural i represents every cultural dimension value in absolute differences from Poland and the rest of the variables are the same as those in equation (3). Negative values of the coefficients indicate that the shorter the distances between the home country and Poland are, the better the project performs, and vice versa. Individual effects that commonly arise in panel data models are considered fixed effects after applying the Hausman test. To avoid potential heteroskedasticity, this study uses robust standard errors following Wooldridge (Wooldridge, 2010). Poland has high scores on the power distance (PDI) and uncertainty avoidance (UAI) dimensions. With respect to PDI, along with France, Poland presents the highest score (68) out of all the countries in the sample, showing that individuals within enterprises expect the establishment of an organizational hierarchy based on relationships with each employee. Regarding UAI, Poland has one of the highest scores (93) in this dimension, just below Portugal (99) and Belgium (94), revealing a relatively high preference among Poles for avoiding uncertainty. Poles are generally very busy with their duties; however, they are willing to accept performing difficult and time-consuming tasks in return for job security. At the same time, they like transparent situations, truthfulness, and clearly established rules and regulations of conduct 2 . Figure 1 displays Poland's values in comparison to the average values of the investor countries. Table 5 shows the cultural differences between Poland and the country of the acquiring bank, specifically the CDI and Pearson's correlation variables. Mediterranean countries, which include shareholders from Spain, Italy and Portugal,  have the strongest relationship between the cultural dimensions of the country of the main shareholder and Poland, with CDIs of 0.4353, 0.5744 and 0.8778, respectively. Likewise, the culture of French and French-speaking Belgian banking corporations is close to that of Poland and is similar to that of Mediterranean countries (CDI 0.6492 and 0.9322). These results contradict the results obtained by Trompenaars and Hampden-Turner (1997) and Schwartz (2004), who assume that these countries, French and French-speaking parts of Belgium, are more similar to Western European countries. All these countries present a CDI of less than 1. However, Scandinavian countries, such as Sweden and Denmark, present major cultural distances, with CDIs larger than 45. These results are confirmed by Pearson's correlations. Table 6 shows the descriptive statistics of the variables used. ROE presents an average value of 2.065, with a wide range of dispersion (standard deviation 87.19). Regarding capital structure, the mean value shows a high debt ratio with a very wide dispersion; the mean values of the distance to default, debt quality and long-term solvency are 7.77, 84.71 and 83.45, respectively. Finally, the mean value of Ln_TA is explained by the medium-small size of the firms in this sector. Table 7 reports the results of models 1 and 2. The first model studies the effect of the CDI and every cultural dimension on ROE. This paper finds that the CDI is negative  and significant (b¼-53.469, p < 0.01), meaning that the lower the cultural distance between the two countries is, the better the performance is. These results confirm Hypothesis 1a and are in line with Beugelsdijk et al. (2018) and Datta and Puia (1995), indicating that cultural distance could impose integration costs on M&As or direct investments in the Polish bank sector. With respect to the role of cultural dimensions, of the six dimensions analysed, four are significant. The power distance (PDI; b ¼ 5.676, p < 0.01), individualism versus collectivism (IDV; b¼ À4.608, p < 0.01) and uncertainty avoidance (UAI; b ¼ 9.443, p < 0.01) dimensions are relevant, confirming Hypotheses 2, 3 and 4. Surprisingly, the masculinity versus femininity dimension (MAS; b ¼ 1.101, p < 0.01) is also significant; therefore, Hypothesis 5 is not confirmed. The long-term versus short-term (LTO) and indulgence versus restraint (IDV) dimensions are nonsignificant, confirming Hypotheses 6 and 7.

Results of the baseline model
The positive signs of PDI and UAI show that the higher the score of these variables in the home country is, the better the project performs. Therefore, the more acceptable the unequal distribution of power is and the higher the level of regulation is, the better the performance is. Conversely, IDV presents a negative sign, showing a negative influence of a high score on this variable on ROE. Thus, nonindividualistic and tolerant societies generate better results in Polish banks. With respect to MAS, which shows a positive sign, the results confirm that competitive societies obtain better results in terms of ROE.
Concerning the bank-level variables, credit risk (ZSCORE; b ¼ 0.193, p < 0.1) is positive and significant, suggesting that the larger the distance to default is, the better the bank's performance. Additionally, larger banks (Ln_TA; b ¼ 8.355, p < 0.01) show a positive relationship with ROE.
The second model analyses the relationship between Pearson's correlation and bank performance. Pearson's correlation is positive and significant (b ¼ 26.973, p < 0.01), suggesting that the closer the distance between two countries is, the better the bank's performance. This result is similar to that of the first model, reinforcing the confirmation of Hypothesis 1a.
This study develops models 3 to 8 (Table 8) to check the influence of the absolute values of the distances of every cultural dimension between the home country of the investor bank and Poland. Thus, a negative sign means that the closer the distance is, the better the ROE. In contrast, a positive sign indicates the opposite. The betas are consistently negative for every cultural dimension, with significant values for PDI (b¼-0.03, p < 0.05) and UAI (b¼-0.09, p < 0.1). These dimensions are precisely those for which Poland has the highest values in relation to the average (see Figure 1). These results suggest that cultural proximity in each of the dimensions contributes to bank profitability and that this effect is more relevant in the extremes.

Robustness checks
This study conducts a number of robustness checks to confirm the previous findings. First, it tests the relationship between cultural differences and a different proxy of bank performance only for listed firms. In this respect, models 9 and 10 evaluate the effect of cultural distance on Tobin's q (TOBINSQ), which is a market-related variable. Defined as the market value of equity plus the book value of debt over the book value of assets, Tobin's q is a well-established measure of the profitability of a firm   (Bernile et al., 2018Guerrero-Villegas et al., 2018Bebchuk et al., 2009;Bhagat & Bolton, 2008). The models are as follows: where TOBINSQ is a market-related variable for value creation and the rest of the variables are defined as in equation (3). Based on the estimation results reported in Table 9, the estimations do not change the previous findings. There is a positive association between cultural proximity and performance, as measured by Tobin's q.
Second, following Triandis (1995), this study triangulates its findings with a different cultural framework by using the GLOBE dimensions (House et al., 2004). GLOBE identifies nine cultural characteristics, and this study selects five that correspond to Hofstede's dimensions (House et al., 2004;House et al., 2002): power distance (PD), institutional collectivism I (COLL) as the inverse of IDV, uncertainty avoidance (UA) as UAI, assertiveness (ASSERT) as MAS, and future orientation (FO) as LTO. The model studies the relationship between cultural distance, as measured across the GLOBE dimensions, and Polish bank performance as follows: where globe_distance_index represents the CDI and PEARSON calculated with the five GLOBE dimensions and the rest of the variables are that same as those in equation (3).
Models 11 and 12 present the results. The beta coefficients for the CDI (b¼ À0.029, p < 0.05) and PEARSON (b ¼ 0.073, p < 0.05) are significant and maintain the same sign as the first and second models, which is consistent with the main findings of this study. The results of the estimations are reported in Table 10. Finally, this study individually tests every GLOBE dimension in absolute distances. Models 13 to 17 examine these variables as follows: where GCultural i represents every GLOBE cultural value in absolute differences from Poland and the rest of the variables are defined as in equation (3). The estimations show that all the dimensions analysed are significant and have negative signs. These results, which are reported in Table 11, confirm a positive relationship between cultural proximity and bank performance.

Summary and conclusions
Commercial banks operating in Poland are presently among the most valuable assets in the capital groups and investment portfolios of their strategic shareholders. Poland's good economic situation, favourable macroeconomic indicators and  (1) Wald chi 2 is a test for the validity of the joint variables. The estimations include year dummies.
prospects for further development mean that the Polish banking sector continues to be viewed as an attractive investment market and a destination for capital allocation. The purpose of this study is to determine the influence of cultural differences on the performance of cross-border investment projects carried out in Polish banking from 1994 to 2010 considering complete historical data since the entrance of foreign capital up to 2018. This paper uses fixed effects models with robust standard errors.
The results confirm that cultural differences influence bank performance: the culturally closer the countries are, the better the performance. Specifically, Polish bank performance is influenced by the dimensions of power distance, individualism, uncertainty avoidance and masculinity. Finally, different robustness checks confirm these findings.
These results can have various applications. First, they can be used by supervisory institutions to evaluate existing transactions in banking and to draw conclusions on the success or failure of such transactions. Second, they may serve as a tool to support decision making by potential investors in regard to carrying out M&As or establishing new banks, providing additional knowledge about the phenomena occurring during the transaction. This knowledge could constitute one of the determinants of allocating funds for stock exchange investorsboth individuals and institutions. Finally, these results could serve to establish personnel policies, adapting them to the cultural characteristics of the country in which the investment is made.
According to the European Central Bank (2019), cross-border consolidation could support greater risk diversification and contribute to financial market integration. Moreover, it could play an important role in removing excess capacity, enhancing cost efficiency, and promoting more focused and credible business models. Future research should be focused on determining the role that cultural differences play in achieving these objectives, both in Poland and extending to a larger range of European countries.