ESG Performance and Shareholder Value Creation in the Banking Industry: International Differences

: The aim of this paper is to study the role of socially responsible activities on shareholder value creation in a sample of 166 banks from 31 countries over the 2010–2015 period. Prior research about this relationship is scarce and limited to the period before and during the global ﬁnancial crisis. In contrast, this research analyzes banks over a period of time when these institutions have increased their social responsibility practices in order to reinforce their credibility and the trust their stakeholders have in them. More precisely, we analyze the relationship between these two magnitudes distinguishing between environmental, social, and corporate governance actions as well as between countries taking into account the level of development, legal systems, and the geographic area. Our ﬁndings are relevant not only for academics, but also for the managers of these companies, policymakers, investors, and society in general.


Introduction
Banks have always played an active role in the economic and social development of countries by having the ability to select investment projects, manage risks, and decide who has access to capital and what activities are financed. By performing this function, these institutions have a huge impact on society and, consequently, on sustainable development understood as "development that meets the needs of the present without compromising the ability of future generations to meet their own needs" [1].
However, in the summer of 2007, one of the deepest economic crises in modern history began, which was caused by the problems of the financial sector in the United States with the outbreak of the subprime mortgage crisis and later, in the fall of 2008, with the bankruptcy of Lehman Brothers. It is from this bankruptcy that many countries saw how their financial markets were infected by the negative climate of the United States and began to have problems. From that moment, the impact of the banking industry on society, especially the large banks, started to become more important for all stakeholders [2][3][4].
As a result, banks are increasing their social responsibility practices, reinforcing their credibility and the trust that their stakeholders have in them [5]. In this context, nowadays there are more banks that may be considered socially responsible due to the fact that they adopt socially responsible actions. These include the publication of sustainability reports following the Global Reporting Initiative (GRI) guidelines, the adoption of the Equator Principles and the Global Compact, and the inclusion of environmental risk assessments in their credit policies, among other practices. This is due, among other reasons, to the recommendations of the European Union. Specifically, in its 2011 directive, the European Commission specified that companies should integrate social, environmental, and ethical

Literature Review
The question of whether companies should incorporate social responsibility practices into their management strategies designed to meet the expectations of their different stakeholders leads to theoretical positions in favor and against. On the one hand, we have the neoclassical position, defended by Friedman [22], which argues that the responsibility of the company is to use its resources on activities aimed at maximizing its profits, acting in accordance with the basic rules of society incorporated into the law and ethical customs. Therefore, the social function of the company should be limited to the value of the results obtained which should not be less than the resources used. This is achieved when companies maximize their profits or when they maximize the capital value of owners or shareholders. Hence, the conclusion is that companies should maximize that value. When this is achieved, the contribution of the company to society will be optimized according to Friedman [22]. This implies that any other activity that prevents the company from maximizing the value for the shareholder will be considered unacceptable since a misallocation of resources will be incurred.
The neoclassical position, therefore, maintains that the management of the company should only be concerned with the interest of its owners or shareholders. In contrast to this position is the so-called stakeholder theory, developed by Freeman [23], which considers that the company does not just belong to one party (owner or shareholder) but must be understood considering the plurality of agents involved in it and, therefore, that make it possible. In this sense, the objective of the company should not be to maximize value for the shareholder, but instead, the company should create value for all the stakeholders including employees, consumers, local communities, natural or environmental resources, etc.
Following this theory, some authors such as Post, Preston, and Sachs [24] argue that companies should apply those social, environmental, and corporate governance aspects that are necessary, regardless of the costs incurred or the income they produce. On the other hand, other authors such as Porter and Kramer [21] argue that the company's objective must continue to be the maximization of shareholder value, while at the same time it being possible to incorporate social, environmental, and corporate governance measures into management, considering these in terms of the creation of shared value for the company and society. It is therefore important to know whether these measures are profitable for the company, in the sense that they allow shareholder value to be maximized.
However, before performing the empirical analysis, it is necessary to expose the different alternatives that could exist in terms of the relationship between ESG performance and value creation, as well as the hypotheses on which these possible links are based. To that end, we resort to the conceptual framework developed by Preston and O'Bannon [25] and later extended by Gómez [26] to summarize the theoretical foundations of the relationship between social and financial performance.
Specifically, according to Preston and O'Bannon [25], there are two alternative hypotheses to explain the possible impact of ESG activities on the creation of value. On the one hand, a positive result would be explained by the social impact hypothesis that states that when the company is able to effectively manage relationships with its stakeholders, it will obtain a competitive advantage that will allow it to generate value for the company over time [23]. On the other hand, a negative result would be explained by the trade-off hypothesis, which argues that the high costs of social responsibility actions mean a significant reduction in the profits of companies and puts them at a disadvantage compared to others that will prevent them from generating value for the shareholder [22,27,28].
Finally, Gómez [26] adds the hypothesis of moderating variables to explain the neutrality or lack of significance found in the relationship between both variables based on other explanatory variables that could be moderating this relationship. Therefore, Gómez [26] considers that the multivariate analysis is the most appropriate framework to discern between the aforementioned hypotheses. In the context of the literature that we have just mentioned, our objective is to analyze what kind of relationship exists between ESG performance and value creation, also for the case of financial institutions.
Previous empirical evidence for the banking industry presents inconclusive results. Initially, Simpson and Kohers [13] provided evidence of a positive and significant relationship between CSR activities and the performance of the company. However, more recently, Soana [14] concluded that there is no statistically significant link between the two measures of performance for a sample of Italian banks. More recently, Wu and Shen [15] observed, on the basis of a sample of 162 banks from 22 countries covering the period 2003-2009, that CSR is positively associated with financial performance in terms of asset performance, performance of the capital, net interest income, and no financial income. On the contrary, CSR is negatively associated with unproductive loans. Shen et al. [16], on the other hand, analyzed 65 socially responsible banks corresponding to 18 countries during the period 2000-2009. The results reveal that socially responsible banks have significantly higher financial performance than non-socially responsible banks. Meanwhile, Cornett et al. [17] analyzed the relationship between CSR and financial performance in US banks during the financial crisis. Their results indicate that larger banks perform significantly more CSR activities than smaller banks. Additionally, Esteban-Sánchez et al. [4] analyzed the effect of different CSR dimensions on the financial performance of 154 banks in 22 countries, before and during the years of financial crisis obtaining mixed results, depending on the dimension analyzed.
As indicated in the introductory section, in this study we take into account the three basic modern dimensions of CSR-environmental, social, and corporate governance. It is necessary to point out the importance of environmental performance for shareholders and other interested parties. Many of the decisions of these interested parties could be based on the environmental aspects of the services offered. Therefore, there is an increasing pressure that forces banks to change their management systems to meet these growing demands. Meanwhile, although it is true that the banking industry cannot generate serious problems of environmental pollution, it is also true that great savings can be achieved in the billing of electricity, water, fuel, and the use of paper, among others, if corrective measures for environmental improvement are adopted [29]. In this sense, it is necessary to highlight the works of Jo et al. [19], Finger et al. [12], and Laguir et al. [18], who strictly analyzed the creation of value through the environmental activities carried out by banks. Specifically, Jo et al. [19] examined whether corporate environmental responsibility plays a role in improving operational performance in the financial services sector of 29 countries. Their results show that the reduction of environmental costs requires at least one to two years before improving financial performance, being the fastest process in developed markets and slowest in the developing markets. Finger et al. [12], who also distinguished between banks from developed and emerging countries, analyzed the effect of the adoption of the Equator Principles on banks' financial performance. Their results show that in developed (emerging) countries, the adoption of the Equator Principles is associated with an increase (decrease) in financing activity and the participation of interest income. They also indicate that these results show that the adoption of the Equator Principles is a strategic decision for banks in emerging countries and a form of greenwashing in developed countries. Meanwhile, Laguir et al. [18] focused on the French banking sector, and document that the environmental and financial performance of banks is mutually reinforcing, suggesting that there is a complex bidirectional relationship between both magnitudes.
Social performance, on the other hand, refers to how the organization treats its employees, the community, and the client, through responsibility in their products and services. In this sense, the works of Simpson and Kohers [13], Soana [14], and Esteban-Sánchez et al. [4] stand out. Specifically, Simpson and Kohers [13] observed that banks that are more involved with the community in which they operate achieve greater financial performance. On the other hand, Soana [14] observed a positive relationship between the cost/income ratio per employee and the book/market ratio of the Italian banks analyzed. Meanwhile, Esteban-Sánchez et al. [4] observed that banks with better relationships with employees have better financial performance. However, they also note that the social responsibility dimension of the product did not positively influence financial performance.
Finally, with the term corporate governance, we refer to how power is exercised and how decisions are made in a bank that guarantees that members of its board of directors and executives act in the best interest of their long-term shareholders [30]. In addition, corporate governance is an important dimension of CSR that guarantees accountability, compliance, and transparency. Therefore, responsible governance involves reducing the agency's problems with financial stakeholders [31], which is expected to have a direct impact on stock prices. In this aspect, the studies for the banking industry are scarcer. We first highlight the work of Soana [14], that found a positive link between corporate governance and the performance of the assets of the Italian banks analyzed. Secondly, we highlight the work of Esteban-Sánchez et al. [4] that also observed a positive relationship between corporate governance and financial performance for a sample of banks from different countries, before and during the financial crisis.
Definitively, we must point out that, unlike the previous empirical evidence, the aim of this research is to study the effect on shareholder value creation of the ESG performance of a group of banks listed in different stock markets in a period after the financial crisis, in which compliance with CSR practices is more consolidated since, following the crisis, all stakeholders have begun to give more importance to the impact of the banking industry on society. In addition, we analyze if the results are maintained when we distinguish by groups of countries according to the level of development of the stock market, the legal system, and according to the geographical area.

Materials
The sample of this empirical research consists of 166 commercial banks which are listed in 31 stock markets from the years 2010 to 2015, and for which exist CSR and financial information extracted from the hugely employed Thomson Reuters Eikon database [32][33][34][35]. These time and cross-section dimensions have been selected in order to obtain a heterogeneous, as well as balanced panel data that allows us to obtain concluding remarks.
Specifically, we employ the environmental, social, and corporate governance (ESG) performance calculated by Thomson Reuters Eikon for each company. This database company calculates these scores as the weighted average of the scores achieved in more than 70 key performance indicators (KPIs) calculated from more than 400 data points. These variables are included in the model individually, with the objective of analyzing the market value of each of these ESG pillars.
More precisely, the social performance variable measures those CSR actions adopted by banks in relation to providing clients with financial products and services that integrate ethical principles, their commitment to the community, and employees creating a healthy and safe work atmosphere, respecting diversity and human rights, and providing equal opportunities. Meanwhile, corporate governance performance measures whether banks include CSR values in their general vision and strategy of the company, in their decision-making processes, and their communication practices in terms of elaboration and disclosure of sustainability reports. Finally, the environmental variable measures the contribution of banks in terms of minimizing resources, reducing emissions, and product innovation.
It is necessary to point out the importance of environmental performance for investors and other stakeholders. Many of the decisions of these stakeholders could be based on the environmental aspects of the services offered. Therefore, there exists more and more pressure forcing banks to change their management systems to meet these growing demands. Meanwhile, although it is true that the banking industry may not generate serious problems of environmental pollution, it is also true that great savings can be achieved in the billing of electricity, water, fuel, and the use of paper, among others, if corrective measures of environmental improvement are adopted [29].
In Table 1 we show the number of commercial banks listed in each stock market for which information about their ESG performance is available. As we can see, Europe is the geographical area with the most socially responsible banks, with 64 of the total commercial banks listed with CSR practices, followed by Asia with 46, North America with 33, South America with 13, Australia with 6, and, finally, South Africa with 4. Consequently, if we filter by market, the country with the most socially responsible banks is the US with 25, followed by Japan with 21, and China and Italy with 10. In the remaining markets, the number of commercial banks listed with CSR practices is less than or equal to 8.
In Table 1 we also present the average ESG performance of each bank for the years 2010 and 2015. It should be noted that Thomson Reuters Eikon provides an annual score for each company of between 0 and 100 points. This allows us to quickly and easily identify the ESG strengths (50-100 points) or ESG weaknesses (0-49 points) of each financial institution [34]. In relation to the average ESG performance of these banks, we have to highlight that in almost all cases, environmental and social parameters are higher than corporate governance. Moreover, in almost all cases, these parameters are higher in 2015 than in 2010. It indicates that there has been an average increase in the implementation of CSR practices in the banking industry during the years under study. Note: Authors elaboration.

Methods
In order to analyze the effect of these CSR practices on shareholder value creation, we used a modified version of the model proposed by Yu and Zhao [36] adapted to the banking industry. To that end, we include control variables at the bank level and the country level based on the banking literature [4,12,[15][16][17]37,38]: where the dependent variable is Tobin's Q, which relates the valuation that the market makes about the ability to generate profitability for the shareholder (measured by the market capitalization or market value of the company) on the replacement cost of the assets (that is, the cost of acquiring the productive capacity of the company) approximated by the market value of the T and the debt between the value accounting of assets. Thus, if the Q ratio is greater than one, the company will be creating value, since the market is carrying out a valuation of the company above the replacement cost of its assets. Otherwise, value is destroyed. As Jiao [39] points out, this ratio is the most suitable for this type of study since it is a variable based not only on historical data but also on future expectations. This is its main advantage with respect to other alternatives such as profitability ratios based exclusively on the information provided by the company's financial statements, which are being gradually displaced in the empirical studies.
Moreover, in order to explain Tobin's Q, we consider a set of financial variables which are very common in the analysis of banking institutions such as bank size [12,[15][16][17]37,38], return on asset [16,37,38], leverage [4,15,16,36], Tier1 [12,17,37], and the Gross Domestic Product of the country where the banks' headquarters are located [15,16,37,38]. More precisely, LogSize is the annual amount of total asset of each company expressed in logarithm which controls the possible effects of scale; ROA refers to return on assets and is calculated from the quotient between the result before interest and taxes and the total assets, controlling the economic moment that the company is going through; LEV refers to the leverage ratio, calculated from the quotient between total debt and total assets, and controls the impact of the bank's capital structure (the higher indebtedness of a bank creates risks that may affect its growth prospects); Tier1 refers to the capital ratio, calculated from the ratio between provision for uncollectible debtors and total loans granted, controlling assets quality (an increase in this ratio favors the overvaluation in the stock market of the banking institutions); LogGDP is the Gross Domestic Product expressed in logarithms of the country where the bank's headquarters are located and therefore are quoted in their respective stock markets. LogSize and LogGDP have been converted to US dollars by applying the Exchange rate of the closing date of the year to which they refer. Finally, we consider three additional variables that are not related to the financial statements of the listed companies but also to their CSR practices, as we described in the previous section.
Before presenting the empirical results, we show in Table 2 the summary statistics of the variables under study. More precisely, Panel A shows the descriptive statistics. Regarding the descriptive statistics of the ESG performance measures, it is observed that, on average, these banks have a score of 60.20 points out of 100 for environmental performance, 60.69 for social performance, and 50.95 for the performance of corporate governance, with a standard deviation of 32.91, 30.52, and 29.21, respectively. The averages of the variables related to CSR remain above 50% and do not exceed 70%, which indicates that banks on average have good social responsibility practices but not excellent. Finally, regarding the information provided by the matrix of correlations (Panel B), it is evident that there is a high correlation between the qualifications of the CSR components, the most relevant being that of the social dimension with the environmental one. Another preliminary result is that these ESG variables have a positive correlation with the financial ones.

Empirical Results
We present below the results obtained from the estimations described in the previous section. Initially, we show the results for the whole sample ( Table 3). After that, we analyze the results of the estimations considering the type of markets where the banks are located, differentiating between developed and emerging markets (Table 4), considering the legal system, distinguishing between those of common and civil law (Table 5), and finally, geographical zones discriminating between European and American banks ( Table 6).
Empirical studies that compare the effect of CSR activities on firm performance in developed and emerging markets are scarce, especially in the banking industry [12]. Certainly, recent literature on CSR notes that little is known about CSR practices in emerging markets and highlight the need for more research in this area, given the differences in developed markets in terms of capitalization, liquidity, or access of foreign investors. It has not been possible in the past due to the fact that, until very recently, sustainable reliable data about companies listed on emerging markets has not been available [34]. However, we follow McWilliams et al. [40] who point out that it is essential to study a global perspective as well as differentiate between developed and developing markets.
Meanwhile, taking into account previous literature [38,[41][42][43][44], we believe it is important to highlight the legislative differences between countries. As La Porta et al. [41][42][43] argue, legal systems based on common law, such as Australia, Canada, Hong Kong, Ireland, the UK, and the US, are systems more oriented to the discretion that support the results of the private market, impose less ex-ante restrictions on the directors, and favor the protection of the shareholders. In contrast, the legal systems based on civil law-such as those of Austria, Belgium, France, Germany, Italy, Japan, Norway, Portugal, Spain, Sweden, and the Netherlands-are characterized by significant government participation in corporate structures and having labor protection legislation generally regulated by law. More recently, Liang and Renneboog [44] found that a company's CSR rating and the legal origin of their country are strongly correlated. More precisely, they note that companies in common law countries have lower CSR ratings than companies in civil law countries, in which managers are more likely to make decisions in favor of employees and other stakeholders, to the detriment of shareholders. Moreover, Chih et al. [38], who analyzed the determinants of CSR in the banking industry, consider the origin of the legal system of each country because empirical results may suffer from endogeneity bias. Therefore, following these authors, we consider it highly relevant to analyze whether our initial results persist when we analyze separately both groups of countries.
Finally, as McWilliams et al. [40] and Finger et al. [12] point out, much of the extant literature about the effect of CSR practices on firm performance is based in North America or Europe. Moreover, Amel-Zadeh and Serafeim [45] indicate, based on a global survey, that significant differences exist between European and North American investors in terms of their valuation of ESG performance. For these reasons, we also consider it interesting to separately analyze these two geographical areas.
As we can see in Table 3, there exists a positive and significant relationship between banks' environmental and corporate governance performance with Tobin's Q and, therefore, on shareholder value creation, after controlling bank size, return on asset, leverage, capital ratio, and the country's Gross Domestic Product. Moreover, as expected, there exists a significant and negative relationship between a bank's size and leverage with shareholder value creation as well as a country's wealth measured by Gross Domestic Product (GDP). On the other hand, there exists a significant and positive relationship between the remaining financial variables with shareholder value creation. Finally, the correct specification of the model has been demonstrated through the F statistic that tests the joint significance of the explanatory variables, and the adjusted R 2 that indicates the percentage of variability of the dependent variable explained by the explanatory variables included in the model. This table shows the results of analyzing the effect of ESG performance on shareholder value creation in a sample of 166 banks of 31 countries over the 2010-2015 period after controlling for bank size, return on asset, leverage, Tier1, and the Gross Domestic Product of the country where the bank's' headquarters are located. The last rows include the adjusted R 2 and F test statistics. In brackets are the p-values, indicative of the significance of each coefficient and the F test. Finally, we present total panel observations. *** and ** represent the 1% and 5% significance levels, respectively.
As indicated above, there is great pressure from financial stakeholders to adapt banks' management systems and incorporate environmental aspects. As a consequence, there is a positive and significant correlation of environmental performance with Tobin's Q as a measure of shareholders value creation.
At the same time, our results support Jamali et al. [30] when they argue that corporate governance is an important CSR dimension that guarantees accountability, compliance, and transparency and implies a reduction in agency costs for financial stakeholders. That is why there is a positive and significant impact on Tobin's Q when the listed bank has a higher score of governance performance.
In contrast, social performance, which measures a bank's commitment and effectiveness in generating trust and loyalty with its employees, customers, and society, does not have a direct relationship with the bank's financial stakeholders. That could be the reason why the valuation that shareholders make in the stock markets of this measure is significant and negative. This result implies that responsible practices in terms of quality, safety, diversity, and equal opportunities in employment, concern for human rights, as well as quality and safety in products and services, among others, reduce the market value of the company.
Definitively, these results reveal that there is no homogeneity in the value relevance of environmental, social, and governance practices adopted by the selected banks over the entire sample period. It is, therefore, necessary to perform additional tests in order to provide some conclusive results.  This table shows the results of analyzing the effect of ESG performance on shareholder value creation, distinguishing between developed and emerging markets and after controlling for bank size, return on asset, leverage, Tier1, and the Gross Domestic Product of the country where the bank's headquarters are located. The last rows include the adjusted R 2 and F test statistics. In brackets are the p-values, indicative of the significance of each coefficient and the F test. Finally, we present total panel observations. *** and ** represent the 1% and 5% significance levels, respectively.
Moreover, we are conscious that our entire sample is very heterogeneous. For that reason, we provide the results after dividing this entire sample in two sub-samples, taking into account whether banks are listed in a developed or emerging market. As we can see in Table 4, in developed markets, corporate governance performance has a positive and significant relationship with the long-term benefits of banks, measured by Tobin's Q, although we do not observe it for the social and environmental variables. Meanwhile, in emerging markets, the environmental and corporate governance variables are the ones that have a positive and significant relationship with shareholder value creation, while the social one has a negative and significant correlation. Regarding the latter, there is a strong difference between banking entities in developed markets and those in emerging markets when we analyze the relationship between the environmental variable and Tobin's Q. These differences may be due to the fact that many banks that are quoted in emerging and transition markets are under pressure to better manage the risk arising from environmental liabilities [46]. Thus, there is expected to be increasing pressure on banks that do not have environmental management systems to begin to take actions to improve the management of the environmental variable. Moreover, as Schmidheiny [47] indicates, in developing countries, financial institutions are limited by an acute shortage of experience in managing environmental risk and a lack of knowledge about eco-efficient technologies. Table 5. The effect of ESG performance on shareholder value creation common law vs. civil law countries. This table shows the results of analyzing the effect of ESG performance on shareholder value creation distinguishing between common law and civil law countries and after controlling for bank size, return on asset, leverage, Tier1, and the Gross Domestic Product of the country where the bank's' headquarters are located. The last rows include the adjusted R 2 and F test statistics. In brackets are the p-values, indicative of the significance of each coefficient and the F test. Finally, we present total panel observations. *** and ** represent the 1% and 5% significance levels, respectively.

Common Law Countries Civil Law Countries
Meanwhile, in Table 5, we present the results of the empirical model based on the legal system of the country where the banks have their headquarters, distinguishing between common law and civil law countries. As Aguilera et al. [48] and Liang and Renneboog [44] argue, the demand for CSR practices by stakeholders can vary substantially with the legal origin of the country. In this respect, social and environmental performances have a positive and significant correlation with those banks that operate in common law countries. As indicated by Liang and Renneboog [44], these are voluntary practices highly valued by the market, but not in civil law countries that have serious government requirements.
Definitively, in the banks of common law countries, there is a greater interest in governance, transparency, and democratization of social, economic, and public actors, as opposed to a greater emphasis on the creation of an economic and productive structure through institutional strengthening locally, as well as a real protection to environmental resources, existing in civil law. Table 6 presents the results obtained from our estimation model considering the geographic zones of Europe and North America. Although there have been important advances in terms of CSR practices, analyzing the results of this table, we observe that social parameters are notable for their negative and significant correlation with Tobin's Q in Europe and North America. This CSR approach seeks a maximizing benefit behavior [26], in such a way that social objectives are mixed with economic objectives, in aspects such as the improvement of efficiency and productivity via an increase in the quality in the factors of production, processes, and products and services, something demanded by society that becomes an economic benefit in itself for the company. According to a survey carried out by the University of Hong Kong [49], which compares parameters of CSR in Europe and North America, in the classification it makes about the main destination of social investments both inside and outside the country, it is appreciated that at an internal/national level, the European and North American banks are very focused on the creation of conditions of non-discrimination, the maintenance of an equality of opportunities, and transparency in salaries.
If we analyze the parameters of governance, where we refer to the entity's own management, shareholders' rights, and its vision and strategy [50,51], Europe stands out against North America with a positive and significant result against the non-significance of the latter. This table shows the results of analyzing the effect of ESG performance on shareholder value creation distinguishing between Europe and North America and after controlling for bank size, return on asset, leverage, Tier1, and the Gross Domestic Product of the country where the bank's headquarters are located. The last rows include the adjusted R 2 and F test statistics. In brackets are the p-values, indicative of the significance of each coefficient and of the F test. Finally, we present total panel observations. ***, **, and * represent the 1%, 5%, and 10% significance levels, respectively.

Conclusions
Nowadays, CSR has become a requirement for all companies, and therefore stakeholders increasingly demand it, regardless of its possible positive or negative impact on firm performance of companies. These demands are critical in the banking industry, as it is a key element for the proper functioning of the economy. In this sense, the socially responsible actions of banks are taken into account not only by their main stakeholders but also by society in general. Additionally, we must consider the discredit suffered by these entities because of the past economic crisis, which has put in serious doubt the reputation of the financial sector. In 2008, we witnessed how one of the most sophisticated financial systems in the world generated the worst global financial crisis in recent decades. Due to the collapse of the markets of some developed countries, other markets, both in developed and developing countries, were inevitably dragged. As a result of this global financial crisis, there is a growing consensus that the financial system should not only be solid and stable but also sustainable.
This work has two main implications in this context. The first one is to contribute to enrich the existing literature on the impact of CSR practices on the company's financial performance by analyzing the specific case of the banking industry, using a large sample of banks in a post-crisis period in which the implementation of these measures are already quite widespread since the CEOs of these companies have taken due conscience on these matters. The second implication has been to analyze the differentiated effect of environmental, social, and corporate governance measures.
Our overall results reveal that there is no homogeneity in the value relevance of environmental, social, and governance practices adopted by the selected banks over the entire sample period. More precisely, we observe that there exists a positive and significant relationship of banks' environmental and corporate governance performance with Tobin's Q and, therefore, with shareholder value creation. On the other hand, there exists a negative and significant correlation of banks' social performance with shareholder value creation.
These results reveal that there is great pressure from financial stakeholders to adapt banks' management systems and incorporate environmental aspects. Meanwhile, our results also reveal that corporate governance is an important CSR dimension that guarantees accountability, compliance, and transparency, and implies a reduction in agency costs for financial stakeholders. By contrast, we also observe that responsible practices in terms of quality, safety, diversity, and equal opportunities in employment, concern for human rights, as well as quality and safety in products and services, among others, reduce the market value of the company. However, these results exclusively persist for banks quoted on emerging markets and banks that have their headquarters in civil law countries. Thus, it is necessary to conclude that the relationship between ESG performance and banks' shareholder value creation is complex and needs more research.
Despite this, we believe that our results legitimize the managers of these companies to implement these CSR measures in their management systems. However, these results also reveal that work must be continued in the search for the creation of shared value for the shareholder and society in the banking sector.
Future lines of research should be aimed at analyzing the direct impact on the creation of value for the shareholder and society of more specific measures of internal and external CSR of banks, such as the actions adopted to contribute to the satisfaction of employees or the actions adopted to contribute to the reduction of polluting emissions, respectively. It is also necessary to analyze alternative differences among banks, such as their international reach. In this way, we could deepen even more in the essential aspects that should be taken into account by these companies for the creation of shared value and therefore contribute to sustainable development.