IMPACTS OF CODE OF ETHICS ON FINANCIAL PERFORMANCE IN THE ITALIAN LISTED COMPANIES OF BANK SECTOR

Over the last two decades academic literature has addressed much attention to the relationship between corporate ethical practices and financial performance. Results however remain contradictory, especially in terms of direction and effectiveness of their connection. Broadly speaking, most theorizing on the link between social and economic indicators assumes that the evidence is insufficient or too contrasting to draw any generalizable conclusions. In this perspective, this study aims to better explain the connection between corporate ethical practices and corporate financial performance, verifying that it is impacted by a large number of key variables. The empirical research is based on a longitudinal study on Italian listed companies operating in the banking sector. The adoption of the code of ethics is considered to measure their ethical practices, while regarding financial performance several accounting indicators are taken into consideration, including some control variables. To process the dataset a panel regression with fixed effect is applied. The paper aims at strengthening recent studies that consider bidirectional causality in the theory that “corporate social responsibility is both a predictor and consequence of firm financial performance”. Thus, the interest of the study lies in the identification of a reverse causality between positive financial performance and ethical orientation of Italian banking services companies.


Abstract
Over the last two decades academic literature has addressed much attention to the relationship between corporate ethical practices and financial performance. Results however remain contradictory, especially in terms of direction and effectiveness of their connection. Broadly speaking, most theorizing on the link between social and economic indicators assumes that the evidence is insufficient or too contrasting to draw any generalizable conclusions.
In this perspective, this study aims to better explain the connection between corporate ethical practices and corporate financial performance, verifying that it is impacted by a large number of key variables. The empirical research is based on a longitudinal study on Italian listed companies operating in the banking sector. The adoption of the code of ethics is considered to measure their ethical practices, while regarding financial performance several accounting indicators are taken into consideration, including some control variables. To process the dataset a panel regression with fixed effect is applied. The paper aims at strengthening recent studies that consider bidirectional causality in the theory that "corporate social responsibility is both a predictor and consequence of firm financial performance". Thus, the interest of the study lies in the identification of a reverse causality between positive financial performance and ethical orientation of Italian banking services companies. INTRODUCTION Several studies show as trust represents the variable with the greatest impact on customer emotional responses in the banking industry (Marinkovic and Obradovic, 2015;McNeish, 2015;Yu, et al., 2015;Ivanauskiene and vilte, 2015), since in financial services companies it is driven far more by emotional than by functional considerations, among investors as well (Ipsos Public Affairs, 2013). Even ethics appears inextricably connected to financial and banking activities as it forms the basis for trust (Boatright, 2011), without which the banking system could become either dysfunctional or unbalanced (Cullen, 2016;Monferrer-Tirado et al., 2016). Both trust and corporate social responsibility initiatives affect relationships with stakeholders that need to be correctly managed, especially in conditions of information asymmetry (Cui et al., 2016). Actually, they constitute two pillars of the corporate reputation construct (Cuomo et al., 2013;Shanahan and Seele, 2015;Mobin et al., 2016), that is very crucial for the banking sector because financial services deal with people's money and eventual problems, i.e. during crises, trigger serious external collectivized costs (Walter, 2013). Despite the different definitions of corporate reputation proposed over time (Walsh et al. 2009;Walker, 2010) nowadays the literature converge in grouping its main components into six basic pillars: emotional appeal; products and services; income performance; vision and leadership; working environment; social responsibility (Fombrun et al., 2000). Indeed reputation can be included quite legitimately among the tools of corporate governance, with reference to the mechanisms of management and coordination of interaction with stakeholders (Cuomo et al., 2014).
In the banking services industry, as in others, enhancing corporate reputation is both an intangible asset and a source of strategic advantage in incrementing a corporation's long term ability to create value (Gupta et al., 2008). From a theoretical point of view, a general construct can be asserted: the reputation of a company and the welfare of distinct stakeholders are fundamental to stockholders' wealth maximization (Logsdon and Wood 2002;van der Laan et al. 2008) and long-term survival (Becchetti et al., 2012;Dell'Atti and Trotta, 2016;Gorondutse et al., 2014). In order to sustain and improve profitability, managers now have to focus especially on social responsibility toward stakeholders (Mobin et al., 2015), that horizontally crosses and influences all the other five pillars (financial performance, in particular). however, despite the evident interest, reputation analysis has been omitted by scholars in the banking sector, as long as fraud cases and scandals have underlined its relevance, in particular its linkages with ethics (Skowron and Kristensen, 2012;Leiva et al., 2014). In this construct, the paper aims at strengthening some studies that consider corporate social responsibility both a predictor and a consequence of firm financial performance.
Starting from these considerations, the paper is structured as follows: in the 'Conceptual framework' section the role of Corporate Social Responsibility (CSR) in the Italian banking industry is discussed and the objectives of research are highlighted. Subsequently, the main tools to apply CSR, with reference to the adoption of codes of ethics in the banking services industry companies are presented. Thereafter in the 'Methodology' section the design research adopted is illustrated, in terms of a longitudinal analysis on Italian listed companies operating in the banking services industry, covering the period 2001-2015. Following, the main findings are illustrated and discussed. Managerial implications referred to relationships (bi-directional causality) between corporate ethical practices and corporate financial performance conclude the paper, suggesting a different perspective in terms of priority of company stakeholders' fulfillment and ethical firm orientation positioning in the market.

II. CONCEPTUAL FRAMEWORK
In the last decades, initiatives related to CSR in the banking services industry have come under a great debate by the academic community (McDonald and Rundle-Thiele, 2008;Goss and Roberts, 2011;Weshah et al., 2012;Basah and Yusuf, 2013;Kilic, 2016;Bae et al., 2016;Francis et al., 2016). Since the bank industry functioning -in terms of vision, rules and operations -strongly affects the economic development of countries, economic players and people, socially responsible banking is becoming a well-established notion (Scholtens, 2009;Ferreira et al., 20016). Starting from the major theories that support the practice of CSR, as the Social contract theory (Garriga and Melé, 2004), the Agency theory (Foote et al., 2010), the Stakeholder theory (Simmons, 2008;Russo and Perrini, 2010) and the Resource-based view of the firm (Bhattacharyya, 2010), it consists of the requests for corporations to make additional efforts to the well-being of society (Carroll and Shabana, 2010). In addition, different connotations can be highlighted with respect to the social, stakeholders, economic, voluntariness and environmental dimension (Dahlsrud, 2008;Lin-hi and Muller, 2013).
Nevertheless, the banking services industry engagement in CSR activities is quite controversial (Chung-hua et al., 2016). As some mechanisms pertaining to the relationship of CSR and financial performance are identified, such as slack resources mechanism, good management mechanism, penance mechanism and insurance mechanism (Kang et al., 2016), whereas the link between responsibility and economic reward or financial performance appears very contradictory in the managerial literature (Weshah et al., 2012). By the way, three major theoretical approaches can be considered.
A "positive association" assumes that improved CSR performance (voluntaries or induced also by regulations) is a potential source of competitive advantage, as it can lead to more efficient processes, improvements in productivity, lower costs of compliance and new market opportunities, competitive disadvantage (Weshah et al., 2012), suggesting that brand value is positively related to CSR (Bouvain et al., 2013), considering an increased profitability and reducing losses (Simpson and Kohers, 2002), an improving of the revenue function (Wu and Shen, 2013), or CSR strategic consistency in turbulent era (Rivera et al, 2017).
A "negative association" refers that firms with a responsible behaviour incur competitive disadvantage due to the higher costs required for upgrading performance, or that they could transfer to other agents, i.e. customers or government. In this respect, the interest of the firm must be in the maximization of profit (Friedman, 1970) rather than social well-being; this perspective can cause conflict between the management and shareholders because of the reductive effects on the financial performance of the firm (Bauer et al., 2005;Brammer et al., 2006;Jensen, 2010).
Finally, a "neutral association" suggests that there is no causal linkage between corporate social performance and financial performance (Soana, 2011) or there is insufficient empirical evidence to show that CSR strictly affects bankers or stakeholders value creation, because of many factors or variables intervening that may have masked this relationship (Ullman, 1985). Therefore, it seems not emerge an automatic economic effect of ethical activities on competitive performance, neither in a positive nor negative sense (Cuomo et al., 2015). CSR strengths and concerns are expected at the same time to have both positive and negative impact on financial performance in the banking services industry. hence, some studies have started to consider a bi-directional causality in their empirical analyses in order to account for the theory that corporate social performance affects and predicts firm financial performance, and at the same time can be considered as a consequence of it (Waddock and Graves, 1997). Within this theoretical perspective, the present study aims to analyze a bi-directional connection between the application of corporate ethical practices and financial performance (with a moderating effect played by cultural values, Shi et al., 2017).

III. CSR IN THE BANK SECTOR
The impact of CSR on firm performance is increasingly important in the international banking services industry, regardless of a positive or negative effect. Neverthless, the complexity of financial commitments and transactions such as innovative products, long chains of intermediation, additional information and so on, can make "ethical behaviour" a highly ambiguous concept to apply (Oates and Dias, 2016). For this reason, in the Italian banking services industry many formal mechanisms and numerous instruments of control and supervising of the conduct of the players are applied by the main financial regulators, as: The Bank of Italy (Banca d'Italia) -the central bank that has the function of banking regulation and supervision; the Italian Securities and Exchange Commission (Commissione Nazionale per le Società e la Borsa -CONSOB); The Antitrust Authority; the Comitato Interministeriale per il Credito ed il Risparmio (CICR); and The Commissione di vigilanza sui fondi pensione (Covip). (www.bancaditalia.it). On a European level, the Basel Committee develops and supervises processes within the banking sector coordinating on a global scale (Opromolla and Maccarini, 2010;Siclari, 2015).
This extensive control, however has led to the misconception that if a practice is legal consequently it is ethically okay (Boatright, 2013, p. 16). Even though some authors support an integration, with ethical principles included into law (Blodgett, 2011), regulation does not cover all the extended aspects of moral behavior in business. In addition, law is often developed as a reaction to amoral or unethical activities. Finally, law is a relatively low standard of a minimal level of acceptable conduct (Boatright, 2013).
Bank lending, investment and asset management operations are the areas of main application of CSR practices,as key elements of anti-corruption efforts, which is a very relevant theme in the banking services industry. however, also other situations need to be taken into consideration, when they have impact, in terms of benefits or damage on other stakeholders, even though they do not influence the banking profit in the short-term (e.g. faulty product development that could cause system-level failures that might destroy the savings of certain household groups). Therefore, the basic principles of CSR could be fixed in voluntary codes of ethics that go beyond the rules, in order to keep the right direction (Lentner et al., 2015). Many Italian banking services companies positively answered to this suggestion, adopting a code of ethics, pursuant to Italian Legislative Decree no. 231 of June 8, 2001, in compliance with the principles set forth in EU legislation on the prevention of corporate crimes and the assessment of companies' liability.
Among other disclosure practices (Rossignoli, 2013;Salvioni et al., 2014), the code of ethics contains a set of internal guidelines, based on the criteria of fairness, collaboration, loyalty, transparency and mutual respect, that should ensure that firms operate legally, promoting honesty, accountability and ethical conduct (Stevens et al., 2005) and avoiding behaviours that could constitute the offences and crimes according to the Italian Legislation (Kaptein and Schwartz, 2008;Lugli et al., 2009;Opromolla and Maccarini, 2010). The adoption of a code of ethics guarantees a valid model of organization, management and control to protect and further the interests of all stakeholders because of their experience and their sense of moral and legal obligations in pursuance of the banking activity. The specific rules of conduct are applicable to parties subject to the code, and with which such parties must comply; it is also explicated the mechanism of communication, training and monitoring of the code, and constitutes a guide to the company policies and to the legal requirements that govern its conduct (Schwartz, 2002).
As it belongs to a voluntary statement, the adoption of the code of ethics is considered one of the most relevant measurement of the ethical practices of banking services companies. Also in the present analysis, it represents a discriminant signal of CSR practices in the bank sector.

IV. METHODOLOGY
By means of a longitudinal study, the Italian Stock Exchange listed companies operating in the banking services industry have been analyzed, covering the period 2001-2015. The analysis has been stopped at 2015, that is the last year of actual availability of the official balance sheets for Italian companies. National legislation establishes the date of May 31 of every year to publish the firms' balance sheet, thus documents referred to 2016 will be available after May, 31 of 2017. The focus on the banking services industry is due to the fact that financial institutions show several specific reporting requirements (Simionescu and Gherghina, 2014) and are more likely to extensively disclose information on their CSR practices (Andrikopoulos et al., 2014). Furthermore, banking stocks have a considerable weight in the Italian Stock Exchange. At the end of 2015, the market capitalization of banks and financial institutions represented 20% of the total listed companies (Bank of Italy, 2016).
In this context, the research hypotheses are the following: hP1: The adoption of code of ethics affects financial performance; hP2: Banking services industry companies with higher financial performance have more proclivity for the early adoption of code of ethics.
To verify the research hypotheses, the accounting variables presented in Table 1 have been employed. As suggested above, the adoption of the code of ethics (CSR variable) has been proxied through a dummy variable, taking on the value 1 if the company i in the year t holds a code of ethics, 0 otherwise. In any case, in compliance with the Italian Legislative Decree no. 231 of June 8, 2001 all the banking services industry companies listed in Italian Stock Exchange have adopted the code of ethics as the main instrument to measure their propensity in applying CSR activities.
In order to assess the companies' performance 8 accounting-based performance measures have been considered: 1 asset quality indicator (Loan loss reserves/gross loans ratio); 2 capital indicators (Tier 1 and Equity/ liabilities ratios); 3 operations indicators (Net interest margin, ROAA and ROAE); and 2 liquidity indicators (Net loans to assets ratio and Liquid assets/total deposits and borrowing ratio).
It has been decided to use the abovementioned accounting-based measures because they capture historical performance (Simionescu and Gherghina, 2014;Moody's, 2011) and appear more highly correlated with CSR than market-based ratios (Orlitzky et al., 2003), despite their susceptibility to differential accounting procedures and managerial manipulation (Branch and Gale, 1983;McGuire et al., 1988;venanzi, 2012). Furthermore, in order to make sure that the results are not driven by bank heterogeneity, two control variables have been included, and namely loans and total customer deposits, that cover the bank's characteristics such as size, liquidity, as well as indebtedness level and risk.
To define the sample, the dataset of all the companies listed in the Italian Stock Exchange in 2016 has been used. Initially the dataset included 34 listed companies operating in the banking service industry according to the Italian Stock Exchange and Italian Bureau of Statistics classification of economic activities. Subsequently, the companies without financial statements -with regards to Italy -available on Bankscope have been removed. hence, the final sample is composed of 27 companies, directly interviewed in order to ask the year of adoption of the code of ethics.
From data collection a panel dataset of 290 observations related to 27 companies of the sample has been obtained. To process the dataset a panel (cross-sectional time-series data) regression with time fixed effect has been applied because of the number of years taken into consideration for each firm and the analysis of the variation of the variables through the years.
The purpose of this study is to better investigate both the direction and the effectiveness of the existing relationship between corporate ethical practices and corporate financial performance in the Italian banking services industry (Cuomo et al., 2016). In this context, it should be clarified that improving corporate ethical practices by means of the code of ethics could have a significant impact on companies' financial performance variability (hP1). At the same time, higher financial performance may increase the probability in improving corporate ethical practices, as testified by a more rapid adoption of the code of ethics (hP2).
Thus, to test the hP1, it is necessary to estimate a panel regression model with time fixed effect (Cuomo et al., 2016).
The model explains the corporate financial performance as a function of the adoption of the code of ethics and of a vector of covariates and time dummies. In this model it is also considered the variable code of ethics lagged of both one year and two years.
Conversely, in order to verify the hP2 companies have been divided into two clusters based on the year of adoption of the code of ethics (Cuomo et al., 2016). The demarcation year has been the year 2006 (average year of adoption of the code of ethics). Then, descriptive statistics based on univariate analysis have been used in order to analyse the financial performance differences registered before the demarcation year between the two clusters. For this reason, the central tendency of the 8 accounting-based ratios has been examined thanks to the computation of the mean of each indicator. In addition, the mean values have been compared with the optimal thresholds of the selected performance indicators (Zani and Cerioli, 2007).

V. MAIN FINDINGS
The sample is composed of 27 listed companies belonging to the banking services industry in Italy. All of them adopted the code of ethics during the period examined, even though in different years, the most after the year 2006. Table 2 (see, Cuomo et al., 2016) provides descriptive statistics aiming to point out the main values of the accounting-based performance measures, in terms of central tendency (mean), dispersion covering variance and standard deviation, and minimum and maximum values. In addition, Table 3 shows the correlation matrix, that provides the levels of correlations between all pair of the main variables analysed. It can be noted a strong correlation between Equity/Liabilities ratio and Tier 1 ratio, as well as between ROAE and ROAA. Moreover, a strong correlation between the total amount of loans and the total amount of customer deposits has been observed.  Thus, in order to exclude multicollinearity problems, the variance inflation factor -vIF -has been inspected (Table 4, Cuomo et al., 2016). values of vIF above the value 5 imply that variables within the model are greatly correlated (Caramanis and Spathis, 2006;Judge et al., 1987;Studenmund, 2006). Thus, vIF presented in Table 4 enables to rule out undesirable situations that could emerge when the explanatory variables in the regression equation are highly correlated.
Subsequently, the results of the panel regression with fixed effects are presented (Table 5), with reference to the influence of code of ethics on accounting-based financial bank performance (as dependent variables). hereby, it is clear that the code of ethics significantly and positively affects Equity/Liabilities ratio and Net Loans/Tot Assets ratio. On the contrary, as regards ROAE, a significant but negative relationship between the adoption of the code of ethics and this operations indicator has been found. Loans (  Notes: Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 Table 6 and table 7 show the panel regression with fixed effects results regarding the influence of code of ethics on accounting-based financial bank performance as dependent variables with the variable code of ethics lagged respectively of one year and two years. In table 6, a significant and positive relationship between the adoption of the code of ethics and Tier1, Equity/Liabilities and Liquid Assets/Tot Dep & Bor ratios have been found. It is worth to notice that with the variable code of ethics lagged one year, the corporate ethical practices seem to affect positively the corporate financial performance. As shown in Table 6, it has been found a positive relationship between the loans and Net Loans/Tot Assets, as well as between the customer deposits and ROAE, and Liquid Assets/Tot Dep & Bor. Analyzing the variable code of ethics lagged two years (Table 7) emerges that the ethical practices significantly and positively affect Equity/Liabilities and Liquid Assets/Tot Dep & Bor ratios. As for code of ethics lagged one year, it seems that the corporate ethical practices impact positively on the corporate financial performance, even if with a slighter effect (Tier 1 is not affected).  Table 8 shows the descriptive statistics results regarding the financial performance comparison between the two clusters of companies selected. Cluster 1 includes 12 companies, so called early adopters of the code of ethics, encompassing corporations that have adopted the code of ethics before 2006. On the contrary, Cluster 2 is made up of 15 banking services industry firms, so called late adopters, that have introduced the code of ethics after the demarcation year. The mean values of the accounting indicators have been estimated and compared within the two clusters taking into consideration two years before 2001 (the starting year of the analysis). As can be displayed in Table 8, the early adopters (cluster 1, Cuomo et al., 2016) had more brilliant performance before 2001 than the late adopter companies (cluster 2, Cuomo et al., 2016). In fact, cluster 1 presents 5 financial ratios out of 8 with more positive performance than cluster 2. This outcome seems to confirm a more propensity toward ethical behaviours of higher performing companies.

VI. DISCUSSION, MANAGERIAL IMPLICATIONS AND CONCLUSIONS
In accordance with the supposed causality linking corporate ethical practices in the bank sector and financial performance, the findings shown above clearly indicate that CSR practices, in terms of adoption of code of ethics, influence economic and financial results achieved by the companies of the sample. Sure enough, the presence of the code of ethics, as expression of corporate CSR activity, increases company financial structure (Equity/ Liabilities) and corporate capability to employ their assets (Net Loans/Total Assets). The financial performance of the companies during the time (after the first and second year of adoption) confirm this result. In actual fact, at the end of the first year of adoption, the applying of the code of ethics it seems to have an amplifying and positive effect over the economic results (Tier1 Ratio, in particular), that are stabilized, kept and reinforced since the second year on, generating an increase of the level of financial performance of the banking services industry companies.
These results suggest that hP1 has been verified.
An ethical conduct improves relations of the firm with its external stakeholders, in terms of both perception and communication. It also allows a better regulation of the internal stakeholders' behavior, creating evident advantages of cost and resource implementation (Leiva et al., 2014). In addition, the higher performing companies accomplish these better ratios because they take much care of the stakeholders' expectations, demonstrating in this way a higher propensity toward honesty, sincerity and trust. The stakeholders reward this fair conduct choosing these companies in place of other ones less responsible (see Table 8). It can be noted the enveloping of a virtuous circuit that leads the companies of the sample to transform the inclination towards the consideration of stakeholders values into ethical practices via the earlier adoption of the code of ethics.
hP2 is confirmed as well.
Nevertheless, as highlighted by the literature on CSR, ethical and philanthropic responsibilities based on voluntary activities become common practices among firms after that economic and legal responsibility are completely fulfilled (Decker and Sale, 2009). A kind of order with wich take into consideration stakeholders expectations emerges fron this analysis too, privileging stockholders. Only after their satisfaction, the company addresses its resources to implementing ethical practices, in order to carry out further market needs. however, a hierarchical order of steps to reach an overall responsible approach may be respected (Carroll, 1991).
These considerations, however, need to be interpreted in the light of the banking services industry, one of the most heavily regulated market with many institutions (regulators), instruments of control and supervising of the conduct of the players and where ethical requests represents a very remarkable variable, also demonstrated by the strong participation of all the companies of the sample to the survey.
Anyway, this managerial effort toward ethical practices could also provoke the restriction by means of formal mechanisms and this is a limit of the research.
Instead, future perspectives of research could verify the concrete use of the code ethics that differs from its formal adoption, in the persuasion that CSR activity can really improve corporate financial performance, provided that it is not a mere communication tool but a corporate value and a practical model of behaviour.