The ethical commitment of independent directors in different contexts of investor protection

García-Sanchez, Isabel María; Frías Aceituno, José Valeriano; Rodríguez Domínguez, Luis Article The ethical commitment of independent directors in different contexts of investor protection BRQ Business Research Quarterly Provided in Cooperation with: Asociación Científica de Economía y Dirección de Empresas (ACEDE), Madrid Suggested Citation: García-Sanchez, Isabel María; Frías Aceituno, José Valeriano; Rodríguez Domínguez, Luis (2015) : The ethical commitment of independent directors in different contexts of investor protection, BRQ Business Research Quarterly, ISSN 2340-9436, Elsevier España, Barcelona, Vol. 18, Iss. 2, pp. 81-94, https://doi.org/10.1016/j.brq.2014.07.001 This Version is available at: https://hdl.handle.net/10419/206319 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-nc-nd/4.0/ BRQ Business Research Quarterly (2015) 18, 81---94 www.elsevier.es/brq BRQ Business Research Quarterly ARTICLE The ethical commitment of independent directors in different contexts of investor protection Isabel María García-Sáncheza, José Valeriano Frías Aceitunob, Luis Rodríguez Domíngueza,∗ aUniversidad de Salamanca, Spain bUniversidad de Granada, Spain Received 21 January 2013; accepted 22 July 2014 Available online 16 September 2014 JEL CLASSIFICATION M14; M48 KEYWORDS Codes of ethics; Business ethics; Board independence; Board of directors; Corporate governance Abstract The purpose of this study is to compare, for countries with different legal environments, the degree to which boards of directors may improve corporate ethical behaviour by designing codes of ethics. These codes address issues such as a company’s responsibility regarding the quality of its products and services, compliance with laws and regulations, conflicts of interest, corruption and fraud, and protection of the natural environment. Using a sample of firms from 12 countries, we obtain evidence that a greater presence of independent directors on the board leads to the existence of more complex codes of ethics. Moreover, there are significant differences between countries with high levels and countries with low levels of investor protection as regards the effectiveness of independent directors in constraining unethical behaviour by managers. © 2014 ACEDE. Published by Elsevier España, S.L.U. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Introduction The effects of scandals involving fraud, corruption, etc. on a company’s image, profitability, and long-term survival have heightened the corporate concerns regarding ethics and led to the development and implementation of ethical codes (Fan et al., 2008). These codes contribute to formalizing the corporate values, institutionalizing the guidelines for ∗Corresponding author at: Facultad de Economía y Empresa, Campus Miguel de Unamuno, 37007 Salamanca, Spain. Tel.: +34 923 29 45 00x3132; fax: +34 923 29 47 15. E-mail address: [email protected] (L. Rodríguez Domínguez). decision making within the organization, and laying down standards for responsible behaviour (McKinney et al., 2010; Singh, 2011). Various international agencies and organizations (New York Stock Exchange, 2003; OCDE, 2004) and researchers (García-Sánchez et al., 2008; Rodríguez Domínguez et al., 2009) have considered the role of the board of directors in business ethics. In this sense, this body is responsible for supervising the senior management and for preventing and/or punishing inappropriate behaviour (Schwartz et al., 2005). The need for the involvement and commitment of the senior management and its delegate bodies to ensure the effectiveness of an ethics programme has been specially emphasized (Weaver et al., 1999). http://dx.doi.org/10.1016/j.brq.2014.07.001 2340-9436/© 2014 ACEDE. Published by Elsevier España, S.L.U. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). 82 I.M. García-Sánchez et al. Among the desirable features of the board’s functioning for business ethics, its independence stands out. It ensures greater effectiveness in the control of the senior management (Hermalin and Weisbach, 1988; Zahra and Pearce, 1989). Furthermore, it has a decisive influence on the design of strategies for corporate responsibility (Jo and Harjoto, 2011) and ethical strategies (García-Sánchez et al., 2008). The final goal is to limit managers’ opportunities for self-benefit and the high direct costs associated with malpractice, such as fines and prison sentences, as well as negative media exposure and the consequent damage to the firm’s reputation (Johnson and Greening, 1999). On the other hand, Ravina and Sapienza (2010) argued that independent directors are economic agents, whose decisions may be influenced by their own interests. Moreover, they are subject to the institutional environment in which they operate (Bebchuk and Weisbach, 2010), which determines the type and functioning of the control mechanisms (La Porta et al., 2000). Therefore, previous studies have suggested the importance of the board of directors regarding its active role in ethical practices, its monitoring role, and the relevance of its independence as a positive feature that may encourage stricter ethical behaviour. Based on these premises, this study examines the level of involvement of independent directors in the development of ethical codes. In other words, we attempt to determine whether independent directors promote the implementation of ethics codes addressing a wider range of issues. It also studies whether this involvement varies according to the level of investor protection present in the institutional environment in which the company operates. Along this line, the level of investor protection is considered to be one of the most important factors among the characteristics of the institutional environment. It may contribute to explaining the presence or absence of opportunistic behaviour by managers (Benos and Weisbach, 2004) and/or their misappropriation of investors’ rights (Dyck and Zingales, 2004). Such expropriation or opportunistic behaviour by internal agents can take the form of the personal use of company assets, acquisition strategies that destroy value, accounting manipulation, bribery, corruption, etc. (Morck et al., 1990). However, there is no clear evidence of interaction with the institutional environment that could influence the behaviour of independent directors. Previous research has produced conflicting results; although a substitutive relationship has been shown to exist between the institutional environment and the characteristics of the board of directors (Aggarwal et al., 2009), many authors have observed a lower degree of opportunistic behaviour by managers, such as accounting manipulation or information asymmetries, in countries where the investor protection is stronger (Ball et al., 2000; Bhattacharya et al., 2003; Leuz et al., 2003; Bushman et al., 2004). On the other hand, there is no empirical evidence of an interaction between the two factors with respect to the design of corporate strategies (Bebchuk and Weisbach, 2010). Therefore, we propose that countries with strong investor protection offer a more suitable context for independent directors in order to implement more complex codes, whereas these directors in countries with less investor protection would face greater reluctance to implement mechanisms that restrain opportunistic behaviour. Taking these considerations into account, our study makes a new contribution that copes with both perspectives: the importance of the board’s independence as a corporate governance mechanism in the implementation of ethics codes and the difference in involvement according to the context in which the company operates. The two perspectives are combined in the proposal of our research model, in which the scope of ethics codes is expressed as a function of board independence, its interaction with the investor protection existing in the environment, and some control variables. To test this model, we use a panel data sample made up of 5380 observations for an average of 760 companies from 12 countries, for the time period 2003---2009. The financial data were obtained from Compustat, whereas the data on corporate governance and ethics were extracted from the EIRIS database. Our findings point out that independent directors positively influence the implementation of codes with a wider scope, with an ethical commitment that extends beyond the discrimination and the adequate relationships with providers and clients. This compromise encompasses a wider range of ethical issues, affecting the sustainable use of resources and the overall relationship with society. The findings obtained are in accordance with previous studies (e.g. Ibrahim and Angelidis, 1995; Johnson and Greening, 1999), which emphasize the potential link between board independence and willingness to show the firm’s ethical behaviour. Additionally, we detect that the investor protection existing in the corporate context in which the company operates influences the extent of the impact of independent directors on the development of more complex codes. Hence, directors of companies in countries with a lower level of investor protection have more difficulty in implementing ethical codes with a wider range of contents. This result reinforces the previous evidence found by Beck et al. (2003) or Shen and Chih (2005), which show a complementary link between internal and external mechanisms of corporate governance. This study is structured as follows. The section ‘‘Ethical codes’’ describes ethical codes as a mechanism for combating unethical practices and the importance of the board in ensuring their effectiveness. The section ‘‘The ethical role of independent directors’’ contains the main research hypotheses, related to the role played by independent directors in implementing more complex ethical codes and the stronger influence expected in those countries with greater investor protection. The fourth section presents the sample analysed, the variables used to test the hypotheses, and the models proposed. The section ‘‘Empirical analysis’’ explains the results obtained after estimating the original models and undertaking some sensitivity analyses, and finally section ‘‘Conclusions’’ summarizes and concludes. Ethical codes Business ethics can be understood as the set of values, norms, and principles that seek to achieve respect for the rights generally recognized within a society. To institutionalize ethical conduct within corporations, specific codes have Board independence and ethical behaviour 83 been developed. According to Kaptein and Schwartz (2008, p. 113), such a code should constitute ‘‘a distinct and formal document containing a set of prescriptions developed by and for a company to guide present and future behaviour on multiple issues of at least its managers and employees toward one another, the company, external stakeholders and/or society in general’’. More specifically, codes transmit ethical values to members of the organization (Wotruba et al., 2001), offering them moral guides or anchors when new and confusing situations are encountered in the workplace (Chua and Rahman, 2011) and in decision making (Urbany, 2005). The existence of a code of ethics was initially considered as preventive medicine against fraud, misappropriation, embezzlement, nepotism, cronyism, favouritism, abuse of influence, abuse of power, the illegal financing of political parties, the misuse of privileged information, workplace mobbing, defamation, false advertising, negligent discrimination, and environmental actions, among other unethical practices. However, several studies have shown that the mere existence of such a code does not guarantee ethical business behaviour (Vethouse and Kandogan, 2007; Kaptein and Schwartz, 2008). In this respect, Ibrahim et al. (2009), Kaptein (2011), and Singh (2011) observed that the existence of training programmes and communication channels to familiarize users with the content and intent of this code will improve its effectiveness. It has been shown that the key factor in successfully influencing the behaviour of company staff is the actual content of the ethical code; this is what determines its impact on managers’ judgements and decisions. Additionally, the factor of corporate governance is of crucial importance in ensuring the effectiveness of codes of ethics. Many previous studies on the practice of corporate ethics have highlighted the need for the involvement and commitment of the senior management and its delegate bodies to ensure the effectiveness of an ethics programme (Weaver et al., 1999). Maintaining organizational integrity and ethics is assumed to be among the skills that directors and managers should actively practice (Vethouse and Kandogan, 2007). Consequently, according to Bonn and Fisher (2005), boards and senior management need actively to promote, manage, and monitor a culture that emphasizes ethical behaviour and integrity within the organization. They should evaluate its current strategies, policies, and procedures and investigate whether ethical behaviour is encouraged and the company’s ethical values are reflected. The primary goal should be to develop an ethical code setting out the recommendations and principles to be followed in a wide range of situations. This is regarded as the most important factor underlying the achievement of more ethical business behaviour (Ibrahim et al., 2009; Kaptein, 2011; Singh, 2011). According to Schwartz et al. (2005), board members are ultimately responsible for the selection, permanence, and discipline of senior officers and must take their ethical obligations into account. Furthermore, the main codes of corporate governance applicable in different countries state that among the board’s functions is that of ensuring ethical conduct within the organization (Aguilera and Cuervo-Cazurra, 2009). The ethical role of independent directors Independent directors have a particular responsibility to safeguard the interests of shareholders and investors. They supervise the senior management and ensure that business ethics form part of the organizational culture (Rodríguez Domínguez et al., 2009). One ethical policy is to encourage the adoption of complex ethical codes to restrain inappropriate actions and maintain the firm’s good image and long-term survival (Singh, 2011). In addition to their stipulated functions, independent directors are argued to be the best equipped and most appropriate board members to take responsibility for compliance with the regulations and to ensure the ethical behaviour of the firm (Ibrahim and Angelidis, 1995). This is due to their greater objectivity and independence in analysing the management process (Prado-Lorenzo and García-Sánchez, 2010), as well as the impact of their success or otherwise, in this respect, on their personal standing (Frías-Aceituno et al., 2012). Fama and Jensen (1983) and Lorsch and Maciver (1989) argued that the main benefits enjoyed by independent directors are prestige, reputation, job openings, and networking opportunities. Directors who perform their duties effectively and efficiently are more likely to be rewarded, while those who work in companies that obtain poor results will tend to lose privileges. Homstrom (1999) observed that concerns about the firm’s reputation can create incentives for directors to avoid risky actions that could have negative consequences for their future as external directors. Any loss of reputation would reduce their chances of being offered another such post (Srinivasan, 2005; Fich and Shivdasani, 2007) and even of retaining their present one (Fahlenbrach et al., 2010). In addition, these directors will have a more critical view of the corporate activities that may be carried out, having greater freedom to defend costly and/or unpopular decisions (Arora and Dharwadkar, 2011). As mentioned previously, their role may be especially outstanding when dealing with ethical issues. They may be less reluctant to investigate/prevent cases of fraud (Beasley, 1996). Furthermore, they will be more receptive to the demands of external groups for improved ethical behaviour by the company and will seek to encourage internal groups to meet the goals of good governance (Fombrum and Shanley, 1990). Given that (1) one of the board’s functions concerns ensuring ethical conduct within the organization (Aguilera and Cuervo-Cazurra, 2009), (2) their ethical functions may be better fulfilled in the presence of independent directors (Ibrahim and Angelidis, 1995), and (3) ethical issues are mainly dealt with through codes of ethics, we posit our first research hypothesis: Hypothesis 1. The presence of independent directors encourages the implementation of ethical codes. A further stage in this analysis is to determine whether the involvement of independent directors in the definition of policies on business ethics is similar in all corporate environments (Bebchuk and Weisbach, 2010). Firstly, this debate is based on the arguments of Ravina and Sapienza (2010), for whom the decisions of independent directors 84 I.M. García-Sánchez et al. are determined by the satisfaction of their own utility function as economic agents. Furthermore, the earlier empirical evidence shows that the functioning of the internal mechanisms of corporate governance (e.g. boards and their independence) cannot be understood without considering the mechanisms associated with the institutional environment proposed by La Porta et al. (1997, 1998, 2000). La Porta et al. argued that the laws protecting the rights of investors and the degree of effective implementation of these laws are the main determinants of the way in which corporate governance develops. Investor protection is understood as the degree to which business legislation and its application protect investors from the expropriation that may be conducted by insiders. Greater legal protection for investors would restrain the behaviour of directors, limiting their chances to engage in opportunistic behaviour and preventing insiders from obtaining private benefit (Hart, 1995; Djankov et al., 2005). Therefore, the extent of investor protection may influence directors’ impact on ethical issues, particularly when the directors come from outside the company, like independent ones. More specifically, there are three possible scenarios regarding the moderating role of investor protection in the ethical commitment of independent directors. In the first such scenario, the least plausible one, the effect would be neutral. These directors would perform their roles in the same way worldwide, with no influence being exerted by the institutional environment. The other two scenarios require us to consider systemic variations in directors’ behaviour, as a result of the incentives provided in different institutional settings. The second scenario assumes a substitutive relationship between corporate governance mechanisms. Along this line, Doidge et al. (2004) suggested a negative relationship between the strength of firm-level governance and countrylevel laws. Thus, the independent directors of companies located in countries with less investor protection would be expected to take a more active role in developing complex ethical codes than other directors. They would seek the adoption of better firm-level governance to offset the weaknesses in the law or in its application and indicate their intention to offer more rights to investors. The third scenario follows the complementary argument proposed by Doidge et al. (2004), according to which the two mechanisms are mutually reinforcing. In this situation, in countries where the regulatory and business environment offers greater investor protection, independent directors are expected to take a more active role in developing complex ethical codes than the other directors. Within this relationship, the combination of interventionism by the independent directors and the strength of the legal and judicial protection of investor rights would ensure the highest degree of investor protection. Empirically, although there is no clear evidence of the interaction with the institutional environment that could impact on the behaviour of independent directors, the first scenario should be rejected. Rather, previous research has revealed the existence of substitutive (Denis and Kruse, 2000; Huson et al., 2001; Aggarwal et al., 2009) and complementary effects (Claessens et al., 2000; Gul et al., 2002; Beck et al., 2003; Leuz et al., 2003; Shen and Chih, 2005) between diverse internal and external mechanisms of corporate governance. In this paper, we argue that the effect of the level of investor protection on managers’ ethical commitment will be complementary, as the consequence of rational decisions made by directors in their role as economic agents. More specifically, we suggest that in the decisions taken as economic agents, managers will realize that the development of ethical codes is the outcome of demands in this respect made by social activists, regulators, environmental legislative pressure, etc. Their implementation is also the result of external pressure from investors and other capital market agents, who seek to protect the company and secure it against potential unethical conduct (Stevens et al., 2005; Robertson et al., 2013). Moreover, managers will consider that the positive outcome obtained from effective codes is not limited to the company’s internal affairs, i.e., achieving more ethical behaviour, developing a comprehensive ethical culture, and increasing employee satisfaction. It will also be transferred to its external representation, enhancing institutional legitimacy and improving the organizational approach to public accountability (Valentine and Fleischman, 2008). All of these aspects promote a positive external image of the firm; they influence shareholders’ perceptions, generate reputational benefits (Stevens et al., 2005), and improve financial performance (Weigelt and Camerer, 1988; Stevens et al., 2005). Both the pressure from investors and the benefits generated by the adoption of a code of ethics are more apparent in capital markets that are more developed and transparent. Given that these characteristics are usually typical of countries with higher levels of investor protection (La Porta et al., 1998), independent directors will obtain greater benefits from their involvement in ethical issues in these countries. In other words, these directors will view a code of ethics as a crucial component of the company’s ethical infrastructure. A code will contribute to the development of a culture and an image that will maintain or restore the firm’s ethical reputation vis-à-vis its stakeholders, especially investors (McKinney et al., 2010). In addition, it will help raise public confidence in the firm’s conduct, influencing external impressions of its directors’ work, and thus help them to achieve greater professional recognition and provide more opportunities for participation in other boards of directors. Therefore, we suggest that the business environment for companies located in countries with strong investor protection will be more suitable for independent directors to assume a more interventionist outlook with respect to the firm’s policy on ethical issues. In those contexts, they enjoy greater flexibility to implement a more complex ethical code, thereby enhancing investor protection. On the contrary, the managers of companies located in countries with less investor protection have greater potential to obtain private benefits (Nenova, 2003; Dyck and Zingales, 2004); thus, they will present stronger opposition to mechanisms that limit opportunistic conduct (Renders and Gaeremync, 2007). On this line, the following hypothesis is proposed: Hypothesis 2. The implementation of ethical codes on behalf of independent directors is more active in countries with stronger investor protection. Board independence and ethical behaviour 85 Table 1 Sampling distribution, by countries. Country Sampling period Total Frequency 2003 2004 2005 2006 2007 2008 2009 Germany 0 0 0 34 44 42 30 150 2.79% Canada 48 54 50 57 54 54 56 373 6.93% Denmark 0 0 0 0 16 16 15 47 0.87% Finland 0 0 0 0 19 19 18 56 1.04% France 0 0 0 0 36 38 38 112 4.98% Italy 34 35 39 41 42 41 36 268 4.98% Netherlands 23 25 26 26 25 24 23 172 3.20% Norway 0 0 0 0 10 10 10 30 0.56% Spain 32 34 37 38 37 36 36 250 4.65% Sweden 0 0 0 0 41 39 39 119 2.21% UK 115 127 129 136 140 140 130 917 17.04% USA 365 382 412 422 424 448 433 2886 53.64% Total 617 657 693 754 888 907 864 5380 Research methodology Population and study sample The study population is comprised of the listed European, US, and Canadian companies for which economic---financial data are available in Compustat, together with the data relating to corporate governance and ethical codes published by Ethical Investment Research and Information Services (EIRIS). Merging the two databases provided a sample of 5380 observations for an average of 760 companies from 12 countries, for the analysis period 2003---2009. The panel data sample is non-balanced, as not all companies were observed for all the financial years in question; thus, the number of firms observed ranged from 617 to 907 for this period. The EIRIS database contains information on the corporate social responsibility practices of the 3000 largest companies worldwide. The information is mainly obtained from public data, although questionnaires are used to ascertain unpublished or unclear data. Subsequently, these data are checked by external experts, unrelated to EIRIS and the companies in question. This methodology has been awarded the AI CSRR Voluntary Quality Standard. Its main customers include the FTSE indices, the stock exchanges in Johannesburg and Mexico, and brokers, asset managers, etc. in Europe, North America, Australia, and Asia. As shown in Table 1, there is a bias in the sample distribution due to the weight of the number of companies in the USA and the UK. This results from the size bias of the companies making up the EIRIS database. From the temporal point of view, the highest number of observations was obtained for 2008. Dependent variable The dependent variable, CElevel, was defined to represent the implementation of a code of ethics in two ways: its existence and its level of application. This is an ordinal variable that takes values between 0 and 600 500 400 300 200 Number of companies 100 0 2003 2004 2005 2006 2007 2008 2009 0 1 2 3 4 CElevel Figure 1 Evolution of the level of application of the ethical codes applied. 4 by identifying the level or inclusivity of the ethical code. • A value of 1 represents limited inclusion, i.e., the code refers to a very limited number of aspects, such as conflicts of interest, corruption, and bribery. • A value of 2 is a basic level, incorporating, in addition to the first level, recommendations on questions such as discrimination, occupational hazards, the work environment, and the confidentiality of information. • A value of 3, the intermediate level, incorporates, as well as the aspects addressed in the two previous levels, the principles and values related to relationships with customers, suppliers, and competitors. • A value of 4, the advanced level, adds references to the sustainable use of resources, relations with society, and any other value that forms part of the corporate culture. • The value of 0 is assigned to companies that do not express any ethical commitment. Table 2 shows that 96.7% of the companies listed had implemented a code of ethics. Regarding the level of these codes, 12.7% were classed as limited or basic, while 27.6% were intermediate and 56.4% advanced. 86 I.M. García-Sánchez et al. Table 2 Level of application and awareness of ethical codes. 2003 2004 2005 2006 2007 2008 2009 Absolute frequency Relative frequency 0 77 11 12 11 18 15 5 149 3.3 1 106 37 30 23 27 24 14 261 5.0 2 60 49 57 59 68 68 54 415 7.7 3 130 213 226 220 231 243 231 1494 2.6 4 244 347 368 441 544 557 560 3061 5.4 Regarding the evolution in this field, Fig. 1 shows that there has been a significant decrease in the number of companies that either do not have a code of ethics or have one of only limited reach. On the contrary, a corresponding increase occurred in the numbers of companies classified as intermediate or advanced in this respect. Furthermore, during the period examined, the number of companies of which the ethical codes were considered basic remained unchanged, but there was a very significant increase in the number of companies that implemented advanced codes of ethics, in terms of the recommendations made. Independent and control variables To test hypothesis H1, we defined as the independent variable %INDEP, representing the percentage of independent directors on the board of directors of each company. A time lag was applied to the variable to avoid potential problems of endogeneity arising from the fact that the more ethical companies may promote a greater presence of independent directors on their boards. In analysing hypothesis H2, we must first define the institutional environment. In this respect, Chong and Lópezde-Silanes (2007), among others, suggested that investor protection should be defined in terms of: (1) the tradition and the existence of laws that safeguard the interests of investors and (2) the characteristics of the judicial institutions created to ensure their implementation and enforcement; in this sense, judicial support for the applicable rules and laws has the power to prevent, or at least limit, the expropriation of investors. The underlying legal tradition is the foundation of basic legal rights, including the protection of property rights, and the bedrock of corporate and securities law. La Porta et al. (1998) classified legal traditions into two families, common law and civil law. Common law countries developed a legal tradition based on customary law, characterized by less reliance on the statutes and a preference for contracts and private litigation to resolve disputes. In contrast, civil law countries are characterized by greater explicit reliance on laws and procedural codes and by the preference for the state to regulate the resolution of private disputes. In this sense, common law environments offer greater protection for investors because of their strong focus on private contracts and the protection of property rights. Therefore, our first variable in investor protection is that of the legal tradition, and this is encoded by a dummy variable that takes the value 1 for countries with a common law legal tradition and the value 0 otherwise. A second level of investor protection is provided by commercial law, in particular the legal mechanisms that protect investors, alleviating the agency problems that may occur. La Porta et al. (1998) developed an index of antidirector rights on the basis of the presence/absence of six specific elements of investor protection. This index uses six values to measure the ease with which investors can exercise their rights in response to opportunistic behaviour. The third level of protection is based on the existence of other parameters of the legal system (Deffains and Guigou, 2002), such as mechanisms to enforce the existing regulations, which can alleviate the firm’s ethical problems. In this respect, Defond and Hung (2004) and Durnev et al. (2004) observed that the strength of the enforcement control mechanisms is more significant than the mere existence of a broad set of laws governing them. To characterize the mechanisms supporting legal provisions, we shall use two indices proposed by La Porta et al. (1998) to assess a country’s legal framework: the index of judicial efficiency and the index of law and order. The first index rates the independence and professionalism of the judiciary in all kinds of proceedings, together with the adequacy of its time scales, especially as regards the reasonableness of judicial delays. The law and order index concerns the generality and non-arbitrariness of rules, their comprehensiveness, their fairness, etc. Both of these enforcement control mechanisms are true determinants of the protection of investors’ rights because they determine the responsibility of company managers and directors (La Porta et al., 1998). To operationalize investor protection and make it interactive with the proxy for independence (following Hillier et al., 2011), we created three sub-indices: (1) DCL, which takes the value 1 if the firm is located in a common law country and the value 0 if it is located in a civil law country; (2) DAR, which takes the value 1 if the firm is located in a country with investor protection rights that are more restrictive on directors than the median level in the sample and 0 otherwise; and (3) DEF, which takes the value 1 if the firm is located in a country where the index of application of the law is above the median and 0 otherwise --- the latter index is obtained as the sum of the indices of judicial efficiency and of law and order. Finally, we take as a proxy for effective investor protection the sum of the three dummy variables --- DCL, DEF, and DAR --- and construct a new dummy variable, DINV PROTEC, which takes the value 1 if the firm is located in a country with above-average investor protection and 0 otherwise. Additionally, in order to test the role of independent directors in each investor protection system, we interacted the percentage of independent directors with the investor protection dummy. This interacted variable is labelled %INDEP*DINV PROTEC and identifies the percentage Board independence and ethical behaviour 87 of independent directors in countries with an above-average level of investor protection. As with the %INDEP variable, a time lag is applied to the variable in order to avoid problems of endogeneity. To eliminate bias from the results, we considered a set of control variables previously shown to be effective in this respect: board diversity, size, and activity, company size, level of borrowing, and growth opportunities. The numeric variable %FEM represents the percentage of female directors on the company board. Gender diversity plays a very important role in the ethical commitment of companies, due to differences in the ethical frameworks used by men and women in their moral judgements (Harris, 1989; Ibrahim et al., 2009). According to previous studies (Stultz, 1979; Ibrahim and Angelidis, 1994; García-Sánchez et al., 2008; Rodríguez Domínguez et al., 2009), female directors are usually more sensitive to ethical issues. The numeric variable BOARDSIZE represents the total number of directors. In general, large boards of directors have more serious agency problems and a greater need for ethical codes (García-Sánchez et al., 2008). The BOARDSIZE2 variable is the square of the above and is taken into consideration because authors such as Diwedi and Jain (2005) incorporated this value into their analyses in order to test for a possible non-linear relationship. In other words, the general behaviour of the board may be modified when a certain board size is exceeded. The BOARDACT numeric variable measures the number of board meetings in each year. The effect of this variable is not clear a priori. On the one hand, Lipton and Lorsch (1992) argued that active directors are more effective, presenting a greater predisposition towards corporate social responsibility. However, a large number of meetings may simply evidence inoperability and the fact that the directors are taking on too much, negatively affecting business management (Vafeas, 1999). The indicator of company size, FIRMSIZE (the logarithm of total assets), is valuable because of this factor’s effect on the processes of corporate social legitimation, as highlighted in studies such as Hackston and Milne (1996), Archel and Lizarraga (2001), Gray et al. (2001), and Archel (2003). The level of company debt, LEVERAGE (the ratio of debt to equity), is another factor associated with the development of codes of ethics, especially as a means of prevention and response to agency conflicts that may arise. With respect to growth opportunities, companies with high MTB values (calculated as the ratio between the market value and the book value of business assets), comparably with expectations for the legitimation process, are expected to develop ethical codes aimed at reducing the problems of asymmetric information; this enables them to regulate employee behaviour (Larrán and García-Meca, 2004; Gandía and Pérez, 2005). Finally, the industry sector variable is introduced in order to control the effect of the firm’s economic activity on the level and impact of ethical codes; some industry sectors are more likely to establish internal rules to prevent potential unethical situations that may occur in relation to the specific activity of the sector (García-Sánchez et al., 2008). Table 3 summarizes the descriptive statistics of the variables proposed for analysis. As can be seen, on average boards are composed of 11 directors, presenting a level of activity of 9 meetings per year. The average proportion of independent directors is 72.59%, with higher levels in institutional environments featuring greater investor protection (81.40%). The average presence of female directors is very low (7.98%). The majority of observations (60.6%) refer to institutional settings in which investor protection is higher than the average level for our sample. Table 3 also shows that the average MTB ratio is 1.80, i.e., the market value exceeds the book value of the business assets, indicating that investors expect to obtain added value in the future, in view of the current value of company assets. Finally, these companies have an average debt ratio of 16.56%, which is notably lower than the 21.58% observed in other studies with international samples (Brockman et al., 2013). This variable has a standard deviation of 55.66, indicating a high degree of dispersion. To test the effect of this disparity, two analyses were conducted, the first with the original values and the second after eliminating outliers. As no changes were observed in the results obtained, we decided to retain the original values in the analysis. Analytic model From the hypotheses set out and the variables defined above, the following models are proposed: CElevelit = ˇ0+ ˇ1%INDEPit−1+ ˇ2DINV PROTECit + ˇ3%FEMit + ˇ4BOARDSIZEit + ˇ5BOARDSIZE2it + ˇ6BOARDACTit + ˇ7FIRMSIZEit + ˇ8Leverageit + ˇ9MTBit + ˇ10Sectorit + ˇn iYear + it + i(1) CElevelit = ˇ0+ ˇ1%INDEPit−1+ ˇ2%INDEP ∗DINV PROTECit−1+ ˇ3DINV PROTECit + ˇ4%FEMit + ˇ5BOARDSIZEit + ˇ6TAMCONS2it + ˇ7BOARDACTit + ˇ8FIRMSIZEit + ˇ9Leverageit + ˇ10MTBit + ˇ11Sectorit + ˇn iYear + it + i(2) where ‘‘CElevel’’ is an ordinal variable that takes a value from 0 to 4, depending on the level of application of the company’s ethical code. ‘‘%INDEP’’ is the percentage of independent directors on the board at t − 1. This is interacted with the variable ‘‘DINV PROTEC’’ to identify the role played by independent directors at t − 1 in companies located in countries where the levels of investor protection are above average. ‘‘%FEM’’ represents the percentage of female directors on the board. ‘‘BOARDSIZE’’ is a numerical variable that represents the total number of directors, both internal and external, on the board. ‘‘BOARDSIZE2’’ is a numerical variable that represents the square of the total number of directors, both internal and external, on the board. ‘‘BOARDACT’’ is a numerical variable that reflects the number of board meetings held each year. 88 I.M. García-Sánchez et al. Table 3 Descriptive statistics. Min Percentile 25 Median Mean Std deviation Percentile 75 Max %INDEP 0.48 0.56 0.70 0.73 0.38 0.89 0.91 %INDEP*DINV PROTEC 0.56 0.68 0.80 0.81 0.11 0.88 0.95 %FEM 0.00 0.00 0.75 0.08 0.14 0.13 0.15 BOARDSIZE 7.00 9.00 11.20 11.27 3.29 13.00 15.00 BOARDACT 4.00 6.00 8.50 8.59 3.80 10.00 12.00 FIRMSIZE 5.20 7.90 8.75 8.81 2.23 10.03 11.56 Leverage 6.23 8.75 16.50 16.56 55.66 24.67 25.49 MTB 1.01 1.45 1.70 1.80 1.70 3.75 4.20 Frequency Absolute Relative DINV PROTEC 3.259 60.6% ‘‘%INDEP’’ is the percentage of independent directors on the board. It interacts with the DINV PROTEC variable to identify the role of independent directors in companies located in countries with above-average levels of investor protection. ‘‘%FEM’’ represents the percentage of women on the board of directors. ‘‘BOARDSIZE’’ is a numeric variable representing the total number of board members, internal and external. ‘‘BOARDACT’’ is a numeric variable expressing the number of board meetings held each year. ‘‘FIRMSIZE’’ is the size of the corporation, measured by the logarithm of its total assets. ‘‘Leverage’’ is the level of corporate debt represented as the ratio external funds/equity. ‘‘MTB’’ represents the market value of the company compared to its book value. ‘‘FIRMSIZE’’ is the size of the company, expressed as the logarithm of its total assets. ‘‘Leverage’’ is the level of company debt, represented by the borrowing/equity ratio. ‘‘MTB’’ is the market-to-book value of the company. ‘‘Sector’’ is a numerical variable that identifies the area of activity of the company. ‘‘Year’’ is a dummy variable vector identifying the time period analysed. The dependent variable CElevel takes values between 0 and 4, such that 0 represents the absence of a code of ethics and 4 represents an advanced ethical code, with the highest rating. Accordingly, the dependent variable is an ordinal value. Each value of ‘‘CElevel’’ generates a continuous evaluation of the company, which is incorporated into an unobserved latent variable, which we term ‘‘CElevel*’’. This variable has a linear shape and is dependent on the same independent and control variables: CElevel∗it = ˇ0+ ˇ1%INDEPit−1+ ˇ2DINV PROTECit + ˇ3%FEMit + ˇ4BOARDSIZEit + ˇ5BOARDSIZE2it + ˇ6BOARDACTit + ˇ7FIRMSIZEit + ˇ8Leverageit + ˇ9MTBit + ˇ10Sectorit + ˇn iYear + it + i(3) CElevel∗it = ˇ0+ ˇ1%INDEPit−1 + ˇ2%INDEP ∗ DINV PROTECit−1 + ˇ3DINV PROTECit + ˇ4%FEMit + ˇ5BOARDSIZEit + ˇ6BOARDSIZE2it + ˇ7BOARDACTit + ˇ8FIRMSIZEit + ˇ9Leverageit + ˇ10MTBit + ˇ11Sectorit + ˇn iYear + it + i(4) As there are a limited number of categories of ‘‘CElevel’’, this variable presents various cut-off points, delimiting each category, as follows: CElevel = ⎧ ⎪ ⎪ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎪ ⎪ ⎩ 0 if CElevel∗ < cut1 1 if cut1< CElevel∗ < cut2 2 if cut2< CElevel∗ < cut3 3 if cut3< CElevel∗ < cut4 4 if cut4< CElevel∗ Wooldridge (2002) proposed two approaches for estimating panel data models with an ordinal dependent variable. The one that is most commonly used assumes that the it and ierrors are normally distributed and is estimated by maximum likelihood. This is the approach implemented in STATA by Rabe-Hesketh et al. (2001) and improved by Frechette (2001a, 2001b). The programme estimates a probit model with random effects. These models are widely used in analyses of rating agencies’ classifications (Afonso et al., 2007). Empirical analysis Basic estimations Table 4 summarizes the bivariate correlations for the variables considered in the analysis. The most significant relationships with the dependent variables are those for firm size, the size and degree of activity of the board, the presence of independent directors, and the proportion of female directors. Table 5 summarizes the results obtained from the two analytic models proposed: Model 1 includes the proxy for independent directors overall, while Model 2 also includes the interaction of this variable with the level of investor protection.