BOARD OF DIRECTORS SPECIFICITIES IN THE CONTEXT OF PORTUGUESE CORPORATE GOVERNANCE AND CORPORATE LAW

How to cite this paper: Vieira, E., & Neiva, J. (2019). Board of directors context of specificities in the Portuguese corporate governance and corporate law. Corporate Law & Governance Review, 1(1), 41 57. http://doi.org/10.22495/clgrv1i1p5 Copyright © 2019 The Authors a This work is licensed under 4.0 Attribution Creative Commons International License (CC BY 4.0). https://creativecommons.org/licens


INTRODUCTION
The development of securities markets has led to a strong debate on the structure and control of listed companies. This issue is related to Corporate Governance (CG), which is a universal question of achieving mechanisms to protect investors in all international markets.
Considering the effects of economic and financial globalization, the development of capital markets, the internationalization of companies, the evolution of information and communication technologies and the introduction of the Euro, this problem can no longer be ignored in Portugal, as well as around the world. Moreover, the financial scandals of some international firms raise a number of issues related with CG, the audit committee and the role of the board on firm's performance that are listed on capital markets (Soltani & Maupetit, 2015).
The CG concept is complex and comprises several dimensions (Silva, 2009). However, although there is a wide range of definitions of the concept of CG, they all emphasize the need for alignment between managers, auditors and shareholders.
The CG system is the set of principles and rules that the companies must implement in their activity.
It is characterized by the inclusion of rules which aim is to make the administration of the company transparent, defining the responsibility of its members, ensuring that the composition of the board of directors and their decisions preserves the interests of the different stakeholders, such as shareholders, lenders and employees. According to the Organisation for Economic Co-operation and Development (OECD), the CG is a system through which organizations are directed and controlled. Its specific structure, the distribution of the rights and responsibilities of the different participants of the companies (the managers, directors, of board stakeholders) dictate the and other shareholders decision the for procedures and rules -making process. For Shleifer and Vishny (1997), "Corporate Governance deals with the way in which suppliers of finance to corporations assure themselves of getting a return on their investment". Denis and McConnell (2003, p. 2) define CG as a "set of mechanismsboth institutional and market based -that induce the self-interested controllers of a company (those that make decisions regarding how the company will be operated) to make decisions that maximize the value of the company to its owners (the suppliers of capital)". Silveira (2002) identifies the CG as a set of restrictions and limitations implemented by the shareholders to managers, with the purpose of guiding the application of the resources. In the same line, Leal et al. (2002) states that CG is based on a set of practices and formal processes to supervise the executive management of a company, which aim to preserve the interests of the shareholders and minimize the conflicts of interest between them and the other stakeholders. Thus, the CG intends to be a set of practices to reduce the conflicts between managers and shareholders.
The CG can also be assumed as a model that favours an environment of confidence, ethics and morality, by giving the board of directors the mission of protecting the interests of shareholders and, at the same time, maximizing the companies value (Crowther & Sefi, 2011). It can also be understood as a system whereby companies are managed and supervised in a legal perspective, essentially as regards the mediation between ownership and management of companies (Vaz, 2013), in order to reduce agency costs (Fernandes, 2014).
The Portuguese Securities Market Commission -Comissão do Mercado de Valores Mobiliários (CMVM), which regulates and supervises the market stock exchange in Portugal, characterizes the CG as a system of rules of conduct regarding the exercise of management and control of shareholders (CMVM, 2013b). It comprises the set of structures of authority and supervision of the exercise of that authority, internal and external, with the purpose of ensuring that the companies determine and carry out activities consistent with the purposes for which it was created, without compromising the social responsibilities underlying to its existence (Silva et al., 2006).
According to the Portuguese Corporate Governance Institute -Instituto Português de Corporate Governance (IPCG, 2018), the CG should promote and enhance corporate performance, as well as the capital market, and strengthen the confidence of investors, workers and the public in the quality and transparency of management and oversight, as well as in the sustainable development of societies.
Nowadays, in the context of the financial markets situation, which tend to strength the separation between who controls the organizations and who has an interest in the success of them, CG is assumed as a system of management and control of companies that seeks to balance the relationships between management and shareholders (Santos, 2009).
Summarising, the CG is a relevant component of sustainable economic growth at the international level, aimed at reducing the vulnerability of the system, and providing incentives for efficient and socially responsible investments (Monks & Minow, 2008).
The genesis of the CG is attributed to Berle and Means (1932), in the consequence of the great crisis of 1929. This work was a reference to the American legislation that has been approved, such as the Securities Act of 1933 and the Securities Exchange Act of 1934, both still in force in the United States of America (USA).
The CG developments of the last decades have responded to companies' crises, seeking to restore transparency and confidence in the markets. The Cadbury Report (1992) reacted to scandals of British companies, such as BCCI and the Mirror Group. The OECD principles were response to the Asian crisis of 1997/1998, and the Sarbanes-Oxley Act and its regulatory development by the Securities and Exchange Commission (SEC) emerged in the aftermath of Enron and Worldcom, among others.
In what regards the Portuguese market, CG issues have a relevant role, since Portugal is a civillaw country, with a weak legal protection of investors, a high concentration of shares ownership, and a limited information transparency (La Porta et al., 1998). In addition, the crisis in the banking and financial systems, as well as the related consequences in the Portuguese capital market, have raised awareness about the CG issues, especially with regard to accountability, business with related parties and the composition and functioning of the management body (Resende, 2017).
Consequently, the CG mechanisms have been improved over time, in order to enhance corporate performance, as well as to protect the interests of all the stakeholders of firms, such as employees, creditors and investors.
The main legal regulations associated with CG in Portugal are the following ones:  Commercial Companies Code -Código das Sociedades Comerciais (CSC), which includes information about investors rights and obligations, firms control and management, managers and members of other committees, and protection of minority investors, among other aspects;  CMVM, that was responsible, from 1999 to 2017, for the regulation on CG. The first code of CG was approved by CMVM in October 1999 (CMVM, 1999), considering a set of recommendations and rules of conduct to be followed in the exercise of the management of companies issuing shares admitted to trading on a regulated market. However, companies whose shares are not admitted to trading on a regulated market are also encouraged to adopt these recommendations;  IPCG, which create a code of CG in 2015, which was changed in the meantime. The new CG code from IPCG has entered into force on January 1, 2018, substituting the CMVM code.
As the recommendations have evolved over time, the level of compliance on the part of Portuguese companies has also been growing.
The CMVM on CG 2 recommendations focus on the following subjects:  disclosure of information;  the exercise of voting rights and representation rights by shareholders;  corporate rules;  board of directors;  institutional investors. The IPCG code includes principles and recommendations regarding several issues, such as:  conflicts of interest;  related parties' transactions;  the role of independent managers;  diversity in the composition of corporate bodies;  risk management;  supervisory functions. According to the European Confederation of Directors Associations (ecoDa, 2015), adopting CG best practices can lead to the following benefits:  improve the access to external financing;  decrease the cost of capital;  improve operational performance;  increase company valuation and improved share performance;  improve company reputation;  reduce risk of corporate crises and scandals.
Although the CG issues are important for all the companies, we will focus our analysis on Portuguese firms with shares listed on the Euronext Lisbon.
The rest of the paper is organised as follows. Section 2 presents the legal framework of CG, including the models applied in Portugal. Next section exposes the corporate board of directors' practices, considering the different types of board members, the role of women on board, the remuneration of the board members, the role of the chairperson, the duality between CEO and chair and the board committees. Section 4 analyse the effects of CG on firm's performance. Finally, section 5 concludes the study.

LEGAL FRAMEWORK OF CORPORATE GOVERNANCE
Legal systems are divided into two main groups: civil law and common law. The civil law system is characterized by legal principles that are organized through comprehensive statutes and codes. It is predominant in the European continent and in countries with European influence. In countries with a civil law tradition, such as Portugal, investor protection is lower, with a negative correlation between ownership concentration and investor protection (La . In this context, the CG is mainly defined by the power of financial institutions and the high concentration of business control in large families. The Portuguese capital market is not expressive in terms of dimension and has low liquidity, which results in as excessive concentration of ownership and dependence on banking (Costa & Santos, 2011).
In countries with a common law system, the law relies mainly on court decisions, and not on legislative or executive acts typical of civil law countries. This approach is typical of countries such as the USA, Canada, the United Kingdom (UK) and other countries influenced by the previous ones (Costa & Santos, 2011). In this case, institutional investors hold a significant percentage of shares, so, the individual position of shareholders in each company has little relevance. Here, the capital markets tend to present high liquidity, being characterized by a market-based system, and not a bank-based system. Djankov et al. (2008) argue that countries with a legal system based on common law have a more effective system of legal protection, compared to civil law, and are more demanding in what concerns the information disclosure system. Cunha and Rodrigues (2018) analyse the  determinants of the level of CG disclosure by  Portuguese firms listed on Euronext Lisbon between  2005 and 2011. The results suggest that foreign  investor ownership, board size, board independence,  external  audit  quality  and  degree  of  internationalization had a significant and positive  influence on the CG disclosure level, whereas  ownership concentration, unitary leadership structure and debt had a significant and negative influence on corporate governance disclosure.
These two legal currents give expression to two major systems, the Continental system and the Anglo-Saxon system (Batista, 2009). Companies from Continental Europe and Japan are aligned with the Continental system and are based on the civil law legal system, whereas the USA, the UK and others of Anglo-Saxon inspiration, are regulated by the common law system (Silva et al., 2006;Costa & Santos, 2011).
CG has always been a central concern of the CMVM, which follows the issuance of the OECD principles on CG. In this sequence, Portugal published, for the first time, in 1999, a set of recommendations, principles and guides, through the CMVM (1999). Although this is a non-binding code, it is seen as a CG code, and, indeed, in 2007, the document changes its name from "Recommendations of the CMVM on Corporate Governance of Listed Companies" to "Code on the Governance of CMVM Companies". In 2013, result of the awareness of the absolute relevance of the theme, the regulatory framework that definitively opened space for private autonomy came to fruition, making possible the effective adoption of CG codes other than that issued by the CMVM. This situation, allied with the contribution of soft law (rules that are not legally binding), were relevant for the development of CG in Portugal. In this context, the IPCG published its own code on CG in 2015.
In what concerns the recommendatory strand, it is now allowed to use a CG code besides the CMVM code (article 2/1). Although the CMVM continues to afford a governance code that encourages the implementation of best corporate practices, the decision of choosing the code is left exclusively to the companies, and no longer to the CMVM. However, the firms must justify and explain carefully their choice. In the context of the "comply or explain" principle, the companies shall explain the recommendation compliance and justify the reasons for the non-compliance (CMVM, 2013b).
In March 2016, the CMVM and the IPCG announced a successor to both codes to be published by IPCG and implemented jointly. IPCG will monitor the new code and review its implementation and content, while CMVM enforces the legal compliance and collaborates in any revisions to the code.
Until recently, the main regularity sources on CG in Portugal are the CMVM and the IPCG. The CMVM is a public agency, and published the more recent documents on CG in 2013, the "Code on the Governance of CMVM Companies" and the CMVM Regulation No. 4/2013. The IPCG, a private association that promotes the best practices in CG, published the last version of his own code on CG in 2018 (IPCG, 2018).
However, the existence of two regulatory sources of CG, that set different rules, caused some problems, namely because the Portuguese capital market is narrow in scope (Resende, 2017). Consequently, and following the drawbacks of the existence of two distinct codes (CMVM and IPCG) in a small capital market such as the Portuguese, a single CG code was developed by the IPCG, which is a successor to the two existing codes. The main novelty of the new code is that it incorporates the transition to a model of self-regulation and constitutes a complement to the legal order and a guide to good CG practices.
In this context, on 13 October 2017, the President of the CMVM, Gabriela Figueiredo Dias, and the President of the IPCG, António Gomes Mota, signed a protocol, which establishes the principles of cooperation between both entities in the framework of the entry into force of the new CG code of the IPCG from January 2018, replacing the CMVM code. The model of self-regulation responds to a request from market agents and follows the trend in most developed capital markets. Consequently, the code of CG of the CMVM (2013a) was repealed on December 31, 2017, and the code of CG of the IPCG has entered into force on January 1, 2018.
The new code includes principles and recommendations regarding, for example, conflicts of interest, related parties' transactions, the role of independent managers, diversity (in particular gender) in the composition of corporate bodies, risk management and supervisory functions.
The CMVM maintains three large responsibilities:  ensure that companies choose a cg code;  ensure that companies make the disclosure of the cg report, which they are obliged according to law, a detailed report on the structure and practices of cg;  ensure that the whole comply is in agreement with the requirements of the securities code, this is, the information must be complete, true, up-to-date, clear, objective and lawful.
On the other hand, the IPCG assumes to define, in conjunction with issuers and other relevant entities, a model and structure for the monitoring of the code of CG. The Institute must promote, in biennial cycles and with the cooperation of the CMVM, the updates that may prove necessary, in line with the international best practices.
In sum, the CMVM remains responsible for the "hard law", which implies the sanctioning part, and the IPCG has "soft law", that is, it monitors the recommendations, but without the sanctioning associated.
In February 23, 2018, a protocol was concluded between the IPCG and the Association of Issuers of Securities Listed on Market -Associação de Empresas Emitentes de Valores Cotados em Mercado (AEM). In this protocol, the IPCG agrees with the AEM the basis and terms of the monitoring and follow-up of the code, in order to incorporate the reality verified in the monitoring, changes in the law and the international dynamics of evolution of best practices of corporate governance.
On Silva (2017) states that Portuguese listed companies have a concentrated structure, belonging mainly to families. According to the OECD (2017, p. 13), "a key feature of the Portuguese listed firms is the dominance of controlling (often family) shareholders. In 24 out of 53 listed companies, a single shareholder owns a majority stake".
Filho and Alves (2018) characterized Portugal as a country with low investor security environment, fragility of dispute settlement mechanisms, concentrated ownership structures, and by capital markets still insufficient to pressure corporate behaviour.

Models of corporate governance
The process of organizational structures of management and supervision varies according to the legal and institutional framework of each country (Rodrigues, 2014). Portugal has a hybrid system, that allows three options and provide an additional statutory body mainly for audit purposes. The Portuguese law, CSC, Article 278, allows companies 6 to choose one of three corporate governance models: 1. Board of Directors and a supervisory board 7 : Latin Model 8 ; 2. Board of Directors, including an audit committee and an auditor 9 : Anglo-Saxon Model; 3. Executive management body, general and supervisory board and official chartered accountant: Two-Tier Model.
In Portugal, all three models comprise two boards (a board of directors and a supervisory board), although subject to different rules.
The difference of management structures in each model can be seen in Figure 1. The board of directors is in charge of management, which has authority over all matters related to the management of the company which are not specifically set forth as falling under the competence of the shareholders' meeting. It may also delegate to one or more directors or to an executive committee the day-to-day management of the company.
The board of directors is a corporate body, so decisions are made in a specific meeting, after analysis, discussion and voting. For the board to validate a decision, most of its members must be present at the meeting. Resolutions are taken by the majority of the votes of the directors.
Although the board of directors carries out the companies' management, at least three elements form an audit committee 10 .
According to Resende (2017), this model has the advantage of providing effective control of management decisions, since supervisors have privileged access to information related to the resolutions to be taken by the board of directors, as well as to the financial situation of companies. The overlapping of management and supervision functions of the audit committee members may lead to the approval of management decisions over the supervisory function; thus, the management and a 9 The auditor can be an official chartered accountant or an official chartered accountant company. 10 The directors that form the audit committee may not have executive powers, may participate in the board of directors' resolutions, may participate in the executive committees' meetings, and shall have fixed remuneration (Resende, 2017).
single corporate body will perform supervisory functions.
The main role of the statutory auditor consists of examining the financial reporting. In this model, the supervisory board and the statutory auditor are responsible for the supervisory activities.
On average, the supervisory board of companies that adopted the Latin model met 6.2 times during the year of 2013, and 6.7 times in 2014. About 96% and 90% of the members of the supervisory board were considered independent, respectively in 2014 and 2013 (CMVM, 2014).
Concerning the legal framework applicable to directors and the executive committee, we want to emphasize that (Resende, 2017):  the appointment of an executive committee, as well as the scope or delegation of its powers to one or more directors is at the discretion of the board of directors, and shall not be decided by the general meeting or determined by shareholders' agreement;  revocation of an appointment or changes to the terms of delegation of powers is the sole discretion of the board of directors, which may make such changes at any time;  the delegation of decision-making power to an executive committee, it does not inhibit or limit the ability of the board to make binding decisions. This is the model most adopted by Portuguese

Anglo-Saxon model
As in the previous model, in the Anglo-Saxon model, the board of directors is responsible for managing the company, but it can delegate some powers on an executive committee, the management board. In this model, the responsibility for the supervisory activities belongs to the audit committee and the statutory auditor.
In this model, the administration or management of the company is carried out by a board of directors. However, some members of the board of directors (at least three) form an audit committee. The authority of the audit committee is similar to that of the supervisory board. This model involves the appointment of an executive committee, made by the board of directors, where members of the audit committee also vote. Revocation or alteration of the terms of the undertaking may occur at any time.
The coincidence of the management and supervisory functions of the audit committee members may lead to the approval of management decisions, to the detriment of the supervisory function, whereby the management and supervisory functions will be performed by a single corporate body.
The Anglo-Saxon model has the advantage of providing more effective control of administrative decisions, since supervisors have information on the resolutions to be taken by the board of directors as well as information on the financial situation of the company. The directors that make up the audit committee may not have executive powers; will have fixed remuneration; may not be waive, except with just cause and may participate in the board of directors' resolutions, as well as in the executive committees' meetings (Resende, 2017).

Two-Tier Model
In the two-tier model, the management of the company is performed by an executive board of directors, which has the power of representation of the company. The executive board of directors may be appointed and dismissed by the general and supervisory board or by the general meeting, in accordance with the company's articles of association.
Although a decision such as the appointment and dismissal of the members of the executive board members may be the responsibility of the general meeting or the general and supervisory board, the powers of the executive board are legally established and are not subject to delegation.
Decisions of the executive board of directors or of the general and supervisory board must be taken by a majority of the votes and, for their validity, a majority of their members must be present at the meetings.
According to the previous evidence that the Latin Model is the most frequent in Portuguese companies, in 2014, about 72.1% of Portuguese companies adopted the Latin model, 25.6% the Anglo-Saxon model, and only 2.3% opted for the twotier model (CMVM, 2014). As we can see in Figure 2, the Latin model is so far the most adopted model (with always more than 70% of firms adopting it). Some examples are Galp Energia (oil and gas sector), Sonae SGPS (food industry and distribution sectors), NOS (telecommunications and entertainment sector), The Navigator Company (pulp and paper sector), Mota Engil (construction sector), Semapa (pulp and paper, concrete and aggregates sectors), and Espírito Santo Saúde (health sector). This model is followed by far by the Anglo-Saxon model, with no more than about 26% of companies adopting it. Large companies usually adopt this CG structure, such as REN (energy grid sector), BCP (banking sector), Altri (paper and pulp sector), CTT (postal services sector) and Jerónimo Martins (food industry and distribution 11 The last year of analysis (2014) in conditioned by the availability of data. sector). Finally, the two-tier model has a very low percentage of adoption by Portuguese companies. EDP (energy sector) is an example.

CORPORATE BOARD OF DIRECTORS' PRACTICES
As we said before, the Latin model prevails in most of the Portuguese companies.

Board of directors
The board of directors is a recognized group of people that supervise the activities of a company.
It is recommended by the CG Portuguese codes (CMVM, 2015; IPCG, 2018) that the board of directors should have a number of members that must ensure the optimization of the decisionprocess making, and that allows the companies to maximize their value and performance. On one hand, if it is too small, the capability to supervise the executive directors in not efficient. On the other hand, if it is too big, it may restrain the speed and efficacy of the decision-making process. It is also advised that the board of directors should include a sufficient number of non-executive directors, whose role is to monitor and assess the management of the company by the executive members of the board.
The formal evaluation of the board of directors is of the general meeting responsibility.
According to the Heidrick and Struggles (2014)

. Executive board members
The executive directors are the members of the board of directors to whom day-to-day management powers have been delegated, under article 407(1) of the CSC. The executive directors must be focused on the aim of achieving the mission and goals of companies, safeguarding the interests of all shareholders, independently of their shareholding structure.
It is supposed that executive board members are updated on matters of interest to the firms, as well as on corporate governance subjects. They must also guarantee the transparency and the professionalism of companies.

Non-executive board members
The non-executive directors are the members of the board of directors who are not part of the executive board or to whom the current management of the company has not been delegated, under article 407(3) of the CSC.
In order to be considered independent, the nonexecutive board member cannot be associated with interest groups in the company, nor be in a situation that could affect his exemption from analysis or decision.
The Portuguese corporate board structure includes non-executive directors, who are supposed to control management decisions in order to protect the shareholders' interests (Alves, 2011). In addition, non-executive directors contribute to the alignment of the interests between managers and shareholders (Gregory, 2002), which reduces the probability that managers act opportunistically and helps to reduce the agency costs.
Although non-executive directors do not perform management functions and, they play a fundamental role in mitigating the potential conflicts of interest between managers and shareholders. In addition, independent nonexecutive directors are a key element in defending minority shareholders against the risk of expropriation (Bertoni et al., 2014). This type of directors assumes various responsibilities, such as supervising management activity, giving advice, exercising the right of veto, or even dismissing the Chief Executive Officer (CEO) (Weisbach, 1988). In Portugal, non-executive directors are seen as advisors of the executive directors and as decisionmakers on matters where the power of decision has not been delegated.
Non-executive directors must be independent of management to effectively perform their duties and to avoid skewed judgments (Bertoni et al., 2014;Fuzi et al., 2016).
Usually, the number of independent members increases with the size of the company; when internal managers have more opportunity to extract their own benefits; or when the CEO has more influence on the board (Linck et al., 2008).
According to the CMVM recommendations, the non-executive members on the board of directors must include a sufficient number of independent members. When there is only one non-executive director, this member must also be independent. The new CG code (IPCG, 2018) recommends that a majority of the board members should be nonexecutive directors and at least one-third (minimum two), should be independent, in order to ensure that their action is effective, as well as that all shareholders must be protected, even the minority ones.
The empirical evidence in the Portuguese context about the benefits of non-executive directors is not consistent. Although Fernandes (2008) concludes that non-executive board members do not help to align the interests between managers and shareholders, Alves (2011) states that non-executive directors protect the interests of shareholders, by monitoring management decisions.
Concluding, the non-executive directors should adopt the role of monitoring and evaluating the executive directors, as well as to ensure that the principles of social responsibility and sustainability considered by the company are enforced. In addition, the independent non-executive directors also have the role of protecting all shareholders, namely ensuring that the interest of minority shareholders is not neglected to the benefit of the remaining shareholders' interests.
In 2014, 52% of the board of directors were executive members, and 48% were non-executive members, as we can see in Figure 4. Although the percentage of non-executive members on board is higher than that of executive members, three of the 43 companies had only executive members on board. The weight of nonexecutive members considered independent was on average 34%, with two companies presenting 100% of independent non-executive members.
In 12 companies, none of its non-executive members could be considered independent.

Independent members on board
According to the IPCG (2011), the size of the board depends on several factors, such as the shareholder structure, the free float level, the size of the company and the presence of institutional shareholders. Consequently, the board must include a number of non-executive members that ensure effective supervision and evaluation of executive members' activity (CMVM, 2013b; IPCG, 2011).
The presence of non-executive directors that are also independent on board is a recent phenomenon in Portugal, being more common in large companies and firms with dispersed capital (IPCG, 2011). The CMVM recommend a minimum of 25% of independent directors in the board in order to ensure an effective supervision and monitoring of the executive members on board.
Based on the agency theory assumptions, it is expected that independent directors influence positively the firms' performance, since they contribute to reduce the agency conflicts between large and minority shareholders (Anderson & Reeb, 2004); promote the interests of other stakeholders (Chen & Roberts, 2010)  However, the empirical studies lead to different conclusions. While some authors find a positive relation between independent members on board and performance (Gama & Rodrigues, 2013), others document a negative relationship (Shukeri et al., 2012), and others report no significant relation (Hermalin & Weisbash, 1991;Wintoki et al., 2012). Romano (2005) alerts for the true independence of these board members, since they are classified as independent, but their recruitment and selection may be done or influenced by personal contacts, or by management.

Role of women on board
Diversity throughout the companies makes business sense, namely because it is easier to understand and meet customer needs if the organisation reflects its customer base.
Diversity on board embraces knowledge and understanding of relevant diverse geographies, people and their backgrounds including gender, race, sexual orientation, religion, belief and age, as well as personality, culture, and work-style, not forgetting that board should be based on merit as well as complementing and expanding the skills, knowledge and experience of the board as a whole. Consequently, requirements for the new members of the board and supervisory bodies should take into consideration general diversity requirements, paying particular attention to gender diversity.
Portugal has introduced demanding quotas for female members in each administration and supervisory body of each company, and it cannot be less than 20% and 33% in 2018 and 2020, respectively (Law n.º 2/2017, from 1 st August).
There are different conclusions in what concerns the effect of women on board performance. There are some authors asserting that the presence of women on board influences positively the firm's performance ( Bart & McQueen, 2013). The arguments of these authors are based on the fact that women have better communication and listening skills, as well as better attendance records, are more likely to bring international diversity to the board, and are less overconfident than men, leading to an increase in the return on their financial decisions. Vieira (2018) examines the relationship between board of directors' characteristics and performance in Portuguese listed family firms between 2002 and 2013, finding that family firm performance is positively related with gender diversity.
However, Olsen and Cox (2001) argue that women on board worse firm performance, because they are more risk adverse and more prone to emotional conflicts than men are. Torchia et al. (2010) analyse the presence of women on corporate boards, considering a sample of 317 Norwegian firms. The results show that at least three women may be beneficial in terms of contribution to board strategic tasks.
Heidrick and Struggles (2014) survey found that 63% of directors of European listed companies believe that a diverse gender and nationality mix on board is relevant. According to the authors, in 2012, the share of women on boards of the largest Portuguese listed companies was 6%, compared to 13.7% in European Union (EU) 27, and, in 2013, there were a mean of 8% of women on board, and there were 30% of boards with no women directors, compared to 17% and 12% in Europe, respectively. This percentages contrast with the ones of the USA, where female represent 36% among directors, for the S&P 500 (Stuart, 2017).
In Portugal, in the year of 2014, from the 422 positions on board, only 40 were held by women (less than 10%), and in 19 companies the management body consisted of only men (CMVM, 2014). Table 1 shows some characteristics of the board of directors. The board of directors has, during the period from 2010 to 2014, a mean of about 10 members. The executive members on board vary between 42% (2012) and 55% (2011). The number of non-executive members is always higher than the executive's ones, with the exception of the year of 2011. The evolution of the proportion of non-executive directors may be related to the recommendations of the CMVM, which encourages the participation of these members on the board of directors. The proportion of independent directors on board has increased progressively, except in the period 2011-2012, reflecting greater compliance with the recommendations issued by the supervisory authorities. The percentage of women on board have increased consistently during the period considered, reflecting a higher degree of diversity in the board. Finally, the mean of annual meetings was between 12.3 and 13.

Remuneration of the board members
Usually, the responsibility to determine the remuneration policy for the members of the board is attributed to a remuneration committee, composed by independent directors. In the Portuguese context, according the CSC, article 399, the remuneration of the board members, which may include a fixed and a variable part, is related to the function that the member executes, as well as the financial and economic situation of the company.
Although the variable component of remuneration has some advantages, it has also potential disadvantages, because less ethical managers can manipulate the results in order to maximize this remuneration component over the short term. Although it seems that a large fixed salary is better than a large variable salary, Rodrigues (2014) and Nunes (2014) argue that a remuneration with a mostly fixed component may attract the less competent managers and lead the more competent to behave negligently for lack of interest and incentive.
In what concerns the financial and economic situation of the company, Hill (2009) argues that one of the causes of some companies' financial adversity was the discrepancy of remunerations between the directors, mainly for the executive directors and top management, who received huge salaries, and the other workers.
The CMVM recommends that the remuneration of members of the board should be structured in such a way as to permit that the interests of board members are in line with those of the company, and should be disclosed annually in individual terms. In addition, the members of the remuneration committee or equivalent should be independent members. The new CG code (IPCG, 2018) reinforces the CMVM recommendations.
The remuneration policies should include an explanation of how the total remuneration paid complies with the adopted remuneration policy, how it contributes to the long-term performance of the company, and how performance criteria were applied. Furthermore, while stopping short of recommendation to disclose CEO pay ratio (the ratio of CEO pay to average worker pay), annual changes in the executive remuneration, performance of the company and average remuneration in terms of the fulltime employees during the last five financial years should be presented together to allow comparison.
According to the EU Commission, companies should benefit from remuneration policies that stimulate longer-term value creation, and executive pay should be related to performance. To encourage shareholder engagement in their investee companies, the EU Commission introduces the concept of "say on pay".
The executive directors' compensation must reward their effort, time and competence put at the companies' service. In addition, it must assure that their interests are aligned with the ones of shareholders. Consequently, a part of the compensation is variable, in order to reflect the achievement of the firm's goals, as well as the performance of each director.
The non-executive directors' compensation must be based on a fixed amount and presence payments, but must not contain any variable component indexed to share prices that may set their interests in agreement with those of the executive directors. Ingolf et al. (2011) concluded that several ways of limiting remuneration could lead to undesirable consequences. If the limitation is to focus on total compensation in a contractual way, the company will have difficulties in maintaining the level of management and will face poor performance with inadequate compensation, which will result in low market value and loss of competitiveness in relation to other companies where there are no CEO salary restrictions.
According to Rodrigues (2014), remuneration is one of the tools that allow the alignment of interests between managers and shareholders. A performance-based compensation system facilitates not only the alignment of the interests, but also reduces the conflict between managers and investors, which increases the promise of managers to maximize value creation, since it will increase their compensations.
Analysing the Portuguese reality regarding the evolution of the highest average salaries of listed firms in Euronext Lisbon, considering the period from 2007 to 2011, the author points to the positive and significant relationship between total compensation and performance because this leads to the selection and maintenance of the managers with more capabilities and motivation. Ferreira (2013) analyses the determinants of the board members remuneration, considering a sample of 122 companies listed on Portugal and Spain, and having as reference the 2010 fiscal year, finding that the board remuneration is negatively influenced by the market return, and positively influenced by firm's social responsibility and size. The positive relationship between size and board remuneration is consistent with the results of other studies, such as Fernandes (2008), for the Portuguese market.
In 2014, all Portuguese listed companies, excepting one, had a remuneration committee or other similar body with the purpose of defining the remuneration of the board of directors. In most cases, this body was composed of three persons. The annual variable remuneration granted through bonus systems, bonuses and participation in company results is a common practice in 90.7% of companies (CMVM, 2014). Table 2 shows some characteristics associated with the remuneration policy of the board.  The remuneration of executive members is significantly higher than the one of non-executive members, as we can see by the low percentage of remuneration of non-executive members 12 . This difference can be justified by the remuneration structure, since the compensation of the executive directors depends largely on elements related to performance, whereas the remuneration of nonexecutive directors is normally based on fixed components.
Finally, we can see that the fixed component of remuneration has increased and the variable component has decreased during the 2010-2014 period 13 .

The role of the chairman
The role of the Chairman of the board is to head the board of directors. The Chairman is expected to promote and supervise the highest standards of CG within the board and the company. This figure is provided in the CSC (article 395).
The main roles of the Chairman are the following ones 14 :  promote and oversee the highest standards of corporate governance within the board and the company;  set an agenda for the board, namely focused on strategic matters, forward looking and evaluating and supervising companies' business;  maintain a proper process to ensure compliance with board policy on matters reserved to the board for consideration;  ensure that board members receive accurate, timely and clear information to enable them to monitor performance, make sound decisions and give appropriate advice to promote the success of the company;  manage board meetings, motivating the discussion of complex issues and ensure that all members' contributions are encouraged and valued;  maintain an effective and balanced team, and inspire active engagement by all members of the board;  create the conditions for overall board and individual director effectiveness, creating the opportunity for maintenance of the relevant skills and knowledge required to fulfil the director role on the board and its committees;  ensure that the board undertakes an annual evaluation of its own performance. 12 We want to emphasize the amount of information not available on the CMVM reports. 13 In the first three years (2010-2012), the total of the fixed and variable components does not sum 100%, as the difference is associated with other components, such as funds and pension plans. 14 https://www.bupa.com/corporate/about-us/corporate-governance.

CEO-Chairman duality
In continental Europe, there is a distinction between the role of the President of the board of directors (Chairman) and the role of the CEO, being a clear division of responsibilities between the Chairman and the CEO.
While the Chairman is responsible for the leadership of the board and for the effectiveness of the overall board, as well as individual directors, the CEO is in charge for the strategic plan's execution and the day-to-day management of the business, in line with the strategy approved by the board. The CEO suggests, implements and reports on the strategic direction of the companies as well as business strategies and initiatives. In addition, it takes lead responsibility for the relationships with the media regulators, governments, local communities, suppliers, customers and other stakeholders.
If the president and the CEO is the same person, it is necessary to guarantee that the conditions are met for non-executive director's work to be efficient and independent from the executive board. In addition, companies must clarify in their annual report how the different functions are coordinated.
CEO  Bank, 2005). In what concerns the American market, more than 51% of S&P 500 boards have a separate chair and CEO (Stuart, 2017).  argue that the companies that separate the functions of Chairman and CEO get, on average, better performance. The separation of these two positions is pointed out as an appropriate procedure to reduce the concentration of power and to avoid less correct practices on the part of those who control companies (Matos & Gois, 2013).

Board committees
In addition to the executive board, it is useful that firms have some committees that contribute to the compliance of CG principles. According to the ICGP (2018), companies must constitute specialized internal committees appropriate to their size and complexity, covering, separately or cumulatively, matters of CG, remuneration and performance evaluation, as well as nominations.
Some of committees, suggested by the code, are the following: 1

. Audit Committee
The Audit Committee has responsibility for monitoring the integrity of the companies' financial statements, since they have the needed expertise, and are independent. In addition, this committee is responsible for the effectiveness of the systems of internal control, as well as for monitoring the effectiveness, performance and impartiality of the internal and external auditors.
In Portugal, the duty to supervise the effectiveness of risk management systems is commonly attributed to audit committee, or, alternatively, to the supervisory board. None of the committee members is part of it for more than two consecutive mandates.
The audit committee of the 11 companies that adopted the Anglo-Saxon model 15 in 2014, were composed, on average, of three members, and met around 9.8 times during the year, and 11.1 times in the previous year (CMVM, 2014).
Related with external auditors, Silva et al. (2019) analyse the effect of corporate governance on external audit fees in Portugal and Spain. For the Spanish sample, the results show that the capital hold by the BD influence negatively external audit fees, suggesting that better CG practices decrease risk and, consequently, audit fees. On the other hand, for the Portuguese sample, the results do not evidence a significant effect of CG characteristics on external audit fees.

Risk Committee
The Risk Committee has responsibility for the leadership and oversight of risk across the companies, including the understanding of current risk exposures and future risk strategy, overall risk tolerance, risk management framework including risk policies, process and controls, and the promotion of a risk awareness culture throughout the companies.
The number of companies that report to have an internal risk control system decreased from 40 in 2013, to 39 in 2014 (CMVM, 2014).

Corporate Governance Committee
The corporate governance committee is one of the most influential standing committees of the board. This committee main role is to recruit new board members and to ensure that each board member is equipped with the proper tools and motivation to carry out his or her responsibilities.

Remuneration Committee
Portugal recommends an independent remuneration committee, but has no specific requirement or recommendation on nomination committees. According the CSC, article 399, the remuneration of the board members may have a fixed component and a variable one, which must be function of the financial and economic situation of the company.
According to Heidrick and Struggles (2014) survey, the average number of board committees in 2013, in Portugal, was 3.3, compared to 3.4 in Europe. In 2014, the number of committees ranged from one to nine. Approximately half of the companies reported one or two committees. The other companies created, on average, 2.6 committees. Of the 15 companies with only one committee, nine had an executive committee, four had a remuneration committee and one had a CG committee (CMVM, 2014).

CORPORATE GOVERNANCE AND FIRM PERFORMANCE
One of the main missions of the CG is to guarantee information that can support the decision-making process and contribute to the maximization of company's value (Campos, 2015). The assumption of good governance practices improves the performance of companies and the exchange capital markets efficiency, namely because it results in confidence of investors, employees and general public in the quality of management, supervision and the company's sustainability (IPCG, 2018). Antunes (2009) states that, since the mid-1980s, a significant percentage of the company's equity has come to be in the hands of institutional investors, among 23 investment funds, pension funds, credit institutions, insurance companies and investment firms. The author concludes that institutional investors appear as a solution to minimize agency costs and safeguard the rights of shareholders. Santos (2011) argues that investment funds, insurance companies and pension funds have been taking the place of reference shareholders of large companies, contradicting the idea that these investors are not active as shareholders. Esperança et al. (2011) state that the weight of the financial assets held by institutional investors in Portugal increased significantly in the last years, with pension funds taking the most important role, followed by the insurers and investment funds.
There are several studies analysing the relationship between CG and companies' performance, both abroad ( (Faria, 2013;Campos, 2015;Carvalho, 2017;Vieira, 2018). Usually, the authors find a positive relationship between CG practices and performance. However, there are few studies that describes some limitations to the results, namely because of the methodology and the proxies used to measure CG practices.
Other studies find different results, concerning the corporate governance characteristics. For example, Danoshana and Ravivathani (2019), analysing sample of institutions in Sri Lanka, find that board size and audit committee size have a positive impact on firm's performance, but meeting frequency has a negative relationship with performance. Ciftci et al. (2019) study the relationship between context, internal corporate governance and firm performance, analysing he case of Turkey, an exemplar of family capitalism, finding that more concentrated ownership lead firms to perform better. They find also evidence that increase in cross ownership did not influence market performance, but is negatively related with accounting performance. On the contrary, the results show that a higher proportion of family members on boards had no significant effect on performance.
Moreover, several studies conclude that wellgoverned European companies perform better than the poorly governed ones (Claessens & Yurtoglu 2012). Iqbal et al. (2019) analyse the financial performance and corporate governance in microfinance firms, considering the market of Asia, concluding that profitability and sustainability of this kind of firms improve with good governance practices and contrariwise that more profitable and sustainable microfinance firms have better governance systems.
In Portugal, because of the intensification of the globalization effect and its consequences, especially in relation to the higher demands on the ethical standards of managers, the last decade was characterised by a significant evolution of the CG practices (Gonçalves, 2007). However, the conclusions about the effect of CG on firms' performance are not coincident. While some authors find no relationship between CG and performance, others find a positive relationship between the two variables. Alves and Mendes (2003) analyse to what extent compliance with the recommendations contained in the CG code of CMVM would imply a better performance in relation to their noncompliance, and concluded that, overall, the impact would be insignificant. However, for some individual recommendations, namely on the structure and functioning of the board of directors, there is a significant impact, with the greatest influence coming from the recommendations on the executive committee.
Regarding compliance with the recommendations of the CMVM to include independent members on the board, the authors concluded that it is not significantly associated with abnormal returns. They conclude that independent managers have a negligible impact on the company's performance. Alves and Mendes find also evidence of a negative relationship between the performance and the recommendations of the CMVM code regarding company rules, as well as a negative correlation between dividend policy and performance.
Using a sample of Portuguese listed companies on Euronext Lisbon for the period between 2007 and 2009, Mota (2011) concludes that there is no relationship between the board members and the performance of companies. Consistent with this conclusion, Silva (2017) investigates the effect of non-executive directors in the performance of nonfinancial Portuguese listed companies for the period between 2012 and 2016, and finds no significant relationship between the two variables.
Carvalho (2017)  Recently, Vieira (2018) finds that the presence of independent members on board does not affect Portuguese firms' performance, suggesting that independent directors are not really performing their assigned function.
There is empirical evidence that companies with a higher level of good practices in CG show higher value creation than the companies that show less evolution at this level (Davis, 2005;Gonçalves, 2007;Faria, 2013;Sardinha, 2014).
Faria (2013) develops a study based on a sample of PSI-20 companies, for the period between 2006 and 2012, in order to analyse the evolution of CG practices in Portuguese companies, through the construction of an index of compliance for the best practices of CG. The author finds evidence that the adoption of good practices is positively related with the financial performance of the sample companies, during the period under analysis. The results suggest that the companies that adopt more CG practices create more value than the other companies did. Rodrigues (2012) find evidence that the board size has a positive and significant relationship with companies' performance. Campos (2015) analyses the relationship between the characteristics of the most widely used CG model in Portugal and the firms' performance, considering a sample of 48 companies listed on the Euronext Lisbon, for the period between 2011 and 2013. The main results of this study are the following ones: 1. The Portuguese companies tend to adopt the Latin model; 2. There is no significant relationship between the CG model and the company earnings; 3. There are differences regarding business performance, measured by the earnings before interest, taxes, depreciations and amortizations (EBITDA), due to the CG model adopted by companies, which is consistent with Ferreira (2013); 4. There is a positive and significant relationship between the board dimension and business performance. This is not supported by the theory, which postulates that there is no positive association between the board dimension and business performance ( 7. There is no significant relationship between the remuneration policy and business performance, which result was also reached by Rodrigues (2014); 8. There is a positive and statistically significant relationship between the CEO independence (boards split the chair and CEO roles between two individuals) and the company's performance, which is in agreement with the results of .
Summarising, the results of Campos (2015) show that there is a positive and significant association among the variables size of the board, the number of independent members on board, the existence of different people on board and CEO role of president, independence of the CEO, and firms' performance.
Particularly, family businesses are very common in Portugal, with about 56% of non-financial Portuguese firms listed on the Euronext Lisbon between 2002 and 2013 being family firms (Vieira, 2018). In this context, Martins (2017) analyses the corporate governance problems common to the Portuguese family businesses, concluding that Portuguese Company Law does not give special attention to family businesses. The author suggests that, since some rules are not mandatory, they may be used as a tool to avoid frequent problems in family businesses. Vieira (2018) finds that, although family firms have higher mean values than nonfamily businesses in what concerns profitability (measured by the return on assets and the market to book ratio), the mean differences are not statistically significant.
Concerning other markets, Martin-Reyna and Duran-Encalada (2012) study whether there are differences in performance between Mexican family and non-family firms, considering the particularities of the Mexican corporate governance system. The authors find that family firms adopt different corporate governance structures than non-family firms, suggesting that these differentials influence firm performance. Minichilli

CONCLUSION
This study gives an overview of the legal framework of CG in Portugal and the evolution of CG practices in the last years, being this analysis, however, conditioned by the availability of information.
The most widespread CG model is the Latin model. Indeed, in 2014, 72.1% of the Portuguese listed companies employ the Latin model, 25.6% use the Anglo-Saxon model, and only 2.3% adopt the two-tier model.
The members forming the Board of Directors belong to two main categories, which is executive and non-executive directors. Among the latter's, there is a share of independent directors. As we observed, the percentage of non-executive members is always higher than the executive's ones, with the exception of the year of 2011. In addition, the proportion of independent directors on board has increased progressively, except in the period 2011-2012, reflecting greater compliance with the recommendations issued by the supervisory authorities.
Referring to the gender diversity characterizing the board of directors, the percentage of women on board have increased consistently during the period considered, reflecting a higher degree of diversity in the board, namely from 5.9% in 2010 to 9.5% in 2014.
With regard to the directors' compensation, data show a decrease, through time, in the total remuneration of the board, being the remuneration of executive members significantly higher than the one of non-executive members.
The relevance of CG has been increasingly focused as the market recognizes the positive impact that CG practices have on the economic growth and on stability of financial markets. CG is related with controlling the business for what is vital to all organizations, regardless of size or structure. Thus, it is expected that CG makes a solid contribution to the stock market and the protection of stakeholders, through changing attitudes in the management of organizations, and following CG good practices (Vicente, 2014).
Heidrick and Struggles (2003) analysed the CG practices in Europe, and conclude that Portugal is significantly behind Europe. Despite this evidence, the country made the most progress. Regardless of the evolution of good practice in Portugal has been positive, there is a way yet to go.
The recent financial sector scandals in Portugal (Banco Português de Negócios -BPN, Banco Privado Português -BPP, Banco Espírito Santo -BES and Portugal Telecom), as well as the 2007 financial crisis, have shown that CG goes beyond corporate models and should have effects on the internal control structures and the adequacy of the people who are members of its management and supervisory bodies, and revealed the weaknesses of internal control, non-executive directors, supervisory bodies and external auditors. In addition, it highlights the need for more effective CG models, in order to avoid further scandals for notoriously ineffective governance and control mechanisms.
The CG did not avoid some manipulations, rather they disguised them, transmitting to the market a wrong feeling that everything would be well in the management of the big companies.
Sometimes, it is not enough to create recommendations and advice companies to comply, ignoring existing norms of positive law. If necessary, compliance must be enforced. This is a role for the regulatory and supervisory bodies, which cannot fail to set an example as regards good practices. As long as there is no adequate educational and cultural literacy, which will only happen with high standards of education, it is necessary to regulate by prevention, by imposing perceptive rules and by being responsible, especially for those who do not enforce them (and not only those who infringe them and are caught).
Maybe, changing the regime of governance for institutional investors, who have management capabilities or the granting of more voting rights, are measures that could be considered for a better longterm intervention in companies.
The literature review raises questions about the true effectiveness of some control mechanisms (Alves & Mendes, 2003), because of the existence of empirical evidence that supports the idea that some measures have the opposite effect on the performance of companies that adopted them. One way to mitigate these problems is to have independent members on board who are expected to question the managers' decisions and thus to put an end to their discretion.
Prior studies results on CG and firm performance are mixed and inconclusive, mostly because of differences in definitions, variables measurement, and analysis periods, which suggests that this subject needs further investigation (Shank et al., 2013;Steigenberger, 2014).
According to Silva et al. (2006), Portugal has a fairly comprehensive and up-to-date set of regulations.
Based on previous studies conclusions, it seems that the evolution of CG practices in Portugal is moving in the desired direction, namely in what respects to the following evidences:  growing concern about transparency, updating and availability of information supplied to the market;  growing concern about the alignment of the interests of managers with those of firms, namely through remuneration policies;  reasonable compliance with the recommendations for the creation of internal risk control;  increase of independent members on board;  reinforcement of committee independence, namely the increase of audit committees with nonexecutive members;  clearly identified the executive, nonexecutive directors, and the independent members.
From our point of view, this study has some limitations. First, we highlight the fact that official reports are outdated. For example, the most recent report of CMVM is from 2014, which conditioned the analysis and the conclusions obtained in this work. Second, the empirical studies done on this subject in Portugal are still scarce. Third, we cannot conclude about the effective role of CG on the stakeholders' protection, in general, and investors, in particular. Fourth, the women representation on the board of management is still rare. Finally, there is a gap between the CMVM rules and the IPCG code enter into force. However, there is still plenty of work to do on the governance front if Portugal wants to reach the higher European standards, but perhaps passing the baton from CMVM to IPCG will strengthen the CG environment in Portugal and allow for more rapid developments in the future.
In order to deal with the lack of data problem, we suggest a longitudinal study, with more recent information, to obtain conclusions that are more updated, and obtain tendencies of attitudes and effects. We can also make a study comparing the two Iberian markets, given their proximity and market characteristics.
Finally, we suppose it would be interesting to develop a robust model to measure the quality of CG that can be used in several countries, and whose results are comparable.