Abstract
In the backdrop of the convergence–divergence debate, the goal of this article is to examine the proliferation of corporate governance codes in the light of a single factor, namely varying corporate ownership structures across countries. While such codes emanated and became popular in the United Kingdom where companies largely display dispersed shareholding, the concept has been disseminated to countries that carry considerably different ownership structures, i.e. mainly concentrated shareholding. This is bound to give rise to incongruities in the implementation of these codes. In order to enunciate the claim made above, the article will consider two aspects that convergence advocates have focused on, namely (i) shareholder empowerment and (ii) self-regulation. For example, corporate governance codes place considerable emphasis on the structure and independence of the boards of directors of companies as a means to ensure shareholder protection. While this approach is meant to produce results in companies with dispersed shareholding, the same cannot be said of companies with concentrated shareholding where the empowerment of shareholders through director independence or other means would only embolden the already dominant controlling shareholders. Moving to self-regulation, a voluntary code operating on a ‘comply-or-explain’ basis can ensure sufficient adherence only if certain factors are present in the jurisdiction where it is applied. Relying upon available empirical evidence, this article finds that use of self-regulation in voluntary codes of corporate governance may generate different results depending upon the ownership structures of companies, thereby exhibiting signs of divergence on this count.
Keywords
Introduction
There cannot be a more opportune time to study the diffusion of corporate governance codes around the world. After all, a quarter of a century has elapsed since the Cadbury Committee Report was issued in the United Kingdom (UK), which accelerated the phenomenon by which several countries began adopting voluntary codes of corporate governance. Although there is a substantial variation in the nature and content of these codes, they are essentially treated as ‘soft law’ by virtue of the flexible and informal nature of compliance requirements that are often not accompanied by legal enforcement mechanisms. This stands in contrast with mandatory rules that are supplemented by penalties and other legal consequences for non-compliance.
The spread of corporate governance codes has been brandished as evidence of convergence not just of corporate governance rules (de jure) but also those of practices across countries (de facto) (Aguilera and Cuervo-Cazurra, 2004; Khanna et al., 2006), although the debate regarding the direction and degree of convergence has failed to abate (Cuomo et al., 2016). More important is the normative claim by theorists that globalization will drive corporate governance systems towards a common endpoint represented by the Anglo-American shareholder-centric model (Hansmann and Kraakman, 2001). Here, the goals of corporate governance norms would be to help attain shareholder primacy through shareholder protection and empowerment promoted through self-regulation. This idea has come under strain due to the path dependence theory that recognizes differences between corporate structures and national institutions, due to which systems are likely to diverge (Bebchuk and Roe, 1999; Hall and Soskice, 2001). Other factors leading to divergence include the quality of the legal system, the depth of the financial markets and even corporate culture (Lau et al., 2007).
In the backdrop of the convergence–divergence debate, the goal of this article is to examine the proliferation of corporate governance codes in the light of a single factor, namely varying corporate ownership structures across countries. While such codes emanated and became popular in the UK where companies largely display dispersed shareholding, the concept has been disseminated to countries that carry considerably different ownership structures, i.e. mainly concentrated shareholding. This is bound to give rise to incongruities in the implementation of these codes, as voluntary codes will be effective only in the presence of specific conditions. While these conditions are present in the UK, they do not operate in their entirety in other jurisdictions that have adopted codes. Moreover, while voluntary codes have largely emanated in the shadow of the agency problems between managers and shareholders (that are paradigmatic of governance concerns in the UK), it is unclear to what extent they would function effectively to address the agency problems between controlling shareholders and the minority (that are replete in several jurisdictions that have adopted and are seemingly implementing corporate governance codes).
In order to enunciate the claim made above, the article will consider two aspects that convergence advocates have focused on namely (i) shareholder empowerment and (ii) self-regulation. It will be argued that while these goals underpinning the convergence debate may be attainable in corporate governance systems like the UK and the United States (US) where companies display dispersed shareholding, the result is not so straightforward in most of the rest of the world where concentrated shareholding is the norm. To illustrate the arguments, the article will focus on the specific components of shareholder empowerment and self-regulation. For example, corporate governance codes place considerable emphasis on the structure and independence of the boards of directors of companies as a means to ensure shareholder protection. Nearly all, if not all, the codes issued thus far focus on director independence as the means by which a monitoring board can be stimulated to act in the interests of the shareholders. While this approach is meant to produce results in companies with dispersed shareholding, the same cannot be said of companies with concentrated shareholding where the empowerment of shareholders through director independence or other means would only embolden the already dominant controlling shareholders. In controlled companies, board independence measures need to be more nuanced.
Furthermore, in order to impose checks and balances on the actions of controlling shareholders, wider measures need to be deployed, especially to impede self-dealing transactions and abusive related party transactions and to prevent expropriation of minority shareholders through transactions such as squeeze outs. Curiously enough, such agency problems that require greater attention in controlled companies are largely untouched by codes of corporate governance. They are usually dealt with through ‘hard law’ in the form of corporate law rules and private enforcement mechanisms (such as derivative suits) conferred upon shareholders. Hence, shareholder empowerment and shareholder orientation engendered by corporate governance codes take on very different connotations in companies with varying ownership structures which, as the article argues, has not received the attention it deserves in the literature and policy debates.
Moving to self-regulation, a voluntary code operating on a ‘comply-or-explain’ basis can ensure sufficient adherence only if certain factors are present in the jurisdiction where it is applied. These include a legal tradition that accepts the power of ‘soft law’, a pool of institutional investors that plays an active monitoring role on the basis of disclosures made by issuer companies regarding their compliance (or explanations of non-compliance) and a robust legal system with strong foundations for protection of shareholder rights (Varottil, 2017). The presence of these conditions in the UK has naturally led to the popularity of corporate governance codes in that jurisdiction. However, the conditions that enable an effective functioning of voluntary codes may not be present in other jurisdictions, particularly in the emerging and developing economies. Moreover, given that codes have largely emanated from jurisdictions with dispersed shareholding where institutional investors have the necessary incentives to monitor levels of compliance, it is arguable whether the use of voluntary codes and associated compliance would witness similar results in countries with concentrated shareholding. Relying upon available empirical evidence, this article finds that the use of self-regulation in voluntary codes of corporate governance may generate different results depending upon the ownership structures of companies, thereby exhibiting signs of divergence on this count.
To be sure, this article does not delve into the question of outcomes with respect to the application of corporate governance codes in jurisdictions with dispersed shareholding nor does it positively evaluate codes under the dispersed model. It is necessary to recognize problems with corporate governance in the dispersed model, including excessive compensation arrangements for managers, insufficient monitoring systems and the several corporate scandals that emanated during and in the wake of the global financial crisis. The principal claim of this article, therefore, is the need for a different approach in case of controlled companies that does not include any qualitative assessment of codes in the dispersed shareholding model.
This article begins with a discussion of the emanation and diffusion of corporate governance codes around the world, focusing on the factors that propelled the phenomenon. It then goes on to examine the importance of divergent ownership structures in countries that have been recipients of codes. Other than the US and the UK, most jurisdictions witness concentrated shareholding, where business families and the state represent a common type of controlling shareholders. The article then delves into the two aspects of convergence. First, in considering shareholder empowerment, it finds that the concept operates differently in dispersed and controlled companies. The monitoring board, underpinned by the requirement for independent directors, may serve different functions in controlled companies. Second, the article explores the use of self-regulation as a means of enforcing code requirements in the context of varying ownership structures. Finally, it concludes with some normative observations and identifies areas for further research.
Proliferation of corporate governance codes as a means towards convergence
Evolution and diffusion of codes
Although the Cadbury report has popularized corporate governance codes internationally, it was not the first one to be devised. The credit for pioneering the concept of codes goes to the Business Roundtable in the US, which issued a report in 1978 titled ‘The Role and Composition of the Board of Directors of the Large Publicly Owned Corporation’ (Aguilera and Cuervo-Cazurra, 2004; Hermes et al., 2006). This was followed by sporadic activity by some regulators to establish codes until the Cadbury report that was commissioned in the wake of scandals that befell the British corporate sector in the 1980s (Aguilera and Cuervo-Cazurra, 2004). Despite the lack of imitation of the Cadbury report in its initial years, it subsequently attained the status of ‘Rolls Royce of corporate governance regulation’ (Nestor, 2008) and ‘went viral’ (Jordan, 2013). The Cadbury-inspired code was found to be useful as it was flexible, adaptable and belied the criticism meted out to ‘one-size-fits-all’ governance mechanisms.
The propagation of codes around the world was aided by a number of factors. Principal among them were the efforts of institutions such as the World Bank and the International Monetary Fund, which called upon countries to enhance their corporate governance mechanisms. Several international codes issued by institutions such as the Organisation for Economic Co-operation and Development (OECD) and the International Corporate Governance Network (ICGN) came up with their own sets of corporate governance principles that were widely adopted by countries (Davies and Schlitzer, 2008; Jordan, 2005). Several crises (global, regional and local) had their role to play in the race by countries to adopt codes. These include the Asian financial crisis of the late 1990s, the Enron corporate governance episode (followed by the enactment of the Sarbanes Oxley Act of 2002 in the US) and the global financial crisis of 2007–2008 (Jordan, 2013; Nowland, 2008; Zattoni and Cuomo, 2008). While governments of several countries engaged in crisis-driven reforms through the introduction of corporate governance codes, there was hardly any opposition to a voluntary ‘soft law’ based approach that would in fact preempt possibly harsh mandatory regulation (Jordan, 2005).
Cuomo et al. (2016) have collected information regarding the dissemination of corporate governance codes. Among individual countries, they find that ‘91 countries issued a code and that a total of 345 codes (91 first codes and 254 revisions) have been developed around the world by the end of 2014’. Moreover, at an international level, they find that ‘14 transnational institutions issued 21 corporate governance codes by the end of 2014’. Interestingly, there is no homogeneity in the nature and focus of the codes. In the context of divergence of ownership structures, codes have been issued for specific types of companies such as family owned companies and state-owned enterprises. While Colombia, Morocco and Switzerland have established specific codes for family owned companies (ECGI, 2017), nine countries have issued 11 codes for state-owned enterprises (Cuomo et al., 2016). In addition, the OECD too has issued its guidelines on corporate governance of state-owned enterprises (OECD, 2015). While the use of specific approaches to address governance issues in companies with concentrated shareholding is welcome, it is clear that the efforts pale in comparison with the total number of codes issued. Given that most countries around the world (to the exclusion of the US and the UK) have concentrated shareholding with a predominance of family owned companies and state-owned enterprises, the miniscule number of targeted codes that consider controlled companies suggests that they are clearly missing the mark.
Convergence through corporate governance codes?
Despite the lofty claims staked by the proponents of convergence, it is abundantly clear from available empirical evidence that dissemination of corporate governance codes has not led to any form of convergence towards the Anglo-American shareholder-centric model. Through their comprehensive review of previous country-level and firm-level studies on corporate governance codes, Cuomo et al. (2016) find that ‘these studies show that divergence prevails around the world and that the content of codes is not converging either in European or emerging countries’. Hence, the convergence story is somewhat nuanced since there is convergence on some counts towards the Anglo-Saxon model, while there is divergence on others (Cicon et al., 2012). This is arguably attributable to differences in legal systems, business environment and even culture. As a result, one witnesses that domestic forces are driving corporate governance systems that lead to significant deviations from a perceived common end point (Hermes et al., 2006).
Unsurprisingly, the lack of extensive convergence is in large part due to the variations in shareholding structures across countries. The relevance of UK-style corporate governance norms practices could be called into question in controlled companies (Bozec and Bozec, 2007). Some have argued that convergence or isomorphism in the backdrop of differing ownership structures may act as a problem rather than the solution (Lazarides and Drimpetas, 2010). The studies resonate closely with the structure-driven path dependence approach asserted by Bebchuk and Roe (1999) whereby interest group politics will prevent corporate structures of an economy from altering, due to which convergence to a common shareholding structure is more in the nature of a pipe dream.
In all, while corporate governance codes have become ubiquitous around the world, the evidence is mixed on whether that has led to convergence towards the shareholder-centric model displayed by Anglo-American corporate governance. If corporate governance codes have played any role at all in engendering homogeneity, they have only succeeded in bringing about de jure (or formal) convergence and not de facto (or functional) convergence (Gilson, 2001; Khanna et al., 2006). While there may be ostensible similarities in some of the substantive rules and systems, significant differences abound in the implementation thereof in terms of practices. The answer to this puzzle might lie in an examination of the motivating factors behind countries adopting corporate governance codes.
Motivating factors behind code diffusion
In their seminal work, Aguilera and Cuervo-Cazurra (2004) seek to provide two accounts for why countries would adopt corporate governance codes. The first is an ‘efficiency’ rationale by which endogenous or domestic factors propel countries to adopt best practices in corporate governance so that the efficiencies in the system can be improved by protecting the interests of various stakeholders. The second is ‘legitimation’ which suggests that increasing globalization and the mobility of foreign investors across jurisdictions, i.e. exogenous factors, would compel countries to establish corporate governance mechanisms that are widely accepted. Juxtaposing these accounts with the convergence debate, one may hypothesize that where countries are driven by forces of legitimation, there is a likelihood of convergence, but when they are motivated by efficiency reasons, one can expect more divergence. Interestingly, the empirical evidence surrounding the analysis of rationale for adoption of codes is rather mixed, thereby indicating the lack of a consensus towards convergence.
Aguilera and Cuervo-Cazurra (2004) find the existence of both ‘efficiency needs and legitimation pressures’ that have led to code adoption. Countries are motivated to adopt codes not only because they improve the corporate governance system by inducing efficiency, but also because they seek to obtain legitimacy by incorporating widely accepted norms regardless of their suitability or level of implementation. For example, the pressures of globalization and the urge to adhere to ‘international standards’ have led countries to adopt governance standards that are similar to those in countries from which they seek to attract foreign investment (Varottil, 2017). Adding to these pressures is the focus of multilateral institutions such as the International Monetary Fund, the World Bank and the OECD to promote global governance standards on a broader basis (Jordan, 2005). Such a mixed approach has been identified in other studies as well (Cuomo et al., 2016). For example, Hermes et al. (2006) find that codes are driven by both domestic and external factors and that differences in institutional settings across countries turn out to be crucial determinants of the manner in which codes are adopted and implemented. Similarly, Zattoni and Cuomo (2008) show that both efficiency and legitimation reasons help explain the diffusion of codes, which ‘support the idea that the characteristics of the national corporate governance system and law explain the main differences among the coverage of codes’.
Interestingly, in the context of the dispersion of corporate governance norms around the world, an OECD Business Advisory Group comprising business leaders including Sir Adrian Cadbury rejected a ‘“one-size-fits-all” approach to corporate governance practices’ and focused on ‘a set of general public policy perspectives and guiding norms in a context of pluralism and adaptability’ (OECD, 1999). Such advice appears to have been ignored by the international institutions propagating corporate governance codes around the world, thereby giving rise to the appearance of convergence, while the reality suggests otherwise.
A study of the rationale supporting the diffusion of codes therefore fails to identify a clear trend of convergence towards the Anglo-American model of corporate governance. If there is divergence even in terms of the nature and content of the codes, more differences are likely to emanate when one delves into the implementation of codes in the context of structural and institutional variations among countries. This leads to an examination of the significance of ownership structures and their impact on the diffusion of codes in light of the convergence–divergence debate.
The importance of ownership structures
Broad types of ownership structures
Apart from the US and the UK, most countries around the world are characterized by companies with concentrated shareholding, which is the more common corporate structure (La Porta et al., 1999). More particularly, this is true of continental Europe (Faccio and Lang, 2002), East Asia (Aronson, 2014; Claessens et al., 2000) and emerging economies such as Brazil, China and India (de Silviera and Saito, 2009: Rajagopalan and Zhang, 2008). Companies in these jurisdictions are epitomized by controlling shareholders, being either business families or the state, who exercise rights over a significant proportion of shares in the companies, with the remaining shareholders, i.e. the minority, being dispersed in nature. The controlling shareholders are able to exercise dominant control over the affairs of the company, including to appoint and remove directors on the board.
In several economies displaying concentrated ownership structures, there tends to be a misalignment between the cash flow or economic rights of the controlling shareholders and their control or voting rights. It is all too common for the controller’s positon to be embellished by mechanisms such as dual class share structures, pyramiding and tunneling, by which the controller is able to exercise greater control rights in comparison with economic rights (Lau et al., 2007; Shleifer and Vishny, 1997). Such a wedge between ownership and control exercised by the controlling shareholders creates corporate governance problems, as there could be a tendency to engage in tunneling and expropriation of minority shareholders.
At the same time, the existence of controlling shareholders is insufficient on its own to give rise to agency problems with minority shareholders. Concentrated shareholding can create potential solutions to agency problems that exist in companies with dispersed shareholding so long as the controlled structures are efficient and are ably supported by an effective legal system (Gilson, 2006). Here, the legal system plays an important role of ensuring that the controlling shareholders are able to monitor managers in a manner that is beneficial to all shareholders.
In any event, controlled companies are less likely to be effectively subjected to market-based regulation. For example, there is a complete lack of a market for corporate control. Due to the strong position and influence of the controlling shareholders, it is nearly impossible for a hostile acquirer to succeed in a takeover offer without obtaining the concurrence of the controller. This considerably dilutes the market for corporate control as a tool to enhance the governance of companies. Hence, the corporate governance issues in controlled companies need to be examined through an altogether different lens.
Agency problems analysis
The agency theory has been a dominant tool applied to analyse the proliferation of corporate governance codes (Cuomo et al., 2016). In countries with dispersed shareholding, i.e. the US and the UK where corporate governance codes emanated, the agency problems operate between the managers of corporations (agent) and the shareholders (principal) (Kraakman et al., 2017). However, in the recipient jurisdictions of codes, where concentrated shareholding is the norm, the agency problems are those between the controlling shareholders (agent) and minority shareholders (principal). In controlled companies, there is a lack of separation between ownership and management as controlling shareholders possess the power to hire and fire management, thereby reflecting an alignment of interests between the controlling shareholders and the managers (Lan and Varottil, 2015). Generally, constituents within the controlling shareholder group tend to be members of the management as insiders, thereby creating a dissonance between their interests and those of minority shareholders.
The mismatch between the agency problems in companies with dispersed and concentrated shareholding plays a significant role in analysing whether the dissemination of codes around the world is likely to be effective in attaining convergence, particularly towards the Anglo-American model. Significant doubts have been expressed about the efficacy of transposing Cadbury-style codes onto countries where companies predominantly display concentrated shareholding. Davies and Hopt (2013) argue that due to differences in ownership structures, corporate governance rules ought to be necessarily different, ‘although they may produce equally efficient equilibria’. They caution against the attempt to impose rules of one system on the other, as that would ‘likely increase production costs’. Similar arguments have been made sharply in the Asian context. Lau et al. (2007) argue that transplanting Anglo-Saxon best practice guidelines in the form of codes can be regarded as a ‘band-aid’ solution, given that ‘the codes and laws developed in Anglo-Saxon countries were intended to address country-specific structural and nonstructural issues, such as business and cultural norms, in those countries’ and that they ‘could therefore be unsuited to Asian business practices’. The observations of Banaji and Mody (2011) are even more striking: … one should note that Cadbury makes several assumptions. It assumes a corporate culture or system where there is already a widespread and well-established separation of ownership and control. Cadbury is not tailor-made to a context where dominant shareholders, e.g. promoters, control management and where the corporate governance problem is chiefly one of the protection of minority shareholder rights. The assumption of dispersed ownership explains why there is little emphasis in Cadbury on the equitable treatment of different groups.
Finally, although the dispersed–concentrated shareholding structure dichotomy is a useful one, it is unsatisfactory on its own. In the case of controlled companies, necessary regard must be had to the identity of the controlling shareholders, as the agency problems analysis may have to acquire another level of nuance in order to address those. For instance, in family-owned companies that are common in several parts of the world, corporate governance concerns may relate to long-term issues and multigenerational matters. Monitoring issues become somewhat complex, given that corporate governance issues become enmeshed into family governance matters (Baron and Lachenauer, 2015; DeMott, 2008). On the other hand, when the state is a controlling shareholder in companies such as state-owned enterprises, a unique set of issues may arise. Unlike a private controlling shareholder, the state is not a unitary actor, and different governmental bodies and agencies may display differing interests that may be difficult to reconcile (Kahan and Rock, 2011; Pargendler, 2012). Moreover, the state’s motivations tend to be different, as it may pursue transactions that enable the realization of political goals and public interest considerations (Milhaupt and Pargendler, 2019). Hence, apart from differences in ownership structures between dispersed and controlled companies, another layer of differentiation within concentrated ownership becomes crucial to the analysis.
With this background on the impact of ownership structures in the dissemination of corporate governance codes in the context of the convergence–divergence debate, it is apposite to develop these issues further by considering the two key features of corporate governance codes, namely (i) shareholder empowerment and (ii) self-regulation.
Shareholder empowerment and the monitoring board
Due to the origination of corporate governance codes from the Anglo-American perspective that bears a shareholder orientation, the codes focus on designing an optimal structure for the board of directors that can help resolve the manager–shareholder agency problems that are replete in companies with dispersed shareholding. But, when such codes are transplanted to jurisdictions with concentrated shareholding, they face challenges in addressing the controller–minority agency problems, as they were not designed for this in the first place. Here, it would be useful to begin with a broader agency problems analysis to address the dispersed–concentrated shareholding dichotomy and then examine the extent to which corporate governance codes in their current form are able to traverse across both. In doing so, emphasis will be placed on the phenomenon of board independence canvassed strongly by the code diffusion process. Thereafter, the limitations of codes in addressing the controller–minority agency problem will be discussed in detail, thereby raising questions of the ability of the code diffusion process to engender convergence towards a universal shareholder primacy model.
Shareholder empowerment in the context of divergent ownership structures
In accordance with the agency theory that has acquired dominance in the discourse, corporate governance codes place considerable focus on the composition and role of the board of directors of a company. In the dispersed shareholding context, efforts such as the Cadbury report and its transplants across the globe seek to use reforms surrounding board design to address the manager–shareholder agency problem. This is intended to enhance the accountability of managers towards shareholders as a whole (MacNeil and Li, 2006). Accordingly, codes of corporate governance stipulate norms relating to board composition (including the number of non-executive and independent directors), nomination and selection of directors, constitution of board committees, splitting the roles of the chief executive and chairperson, remuneration of directors, audit and information disclosures and similar matters relating to board practices (Cuomo et al., 2016; Hermes et al., 2006; Zattoni and Cuomo, 2010). Since the origination of codes in jurisdictions with dispersed shareholding structures focuses on addressing the manager–shareholder agency problem, their diffusion around the world to countries with concentrated shareholding structures (where different agency problems operate) is somewhat curious.
Several strategies may be adopted to enhance shareholder protection. Some, as discussed above, are targeted at the board of directors, while others may be targeted at shareholders. The measures relating to shareholders essentially involve aspects such as the power of shareholders to elect and remove directors and thereby have an influence over the composition and performance of the board of directors. Such forms of shareholder empowerment can be crucial in companies with dispersed shareholding because collective decision-making on the part of the shareholders and appropriate design of board structures would help minimize the manager–shareholder agency problem. However, these will arguably have little or no impact in addressing the controller–minority agency problem in controlled companies (Lan and Varottil, 2015).
The empowerment of shareholders in controlled companies as if it is a homogenous body of interests will do no more than to perpetuate the power and dominance of the controlling shareholders. If shareholders as a whole are conferred powers, the majority rule principle of corporate democracy will enable the controlling shareholders to exercise those powers, possibly to the detriment of minority shareholders (Bozec and Bozec, 2007). Moreover, barring certain exceptional situations, shareholders can exercise their powers in their own interests which, in some cases, could have a detrimental impact on the minority shareholders. As Bebchuk and Hamdani (2009) observe in relation to controlled companies: ‘measures that increase the board’s adherence to the majority shareholder’s preferences would exacerbate – rather than alleviate – the risk of controller opportunism’ and that ‘giving the majority of shareholders more power vis-à-vis the board would operate to weaken – not enhance – the protection of outside investors’. Hence, matters of board design and shareholder empowerment that are paradigmatic features of corporate governance codes may not be sufficient to address the controller–minority agency problem.
On the other hand, in controlled companies, it is the risk of self-dealing and expropriation by controlling shareholders that needs to be addressed, as such transactions tend to illegitimately transfer value from minority shareholders to the controlling shareholders who reap the private benefits of control. These include squeeze out transactions whereby the controlling shareholders may forcibly acquire the shares of the minority on terms that may not be favourable to the minority selling shareholders. In such cases, argue Lan and Varottil (2015), it is necessary to adopt a controlling strategy, whereby the law can moderate the powers of the controlling shareholders ‘such that checks and balances can be imposed on their private benefits of control, thereby minimizing the agency problem’. In other words, the focus on shareholder empowerment and board design is only a partial solution, if at all, to the issues encountered in controlled companies, where the conduct of controlling shareholders must receive legal attention. However, as discussed later, corporate governance codes do not address strategies aimed at controlling shareholders and hence fail to fill a big gap in the governance requirements of controlled companies. The controlling strategy is generally driven by mandatory corporate legislation and listing requirements and not by Cadbury-style voluntary codes of good governance, thereby exposing some limitations of corporate governance codes in the context of convergence across different ownership structures.
With this background, it would now be useful to first examine the most popular element of board design (aimed towards shareholder protection through the monitoring board) that has been propagated around the world through codes, namely the independent director, and then to consider the elements of the controlling strategy that receive minimal attention of codes.
Board independence: A key thrust of corporate governance codes
The spread of the concept of independence directors around Europe (Davies and Hopt, 2013) and indeed the rest of the world through corporate governance codes has been considered to be an indication of convergence towards the shareholder model. What is interesting is that all codes of corporate governance deal with principles of board composition and independence (Zattoni and Cuomo, 2008). In their survey through hand-collected data of 245 codes, Puchniak and Lan (2017) find that all of the codes had some provisions relating to independent directors, thereby demonstrating the ubiquity of the concept and its popularization through corporate governance codes. Even an economy like Japan that has hitherto demonstrated resistance to the concept of board independence incorporated the roles and responsibilities of independent directors in its round of significant corporate governance reforms in 2015 (Curtiss, 2017; Tokyo Stock Exchange, 2015).
Like corporate governance codes themselves, the theoretical underpinnings of the monitoring board and the concept of independent directors emanate from the dispersed shareholding context. The history of the independent director can be traced to the 1970s in the US. In addition to the advisory roles that boards were already performing, a need was felt to induce a monitoring role. The idea caught on in the UK with the Cadbury Code and the subsequent Combined Code on corporate governance. Such a concept steeped in the Anglo-American context then spread to other countries in the form of codes based on the principle of ‘comply-or-explain’. What is more surprising is that almost all of the recipient jurisdictions carry concentrated shareholding where independent directors’ role may have to significantly vary in light of the need to address the controller–minority agency problem (Davies and Hopt, 2013).
Puchniak and Lan (2017) analyse a simple, but important, implication of transplanting the independent director concept from the dispersed shareholding context to a concentrated shareholding one. That relates to the definition of an ‘independent director’. The historical American approach defined an independent director as one who is only independent from management. This resonates entirely with the manager–shareholder agency problem. The initial version in the UK arising from the Cadbury report too followed a similar approach. However, such a definition would make board independence totally ineffective in the context of controlled companies because an independent director may be independent of management but not the controlling shareholder. Hence, several countries (including the UK) modified the American definition of independent directors to require independence from controlling shareholders (in addition to management) so as to reflect the underlying realities in controlled companies that takes account of the agency problems between the controlling shareholders and the minority. Puchniak and Lan (2017) find in their study that a majority of the companies that have adopted codes (55.2%) define independent directors with reference to controlling shareholders as well. They then go on to discuss the curious case of Singapore where the American definition held sway until 2015 even though Singapore in a jurisdiction that is populated by companies with concentrated shareholding, which they argue ‘was the product of strategic regulatory design (not ignorance) and was surprisingly effective’.
Another significant aspect of independent directors that varies along ownership structures relates to the precise roles they are to perform, whether advisory, monitoring or otherwise. In controlled companies, independent directors may be required to play a specific role in protecting the interests of minority shareholders as a whole. While several codes prescribe the independent directors’ role in the context of dispersed shareholding, some codes such as in India impose a specific role on them to consider the impact of their decisions specifically on minority shareholders (Khanna and Varottil, 2017). Such codes directly embrace the need to address the controller–minority agency problem. Such situations include the independent directors’ oversight of related party transactions with a view to ensure that they do not lead to wealth tunneling by the controlling shareholders. While the UK and US-style codes seek board independence as a means towards shareholder protection, in other jurisdictions with controlled companies, there is a specific need to fine-tune their role towards minority shareholder interests as controlling shareholders are able to exert their influence through their control rights and do not require any additional protection.
Although adjustments have been made in some codes to expressly recognize the controller–minority agency problem, they are unlikely to function in the shadow of director election mechanisms that are usually prescribed by the corporate law of a jurisdiction rather than a voluntary code. For instance, independent directors are elected in the same manner as other directors, by which a majority of the shareholders are able to determine the substantial composition of the board including independent directors. Similarly, in several jurisdictions, shareholders (acting through a requisite majority) are in a position to remove directors, including independent directors. All of these situations make the director election and removal subject to capture by the controlling shareholders. If independent directors in controlled companies ought to act in a manner that takes into account the interests of the minority shareholders, they must not be beholden to the controlling shareholders. But, as the independent directors owe their appointment to, and operate in the shadow of a potential removal by, the controlling shareholders, considerable doubts emerge in the extent to which independent directors would be accountable to minority shareholders.
For this reason, there have been calls for a review of the nomination and election mechanisms for independent directors in countries with concentrated shareholding (Khanna and Varottil, 2017). It is necessary to ensure that the process of election and removal of independent directors is such that they weaken directors’ incentives to toe the line of the controlling shareholders, by effectively imposing checks on the ability of controlling shareholders to appoint and remove independent directors. In this vein, Bebchuk and Hamdani (2017) propose the election of some directors – ones they refer to as ‘enhanced independence directors’ – in a manner that eliminates the involvement of controlling shareholders, so that such directors can play a significant role in transactions involving conflict of interest of the controlling shareholders. Several alternative mechanisms have been proposed for the election of independent directors to minimize or eliminate the influence of controlling shareholders. These include voting by a ‘majority of the minority shareholders’ wherein the controlling shareholders abstain from voting, or other mechanisms such as cumulative voting. Similar protections need to be made available for reelections and renewals. Removal of independent directors too must be placed outside the hands of controlling shareholders by imposing a higher threshold (e.g. 75% majority) or, alternatively, by providing veto rights against removal to minority shareholders. All of these mechanisms would ensure that independent directors’ appointment, continuation and removal are placed outside the purview of the controlling shareholders, so that such directors can exercise their functions in the interests of minority shareholders, thereby addressing the controller–minority agency problem.
Given this scenario, it is crucial to note that voluntary codes of governance prescribe the requirement of independent directors and establish the roles and practices to be adopted by them. On their own, codes do not address the concerns of appointment and removal as articulated above, which are essential for independent directors to function effectively in controlled companies. At the same time, issues of appointment and removal of directors, including independent directors, are the subject matter of corporate law in different countries rather than codes. Transplanting the concept of independent directors through codes without introducing reforms to the underlying board mechanisms and processes contained in corporate law could arguably be an incomplete exercise. Hence, the diffusion of the independence director concept through codes suffers from significant limitations in bringing about convergence in the background of the dispersed–concentrated shareholding dichotomy.
Reining in controlling shareholders
Given the inadequacies of the Anglo-American voluntary codes of conduct, including through mechanisms such as board independence, to deal with the controller–minority agency problem, one may envision alternative strategies that may be more targeted. Referring to as the controlling strategy, Lan and Varottil (2015) discuss a system of greater monitoring of controlling shareholders, especially on matters such as self-dealing transactions and squeeze outs of minority shareholders, which would enure to the benefit of minority shareholders. Such a controlling strategy may be explored in two parts: ex ante and ex post. The ex ante mechanism seeks to address the controller–minority agency problem through substantive regulation. For example, in the context of related party transactions, this could include appropriate disclosures, their approval by a committee of independent directors or the approval of the ‘majority of the minority shareholders’. In certain jurisdictions such as Delaware, transactions such as squeeze outs are regulated by imposing fiduciary duties on controllers. The ex post mechanism includes remedies available to shareholders under corporate law, such as oppression actions brought by minority shareholders against the company or controlling shareholders as well as derivative actions brought by shareholders on behalf of the company against directors or other wrongdoers. Through both the ex ante and ex post mechanisms, the controlling strategy attacks the crucial part of the controller–minority agency problem. By curbing the power of the controlling shareholders, it curbs their incentives to indulge in value-reducing transactions that may confer them private benefits of control at the expense of the minority shareholders.
While the controlling strategy is arguably more suitable for controlled companies, the Anglo-American version of the corporate governance code transplanted to various countries does very little to further that strategy. With their single-minded focus on shareholder protection and empowerment in general, they have failed to play any role in inhibiting the powers of the controlling shareholders vis-à-vis the minority shareholders. At most, there is some evidence of ex ante regulation, often in the form of listing rules, which deals with how related party transactions or similar arrangements may be regulated. But, apart from that, most other measures, including the ex post shareholder remedies, are dealt with in the form of ‘hard law’ dispensed by corporate law rather than through voluntary codes. Although there is some formal convergence in terms of the contents of codes, the enforcement of shareholders rights through remedies available to them under corporate law varies across jurisdictions thereby indicating a great deal of divergence from a functional point of view.
In all, the concept of shareholder protection and empowerment (the mantra driving corporate governance codes) must be viewed with a different lens in companies with concentrated shareholding. The approach of dissemination of the concept through codes in a manner that is agnostic to corporate ownership structures is likely to bear less fruit. This is because shareholder empowerment per se only leads to the dominance of the controlling shareholders. By seeking to address the manager–shareholder agency problem, such shareholder-oriented voluntary codes may have the unintended effect of emboldening the controlling shareholders and thereby exacerbating the controller–minority agency problem. At most, a combination of shareholder empowerment and controlling strategies may have to be combined to generate an optimal outcome. To that extent, corporate governance codes perform a limited role in companies with concentrated shareholding. To have any effect through the controlling strategy, codes must be supplemented by ‘hard law’ mechanisms.
At the same time, the political economy literature in corporate governance would suggest that the texture of the law (e.g. soft or hard) depends upon the interplay of various interest groups that influence the corporate reform process in a given jurisdiction (Armour et al., 2011; Gourevitch and Shinn, 2005). The rules are reflective of the interest of the dominant groups, whether it be shareholders (controlling or minority), managers or employees (Pagano and Volpin, 2005; Roe and Vatiero, 2018). The influence of incumbent interest groups is also found in the path dependence theory whereby there is considerable resistance to changes in prevailing corporate ownership structures (dispersed or concentrated) (Bebchuk and Roe, 1999; Coffee, 2001). The role of political economy is a significant factor in the debate between corporate governance codes and mandatory legislation as codes tend to be acceptable to incumbent interest groups if they help ward off more stringent hard law.
In the context of companies with concentrated shareholding, the influence of the controlling shareholders in establishing the governance framework is palpable. This may explain the shape of corporate governance codes in those jurisdictions. While there is a great deal of enthusiasm in laying down norms that comport with globally acceptable standards that help attract international investors, there is far limited movement when it comes to governance practices in various jurisdictions. In that sense, corporate governance codes tend to bring about formal convergence that lacks a corresponding functional impact. The political economy of corporate governance also accounts for regulatory responses to scandals and crises. Quite often, populist demands compel governments and regulators to respond swiftly to the situation, but the responses are not necessarily effective as they are subject to interest group pressure to forestall more stringent regulation. Voluntary corporate governance codes are more acceptable to incumbents such as controlling shareholders than mandatory regulation. The political economy analysis buttresses the legitimation motive for the proliferation of corporate governance codes that disregards ownership structures and the influence of controlling shareholders.
After considering the shareholder-orientation of corporate governance codes whose origins can be traced to the Anglo-American concept, it would be useful to reflect upon the mode of implementation of such codes through self-regulation on a ‘comply-or-explain’ basis.
Self-regulation in the concentrated shareholding context
Voluntary codes of corporate governance have been popularized as they follow the ‘comply-or-explain’ approach that is enabling in nature and allows for flexibility in being able to encompass various types of companies, irrespective of size. It counteracts the rigidity of mandatory rules. However, voluntary codes are optimal only in the presence of several factors that have made it the common form of corporate governance regulation in countries such as the UK. The transplant of such voluntary codes to other countries, where similar conditions do not operate, raises questions about the viability of codes in those countries.
Here, it would be useful to consider the reasons why UK has epitomized the voluntary code of corporate governance and why other countries may find it difficult to emulate the extent of adherence that the UK has demonstrated with the code (Varottil, 2017). First, voluntary codes have represented the hallmark of regulating the corporate sector in the UK. For example, even prior to the emergence of the Cadbury Code, takeovers in the UK had been regulated through the City Code on Takeovers and Mergers that set forth a flexible mechanism to resolve matters relating to the market for corporate control, to be administered by an industry body in the form of the Takeover Panel (Kershaw, 2016). Second, the triumph of voluntary codes in the UK has been attributed to the presence of influential institutional investors who have demonstrated a preference for ‘soft law’ rather than government regulation of corporate governance matters. Third, such institutional investors have played an active role in monitoring disclosures made by companies under the ‘comply-or-explain’ approach, thereby enabling a market-based outlook. For this reason, companies are likely to pay closer attention to compliance requirements, failing which they have to explain the reasons for non-compliance. Fourth, such a market-based approach would function effectively only in the presence of a robust legal system with sophisticated legal and supportive institutions. These include professionals such as independent directors, auditors, secretarial and compliance specialists. Although these conditions have enabled the use of voluntary codes in the UK, concerns have been raised either regarding the levels of compliance or the quality of the explanations provided for non-compliance (Arcot et al., 2010; MacNeil and Li, 2006). Moreover, some have argued that there is a need to ensure greater regulatory oversight, including a monitoring mechanism by which compliance levels and explanations can be scrutinized (Keay, 2014).
These rather specific factors that led to the emergence and continuance of the voluntary code in the UK are arguably not present in other jurisdictions to which it has been transplanted (Varottil, 2017). First, unlike the UK that has relied strongly on informal mechanisms and traditions as means of compliance through ‘soft law’, several other countries depend entirely on mandatory rules in the form of ‘hard law’. Second, the institutional investors who were responsible for the UK voluntary code may not operate similarly in other jurisdictions. Third, the voluntary code is founded on the principle of monitoring by investors, which makes disclosures regarding compliance as well as explanations for non-compliance more meaningful. However, the type of activist investors present in the Anglo-American context may not operate to the same extent around the world, especially in controlled companies. Although the phenomenon of shareholder activism has taken shape in several countries, the influence of activist shareholders may be minimal in light of the dominance of controlling shareholders in companies with concentrated shareholding. For instance, even in Japan, which is considered to have the most dispersed shareholding structure (after the US and the UK) (Armour et al., 2011; Kraakman et al., 2017), the role of institutional investors has varied considerably in influencing the shape of corporate governance (Brodeur, 2017). Due to the contrasting results, commentators have raised significant doubts as to whether the Japanese corporate governance system will converge to that of the Anglo-American approach (Mizuno and Tabner, 2009). Finally, legal institutions in several countries may not be robust enough to stimulate an ethos of voluntary compliance. Hence, even in a country such as Japan that displays dispersed shareholding, the acceptability of corporate governance codes and the movement towards convergence is in doubt.
If the evidence regarding the adherence to voluntary codes in the UK is mixed, it is difficult to be optimistic about their accomplishments in countries with concentrated shareholding where the combination of factors that promote self-driven compliance do not exist in its entirety. Available empirical evidence bears out the pessimism in the utility of the self-regulatory mechanism in corporate governance codes as they apply to companies with concentrated shareholding. One study focusing on the impact of voluntary codes on companies with concentrated shareholding finds that ‘ownership concentration appears to play a larger role’ for compliance with Canadian corporate governance practices, ‘with increased voting power associated with lower compliance rates regardless of whether the blockholding represents executive, family, or outside investors’ (Anand et al., 2012). Moreover, there is evidence to suggest that disclosure and transparency, which forms the bedrock of the ‘comply-or-explain’ approach is not that robust in the concentrated shareholding environment. The findings of Arcot and Bruno (2009) that ‘companies with a large shareholder are less inclined to governance transparency suggests that voluntary regulation may have drawbacks in countries with large shareholders’. Studies have shown that companies with controlling shareholders are less likely to promote voluntary disclosure due to their ability to generate private information and benefits (Ajinkya et al., 2005). Probing further into the precise identity of the controlling shareholders, other studies have found a greater incidence of non-compliance with voluntary norms among companies owned by business families, founders or heirs (Anderson et al., 2009; Arcot et al., 2010; Chen et al., 2008). In that sense, there is a correlation between ownership structure and corporate governance practices (Bozec and Bozec, 2007).
In a similar vein, the implementation of corporate governance codes through the comply-or-explain approach has been subjected to different forms of empirical testing around the world, and they reveal significant problems and discrepancies, particularly in countries where concentrated shareholding is the norm. A number of studies have emanated from continental Europe, which identify deficiencies in the promotion of corporate governance (Cuervo, 2002). More specific country analyses of corporate governance codes point to a similar direction in the Netherlands (Akkermans et al., 2007; de Jong et al., 2005; Hooghiemstra and Van Ees, 2011), Germany (Seidl et al., 2013), Slovenia (Cankar et al., 2010), Poland (Campbell et al., 2009) and Greece (Nerantzidis, 2015). Hence, the use of voluntary codes in continental Europe seems rather complex and involves an interplay of several factors, including ownership structures, institutional considerations and even cultural factors, due to which ‘the compliance rate seems to convey rather mixed messages’ (Sergakis, 2015).
From the point of view of self-regulation, corporate governance codes are most suitable for a country like the UK with its general reliance on ‘soft law’ in certain corporate and commercial aspects, a dispersed shareholding structure with influential institutional investors and a robust legal regime with capable institutions that engender compliance with voluntary codes. Such codes are unlikely to function efficiently in the absence of this combination of factors, and hence it is clear that countries to which codes have been transplanted have enjoyed varying degrees of compliance with codes, as the empirical evidence discussed above demonstrates. Of these, there is a clear indication that ownership structures in the different jurisdictions have a considerable role to play in the utility of self-regulation, and companies with concentrated shareholding, which experience the controller–minority agency problem, are less amenable to regulation through ‘soft law’ rather than those with dispersed shareholding. Hence, even from a functional perspective, the claims towards convergence have not received sufficient backing.
Conclusion and way forward
Entrenched in the convergence–divergence debate, this article considered the effectiveness of the dissemination of voluntary corporate governance codes around the world. It found that such codes in the form of ‘soft law’ are more appropriate to countries like the UK where they originated. This is due to be presence of several specific factors in the UK, not least the corporate ownership structure which largely displays dispersed shareholding in companies. However, the article finds that when such voluntary codes are transplanted to jurisdictions where those factors do not operate in the same way as in the UK, results are likely to be different. Notably, this could be due to the concentrated shareholding structures that are prevalent in most countries other than in the Anglo-American context. This raises some doubts about whether the propagation of voluntary codes can be taken as indicia of convergence in corporate governance.
At the same time, there are some indications that voluntary codes and similar arrangements have begun to take into account the dichotomy between dispersed and concentrated shareholdings while designing corporate governance mechanisms. Although only a few in number, specific codes have been earmarked for companies with concentrated shareholding, such as family-owned companies and state-owned enterprises (Cuomo et al., 2016; ECGI, 2017). Such a focused approach will help address the controller–minority agency problem in such companies. Prescribing different sets of corporate governance rules for companies with varying shareholding structures could ultimately result in an efficient equilibrium (Davies and Hopt, 2013) that might display de facto convergence despite showing signs of de jure divergence. This can be achieved by advocating for the protection of minority shareholders.
A few illustrations are worthy of mention and could serve as effective examples of trends leading towards a targeted treatment of the controller–minority agency problem by using mechanisms that fall short of mandatory ‘hard law’. In 2014, the UK’s Financial Conduct Authority introduced rules for [premium-listed controlled companies] by which the election or re-election of independent directors must receive the approval of (i) the shareholders of the listed company as whole and (ii) the independent shareholders of the company, i.e. persons other than the controlling shareholder (FCA, 2014). By ensuring that independent directors are elected by both a majority of all the shareholders as well as a majority of the minority shareholders, it seeks to address the controller–minority agency problem that affected controlled companies (Bebchuk and Hamdani, 2017). The rules for the premium listing segment also require issuers to enter into agreements with their controlling shareholders to ensure that transactions with controlling shareholders are entered into on an arm’s length basis and that neither the controlling shareholders nor its associates will take any action, including a proposal for a shareholder resolution, that will have the effect of preventing the compliance with the listing rules (FCA, 2014). All of these measures are evidence of the use of the controlling strategy to address agency problems in companies with concentrated shareholding, although it remains a matter of some surprise that such steps are being undertaken in the UK that largely hosts companies with dispersed shareholding.
Separately, through an element that resonates with the controlling strategy, the special listing segment of ‘Novo Mercado’ on the BM&FBovespa, the Brazilian Stock Exchange, has required controlling shareholders of controlled companies to enter into the listing agreement with the stock exchange to ensure compliance with the various listing obligations (BM&FBovespa, 2011). By requiring the controlling shareholders to undertake legal obligations with the stock exchange, it ensures the enforceability of those contractual provisions against the controlling shareholders. The Novo Mercado experiment has been met with success and has also led to the gradual dispersion of shareholding in the Brazilian context (Gorga, 2014). The premium listing segments in both the UK and Brazil are interesting illustrations of the use of a ‘soft law’ approach to address agency problems in controlled companies. These efforts explicitly take into account divergences in ownership structures across the world and maintain targeted solutions to address corporate governance problems.
In terms of future research and policymaking efforts, the discussions in this article open up several avenues. First, regulators and stock exchanges may pursue the approach of an increasing number of countries that are inscribing codes for companies with specific types of shareholding structures, including for family-owned firms and state-owned enterprises. The same would hold good for international bodies such as the OECD and ICGN. Second, the level of compliance with corporate governance codes in the form of ‘soft law’ in companies with concentrated shareholding continues to remain an empirical question that needs to be tested further. While empirical studies have been focused largely on Europe, evidence from other significant jurisdictions, including economies such as China, India and Brazil, need to be further developed. Third, given that compliance with voluntary codes is best monitored by institutional investors, it is necessary to place emphasis on matters such as shareholder activism in the context of companies with concentrated shareholding, which will ensure a greater involvement by such investors. This will help determine the success in compliance or, alternatively, the meaningful explanations of reasons for non-compliance by companies. Such efforts can be enhanced by an independent monitoring and verification of disclosures by companies (Keay, 2014). Finally, a more granular analysis is required regarding the enforcement of codes, including the level of sophistication of the legal system and its institution, and the capabilities of various enforcers such as regulators, industry bodies and stock exchanges (Wymeersch, 2006). These endeavours will enable a more in-depth analysis of the impact that shareholding structures have on the dissemination of corporate governance. Until further results emerge, it would be imprudent to conclude that codes have led to a convergence in the direction of the Anglo-American shareholder-oriented regime.
Footnotes
Acknowledgements
I thank Dan Puchniak for helpful discussions on the subject matter of this article, the editors and anonymous referees for comments on previous versions and Joshua Kwan for research assistance. Errors or omission remain mine alone.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.