Independent Directors and Team Production in Japanese Corporate Governance

Abstract

Independent directors (IDs) in listed Japanese companies have gradually increased with the transplant of the Western model of the monitoring board. In practice, however, IDs act more like the mediating hierarch in team production theory than the agent of the shareholders, albeit with a number of differences from Blair and Stout’s seminal model. Japanese IDs mediate formally and informally, resolving vertical disputes between groups of executives as they contest control of the company. Given the norm of lifetime employment, such vertical disputes are common in Japanese companies and are economically significant, since failure to resolve them can result in destruction of firm-specific human capital. The article explores the scope for mediating hierarchy in Japanese law and corporate governance practice, then develops three case-studies which highlight the role played by IDs. Their practice is shaped by and supports social norms that emphasize the importance of continuity in team production.

1. Introduction

Independent directors (IDs) have gradually been accepted into Japanese company law and corporate governance. ¹ They are formally expected to perform a number of different functions that reflect the demands of agency theory—the approach that dominates corporate governance debates in the West ² and justifies appointing IDs to boards on the basis that they will control the decisions of management in the interests of the shareholders. ³ Those functions, which mirror Western corporate governance codes, include monitoring management and drawing up plans to deal with Chief Executive Officer (CEO) succession. ⁴ They are also expected to perform functions that have no counterpart in the West, such as encouraging management to take risks, reflecting Japan’s current economic situation.

In this article, we develop three case-studies to highlight how IDs in practice also perform functions that are unanticipated by the Japanese Corporate Governance Code (JCGC). More specifically, we show that, in some cases, IDs have acted in ways that are strongly redolent of the mediating hierarchs in Blair and Stout’s team production theory. ⁵ Disputes that arise among the various groups that come together within the company are often deep-seated, but it is essential to resolve them in a way that protects the significant investments in firm-specific human capital made by senior executives, managers, and employees.

At the same time, we also show that the mediating hierarch function they perform in Japan is subtly different from that described by Blair and Stout in their seminal analysis of the governance of team production in US corporate law.

First, Blair and Stout expected that directors would resolve “horizontal” disputes between, for example, shareholders and the CEO, or shareholders and creditors, but that “vertical” disputes within the managerial hierarchy would be resolved by the CEO. Such vertical disputes are highly salient in Japanese companies, which witness power struggles as different groups of executives strive for control of the company. The case-studies we develop in this article suggest that, as the number of IDs in Japanese companies has increased, this has given the board, for the first time, a degree of factual independence from management, allowing it to mediate such “vertical” disputes.

Second, Blair and Stout focused on the mediating function of the board as a collective body, and did not really explore individual directors’ roles. In Japan, in contrast, individual IDs may play a key role in mediating disputes. This has been necessary because IDs do not typically constitute a majority on the board. It has been possible in scenarios in which two management factions on the board offset each other, leaving the ID(s) holding the balance of power. It has also been possible where the IDs have been able strategically to leverage support from actors with an interest in the continued integrity of the team, such as an opposing managerial faction, outside shareholders, or even trade unions. Hence, our case-studies highlight for the first time individual IDs, rather than the board of directors, playing a key role in mediating disputes within team production.

Third, our case-studies show that IDs have acted as genuine mediators, in the sense of encouraging or enabling managerial factions to find their own solution to disputes, rather than imposing a solution upon them. It is only where the factions fail to find a solution to the conflict that IDs resort to the use of their formal authority. This provides a contrast to Blair and Stout’s model, in which the role of mediating hierarchy appears to be rather closer to that of an arbitrator, relying on formal authority where team members cannot resolve disputes themselves.

We think our analysis breaks new ground because the possibility of boards or IDs acting as mediating hierarchs has received little detailed consideration in the Japanese context. Primarily this was because boards of directors at Japanese companies tended, until fairly recently, to be managerial bodies with no—or very few—IDs. As such, the board was simply the apex of the managerial hierarchy, and team production theory seemed inapplicable because the board was not independent from one group of team members—that is, the executives. This permitted resolution of horizontal disputes: as Aoki highlighted in 1984, the role of managers in the Japanese “corporative managerialism” model was to act as “referees” who “excel in the skill of interest mediation,” integrating and mediating the interests of shareholders and employees. ⁶ Otherwise, team production theory perhaps seemed unnecessary. As we discuss in Section 4.2 below, strong coalitions, termed “company communities” by Shishido, developed at the firm level between management and employees in response to the social norm of lifetime employment, the absence of liquid external labour markets, and the existence of strong employment protection. ⁷ As management bodies, boards of directors acted to protect these coalitions against disruption, and were held accountable for this by the other members of the “company community.”

The concerted push for more IDs from the early 2000s raised doubts about the future of the Japanese model. Deakin and Whittaker noted in 2009 that moves to enhance shareholder power and, in particular, the introduction of IDs “as the representative of shareholder interests” called into question whether “the organizational practice of the community firm” would continue. After all, these practices were underpinned by social, rather than legal, norms. This led them to wonder whether, “in the long run at any rate,” growing reliance on IDs would threaten “organizations which depend on the long-term value created by firm-specific physical and human assets.” ⁸ Even the most innovative analysis, by Shishido himself, of the “incentive bargain” between monetary and human capital providers tended to assume that IDs would ratify the CEO’s plans, review their execution, and act as monitors of management on behalf of the shareholders. ⁹

The case-studies we develop in this article suggest that, as IDs have increased in number, they have acted, either individually or in collaboration with others, in ways that take account of shareholder interests, as conventional agency theory suggests, but also that they act to preserve the integrity of team production. Our case-studies show IDs acting as mediators where “vertical” conflicts emerge within the managerial hierarchy itself. In particular, they mediate the power struggles that develop between different groups of executives as they contest control of the company.

Hence, our case-studies and analysis suggest that the transplant of the notion of the ID from the US and UK to Japan provides an instance of Teubner’s notion that unifying law produces new divergences, as a transplanted institution is forced to work with the logic of the system into which it has been inserted. ¹⁰ Similarly, the widespread adoption of IDs by Japanese companies is a good example of the “faux convergence” identified by Goto, Koh, and Puchniak, according to which the adoption by a country of a tool of good governance turns out, on closer analysis, to be “the adoption of a different tool with different functions.” ¹¹ Indeed, our analysis suggests that IDs in Japan perform a function somewhat similar to that played by their counterparts in ethnic Chinese family firms in Singapore. ¹² Using case-studies, Ng and Roberts identified IDs playing a mediating role specific to the context: working within existing power structures, and family control in particular, outside directors worked to develop trust relations with the patriarch, which then allowed them to advise and persuade the board to resolve disputes in ways that preserve the viability of the family-controlled firm. This was a particularly delicate task where the solution involved replacement of a managing director appointed from the founding family, for example. Whilst there are clear parallels with the practices of IDs in Singapore, our case-studies also highlight that IDs in Japanese companies are confronted with a wider range of disputes than the directors in Ng and Roberts’s case-studies, and also that they use both formal and informal influence to mediate.

The rest of this article is structured as follows. In Section 2, we explore the emergence and institutionalization of the practice of appointing IDs to the boards of companies in the US and UK, and its subsequent transplantation to Japan. In Section 3, we introduce the team production theory of Blair and Stout, and examine whether it can be accommodated within the main features of Japanese company law and corporate governance. Section 4 develops a typology of the very specific disputes that have fallen to be mediated by IDs in Japan, and highlights how the role played by IDs in Japan differs from that of mediating hierarchs in the US. Section 5 uses three case-studies to illustrate the role played in practice by IDs. These case-studies show that IDs in Japan have mediated deeply entrenched disputes between groups of executives at the top of managerial hierarchies, and have done so with a view to keeping the team together as far as possible. A brief conclusion, including a couple of normative suggestions, follows.

2. The development and institutionalization of IDs

2.1 In the US and the UK

IDs have long been members of boards of US corporations. In 1927, Ripley argued that the “informed and more or less expert onlookers,” who sat on the board alongside executives, tended to act as “dummy directors … put there and kept there by the inner circle.” Instead of “knowing the master’s voice and acting harmoniously therewith,” they should, “so far as is possible, be independent of and serve as a check upon the inner group.” ¹³ By 1940, Ripley’s prescription was becoming reality as investment companies were, as part of the response to the 1929 stock market crash, required to have boards consisting of at least 75% IDs. In 1956, the New York Stock Exchange (NYSE) encouraged listed companies to have at least two outside directors. In 1977, with encouragement from the Securities and Exchange Commission (SEC), the NYSE proposed a rule requiring listed companies to have independent audit committees composed of outside directors (following an already well-established practice among listed companies). ¹⁴ These efforts on the part of regulators bore fruit: by 1977, outside directors were a majority on the boards of many large US corporations. ¹⁵ However, nominating committees only began to spread in earnest from the 1980s. It was not until 2003 that it became an NYSE listing requirement that listed corporations form a “nominating/corporate governance committee” comprising entirely IDs, and CEOs are still considered to exercise influence over the nomination process. ¹⁶

As for the function of these outside directors, a 1971 study by Mace concluded that they provided advice and counsel to the CEO, disciplined and constrained the activities of management, and acted in crisis situations such as where the CEO must be suddenly replaced through ill health, for example. Mace found no contribution to strategy or policy; and their disciplining role was akin to a “corporate conscience,” putting pressure on officers to identify problems and give explanations, but not asking “discerning or penetrating questions.” ¹⁷ Eisenberg’s blueprint placed more emphasis on their role in monitoring performance and replacing those found to be incompetent. According to his monitoring model of corporate boards, while executives manage companies, boards with a majority of IDs “hold the executives accountable for adequate results by selecting and dismissing executives and monitoring their performance.” ¹⁸

Looking back over the second half of the twentieth century, with its shifting patterns of corporate control, changes in management’s self-description, and seismic changes in theories of corporate governance, one thing has remained constant: the prescription that companies should appoint more IDs. This demand received broad political support during the 1970s, in no small part because the precise function of IDs had not been determined, allowing them to be justified as public interest directors ¹⁹ or as monitors responsible for “hold[ing] the executives accountable for adequate results (whether financial, social, or both).” ²⁰ Even by 1982, as agency theory began to dominate, Brudney argued that the move to a monitoring board through the appointment of outside directors was part of an “effort … to make the large, publicly held corporate both a more faithful instrument of wealth maximization for its shareholders and a more responsible citizen of society.” ²¹ However, as corporate governance moved towards “unalloyed shareholder value maximization in the 1990s and 2000s,” independent outside directors were seen as unconflicted and ideally placed “to insist on the primacy of shareholder interests; the expectations of director independence became increasingly stringent.” ²² Gilson and Kraakman insisted that these outside directors should be both independent and accountable to shareholders. ²³

A similar story could be told in relation to the UK, which began to push for the inclusion of IDs (there referred to as “non-executive directors” (NEDs)) on the boards of public companies from 1973. ²⁴ This was strongly encouraged by the Bank of England, which took the view that something had to be done to prevent further decline of the UK’s manufacturing industry. It was within reach of companies themselves to appoint “a good leavening of NEDs,” who could prevent drift and provide advice to the chairman based on their experience. ²⁵ From behind-the-scenes encouragement during the 1980s to institutionalization in the Cadbury Report and subsequent iterations of the UK’s Corporate Governance Code, the UK ultimately adopted the same solutions as the US. From the perspective of transplantation to Japan, it is important to note the UK’s approach because it embedded the requirement to appoint NEDs in a soft law code, with listed companies required by the London Stock Exchange to “comply or explain,” allowing capital markets to determine whether the level of compliance was satisfactory. ²⁶ That technique was ultimately adopted by policy-makers in many countries around the world, including Japan, as they sought to move towards a monitoring board. ²⁷

This brief historical tour has shown that IDs have long and consistently been advanced as an important part of the US and UK corporate governance systems. In recent decades, they have been accommodated as a mainstay of the shareholder value system of corporate governance. It is to the transplantation of this quintessentially Anglo-American institution to Japan that we now turn.

2.2 Transplantation to Japan

IDs were rarely seen in Japanese corporate governance before the early 2000s. The revision of the Special Provisions Act ²⁸ in 2002 was the first legislative attempt to increase the use of IDs. Inspired by recent structural reforms at Sony, as well as by US practice, the revision enabled companies to take a new structure that was similar to that adopted by companies in the US. It aimed at enhancing the monitoring of management and restoring company profitability after the long recession that followed the bursting of the bubble economy. ²⁹ It is clear that the drafters of the reform envisaged a transformation of the board from a managerial body to one in which shareholder interests were more clearly articulated. ³⁰ Policy-makers were clearly inspired by US corporate governance practice, with one member of Hōsei Shingikai Kaishahō Bukai (the Company Law Subcommittee at the Legislative Council of the Ministry of Justice) explaining that the new structure was needed because most directors were executive directors and therefore under the control of CEOs. ³¹ The Company Law Subcommittee did discuss a mandatory requirement for listed companies that took the conventional structure of kansayaku-kai-secchi-kaisha (company with board of statutory auditors) to install a shagai-torishimari-yaku (an outside director). Strong opposition from business associations, however, ultimately led to the failure of this initiative. ³²

In the conventional company with a board of statutory auditors structure, the board of directors has authority to appoint executives and make decisions on significant managerial matters. ³³ Kansayaku (statutory auditors) have a duty to supervise management to ensure that managerial authority is exercised in accordance with applicable laws. ³⁴ Kansayaku-kai (the board of statutory auditors) is expected to assist the statutory auditors in performing this function. ³⁵ Although the 2002 legislation required companies adopting this form to appoint two or more outside statutory auditors, ³⁶ there was no requirement regarding the appointment of independent or outside directors in this structure.

The 2002 legislation, instead, introduced a new optional governance structure called iinnkai-tō-secchi-kaisha (company with three committees), ³⁷ as an alternative to the conventional kansayaku-kai-secchi-kaisha (company with board of statutory auditors). ³⁸ In the new structure, day-to-day management of companies was delegated to shikkō-yaku ³⁹ (executive officers) while the board of directors, a majority of whose members would still have been drawn from management, had a legal duty to monitor the management. ⁴⁰ The monitoring function of the board was supported by three iinkai (committees) ⁴¹ —shimei-iinkai (nomination committee), hōshū-iinnkai (remuneration committee) and kansa-iinkai (audit committee). Each committee was required to consist of three or more directors and a majority of members of each committee were required to be outside directors. ⁴² This meant that companies choosing this structure had to appoint a minimum of two outside directors, then defined as directors who were not, and had not been, executive directors, executive officers, or employees of the company or its subsidiaries. ⁴³ However, out of fear that it might deter companies from adopting the voluntary structure, ⁴⁴ there was no requirement that outside directors form a majority on boards nor was there any formal independence criterion.

Policy-makers hoped that the old and new structures would compete with each other and that many companies would choose the new system in order to improve performance. ⁴⁵ Ultimately, that hope was dashed and, despite the flexibility offered, only a small number of companies chose the new structure; at its highest point, in 2006, it had been adopted by a mere 2.5% of all listed companies. ⁴⁶ One explanation for the lack of uptake is that, since committee decisions cannot be overruled by the board of directors, incumbents are reluctant to hand over control of nomination and compensation decisions to outsiders. ⁴⁷ Those interviewed by Buchanan and Deakin viewed the committee structure as allowing quicker and more efficient decision-making, rather than indicating a wholesale change in corporate governance towards “global standards.” ⁴⁸ Deakin and Whittaker note that, in those companies that had adopted the new structure, “external directors were treated as advisers and associates, very much as before,” ⁴⁹ rather than playing the role of agents of the shareholders.

Further reforms were introduced in the second decade of the twenty-first century. Three main changes were made in 2014 to Kaishahō (the Companies Act). ⁵⁰ First, companies were permitted to adopt (yet) another governance structure called kansa-tō-iinkai-secchi-kaisha (company with audit committee), ⁵¹ with a majority of kansa-tō-iinnkai (“audit-plus” committee) consisting of outside directors. ⁵² In addition to auditing the accounts, the committee can potentially oppose management, being appointed by the shareholder meeting ⁵³ and is entitled to express its views on election, dismissal, resignation, and compensation of directors. ⁵⁴ With the shareholders setting the aggregate remuneration of this committee, its members look very much like the “agents” of the shareholders. ⁵⁵ Second, the reform offered a stricter definition of outside directors, ⁵⁶ disqualifying three further groups from serving with a view to ensuring greater independence: controlling shareholders, defined as a person who controls determinations of the financial and business policies of the company; directors, executive officers, and employees of the company’s parent or sister company; and spouses or close relatives of the directors, executive officers, or important employees of the company or its controlling shareholders. ⁵⁷ Third, the reform imposed a “comply or explain” requirement on all large listed companies, whatever their governance structure, requiring them, where they do not appoint even a single outside director, to offer an explanation of why doing so would be detrimental to the company. ⁵⁸ Since making a credible explanation has become onerous, this was often viewed as a de facto obligation to appoint a single outside director. ⁵⁹ Ultimately, when the Companies Act was amended in 2019, the “comply or explain” requirement was upgraded to a mandatory requirement that at least one outside director be appointed. ⁶⁰

Alongside these reforms to the Companies Act, both the Tokyo Stock Exchange (TSE) and JCGC have advocated the use of IDs since 2013.

TSE revised its Yūka Shōken Jōjō Kitei (Tokyo Stock Exchange Listing Rules) in 2014 to require listed companies to make an effort to appoint a torishimari-yaku-dearu-dokuritsu-yakuin (independent outside director), ⁶¹ although it was not mandatory actually to appoint one. According to the TSE Listing Rules, “independent outside directors” are outside directors “who have no conflict of interest with general shareholders.” ⁶² Directors are disqualified from acting as independent outside directors when they: hold senior positions in entities that are major trading partners of the company; act as consultants, accountants, and lawyers who receive a large amount of money from the company; are executive directors or officers of the company’s parent company or sister company; and where they are family members of executive directors or officers of the company or its subsidiary or of any of the people mentioned above. ⁶³

The origins of the JCGC can be traced to the 2013 Cabinet announcement of its Nihon Saikō Senryaku (Japan Revitalization Strategy), ⁶⁴ which aimed to improve the productivity of companies and encourage corporate investments in order to reverse the economic downturn. The appointment of outside directors was included among its policies to improve corporate governance. A revised strategy, published in 2014, ⁶⁵ encouraged companies to make use of independent outside directors to improve management strategy. Following its recommendation that a JCGC be drafted and implemented, the Financial Services Agency (FSA) and the TSE set up Kōporēto Gabanansu Kōdo No Sakuteis Ni Kansuru Yūshikisya Kaigi (the Council of Experts Concerning the Corporate Governance Code), which in turn, in March 2015, published a draft requiring, on a “comply or explain” basis, companies to appoint an independent outside director. ⁶⁶

The first formal version of the Code (JCGC 2015) stated that, in order to “stimulate healthy corporate entrepreneurship, support sustainable corporate growth and increase corporate value over the mid- to long-term,” companies should “make effective use of independent directors.” ⁶⁷ They were expected to provide advice, monitor management, monitor conflicts of interest, and represent minority shareholders and other stakeholders “from a standpoint independent of the management and controlling shareholders.” ⁶⁸ JCGC 2015 required listed companies to appoint two or more dokuritsu-shagai-torishimari-yaku (independent outside directors). ⁶⁹ However, as the preamble to the Council of Experts’ Final Proposal, which was appended to JCGC 2015, made clear, this was a “comply or explain” obligation, with companies not obliged to comply, provided they explained fully their non-compliance. ⁷⁰ JCGC 2015 also encouraged listed companies, where they considered this appropriate, to disclose a roadmap to appointing independent outside directors making up one-third of the board. ⁷¹ In its next iteration, JCGC 2018 ⁷² maintained the requirement of two IDs but encouraged listed companies to “appoint a sufficient number of independent outside directors.” ⁷³ In its most recent version, JCGC 2021 ⁷⁴ required companies whose shares are listed on the Prime Market, which is to be established in 2022, to appoint IDs making up one-third of the board (whilst companies listed on other markets should appoint at least two). ⁷⁵

The JCGC has some distinctively Japanese characteristics. For example, it expects IDs to encourage risk-taking, ⁷⁶ which is almost the opposite of Western codes that put emphasis on systems of risk management. At the same time, with its reliance on “comply or explain” to encourage companies to appoint IDs, giving companies scope to tailor their governance structures to their own circumstances, the JCGC was, as noted above, clearly inspired by the pioneering approach of the UK Corporate Governance Code (UKCGC). ⁷⁷ Likewise, following the lead of the London Stock Exchange, which had appended the Cadbury Report (which formed the foundation of the UKCGC) to the Listing Rules, ⁷⁸ the TSE incorporated the 2015 iteration of the JCGC as an appendix to the Listing Rules in 2015. ⁷⁹ Each subsequent iteration of the JCGC has been similarly incorporated into the Listing Rules.

Combined with the legal changes discussed above, requiring listed Japanese companies to explain any non-compliance with the JCGC to the market has clearly driven the appointment of more IDs. Whilst listed companies with a majority of IDs are still rare, a majority of listed companies now have boards consisting of one-third or more IDs. In 2012, 65.6% of listed companies had no ID ⁸⁰ ; by 2021, 81.5% of those companies and 97% of those companies listed in the first section had appointed two or more IDs, whilst in 58.2% of listed companies and 72.8% of companies listed in the first section, one-third of the board consisted of IDs. ⁸¹

The above discussion has shown that the Japanese reforms have drawn on both US and UK regulation and practice in an effort to encourage companies to appoint IDs. However, our argument in this article is that, rather than acting as mere agents of the shareholders, IDs in Japan have come to play a distinctive mediating hierarch role, resolving vertical disputes that arise in team production between groups who have acquired highly specialized skills with a view to holding the team together. In the next section, we begin by outlining team production theory. After that, we canvass whether mediating hierarchs are needed in Japan, and explore how much scope there is in Japanese company law and corporate governance practice for IDs to play the role of mediating hierarchs.

3. Team production theory and the potential for mediating hierarchy in Japan

3.1 Team production in theory

Although mainstream corporate governance discourse has, as we noted in the introduction, for the most part conceptualized IDs as a solution to the agency problem facing shareholders, it is possible to conceptualize their functions in a number of different ways. ⁸² Margaret Blair and Lynn Stout rely on the economic theory of team production to explain the legal autonomy of the board of directors. ⁸³ They argue that the law applicable to boards of public companies is better understood not as a response to principal-agent problems between shareholders and executives, but as a solution to team production problems that arise where multiple team members make firm-specific investments. Each of the groups involved in team production surrenders control over their assets to a legal entity. That control is then exercised by a managerial hierarchy, at the top of which sits a board of directors “whose authority over the use of corporate assets is virtually absolute and whose independence from individual team members … is guaranteed by law.” ⁸⁴

In addition to shareholders and creditors, employees and executives are encouraged to enhance their productivity within the team production process by making investments in firm-specific human capital (FSHC) that are of far greater value to their present employer than to any other. ⁸⁵ Such FSHC includes knowledge of the firm’s routines and relationships, skills that are specialized to the firm’s needs, and so on. Team production therefore “involves several groups with potentially conflicting interests over the output of the production by companies.” ⁸⁶ Serious problems can arise in determining how any economic surpluses (or “rents”) generated by team production should be divided up. That surplus cannot be divided up by ex ante binding contract, because that would “invite shirking,” while waiting until ex post to divide by means of authoritative decision will “create incentives for opportunistic rent-seeking.” Neither of these is desirable, yet leaving team members without any protection for their expectations would discourage them from bearing residual risk by committing to tying “their economic fortunes to the firm’s fate” and investing in FSHC. In order to resolve this dilemma, say Blair and Stout, the various team members can put their assets and expectations under the control of a legally autonomous board in the expectation that they will ensure that the contractually unprotected economic gains that result from team production will be shared fairly.

Team production explains, argue Blair and Stout, why the board of directors is invested by law with very wide authority, free from the control of any team member, whether shareholders, creditors, executives, managers, or employees. That authority enables them to act as mediating hierarchs, with a view to “maximiz[ing] the joint welfare of the team as a whole.” The role of the mediating hierarchs is to determine the use of corporate assets and resolve disputes between team members over the allocation of duties and rewards. The board of directors usually appoints the executives who head the managerial hierarchy, and most decisions are expected to be “made collegially among team members at lower levels;” it is only where disputes cannot be resolved horizontally that the problem is “kicked upstairs” to a “disinterested—but potentially erratic or ill-informed—hierarch.” ⁸⁷ Hence the background threat of intervention by the mediating hierarchy provides strong incentives for horizontal dispute resolution among team members, and the board only rarely needs to intervene in relation to resource and output allocation. But if necessary, the board may intervene and fire the CEO, which is something that becomes particularly apparent where—as is the norm for the largest listed companies—the majority of the board consists of outside directors. ⁸⁸

This broad discretion may even exacerbate agency costs: mediating hierarchs have legal latitude to use companies’ resources to benefit managers, employees, and other stakeholders at shareholders’ expense. ⁸⁹ Whilst this may be contrary to the shareholder wealth maximization norm supported by the principal-agent model, Blair and Stout claim that their model fits better with key features of US corporate law. Millon counters that while “there may in fact be enough play in the joints” of corporate law to allow boards to act as mediators, this is not mandated by the law, and considerable further steps would need to be taken to establish “this state of ivory tower autonomy,” including making changes to hostile takeover rules, board election, and remuneration. ⁹⁰ In essence, Millon’s contention is that company law and corporate governance in the US have too many features aligning director and shareholder interests for mediating hierarchy to be a realistic possibility. ⁹¹

It is worth noting that, aside from briefly mentioning their presence on the boards of listed and pre-IPO (Initial Public Offering) companies, and noting that the board “may also include several outsiders,” ⁹² Blair and Stout paid little attention to the role of IDs. This is unsurprising since US listed companies have had a majority of IDs for several decades now, and so their focus was on the more controversial topic—from the perspective of agency theory—of why directors have legal autonomy from shareholders. Millon noted that team production theory offers “a compelling reason to liberate the board from management’s control,” but the board would also have to be “fully protected from shareholder efforts to influence its behaviour.” ⁹³ Perhaps boards should be “self-perpetuating, rather than subject to selection by shareholders or senior managers,” he suggested. ⁹⁴ Alternatively, Kaufman and Englander suggested that, in team production, boards needed to move away from representing shareholder interests and towards replicating the various team members who add value and assume risks. ⁹⁵ In any event, it is clear that a considerable degree of legal and factual independence is a necessary, if not sufficient, condition for a board to act as a mediating hierarch.

In what follows, we examine first whether, in theory and practice, mediating hierarchs are needed in Japan and, second, whether, as a descriptive matter, IDs and boards have sufficient autonomy in Japanese company law and corporate governance practice to perform this mediating role.

3.2 Mediating hierarchy in Japanese company law and corporate governance

The changes in the post-bubble economy created a need for mediating hierarchs in corporate governance. Before the bubble economy burst, the Japanese corporate governance system was characterized by complementarity of social norms such as cross-shareholding, relationship banking (also referred to as the “main-bank system”), lifetime employment, and internal promotion of management. ⁹⁶ Companies held shares in each other, reciprocally insulating executives from stock market pressure. Companies also built long-term relationships with specific banks, which not only provided capital, but also monitored management. Lifetime employment and internal promotion of executives from amongst the company’s workforce were also commonly discussed features of Japan’s co-ordinated market economy. ⁹⁷ Under the practice of lifetime employment, companies recruited new graduates and, once hired, these employees typically worked for the same company until they reached mandatory retirement age. Similarly, once selected from the ranks of senior employees, executives typically remained in position until they retired. ⁹⁸

As a whole, this system allowed executives to focus on the long-term prosperity of companies and enabled companies to develop employee and executive investments in FSHC. With a relatively small number of fixed team members, disputes were, in theory and practice, solved by the team members themselves, with management given broad autonomy subject only to the need to keep the team together under the umbrella of the “company community.”

After the bubble burst, however, the number of outsiders began to increase: cross-shareholding started to decline with foreign shareholders taking up the shares ⁹⁹ ; the scale of lifetime employment declined and it became more common to hire employees on short-term contracts. ¹⁰⁰ These changes made it harder for the team to co-ordinate interests and resolve disputes by themselves, creating a need for IDs.

In the theory of team production, team members cede authority to the board of directors. Until recently, management, perhaps the most influential team member, were reluctant to give up their dominant position on the board and strongly resisted the appointment of IDs. ¹⁰¹ When the corporate governance changes discussed in Section 2 above were introduced, management were forced to accept the presence of IDs, but learned to work around the change by nominating as IDs individuals who could be expected to confine themselves to modest functions, such as acting as advisers rather than strong external monitors. Indeed, according to research conducted by the Ministry of Economy, Trade and Industry in 2000, more than half of outside directors viewed their role as advisory rather than monitoring. ¹⁰² However, as time passed, companies, on average, started to appoint more IDs than required by the Companies Act and JCGC. Moreover, some management teams even gave up their authority to IDs. An example of this can be found in the LIXIL case-study in the final section of this article. There, management voluntarily nominated sufficient IDs to constitute a majority of the board, hoping that they would thereby gain shareholder support and that those IDs would mediate the dispute in their favour. In other words, management seems increasingly to have recognized that IDs can help support the continuity of team production. The research conducted in 2000 also revealed that more than half (50.9%) of outside directors prioritized the interests of “various stakeholders beyond the shareholders such as employees, customers, trade counterparties, creditors and local communities (but not shareholders),” whilst 37.9% prioritized general shareholders and 11.5% prioritized “executives such as shachō [president] or kaichō [chairman]).” ¹⁰³

While there are now more IDs on boards, they also have considerable autonomy and power that they can mobilize in mediating disputes between team members. It is true that shareholders have the formal power to appoint IDs ¹⁰⁴ but, once they have been appointed, they are free from shareholder control. Beyond shareholders, neither the employees nor any other corporate constituency have appointment rights in relation to IDs and so have no mechanisms available to them in company law to influence them either.

In mediating disputes between team members, IDs can leverage the board’s autonomy and authority. Boards have authority to decide the company’s business affairs, ¹⁰⁵ with the exception of certain matters that are reserved for the shareholder meeting, such as approving dividends ¹⁰⁶ or mergers. ¹⁰⁷ Whilst the board can delegate some of their decision-making authority to management, they must retain authority in relation to key decisions. For kansayaku-kai-secchi-kaisha (companies with board of statutory auditors), these include: appointment and dismissal of important employees ¹⁰⁸ ; appointment and dismissal of representative directors (who act on behalf of companies to make contracts or bring litigation) and executive directors ¹⁰⁹ ; disposal and acceptance of transfer of important assets ¹¹⁰ ; and other important matters. ¹¹¹ For shimei-iinkai-to-secchi-kaisha (companies with three committees) or kansa-to-iinkai-secchi-kaisha (companies with audit committee), these include: appointment and dismissal of representative directors, executive directors, and officers ¹¹² ; and basic management policy. ¹¹³ Perhaps most importantly, in all three types of company, boards retain the authority to appoint the CEO. ¹¹⁴

Looking beyond the law, the passivity of shareholders was traditionally an important factor in empowering boards to act in practice as management bodies. Conventionally, this passivity was explained by reference to intercorporate cross-shareholdings, a hallmark of “coordinated market economies” in the Varieties of Capitalism literature. ¹¹⁵ Historically, cross-shareholdings provided support for long-term relationships between companies, as well as insulation of managerial discretion from the influence of outside shareholders. ¹¹⁶ But with the decline in size and significance of cross-shareholdings since the mid-1990s, ¹¹⁷ the picture has become much less clear.

On the one hand, activism appears to be on the rise, with Japan witnessing both “confrontational” and “quiet” shareholder activism during the first two decades of the twenty-first century. One potential effect of this is to strengthen the emergent social norm of shareholder wealth maximization and undermine the mediating function of IDs. ¹¹⁸ Foreign institutional investors have increased their shareholdings ¹¹⁹ and are more likely to vote against proposals from executives at shareholder meetings. ¹²⁰ Similarly, domestic institutional investors, dissatisfied by poor stock performance, which they attribute to poor corporate governance, have taken a more active approach to voting. ¹²¹ This is arguably, in part, because the 2014 Japanese Stewardship Code encourages this and, in part, because they have been assisted by proxy advisory firms. ¹²² The latter exercise considerable influence over voting by institutional shareholders: research conducted in 2017 revealed that domestic investors, and foreign investors in particular, are much more likely to vote against proposals for director appointments when advised to do so by ISS. ¹²³

Whilst these shareholders may be ready to oppose management, they do not yet seem willing to confront IDs. In a few rare cases, they have refused to reappoint IDs. One example is the recent, but exceptional, case of Toshiba. ¹²⁴ Such cases, however, remain rare and it remains unlikely that shareholders will intervene when IDs perform mediating functions, given the wide discretion of the board.

On the other hand, despite shareholder dispersal and a not unfavourable regulatory regime, hostile takeovers remain rare in Japan. ¹²⁵ Several hostile takeover attempts have been witnessed in Japan since the 2000s but, until recently, they had all failed: the first successful hostile takeover was seen in 2020 when Colowide succeeded in its bid for control of Ootoya Holdings. ¹²⁶ It remains uncertain whether hostile takeovers will emerge as a strong driving force for shareholder wealth maximization. Similarly, strong rights for shareholders to make proposals and use proxy machinery ¹²⁷ have not translated into hedge fund activism, apart from a brief period in the early 2000s. ¹²⁸

Further complicating the picture, 17.3% of all listed companies on the TSE have controlling shareholders, in the form of either a parent company or some other kind of controlling shareholder. ¹²⁹ This could potentially undermine the mediating function of IDs. Whilst, IDs are required by the Companies Act and TSE regulations (and expected by JCGC 2021) to be independent of and to act independently from such controlling shareholders, shareholders can use their strong rights in law to exercise control over IDs. One rare instance of this is the case of Askul, in which a confrontation arose between Yahoo Japan, which owns 45% of the shares in Askul, and Askul’s management. When Yahoo Japan demanded the sale of one of Askul’s businesses and Askul’s management refused, Yahoo Japan voted against the reappointment of three IDs as well as the CEO of the subsidiary. ¹³⁰ Yahoo Japan’s behaviour was criticized by corporate governance advocates ¹³¹ and authorities are seeking to implement regulations or guidelines to secure substantial independence of IDs from controlling shareholders, thereby protecting minority shareholders. ¹³² Furthermore, both JCGC 2021 ¹³³ and proxy advisory firms ¹³⁴ encourage companies with controlling shareholders to appoint additional IDs—a move intended to give IDs greater autonomy. But, for the time being, it remains uncertain how far controlling shareholders will allow IDs to act autonomously.

Overall, then, there is considerable space for IDs to perform mediating functions. The limits on shareholders’ legal powers and continuing shareholder passivity just discussed combine with directors’ legal authority over management to allow IDs to mediate disputes among team members.

This brief review suggests that Japanese company law, as well as corporate governance and practice, creates space for companies to operate along the lines set out in team production theory in the previous subsection. Boards have a degree of legal and factual independence from both shareholders and other stakeholders, and they can also assert legal authority over management. In the next section, we explore how social norms give rise to very particular types of disputes within the Japanese system, and how IDs have accordingly been required to perform a mediation function that differs in scope from the one in Blair and Stout’s model.

4. Mediating disputes in Japanese corporate governance

4.1 A typology of disputes in Japanese corporate governance

While social norms in the Japanese corporate governance system, such as lifetime employment and internal promotion, made it possible for team members to create productive coalitions as discussed in Section 3, these practices also give rise to very specific disputes and problems. In particular, power struggles may develop between different groups of executives—and sometimes employees—as they strive for control of corporate hierarchies. These struggles can come to the surface in the contest to succeed a retiring CEO, but serious problems may also develop where the CEO is dismissed. Following Blair and Stout, we refer to these disputes as “vertical disputes.” Whilst they are rarely the focus of corporate governance analyses, ¹³⁵ our case-studies highlight how vertical disputes may pose a serious danger to the continued integrity of the team. Based on our case-studies, we divide vertical disputes in Japan into four separate, but sometimes overlapping, categories: between loyalty groups or factions; between different founder families or between founder families and executives; between groups of executives from different companies that have merged; and between executives and employees.

The first type of dispute (between loyalty groups or factions) develops directly out of the social norms of lifetime employment and internal promotion. With executives being drawn from among senior employees, enduring networks of loyalty and patronage develop among those who have worked together for a long time. Those networks often fight to gain and maintain control and influence within the company in various ways. This may be obvious, as where they contest the nomination of the CEO, but it may also occur in more subtle ways, such as by appointing retired executives as “advisers.” Retired executives often continue to occupy an honorary position such as komon or sōdanyaku (both meaning “adviser”). These honorary positions do not carry any formal legal authority but enable their holders to maintain significant continuing influence in practice—something that has given rise to significant problems in recent times. TSE has recently developed regulations requiring listed companies to disclose the appointment of retiring executives to such advisory positions. ¹³⁶ This is a response to practical guidelines provided by the Ministry of Economy, Trade and Industry acknowledging that these advisers may perform valuable functions as a result of their personal ties, developed during their long service to companies to a range of stakeholders. However, the guidelines also note that these retired executives may have a harmful influence over the management of the companies, even though they have no managerial authority and bear no legal responsibilities. ¹³⁷

The second type of dispute arises between founder families and other executive factions or between different founder family members. Founder families frequently retain considerable (and even controlling) shareholdings, whilst individual members of that family may hold executive positions, and so have developed FSHC over time. Saito shows that founder family members held the largest share in 25% of listed companies during the period from 1990 to 1998; 36% had a founder family member as either kaichō (chairman) or shachō (president). ¹³⁸ Founder family members who find themselves in this position can exercise substantial influence over corporate affairs. They often pay attention not only to shareholder returns, but also to continuity of the business on which their human capital depends, as well as to more psychological benefits such as a reputation for quality. Whilst a focus on long-term corporate success rather than short-term profitability might be welcomed, ¹³⁹ family influence over the managerial succession process might still create disputes with other executives and their associated factions.

Conflicts involving founder family members might arise in a number of ways. The founder family member may have handed the managerial reins over to one or more family or non-family successors, but subsequently, for some reason, attempted to take back control, either formally or informally. Alternatively, they may retire and bring in successors from outside, giving rise to conflicts over management policy between founder family members (who normally remain significant shareholders and may also be directors) and their chosen successors. Even when founder family members do not retain their positions as directors, they often continue to occupy the honorary positions discussed above, allowing them to retain significant influence.

The third type of dispute arises after a merger has taken place between two or more companies. The post-merger setting sees two or more groups of executives, each having had careers with a different company before the merger, struggling for control of the enlarged corporate group. This type of dispute is often cited as a problem to be solved after mergers have occurred. ¹⁴⁰

These three types of disputes occasionally give rise to the fourth type, which arises between executives and employees. Formally, and in line with the nature of the firm, executives have hierarchical authority over employees. But one effect of the social norms of lifetime employment and the “company community” is that employees can, belying their position in the formal hierarchy, wield considerable influence. For example, Yoshimura asserts that employees have exercised influence over decision-making by boards. He reports several cases in which trade unions and senior employees contributed to decisions on significant matters such as dismissing or appointing CEOs, or rejecting mergers at board meetings. This influence is arguably exerted because employees are dependent on the long-term stability and growth of companies. ¹⁴¹

Our case-studies show that IDs who are unaligned with any of the executive or family factions may offer a way out of these disputes and the impasses to which they may lead. Such IDs can intervene where the parties cannot co-ordinate their interests by themselves. It is true that, technically, IDs could be drawn from family factions, provided the faction’s voting share is less than 40%. ¹⁴² Such appointments, however, are discouraged by the TSE disclosure requirement ¹⁴³ and the provisions of JCGC 2021. ¹⁴⁴ Moreover, many companies articulate and disclose their own standards of director independence (as required by JCGC, Principle 4.9) in order to establish substantial independence, often providing that neither those who own 10% or more of the company’s shares nor their close relatives can be IDs. ¹⁴⁵

4.2 The scope of mediating hierarchy in Japanese corporate governance

The second point to be made in this section is that the introduction of IDs in the Japanese system has resulted in a mediating hierarchy that differs in scope from that described by Blair and Stout in the US context.

First, Blair and Stout expect interventions from the board of directors only where there is a horizontal dispute between the CEO and the shareholders, for example, or the CEO and other constituents. ¹⁴⁶ So, for example, shareholders can ask the board to fire the CEO, but the board is free to decline to do so; similarly, the board determines how much the balance sheet should be leveraged, and therefore the balance of risk between shareholders and bondholders; and the board ultimately decides whether to continue operating a marginally profitable plant for the benefit of communities and employees. ¹⁴⁷ They do not appear to envisage vertical disputes between the CEO and other executives falling to be mediated by the board of directors. Rather, the CEO has hierarchical authority over the executives below her and executives who are unhappy about the CEO’s decision must either leave the company or abide by her decision. Similarly, executives have authority over lower-level managers, who in turn have authority over employees. As a result, vertical disputes are not mediated by the board. ¹⁴⁸ This can be seen clearly from Blair and Stout’s diagrammatic representation, ¹⁴⁹ shown in Figure 1.

Figure 1. Blair and Stout’s team production corporate structure

It is of course incorrect to say that US companies do not witness contests for corporate control between the CEO and other executives, or to say that the board of directors never exercises its authority when vertical disputes arise. One example is the well-known power struggle between Steve Jobs and John Sculley at Apple, which, it is commonly claimed, had to be resolved by the board. ¹⁵⁰ Such disputes, however, occur far less frequently in the US and UK than in Japan because executives can find other jobs via a well-developed external managerial market (and as a result their investments in human capital will tend to be less firm-specific than their Japanese counterparts). This means they do not have to commit themselves to costly struggle within companies and the board of directors can normally leave this kind of dispute to the CEO, knowing that this will neither seriously discourage investments in human capital nor cause major disruptions to team production.

In contrast, Japanese executives, who cannot rely on an external managerial market, have much more at stake. We suggest that this means that serious contests among executives for corporate control are considerably more common and must be resolved in ways that preserve the executives’ FSHC. Executives have typically not only invested in FSHC throughout their career as employees at the companies, but also make further investments during the early stages of their executive career. Whilst FSHC developed by employees can arguably be protected through the social norms of the community firm or “company community” (discussed next), once an employee is promoted to an executive position, further investments in FSHC must be protected in the boardroom.

When the board of directors in Japan was a pure management body, vertical disputes were generally resolved in line with the social norms of the community firm, with the CEO exercising a dominant—but informal –influence. Kawamura, based on his experience as a company lawyer, described how shachō (president) and kaichō (chairman) had complete autonomy in making personnel decisions. ¹⁵¹ On the other hand, Shishido emphasized that the CEO’s autonomy is constrained in practice by the members of the “company community.” In Shishido’s account, “[e]mployees are, at least within their own mindset, not employed by the company, but rather belong to the ‘Company Community’.” This “company community” consists of “management, board members, and core employees who share an identity as ‘company men’.” Those core employees are “male white-collar full-time employees, [who] maintain a remote possibility to be the president or a board member.” ¹⁵² The CEO sits at “the top” ¹⁵³ of the company community, but management is appointed on the basis of informal consensus among employees. This consensus is, according to Shishido, established “through the long-term process in which employees are continually selected for promotion by the informal consent of fellow employees.” ¹⁵⁴ Shishido construed this process as continuous monitoring by the company community over the management, ¹⁵⁵ sometimes leading to, ideally voluntary, management turnover. ¹⁵⁶ Whereas he anticipated that such turnover would occur when members of the management team perform poorly, ¹⁵⁷ we submit that the same informal process can in principle be (and indeed, commonly is) used to resolve power struggles between different management factions. However, sometimes these power struggles become too entrenched, leading IDs to play a crucial mediating role in the interests of the company, the shareholders, the management, and the employees.

Second, whereas Blair and Stout focused on the mediating function of the board and did not really consider the role of individual directors, our case-studies suggest that, in Japan, individual IDs sometimes play a key role in mediating disputes. This is because IDs do not typically constitute a majority on the board. In the Japanese system, rather than acting as mediating bodies, boards can become battlegrounds in which vertical disputes between groups of executives for control of the company become entrenched. Where the board becomes polarized in this way, mediating functions may fall to be performed by IDs individually or collectively. Our case-studies suggest that this may occur even where IDs are a small minority but hold the balance of power between two competing factions, allowing them to mediate vertical disputes within the board itself. Their authority to do this may be enhanced where they are supported by, for example, the trade union that represents the company’s employees, as our Seiko case-study highlights.

Third, our case-studies highlight how IDs in Japan may perform their mediating functions in an informal manner. The informal nature of this mediating function is similar to the way management used to co-ordinate the interests of various team members in the past. While Blair and Stout do not describe in detail how the board mediates horizontal disputes between team members, it is supposed to be done through the exercise of formal decision-making authority, such as declaring dividends or approving key strategic decisions. In contrast, our research suggests that, when they act to resolve vertical disputes, IDs in Japan sometimes act as true mediators or facilitators, putting in place the conditions for team members to co-ordinate their interests by themselves. They try to foster agreement between managerial factions as to the way forward. Reliance on more formal authority remains a last resort to be exercised only where one faction of executives rejects the IDs’ proposed resolution of the dispute. One formal legal power that may be used in this situation is the board’s power to appoint and dismiss the CEO and other executives. It is only when these informal efforts to foster agreement between competing factions fail that serious disputes and the mediating roles of IDs become visible to outsiders.

Whilst our case-studies, of necessity, consider only the handful of publicly visible disputes, we expect that IDs will, in most cases, continue to mediate vertical disputes in an informal manner very similar to how management used to mediate them in the company community discussed above. Even after the bursting of the bubble, many companies maintained their “company community” practices, keeping existing employees and reducing the number of new employees in order to reduce excess employment. ¹⁵⁸ Indeed, the background of many IDs makes this continuity of approach more likely. According to TSE, ¹⁵⁹ 58.5% of IDs came “from other companies” as of 2020. With a majority of IDs being deeply familiar with the way decisions are made in Japanese companies, we would expect them to continue to follow the same path, only resorting to formal authority where other means fail.

In the next section, we develop three case-studies that illustrate the four types of dispute discussed above. They show that, in practice, IDs in Japan play a mediating role that sometimes extends to resolving “vertical” disputes between the CEO and other executives lower down the managerial hierarchy. Hence, whilst the CEO might have formal authority to resolve disputes within the managerial hierarchy, IDs may intervene to mediate (or foster viable compromise in) those disputes in order to prevent the dissipation of FSHC as a result of other executives lower down the hierarchy choosing to leave the firm.

5. Case-studies on the mediating function of IDs in Japan

In this section, we construct three case-studies on the basis of newspaper and other reports that highlight how IDs—acting individually or collectively—have mediated disputes falling into one or more of the four categories discussed above. We recognize that most vertical disputes remain invisible to outsiders like us, being resolved internally and informally. This makes it very difficult to develop a picture of what is occurring inside companies, and indeed whether IDs are playing a role in steering that resolution. Occasionally, however, serious disputes emerge into the public realm, either where board-level executive appointments are contested or, more rarely, where disputes between executives and employees cannot be resolved informally. These disputes may pose such a threat to the company and the integrity of team production that they cannot be left to internal resolution, leading the IDs to mediate them with a view to protecting the interests of the company, the shareholders, and the different groups that have made firm-specific investments. We are conscious that this is a small sample size, but the presence of IDs on the board is a relatively new phenomenon. These three cases are the only occasions on which the mediating role played by IDs has become visible to the public.

5.1 Seven & i Holdings

Seven & i Holdings is the holding company for one of the two largest retail businesses in Japan. It holds the shares of multiple companies including Seven Eleven Japan, which is now the largest chain of 24-hour convenience stores and the largest division in Seven & i Holdings. It adopted the form of a company with the board of statutory auditors. In 2016, at the time of the dispute, there were 15 directors: 11 executives and four IDs. While the IDs made up less than one-third of the board, they gained additional influence through their presence on the nomination and remuneration committee. That committee had been established in March 2016 as an advisory body in line with the recommendations of the JCGC, with two executives and two IDs as its members. The committee was supposed to provide advice to the board on the nomination and compensation of executive officers and directors. ¹⁶⁰ Apart from founder family members holding 10%, the company’s shares were widely held, with foreign investors holding 35%. ¹⁶¹

A dispute arose in 2016 between two factions of executives. One faction was led by founder family members: they held 10% of the shares, founder Masatoshi Ito held an honorary position, and his son, Junro Ito, was one of the directors. The other faction was led by the CEO and chairman, Toshifumi Suzuki. The dispute arose when Suzuki proposed a management reshuffle that involved replacement of a senior executive, Isaka. Isaka was both a director of the holding company, Seven & i Holdings, and the president and chief operating officer of its subsidiary, Seven Eleven Japan. The founder family faction was strongly opposed to this plan and they supported Isaka when he confronted the CEO, Suzuki. ¹⁶²

IDs performed a critical role in mediating the dispute. The first event was when Suzuki failed to get the plan approved by the Nomination and Compensation Committee. That committee consisted of: the CEO, Suzuki; the president and chief operating officer, Murata; and two IDs, Kunio Ito and Yonemura, one of whom acted as chair. ¹⁶³ The two IDs opposed the reshuffle plan, which therefore failed to obtain the committee’s approval. The board of directors then also rejected the plan with seven out of the 15 directors voting in favour, but six voting against and two abstaining. Again, the four IDs on the board played a critical role, not only by voting against the plan, but also by insisting that the vote should be conducted through a secret ballot rather than following the ordinary practice of a show of hands. This voting style was suggested by the ID, Kunio Ito, ¹⁶⁴ reportedly on the basis that if a show of hands were required, directors would not vote based on their own judgement out of a desire to avoid displeasing the CEO. ¹⁶⁵ There are no statutory rules regulating how boards vote and a show of hands is the usual practice. This practice combines with another: most business decisions are discussed and made by groups of senior executives, and only then brought to the board for approval. ¹⁶⁶ The effect of these combined practices is that it is very rare to see directors opposing proposals at board meetings.

Following the board’s rejection of his reshuffle plan, CEO Suzuki resigned. President Murata, who had been the second-highest-ranked manager behind Suzuki, led the discussion about the new management hierarchy. His proposed plan to stabilize the board would see all the other directors retaining their positions. Murata would take over leadership of the company and act as CEO going forward; Isaka, the executive whom Suzuki had sought to remove, would retain his positions as president and Chief Operating Officer (COO) of the subsidiary, as well as remaining a director of the holding company. However, before the meeting of the Nomination and Compensation Committee, two of the IDs, Ito and Yonemura, expressed opposition to Murata’s plan. Their view was that president Murata, who had supported CEO Suzuki throughout the attempted reshuffle process, should not remain in position either, presumably because they considered that Suzuki would continue to influence Murata behind the scenes. As a result of their opposition, Murata also resigned and these IDs, together with the remaining executives, agreed on a new management plan. This saw Isaka taking over as president of the holding company, Seven & i Holdings, whilst continuing as a director of the subsidiary, Seven Eleven Japan. Another executive became president of the subsidiary, Seven Eleven Japan. ¹⁶⁷ The founder’s son, Junro, took up a senior executive position and is reportedly expected to succeed Isaka as the president. ¹⁶⁸

With the new management plan agreed, the IDs also sought to eliminate any continuing influence of Suzuki in the new management structure. As noted above, the practice of appointing advisers from among retiring managers has been very controversial, because they go on to influence current management. At the Nomination and Compensation Committee, the IDs had already opposed Murata’s proposal to give outgoing CEO Suzuki the title of saiko-komon (chief adviser) on the basis that he would continue to exercise inappropriate influence over the company’s affairs. ¹⁶⁹ Ultimately, he was allowed to retain the title of meiyo-komon (honorary adviser), but the new president demanded that Suzuki’s office be located outside the headquarters building. ¹⁷⁰

In mediating this dispute, the IDs arguably balanced the competing interests of various team members, rather than allowing one particular faction to triumph. It is clear from his background that the ID, Kunio Ito, considered the interests of shareholders. He was a professor of accounting and corporate governance, and had led a government project, whose final report emphasized that Japanese companies should increase their focus on the capital market and on creating shareholder value. ¹⁷¹ According to the so-called “Ito Report,” Japanese managers had focused primarily on supplying high-quality goods and services to their customers, as well as providing employees with long-term employment, but had given the interests of shareholders insufficient attention. The report argued that, in order to attract capital investment and to achieve persistent growth, companies needed to set a target return on equity of at least 8% and encouraged companies to make good use of IDs to achieve this target. Interviewed about his role in the Seven & i Holdings dispute, Ito remarked that he had opposed the management reshuffle plan because, under Isaka’s leadership, the Seven Eleven subsidiary had marked record operating profits. ¹⁷² At the same time, we can infer that the IDs considered the interests of both executive factions and placed importance on keeping most of the management team together. Hence, the IDs did not award an outright victory to one faction over the other, allowing five executive directors who had backed Suzuki’s reshuffle plan to remain on the board, and even permitting former CEO Suzuki to retain an honorary position, albeit one that sounded weaker than originally proposed. ¹⁷³

Three features are worth noting. First, this “vertical dispute” had developed between two management factions: one led by the CEO and the other led by another executive who had support from the founder family. As we have seen, Blair and Stout envisaged this type of dispute being resolved by the CEO at the top of the management hierarchy, with more junior executives either abiding by the CEO’s decision or leaving the company. In the Japanese context, this exercise of strict legal powers would be inappropriate because it will destroy the FSHC of executives who are forced to leave. Instead, the IDs had to step in to prevent the CEO from using his power to remove another executive who had spent his whole career at the company. By doing so, the IDs were able to hold the team together for the benefit of both managerial factions, the employees, the shareholders, and the company.

The second notable feature is the way in which the IDs mediated the dispute using both formal and informal methods. In one sense, this dispute was unusual because it was put to a formal vote at a board meeting, with the IDs exercising their voting rights. At the same time, however, the IDs also exercised influence informally over the outcome of the dispute through a number of channels. First, while the Nomination and Compensation Committee had no legal authority to determine the matter, their rejection of the reshuffle plan arguably influenced the board’s decision to reject it. Introduced only a month earlier to comply with the recommendation of the JCGC 2015, ¹⁷⁴ the committee was merely an advisory body. Second, the IDs not only voted against the reshuffle in the board meeting, but also advocated the use of a secret ballot, which was similarly beyond their formal authority. Likewise, informal communications between executives and IDs caused president Murata to give up on becoming CEO and paved the way for his voluntary resignation. Third, even the change of top management personnel occurred in an informal way: the board did not have to dismiss CEO Suzuki because he voluntarily resigned when his reshuffle plan was rejected.

The final notable feature of this case is that the IDs made up only a small fraction of the board (less than one-third). However, despite their minority position, IDs held the balance of power because the two rival executive factions perfectly offset each other. This meant that, provided the IDs co-ordinated their votes, they effectively held a casting vote, allowing them to resolve a “vertical” dispute within the management team.

This type of compromise is more appropriate than a winner-takes-all solution to intra-management disputes at Japanese companies. Continued successful team production depends on the company’s continued access to FSHC, developed through whole careers with the company and embodied in the executive team members. Whilst it could be argued that this case fits with the principal-agent model, which expects IDs to replace poorly performing CEOs, here they did more than that. Certainly, they contributed to the CEO’s resignation and sought to restrict his continued informal influence; but they also balanced the interests of the two factions by allowing directors from both sides to remain in the management team. This outcome ensured the company’s continued access to the FSHC embodied in both factions, and is clearly compatible with the theory of team production, which emphasizes keeping the productive coalition together.

5.2 LIXIL Group

LIXIL Group is another good example of a vertical dispute between two factions of executives, and also highlights how disputes can be driven by post-merger dynamics. The company was founded in 2001 by consolidating two separate companies, TOSTEM and INAX. The company had been led from formation by Ushioda, who had been the CEO of TOSTEM and was a founder family member. He also held 3% of LIXIL Group’s shares. ¹⁷⁵ In 2015, Ushioda invited Seto to join the company from another company and run LIXIL Group as his successor, and Seto had been the CEO and president since 2016. LIXIL Group took the form of a company with three committees, having 11 directors: seven executives and four IDs. Its nomination committee consisted of five directors: three of them were IDs, whilst Ushioda, who was chairman of the board, was one of the two other members. In addition to its statutory power to nominate directors, the nomination committee had an advisory role regarding the appointment and dismissal of CEO. ¹⁷⁶

In 2018, two years after Seto had succeeded him as CEO, Ushioda used his control of the board to take the role of CEO back from Seto. This gave rise to a dispute between two factions—one led by Ushioda, the other led by Seto. Seto’s faction also included two other directors of LIXIL Group who had come from INAX: founder family member Keiichiro Ina and former president Ryuichi Kawamoto. ¹⁷⁷ The dispute played out through the process of board nominations. Seto nominated eight directors (himself, three other executives, and four new IDs); if appointed, this would have allowed his faction to wrest control of the board from Ushioda’s faction. In response, Ushioda nominated ten new directors (one executive and nine new IDs). Two of the independent candidates were nominated by both factions. These two candidates were first nominated by Seto, whilst Ushioda nominated them later, supposedly because he could not find allies who would qualify as IDs.

As the law requires, ¹⁷⁸ the slates of directors proposed by Ushioda and Seto were put to the shareholders for approval. As foreign investors (38.1%) and financial institutions (27%) held the majority of the company’s shares, ¹⁷⁹ each faction emphasized, as they solicited votes, that their nominees would improve corporate governance. ¹⁸⁰ In principle, it was possible for all 16 nominees to be appointed as directors since LIXIL Group’s articles fix the maximum number of directors at 16. ¹⁸¹ However, it is evident that each faction sought to exclude the other’s nominees by asking shareholders to vote for their nominees and against the other faction’s.

Seto’s faction criticized Ushioda for dominating the company; it was reported in an investigative report written by two lawyers appointed by the company ¹⁸² that, during his campaign to have Seto dismissed as CEO, Ushioda had misled both the nomination committee and the board. At the nomination committee meeting, Ushioda had falsely claimed that Seto had offered to resign as CEO; Ushioda then persuaded Seto to resign, misleadingly telling him that the nomination committee had unanimously approved his resignation. In fact, the committee’s approval had been conditional upon confirmation that Seto had actually stated his intention to resign. It is evident that the manner of Seto’s removal triggered a significant dispute between pre-existing factions, creating a crisis that the IDs subsequently had to mediate. Director Ina, who supported Seto, publicly claimed that Ushioda would influence the nomination committee in relation to the candidates to be put to the shareholder meeting. ¹⁸³ In addition, ten lower-ranking executives expressed support for Seto by demanding that the nomination committee should eliminate Ushioda’s influence. ¹⁸⁴

In the end, the shareholders approved all six directors nominated by Seto’s faction, six of the eight directors nominated by Ushioda’s faction, and the two jointly nominated directors who had by then announced their support for Seto. ¹⁸⁵ The result was a board with a majority of IDs, nine out of a total of 14, which then set about mediating the dispute. It placed executive directors from both factions in key positions, giving them influence over new managerial appointees: the board reappointed Seto as CEO and president; the board also appointed as vice president an executive director Ōtsubo, who had been nominated by Ushioda. Matsuzaki, one of the IDs who had been nominated by Ushioda, was appointed as chairman of the board.

This case marked an important moment in the evolution of Japanese corporate governance because IDs, who made up a majority of the board, actually had formal legal authority to resolve the dispute. Before the contest between the two factions to nominate new directors, it appeared that the IDs were ineffective. Despite consisting of a majority of IDs, the nomination committee had simply rubber-stamped the decision of the chairman of the board and a member of a founder family to approve the CEO’s resignation. At the newly elected board, however, the IDs stepped up and performed a mediating function. Rather than simply supporting the faction that had nominated them, it seems likely that, once they made up a majority, the IDs identified a solution that kept the team together. Even if that is not what happened, it is reasonable to infer that, faced with a board consisting of a majority of IDs, the executives had an incentive to negotiate and solve the dispute among themselves (as Blair and Stout suggested). After all, neither faction could take control of managerial matters without the board’s consent. Comments made by ID Matsuzaki after his appointment as chairman showed that the board had worked to mediate the dispute and preserve the integrity of the team. He said that the directors would overcome the dispute between the two groups and that the CEO Seto had understood both sides’ views. ¹⁸⁶

As in the case of Seven & i Holdings, IDs acted in the interests of various stakeholders. By acting strongly to reinstate Seto, who had improved the profitability of the group, and promising more monitoring in the future, they acted in the interests of shareholders. But it appears that the new board also took account of the interests of executives from both factions and the managerial hierarchies below them. Some support for this view can be drawn from the fact that two of the directors nominated by Ushioda took up important positions within the company. This balancing approach to the two factions highlights that executives with high levels of FSHC have a lot at stake in these disputes, and that post-merger disputes between executives originating in different companies can be deep-seated and particularly difficult to resolve. Given that these executives have specialized their skills to different parts of the post-merger group, it is essential that the team be held together following the merger in order for the new business to prosper. We submit that the importance of FSHC would have been well understood by the IDs, given their background as executives in other companies, ¹⁸⁷ and that this explains why the IDs acted to bring about a compromise solution to the dispute.

5.3 Seiko Holdings

Seiko Holdings provides another example of a dispute between founder family and executive factions, which was reported as “a boardroom coup” by the UK business press. ¹⁸⁸ However, the essence of the dispute was a conflict between a senior executive allied with the founding family and a group of employees and managers who had complained about harassment. The case is important because it illustrates how employees and more junior managers can work with IDs to exercise significant influence over the company’s affairs, and management succession in particular.

Seiko Holdings is the holding company of a group that includes a watchmaker, an electronics component maker, and other businesses. One of the subsidiaries is Wako, a luxury department store. The holding company took the form of a company with a board of statutory auditors. It had seven directors: six executive directors and one ID. Its board of statutory auditors consisted of five statutory auditors: two were former employees of the company’s subsidiary; three were outside statutory auditors. ¹⁸⁹ While founder family members held more than 15% of the shares in total, the remainder of the shareholders in Seiko Holdings were dispersed: financial institutions held 13%; other types of companies held 28%; foreign investors held 9%; and individuals (excluding the founder family members) held more than 31%. ¹⁹⁰

At the time of the dispute, the company was officially led by chairman and president Koichi Murano. The second-most-senior manager was vice president Shinji Hattori, who was a founder family member and also held around 6% of the company’s shares. ¹⁹¹ His uncle Reijiro Hattori, another founder family member who held around 10% of the company’s shares, was meiyo-kaichō (an honorary chairman) without official directorship or executive authority. He was also chairman and president of the subsidiary, Wako. One of the executive directors of Wako was Noriko Unoura, who was Reijiro Hattori’s protégée. She was also an executive director of Seiko Holdings. ¹⁹²

The dispute became public in 2010 when Seiko Holdings announced that the president, Koichi Murano, had been dismissed at a board meeting: out of six directors, five of them had voting rights, excluding Murano who was prohibited by law from voting on his own dismissal ¹⁹³ ; three out of the five eligible directors voted for Murano’s dismissal. ¹⁹⁴ Newly appointed president Shinji Hattori explained that Murano had been blindly following the instructions of honorary chairman Reijiro Hattori. ¹⁹⁵ Following Murano’s dismissal, both the chairman and president of Wako, Reijiro Hattori, and an executive director, Noriko Unoura, were also dismissed from the board of Wako. ¹⁹⁶ Shinji Hattori explained that a number of employees had complained about harassment by Unoura ¹⁹⁷ and that four managers, including two very senior ones, had been forced to resign in the middle of their terms of office. ¹⁹⁸ It was reported, based on the company’s independent investigation committee report, which is not publicly available, that more than 50 employees suffered from depression because of the harassment. ¹⁹⁹

The trade union, which held shares in the company, played a key role in the process that led to Reijiro Hattori’s dismissal. It demanded that the company’s statutory auditors sue the company’s directors, including Murano, Unoura, and Shinji Hattori, for extensive damages arising out of abusive and inconsistent management, as evidenced by the executive resignations, as well as the harassment allegations that had been made against Unoura. ²⁰⁰ The aim of the demand was to put pressure on the directors to oust Reijiro Hattori. Whilst Kaishahō (the Companies Act) allows shareholders to bring a derivative action if the statutory auditors refuse to do so, ²⁰¹ it would not technically have been possible for the trade union, as a shareholder, to sue Reijiro Hattori, because he was the honorary chairman rather than a director. ²⁰² Reports suggest that this strategy was effective and that Shinji Hattori sought to eliminate Reijiro Hattori’s influence in line with the trade union’s demand. ²⁰³ As such, a serious vertical dispute developed between the honorary chairman and his followers on one side and another senior executive, his allies, and the company’s employees on the other.

Besides the trade union, Seiko’s sole ID in 2010, Akio Harada, also played a critical role in finding a solution to the dispute. A former prosecutor, Harada proposed that the board should dismiss president Murano because he had allowed Reijiro Hattori and Unoura to influence executive appointments. ²⁰⁴ Harada voted with Shinji Hattori and his ally Nakamura for the dismissal of Murano, and to appoint Shinji Hattori in his place, by a margin of three to two. ²⁰⁵ In this case, as in the case of Seven & i holdings, the mediating function of the ID was made possible because the two factions offset each other, giving the decisive vote to the only ID. At the request of Shinji Hattori, Harada also asked Reijiro Hattori to resign as the chairman and president of Wako prior to the board meeting. When Reijiro Hattori refused, the dismissal of Murano as president of the company and of Reijiro Hattori and Unoura as executives of Wako followed. ²⁰⁶

Unlike the two other cases discussed above, here, the ID ultimately resolved these disputes in a more formal way: he proposed and voted for the dismissal of the president. We suggest that he did this not only because the honorary chairman and his followers had undermined the productive coalition within the company by abusing their (formal and informal) power, but also because the honorary chairman refused to step down following informal communications from the ID. ²⁰⁷ This left Harada as the ID with no option but to exercise his formal authority so as to resolve the disputes between management and employees.

It is recognized in the West that IDs can play a role in preventing the CEO from dominating the board by, for example, playing the role of non-executive chairman. However, it is much less likely that IDs in the West would intervene to resolve disputes between the top level of management and the company’s employees. Harada’s intervention is a clear indicator of the importance of social norms of lifetime employment, as well as the danger posed by management that violates the social norms of the “company community” by failing to pay attention to the views of the employees. Reportedly, one of the major reasons for Shinjiro Hattori to take action was to prevent the harassment of employees, driven by the trade union’s actions. Although the ID, Harada, did not explain his role in mediating the dispute, it is reasonable to infer that he advocated for the employees by supporting Shinji Hattori.

6. Conclusion and the future evolution of Japanese corporate governance

This article has suggested that, in practice, and in addition to the formal functions assigned to them by the JCGC, IDs in Japan often play a vital mediating role. In our typology of disputes that arise within Japanese companies, two or more groups of executives often end up competing with each other for control of the company or corporate group, leading to deep-seated factionalization. Such disputes have potentially very serious implications for companies since executives and employees have normally invested heavily in FSHC. This means that they want to retain their positions, they want their companies to prosper and be sustainable, and they want to be able to exert sufficient control to ensure they can achieve these goals.

The case-studies we developed in the final section highlight the actual mediating functions performed by IDs when they are facing such internal disputes. We showed that this mediating hierarch function is somewhat similar to that identified by Blair and Stout in the US system, but that it differs in certain key respects, especially in that it extends to “vertical” disputes within the managerial hierarchy.

We recognize that there are limits to the case-study method. Case-studies cannot conclusively demonstrate that IDs invariably play this mediating role in relation to vertical disputes; they can only demonstrate that IDs have played this role, and that this fits with other aspects of the Japanese corporate governance system. We also accept that our case-studies are outliers in the Japanese corporate governance system in the sense that they were witnessed by the public; it is surely more normal for team members’ interests to be co-ordinated, and disputes resolved, behind closed doors. This will normally be done by the team members themselves, but we strongly suspect that IDs may be playing an informal mediating role in private too. Perhaps our examples and analysis will serve as a starting point and inspiration for larger-scale, qualitative empirical research that opens up what happens outside the public gaze.

Moving forward, this question is only going to become more important. These types of dispute will arise more frequently as share ownership by institutional investors increases, allowing executives who oppose incumbent CEOs to compete for corporate control by soliciting support from shareholders: one such case recently reported is Sekisui House Ltd in which the former chairman attempted to regain his power by nominating directors. ²⁰⁸ At the same time, as our case-studies show, the need to gain shareholder approval of any boardroom “coup” is likely to result in the appointment of more IDs. However, they also show that those IDs are likely to take an approach that seeks to keep the team together and preserve FSHC, rather than simply acting as the monitors and strategic advisers expected by principal-agent theory.

More generally, the mediating role of IDs can create the conditions for executives themselves to find appropriate solutions to their disputes. IDs certainly consider which of the factions is more likely to enhance the interest of shareholders, either because they are attuned to the interests of founder families who still own large blocks of shares or because they are dependent on institutional investors who collectively have control over their appointments. However, they also consider the interests of other stakeholders, in particular executives and employees, who, pursuant to the social norm of lifetime employment, have invested significantly in FSHC. Those interests must be protected, alongside those of shareholders, if team production is to continue to contribute to the success of the company. The career background of the typical ID discussed in Section 4 makes it highly probable that they take account of this. A majority of IDs have been socialized and have typically spent their working lives in a single company, and so are likely to understand the importance of FSHC to executives, employees, and companies alike. Likewise, they are able to recognize opportunistic behaviour that poses a threat to FSHC and their appointment to the position of ID allows them to take steps to prevent it.

While this article has provided a descriptive account of how IDs mediate vertical disputes in Japanese companies, we conclude with two normative suggestions as to how Japanese corporate governance might evolve.

First, as regards the composition of the board, most listed companies appoint at least one ID, but they still make up a relatively low proportion of the board, at least compared with the West: as of 2021, only 6.3% of listed companies had a board with a majority of IDs, whilst in around half of listed companies, IDs made up more than one-third of the board. ²⁰⁹ As this article has shown, IDs can perform mediating functions even when they are in a minority, especially when they are faced with an internal dispute between two relatively evenly matched factions. This observation provides a degree of normative support for the current Japanese provisions, which do not require companies to have a board with a majority of IDs. Indeed, once the mediating role of IDs is recognized, it might even be argued that the board should be required to include a certain number of executives. This would run counter to the emerging practice in American listed companies of appointing so-called “supermajority independent boards” that include only one executive (the CEO), with the remainder being IDs. ²¹⁰ This American practice seems inappropriate for the Japanese context, where IDs need clear routes of communication to all management factions in order to perform their mediating function. At the same time, we do not expect that we will witness more disputes and mediation in the future. Feuding executives would rather negotiate with each other behind closed doors, with IDs only acting formally where a negotiated solution is not forthcoming.

Second, the JCGC and other sources of regulation should arguably articulate the mediating function of IDs more clearly. As we have seen, the JCGC explicitly expects IDs to provide advice; monitor, appoint, and remove senior management; monitor conflicts of interest; and appropriately represent minority shareholders and other stakeholders independently from the standpoint of management or majority shareholders. These provisions treat executives as the object of monitoring by IDs, but do not explicitly identify executives as having an interest in the continuity of team production. It would be desirable for the JCGC to reference executives’ (and other managers’ and employees’) investments in FSHC. ²¹¹ Similarly, the role of IDs in maintaining and fostering consensus between different groups in the long-term interests of the company should also be made explicit. JCGC 2021 already articulates the relationship between listed companies and a wider range of stakeholders. First, listed companies are expected to recognize the contribution of, and endeavour to co-operate appropriately with, stakeholders such as employees, customers, business partners, creditors, and communities. ²¹² Second, the board is responsible for drafting a code of conduct that employees are expected to abide by as regards co-operation with stakeholders. ²¹³ Third, boards are expressed to be responsible for considering sustainability issues including environmental problems, the fair and appropriate treatment of employees, and fair transactions with customers and suppliers. ²¹⁴ However, JCGC 2021 does not mention the role of IDs in mediating disputes and maintaining the integrity of the productive coalition. Explicitly articulating IDs’ mediating function would legitimize, and encourage them to play, this essential role in supporting team production.