Public Comment to the Proposed Revisions to Japan’s Governance Code – Nicholas Benes

by Nicholas Benes (as an individual)
April 27, 2018

1. Regarding the Overall Revision Process
2. Regarding Principle 2-6 (Activating the Function of Corporate Pension Funds as Asset Owners)
3. Regarding Principle 1-4 (“Policy Shareholdings”)
4. Regarding Principles 4-1③,4-3② and 4-3③ (Appointment and Termination of the CEO)
5. Regarding Principle 4-10① (The Use of Optional Structures)
6. Regarding Principle 4-14 (Training of Directors and Kansayaku)
7. Regarding Revision of the Machine-Readable Format of Corporate Governance Reports

(Note: This is a translation of a public comment which was originally written in Japanese and submitted in that form to the JPX/TSE.  The original version of the public comment is available here.)

1. Regarding the Overall Revision Process

I would like to express my thanks and appreciation for the hard work of the members of the Follow-up Committee with respect to this review of the Corporate Governance Code (the “CG Code”) . However, I would note that four years have elapsed since the initial drafting of the Code. As you know, in Germany there is a commission which monitors the effectiveness of the governance code on an ongoing basis, and proposes changes on a yearly basis if and as necessary. Following the same philosophy that has been cultivated in Japan’s manufacturing sector to “create the best possible products” and strive for “continuous improvement” (kaizen), I believe that the timing and process for future reviews and revisions of the Code should set in advance (that is, should be set at this time). If this is not done, we will wake up one day and discover that the review process has become completely dependent on the initiative of bureaucrats. Therefore, the introductory section of the CG Code (“Regarding the Corporate Governance Code”) should specify a schedule for the subsequent revision(s) of the Code.

2. Regarding Principle 2-6 (Activating the Function of Corporate Pension Funds as Asset Owners)

I am very pleased that Principle 2-6, “Activating the Function of Corporate Pension Funds as Asset Owners”, has been added to the CG Code. However, it is unfortunate and rather strange that there is absolutely no mention of the Stewardship Code of Japan (the “SC Code”) in this section, despite the wording in the draft about the operation of corporate pension funds and their “stewardship activities to monitor fund managers”.

Looking back, my own proposal (Footnote 1) made to the Ministry of Health, Labour and Welfare (MHLW) in August of 2016 was the catalyst that brought about the formation of the “Stewardship Study Group” by MHLW, the Pension Fund Association, the Financial Services Agency (FSA; as observer) and experts in the field. As a result, when the Study Group released is voluminous report and attached materials in March, 2017, for the first time the MHLW made an announcement clarifying that its policy was to encourage corporate pension funds to sign the Stewardship Code. (Translator’s note: this was a full three years after the SC Code was promulgated.)  

As is well known, it is expected that that CG Code and the SC Code will function as the “two wheels of a cart” (Footnote 2). Yet up until the MHLW’s announcement in 2017, few corporate pension funds had signed the SC Code. (In fact, among the pension funds of non-financial corporations only Secom’s had signed the SC Code.) Following the MHLW’s decision, at long last more corporate pension funds have now signed the SC Code, even if the total number is still small.

Considering these developments, now is the clearly a time when the opportunity in this area looms large. Therefore, it is extremely strange to me that it was not made clear in the draft revised CG Code that corporate pension funds should sign the SC Code. If this were to be clarified, those companies with pension funds that have not yet signed the SC Code would have to explain the reason why they have not signed it, in accordance with the “comply or explain” principle. As a result, one of the important stakeholder groups for companies – their employees – would be able to confirm the important stewardship activities of the company managing the assets that will pay for their living expenses after they retire.

For the revisions to encourage “stewardship activities” without explicitly mentioning the SC Code itself, is tantamount to superficial “encouragement” without setting forth any real substance.

3. Regarding Principle 1-4 (“Policy Shareholdings”)

In Japan, “policy shareholdings” that have the effect of undermining governance are still at a high level (more than 30%; see Footnote 3). I agree with the thrust and content of the revised version of Principle 1-4 (“Policy Shareholdings”), but would be grateful if the following two suggestions could be considered.

The first point relates to the definition of “what constitutes a policy shareholding”. In their Securities Reports, companies are required to disclose details about the equity securities that they “hold for purposes other than pure investment”. (Please refer to “Instructions for Filling out Form No.2 as Required by the Ministerial Order Regarding Disclosure by Corporations, Item (56)a (e).)  I believe that the term “policy shareholdings” in the CG Code refers to these holdings. However, in order to prevent individual companies from applying their own selective interpretations, this definition should be explicitly set forth in the CG Code.

The second point relates to the content of such disclosure. In order for investors and companies to engage with each other at a deeper level than at present, and in order to enable concrete discussion focusing on the necessity of individual stock-holdings, more detailed disclosure than at present is needed. For this reason, I would like to request that the following three aspects be disclosed with respect to each stock-holding [of a listed company]:

(1) Whether or not a concrete business alliance with that company exists.
(2) Whether or not the total revenue from transactions received from that company during the preceding fiscal year exceeded the book value of the stock-holdings.
(3) Which of the following categories describes the stock-holding: (a) ongoing customer; (b) potential customer; (c) group company; or (d) other.

I would like to explain a bit more about (2) above. From an investor’s point of view, one can determine that there is a certain degree of logic or rationality to holding a stock if that [investee] company is the source of enough profit to compensate for the cost of capital and the investment risk. Having said that, requiring disclosure of the amount of that amount of profit (or the revenue from which it arises) could be problematic in a number of ways. But we should at least set a rough yardstick and require disclosure of whether or not that level has been exceeded. I set forth (2) above as it reads because I believe that companies should at least be receiving annual revenues (the actual profit from which will of course be smaller) greater than the book value of any related shareholdings. (See Footnote 4.)

The aforementioned three items (1) through (3) should of course be disclosed in “Yuho” Securities Reports, and one would hope that ultimately the related laws would be amended so as to require that. In the meantime, I would request that the CG Code should specify that disclosure should be made simultaneously in both Securities Reports and in another appropriate manner.

4. Regarding Principles 4-1③,4-3② and 4-3③ (Appointment and Termination of the CEO)

Principles 4-1③,4-3② and 4-3③ are revisions that reflect the importance of appointment and termination of the CEO at companies. Moreover, the attached draft “Guidelines for Investor and Company Engagement” repeatedly stresses the importance of CEO appointment and termination. I am of course in agreement with these revisions as far as they go, because it is absolutely true that CEO appointment and termination is vital.

However, when malfeasance occurs, and it becomes necessary to replace the CEO, in reality a difficult problem arises. Namely, at Kansayaku-style companies and Audit Committee-style companies, under current Company Law the CEO can only be selected [by the board] from among the directors who have already been approved by shareholders. Therefore, even though the company can respond to the malfeasance by terminating the current CEO, until such time as it can hold an extraordinary shareholders meeting (Footnote 5) in order to elect new directors, the company can only appoint a new CEO from among the currently serving directors – that is, from among a group of persons who are colleagues of the fired CEO, who may have been involved in the same malfeasance.

Under such conditions, the board will be unable to rapidly reform management, and its response to the malfeasance or scandal will be significantly delayed. Principle 4-3② of the draft revision of the CG Code requires the board to appoint the “a quality CEO” using procedures that are “timely”, but in practical reality it is extremely difficult to actually do this. At companies facing strategic crisis because of “legacy” issues, the same sort of problem occurs.

The most effective way to address this problem would be to make it possible to appoint persons other than directors as the CEO. METI recommended this very solution (Footnote 6) in its memo entitled “Opinion Concerning the Issues We Believe Affect Discipline in Corporate Governance”, which it submitted to the Corporate Legal System Subcommittee of the Ministry of Justice (MOJ) on April 26th, 2017. METI’s opinion proposed: “at Kansayaku-style companies and Audit Committee-style companies, to make it possible for boards to appoint executive officers (shikkoyaku or shikko yakuin) having executive responsibilities, and to select an executive officer representing the company [a CEO] from among such persons.” (See Footnote 7.)

If the law were amended in this manner, it would become possible to reform or change management as necessary, and boards could focus more on monitoring senior executives. Moreover, it would fix the legal problems arising from “shikko yakuin” that have been criticized in the past, such as the fact that shikko yakuin do not carry fiduciary duties and cannot be sued by shareholders. (Translator’s note: at present, without the aforementioned legal amendment, “shikkou yakuin” are simply employees that have a misleading title that makes its sound as if they are directors, when in fact they are not.)  

I understand that the draft revisions to the CG Code were prepared by the JPX and the FSA, but I would greatly appreciate it if the JPX and the FSA would closely consider and actively voice their support of METI’s proposal to amend the Company Law, in order to ensure that the “timeliness” stressed so much in the revised CG Code is indeed possible in practical reality.

5. Regarding Principle 4-10① (The Use of Optional Structures)

As is correctly set forth in Principle 4-7, independent directors are expected to serve many different roles and responsibilities. One of the very most important of those responsibilities is to ensure the objectivity and transparency of decisions regarding the nomination and compensation of executives. (This is the general view in foreign countries as well.)

It might be said that this aspect is at the very heart of the current revision of the CG Code.  For instance, Supplementary Principle 4-2① has been modified to make it clear that decisions about the compensation of executives should be made “according to procedures that are objective and transparent”, and Supplementary Principle 4-3② has been added to make it clear that the appointment or termination of the CEO should be made “following procedures that are objective, timely, and transparent”. I much agree with these revisions of the Code which require objectivity and transparency with respect to the nomination or compensation of senior executives.

However, notwithstanding this appropriate emphasis, the draft leaves ample room for the direct and active involvement of senior executives in decisions about the nomination or compensation of senior executives.  Moreover, Supplementary Principle 4-10① states nothing more than “[committees] in which independent outside directors make up the principal members”, and therefore does not deny the possibility that the CEO or other senior executives can be members of such committees.

That sort of participation might be deemed acceptable [as not posing a problem] in countries such as the UK, where the labor market is more mobile than in Japan, and where the evaluation of performance based on results is firmly rooted. But in Japan, although there have been minor recent changes, the “seniority system” (promotions and hierarchy according to age) is still the dominant paradigm, and the external market for labor and executives is not active and flexible. As a result, executives and directors who are promoted internally have a strong sense of dependence on the company, and frequently face an authoritative, strict hierarchy. Moreover, because practices to develop and train personnel so that they can appropriately fulfill the role of independent director are not very advanced, in many cases the independent directors who are appointed have some kind of pre-existing relationship with senior executives.

Under these conditions, if senior executives can directly participate in decisions about nominations or compensation as regular members of such committees, it will be impossible to ensure the objectivity and transparency of decision-making processes. How many committee members will not hesitate to ask a CEO serving as such a regular member to leave the room in order to discuss that CEO’s own re-nomination or compensation?

In order for the revisions to the CG Code to have a more positive impact, I believe that that the Code should clarify that executives should not directly participate in the decision-making process regarding the nomination or compensation of executives, irrespective of which of the three types of corporate governance structure defined in the Company Law is used by the company. This means that “companies with a nominations committee, etc.” (“three-committee-style companies”) should be added to the scope of Supplementary Principle 4-10①. It also means the CG Code should specify that regardless of the “legal” form of corporate governance that is chosen (and even if independent directors make up a majority of the board), matters concerning nominations or compensation of executives should be decided by nominations committees and compensation committees that only include independent directors. (In the case of a “three-committee style” company, this refers to the committees that are legally required; and in the case of Kansayaku-style companies and audit committee companies, this refers to extra-legal “advisory” committees.)

In order to enable these various committees to be composed solely by independent directors, the sentence in Principle 4-8 which presently states that “at least two independent directors should be appointed” should be changed to read “at least three independent directors should be appointed.”

Irrespective of whether the aforementioned revisions are made or not, I would also like to request a change in the wording of Principle 4-10①. Principle 4-10① currently includes the words, “…by means of setting up independent advisory committees such as ‘optional’ nominations committees and compensation committees that are mainly composed of independent directors, etc”. I understand that this clause is calling for the establishment of “an advisory committee that is mainly composed of independent outside directors”, regardless of the exact name of that group.  However, because of the present wording it is possible for companies to misinterpret things and claim that they comply with the CG Code if they merely set up a committee that is called an “independent advisory committee”, even if independent outside directors are not its principal [majority] members.

Lastly, the fact that the word “optional” is used in the wording of this Principle creates the risk that since the Code itself clearly says “optional”, some companies will make their own “selective interpretation” that they do not violate the Code’s Principles even if they do not set up any committee at all. Therefore, I would appreciate it if wording like this were changed so that it does not bring about misunderstandings based on such misinterpretation. Along the same lines, in the upcoming revised Code it makes all the more sense to delete words such as “for example” and “etc.”.

6. Regarding Principle 4-14 (Training of Directors and Kansayaku)

In Section 3 of “The Thinking Underlying Drafting of the Revision of the Code and Engagement Guidelines” related to the upcoming CG Code revisions, issues are raised such as “it has been pointed out that at many companies, the system for developing and appointing CEO talent is insufficient” or “few companies’ boards sufficiently engage in monitoring, even regarding succession planning.” Principles 4-3② and ③ have been added to address these points, but unfortunately, despite the fact that the promotion and strengthening of director training could contribute greatly to the solution of these problems, Principle 4-14 (Training of Directors and Kansayaku) was not revised at all.

For example, the very policy of requiring employees (before they become directors) who later may become candidates for the position of CEO to acquire knowledge that is essential in order act as a CEO or a director, is a valuable way to prepare for the nomination and development of future CEOs. Moreover, through such training processes, it is possible to better discern the abilities and “quality” of each such employee. By preparing management in advance in this way, it becomes possible to facilitate smooth transitions between CEOs.

For these reasons, in order to achieve the goals of Principles 4-3② and ③ in practical reality, it would be logical to add more detail to Principle 4-14.

First, I recommend that directors and kansayaku at significant subsidiaries, and “shikkou yakuin” executive officers, should be included within the scope of Principle 4-14. To provide training to persons at the level of “shikkou yakuin” is completely in line with the issues and thinking described above. Moreover, in view of the fact that there are many holding companies, and considering the reality that many recent cases of malfeasance have occurred at subsidiaries, I believe it goes without saying that board members at significant subsidiaries should be included as a group that receives director training,

Second, it should be clarified that disclosure is required about concrete aspects of training.  As the Representative Director of The Board Director Training Institute of Japan (BDTI; a certified “public interest” nonprofit), I have met with many companies to discuss the topic of training for board members, and have provided training in many cases. From these activities I have realized with dismay that persons who do not have the minimum knowledge needed to fulfill their role as board members are on the boards of listed companies (or their subsidiaries) in this nation which is the third largest economy in the world.  In the feedback surveys that are filled out by participants at BDTI’s one-day training sessions, we often receive responses such as “I now understand the role of board member” or “I now realize how much I lack sufficient knowledge about finance and financial statements”. Such responses vividly reveal that even now, director training is insufficient.

Similarly, the article “After Implementation of the Corporate Governance Code – Looking Forward: Issues Raised by the Results of a Survey about Director Training “, (in the magazine “Keiei Houyuukai Report”, February 2016), came to the conclusion that: “the incidence of director training was 53.9% for internally promoted board members at respondent companies, and 28.9% for outside directors. Considering that 73.3% of the respondent group consisted of companies with consolidated revenues of at least 100 Billion Yen, it is hard to escape the impression that these levels are low” . (See Footnote 8.)   The survey also showed that at 50% of respondent firms, the length of director training [even when it existed] was less than three hours, and that training about important topics(especially those related to “governance practice” and “finance”) was not common. For example, at only 16.2% of respondent firms of did training cover competition law; at only 25.3% did it cover risk management and crisis management; at only 13.1% did it cover the CG Code; at only 21.2% did it cover finance and accounting; and at only 7.1% did it cover corporate value and valuation topics.

Despite the fact that director training is required by the CG Code, director training at most companies is completely insufficient. One reason for this is that Supplementary Principle 4-14② merely requires companies to disclose their “policy with regard to training”.  Unfortunately, it seems clear that to many companies, the stance taken by the Code signaled that “because all the CG Code requires is disclosure of ‘a training policy’, all the Code expects us to do is set forth any sort of ambiguous policy.”

In order to break this impasse, the CG Code should require more disclosure about concrete aspects of training. Specifically, Supplementary Principle 4-14② should be revised (or another Supplementary Principle added) to require companies to disclose each year with respect to each type of board member or “shikkou yakuin“: (a) the number of persons who received training; (b) the average number of hours of training per person (for each of interactive methods and non-interactive methods such as e-Learning); (c) the subjects (topics) covered by the training, and a summary of the content; (d) whether or not external specialists or experts were used; and (e) the number of persons who were reimbursed for training supported by the company, and the total amount of such reimbursements.

7. Regarding Revision of the Machine-Readable Format of Corporate Governance Reports

The CG Code was proposed with the clear goal of enabling the Stewardship Code to function efficiently by means of requesting each company to disclose its governance framework and the concrete practices it uses, thereby enabling investors to easily conduct comparative analysis of this information (Footnote 9). This was necessary and essential in order for constructive engagement to take place between shareholders and corporations. In point of fact, in the CG Code many Principles were set forth for the express purpose of requiring disclosure.

However, in large part this goal has not been achieved. The reason for this is that the Corporate Governance Reports submitted by each company do not enable the easy collection of the information contained in them, and its comparative analysis.  Compared with the codes [or listing requirements] in the UK or the US, the language of Japan’s CG Code is considerably more ambiguous, partly due to the fact that in Japan companies can choose from three different legal frameworks for their governance, – the Kansayaku-style company, the Audit Commitee company, and the Three-Committee-style company. Because of this ambiguity, if a company merely discloses that it “complies with the CG Code”, it is difficult to know what kind of practices are actually being utilized at that particular company.  Therefore, it is exceedingly important to require disclosure about what exactly is occurring with respect to each item and thus facilitate the comparison and analysis of that information.  Only by doing so will it become possible – for the first time – to understand which governance frameworks and practices correlate with firm performance and to discern which companies’ structures are superior and which are insufficient, thereby leading to constructive engagement between investors and companies. If inter-firm comparisons are not made and such analysis is not done, companies will have no incentive to seek to identify “best practices” in order to further improve their governance, and their corporate governance quality will necessarily decline.

From my point of view, as the person who originally proposed the CG Code to the government with this vision of comparative analysis between firms in mind (Footnote 10), it is exceedingly disappointing that it is not easy to collect, compare, and analyze the information contained in Corporate Governance Reports.  In order for the “virtuous cycle” between the Stewardship Code and the Corporate Governance Code to arise as it was originally conceived, I strongly request adoption of the following measures in order to facilitate the easy collection, comparison, and analysis of all the data in the Corporate Governance Reports submitted by each company

First, currently the 11 so-called “disclosure items” set forth in the CG Code (explanations of 11 different and distinct policies) are not treated as information pertaining to each policy.  Instead, in CG Reports they are all just bundled together as if they were all a single information item.  (There is only one XBRL tag for all of them, as single group.)  This format for providing the data must be modified.  It would be a simple matter to provide a separate XBRL tag for each separate disclosure item.  This would not require much cost, and no one would be harmed as a result.  This should be done immediately.  (The current usage of XBRL is extremely unusual and queer. Despite the fact that XBRL tags exist in order to separate out distinct data items so they can each be identified individually, in this case the usage does the exact opposite:  XBRL is currently being used to make the data UN-readable by a computer because it mashes together completely different data items.

Second, presently the rules for the preparation of Corporate Governance Reports permit firms to “disclose” by simply referring to company web site URLs, or to set forth “disclosure item” information in sections of Reports other than the “disclosure item” section (again, without any specific XBRL tags), rather than directly explaining. This also makes the automated collection of this data using computer systems extremely difficult. The “Instructions for Preparing Corporate Governance Reports” should be amended so as to require that information must be disclosed in sufficient detail and in the appropriate section of the Report.

Because we are now in the age of AI, “making it easy to to collect data and do comparative analysis” necessarily means “providing data in a format that enables that to be done by a computer, in order to make the process vastly more efficient.” Unless this is done, not only will JPX/TSE not have achieved the original objective of the CG Code, but Japan will fall behind compared to the rest of the world.

( The original Japanese version of the public comment is available here.)



1)   “A Concrete Proposal to Strengthen Pension Governance in Order to Make the Stewardship Code Function More Effectively”, by Nicholas Benes, August 20, 2016 (「スチュワードシップコードが実効的に機能するために、年金ガバナンス強化の具体策を提言する」、ニコラス ベネシュ、 ).

2) “The Third Arrow that Will Not Require a Budget, But Will Be Most Highly Evaluated — Promulgation of a ‘Corporate Governance Code to Protect Investors by Establishing a Framework for Information Disclosure”, by Nicholas Benes, October 24, 2013 (「予算が要らない、最大に評価される三本目の矢 ― 投資家保護のため、情報開示の枠組みである『コーポレート・ガバナンス・コード』の制定」、ニコラス ベネシュ、平成25年10月24日、 ).

3) ”Allegiant Shareholders in Japan”, by Dr. Ryoko Ueda and Ken Hokugo, Governance, July 16 #265, p.7-8、 .

4)This is based on the following logic: given that on gross margins in Japan on average run in the range of 15-20%,and rarely exceed 50-60%, unless annual revenue exceeds the book cost of a stock-holding, in almost all cases the probability will be close to zero that the contribution to net income could be enough to cover for the cost of capital and the risk of making a minority investment.

5) Because convening an extraordinary shareholders meeting entails costs, management time, confusion, and a sense of shame, it goes without saying that any board of directors will want to avoid it.  For this reason, the chance that the decision will simply be delayed [until the next AGM] is high.

6) See the document submitted by METI at: , pages 1 and 2.

7) Recently, METI repeated its proposal in a memo submitted to the Ministry of Justice on April 13,2018, “Opinion Regarding the Interim Proposal Regarding the Corporate Legal System (Related to Corporate Governance), (「経済産業省は最近には「会社法制(企業統治等関係)の見直しに関する中間試案に対する意」).

8) “After Implementation of the Corporate Governance Code – Looking Forward: Issues Raised by the Results of a Survey about Director Training”, by Takayuki Kimura, in Keiei Houyuukai Report, February 2016, (「役員研修に関するアンケート結果の分析と今後の課題 CGコードの適用を受けて」、木村孝行氏、経営法友会リポート2016年2月号).

9)”To Revive the Japanese Economy, Promulgate a Corporate Governance Code Soon”, page 2 (presentation by Nicholas Benes to the Liberal Democratic Party of Japan’s “Japan Economic Revival Headquarters” (Financial Research Group Joint Session), February 6, 2014),  (「日本経済の復活のため、コーポレート・ガバナンス・コードの早期制定を」、(ニコラス ベネシュの日本経済再生本部(金融資本市場・企業統治改革グループ)・金融調査会 合同会議へのプレゼン資料、2014年2月6日)、ページ2、 ).

10) ”To Revive the Japanese Economy, Promulgate a Corporate Governance Code Soon”, page 2 (presentation by Nicholas Benes to the Liberal Democratic Party of Japan’s “Japan Economic Revival Headquarters” (Financial Research Group Joint Session), February 6, 2014),  (「日本経済の復活のため、コーポレート・ガバナンス・コードの早期制定を」、(ニコラス ベネシュの日本経済再生本部(金融資本市場・企業統治改革グループ)・金融調査会 合同会議へのプレゼン資料、2014年2月6日)、ページ2、 ).

Leave a Reply