Putting rights in reverse, Part 2: The data do not support change

Background

In early June, Firefly wrote about a major series of proposed amendments to company law in Japan titled, “Putting rights in reverse -- where Japan is heading on Companies Act changes”. Our analysis covered only the shareholder meetings section of the omnibus amendments, as we felt investors needed a clearer picture of important rights they could lose if the proposals were adopted.

Since the consultation concluded in May, the Subcommittee on Company Law under the Ministry of Justice (MOJ) Legislative Council has held five meetings to discuss the findings and make preliminary decisions. The amendments have multiple objectives some of which are favourable for shareholders, such as removing duplication in business reports and the annual securities report (Yuho) and opening shareholder meetings to investors whose shares are held in nominee accounts (hence their names are not on the register). Yet on shareholder meetings more broadly there is a concerted effort to reduce the proposal rights of minorities and take action against “concealed activism”, among other things.

While many details remain to be finalised, the general direction of travel is in line with what was proposed in the consultation—with one major exception relating to privatisation-related “squeeze-outs”. This outcome is not surprising given the pressure exerted by leading business chambers, the ruling Liberal Democratic Party, and other arms of government for more flexibility and lower costs for companies.

We argued in our original article that much of the data provided by the MOJ did not provide clear support for several of the key amendments proposed. New numbers from the ministry and the Tokyo Stock Exchange (TSE), as well as a further round of research by Firefly, indicate that the need for legislative change on shareholder-proposal thresholds has been exaggerated, while claims that concealed activism is posing a systemic risk to the market are over-stated too. At the same time, policymakers are under-playing the risk of unintended consequences from these amendments and the likelihood of negative media coverage that could undermine Japan’s hard-won reputation for improved corporate governance.

First the good news: What has been dropped

The main proposal no longer on the table was a plan to make it much easier for acquirers to forcibly buy out minority shareholders following a tender offer aimed at taking a listed company fully private. Current rules allow acquirers to automatically trigger such a “squeeze-out” once they have secured 90% of the voting rights. In this sense, Japan is like many other developed markets. Where it is more permissive, however, is that it also allows acquirers with just two-thirds of the voting rights to force through a squeeze-out if they hold an EGM and pass a special resolution. Since these resolutions only require a two-thirds majority, the outcome is effectively a foregone conclusion.

The MOJ’s proposal would have lowered the automatic squeeze-out threshold from 90% to 66.6%, thus eliminating the need for an EGM, if certain procedural protections had been followed. The main litmus test would have been a “majority of minority” (MoM) condition: a majority of independent general shareholders tendering their shares in the offer, with the process carried out in a fair and non-coercive way.

This proposal failed because there was “almost no public support” for MoM, according to Subcommittee minutes. Some respondents didn’t like it in principle. Others worried that the validity of the process could be challenged because there were still no established rules on MoM in Japanese law. And the ministry noted it was hard to define MoM and no alternative suggestions had been made by respondents. Meanwhile, a few Subcommittee members recognised that shareholder meetings still served a useful purpose even if the outcome of resolutions was known in advance—a point we made in our original piece.

Current rules therefore stay in place: only acquirers with stakes of 90% can enforce the purchase of all remaining shares without an EGM.

While this is a good outcome for minority shareholders, it is unlikely the need to define MoM will go away. As the take-private of Toyota Industries earlier this year showed, the general language on MoM in the “Fair M&A Guidelines” from the Ministry of Economy, Trade and Industry (METI) has led to some illogical interpretations. (See “Toyota tests the system, falls short”.) Since business restructuring is a core objective of CG reform in Japan, as well as the new Japan Growth Strategy, many more privatisations are expected.

What has been retained

This leaves four out of the five proposals we analysed earlier still on the table and almost certain to be included in any bill sent to Parliament. They are:

1.         The abolition of the 300 voting-right threshold for making shareholder proposals.

2.         The right of boards to suspend the voting rights of beneficial owners and large shareholders who do not disclose their stakes.

3.         The pre-adoption of shareholder meeting resolutions effectively decided in advance through proxy voting.

4.         Allowing fully virtual AGMs.

What the data says

The remainder of this article surveys the current state of play on these issues and explains why we believe the evidence does not support a strong case for amending the law. Our conclusions are based in part on new and original research by Firefly.

1. 300 voting rights

In addition to holding shares for at least six months, there are two other criteria for making a shareholder proposal in Japan: ownership of 1% of the total voting rights in a company or at least 300 “voting rights” (equivalent to 30,000 shares as each voting right equates to an investment unit of 100 shares).

A long campaign has been waged against the 300 voting-right threshold in Japan and finally appears to be winning. A core complaint is that it is “too low” in absolute terms, especially for large-cap issuers. Business groups and certain politicians like to emphasise that it encourages nuisance proposals from so-called gadfly shareholders, with the LDP, Keidanren and METI all keen to see it go. Some argue that shareholders should be able to use modern technology to aggregate their holdings and achieve the 1% threshold without too much difficulty, rendering the 300 criterion redundant. And there is a view that recent efforts by the TSE to boost retail investment through lower cost investment units makes reaching 30,000 shares even easier. The exchange has been doing this by encouraging stock splits.

Not only do these arguments lack firm and broad-based evidence, the data that does exist largely supports the opposite conclusion.

300 voting rights not cheap

First, the claim that it is somehow easy or inexpensive to buy 30,000 shares in a typical Japanese listed company is not correct for a large number of issuers on the Prime Market, the main target for share proposals.

According to Subcommittee minutes from July 2026, and carrying out a simple extrapolation, the “median average” cost of buying such a stake in a Prime Market company is around US$426,000. Hardly a small amount. While half of the roughly 1,550 issuers in this segment would be lower, half would be higher. The “mean average” is around US$600,000, while the maximum is around US$17m. These figures are based on data provided by the MOJ on the cost of buying a 500 voting-right stake (see table below), one of the alternative proposals that the Subcommittee considered and rejected.

Source: TSE data quoted by MOJ in Subcommittee on Company Law minutes, 22 July 2026; conversions to USD done by Firefly on 6 October 2026 and rounded.

The numbers also do not support the argument that the TSE’s campaign to lower the average cost of investment units (100 shares) makes acquiring 300 voting rights inexpensive for the average retail investor. In March 2026 the exchange released data on 266 companies that had conducted stock splits over the previous 11 months, with 70% achieving investment units priced around the ¥100,000 range (from 14% before this exercise). This is the level that many respondents to a TSE survey, in particular younger investors, said was their preferred cost of an investment unit.

Extrapolating to 300 voting rights produces a total cost of ¥30,000,000 (US$190,000). Still not a small amount.

What are implications for investors of retaining only the 1% criterion? As the Subcommittee minutes also show, this would require a very substantial investment:

Source: TSE data quoted by MOJ in Subcommittee on Company Law minutes, 22 July 2026; conversions to USD done by Firefly on 6 October 2026 and rounded.

Large-cap reality check

Looking at actual corporate examples, Firefly calculated the cost of 300 voting rights and 1% stakes in several large issuers which have received shareholder proposals in the past three years. The figures indicate the high to extremely high levels of investment required to reach the two thresholds, with only two firms (Nippon Steel and NTT) being relatively lower on the 300 threshold (though still significant for an individual investor).

Source: Firefly calculations. Share prices as of 12, 13 and 19 October 2026. Yen/USD exchange rate 155:1. Figures rounded.

Most of the nine companies have undertaken stock splits in recent years, except for Mitsubishi UFJ Financial Group (MUFG), Mizuho FG and J-Power. The most aggressive was NTT, which undertook a 25-for-1 split in July 2023, having done smaller splits in 2015 and 2020. The next most aggressive have been Nippon Steel, which did a 5-for-1 split in October 2025, and Toyota Motor, which did a similar split in October 2021.

The irony of Nippon Steel’s split will not be lost on corporate governance mavens in Japan, since it is that company’s chairman, Eiji Hashimoto, who has been vocal about the need to abolish the 300 threshold. Has the company attracted frivolous proposals as a result? Not recently. Proposals put forward in 2025 and 2026 addressed the management of its listed subsidiaries, which the proposer believes have been underperforming, and executive compensation matters. Clearly the company finds these proposals a thorn in its side and made a point of stating in its AGM convocation notices that the two shareholders who made the proposals owned less than 0.01% of its shares.

The message from the NTT case is quite different: the company has actively sought to boost its shareholder base, possibly because its controlling shareholder with more than 33% of shares is the Ministry of Finance, and is happy to live with more shareholders. Indeed, of the 12 proposals it attracted from three different shareholders at its June 2026 AGM, one argued for a stock consolidation. The NTT board opposed this proposal along with the other 11, saying it “would disadvantage certain shareholders who would lose their voting rights and other advantages”. Meanwhile, it is using digitisation to lower communication costs with its shareholders.

IT to the rescue?

What of the argument that small shareholders should use information technology to aggregate voting rights and put forward proposals? They could if a convenient platform existed to service the retail market, but this does not exist in Japan. Unlike Korea, which has developed a digital application called “ACT” that coordinates retail shareholders, or Australia, where the climate advocacy group Market Forces operates a website registration system for investors wishing to lend their names to shareholder proposals (a minimum of 100 shareholders is required), Japan offers nothing similar today.

It was somewhat easier in the past. In the late 1990s and early 2000s, a retail investor group called “Shareholder Ombudsman” put forward proposals at Japan Housing Loan Company, Sumitomo Bank, Sony, Toyota, and Obayashi Corporation. Although it achieved some quite high levels of voting support, it has since disbanded.

Some collaborative efforts among retail shareholders are taking place today, mostly led by established anti-nuclear groups putting forward proposals urging Japan’s nine electric power companies to withdraw from nuclear energy. In 2026 these groups collated voting rights ranging from a low of 471 at Kyushu Electric (0.01% of shares) to 2,814 at Tohoku Electric (0.056% of shares). But these groups do not have scalable systems, rely on long-standing networks, and have to use the postal service to collate documentation: securing proxies from other shareholders requires paper documents that have been signed and sealed. If the 300 voting-right threshold is abolished and only the 1% criterion remains, this entire proposal effort will disappear.

As we reported in “Putting rights in reverse”, academic research shows that eliminating the 300 threshold would likely result in a 60% drop in the number of shareholder proposals in future. Not only would this do nothing to contain the activists that some politicians and business groups dislike so much, since such funds typically own stakes of 3-5% or more, but the proposal also ignores the reason it was introduced in the first place in 1981 (Showa 56). The fear at the time was that a single criterion, namely a percentage threshold, would ensure that large-cap companies could forever escape shareholder proposals. This is as true today as it was then.

While it is understood that the Subcommittee may consider an alternative to 300 voting rights, such as a fixed amount of investment, this would have to be set at an even higher cost level to satisfy politicians and other lobbyists.

1a. Non-binding resolutions

A related issue being considered by the Subcommittee is whether non-binding advisory resolutions should be allowed at shareholder meetings.

This has arisen in response to an LDP proposal that seeks to abolish the right of shareholders to propose Article amendments on matters of business execution (ie, things that companies do in the ordinary course of business). Because only boards can propose non-binding advisory resolutions in Japan, shareholders who wish to influence corporate behaviour must propose Article amendments, a cumbersome solution and one which typically receives limited support from institutional investors. The latter often have policies to vote against such amendments unless they are proposed by management.

Some of the highest profile proposals in this category in recent years have involved environmental and civil society groups trying to encourage issuers to increase climate-related disclosures and action. A trio of non-governmental organisations, including Kiko Network of Japan, Market Forces of Australia, and the Rainforest Action Network ran a campaign for several years starting in 2022 against Japanese megabanks, power utilities, and trading houses.

But a more direct threat against management are Article amendments targeting such things as business restructuring, the management of listed subsidiaries, and capital policies. In 2026, for example, NTT received three proposals seeking to add text to its Articles on the definition of corporate value, the calculation of cost of capital, and overall capital management. In fact, nine of the 12 proposals it received this year were advisory in nature and had to use the Article amendment method. A similar situation was seen at MUFG in 2025, when five of the seven proposals put forward involved Article changes.

Not surprisingly, corporates almost always oppose these amendment proposals as an interference in their ability to manage business operations, hence the LDP’s push to remove this right. But this idea is probably a non-starter, since it would be technically difficult to achieve from a law-drafting perspective and could limit management’s ability to amend Articles in future, as Subcommittee minutes note. It is also understood that the committee is uncomfortable about abolishing both the 300 voting-right threshold and proposals related to Articles.

A sensible solution would be to allow shareholders to put forward non-binding advisory resolutions. “I think we should accept them. Asking companies to change their Articles on concrete business matters is nonsense. We should have non-binding resolutions,” says Professor Manabu Matsunaka of Nagoya University Graduate School of Law and a secretariat member of the Subcommittee.

2. Suspending the voting rights of beneficial and large owners

Amendments being proposed on beneficial-ownership disclosure address two different issues. One is straightforward and aims to promote dialogue: permitting companies to ask investors if they have an interest in their shares. This would apply largely to shareholders the company cannot see, namely those holding stakes of less than 5% and which are not subject to statutory ownership disclosure. The other is more complex and may have unintended consequences: giving boards of directors the power to suspend the voting rights of shareholders which fail to disclose ownership positions.

The MOJ’s original plan had been to apply suspensions only for “large shareholders” which failed to disclose stakes of more than 5%. Following an intervention from the ruling LDP, however, suspension has been extended to failure by intermediaries such as banks or other financial institutions to disclose the holdings in their custody. This widens the net considerably and would include shareholders with stakes of less than 5%.

Chasing the wolf pack

Leading the charge is an LDP policy sub-committee called the Judicial System Research Council. It has a working group called the “Growth-Oriented Corporate Governance Task Force”, which held a series of hearings over June to July this year before publishing its conclusions in a report titled, “Recommendations for Governance Reform to Sustain Corporate Value Growth”.

The target of the Task Force is “concealed activism”, which refers to suspected activist funds hiding near-5% stakes in a firm and using “wolf pack” tactics to collude and vote at shareholder meetings, the aim being to influence or, in some cases, gain control of companies. 

The Task Force says it is not against all shareholder activism and admits much of it has had a positive effect on the governance and performance of listed firms in recent years. Yet it stoutly opposes “de facto collusive behaviour lacking necessary disclosure” and the abuse of rights that “deviates from the intent of the system”. It lists a few anonymised examples of wolf-pack behaviour and urges the enforcement arm of the Financial Services Agency (FSA), called the Securities and Exchange Surveillance Commission (SESC), to use its recently enhanced sanctioning powers to strictly enforce the law. It is also strongly critical of the regulator for its “lack of enforcement capacity” and said “it is undeniable that, compared to authorities in other countries, there are weaknesses in terms of staffing and expertise”.

A key question is whether concealed activism has created such systemic harm to the market that a legislative amendment is required? It appears not, since the Task Force report describes such collusive behaviour as being limited to a “very small number” of cases. This suggests a risk of legislative over-reaction.

Suspension plans

In addition to fines that could be imposed by the MOJ for low-level disclosure failures, the focus of suspension penalties is two-fold:

First, suspensions for large shareholders could last up to five years or until the conclusion of the first shareholder meeting after the shareholder makes a correct report (“cure”), whichever is shorter. This sanction is only intended to apply to disclosure violations that are intentional or grossly negligent. (Note: An earlier version of this proposed amendment was broader and would have allowed companies to suspend voting rights where a shareholder had violated the large shareholding report rule in “good faith and without gross negligence”. In this case, the suspension would have lasted only three weeks after the cure. This has since been deleted.)

Second, suspensions for intermediary disclosure failures would apply only to intentional or grossly negligent breaches as well and would be lifted three weeks after a correct report had been made. The rules also differ from the large-shareholder rules in three ways: the voting rights of intermediaries would be suspended only for shareholder meetings at which a shareholder (not the board) has made proposals; the suspension only applies to voting rights attached to shares for which intermediaries have not provided the necessary information; and the shares in question must account for at least 0.5% of total issued shares.

Weak points

These proposals remain problematic on several levels:

·       Voting rights are enshrined in company law and giving boards the power to suspend them is unsettling. If these amendments are approved, Japan would be going further than other major markets except the UK in giving boards such broad discretionary powers. Some markets require courts or regulators to order suspensions (Australia, Hong Kong, US) or make them automatic under company law (Germany, France). Only the UK allows companies to amend their articles to give boards a suspension right; otherwise, companies must apply to courts. Our preference is that boards should be required to apply to a court or regulator for a suspension order.

·       There is a presumption among proponents, including politicians, that Japanese companies need these powers because the SESC is weak at enforcing the large-shareholder reporting rules. While this may be true, shouldn’t the first step be to enhance the integrity and effectiveness of the regulator? Indeed, changes to the Financial Instruments and Exchange Law in May 2026 mean that the SESC is now able to issue much stronger monetary sanctions against violators of large-shareholding disclosure. It would make more sense for the government to support the SESC to the fullest extent possible, including giving it more resources, rather than handing out new private-enforcement powers that could be implemented haphazardly and backfire.

·       Suspending the voting rights of shareholders because of failures at the intermediary level is inherently unfair. The custodial banking system involves complex chains of relationships and may not always respond as efficiently as expected. It is instructive, for example, that the International Bankers Association of Japan supports the MOJ issuing fines but is opposed to suspensions.

·       The proposals envisage that an intermediary’s voting rights can be exercised three weeks after it makes a correct report. Yet this three-week wait seems excessive. If an intermediary has made a correction report, shouldn’t the underlying beneficial owner be allowed to exercise its voting rights immediately? Such a long delay could create problems for foreign shareholders if a company’s request for disclosure is made close to a shareholder meeting. This is because they need to vote according to deadlines set by global custodian banks and these are typically several days earlier than deadlines for domestic shareholders.

We recognise that the MOJ has also amended the draft wording on beneficial ownership disclosure to make enhancing constructive dialogue between companies and shareholders the “core” purpose of the amendment, while obtaining information about control of the company in “contingency situations” is now a “secondary purpose”. We also appreciate that the policy intention is that boards should use these powers judiciously, not maliciously. Nevertheless, the risk of bad actors misusing these powers remains.

3. Pre-adoption of meeting resolutions

The MOJ’s original proposal here still stands and is supported by the Subcommittee: to allow companies to declare shareholder meeting resolutions adopted before the meeting if they have been overwhelmingly approved in advance (ie, any voting in the meeting could not change the result). The somewhat cynical spirit underlying this idea is that “most decisions in the AGM are a kind of fiction” and any way to reduce costs must be good.

Two alternative safeguards are still under discussion. The first would require companies to amend their Articles to allow pre-adoption. The second would permit the chairman of the meeting simply to announce resolutions had been pre-adopted, but allow this to be neutralised if 10% of shareholders demanded a full vote prior to a meeting.

One other amendment still on the drawing board is whether to remove the right of shareholders to demand a vote be annulled if management failed in their “duty to explain” (ie, fully answer questions in a meeting on the relevant resolution).

As we argued in “Putting rights in reverse”, there are both principled and practical reasons for opposing the pre-adoption idea. While it is true that most resolutions are adopted in advance, this is a fact of listed-company life: dispersed ownership means that most shareholders who wish to vote must do so before meetings by proxy. In a modern capital market, the majority of absent shareholders will usually outnumber the minority who attend in person. Ignoring votes cast in a meeting not only penalises the latter, but undermines a hard-won fairness principle in modern corporate governance over the past two decades—voting by poll. This proposal also contradicts one of the goals of Japan’s Stewardship Code, which aims at bringing companies and shareholders closer together. The same spirit infuses METI’s new “Growth Investment Guidance”, which supports the Sanae Takaichi administration’s strategic Growth Strategy. All in all, we think this is a prime example of uncoordinated government policymaking.

On a more practical level, it is hard to see how the purported cost savings from pre-adopting a few resolutions will be significant. Companies will presumably still need to set up voting infrastructure for their AGMs. Meanwhile, the ministry has not provided compelling cost evidence to support its proposal.

The other risk is that management teams may not feel compelled to properly answer questions on pre-adopted resolutions—especially in less well-governed companies. Removing the annulment right will likely make matters worse if directors face no consequences for failing in their duty to explain (we have changed our mind on this point since our June article). As Professor Matsunaka of Nagoya University noted, “good firms will continue to explain, while bad firms do not need to explain”. He thinks this could become a particular problem in unlisted companies. He also believes the law should require shareholder approval of an amendment to the Articles if a company moves to use the pre-adoption mechanism.

4. Virtual AGMS

This has been one of the Subcommittee’s simpler issues to opine on. It supports the MOJ’s plan to allow fully virtual meetings on the proviso that companies amend their Articles before doing so. This may be less of a safeguard than hoped, however. As a result of Covid-era laws almost 480 issuers have already amended their Articles to allow them, indicating the ease of securing shareholder support. Part of the reason is that the threshold for such “special resolutions” in Japan is only 66.6%, not the 75% common in other developed markets. 

The most interesting point from the Subcommittee minutes is that only slightly more respondents to the MOJ consultation supported this proposal than were opposed—further evidence that investors and other market participants who care about governance still value physical AGMs.

One reason is the quality of the Q&A. As Professor Ken’ichi Matsuo of Kyoto University Graduate School of Law and a secretariat member of the Subcommittee noted, “Shareholder questions (in a hybrid meeting) are conveyed word for word. But in a virtual AGM the company may summarise the questions. This difference is significant.”

jallen@fireflyreads.com

Copyright: Ninepin Ltd, 2026