Japan's governance reform has not gone soft. It has moved its pressure point. The restrictions on shareholder tools that read like a retreat are paired with a new obligation that cuts deeper: boards must now explain, publicly and in their own words, why their capital earns more than it costs. The government's own arithmetic suggests most of them cannot. METI estimates that roughly 65% of Japanese corporate assets sit in segments that destroy value, against 39% in the United States and 42% in Europe. Once a board is required to state a path to growth and defend its cash, that figure stops being an academic finding and becomes a disclosure liability spread across most of the market.
This changes what kind of trade Japan is. The first phase of reform paid almost any holder of an overcapitalized stock: buybacks, higher dividends, cross-shareholding sales. Owning the basket was enough. The second phase pays only where a company's explanation of its capital holds up against its own segment economics. The market still prices much of Japanese value as one cohort. The disclosure regime now being built is designed to split it.
The Compulsion Tools Are Narrowing
The headline case for a retreat is real and should be stated precisely. Cash-settled equity swaps already count toward large-shareholding disclosure. The Ministry of Justice is weighing changes to the 300-voting-unit route for shareholder proposals, limits on proposals to amend the articles of incorporation, and a mechanism to suspend voting rights when a large shareholding report is missing or materially wrong. The government and ruling party have proposed lifting the threshold to call an extraordinary meeting from 3% to 5%. In July the ruling party warned against undisclosed "wolf-pack" coordination.
The proposal limits matter more in Japan than an outside reader would assume. Japanese shareholders cannot put nonbinding proposals to a vote, so an amendment to the articles is the main formal channel for one holder to rally others. Narrowing it removes a real lever. The vote-suspension idea carries a quieter risk that GMO's Usonian team flags directly: a company could try to disqualify a dissenter's votes over a reporting error that originated somewhere in the custodial chain, outside the shareholder's control.
None of this is law. No bill has been drafted, one is not expected before the 2027 ordinary Diet session, and the final form may differ materially. Investors pricing a hostile-activism discount into Japanese equities today are pricing a plan.
What is being removed, in substance, is procedural leverage: the ability of a small stake to compel a board it has not persuaded.
Disclosure Is the Replacement Weapon
The replacement arrives through three channels, each with a firmer date than the restrictions.
The revised Corporate Governance Code, finalized in July, asks boards to set out "a path for growth" covering capex, R&D, human capital, intangibles and portfolio review, and to explain capital policy and reasons for holding cash. METI's growth-investment guidance puts the diagnosis in numbers: growth investment runs at 17.7% of sales in Japan, roughly half the US and European level. The emphasis is shifting from return on equity toward economic profit, meaning whether each segment earns more than the cost of capital.
The Tokyo Stock Exchange has adopted minority-protection rules that apply from fiscal years ending December 2026. Companies with a controlling or major shareholder must disclose how minority holders voted on each director and, where a majority of them opposes, respond publicly. That is a mechanical, dated catalyst. For a parent-controlled listed subsidiary, the minority vote has always been arithmetically irrelevant to outcomes. It will now be visible, and a majority-of-minority rejection will demand a written answer.
The Financial Services Agency also rewrote the joint-holder tests so that substantive dialogue, including among investors, does not by itself make them a group. The contrast with Washington is sharp: US regulators have recently warned that coordinated engagement could cost asset managers their passive status. Japan is narrowing the tools to compel while widening the space to talk.
Put together, the mechanism is straightforward. A board publishes its capital story. Outside directors, whom the TSE describes as representing ordinary shareholders, are expected to engage directly when concerns arise and to bring that feedback to the board. Informed owners then test the story against segment returns and vote accordingly. Departing from the Code requires an explanation, and an explanation that contradicts the segment data becomes a public record of a gap.
Where the Explanation and the Filing Diverge
The equity-fundamental blind spot sits here. Consensus will read the new disclosures the way it reads guidance: as narrative. The value lies in the gap between the narrative and the segment numbers.
METI's framework is explicit about how economic profit is measured. Segment EP equals segment NOPAT minus company-wide WACC multiplied by segment invested capital, with company-wide profit and capital allocated across segments by the composition of segment profits and assets, or by revenue when that data is missing. Every input after the first is partly a management choice: how assets are assigned, how overhead is spread, which WACC is used. The source authors acknowledge the same point bluntly. Segment returns depend on assumptions that management controls, and "growth investment" can describe disciplined expansion as easily as an empire-building acquisition.
The policy has therefore handed cash-rich boards two scripts. One is to exit value-destroying segments and redeploy into the core. The other is to spend accumulated cash on growth under official sanction. METI's own data implies the second script carries more risk than it advertises. If about two-thirds of assets already earn below their cost of capital, a cohort of boards that has historically allocated poorly is now being encouraged to allocate more. Poorly priced acquisitions could become more common, and the Code that is meant to discipline them was streamlined from roughly 83 items to about 30, with many requirements moved outside comply-or-explain. Investor groups criticized the cut; Keidanren had lobbied for a minimal Code.
This is why the dispersion call follows. Two companies with identical cash piles and identical buyback histories can now diverge on a single variable: whether the capital they redeploy clears the hurdle in segments that already clear it. The filing, not the press release, tells which is which. GMO's own engagement mix tracks the same shift. Strategy, portfolio and control questions rose from roughly 10% of its engagement actions in 2021 to 32% in 2026 to date, across 619 recorded actions since January 2021. The work is moving from "return the cash" to "show the segment math."
The Countercase: Boards Have Gained Room Too
The strongest objection is that disclosure without compulsion is cheap talk, and that the balance of power has tilted toward management.
There is evidence for it. METI's July Q&A on its takeover guidelines confirms that a high price alone does not make a bid desirable and that a board may favor the proposal it considers most conducive to "corporate value." Takeover defenses have become more common. If the extraordinary-meeting threshold rises and amendment proposals are curtailed, a determined board facing only written criticism can simply absorb it. Explanation imposes a reputational cost. It does not force a sale, a spin-off or a capital return.
The counter-evidence is narrower but pointed. The same METI Q&A limits non-price factors to those reasonably expected to raise future cash flows and says they should not be used to protect management. In September the Tokyo District Court blocked Toho Holdings' poison pill as aimed at entrenchment despite earlier shareholder approval, a ruling now subject to appeal. Courts are reading "corporate value" as a cash-flow test, not a shield.
The second objection is macro, and here the rates channel does genuine causal work. Further Bank of Japan increases would lift the cost of capital and with it the hurdle every new investment must clear, precisely as companies begin deploying balance sheets. A higher WACC also mechanically enlarges the share of segments that destroy value under METI's own formula. That cuts both ways: it raises the risk of failed growth spending, and it raises the visible cost of keeping marginal segments. Weak real wages in August, foreign net selling late in the quarter and a deterioration in relations with China at quarter-end add noise that could swamp stock-specific signals for a period.
The thesis breaks if the disclosures prove uniform. Should boards converge on boilerplate growth paths that no investor can test against segment data, the regime becomes paperwork, and the basket trade, or no trade, is all that remains.
The Dates That Settle It
Three observable conditions will confirm or invalidate the read.
The first is the minority-vote disclosure cycle for controlled companies, triggered by fiscal years ending December 2026 and therefore visible from the spring 2027 meeting season. Majority-of-minority rejections, and the quality of the written responses, will show whether the TSE rule has teeth or merely publishes dissent.
The second is the shape of the Companies Act bill if it reaches the 2027 ordinary Diet. A vote-suspension mechanism without a safe harbor for custodial errors would shift leverage to boards in a way disclosure cannot offset. A narrow version would leave the dialogue-first framework intact.
The third, and the most investable, is whether capital deployment announcements begin to separate along segment lines. Companies that pair growth spending with exits from below-hurdle businesses are executing the policy's actual intent. Companies that announce acquisitions while carrying large value-destroying segments are spending the reform's credibility. If valuations within Japanese value start to discriminate between those two groups, the disclosure regime is working, and the edge belongs to investors who read segment economics rather than capital policy slogans. If they do not, procedural leverage was the reform, and its repricing is a loss.
