Tokyo’s cherry blossoms were declared at full bloom on 28 March, three days ahead of schedule. The weekend was clear and warm – a perfect hanami Sunday. Then the weather turned. A low-pressure system brought heavy rain and gale-force winds on the 31st, with parts of western Japan recording near-record March rainfall. April opened with more showers and a sharp cold snap. The blossoms held on, battered but intact, waiting for a break in the clouds.

Japanese bank stocks are having a similar spring.

In 1998, with the Japanese financial system on the edge of collapse, regulators drew a hard line between banks and industry. Cross-shareholdings were to be unwound. Equity exposure was to be curtailed. Investment subsidiaries at bank holding companies were confined to non-listed ventures and distressed firms. The message was clear: banks lend; they do not own.

On 31 March 2026, the Financial Services Agency began dismantling that wall.

An amendment to the relevant cabinet office ordinance will allow bank investment subsidiaries to take equity stakes in listed companies undergoing MBOs or carve-outs. A separate relaxation of large-exposure limits will permit megabanks to extend bridge financing beyond the existing ceiling when a borrower temporarily needs capital for a major acquisition. Both measures are expected to be folded into the Honebuto no Houshin (骨太の方針) – the government’s annual fiscal blueprint – in June. (Asahi Shimbun, 31 Mar 2026)

The Asahi Shimbun reported the changes as a technicality. They are not. What the FSA has set in motion is a structural reversal of the post-crisis consensus that kept Japanese banks at arm’s length from the corporate sector for nearly three decades.

The paradox nobody has spotted

Two reforms are running in parallel, and they pull in opposite directions.

On one side, the Tokyo Stock Exchange’s PBR campaign – now entering its fourth year – is pressuring listed companies to shed cross-shareholdings. The three largest insurers have committed to selling theirs entirely. MUFG plans to offload ¥700bn in strategic shares by March 2027; SMFG targets ¥600bn over five years; Mizuho, ¥300bn by March 2026. (S&P Global, Nov 2024, Feb 2025) Nomura estimates that the narrowly defined cross-shareholding ratio has fallen to around 8% and continues to decline by 0.3–0.5 percentage points a year. Share buybacks, running at ¥17–18.5tn annually, are absorbing much of the freed-up supply. (Nomura, “The Era of Disappearing Shares”)

On the other side, the FSA is opening a new channel for banks to acquire equity positions through investment subsidiaries. The legal wrapper is different: these are not mutual holdings between business partners but directed investments through regulated vehicles. The economic function, though, rhymes. Capital flows out the front door of the old cross-shareholding structure and walks back in through the side entrance of the new one.

Nobody in the English-language financial press has connected these two moves. They should.

A boom that needs financing

The timing is not accidental. Japan’s M&A market hit $385bn in 2025, a record. Public activist campaigns rose roughly 90% year on year, making Japan the second most targeted country in the world after America. Take-private transactions now account for the majority of PE-led deals. (J.P. Morgan, 2026 M&A Outlook)

The MBO wave alone tells the story. KKR’s ¥421bn buyout of Topcon. Bain Capital’s ¥200bn take-private of MCJ. Raksul’s ¥120bn MBO backed by Goldman Sachs. Mandom’s ¥125.6bn tender offer, completed after three price increases. The TSE tightened its MBO disclosure rules in July 2025 (TSE Code of Corporate Conduct revisions), but the pipeline kept growing because the underlying logic is sound: hundreds of listed companies still trade below book value, governance pressure is intensifying and private capital sees the gap between market price and intrinsic worth.

Every one of these deals requires bridge financing. The sums involved are large enough to bump against individual-borrower exposure limits at the megabanks. Under the old rules, a bank had to syndicate quickly or pass the mandate to a foreign competitor unconstrained by Japanese prudential caps. The FSA has now removed that bottleneck. Megabanks can hold larger positions, earn wider spreads on bridge loans and capture advisory fees that previously leaked offshore.

Economic security and the policy put

There is a second motive, less politely stated. Tokyo is frightened of foreign takeovers.

The scars are fresh. Nippon Steel’s $14.9bn bid for US Steel was blocked by President Biden in January 2025 on national security grounds (Foreign Policy, Jan 2025) – a decision that stung deeply in Tokyo, given that Japan is America’s closest military ally in the Pacific. Alimentation Couche-Tard’s pursuit of Seven & i Holdings dragged on for nearly a year before the Canadian firm withdrew (Reuters via Zawya, Jul 2025); during the fight, Japan’s finance ministry designated the convenience-store operator as “core” to national security, a classification usually reserved for nuclear and semiconductor firms. (AFP via Tribune, Jan 2025)

The lesson drawn in Kasumigaseki was blunt: if Washington can block a Japanese acquisition of an American company on security grounds, Tokyo had better be ready to do the same in reverse. The FSA’s amendment explicitly references the need to defend against hostile foreign bids for Japanese companies. The 17 “strategic sectors” nominated by the Takaichi administration – AI, semiconductors, aerospace, shipbuilding and the rest – are the intended beneficiaries.

Read between the lines and the picture is stark. Japanese megabanks are being repositioned as a domestic capital buffer: a standing reserve of financing capacity that can be mobilised to outbid foreign acquirers or fund defensive MBOs when a listed company faces an unwanted approach. That is an industrial-policy function, not a banking one. And it comes with an implicit policy put under the sector that no analyst’s model currently captures.

The rate arithmetic

All of this lands at a moment when the megabanks’ core earnings power is already accelerating.

The Bank of Japan has raised rates three times since abandoning negative-rate policy in March 2024. The policy rate sits at 0.75%. Swap markets price a 68% probability of a fourth hike on 28 April, to 1.00%, with a second move by autumn priced as near-certain. Even Nakamura Toyoaki, the board’s most consistent dove, said this week that he would support a hike if the economic outlook held.

The earnings sensitivity is enormous. MUFG has disclosed that each 25-basis-point increase contributes roughly ¥100bn to annual net interest income. SMFG puts the figure at a similar level. Mizuho estimates the cumulative effect of the three hikes so far at ¥225bn for the coming fiscal year. (S&P Global, Feb 2025) Net interest margins at the megabanks remain between 0.44% (Mizuho) and 0.87% (SMFG) – a fraction of the 3.0–3.5% that American banks earn. The gap is closing, but from such a low base that every incremental hike flows almost directly to the bottom line.

Meanwhile, the Fed is frozen. SOFR futures price no rate movement through at least September 2026. The ECB is similarly stuck. The carry spread between US Treasuries and JGBs has fallen below 200 basis points for the first time in this cycle. When that spread narrows because the BOJ is tightening rather than because the Fed is easing, the unwind is harsher: there is no supportive macro environment on the other side of the trade to cushion the adjustment.

Three tailwinds, one price

The market is pricing Japanese megabanks as though the GDP average represents their customer base. It does not.

MUFG’s corporate loan book is concentrated among large exporters printing record profits. Its fee income comes from M&A advisory, asset management and cross-border finance. Its mortgage exposure is in Tokyo and Osaka, where property values are rising. The struggling small firms in rural prefectures borrow from regional banks, not from megabanks. The bifurcation of the Japanese economy, between a prosperous, globally competitive corporate sector and a squeezed, ageing consumer base, flatters the megabanks’ actual risk profile while depressing their valuation.

Three independent forces are now converging on the sector. Rate normalisation is expanding net interest margins from the lowest base in the developed world. FSA deregulation is opening new revenue streams in M&A financing and principal investment. And the implicit policy role as a domestic capital buffer carries a political premium that the market has not begun to discount.

Oliver Wyman noted earlier this year that even at their recent highs, the megabanks traded at a material discount to North American peers on price-to-tangible-book. Global investors, the firm argued, were waiting for evidence of structural change before re-rating the shares. (Oliver Wyman, 2024) The FSA’s move is structural change. Whether it is recognised as such is a different question.

The entry window

Foreign investors are currently heading for the exit. MOF data for the week of 22–28 March showed ¥4.4tn in net equity selling by non-residents – the largest weekly outflow in the dataset – with a further ¥6.8tn sold across bonds and short-term instruments. The pace has accelerated for three consecutive weeks. Carry-trade positions are under pressure as the US-Japan yield spread compresses below the 200-basis-point threshold where hedging costs erode returns.

Short-term pain is real. An oil shock driven by the Hormuz closure is feeding into Japanese inflation, giving the BOJ additional cover for an April hike. A rate increase delivered into a supply-side shock is the worst possible combination for equities in the near term. Volatility will spike.

But the dislocation between flow-driven selling and structural earnings improvement is exactly the kind of gap that rewards patience. Foreign money is panicking out on the headline. Tokyo is quietly rewriting the rulebook to make banks more profitable on the other side.

The last time Japan rebuilt the relationship between its banks and its industrial base, it produced the highest-growth economy in the developed world for three decades. The comparison is imperfect; this is a narrower, more surgical intervention than the old MITI-era administrative guidance. But the direction of travel is the same: the state is pointing bank capital at strategic sectors and removing the regulatory barriers that stood in the way.

Whether the market notices before the Honebuto is published in June is an open question. The megabanks’ FY2026 earnings releases in May will offer the first hard data on how the new rules affect their deal pipelines. Until then, the thesis is simple enough. Three tailwinds. One price. And a window that the carry-trade unwind is holding open.

The forecast for Tokyo says the rain returns on the 4th. But cherry blossoms have weathered worse and still bloomed.