NextFin News - Shares of Japan's trading houses jumped on Thursday after Greg Abel, chief executive of Berkshire Hathaway, said the conglomerate intends to hold its stakes in the five sogo shosha "for many decades." The remark, made during Abel's first year as Berkshire's CEO, answered the question Tokyo investors have been asking since Warren Buffett handed over the reins: would the successor keep Buffett's signature Asian bet? The market's answer was immediate — Mitsubishi Corp. climbed as much as 4.5% to its highest level since May, while Sumitomo Corp., Mitsui & Co., Itochu Corp. and Marubeni Corp. each added more than 2.5%, making the trading houses the biggest gainers on the broader Topix benchmark.
The rally is more than a relief trade. It is a referendum on whether Buffett's Japan thesis — yen-funded stakes in undervalued, cash-generative conglomerates — survives the man who built it. Berkshire now owns roughly 10% of each of the five houses, a position that has grown from about $6.3 billion when Buffett first disclosed the stakes on his 90th birthday in August 2020 to more than $30 billion by October 2025. The commitment has also widened beyond the trading houses: in March 2026, Berkshire agreed to buy a 2.49% stake in insurer Tokio Marine Holdings for about ¥287.4 billion ($1.8 billion), acquiring treasury shares at ¥5,962 apiece. That stake was financed largely with yen-denominated bonds, a carry trade that printed money while Japanese rates sat near zero. With the Bank of Japan now pushing the 10-year government bond yield above 3% for the first time in three decades, the second-order question beneath Thursday's pop is whether the funding advantage that made the trade work is entering its final act.
The Event: A Succession Signal, Not a New Purchase
Abel did not announce fresh buying. What moved the market was the time horizon he attached to the existing one. Berkshire's commitment to the trading houses predates Abel's tenure — Buffett began accumulating the positions in the summer of 2019 and revealed them publicly on August 30, 2020, when the five stakes totaled roughly $6.3 billion, or about 5% of each company. At the time, Buffett promised the companies Berkshire would not lift its holdings above 10% without their permission.
That ceiling has since been tested. In his February 2025 shareholder letter, Buffett wrote, "As we approached this limit the five companies agreed to moderately relax the ceiling." Berkshire then crossed the 10% threshold across the group, with Mitsubishi rising to 10.23% from 9.74% and Mitsui confirmed above 10% after earlier sitting at 9.82%; Sumitomo and Marubeni also moved past the line in filings dated May 7, 2026. In the same letter, Buffett handed Abel's successors a mandate as much as Abel himself: "I expect that Greg and his eventual successors will be holding this Japanese position for many decades."
"In the next 50 years, we won't give a thought to selling those," Buffett said of the Japanese holdings at the 2025 annual meeting in Omaha. "We will not be selling any stock. That will not happen in decades, if then."
Thursday's gain therefore prices a continuity premium, not new capital. Investors are paying for the removal of succession risk — the fear that a post-Buffett Berkshire might treat Japan as a Buffett-era relic to be trimmed. Abel's phrasing closes that door. It also fits a broader pattern: the March 2026 Tokio Marine purchase, structured as a strategic partnership with Berkshire's National Indemnity reinsurance unit, signals that Japan is a living allocation under the new chief executive, not a museum piece. And in September, the chief executives of the five trading houses are scheduled to meet Abel in Japan — their first such gathering with the new chief executive.
The succession backdrop gives the remark its weight. Abel took over as CEO on January 1, 2026, inheriting a conglomerate sitting on a record cash pile that exceeded $397 billion at the May annual meeting. With U.S. equity valuations stretched — the market-cap-to-GDP ratio, known as the Buffett indicator, touched an all-time high in mid-August — the pressure on a new chief executive to deploy capital is intense. In that context, reaffirming a position that already returned several hundred percent is not a neutral act. It tells investors that the Japan book is core to the Abel era, not a legacy holding waiting to be trimmed when a better idea appears.
Why the Trade Worked: Three Engines, Not One
To judge whether the commitment holds, it helps to separate what drove the returns. The trading houses have not simply ridden a rising Japanese market. Since September 2021, Mitsubishi has returned about 329%, Sumitomo roughly 355%, Mitsui around 314%, Itochu about 223% and Marubeni approximately 468%. Over the same period the Topix ETF returned about 104%. Even measured over the past twelve months, the five names have gained between 27% and 70%, against roughly 34% for the benchmark. That is not a single-factor trade.
Three distinct engines produced that outperformance. First, the currency engine: Berkshire borrowed in yen at near-zero rates to buy globally diversified assets. Buffett noted in his 2023 shareholder letter that Berkshire had issued about ¥1.3 trillion of yen bonds — more yen-denominated debt than any other American company — and that the weak yen had already produced billions in accounting gains. The mechanics are elegant: borrow in a currency that is cheap and, if anything, likely to weaken further against the assets' earnings base, then let the liability shrink in real terms. Second, the commodity engine: Mitsui and Mitsubishi, Berkshire's largest Japanese positions, carry heavy exposure to energy, metals and coal, which re-rated sharply after 2021. Third, and most durable, the governance engine: Japan's stock exchange push for companies to address price-to-book ratios below one forced the trading houses to lift capital efficiency, buy back shares and sharpen disclosure.
The first two engines are cyclical. The third is not. That distinction matters because Abel's "many decades" pledge implicitly bets that engine three can carry the trade once engines one and two stop helping.
The Cyclical Leg: The Cheap-Yen Arbitrage Is Expiring
Here is the uncomfortable part of the story that Thursday's rally glossed over. The funding structure that made the sogo shosha trade a near-perfect arbitrage is under pressure. Japan's 10-year government bond yield touched 3.01% this week, a 30-year high, as the Bank of Japan continues its slow exit from ultra-loose policy. Abel has said he still views borrowing in Japan as an appropriate strategy, but "appropriate" is not the same as "free." Every basis point of yen yield that rises narrows the carry that turned Berkshire's Japanese equity stakes into a leveraged, currency-hedged position.
This is a cyclical headwind with a clear mean-reversion logic — except the mean is moving. For three decades after Japan's asset bubble burst, the reference rate for this trade was zero. If the BOJ's normalization is a genuine regime shift rather than a false start, the trade's cost of capital resets permanently higher. That does not break the thesis; the houses earn returns on equity well above borrowing costs. But it changes the character of the investment from a structurally advantaged carry position to a plain, if still cheap, equity bet.
The currency side cuts both ways. A stronger yen would reduce the dollar value of the houses' overseas earnings when translated back — a reported headwind for Mitsubishi and Mitsui — while simultaneously shrinking the mark-to-market gain on Berkshire's yen liabilities. Buffett's structure was deliberately approximately currency-neutral on the funding side; the equity side is not. Investors cheering Abel's commitment should watch the yen as closely as the share price.
There is also the question of what else that yen funding could buy. With Japanese yields no longer at zero, the opportunity cost of keeping capital in the trading houses has risen. The very instrument that made the trade cheap — yen debt — now carries a real coupon. For Abel, the pledge to hold "for many decades" is also a pledge not to chase yield elsewhere when the funding leg stops subsidizing the position. That is a test of discipline as much as conviction.
The Structural Leg: Governance Reform Is the Real Story
If the yen arbitrage is the trade's past, corporate governance is its future. The sogo shosha spent the lost decades as textbook conglomerate discounts: sprawling, opaque, undercapitalized and indifferent to shareholders. For years, investors accepted that the houses' vast networks — spanning energy, food, machinery, retail and finance — came with an inevitability discount. The Tokyo Stock Exchange's 2022-2023 push on price-to-book ratios forced a reckoning. Management could no longer hide behind diversification; if the market valued the whole below the sum of the parts, the exchange wanted an explanation and a plan.
Trading houses responded with record buybacks, higher dividends and a wave of portfolio pruning — selling non-core assets, exiting low-return businesses and reallocating toward higher-margin sectors such as data centers, aircraft leasing and India's industrial buildout. Itochu, Berkshire's most consumer-and-services-oriented house, has been the most aggressive in treating shareholder value as a management objective rather than a slogan. Mitsui and Mitsubishi trade at lower valuations but carry more runway for portfolio improvement, with management teams that have shown they will act. Marubeni and Sumitomo, historically the laggards, have the furthest to go — which is precisely why Berkshire crossed the 10% line in those two first. A 10% stake is not passive; it is the threshold at which Berkshire can exercise real influence behind closed doors.
This is the structural core of Abel's bet. Even if commodity prices soften and the yen stabilizes, a trading house that compounds book value at double-digit rates while buying back stock below intrinsic value keeps creating shareholder returns. That mechanism does not depend on zero rates. It depends on management incentives, and Japan's corporate-governance code has rewired those incentives in a way that is unlikely to reverse. Once a board has internalized that a sub-one price-to-book ratio is a failure requiring a public plan, the genie does not go back in the bottle.
The Counter-Thesis: A Value Trap in Conglomerate Clothing
The strongest argument against the bullish read is not that Abel will sell — he has said he will not. It is that the market has already priced the governance re-rating, leaving little margin of safety for new money. The five houses now trade at a combined market value many times their 2020 levels, and the Topix itself sits at record highs. If the re-rating is complete, "holding for decades" becomes a description of a fully valued position, not an endorsement of future returns.
There is also the concentration risk that a diversified house is supposed to eliminate. Mitsui and Mitsubishi remain heavily exposed to commodities and to China's demand for steel, coal and copper. A hard landing in China or a sustained drop in resource prices would hit earnings even as the governance story stays intact. And the houses' famed "information networks" — the competitive advantage Buffett has cited — face their own disruption as artificial intelligence compresses the value of intermediation.
The bull case answers this by pointing to valuation discipline: even after the rally, the houses trade at single-digit to low-double-digit earnings multiples with free-cash-flow yields that exceed Japanese bond yields by a wide margin. But that answer only holds if earnings do not normalize lower. The falsifying signal is specific: if the five houses' aggregate return on equity falls back below their cost of capital for two consecutive fiscal years while buybacks continue, the governance thesis is broken — capital is being returned not because management has turned owner-minded, but because there are no returns to be had. That is the line between a structural re-rating and a cyclical sugar high.
There is a second, quieter counter-thesis worth naming: the succession pledge itself may be over-interpreted. Buffett's "many decades" language came from Buffett, and the market may be crediting Abel with a constraint he did not impose on himself. If Abel's own words on Thursday were read as a commitment to never reduce the position under any circumstances, investors could be paying for an optionality they were never actually sold. Berkshire's culture prizes flexibility above doctrine; a pledge to hold forever is only credible while the businesses keep earning it.
What to Watch: Three Horizons
Short term (sentiment and flows): The September meeting between Abel and the five chief executives is the first tangible test. Any hint of coordinated capital allocation — joint ventures, cross-shareholding unwinds, or guidance on buybacks — would extend the rally. Conversely, a stronger yen or a pullback in commodity prices would pressure the group regardless of Berkshire's stance.
Medium term (fundamentals): Earnings for the fiscal year ending March 2027 will show whether the houses can sustain double-digit returns on equity as commodity prices normalize. Watch free cash flow, buyback pace and dividend policy more closely than headline profit, which remains commodity-sensitive.
Long term (structural): The Bank of Japan's policy path is the wild card. If the 10-year yield settles sustainably above 3%, the cheap-funding leg of Berkshire's trade is over, and the investment thesis rests entirely on governance-driven compounding. That is a thinner, though still credible, margin of safety.
Base case: the houses grind higher as governance reforms compound, with volatility tied to commodities and the yen. Upside case: Abel uses the September meeting to signal deeper Berkshire involvement, and the houses accelerate buybacks as China stabilizes. Downside case: a China hard landing or a sharp yen appreciation compresses earnings just as funding costs rise, turning the re-rating into a multiple contraction.
Thursday's rally was a vote of confidence in Berkshire's staying power. The harder vote comes next: whether Japan's trading houses can keep compounding value when the cheap money that carried them there is gone. Abel has promised to hold for decades. The next few quarters will show whether the businesses can earn that patience.
Explore more exclusive insights at nextfin.ai.
