The Sogo Shosha and the Buffett Japan Trade, Explained
Kabu Research · Japan equity research
Short answer: a 総合商社 (sogo shosha) is a Japanese general trading house — a diversified conglomerate spanning commodities, energy, metals, machinery, chemicals, food, and finance, and increasingly investing as a principal. The five largest are Mitsubishi Corporation, Itochu, Mitsui & Co., Marubeni, and Sumitomo Corporation. They became the global shorthand for Japan after Berkshire Hathaway made roughly $24 billion on its five trading-house positions. The trade matters beyond those five names because it is the most visible version of a broader Japan re-rating story.
Below: what the sogo shosha actually are, why Buffett bought them, and how the trade connects to the rest of Japan's re-rating — including the uncovered names no English analyst writes up.
What is a sogo shosha?
The sogo shosha are unlike anything in U.S. markets. They began as trading intermediaries — buying, selling, and financing the flow of goods in and out of a resource-poor island economy — and evolved into sprawling holding companies that own stakes across the entire value chain. A single trading house may hold interests in liquefied natural gas, copper mines, convenience-store chains, food distribution, real estate, infrastructure, and financial services at once.
The "big five" are Mitsubishi Corporation, Itochu, Mitsui & Co., Marubeni, and Sumitomo Corporation. Their defining traits, from an investor's standpoint: durable and diversified earnings streams, conservative balance sheets, and — increasingly, under reform pressure — shareholder-friendly capital allocation through buybacks and rising dividends.
Why Buffett bought the trading houses
Berkshire Hathaway began building its positions in the five trading houses in 2019 and disclosed them in 2020. Since then the positions have returned roughly $24 billion in profit, and Buffett has publicly signaled Berkshire intends to hold for the very long term — framed as the next 50 years.
The publicly stated logic rhymes with how Berkshire is itself built:
- Diversified, durable earnings. Each trading house is a portfolio of cash-generative businesses, not a single-product bet — much like Berkshire's own structure.
- Conservative balance sheets and improving capital discipline. The houses have leaned into buybacks and dividend growth as Japan's reform wave pushed capital-allocation discipline up the agenda.
- Reasonable valuations relative to earnings power. When Berkshire built the positions, the trading houses traded at modest multiples for the cash flow they produced.
The endorsement effect is hard to overstate. A foreign investor who had never looked at Japan suddenly had the most credited long-term investor in the world pointing directly at it — and at a specific, legible corner of it.
How the sogo shosha trade connects to Japan's wider re-rating
The trading houses are the entry point, not the whole story. They are the visible, liquid, English-covered expression of a theme that runs far deeper into the market: Japanese companies with strong balance sheets and improving capital discipline re-rating as governance reform takes hold. The same drivers that lifted the sogo shosha — idle-cash discipline, buybacks, dividend growth, cross-shareholding unwinds, and below-book valuations correcting — operate all the way down the market-cap scale.
That is the key insight for anyone who has noticed the Buffett trade and wants to go further. The trading houses are already covered, already liquid, and already re-rated in part. The larger pool of the same characteristics — strong balance sheets, idle cash, below-book valuations, a reform mandate — sits in the small- and mid-cap names that no English analyst covers. (See Japanese stocks below book value for the scale of that pool, and the orphan cohort for why it stays overlooked.)
The "down-cap" version of the trade
One way to think about the next leg is replication at smaller scale. The features that made the trading houses attractive — diversified cash generation, fortress balance sheets, capital returns accelerating under reform — are not unique to the big five. They recur in less-followed names trading at lower multiples, where the re-rating has not yet happened because the market has not yet looked. Reading those names, however, runs straight into the language wall: their disclosures, like most of the Japanese market's, are filed in Japanese.
This is where access becomes the binding constraint. The sogo shosha can be researched in English. The down-cap version of the same trade generally cannot — not without translating and structuring the underlying Yuho and exchange filings. That is the work Kabu Research does: starting from the names everyone knows and extending the same analytical lens into the cohort almost nobody can read.
How Kabu covers the trade
Kabu Research applies a consistent, source-cited lens across the market — from the liquid, Buffett-benchmark trading houses to the uncovered small-caps that share their balance-sheet and capital-return characteristics. Every claim is tied to a translated source document with claim-level citations, so a reader can verify the work against the original filing rather than take a headline on faith. See the methodology for the full process, or open the demo screener to see the cohort in English.