A year's profit, spent on an orebody that produces nothing until 2030

Group capital expenditure was ¥1,110 B, against ¥359 B the year before, a 3.1× increase. It was also 133% of the group’s net income for the year, which is to say the company spent more on capital account than it earned. And 73% of it went to Mineral & Metal Resources, whose own spending rose 9.0× from ¥90 B to ¥813 B.

A company can do that for two reasons. It can be defending a position that is eroding, in which case the capital is maintenance dressed as growth. Or it can be buying something it intends to own for a very long time. Mitsui says which one this is, and it is worth taking at its word for long enough to test.

The figures on it are worth separating from the language around them. Mitsui’s equity share of iron ore production is 63.8 million tonnes now and 65.3 million at the end of the plan period — which is to say Rhodes Ridge contributes nothing inside the plan this document sets out. The growth the slide is drawn to illustrate is about 1.4 times FY2026 by FY2036, two plans after this one. Copper is the nearer story: 91 thousand tonnes rising to 145 by the end of the plan and 250 by 2036, through Collahuasi and through higher volumes at Anglo American Sur.

Put beside the profit the two commodities earn together today — ¥240 B — the ambition is ¥280 B by 2030. So ¥813 B of capital expenditure is not buying earnings inside this plan. It is buying a position, and the return on it cannot be assessed from anything in these accounts — which is exactly why return on equity reads the way it does this year. The capital is on the balance sheet and the profit is four years out. An investor who marks the company on this year’s 10.2% is marking the cost without the asset.

The same mine, on two of these pages

Anglo American Sur is a Chilean copper business — the Los Bronces and El Soldado mines and the Chagres smelter — and it appears twice in this section. Mitsui counts it among the assets taking its copper production to 250 thousand tonnes by 2036: Integrated operations with the adjacent Los Bronces mine, to start in 2030, alongside expansion at Collahuasi. Mitsubishi holds 20.4% of the same company, carried at ¥217 B against a cost of $4.51 billion — about 30% of what it paid.

Two of Japan’s five trading houses own pieces of the same orebody and describe it in opposite terms: one as a growth asset with a date on it, the other as a holding written down to a fraction of its cost. Both are right. They are answering different questions — what it will produce, and what it has returned.

Mineral & Metal Resources earns 30.2% of the profit on 19.8% of the assets, a spread of 10.3 points in its favour — and it is also the segment that has just absorbed 73% of the year’s capital. Wellness Ecosystem does the reverse: 6.2% of the profit on 14.2% of the balance sheet.

Wellness Ecosystem deserves its own line. It is the largest revenue line in the group at ¥3.4 tn — hospitals, healthcare and protein — and it earns 1.7% on ¥3.1 tn of assets, the lowest return of the seven. Its profit has fallen in each of the three years, from ¥94.1 B to ¥52.0 B, while its assets rose ¥189 B. Revenue is not the measure for a business like this and Mitsui does not claim it is, but a segment of that size earning under two per cent is the largest single drag on the group return.

The record runs from 8.0% in 2021 to 18.9% in 2023, with no loss-making year. 3 of the ten cleared the 12% the current plan asks for. The share count has fallen from 3,528 million to 2,834 million, a reduction of 19.7% — more than Itochu’s 10.8% and less than Mitsubishi’s 23.0%. The shares traded below book value in each of the first seven years of the record, as low as 0.7×, and stand at 1.9× — the highest multiple in the ten years, struck in the year return on equity came in at its lowest since 2021.

Read together, the four say something consistent. The two cash-return targets were beaten comfortably — the payout against core operating cash flow came in over 53% against 37% planned — and the two absolute earnings targets were missed. A company that returns more than it promised while earning less than it promised is one whose shareholders did better than its forecasts, which is an unusual way round.

The three-year average return on equity of 12.5% cleared the 12% target, and every figure in the column fell. That is a target met by its first year and missed by its last, which is the same shape the dividend column does not have: the dividend rose in each of the three.

MTMP2029

The new plan runs three years to 31 March 2029 and targets ¥1.2 tn of profit, 12% on equity, and core operating cash flow of ¥1,100 B a year. The three-year cash plan is below.

¥1.9 tn of the ¥4.7 tn coming in — 40% of it — is not allocated to anything. Mitsui calls this Management Allocation and says it will go to further growth investment or further shareholder returns depending on what turns up. It is a residual rather than a plan, and it is the honest way to present three years of cash that has not been committed yet. The alternative, which is to assign every yen to a named bucket and revise later, is how most of these documents read.

The cash flow target itself is a modest step: ¥3.4 tn over three years against ¥3.0 tn actually generated over the last three, a 1.1× increase. The dividend is a floor rather than a ratio — ¥140 per share as the minimum for the period. And ¥1.7 tn of the growth spending is tagged to three named initiatives: Industrial Business Solutions at ¥780 B, Global Energy Transformation at ¥690 B and Wellness Ecosystem Creation at ¥180 B.

Per share, which is the whole point

One slide in the plan does something none of the other four houses’ documents do: it tracks cash flow per share against the share count over sixteen years, and sets the target on the per-share figure.

The bars fall and the line rises. That is the entire argument for a buyback drawn once, and it is the measure this site is named after: what matters to an owner is not what the company generates but what it generates for each share with a claim on it. Mitsui targets 8% a year compounded on that figure from 2013 to 2029, and has retired 19.7% of its shares over the ten years this page covers.

It also sets the capital decision in context. Spending 133% of a year’s earnings on an asset that produces nothing until 2030 depresses every per-share measure in the meantime. A company that publishes a sixteen-year per-share chart is inviting you to judge that decision on its horizon rather than on this year’s.