Komatsu

Phase A — Understand the business

Company Overview

Komatsu is the world's #2 maker of construction and mining equipment, behind Caterpillar and ahead of the pack (Deere, Hitachi Construction Machinery, Sandvik, Epiroc, Liebherr, Volvo CE). It is a ~120-year-old Japanese industrial that has quietly become one of the two global champions of autonomous, electrified heavy iron — the frontier that makes it interesting to a robotics/automation lens rather than a sleepy cyclical.

How it makes money. Three reported segments (year ended 31-Mar-2026, "FY2025"):

• Construction, Mining & Utility Equipment (CMU) — the engine. Net sales ¥3,806B (~92% of consolidated sales), segment profit ¥491.1B, but profit fell 18.0% YoY even as sales edged +0.2%. This is excavators, wheel loaders, dozers, and — the crown jewels — ultra-class mining haul trucks, drills, and the FrontRunner autonomous-haulage system.
• Retail Finance — the captive lender that finances customer equipment purchases. Net sales ¥126.1B (+2.4%), segment profit ¥36.6B (+24.4%). Small but a margin-and-lock-in flywheel.
• Industrial Machinery & Others — large stamping presses for automakers + excimer-laser service revenue for semiconductor lithography (via Gigaphoton). Grew on press and semi-laser maintenance demand.

Customers. Global mining majors (copper, iron ore, coal, gold — the FrontRunner fleet anchors major copper mines in Chile and Australia); construction contractors and rental fleets, heavily weighted to North America (~25% of revenue), Japan, Asia/Oceania, and Latin America. Distribution runs through captive dealer networks (e.g. Santiago, Lima) plus independent parts depots.

Contract structure & payment terms. Capital-goods model: large up-front equipment sales (cyclical, order-driven) plus a growing recurring aftermarket (parts, service, retrofit, autonomy subscriptions). Management targets parts-and-services at ~50% of CMU segment sales in the medium term, up from the low-40s% pre-2020 — the deliberate shift from box-selling toward annuity revenue. Retail Finance adds take-or-pay-like financed-receivable streams.

Supply Chain

Komatsu is vertically deep — it makes its own engines, hydraulics, and now its own batteries — which is both a moat (Lens 3) and a tariff liability (Lens 5). The named chain:

Upstream inputs →

• Steel & castings — the dominant raw input; global steel prices drive COGS. Japanese and regional mills.
• Powertrain — Komatsu builds much of its own (diesel engines, transmissions), unusual among peers; reduces third-party dependency but concentrates cost in-house.
• Batteries — brought in-house via the American Battery Solutions (ABS) acquisition (Detroit, closed Dec 2023); capacity being quadrupled from 0.6 GWh/yr to 2.6 GWh/yr by 2026.
• Electronics / sensors / autonomy stack — GPS, LiDAR, radar, controllers for FrontRunner AHS; Sony Semiconductor Solutions is a named imaging-sensor partner via the EARTHBRAIN JV. Chip/sensor supply is the single-source-risk chokepoint for the autonomy franchise.
• Semiconductor lithography inputs — the Industrial Machinery arm consumes and services excimer-laser components (Gigaphoton).

→ Komatsu (manufacturing) — plants in Japan, China, the US, and (post-GHH) Europe / Southern Africa / India / Chile. ~50% of US-bound equipment is manufactured in Japan and China — the exposure that the 2025 US tariff regime turned into a ¥140B/yr cost line.

→ End customers — mining majors, construction/rental fleets, and OEM press buyers (automakers).

Chokepoints / single-source dependencies: (1) autonomy sensor/compute supply (Sony + specialist vendors); (2) in-house battery ramp execution (ABS is unproven at 2.6 GWh scale); (3) the Japan/China → US manufacturing corridor, now a tariff and geopolitical chokepoint. Distribution partners: captive dealers in LatAm, independent parts depots in the Brazilian iron-ore corridor.

Competitive Advantages (moats)

Komatsu's moat is real but not the widest in the group — it is the clear #2 to Caterpillar in most categories, and #2 is a structurally harder place to earn super-normal returns.

• Autonomy scale (the durable edge). Komatsu is the first OEM to commission 1,000 ultra-class autonomous haul trucks (Apr 2026), with the FrontRunner fleet having moved >3 billion metric tons and logged >100 million cumulative km across 650+ trucks by 2024. Autonomous haulage is a switching-cost and data moat: once a mine re-architects its pit around one OEM's AHS, ripping it out is a multi-year, safety-critical project. The 14.5% tire-wear reduction from path-planning is the kind of accumulating-data advantage that compounds.
• Vertical integration (engines + batteries + controls). Owning the powertrain and now the battery lets Komatsu pursue a "power-agnostic" architecture (one truck platform that runs diesel, battery, trolley, or hydrogen) targeted for 2030 — a genuine engineering differentiator.
• Aftermarket lock-in. KOMTRAX telematics on the installed base + the parts/service annuity (target ~50% of CMU sales) create recurring, high-margin, sticky revenue and a data feedback loop.
• Brand & installed base. A century of reliability reputation and one of the two largest global installed bases → parts demand and trade-in gravity.

Bargaining power — mixed. Over customers: strong once autonomy is installed (switching costs), weaker at the point of initial sale where Cat, Hitachi, Sandvik/Epiroc compete hard on price and financing. Over suppliers: strong on commoditised inputs; weak on the autonomy sensor/compute stack and on steel (price-taker).

The key competitive vulnerability (see Lens 13): Caterpillar's MineStar takes a more open architecture — it will integrate with non-Cat loading equipment — whereas Komatsu's FrontRunner is more of a closed Komatsu ecosystem. If mines increasingly demand mixed-fleet interoperability, Cat's openness is the more defensible position and Komatsu's closed integration could become a liability rather than a moat.

Segments

Segment detail (year ended 31-Mar-2026, "FY2025"), all:

Source:. (Industrial Machinery exact figures not disclosed in the summary release — the two big segments are precise; the third is n/a at the line-item level, sales "increased.")

Prior-year comparison (year ended 31-Mar-2025, "FY2024"): consolidated net sales ¥4,104.4B (+6.2%), OP ¥657.1B (+8.2%), OP ratio 16.0%, net income ¥439.6B (+11.7%); CMU sales ¥3,798.2B, CMU profit ¥598.9B. So the one-year swing is stark: OP ratio 16.0% → 13.7% (−2.3pt) and CMU profit ¥598.9B → ¥491.1B (−18%) on flat sales.

Geography (year ended 31-Mar-2024, most recent granular split available): Americas ~¥1.65T (the largest region), Asia/Oceania ex-Japan ~¥808.7B, Japan ~¥340B, rest Europe/CIS/ME/Africa. The Americas dominance is why the US construction-demand cycle and US tariffs both hit Komatsu disproportionately.

Trend & cause: the mix is decelerating in construction (North America soft, China property weak, Asia generally weak), holding/improving in mining (copper/energy-transition capexCapital expenditureMoney spent on long-lived things — buildings, machines, servers — rather than on running costs., aftermarket pricing). The margin story is the whole story here — a record top line masking a volume-and-cost-driven profit decline.

---

Phase B — Measure performance

Earnings Result (FY ended 31-Mar-2026)

The headline is a "record revenue, falling profit" print:

• Net sales ¥4,132.8B (+0.7% YoY) — a record, but the growth is almost entirely weak-yen translation + pricing, not volume.
• Operating income ¥567.3B (−13.7%), OP ratio 13.7% (−2.3pt).
• Net income ¥376.4B (−14.4%); EPS ~¥413.92 TTM.
• USD-equivalent: sales ~$27.6B, OP ~$3.78B, NI ~$2.51B. (Cross-check: ADR-basis annual revenue reported ~$27.1B — consistent.)

vs. its own guidance — a material beat. Komatsu's initial guidance for this year (issued ~Apr 2025) was deliberately conservative — net sales ¥3,745B (−8.8%), OP ¥478B (−27.3%), NI ¥309B (−29.7%), EPS ¥334.83 — built on a strong-yen and full-tariff assumption. It was later revised up to sales ¥3,888B / OP ¥500B / NI ¥320B as the yen weakened and the US-Japan trade deal cut tariffs. Actuals (¥4,132.8B / ¥567.3B / ¥376.4B) came in well ABOVE even the revised guide — a ~¥67B OP beat vs. the revised number. This conservative-guide-then-beat pattern is a recurring Komatsu tell and matters for how to read the FY27 guide-down (Lens 11).

What drove it: price hikes and mining strength (mining-equipment sales +15% in one interim quarter; aftermarket-parts pricing) were more than offset by lower unit volumes (North America construction, China) and a ~¥140B/yr tariff cost hitting COGS. Retail Finance (+24% profit) was a bright spot.

Guidance for the year ending 31-Mar-2027 ("FY2026"): management guides "a decrease in both consolidated sales and profits" with FX assumptions ¥150/USD, ¥174/EUR, ¥106/AUD; the press release did not quantify the new-year forecast. n/a at the line-item level for FY27. Tone: cautious/defensive, citing yen, US tariffs, and soft construction demand.

Balance-sheet flags: ROE fell to 11.3% (−2.9pt) — still double-digit, near its 10-yr median of ~11%. Net-debt / goodwill / inventory line items are n/a (no filing on shelf), but two qualitative flags: management cited a ~five-month inventory buffer (built ahead of tariffs), and the retail-finance book carries the usual captive-lender leverage. Dividend ¥190/share, payout 53.8%.

Market reaction: muted-to-negative — the stock has underperformed peers because "underlying volume weakness is outweighing forex-boosted reported numbers". The tape is pricing the volume/tariff air-pocket, not celebrating the record top line.

Earnings Calls (sentiment trend)

No transcripts on the shelf; this is -sourced sentiment across the last ~4 quarters:

• Trajectory of tone: cautiously optimistic (early 2025) → defensive (mid-2025, tariff shock) → guardedly stabilising (early 2026). The dominant word shifted from "pricing" to "tariffs" to "mitigation."
• What management keeps saying: mining resilience (energy-transition/copper demand), aftermarket-and-services expansion toward ~50%, autonomy/electrification leadership, disciplined pricing, shareholder returns (buyback + ≥40% payout).
• What they added: heavy emphasis on tariff mitigation — US price hikes, supply-chain/production-footprint adjustment, and quantified savings from the July-2025 US-Japan trade truce (~$150–200M, and a further ~$140M from a US-China truce per the CEO).
• What they stopped emphasising: aggressive volume-growth framing for North American construction — replaced by "divergence" language (US vs. Indonesia, region-by-region).

Net: a management team managing a cyclical/policy trough with cost and price levers, leaning on the mining-and-aftermarket structural story to bridge to the next up-leg. Credible, not promotional.

Comps

Peer table — Komatsu vs. global construction & mining-equipment peers. Multiples are, late-June 2026 unless noted; where I could not source a figure it is n/a. Do not treat any blank as zero.

Sources:.

The single most important line in this dossier is the valuation gap. Komatsu trades at ~15x earnings — right alongside the other big Japanese name (Hitachi CM, ~15x) but at a massive discount to the Western/Nordic peer set: Caterpillar 47x, Epiroc 36x, Sandvik 32x, Deere 33x. Even allowing that CAT's ROE (51%) is ~4.5x Komatsu's (11%), and that Epiroc/Sandvik are pure-play higher-margin aftermarket-heavy mining names, a 3x P/E gap to CAT and 2x+ to the Nordics is a very wide "Japanese cyclical discount." Part is deserved (lower ROE, JPY/governance discount, #2 position, construction cyclicality); part is the setup for a re-rate if Komatsu executes the aftermarket-mix shift and the tariff line laps. Komatsu is, on these numbers, the cheapest liquid way to own global mining-autonomy.

Stock-Price Catalysts

What moves 6301 (mostly, illustrative of the reaction function rather than an exhaustive >5% log — precise daily moves not on shelf):

• JPY/USD. The dominant swing factor. A weak yen flatters reported sales/profit and the stock; yen strength does the reverse. The FY-end guidance is explicitly built on a ¥150 assumption. This is as much a yen trade as an equipment trade.
• US tariff / trade-policy headlines. Komatsu, Kubota and Hitachi each gained ≥3% on 23-Jul-2025 when the US-Japan deal cut tariffs 25%→15%. Tariff escalation is the symmetric downside catalyst.
• Mining-major capex & commodity prices (esp. copper). Copper's energy-transition/AI-datacenter demand narrative is a structural tailwind to mining-equipment orders.
• China / North America construction data. Property-market and construction-demand prints move the CMU volume story.
• Autonomy/electrification milestones (1,000th autonomous truck, trolley milestone) — narrative catalysts that support the multiple but rarely move the tape alone.
• Capital-return announcements (¥100B buyback + treasury cancellation) — supportive.

52-week range ¥4,615–¥7,840 — a ~70% high-to-low span in one year tells you this is a high-beta, macro-driven name, not a low-vol compounder.

---

Phase C — Judge people & books

Management

• CEO: Takuya Imayoshi — assumed President & CEO 1-Apr-2025. Hiroyuki Ogawa moved to Chairman; Tetsuji Ohashi (long-time architect of the Smart Construction/DANTOTSU strategy) to Senior Adviser. So the current chief is new in seat (~15 months) and is being tested by a tariff-and-cycle trough on day one — his first full-year print (this one) came in above his own conservative guide.
• Track record / archetype: career-Komatsu professional manager (the Japanese-industrial norm), not a founder. Continuity of the DANTOTSU ("uniquely superior") + Smart Construction strategy set under Ohashi/Ogawa. Steady-hand profile; the risk with professional-manager successions at this stage is strategy-by-inertia rather than bold re-allocation.
• Capital allocation — improving, shareholder-friendly:
• Buyback: ¥100B / up to 25M shares (~2.8% of shares out) authorised for Apr–Sep 2026, with treasury-stock cancellation (a real return, not a warehousing gimmick); ~¥61.7B executed by June 2026. The prior Apr-2025 program repurchased ~15M shares for ~¥70B.
• Dividend: ¥190/share, 53.8% payout, with a stated floor of ≥40% consolidated payout.
• M&A discipline: two well-aimed bolt-ons — GHH (underground mining, closed Jul 2024) and American Battery Solutions (electrification, Dec 2023) — both plug genuine capability gaps (underground; in-house batteries) rather than empire-building.
• ROE ~11.3% — respectable for a heavy industrial but a fraction of CAT's 51%; the aftermarket-mix and buyback strategy is precisely aimed at closing this.
• Skin in the game / insider ownership: n/a (no our figures; Japanese disclosure differs). Expect the typical low direct-insider ownership of a large Japanese industrial with cross-holdings.
• Red flags: none promotional or governance-alarming surfaced. The historical concern is the Joy Global (2017) acquisition price — the purchase price rose ~66% during negotiations and drew shareholder litigation (settled $20M) — but that's a legacy item under prior management, now resolved.

Assessment: a competent, capital-disciplined, shareholder-aligned professional-manager team running a well-defined strategy, freshly handed to a new CEO in a hard year. The bull needs them to keep executing the aftermarket-and-autonomy pivot; the bear worries the professional-manager culture won't move fast enough against Cat's open-architecture push.

Forensic Red Flags

Big caveat: with no filings on the research shelf (no CIK/EDGAR), I cannot do line-item forensic work on the balance sheet or cash-flow statement — those checks are n/a and flagged as a genuine gap, not a clean bill of health. What can be assessed qualitatively:

• Revenue quality / recognition: capital-goods point-of-sale plus a growing Retail Finance book — watch that finance-segment growth (profit +24.4%) isn't quietly juicing group results via aggressive financed-receivable terms or channel-stuffing to dealers. Not evidenced, but the mechanism exists; n/a on receivables/DSO.
• Inventory: management explicitly built a ~5-month inventory buffer ahead of tariffs. Pre-buying inventory is defensible strategy, but elevated inventory into a demand-soft year is a classic write-down/margin risk if the cycle worsens — the one hard forensic flag I'd put on this name from public info.
• FX / translation: "record" sales are yen-translation-flattered — a quality-of-revenue caveat, not fraud. Real-terms (constant-currency) growth is weaker than the headline.
• SBC / non-GAAP games: not a feature of Japanese-industrial reporting the way it is in US tech; low concern.
• Goodwill/intangibles: GHH + ABS + legacy Joy Global sit on the balance sheet; impairment risk exists if underground/battery bets underdeliver. n/a at the carrying-value level.

Regulatory findings (required sub-section) — per regulatory/regulatory-findings.md (fetched 2026-07-06) and web:

• SEC (EDGAR EFTS — LR + AAER): zero findings. Komatsu has no CIK and is not an SEC registrant, so no EDGAR enforcement search is possible.
• Non-SEC enforcement (web search — FTC/DOJ/FDA/consent-decree/settlement/fine): no material regulatory enforcement action found for Komatsu Ltd. in 2023–2026. The only material litigation surfaced is the legacy Joy Global (2017) acquisition shareholder suit — $20M settlement, preliminarily approved, plus a related insurer-coverage dispute (Columbia/Travelers) — all pre-dating current management and now resolved. A minor NDTX dealer contract case (Texas Construction Aggregates v. Komatsu America, 2022) exists but is immaterial.
• 10-K Item 3 (Legal Proceedings): n/a (no 10-KForm 10-KA company’s audited annual report to the US regulator. The most complete thing it publishes. on shelf; Komatsu files Japanese securities reports, not a 10-K).
• Verdict: No material regulatory or forensic red flag verified via SEC EDGAR EFTS (nil — no registrant), web enforcement search, and legal-news search as of 2026-07-06 — with the explicit caveat that no financial statements were on the shelf for line-item testing. The inventory buffer is the one item to watch.

---

Phase D — Project & stress-test

Forward Projection

Anchor (actuals, FY ended 31-Mar-2026): net sales ¥4,132.8B, OP ¥567.3B, NI ¥376.4B, EPS ~¥413.92.

Management's own frame for FY ending 31-Mar-2027 ("FY2026"): "decrease in both sales and profits," FX ¥150/USD — deliberately not quantified in the release. For reference, the pattern of the prior year's initial guide was OP ~−27% (a ¥478B initial OP guide that actuals then beat by ~19%).

Three-year EPS path — all, arithmetic shown, JPY/share. Base share count ~910M, reduced ~2.8%/yr by buyback.

• Bear (FY27E, y/e Mar-2027): management guide-down proves real + a cyclical construction leg down. OP ~¥478B, NI ~¥310B, EPS ~¥345. Roughly the level management's conservative machinery implies.
• Base (FY27E): tariff line laps into price hikes, mining/aftermarket hold, yen ~¥150 (no further tailwind). OP ~¥520B, NI ~¥345B, EPS ~¥385. (This assumes Komatsu again guides low and lands above — its established tell.)
• Bull (FY27E): copper-capex up-leg + North America construction stabilises + aftermarket mix accelerates. OP ~¥600B, NI ~¥400B, EPS ~¥446.

FY28–29E (y/e Mar-2028/2029), base case: with the tariff drag fully in the base, aftermarket-mix approaching 50%, and buyback shrinking the count, EPS compounds ~mid-single-to-high-single-digit — base ~¥415 (FY28E) and ~¥445 (FY29E). Not a hypergrowth name — a low-teens-P/E cyclical with a slow structural margin/mix upgrade and a shrinking share count.

Key swing inputs (each an independent bet): (1) JPY/USD; (2) US tariff regime post-truce; (3) copper/mining-major capex; (4) China + North America construction; (5) pace of the parts-and-services mix shift; (6) battery/autonomy ramp execution.

Per --watchlist rules, I am NOT logging a our model Brier forecast in this unattended sweep — the base EPS call is committed here for the record but not registered.

Bull vs Bear

Bull case. Komatsu is the #2 global franchise in an oligopoly with the two clear autonomy leaders, trading at ~15x vs. peers at 32–47x. Structural tailwinds: (a) mining super-cycle optionality — copper/energy-transition/AI-datacenter electricity demand drives multi-year mining capex after a decade of underinvestment; (b) autonomy + electrification moat compounding (1,000 autonomous trucks, power-agnostic platform, in-house batteries) with switching-cost lock-in; (c) aftermarket-mix upgrade toward ~50% of CMU sales = higher, stickier, less-cyclical margin; (d) shareholder returns (¥100B buyback + cancellation, ≥40% payout, 3% yield). The FY27 profit dip is ~90% mechanical (tariff + FX), ~10% demand — as tariffs lap into price hikes and if the yen holds, earnings inflect and the discount narrows. A re-rate from 15x toward even 20x on a recovering EPS is a plausible 50%+ move.

Bear case (permanent-impairment risks). (1) Caterpillar's open-architecture AHS wins the mixed-fleet future and Komatsu's closed FrontRunner ecosystem loses its autonomy edge — the moat erodes structurally, not cyclically. (2) China construction never recovers and the North American construction cycle rolls over hard, turning the 92%-of-sales CMU segment's volume decline from cyclical into secular in the construction half. (3) Perpetual yen/tariff whipsaw — as a Japan/China→US exporter, Komatsu is structurally short the trade-policy and FX lottery, capping the multiple no matter how good the equipment is.

Pre-mortem (18 months out, thesis broke): the yen strengthened back toward ¥130, wiping out the translation cushion; US construction demand fell faster than mining rose; the ~5-month inventory buffer forced discounting and a write-down; a fresh tariff escalation hit the Japan/China→US corridor; and a couple of marquee mines chose Cat's open MineStar for their next autonomous fleet expansion. EPS came in at the bear ~¥345 and the stock de-rated further rather than re-rating — because the market decided the 15x was correct, not cheap.

Are multiples too high? No — the opposite risk dominates. At ~15x, Komatsu is priced for stagnation. The asymmetry is that a little good news (tariff lap, yen stability, one up-cycle quarter in mining) re-rates it, while the bad news is arguably already in the tape (52-wk low ¥4,615).

Contrarian view (what the market refuses to see): the Street is trading 6301 as a yen/tariff cyclical and pricing the FY27 guide-down at face value — ignoring that ~90% of that guide-down is a transient policy/FX line item, that management systematically guides low and beats, and that Komatsu owns one of only two global mining-autonomy platforms at a 3x P/E discount to the other one. The market is anchored on the construction cyclicality and blind to the autonomy/aftermarket annuity building underneath.

Devil's Advocate (short-seller)

Dismantling the bull:

• The moat is narrower than #2-of-two implies. Being permanent runner-up to Caterpillar means Komatsu is a price-follower at the point of sale and earns a structurally lower ROE (11% vs CAT's 51%). "Cheap vs CAT" can persist for a decade because Komatsu deserves a lower multiple — lower returns, JPY discount, weaker pricing power. The gap is not obviously closing.
• Revenue concentration is a double bind. ~92% of sales sit in one segment (CMU), and within it the construction half is hostage to China (property bust) and North America (rate-sensitive, softening), while the mining half is hostage to a handful of majors' capex swings. There is no diversifying business to cushion a synchronized down-cycle.
• "Record revenue" is a currency illusion. Strip the weak yen and constant-currency growth is weak; the top line flatters a business whose volumes are falling. When the yen mean-reverts, the illusion reverses.
• Autonomy could be a trap, not a moat. Cat's MineStar is open (integrates non-Cat loaders); FrontRunner is comparatively closed. If mines standardize on interoperability, Komatsu's closed stack becomes a reason to switch away. The "1,000 trucks" milestone is impressive but backward-looking — the next 1,000 fleet decisions are what matter, and Cat is logging the most autonomous hours globally.
• The inventory buffer is a landmine. A ~5-month stock built for tariffs, into a demand-soft, guide-down year, is exactly how you get discounting and write-downs.
• Capital-allocation history has a scar. The Joy Global purchase price rose ~66% mid-negotiation and drew (settled) shareholder litigation — evidence the team can overpay at the top of a cycle.

What must hold for today's ~15x/¥6,329 price: yen stays ~¥150 (no re-strengthening), tariffs stay contained post-truce, mining capex holds, and North America construction doesn't fall off a cliff. If growth disappoints 20–30% (EPS to ~¥290–330), a 15x multiple on lower earnings is ¥4,350–4,950 — i.e. back to the 52-week low, ~25–30% downside, with little valuation support below because the multiple is already "cheap."

Single scenario that permanently impairs the business: Caterpillar's open AHS becomes the industry interoperability standard and Komatsu's closed FrontRunner is relegated to a legacy installed base — the autonomy moat, the whole reason a robotics lens cares, inverts. Plausibility: low-to-moderate over 3–5 years, but it is the scenario that would justify a permanently lower multiple rather than a cyclical one.

Management Questions (ordered by information value)

1. As mines increasingly run mixed OEM fleets, will FrontRunner open up to control non-Komatsu haulage/loading equipment — and if not, how do you defend against Caterpillar's open MineStar architecture?
2. Of the FY-ending-March-2027 profit guide-down, how much is tariff/FX (transient) vs. underlying volume (structural) — and at what quarter do you expect the tariff line to be fully offset by price?
3. What is the realistic timeline and margin bridge for parts-and-services to reach ~50% of CMU segment sales, and what does that do to through-cycle operating margin?
4. You built a ~5-month inventory buffer ahead of tariffs — what is the write-down risk if the construction cycle stays soft, and how fast can you unwind it without discounting?
5. What return on invested capital do you underwrite for the American Battery Solutions 0.6→2.6 GWh ramp, and at what mining-electrification adoption rate does it pay back?
6. How dependent is the autonomy franchise on single-source sensor/compute suppliers, and what is the mitigation if that supply is constrained?
7. Given ROE of ~11% vs. Caterpillar's ~50%, what is your explicit ROE target and the timeline — and why should the market close the valuation gap?
8. How structurally exposed is the CMU segment to China construction on a 3–5 year view, and what is the plan if Chinese demand is permanently lower?
9. What is your through-the-cycle capital-allocation priority — buyback vs. M&A vs. capex — now that GHH and ABS are integrated?
10. How do you think about hedging or de-risking JPY/USD so reported results are less of a currency trade for shareholders?
11. What is the aftermarket attach and retention on the autonomous/electrified installed base vs. conventional equipment — is the annuity genuinely stickier?
12. Where are you on power-agnostic commercialization (battery pilots Q2-2026, trolley, hydrogen) and what is the customer TCO case vs. diesel today?
13. What did you learn from the Joy Global acquisition price escalation, and how does that shape M&A discipline under the new leadership?
14. How is the North American manufacturing footprint evolving to reduce the Japan/China→US tariff exposure — will you localize more production?
15. What are the two or three risks on the horizon that you believe the market is currently mispricing for Komatsu?

---