Sumitomo Rubber Industries5110.T
The recovery is real and certified: Business Profit margin back to 7.52% from a 1.31% tyre-segment trough. The inherited reading was a balance-sheet discount waiting to protect the downside. Priced against the two metrics this archetype allows, that discount was consumed in a single year — the tangible net-asset gap to peers collapsed from −46% to −80% down to −12.5% when a debt-financed brand right was capitalised at market. What is left to decide is narrower: whether a margin already above its own cycle average has anywhere left to go.
Rebuilt geography by geography at the pod's 130/150 rates, the FY2025 top line loses ¥112bn — about 9.3% — of currency it does not control; at the decade EV/sales median of 0.628x that normalised revenue supports an enterprise value of ¥688bn.
Net of ¥256bn of debt at 31 March 2026 and ¥21bn of minorities, equity is ¥410bn, or 1,559 JPY per share on the 262.8m shares net of treasury — the base-case point on the primary metric.
The market capitalises Sumitomo Rubber at ¥584bn, 2,222.5 JPY.
To close that gap at the decade EV/sales median, the yen has to sit at 179 to the dollar in perpetuity — weaker than spot, weaker than the company's own 157 guidance, weaker than any level in the eleven-year series. The discount the inherited thesis was built on was spent in FY2025.
The interesting thing about Sumitomo Rubber is not that the recovery happened, it is that the recovery is already behind the company rather than in front of it. Consolidated Business Profit margin is back to 7.52%, up from a 1.31% tyre-segment trough in FY2022, and the market has filed the name under a familiar heading: cheap cyclical on the way up. Rebuilt cellularly across eleven years, the cycle-average Business Profit margin is 6.79%. Realised is 7.52%, guided is 8.48%. The company is sitting 52 to 148 basis points above its own normative level, not climbing toward one that is higher. The dossier turns on that single fact, because the narrative supporting the price assumes the destination is above today, and the reconstruction says it is below.
There is a second point that reframes the whole case. The name was inherited as a balance-sheet story: a group trading below book, protected by an asset discount that would floor the downside. Read against the two metrics this archetype actually allows, that premise does not hold. The discount to peers on tangible net assets ran between −46% and −80% for a decade. In FY2025 it collapsed to −12.5% in a single year. The protection everyone was relying on was consumed, and the year it was consumed is documented.
The mechanism is what makes it a Piège rather than a value case. In FY2025 the group capitalised ¥132,777m of intangibles — of which ¥103,893m are the DUNLOP four-wheel brand rights, bought at market and financed by debt — at the precise moment its US capacity was destroyed, with Tonawanda closing in February 2025. Management bought a brand right instead of rebuilding capacity, on the very geographies where the capacity is now gone. Tangible net asset per share had risen from 1,467 to 2,158 JPY across nine years — the only measurable value creation of the decade — and fell 12.8% in one exercise, while net debt rose.
What that leaves is a recovery that is genuine, certified, and largely spent. The real question is the destination, not the reality. Revenue is falling 3.4% at constant currency since FY2023 while the published line rises 2.5%; currency explains 60.2% of the five-year growth; only Japan grows in local terms, at +5.07%, on the most structurally shrinking market in the bucket. Strip the yen out and the top line is contracting. The price does not hold that.
The position framing is no entry at this level. The weighted asymmetry is −28.3%, and its sign survives dropping FX normalisation entirely (−19.9% on published medians); the bull case pays +2.3%, an asymmetry ratio of 0.05. Conviction is moderate — a 3.78% dividend carry, 36.18% volatility, and an amplitude that depends on an FX convention whose timing is unknowable all cap it. The dossier passes to full modelling, sequenced after the 6 August 2026 half-year print. The diagnostic is the Q2 Business Profit and the semestrial gross margin, both on the published calendar.
The cleanest way to read the last decade is as growth that never converted. Revenue rose 51.2% across eleven years; operating profit fell 7.4%. The sequence runs through three regimes. A silent erosion between FY2015 and FY2019, when a price-led European push lifted revenue and lowered tyre-segment margin every single year. A collapse and a purge between FY2020 and FY2024, when a demand shock, an input shock and a wave of impairments took the reported margin to near zero and rebuilt the asset base by writing it down. And a bought recovery from FY2025, when the margin came back and the group spent its balance sheet on a brand. Any valuation anchored on a ten-year average earnings multiple is meaningless here — the five-year mean P/E is 77.6x against a 13.8x median, inflated by the near-zero-profit years in the middle.
| Inflection | FY 2015Pre-erosion | FY 2019Erosion | FY 2022Input trough | FY 2024Tangible peak | FY 2025Bought recovery |
|---|---|---|---|---|---|
| Revenue (¥bn) | 798.5 | 893.3 | 1,098.7 | 1,211.9 | 1,207.1 |
| EBIT margin | 11.17% | 3.70% | 1.36% | 0.92% | 6.84% |
| Gross margin | 33.83% | 28.58% | 23.05% | 29.57% | 30.52% |
| Tyre segment margin | 10.85% | 6.08% | 1.31% | 7.28% | 7.65% |
| Net income (¥bn) | 72.0 | 12.1 | 9.4 | 9.9 | 50.4 |
| Tangible net asset / share (¥) | 1,467 | 1,486 | 1,802 | 2,158 | 1,881 |
| Net debt (¥bn) | 204.4 | 264.9 | 298.9 | 230.8 | 308.0 |
Source: certified DataDesk addenda (S28/S29) and Tanshin Q1 FY2026, pre-split basis. EBIT = reported operating income; tyre segment margin = segment Business Profit margin. Net income FY2024 carries ¥44,963m of asset writedowns (Sumitomo Rubber USA plus Sports goodwill). Tangible net asset per share excludes goodwill and intangibles; the FY2025 fall is the DUNLOP capitalisation.
Three management decisions explain the shape. The FY2015–FY2019 expansion was led by price: European revenue rose 127.5% while tyre-segment margin fell every year, −477 basis points in four. Tonawanda was closed in February 2025, at the exact quarter the tariff regime turned US capacity into a rent — a self-guided ¥28,800m FY2026 hit, 31.2% of the normative margin, against three to four years to rebuild what was shut. And the domestic workforce was held up: headcount rose 13.5% over ten years against −19.8% at Bridgestone, for a Business Profit per employee that is identical a decade apart, ¥2.38m then ¥2.41m. The one accumulation that did happen — tangible net asset per share, up 47.1% from FY2015 to FY2024 — was partly reversed in FY2025 by a single debt-financed transaction.
The engine only makes sense at the level of geography crossed with the currency it is billed in, because the product portfolio is 86.5% homogeneous and tells you nothing on its own. Demand sits in three pockets. Domestic replacement, 30.0% of revenue, is the only one growing in real terms — on the most structurally declining volume market in the bucket. North America, 23.1%, is falling in local currency and is now served entirely by import under 15% to 25% tariffs, with no local plant. Europe, 18.6%, grew +4.86% as published but +0.85% in euros — 83% of the reported growth is monetary. There is no contractual revenue anywhere in the mix.
Where the value actually comes from is the point the consolidated line hides. It is not price and it is not mix. It is the yen: currency explains 60.2% of the five-year revenue growth. Reconstruct the top line geography by geography at constant rates and it contracts — ¥1,249,871m in FY2023, ¥1,219,317m in FY2024, ¥1,207,061m in FY2025, −3.4% — while the published line advances 2.5%. Only one of the five geographies grows in local currency, and it is Japan, the shrinking market. The quality of the revenue is therefore lower than the aggregate suggests, and the aggregate is the only thing the market sees.
The cost that governs the margin is input absorption, and it arrives with a three-to-five-month lag, so the shock lands before any price response catches up. The vulnerability is measured, not assumed: in FY2022, on the same accounting base and the same input shock, the tyre segment earned 1.31% while Bridgestone's passenger and light-truck line earned 11.9% and Yokohama's off-road 14.2% — a 1,060 basis-point gap in pass-through power. The 2026 proof is fresher still. US prices rose up to 25% from May 2025, a +14% average covering 80% of the tariff impact — then North American replacement volume fell 11% in Q1 FY2026 and the Falken price was cut 3% to 4%. A 10% input shock is worth 176 basis points of gross margin unmitigated, the highest sensitivity in the bucket, on the thinnest margin base.
The cash conversion looks like the best in the bucket and is not what it appears. Reported conversion runs 83.9% of cumulative EBITDA, but that is produced by a crushed EBITDA denominator and chronic under-investment — capex has run below depreciation on nine of eleven years, a cumulative ¥103bn shortfall. After total investment, real conversion is 23.6% of cumulative EBITDA; the published free cash flow overstates eleven-year generation by ¥162,590m. The value vector is the yen, and the price treats a currency the company cannot control as if it were a franchise it owns.
This pillar carries the thesis because it decides whether the recovery creates value or merely restores a line. It does not create value. Return on invested capital ex-cash is 6.77% against a cost-of-capital coverage threshold of 7.08% — the working-capital burden, 45.4% of invested capital, lifts the threshold above the 6.84% published margin. The marginal decade margin on the core business is 1.60%: ¥361bn of tyre revenue for ¥5.8bn of Business Profit. A model whose return on capital does not clear its cost of capital, and whose incremental revenue converts at 1.60%, does not compound as it grows — it dilutes. The normalised margin, 5.76% after a currency drag and a recurring-charge provision, sits 132 basis points below the threshold. The only discipline that held was on structure costs, and that ratio has already turned back up, from 21.05% to 23.00% since FY2022.
The moat is the second cardinal because its absence is what makes the recovery banalised and the downside unprotected. The one real position is Dunlop in domestic Japan, where pricing is least contested — but it covers 30% of revenue and stops there. No segment earns above 11.0%, and the best of them is 3.1% of revenue. The pass-through gap of 1,060 basis points in FY2022 is the cleanest possible measure of the missing rent: same shock, same accounting, no ability to pass it on. There is no US capacity left, and the DUNLOP rights bought in 2025 are a legal asset acquired at market price, not a historical advantage — European revenue grew 0.85% in euros in FY2025, so the rent they were meant to buy is unproven. Deep enough to matter at home, absent everywhere it would need to be.
Constrained. The only real-growth pocket is domestic replacement, on the most structurally shrinking volume market in the bucket. Zero contractual recurrence; the North American pocket falls in local currency.
The FY2023–FY2025 recovery is real, 1.31% to 7.65% at segment level. The decade record is not: profit per employee flat over ten years, headcount up 13.5%, and a brand bought where capacity was needed.
No dilution ever — zero buybacks in eleven years, a rigorously stable share count. That is decisive for what it forbids: the register is locked at 28.85% by Sumitomo Electric, so the capital-return lever is not in management's hands. The dividend is amount-based, payout ranging 20% to 155%.
A model with no identifiable value pocket and no accessible correction mechanism. The lowest score in the bucket — Bridgestone 15.0, Toyo 14.0, Yokohama 13.5 — and no pillar of operational excellence anywhere. The decisive pillar is the economic model: a business that destroys value as it grows. Governance follows, not for what it measures but for what it forbids. The grade is consistent with the read: a trap dressed as a discount, not a turnaround.
Is the margin still climbing toward a normative level above today, or is it already above its own cycle average ?
Is the FY2026 guidance reachable ?
The near-term diagnostic is dated. Q2 has to deliver +49.6% Business Profit as the input shock enters the accounts, and the company's own consensus sits 5.6% below the guidance it is asked to price. The guidance itself implies a 31.48% gross margin — 52 basis points under the inherited 32% threshold, which it arithmetically contradicts. And the FY2026 bridge carries a 22,000m contradiction: raw materials at +22,900m in the annual bridge against +900m in the same issuer's semestrial split, a gap larger than the entire guided profit growth. At the low end, FY2026 Business Profit is −0.9% year on year, not +23.4%.
The DUNLOP rights : rent, or a capitalised survival expense ?
The rights are carried at ¥103,893m, perpetual, unamortised, 17.8% of the market cap, and financed by debt. The annual impairment test rests explicitly on a plan to lift tyre volumes in Europe, North America and Oceania — and the issuer itself flags that the volume-uplift assumptions carry uncertainties that could materially affect the cash-flow estimates. European revenue grew 0.85% in euros in FY2025, which is not yet a rent. The counter-intuitive point is that it barely matters for the floor: a full impairment worth 395 JPY per share hits equity and the reported result but does not touch the tangible net asset, from which the intangible is already excluded.
At 2,222.5 JPY the market is pricing the full FY2026 guidance the issuer's own analysts sit 5.6% below, an operating margin roughly 211 basis points above the normalised level, and a yen weaker than anything observed. Solve the price back to the decade EV/sales median and it requires 179 to the dollar in perpetuity, against a 157–160 spot and the pod's 130 normative. The tell is behaviour: the stock returned +37.4% over twelve months while its constant-currency revenue was falling — the currency is the only coherent driver of that move. The forward P/E of 10.3x reads cheap, but that low multiple belongs statistically to the most degraded earnings denominator, which is exactly the error MR2 forbids. On the two prescribed metrics the name is the cheapest of the four in the bucket — 35.3% below the peer floor on EV/sales, 7.2% below on tangible net assets — but that discount has never resolved in eleven years. It is neither an argument for upside nor a protection; it only forbids building the thesis on relative valuation. The FX sensitivity is explicit: fair value runs from 1,290 JPY at 110 to 1,971 JPY at 160, and only reaches the spot at 179.
Rubber, up more than 15% since January, enters the accounts in the second half with the documented lag and cannot be passed on — the FY2022 proof gives 1,060 basis points of missing pass-through, and the 2026 attempt already cost 11 points of volume. A 20% input shock at the gross coefficient is −352 basis points of margin. The yen normalises past the pod's level to 110, compressing the transaction margin, and the multiple compresses toward the bottom of the corridor. This is a timing disappointment and a multiple compression — both reversible — not a permanent loss. The defensible floor is 1,412 JPY, the five-year tangible-asset median; the absolute floor 1,139.
Normalisation without the recovery. The group delivers between 90,000m and 112,000m of Business Profit — the range, not a point, because the 22,000m raw-material contradiction is unresolved and the second half carries the input turn. The yen returns toward the pod's normative over the underwriting horizon; North American volume stabilises after the 11% fall without reconquest; the European DUNLOP launch offsets the domestic volume decline without exceeding it. No impairment is taken. On normalised revenue of ¥1,094,841m at the decade EV/sales median, the sum lands at 1,559 JPY; on tangible net assets, 1,709. The fair value does not need a re-rating; it lands 26.5% below the spot.
The invisible catalysts fire together. The European DUNLOP launch produces euro revenue growth above 10%, validating the rent rather than the survival expense; the shift to the Thai hub delivers the announced cost reduction and neutralises a material part of the ¥28,800m tariff impact; the ARK project delivers its ¥9,500m; the yen stays weak at 150. Even with all three materialising simultaneously and the currency holding, the upside is +2.3%. The bull does not pay — which is the observation that governs the whole decision.
| KPI | Latest value | Status | What it tells us |
|---|---|---|---|
| Q2 FY2026 Business Profit | 6 Aug print | Cardinal | The dated diagnostic. Guidance requires +49.6% year on year. At or above 21,205m with semestrial gross margin at or above 31.5% invalidates the thesis and the consensus together; below 18,000m confirms the ceiling. |
| Consolidated Business Profit margin | 7.52% FY2025 | Above normative | The cellular cycle average is 6.79%. Realised is 52 basis points above it, guided 148 above. The recovery is already past its own normative level, not climbing toward it. |
| Semestrial gross margin | 31.5% threshold | Watch | The FY2026 guidance implies 31.48%, 52 basis points below the inherited 32% threshold it arithmetically contradicts. Below 31.0% for two exercises is the operational leg of the permanent-loss path. |
| North America replacement volume | −11% Q1 FY2026 | Signals | The first direct measure of price–volume elasticity: after +14% price, volume fell 11% and the Falken price was cut 3% to 4%. The pass-through was executed and it failed on volume. |
| Constant-currency revenue | −3.4% since FY2023 | Signals | Rebuilt at normative rates, revenue is contracting while the published line rises 2.5%. Only Japan grows in local currency. No company disclosure presents this series. |
| Implied perpetual USD/JPY | 179 | Reference | The rate that justifies the spot at the decade EV/sales median. Against a 157–160 spot, 157 guidance and a 130 normative — the market prices a yen weaker than any observed level. |
| Tangible net asset / share | 1,905 JPY | Floor | Excludes the intangible, so it is insensitive to a DUNLOP impairment. Fell 12.8% in FY2025 on the brand capitalisation. Five-year median floor 1,412; decade-low floor 1,139. |
| DUNLOP impairment exposure | ¥103,893m · 17.8% of cap | Reference | Perpetual, unamortised, worth 395 JPY per share on a full write-down. Hits equity and the result, not the tangible floor. Closing test 12 February 2027, the first post-launch. |
| Shareholder yield | 3.78% | Reference | All dividend, zero buyback, register locked. Covered 2.0x by normative cash flow and not at risk — but a 3.78% annual carry cost against any short expression of the thesis. |
The negative read breaks if the 6 August print delivers. A Q2 Business Profit at or above 21,205m — up 49.6% year on year — with a semestrial gross margin at or above 31.5% invalidates the thesis and the consensus in the same line, and returns the dossier to R3 for re-underwriting before any modelling opens. It is the single dated, discriminant trigger, and it cuts both ways: it would confirm a normative level above the cellular average and defeat the reading that the recovery is spent.
The read hardens if the narrow engine confirms the ceiling. Q2 Business Profit below 18,000m with gross margin under 31.0%, or North American replacement volume down more than 10% for a second consecutive quarter, pulls fair value toward the 1,115 bear centre. The consensus is already falling — 39.4 JPY between February and July 2026, the steepest downward leg in the series — while the multiple has not compressed, which is the configuration in which the price has further to travel than the estimates.
The permanent-loss path is the one to watch, and it is narrow by construction. It requires a gross margin held below 31.0% for two consecutive years — defeating the inherited 32% threshold outright — and a DUNLOP impairment taken at the FY2026 closing test on 12 February 2027, at which point the floor drops toward 1,139. But the tangible net asset floor at 1,905 is structurally insensitive to that impairment, which is why the most visible permanent loss in the dossier is invisible on the metric that governs its valuation. Currently not signalled.
The information provided on this website is for informational and educational purposes only and should not be construed as financial, investment, legal, or tax advice. All content reflects the personal opinions, interpretations, and analyses of the author at the time of writing and is subject to change without notice. Nothing contained herein constitutes, or should be interpreted as, a recommendation, solicitation, or offer to buy or sell any securities, financial instruments, or other investment products. The author is not a licensed financial advisor, broker, or investment professional. Any references to specific assets, markets, or strategies are illustrative in nature and do not constitute personalized investment advice. Investing in financial markets involves risk, including the potential loss of capital. Past performance is not indicative of future results. Readers are solely responsible for their own investment decisions and should conduct their own independent research and due diligence before making any financial commitments. You are strongly encouraged to consult with a qualified financial advisor, legal professional, or other relevant specialist before making any investment or financial decisions. By accessing and using this blog, you agree that the author shall not be held liable for any direct or indirect losses, damages, or consequences arising from the use of, or reliance on, the information presented herein. All content is provided "as is" without any warranties of completeness, accuracy, or reliability.