Toho Co., Ltd.9602.T
Toho's Film segment prints a 20.4% business-profit margin on a ~47% pre-tax return on segment assets at FY February 2026, against a Real Estate book of ¥292bn carried at a ~6.5% operating yield. The ~20x trailing consolidated P/E blends an asset-light content compounder with a dilutive property rent. The cellular SOTP reconstructs fair value to ~¥1,266 base ; spot is ¥1,229. The diagnostic is a foncier crystallisation signal or a net-cash reduction above ¥30bn. The ¥120bn TOHO VISION 2032 M&A envelope is the offsetting allocation risk.
Content at a mid-cycle ¥51bn normalised segment OP, on differentiated multiples for the asset-light Film leg and the capital-intensive IP & Anime leg, plus Real Estate on a ~1.4x book NAV, reconstructs gross enterprise value of approximately ¥970bn.
Net of ¥124bn net cash and ¥17.6bn minorities, equity value is approximately ¥1,076bn, or ¥1,266 per share base.
Consolidated market cap at spot is ¥1,044.7bn.
The Real Estate book of ¥292bn plus net cash of ¥124bn is an asset floor. The latent property NAV above book and the international IP optionality carry no implicit value at this price.
The dossier rests on one question. Does the segmental dispersion the FY February 2026 reform finally made visible — a Film segment at a ~47% pre-tax return on segment assets against a Real Estate book at a ~6.5% operating yield — reprice as the sum of the parts, or does the holding discount hold as a structural state. The SOTP arithmetic moves with the property NAV applied and the content multiple assigned. The fair-value dispersion is largely a dispersion of those two inputs.
The consensus reading takes Toho as a mature domestic studio fairly valued on a ~20x consolidated trailing P/E after the Godzilla normalisation. Property carries at book, the international IP carries nothing, the ¥124bn net cash sits idle.
The variant reading decomposes the consolidated line into three economies the multiple confounds. An asset-light content compounder — Film plus IP & Anime, ¥54.6bn combined segment OP at 20–23% margin. A prime Tokyo property rent carried at ¥292bn book against a materially higher latent NAV. A net-cash balance now actively returned, 30m shares cancelled and the count cut from 880m to 850m in April 2026.
Where the two readings separate is the property NAV and the content mid-cycle. The probability-weighted SOTP reconstructs to ~¥1,246 against spot ¥1,229. The dislocation is wide in dispersion — Bear −27%, Bull +34% — and modest at the centre. The downside is a timing disappointment with an asset floor ; the upside is a crystallisation optionality that is real and unpriced.
Position framing is deferred ownership. The fair value sits at spot, so the name is held in reserve against a crystallisation signal rather than carried in the book. Conviction is moderate. Sizing at entry is zero until a foncier signal or a net-cash reduction above ¥30bn converts the bias into a long.
The window from FY February 2016 to FY February 2026 reads as three regimes. Regular growth of the integrated studio-property operator through FY February 2020, EBIT scaling from ¥31.8bn to ¥52.9bn at a 20% margin. The COVID shock at FY February 2021, exhibition revenue −27% and EBIT to ¥22.4bn at a 4.1% ROC, while anime distribution carried the rebound and moved the value engine from the screen to content. Normalisation and repositioning from FY February 2022 to 2026 — Godzilla Minus One, record revenue ¥360.7bn, the IP & Anime segment carved out, the 1-for-5 split, the 30m-share cancellation, the ¥190bn TOHO VISION 2032 plan.
| Inflection | FY Feb 2020Pre-COVID peak | FY Feb 2021COVID trough | FY Feb 2022Recovery | FY Feb 2024Godzilla Minus One | FY Feb 2026Record |
|---|---|---|---|---|---|
| Revenue (¥bn) | 262.8 | 191.9 | 228.4 | 283.3 | 360.7 |
| EBIT (¥bn) | 52.9 | 22.4 | 39.9 | 59.3 | 67.9 |
| EBIT margin | 20.1% | 11.7% | 17.5% | 20.9% | 18.8% |
| EBITDA margin | 23.9% | 16.3% | 21.4% | 24.7% | 23.0% |
| FCF (¥bn) | 45.2 | 3.0 | 26.5 | 21.7 | 49.9 |
| Capex (¥bn) | −10.7 | −9.5 | −26.9 | −21.7 | −15.4 |
| Net income (¥bn) | 36.6 | 14.7 | 29.6 | 45.3 | 51.8 |
| Diluted EPS (¥) | 40.75 | 16.51 | 33.45 | 51.90 | 61.20 |
| ROC | 10.0% | 4.1% | 7.7% | 10.2% | 10.3% |
Source: Data pack 9602 v1.1.0, pull 29 May 2026. EPS and per-share series reflect the 1-for-5 split of February 2026. FY February 2026 = year ended 28 February 2026. The Film and IP & Anime segments are split out only from FY February 2025 ; prior years reported a single Cinema Business aggregate.
The capital return since FY February 2025 — a ¥20.06bn buyback, an October–November 2025 tender, a ¥13bn buyback, and the 30m-share cancellation against an idle balance — is corrective under the TSE reform rather than structural. Cancellation over treasury detention is the more telling signal of nascent shareholder alignment. The Management pillar at 3.0/5 captures this directly.
The engine runs on the dispersion between segments. Film operates at a 20.4% segment margin and a ~47% pre-tax return on ¥79.6bn of segment assets — asset-light distribution and production. Real Estate operates at a 24.0% margin but a ~6.5% return on ¥292bn of segment assets, 41.6% of the balance sheet. IP & Anime sits between, a 23.0% margin on a ~15.3% return on ¥112.9bn carrying capitalised content and goodwill. The four legs run at different multiples economically. The consolidated ~10% ROC is a weighted average that prices none of them correctly.
Demand is not box-office gross. Four pockets command the revenue. The distribution commission, captured at the domestic distribution committee, is the structuring claim. Exhibition occupancy at TOHO Cinemas is the cyclical leg, a 15.5% segment margin. The IP and anime licensing run-rate is lumpy, commanded by hits — Godzilla Minus One, the Demon Slayer films in distribution. Contractual property rent is quasi-bond, ~22% of revenue and decorrelated from the box-office. Half the OP comes from recurrence higher than a "cyclical studio" read implies, which is what the ~0.02 Beta brut records.
The critical cost is content amortisation and production, not a commodity. There is no commodity lag, no tariff exposure, and no FX illusion — overseas is ~10% of revenue and reported EBIT equals managerial OP without distortion. Margin volatility comes from the gap between committed content cost and realised box-office, a hit risk rather than an exogenous input. This separates Toho from the rest of the consumer bucket, where the input cost does the work.
Cash conversion is structurally clean and cyclically masked. FCF was ¥49.9bn at FY February 2026 against ¥19.1bn the prior year, the swing driven by pro-cyclical property capex — ¥32.5bn at FY February 2025, ¥15.4bn at FY February 2026. Five-year average FCF is ~¥29bn. The valuation floor is the asset base. The Real Estate book of ¥292bn plus net cash of ¥124bn anchors the downside. A ¥120bn M&A deployment below the ~5.2% WACC is the one mechanism that erodes that floor by converting cash into goodwill at an insufficient return.
Asymmetric. Film is an authentic compounder, asset-light at a ~47% pre-tax return on segment assets, and consolidated ROC of ~10.3% clears the ~5.2% WACC by some +5 points with FCF positive every year of the decade, including +¥3.0bn at the COVID trough. The consolidated return is diluted to ~9.7% by two dead weights — Real Estate at ~6.5% on 41.6% of assets, and ¥124bn of idle cash. The model is excellent at the Film level and average at the aggregate capital level. The dispersion is the SOTP material.
The lowest and most decisive pillar. The historical record is sustained undermonetisation — cash hoarding, segmental opacity to FY February 2025, ~10% overseas on global IP. The repositioning is credible and concrete: the segment reform that makes the SOTP readable, TOHO Global, share cancellation over treasury, a prudent guidance materially beaten at ¥360.7bn against ~¥300bn guided. The test sits ahead. The ¥120bn M&A envelope at unproven ROIC either crystallises the thesis or destroys it. This pillar is why the dislocation exists and why it has not closed.
Bi-modal. A quasi-bond property rent and a dominant domestic distribution base, against a lumpy hit-driven content layer. Beta brut ~0.02 confirms the decorrelation.
Prime Tokyo property irreplaceable, the distribution-committee position quasi-oligopolistic, Godzilla a global IP. The IP has been undermonetised ; exhibition is exposed to multiplex competition.
Capital return accelerating under TSE reform, cancellation ahead of treasury detention. From a low base — ~28–30% payout, ¥17.6bn minorities, an ISS QualityScore of 6.
Comparable to Toei Animation (17.5/25, quality already priced) and well above KADOKAWA (10.5/25, capital-destructive conglomerate). The grade does not justify a premium multiple on a consolidated basis. It justifies a differentiated multiple by segment. The cellular SOTP reading.
Does the sum of the parts crystallise, or does the holding discount hold ?
Is the content margin a structural floor or a hit peak ?
Content OP at a 20–23% margin reflects a sequence of hits — Godzilla Minus One, the Demon Slayer films in distribution — and consensus prices ~−6% net income for FY February 2027. The structural layer is the domestic distribution base, which captures value independent of any single hit ; the event layer is the release calendar, which does not repeat. The consolidated line does not separate the two.
Does the ¥120bn M&A create value or destroy it ?
The TOHO VISION 2032 envelope directs ¥120bn to acquisitions and multiplexes. The optimistic read is accretive consolidation and IP internationalisation ; the sceptical read is the cash-rich Japanese conglomerate deploying its balance sheet to do something at an unproven ROIC. Consolidated ROC is already diluted to ~9.7% by property and idle cash, and a deployment below the ~5.2% WACC deepens that structurally.
At ¥1,229 and a ~20x trailing P/E, the market prices an execution path without crystallisation. Content normalises on the consensus −6% net income for FY February 2027, the property carries near book with a minimal uplift, and the international IP optionality carries nothing. The −35% collapse from the 2025 melt-up returned the name to its current asset value. The consolidated EV/EBITDA reads ~11x on ¥82.8bn of EBITDA, or ~15.5x on a narrower vendor-screen definition — a definition artefact to reconcile rather than a valuation signal. The valid anchor is the asset value ; the historical multiple's 5-year average is contaminated by the COVID-trough prints.
Content retreats to a low mid-cycle on a disappointing hit calendar, the market compresses multiples in a regime hostile to domestic value, and the property stays frozen at book with no crystallisation signal. The M&A deployment begins dilutively. The downside is a timing disappointment, not destruction — the Real Estate book of ¥292bn and net cash of ¥124bn form an asset floor. Permanent loss appears only if the M&A is deployed below WACC on overpaid targets.
Toho executes without surprise. Content normalises modestly off the hit halo, the property carries near book with a ~1.4x uplift, the capital return continues at the current pace, and no destructive M&A or NAV crystallisation materialises. The cellular SOTP delivers ~¥1,266. The name remains a stable patrimonial anchor that the market has not revealed.
The crystallisation catalysts materialise together. A foncier signal reveals the NAV at a ~1.8x book, the international monetisation of the IP accelerates the overseas mix above ~13%, the capital return intensifies, and the content multiple expands as the market recognises the structural layer. The sum of the parts is finally recognised. Three of four in sequence over 18–24 months.
| KPI | Latest value | Status | What it tells us |
|---|---|---|---|
| Foncier crystallisation signal | none | Cardinal | The single trigger governing the dossier. A revaluation, disposal or REIT structuring, or a net-cash reduction above ¥30bn, converts the bias into a long and lifts the property toward full NAV. |
| Content (Film + IP & Anime) OP margin | 21.0% (FY Feb 26) | Watch | Sustained above ~18% over two prints without a major hit confirms the structural layer ; below ~15% confirms hit dependence and triggers bear. |
| First M&A ROIC vs WACC ~5.2% | not yet | Trigger | Above WACC confirms allocation discipline. Below WACC, or no disclosure on expected return, is the permanent-loss vector that erodes the asset floor. |
| Real Estate book vs latent NAV | ¥292bn book | Asymmetric | 41.6% of assets at a ~6.5% operating yield. Book is the prudent floor ; latent prime-Tokyo NAV is the unpriced upside, each 0.1x of book ≈ ¥34 per share. |
| Net cash | ¥124.1bn | Holding | ~11.9% of cap. Caps consolidated ROE at ~10% while idle ; a reduction above ¥30bn is itself a crystallisation signal. |
| Consolidated ROC vs WACC | 10.3% / 5.2% | Holding | A ~+5 point spread, diluted from the Film leg by property and idle cash. The aggregate hides the dispersion the SOTP isolates. |
| FCF (latest / 5-yr avg) | ¥49.9bn / ~¥29bn | Reference | Positive every year of the decade. Volatility is pro-cyclical property capex, not operating quality. |
| Trailing P/E | ~20x | Reference | Below a 5-year average contaminated by COVID-trough prints. Decoded by segment, not by history. |
| Performance YTD | −23% | Reference | Derating on a record EPS print. The risk is multiple and regime, not model. |
A first acquisition at a ROIC manifestly below the ~5.2% WACC, combined with a content OP margin below ~15% across two consecutive prints, reframes the dossier from value-consolidation toward value trap. The ¥190bn plan would become a capital-destruction vehicle and the holding discount a permanent state ; the asset floor itself erodes below the property book. Currently not signalled.
A foncier crystallisation signal — a published revaluation, a disposal, or a REIT structuring — or a net-cash reduction above ¥30bn converts the documented bias into a long, lifting the property toward full NAV. A material acceleration of the overseas mix above ~13% with documented international licensing milestones prices the IP optionality the consolidated line ignores.
A re-rating above ~¥1,430 or a material revision of the weighted fair value would drift the asymmetry beyond ±15% and force a return to the cellular SOTP for a fresh underwrite. The validation calendar is each FY February 2027 print, semi-annual at minimum.
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.