Capcom9697.T
Digital Contents prints a 48.9% segment business-profit margin at FY March 2026, on 74% of revenue and 81% of pre-corporate segment profit. The consolidated 38.5% EBIT margin carries a weak-yen translation component of roughly 3 points and a peak-cycle Amusement contribution ; normalised at USD/JPY 130 and mid-cycle Amusement, AOP reconstructs to ~¥67bn against ¥75.3bn reported. SOTP fair value on 418.3m shares net of treasury is ¥3,177 ; spot is ¥3,069. The diagnostic is the Digital Contents segment margin over the coming prints.
Digital Contents generates ~¥70bn of normalised AOP at USD/JPY 130, on a 49% segment margin. At 18x — a compounder multiple on a catalogue annuity — the segment is worth approximately ¥1,268bn.
The cyclical and corporate poles net out. Amusement at mid-cycle, Arcade, Others, and a corporate structure capitalised at 12x sum to approximately −¥82bn.
Gross enterprise value is ~¥1,186bn. Net cash of ¥142.9bn lifts equity to ~¥1,329bn.
On 418.3m shares net of treasury, fair value is ¥3,177. Spot is ¥3,069.
The dossier rests on one question. Is the 48.9% Digital Contents segment margin at FY March 2026 the visible floor of a perpetual catalogue annuity, or the release-cycle peak of a publisher whose record print normalises as the slate thins. The SOTP fair value moves almost linearly with the multiple applied to that segment. The consolidated AOP trajectory tracks it. The fair-value dispersion is a dispersion of one number.
The consensus reading takes the consolidated 38.5% EBIT margin as a release-cycle peak. It raises forward EBIT toward a 43% margin while compressing the multiple. The stock de-rated roughly 30% over the past year on a record earnings line.
The variant reading reads the Digital Contents margin as a plateau. It has held 49–52% since FY March 2021, across years with and without a major release, on back-catalogue digital sales at near-zero marginal cost — roughly 95.8% of units sold come from the catalogue. The lumpiness sits in the new-release flow that feeds the stock ; the back-catalogue carries the recurring profit.
Two distortions sit inside the record consolidated margin. Roughly 3 points are monetary — the weak yen translates ~60% foreign revenue into reported yen profit. The Amusement segment printed a 56% margin in FY March 2026 on a Monster Hunter pachislot cycle that does not repeat. Normalised at USD/JPY 130 and mid-cycle Amusement, AOP reconstructs to ~¥67bn against ¥75.3bn reported, and the SOTP on 418.3m shares lands at ¥3,177 against spot ¥3,069.
Position framing is monitoring at current levels with a documented long bias. Fair value sits in line with spot once the share count and the yen are normalised. A pullback below ~¥2,700 opens a margin of safety ; sustained Digital Contents prints above 45% with a material buyback would confirm the re-rating. Conviction is moderate. Sizing at entry is zero.
The eleven-year window from FY March 2016 to FY March 2026 reads as three sequential regimes. A sub-margin physical publisher through FY March 2019, EBIT margin held between 15.6% and 18.1%, heavy in packaged software and a volatile Amusement segment that ran operating losses in FY March 2018 and FY March 2019. A digital-catalogue pivot from FY March 2020 to FY March 2023, EBIT margin climbing from 28% to a 40.3% peak as the digital sales mix passed ~90% and the back-catalogue became the dominant engine. A de-rated cruise from FY March 2024 to FY March 2026, margin holding 37–39%, a thirteenth consecutive year of operating-profit growth delivered while the stock fell roughly 30%, capex stepping up to ¥13.6bn on a new headquarters and R&D centre.
| Inflection | FY Mar 2016Pre-digital | FY Mar 2019Sub-margin floor | FY Mar 2021Digital pivot | FY Mar 2023Margin peak | FY Mar 2026De-rated cruise |
|---|---|---|---|---|---|
| Revenue (¥bn) | 77.0 | 100.0 | 95.3 | 125.9 | 195.4 |
| EBIT (¥bn) | 12.0 | 18.1 | 34.6 | 50.8 | 75.3 |
| EBIT margin | 15.6% | 18.1% | 36.3% | 40.3% | 38.5% |
| Digital Contents seg. margin | 23.1% | 28.1% | 49.1% | 54.5% | 48.9% |
| Net Income (¥bn) | 7.7 | 12.6 | 24.9 | 36.7 | 54.6 |
| EPS (¥, split-adjusted) | 17.2 | 28.9 | 58.4 | 87.4 | 130.5 |
| Net cash (¥bn) | 13.2 | 43.9 | 64.4 | 90.4 | 142.9 |
| FCF (¥bn) | −1.5 | 17.8 | 12.3 | 14.7 | 17.8 |
Source: Data pack and workbook 2026-05-31. EPS split-adjusted for the three 1-for-2 splits (March 2018, March 2021, March 2024), on a ~418m weighted-average count. FY March 2026 = year ended 31 March 2026. Net cash = net of total debt ; net debt is negative throughout.
Net cash built from ¥13.2bn to ¥142.9bn across the window on a ~1.5% payout and no material buyback, reaching 53% of equity. Reported ROE sits at ~22%, roughly ten points below the ~42% ROIC ex-cash — the cost of the dormant balance sheet. The Shareholder-alignment pillar at 3.0/5 captures it directly.
The engine runs on the catalogue annuity. Digital Contents carries 74% of revenue and 81% of pre-corporate segment profit at a 48.9% margin. Arcade Operations runs at 12.5%, Amusement Equipments swings from operating losses to a 56% peak depending on the pachislot cycle, and unallocated corporate structure has grown to −¥12.2bn, 6.2% of revenue. The consolidated 38.5% is the core diluted by roughly ten points. The SOTP separates the poles because they earn different multiples.
The IP catalogue — Resident Evil, Monster Hunter, Street Fighter, Devil May Cry, Ace Attorney — is a near-perpetual asset that re-sells across platforms and console generations at near-zero marginal cost. The RE Engine, a shared internal development infrastructure, holds unit production cost below the AAA budget inflation the peer set carries. The first is structural. The second is conditional on development cost staying contained.
The critical cost is capitalised development content, not a commodity input. A title's development cost sits in inventory for two to four years before flowing into COGS at release ; inventory days run at 247 and the cash conversion cycle at 285.6 days, while segment D&A is 1.1% of Digital Contents revenue. The work-in-progress balance doubled from ¥27bn to ¥56bn across the window. That balance is the channel through which development-cost inflation would show. If capitalised content outgrows catalogue revenue, the operating profit per title falls and the cost moat closes.
Reported FCF conversion is lumpy. FCF/EBIT ran from 24% to 98% across the decade and prints 24% in FY March 2026 on the capex step-up. Once the content build is modelled as a working-capital movement rather than capex, FCF/AOP normalises toward ~60% — the weak headline conversion is an artefact of content capitalisation. The balance sheet is a fortress : ¥142.9bn net cash, ¥5.1bn gross debt, EBIT/interest at 1,298x. Capcom is a compounding thesis on operating profit, not a free-cash-flow yield.
Scalability is cellular. Digital Contents earns 48.9% on 74% of revenue at near-zero marginal cost ; reconstructed ROIC ex-cash runs ~42% against a 7% cost of capital, and incremental catalogue sales drop at ~65% margin. The balance sheet carries ¥142.9bn net cash and a −1.78x Net Debt/EBITDA. The limit is cash conversion — lumpy at 24–98% of EBIT before the content build is separated from capex, and a 285.6-day conversion cycle. This is the pillar that carries the thesis.
The annuity is decoupled from the release calendar. Thirteen consecutive years of Digital Contents operating-profit growth, from ¥12.2bn to ¥70.6bn, against an evidently lumpy slate ; roughly 95.8% of units sold come from the catalogue. The beta is 0.41. The constraints are franchise concentration — Monster Hunter is the heaviest pillar — and recurrence by repeat purchase rather than contract. No subscription layer carries a SaaS premium under the segment’s own pricing.
IP depth is near-perpetual and not replicable by the peer set at short notice. The RE Engine cost moat is real but conditional on development cost staying below AAA inflation. No contractual switching cost ; distribution depends on third-party stores.
Execution is first-order — thirteen years of OP, the digital transition carried to completion, reported net margin 27.9% against a 29.1% normalised. The late pivot (five years at 16–18%) and passive capital allocation cap the pillar.
Net cash at 53% of equity holds reported ROE ~10 points below ROIC ex-cash. Payout ~1.5%, no material buyback. The cash dilution is less extreme than peers, and the TSE regime raises the pressure to act.
A compounder with a proven structural margin, capped by allocation. Above Koei Tecmo (14/25), the cleanest margin profile in the sub-industry, comparable on quality to Konami (18/25). The grade does not justify a premium on a consolidated basis. It justifies a premium on the Digital Contents segment. The cellular SOTP reading.
Is the Digital Contents catalogue a perpetual annuity, or a hits publisher at a release-cycle peak ?
How much of the record margin is structural mix versus weak-yen translation ?
Roughly 60% of revenue is earned abroad while development cost is incurred in yen. At a realised USD/JPY near 150 in FY March 2026 against a 130 normative, ~3 points of the consolidated margin are monetary. Normalised at 130, EBIT reconstructs to ~¥61.7bn against ¥75.3bn reported, a structural margin near 33–35%. The valuation neutralises the translation at 130 ; the yen has held 130–162 for four years, so 130 is a conservative reference rather than a forecast.
Does the dormant cash convert into a re-rating ?
Net cash of ¥142.9bn, 53% of equity, holds reported ROE ~10 points below the ~42% ROIC ex-cash. Payout is ~1.5% with no material buyback. The base case assumes a ¥20–30bn annual repurchase that compresses the share count toward 378m by FY March 2031 ; none is announced. Konami re-rated P/B from 1.6x to 4.55x on a balance-sheet conversion that the TSE regime now pushes.
At ¥3,069 and 18.4x forward earnings on the certified 418.3m share count, the market prices Capcom as a quality compounder near fair value. Two anchors do the work. The reported 38.5% EBIT margin is treated as the run-rate, carrying its ~3-point weak-yen component into the forward line. The 56% Amusement margin of FY March 2026 is read as repeatable rather than a pachislot-cycle peak. Normalised at USD/JPY 130 and mid-cycle Amusement, AOP is ~¥67bn and the SOTP lands at ¥3,177. The de-rating of the past year took a rich compounder to fair, not to a discount. Per-share metrics here use the certified 418.3m count net of treasury, not the 533.0m gross register some feeds report.
Development-cost inflation and a thin slate compress the Digital Contents margin and its multiple together toward 14x. Amusement falls back to a cycle trough. The yen reverts toward 130 and removes the monetary margin. SOTP recompresses to ¥2,222. The downside is timing, supported by ¥143bn net cash and the catalogue annuity ; permanent loss requires the cost moat to close over several years.
The catalogue compounds ~9% a year, the digital mix progresses, the slate executes without a major surprise, Amusement runs mid-cycle. Digital Contents holds an 18x multiple on normalised forward AOP. The SOTP delivers ¥3,177. A consolidated re-rating happens or does not ; the fair value does not require it.
A major hit lands, the ~65% incremental Digital Contents margin is realised, the catalogue re-rates toward a 21x perpetual-annuity multiple, Amusement recovers, and a material buyback converts the dormant balance sheet. Three of four in sequence over 18–24 months.
| KPI | Latest value | Status | What it tells us |
|---|---|---|---|
| Digital Contents segment margin | 48.9% (FY26) | Cardinal | The single number governing the dossier. The coming prints close Debate 1. Sustained ≥45% validates the annuity ; two consecutive prints below 45% trigger the bear case toward ¥2,222. |
| Two consecutive DC prints below 45% | not yet | Trigger | Bear-confirmation invalidation. Reframes the dossier toward ¥2,222 fair value. |
| Capitalised dev content (WIP) vs catalogue revenue | WIP ¥56.2bn | Signals | The cost-moat channel. WIP growing >15% YoY without matching catalogue revenue confirms development-cost inflation and OP-per-title decline. |
| Buyback announcement ≥5% of float | none | Trigger | Bull-confirmation materialisation. Converts the ¥142.9bn dormant balance sheet and closes part of the discount on its own. |
| Consolidated EBIT (normalised, FX 130) | ~¥61.7bn | Holding | Against ¥75.3bn reported. The structural earnings line, ~3 points below the yen-inflated print. The line valuation anchors on. |
| AOP normalised (SOTP) | ~¥67.4bn | Asymmetric | Digital Contents ¥70.4bn plus cyclical and corporate poles. Anchors the SOTP base case at ¥3,177. |
| USD/JPY (spot / normative) | 159.3 / 130 | Watch | Neutralised in valuation, drives the reported margin. A move toward 130–140 reveals the monetary component of the print. |
| Net cash / equity | 53.4% | Watch | Holds reported ROE ~10 points below ROIC ex-cash. Capacity for a buyback or strategic action exists. |
| ROIC ex-cash (reconstructed) | ~42% | Reference | Against a ~7% cost of capital. The spread is the intrinsic value engine, confirmed by the cellular reading. |
| Performance 1Y | ~−30% | Reference | The quality-purge de-rating that took the stock from rich to fair on a record earnings line. |
Two consecutive Digital Contents segment-margin prints below 45%, with the work-in-progress balance growing faster than catalogue revenue, reframe the dossier toward bear and compress fair value toward ¥2,222. A yen reversion toward 130–120 removes the monetary margin component as the second-order confirmation.
A material buyback above ~5% of the float closes part of the discount and confirms the shareholder-alignment pillar. Digital Contents margin sustained above 50% across release and non-release years confirms the perpetual annuity at the upper multiple. A structurally weak yen near 145 lifts normalised AOP toward ~¥67bn and the stock reads closer to cheap.
A structural break in IP monetisation — piracy, a platform shift, or durable franchise fatigue — would invalidate the catalogue annuity and force a complete re-underwrite this initiation does not attempt. A transformational acquisition deploying the net cash at a destructive price would do the same. 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.