Bandai Namco7832.T

The interesting thing about Bandai is not the consolidated multiple, it is what that multiple averages. The group earns a 14.1% operating margin and trades at roughly 8.4x forward EV/EBITDA, which is the price of a cyclical toy-and-games conglomerate, and the market has filed it under exactly that heading. Underneath, the picture is far less uniform. One segment — Toys & Hobby — earns a 19.6% segment margin and a 21.3% ex-cash return, on a multi-generational annuity anchored by Gundam. A games business swings from loss-adjacent troughs to blockbuster peaks. An arcade arm earns roughly its cost of capital. The question the dossier turns on is whether the block discount is a correctable false-negative — recognition of the annuity plus deployment of the cash, on the Hasbro/Magic template — or the permanent captivity of a quality annuity inside an opaque conglomerate.

That distinction matters because the recent move is concentrated. Toys & Hobby operating profit doubled in two years, from ¥78.7bn in FY March 2024 to ¥126.9bn in FY March 2026, a +61% lift, while the stock de-rated from roughly 11.7x to 8.2x EV/EBITDA. Over the same window Digital swung from a ¥6.3bn trough — games written down after the ELDEN RING peak — back toward mid-cycle, with a coefficient of variation of 0.42 that makes any straight-line extrapolation of its margin a mistake. The consensus reads the post-ELDEN RING Digital reflux as a block-wide signal and the soft FY March 2027 guidance, operating profit −2.4%, as a deserved discount on the whole group.

There is a second point that reframes the case. The dossier was inherited as a deep block-discount story: a premium IP annuity supposedly trapped at the multiple of a hit-maker, the cash ignored, a gap of roughly 13% to a cellular sum-of-the-parts. Read against the certified segment data at ¥3,698, that premise no longer holds at full strength. The sum reconstructs to ¥3,938 on the Base case, +6.5%, and the probability-weighted fair value to ¥3,885, +5.0%. The deep value was at the one-year trough — the stock is down −25.6% over twelve months — not here. What is left is a favourable but modest asymmetry resting on a cash floor.

What that leaves is a recovery in recognition that is real, partly consumed, and resting on two things the price does not yet hold. One is whether Toys & Hobby holds its margin as the Overseas Gundam mix rises — Gundam Overseas grew +90% and is now 58% of the franchise, booked at distribution margins near 8% against a Japanese annuity near 18.6%. The other is whether the dormant balance sheet — ¥416.6bn of net cash, 17.6% of the market capitalisation, plus ¥161.4bn of investment securities — gets mobilised faster than the slow drip the market assumes. The return ratio actually fell last year, from 62.7% to 51.0%.

The position framing is patient observation, not ownership at this level. The fair value sits on the spot, the weighted asymmetry is slightly positive, and the shareholder yield of 3.3% pays for the wait. Conviction is moderate. The things worth watching are the Toys & Hobby margin under the mix shift and any signal on capital ; both are observable on the published calendar.

The cleanest way to read the decade is as a shift in what drives the margin. Bandai entered it as a multi-format hit-maker — mobile games and audiovisual content, operating margins in the 8–11% band, valued as a cyclical entertainment conglomerate. The gaming super-cycle that peaked with ELDEN RING in FY March 2022 lifted the operating margin to 14.1% and re-rated the stock toward 14x EV/EBITDA, then faded into a games write-down that pulled the margin back to 8.6% by FY March 2024. The recovery since has been genuine, but its source has changed: the consolidated margin is back to 14.1% even though Digital is nowhere near its peak, because Toys & Hobby has quietly taken over as the engine. Revenue grew 2.3x across the decade ; the margin did not grow with it, it oscillated with the games cycle. Any valuation anchored on a ten-year average multiple is therefore reading a denominator distorted by the peak-and-trough years in the middle.

Three management decisions shape the record. The cash has been hoarded: net cash grew +¥167bn in four years despite a rising buyback, ¥416.6bn now idle at 17.6% of the market capitalisation, with no explicit deployment thesis attached. The return ramp was both late and uneven — buybacks only began in FY March 2024, the first share cancellation came in FY March 2026, and the total return ratio actually receded from 62.7% to 51.0% last year. And the Digital over-extension into the post-ELDEN RING peak set up the FY March 2024 trough, a 1.7% segment margin with write-downs, a capital-allocation error in its own right. The discipline since — an equity-spread framework, EPS as a stated KPI, a 5-million-share cancellation, the Toei stake unwinding — is corrective and real, but the same dormant cash that could fund a faster return could also fund the next over-extension.

The engine only makes sense once you stop looking at the consolidated line and split it, because the segments are economically different businesses pretending to be one. The certified FY March 2026 data shows the spread plainly. Toys & Hobby earns ¥126.9bn of operating profit on ¥646bn of revenue, a 19.6% segment margin and a 21.3% ex-cash return — roughly 15 points above the domestic cost of capital. Digital earns 11.9%, a mid-cycle figure on a stream that has run from 1.7% to 18.6% depending on the games line-up. Visual & Music earns 12.8%, Amusement 6.6% on a 7.9% return that barely clears its cost of capital. A single consolidated multiple averages a genuine annuity with several businesses that are cyclical or ordinary — which is exactly why the sum-of-the-parts has to be built segment by segment, not in one block.

Where the annuity actually sits is in the owned IP. Gundam alone is ¥254.3bn of revenue, 18.9% of the group, a 45-year-old franchise whose adult-collector base makes the demand recurring rather than event-driven — the Toys & Hobby quarterly margins run 20–22% outside the fiscal-fourth-quarter seasonality. That is the part of the business that earns an annuity multiple. The contrast is Dragon Ball and One Piece, which are licensed: real cash flows, but revocable, a rented moat rather than an owned one. So the moat is deep and confined — owned and durable on Gundam, leased on much of the rest — and the concentration cuts both ways, because a single owned franchise carrying 19% of revenue is also the dossier's structural fragility.

The cost that governs the margin volatility differs by segment. Toys & Hobby carries IP-access cost and gunpla production and logistics, the latter exposed to tariffs — the compression of the Americas booking margin from 9.8% to 7.6% over the last year is the visible signal. Digital carries title-development cost amortised against success: a miss becomes a write-down, which is what the FY March 2024 trough was. The integration is genuine but two-edged — a tentpole anime pulls hobby, game, licensing and arcade in cascade, but it also means the consolidated margin reads the games cycle, not the annuity. The geography compounds the opacity: profit is booked roughly 86% in the Japanese IP/holding entity through transfer pricing, so the overseas margins are distribution margins, not a read on where the demand or the value actually lives.

The cash conversion is high and is part of what protects the downside. Free cash flow ran ¥128.9bn in FY March 2026, about 92% of net income, with effectively no interest-bearing debt against ¥416.6bn of net cash and a further ¥161.4bn of investment securities. The balance sheet does very little: the reported return on equity, 16.3%, sits more than 13 points below the 29.8% the operating capital earns ex-cash, and that gap is precisely the drag of the dormant pile. The most immediate lever here is allocative rather than operational — redeploying the dormant ¥416.6bn would re-rate the equity mechanically, and the price gives that optionality almost no weight.