Footwear Distribution Dashboard.

The two names sit inside the same TSE bucket and trade on opposite logics. ABC-Mart (2670.T) and United Arrows (7606.T) share a business model — multi-brand specialty distribution with an own-label layer — over two orthogonal economies. One is a footwear-oligopoly compounder whose value is locked by a fortress balance sheet; the other is a premium-apparel cyclical whose value turns on a margin expansion not yet proven. A sector EV/EBITDA or a blended P/E is an analytical error of the first order. The dispersion is the only read.

The decade settled the rest. Over ten years both names structurally underperformed the TOPIX — ABC-Mart roughly flat on price, United Arrows down close to half, against a market up around 160% — in every regime, without exception. The return profile of each is a multiple phenomenon, not an earnings one, but the verdict inverts with the cause. ABC-Mart earned a +73% rise in earnings per share over the decade and saw all of it erased by a P/E that compressed from ~18–20x to 14.7x — a mispricing. United Arrows printed flat per-share value on a near-pure de-rating from a 4.2x price-to-book in 2015 to ~1.6x today, as the operating margin slid from 7.9% to 5.5% — a justified re-rating of an unconfirmed growth premium. The market de-rated both toward a near-identical price-to-book for reasons that point in opposite directions.

The retreated return on capital makes the inversion legible. ABC-Mart's reported 12.1% return on equity sits below United Arrows' 14.8% return on capital; strip ABC's dormant cash and capitalise United Arrows' off-balance-sheet leases and the order reverses — ~24% ex-cash for ABC-Mart against ~8–11% lease-adjusted for United Arrows. What follows below sorts the pair on that retreated basis. The margin engine and the cross-operator inputs describe what is shared; the archetype map and the names section separate them; the mispriced variables and the structural watchlist track what each consensus is reading wrong and what would force a reframing.

The decisive structural fact across the pair is that the entire EBIT-margin gap — 16.7% for ABC-Mart against 5.5% for United Arrows in FY 2026 — is a cost-intensity gap, not a gross-margin gap. Both earn a gross margin in the low-fifties: ABC-Mart 50.7%, United Arrows 52.4%. What separates them is the wall of cost below the gross line. ABC-Mart runs an SG&A ratio near 34%, the signature of a dense, self-service store network at scale; United Arrows runs 46.8%, the cost of a high-service select-shop model — staff, well-sited stores, logistics, an omni-channel layer. That service is the brand and it is also the cost that caps the margin. Pricing cannot move it; only volume on a semi-fixed base can.

The gross margin itself is where the quality divides. ABC-Mart's is structural: it held inside a 50.1–53.9% band through COVID and a ¥160 yen, near-complete oligopoly pass-through demonstrated under stress. United Arrows' is cyclical: it ceded to a 45.2% trough in the COVID year and in warm winters, because seasonal markdown is the most violent gross-margin lever and the fashion assortment carries obsolescence (~121 inventory days). The same low-fifties headline is a floor for one and a mid-cycle point for the other.

The cost-base architecture also runs both ways under volume. ABC-Mart's replacement-staple demand absorbs the volume shock — operating profit stayed positive through the COVID trough at an 8.9% margin where the apparel names fell into loss. United Arrows carries a −43% decremental: the semi-fixed wall drops almost half-weight onto operating profit when revenue falls, so margin is destroyed roughly 2.7 times faster than it rebuilds, and a markdown cycle produces an outright loss — FY March 2021 printed a ¥7.2bn net loss on a 22.7% revenue fall. The same SG&A leverage is a buffer for the footwear staple and a weapon against the apparel cyclical.

The cash bridge completes the picture, and here both fail differently. ABC-Mart generates healthy free cash flow on a low-capex model but traps it — net cash roughly quadrupled to ¥213.9bn and has sat above half of equity for nine years, a non-restitution pathology rather than a generation one. United Arrows converts poorly: free cash flow collapsed to 0.1% of operating profit in FY March 2026 on a ¥5.5bn capex spike, the dividend was part-funded by ¥2.67bn of short-term borrowing, and net cash fell from ¥6.3bn to ¥1.6bn in two years. One name has too much cash and will not return it; the other has too little and is straining four claims on it.

The first cross-operator input is the yen, and it runs inverted relative to the exporter universe. Both names are importers: a weak yen at ~150–160 is a cost headwind in the gross margin, never an inflated revenue line waiting to mean-revert. The corollary is the part the market misreads as a risk — a stronger yen toward 130 would relieve the cost of goods, an upside, not a discount. ABC-Mart passes the cost through almost in full under its oligopoly pricing; United Arrows passes it through partially via precision pricing. Any bearish FX normalisation is proscribed across the bucket. The one nuance is ABC-Mart's inbound overlay: a weaker yen also carries tourist demand, so a yen reversion thins inbound even as it relieves cost.

The second input is the re-rating mechanism, identical in form and opposite in lever. In this bucket the multiple re-rates through the return on equity via the price-to-book, never through an autonomous P/E expansion — the price-to-book of each has tracked its return on equity across every regime, under the TSE governance discipline. For ABC-Mart the lever is restitution: returning 25% or 50% of the cash lifts the return on equity from 11.6% toward 13.4% or 15.9%, pulling it back toward the ~24% ex-cash operating return. For United Arrows the lever is margin: carrying the operating margin toward the ~6.4% medium-term target without markdown. The first is an act of governance, dateable but not signalled; the second an act of execution, not yet delivered. Any thesis that models a P/E re-rating without a return-on-equity move is inoperative.

The third input is the demand split, which sorts the defensiveness. ABC-Mart rests on a domestic replacement staple — low price-elasticity, decorrelated from the fashion cycle, beta 0.29, operating profit positive through COVID — overlaid with a cyclical, FX-dependent inbound stream that the company does not isolate. United Arrows rests on fundamentally discretionary demand — revenue fell 22.7% in COVID — dampened but not annulled by the 1.64m-member UA Club, where member sales grow double-digit and repeat purchase rises. Member loyalty is real but never contractual; there is no hard switching cost in either name, and the hero brands in ABC-Mart's case belong to Nike and adidas, leaving it exposed to their direct-to-consumer shift. There is no aggregate macro tailwind to lean on — each re-rating has to come from its own specific catalyst.