Financials

Figures converted from JPY at historical FX rates — see data/company.json.fx_rates. Ratios, margins, and multiples are unitless and unchanged.

Financials

Disco is a small-revenue, mega-margin franchise. Sales went from roughly $0.59B in FY2001 to $2.74B in FY2026 (about 7.4% annual growth in JPY; the USD path also includes yen translation effects), but the real story is what those sales now drop to the bottom line — operating margin moved from 28% to 42%, gross margin from 52% to 70%, and ROE has held above 22% for four straight years. The balance sheet carries effectively no debt and $1.78B of cash; capital is returned almost entirely through a payout-ratio-linked dividend rather than buybacks. The single metric to watch is the cash-conversion cycle: FY2026 operating cash flow set a record of $837M, but free cash flow turned slightly negative because management parked roughly $850M into long-dated time deposits — a balance-sheet allocation, not a deterioration in the business. Valuation is the friction point: at roughly 46–60× trailing earnings and ~35× EV/EBITDA, the market has already priced "best-in-class semi-cap consumables compounder."

"Quality Score" and "Fair Value" composite scores were not available in the rankings probe, so this page leans harder on raw margin, return, and cash-conversion evidence to make the quality case.

1. Revenue, Margins, and Earnings Power

Disco's earnings statement reads like a slow-build compounder hidden inside an equipment supplier. Volumes are cyclical with semiconductor capex; margin structure has stepped up four times in 25 years (2003 recovery, 2010 recovery, the FY2015–18 step-change, and the AI-era step-change since FY2022).

What this chart actually proves: every cyclical downturn (2002 dot-com, 2009 GFC, 2012–13 PC/handset, 2019 trade war, 2024 mid-cycle pause) is shallower and shorter than the last, while the upcycle peaks keep stepping higher. Operating income on the most recent print ($1,160M) is roughly 7× the level Disco achieved at its FY2001 peak on only 4.6× the revenue (USD basis; JPY-basis multiples are 9× / 6×) — the leverage is real, not just a cycle.

The right way to read this picture: Disco's gross margin used to be 47–52% in the 2000s. From FY2018 on, it sits structurally above 58%; on the latest two prints it is at 70%. That is a consumables-rich signal — high-margin blades, wheels, and recurring service revenue scale with the installed base, dragging the company's blended gross margin up even when equipment shipments grow faster. The operating-margin trajectory is the same story with a 5-year lag: SG&A and R&D are roughly fixed, so each step up in gross profit drops almost entirely to the operating line.

Recent quarterly trajectory — has the momentum already turned?

The Q4 FY2026 (Jan–Mar 2026) print of $834M revenue and a 44.2% operating margin is the highest quarterly margin in the dataset and slightly above the prior peak Q4 FY2024 (44.2%). Sequentially the upcycle has shown a clean step pattern: Q1 dips on calendar-quarter shipments, then climbs into Q4 fiscal year-end on equipment acceptance testing. Earnings power is still expanding, not normalizing.

2. Cash Flow and Earnings Quality

Free cash flow is cash generated after the capex needed to keep the business running. For Disco it's the cleanest test of whether the headline margins are real or just timing.

For most of the 25-year history, operating cash flow comfortably exceeds reported net income — the green bars sit above the dark blue bars. That is the signal you want: working capital is funded out of cash earnings, not stretched payables. Three years stand out for noisy free cash flow:

FY2022: OCF $687M but FCF only $329M because capex jumped to $375M (Kuwana facility build). This is growth capex, not maintenance.

FY2025: OCF $805M; FCF dropped to $350M as capex hit $467M (17.8% of revenue) for the Kuwana-2 expansion. Again, deliberate.

FY2026: OCF $837M (record); FCF turned slightly negative at -$14M because cash flow from investing was about -$851M. Per the FY2026 earnings call, most of that investing outflow was long-dated time deposits, not capex — capex itself fell back to $205M (7.5% of sales). The reported FCF figure is therefore misleading: cash earnings did not collapse, the treasury just moved cash out of "cash & equivalents" into "time deposits" reported under CFI.

Operating-cash conversion (the dark teal line) sits at or above 1.0× in nine of the last twelve years — high quality. FCF conversion swings because Disco's capex is lumpy facility-build capex, not maintenance. The right lens: smoothed FY2022–FY2026, OCF averaged ~$717M and FCF averaged ~$343M; over the same five years the company spent about $1.27B on capex, of which Disco has stated roughly two-thirds was the Kuwana facility build (a multi-year capacity step-up tied to AI/HBM demand).

3. Balance Sheet and Financial Resilience

The balance sheet is, frankly, embarrassingly strong. Equity ratio (shareholders' equity ÷ total assets) has been above 70% every year since FY2002 and reached 78.9% in FY2026. Interest-bearing debt is immaterial; the company has been net cash every year in the dataset.

Three signals worth flagging:

Current ratio is roughly 3.2×. Disco is structurally over-liquid. Long-dated time deposits are now showing up as the management response — capital that yielded near zero in the cash line is being termed out for incremental basis points.

Goodwill is immaterial. Disco has built the franchise organically, so the balance sheet is not propped up by acquisition accounting.

PP&E has more than doubled in five years, from $992M (FY2021) to about $1,400M (FY2026). That is the Kuwana-2 and Hiroshima R&D facility build. The capex cycle that depressed FCF in FY2025 is now finishing, and depreciation will catch up slowly given Japanese real-estate accounting (long useful lives).

4. Returns, Reinvestment, and Capital Allocation

Disco's returns trajectory mirrors the margin story. ROE re-rated from a 7–15% band (2007–2018) into a 22–28% range from FY2022 onward. Because the company is virtually all equity, ROE ≈ return on assets here — there is no leverage juicing the number.

The capital-allocation pattern is unusual for a high-quality compounder:

Disco does not buy back stock at scale. Share count has barely moved — 96M shares in FY2001, 108M in FY2026, mostly from option exercise long ago. There is no aggressive repurchase program reducing the denominator the way US peers (AMAT, LRCX, KLAC) do.

Dividends are variable and policy-driven, with a target payout ratio in the 35–45% range. Headline DPS swings (reported as $6.88 in FY2023 vs $2.03 in FY2024) reflect the policy, not a cut — payout-ratio percent has remained in the 36–62% range.

Reinvestment is in plant, not deals. Five-year capex (FY2022–FY2026) totals about $1.27B, all internal capacity build. Goodwill remains immaterial.

The honest read: Disco is letting cash build on the balance sheet rather than buying back stock when shares look cheap. That is a missed opportunity in cyclical troughs (FY2019, FY2024) when valuation compressed. But it is also why the equity ratio sits at 79% — management treats balance-sheet armor as a feature for the next downturn.

5. Segment and Unit Economics

Disco discloses revenue by product line and by region in its IR English data pack. The product mix tells the franchise story; the geography mix tells the customer-concentration story.

What the segment chart actually shows: precision processing equipment (PPE) — the cyclical part — stepped up from ~50% of revenue to ~63% as the AI/HBM capex wave landed. Consumables held steady at 22% of a much larger revenue base. That means consumables revenue in absolute dollars has roughly doubled in five years, even though its share looks flat. Consumables-plus-maintenance is now ~32% of revenue, locked-in by Disco's massive installed base — and is widely understood to be substantially higher margin than the equipment line.

Overseas sales were 89.6% of revenue in FY2026 (Japan at just 10.4%). The dominant slice is "Asia ex-Japan" — TSMC, SK hynix, Samsung, Micron, and Chinese OSAT/IDM customers. This is where the geopolitical-risk and customer-concentration topic in the company write-up gets its weight; financial performance is heavily tethered to TSMC/HBM capex.

6. Valuation and Market Expectations

The valuation question is the only place the financial picture gets uncomfortable. The company is exceptional, the business is exceptional — but the price already reflects that.

The multiple has traded a wide range: P/E was 18–32× in the FY2021–FY2023 period, exploded to 73× in FY2024 on the AI-frenzy spike, collapsed to 26× during the FY2025 correction, and now sits near 49× FY2026 ending and 39.5× on FY2027 consensus. Anchor read: the multiple is currently mid-to-high in its own historical range, not at the absolute peak.

What the price implies, if you triangulate:

At ~39.5× FY2027 consensus EPS, the market expects EPS growth to continue compounding at 15–20%. WSJ shows FY2027 consensus annual EPS at $10.30 (vs $7.84 in FY2026), implying ~30% EPS growth — i.e., the multiple normalizes if the AI/HBM capex cycle remains intact through FY2027.

At ~35× EV/EBITDA, Disco trades at a meaningful premium to AMAT (20×), LRCX (19×), KLAC (23×), and even ASML (25×). That gap exists despite Disco being smaller in absolute revenue — the premium is paying for higher margins and the consumables annuity.

At ~14× P/B, the multiple already capitalises a return on equity that would need to stay above 25% for many years. Any reversion of ROE toward the historical 14–18% band would compress P/B substantially.

The bear / base / bull frame is intentionally crude. Current share price near $470–500 (the 6146.T close near ¥75,000 × ~$0.00631/¥ ≈ $473; DSCSY ADR near $440) sits between the Base multiple of 35× and the Bull multiple of 50× — closer to bull. To justify the current multiple, FY2027 EPS needs to clear roughly $9.50–10.30, which is in line with the analyst consensus the WSJ shows.

7. Peer Financial Comparison

Disco's peer set is unusual — there is one true head-to-head (Accretech in dicing/grinding) and a set of valuation/franchise benchmarks (AMAT, LRCX, KLAC, ASML, TEL) that are larger, broader, and not direct product competitors.

Note: Disco / Accretech / TEL revenues converted at FY2026-end (March 2026) FX. Operating-margin figures use the latest available fiscal year for each. Accretech and TEL trailing balance-sheet and multiple data were not retrieved in the peer pull. Disco's "ROIC" approximated as net income / (equity + debt) given immaterial debt.

The peer gap that matters: Disco's operating margin (42.3%) is the highest in the set, ahead of KLA (39.3%) and ASML (36.9%), while its capital structure is the most conservative (78.9% equity ratio vs. KLA at 29% and AMAT at 56%). Disco's forward P/E (~46×) and EV/EBITDA (~35×) sit above every peer except a brief overlap with ASML. The market is paying a premium for what it sees as the highest-margin, lowest-balance-sheet-risk operator in the set. The discount to Accretech (the only real product competitor) is 20+ margin points and a higher multiple — both deserved, given the franchise data.

Note that Accretech and TEL ROIC are not shown (data unavailable in this pull). The point of the chart is positioning, not absolute size.

8. What to Watch in the Financials

Closing read

The financials confirm a high-quality, capital-light, consumables-rich franchise with a level of margin and balance-sheet defense rarely seen in semi-cap equipment. Earnings convert to cash — the FY2026 negative FCF is an accounting optics issue tied to time-deposit reclassification, not a deterioration in operating cash generation. Returns are durable: ROE has now exceeded 22% for four years on an essentially all-equity capital base, which is the cleanest signal of franchise economics.

The financials contradict the idea that the cycle has already turned. Quarterly margins, sequential revenue, and order-flow guidance for Q1 FY2027 all suggest the upcycle is still extending, not normalising — yet sell-side consensus has only modest EPS growth penciled in beyond FY2027.

The first financial metric to watch is FY2027 operating margin. If Disco prints Q1 FY2027 (July 2026) at the guided $263M operating income on $665M revenue (39.6% margin) and holds 40%+ through the seasonal trough, the structural margin step-up is confirmed and the current 39–46× P/E remains defendable. If margin compresses below 35% on softer mix, the multiple has more downside than the absolute EPS print, because the entire bull case rests on consumables-driven margin durability.