Financials

Financials — What the Numbers Say

TOWA reports in Japanese yen and closes its fiscal year in March. Throughout this page, "FY2026" means the year ended March 31, 2026 (the company's own label is "FY2025"). All amounts are in ¥ millions or ¥ billions unless stated otherwise. Ratios, margins, and per-share figures are unitless.

1. Financials in One Page

TOWA is a small Japanese semiconductor equipment company — ¥194B market cap, ¥54B of revenue — that dominates one narrow but strategically important niche: compression molding equipment used to encapsulate chips during back-end packaging. Through-cycle the business earns mid-teens operating margins, converts most of its profit to cash, and runs a fortress balance sheet (net cash, equity ratio over 65%). FY2026 broke that pattern: orders snapped back to a record ¥54.4B (+1.7%), but gross margin fell 340 bps, operating margin compressed from 16.6% to 12.7%, and free cash flow turned negative ¥1.4B as the company built inventory and stepped up capex to ¥5.5B for the FY2027 ramp. Management's own FY2027 guidance — sales ¥64B (+17%), operating profit ¥10.2B (+47%) — implies the margin damage was transitory; the market's 42× trailing P/E is pricing it that way. The financial metric that matters most right now is gross margin recovery, because everything below it (operating leverage, FCF, multiple compression) keys off whether 33.8% was a one-quarter mix problem or a structural ceiling.

2. Revenue, Margins, and Earnings Power

How TOWA makes money. Roughly 92% of revenue comes from the Semiconductor Manufacturing Equipment segment — selling encapsulation molding systems, singulation/dicing equipment, plating equipment, and the consumable precision molds that go inside them — to OSATs (outsourced assembly and test houses) and IDMs (integrated device manufacturers) in Taiwan, China, Korea, Japan, and Singapore. The remaining ~8% is fine-plastic medical devices and laser processing machines. Equipment revenue is order-driven and cyclical; aftermarket molds and parts are recurring and steady. The 5-year revenue arc reflects a single big secular event: TOWA's HBM/advanced-packaging franchise emerged in FY2022 and pushed the run-rate from ¥30B (FY2021) to ¥50B+ (FY2022 onward).

Revenue has held a ¥50–54B band for four consecutive years even as orders, mix, and end-customer urgency have swung dramatically. That stability is the first thing to internalize: TOWA isn't a growth story right now — it is a high-share, normalized-margin business waiting for the FY2027 advanced-packaging cycle to lift volumes by another ~17%.

FY2026 is the chart that should worry you. Gross margin fell from 37.2% to 33.8% (−340 bps), operating margin fell from 16.6% to 12.7% (−390 bps), and net margin nearly halved. The company attributes this to upfront costs for new products and an unfavorable product mix as customers paused HBM-related orders during the Q1 air-pocket. The bullish read is that the FY2027 guidance restores operating margin to ~16% on ¥64B sales — i.e., classic operating leverage as volumes recover. The bearish read is that competitive entrants in compression molding (Apic Yamada, ASMPT) and a stalled push into TC-bonding (where Hanmi dominates) are compressing the through-cycle margin ceiling.

The quarterly print explains the full-year picture: Q1 FY2026 was a disaster — sales of ¥8.1B with a ¥581M operating loss — as customers digested inventory after the AI-driven advanced-packaging pull-in of FY2025. Q2 and Q4 recovered to normal run-rate; Q3 stayed soft because of product-mix headwinds (the FY2027 ramp had not yet hit revenue). For underwriting purposes, Q4 FY2026 — ¥17.4B sales at an 18.5% operating margin — is the most relevant data point because it shows the business already operating near its through-cycle margin profile.

3. Cash Flow and Earnings Quality

Free cash flow is the cash a company generates after paying for the operating working capital and capital expenditures needed to run the business. For an equipment maker like TOWA, FCF moves with two things: working capital (inventory + receivables) and capex (factory and tooling investment). It is normal for FCF to be lumpy.

Over the six-year window, TOWA generated cumulative net income of ¥37.3B, operating cash flow of ¥38.7B, and free cash flow of ¥13.5B — meaning roughly one-third of reported profits ended up as cash for shareholders after working-capital build and capex. That is below the 50–70% conversion you'd expect from a mature capital-equipment maker and reflects two structural realities: TOWA's customers (OSATs) buy in lumpy cycles that swing inventories sharply, and the company has been reinvesting heavily in factory capacity for the HBM/advanced-packaging ramp.

The FY2026 cash deterioration is two-sided. Operating cash flow dropped from ¥10.4B to ¥4.1B because (a) profit before tax fell from ¥11.2B to ¥7.0B and (b) inventory swelled to ¥19.8B (from ¥15.8B) as TOWA pre-built equipment for FY2027 deliveries. Capex stepped up to ¥5.5B (from ¥4.8B) for the same ramp. Both are deliberate, growth-related decisions rather than distress signals — but until the FY2027 sales materialize, the cash hole is real.

Distortion	FY2025 (¥M)	FY2026 (¥M)	Direction
Inventory build	included in OCF	+¥3.9B drag	growth-related
Receivables	+¥1.5B	+¥4.6B drag	timing
Capex (PPE + intangibles)	4,758	5,525	rising
Tax cash outflow	3,087	2,412	falling with profit
Acquisitions / goodwill	none	none	clean
Stock-based comp	de minimis	de minimis	clean

The reassuring news: there are no acquisitions, no goodwill write-downs, no aggressive revenue-recognition pulls, and stock-based compensation is immaterial. The FY2026 FCF gap is straightforward working-capital + capex, not accounting gymnastics.

4. Balance Sheet and Financial Resilience

TOWA carries one of the cleanest balance sheets you'll find in semiconductor equipment.

The company ended FY2026 with ¥28.4B of cash against ¥18.0B of total debt (including ¥0.4B of lease liabilities), leaving net cash of ¥10.5B — about 5% of market cap. Total debt nearly doubled YoY (from ¥9.9B to ¥18.0B) because TOWA drew down short-term working-capital lines to fund the inventory build and capex. Equity ratio still sits at 66.4% (down from 73.8% but well above the 50% threshold investors typically demand for cyclical industrial businesses). Interest expense is trivial — implied borrowing cost is well under 1% on Japanese floating-rate bank debt.

The balance sheet's role in the investment case is permissive rather than additive: it gives TOWA the flexibility to invest through a downturn without dilution and absorbs working-capital swings without distress. It does not, by itself, justify the multiple — that has to come from the income statement.

5. Returns, Reinvestment, and Capital Allocation

Return on equity measures how much profit the company earns on each yen of shareholder capital. TOWA peaked at 22.6% in FY2022 — extraordinary for any industrial — and has since drifted lower as the equity base expanded (retained earnings + OCI) faster than profit. The FY2026 print of 7.0% is the lowest in five years and is the single biggest reason the stock derated from its FY2024 highs.

Capital allocation is conservative and predictable. Capex is the dominant use of cash (¥5.5B in FY2026), funding the Kyushu and Malaysia capacity that supports the FY2027 ramp. Dividends are paid on a stable ~30% payout-target basis (~¥2.2B in FY2026). The company does not buy back stock in any meaningful size — share count has been flat at ~75.1M (post-3-for-1 split, October 2024) — and has made no acquisitions in the window. There is no obvious value-destruction risk from acquisitions and no dilution from compensation. The criticism that can be made: by not buying back stock when it traded at 7× earnings in FY2022/23, TOWA missed a window that would have been highly accretive. The current ¥194B market cap at 42× trailing earnings is not the price at which to start.

6. Segment and Unit Economics

The segment table answers the only segment question that matters: all the economics are in Semiconductor Equipment. That segment generated ¥49.9B of sales (~92% of total) and ¥6.5B of operating profit at a 13.1% margin in FY2026. Medical Devices contributes a few billion yen of revenue at low double-digit margins and Laser Processing (the TOWA LASERFRONT subsidiary acquired from Omron in 2018) is even smaller. For modeling purposes, treat TOWA as a pure semiconductor packaging equipment company with two small optionality businesses attached.

Geography drives the cycle. Per FY2026 disclosures, Taiwan and China led the YoY sales growth, driven by general-purpose memory investment and singulation equipment demand. The dependence on Asian semiconductor capex (Taiwan + China + Korea + Singapore = roughly 70%+ of revenue) means TOWA's quarterly print swings with TSMC's, Samsung's, and SK Hynix's order patterns — not with the broader Japanese industrial economy.

7. Valuation and Market Expectations

The most important valuation point about TOWA is that it has spent the last five years trading in a range from 7× earnings (FY2022/23) to 41× earnings (FY2024/26). There is no "fair" historical multiple — there is a cyclical multiple that compresses when earnings rip and expands when earnings collapse.

The FY2026 print of 42× reflects the same dynamic as FY2024's 41×: a compressed-earnings denominator. Both periods saw the multiple expand precisely because investors looked through a weak year to a recovery. If management's FY2027 guidance is accurate (¥64B revenue, ¥10.2B operating profit, implied EPS in the high-¥80s assuming a similar tax rate), the forward P/E at the current ¥2,579 price would be roughly 29× — still rich relative to the FY2022 cycle peak's 7× but not absurd for a niche dominator with HBM exposure.

Sell-side coverage carries a Buy consensus with an average ¥3,250 target (+26% upside) and a wide range: Morgan Stanley sits at Hold/¥2,100 (worried about TC-bonding share loss to Hanmi and structural margin pressure), Jefferies sits at Buy/¥4,000 (looking through to FY2027 advanced-packaging volumes). The spread captures the binary nature of the call.

The valuation is asymmetric to the upside if you believe management's FY2027 guidance is achievable and the margin recovery is real; flat-to-down if you believe FY2026 represents a new through-cycle margin ceiling. EV/Sales of roughly 3.4× is the cheapest the stock has been on a forward-revenue basis since 2022, which is the simpler way to frame the bull case.

8. Peer Financial Comparison

Peer multiples below are each in the company's own reporting currency; comparing multiples (P/E, EV/Sales, EV/EBITDA) across currencies is valid because the ratios are unitless. Absolute market caps and revenue figures are not directly comparable.

TOWA is the cheapest member of this peer group on every multiple — by a wide margin. EV/Sales of 3.4× compares with 5× (ASMPT, also a packaging equipment specialist with HBM exposure), 6× (KLIC), 18× (DISCO, the Japanese precision back-end peer), 35× (BESI, the European hybrid-bonding pure-play), and 64× (Hanmi, the Korean TC-bonder leader at the height of the HBM theme). The cheapness is partly deserved: TOWA's FY2026 ROE of 7% lags BESI's 25%+ and Hanmi's 30%+, its FY2026 gross margin of 34% sits below ASMPT's high-30s and DISCO's mid-60s, and narrative leadership in TC-bonding has gone to Hanmi. The cheapness is partly opportunistic: TOWA still owns 60%+ of the compression molding niche, and the multiple does not require heroic assumptions to clear.

9. What to Watch in the Financials

What the financials confirm: TOWA is a high-quality industrial niche dominator — high gross margin through the cycle, low leverage, no acquisition baggage, no dilution, predictable capital allocation. The 5-year scoreboard (ROE peaked at 22.6%, equity ratio above 60% throughout, no goodwill, no SBC, no buyback gymnastics) is exactly what you want to see before underwriting a cyclical.

What the financials contradict: The "structural" margin story. Gross margin compressed 340 bps in a single year, FCF turned negative, and ROE fell to 7%. None of that is consistent with a business earning durable economic rents. Either the FY2027 recovery is real and FY2026 was a transitory mix problem (the management view), or the through-cycle margin ceiling is structurally lower than the FY2022 peak implied (the bear view).

The first financial metric to watch is FY2027 gross margin — the first-half print in November 2026. A clean snapback above 37% on rising volumes would validate the ¥4,000 Jefferies target and is the condition under which a re-rating toward 25× forward earnings becomes consistent with the data. A stuck-at-34% print on the same volumes confirms structural margin compression, supports the Morgan Stanley ¥2,100 view, and breaks the bull case.