2492 — Walsin Technology Corporation
Thesis
Verdict: PASS at TWD 262 / WATCH for entry TWD 200–225. Cautious participate — cyclical trade, not a compounder hold. Conviction: Medium-Low. Walsin is Taiwan's #2 MLCC (multilayer ceramic capacitor) pure-play behind Yageo, and roughly global #6 by volume. The investment case is the cyclical second-derivative trade on the 2025–2027 MLCC upturn, with a deliberate but unfinished pivot from commodity-tier MLCC into auto-grade and AI-server high-cap parts. The stock has tripled off its 2025 low (TWD 76.30 → 266.00, a 3.4–3.5x move) and now trades at a forward P/E of ~15x that requires EPS to roughly 3.7x off trailing — the cyclical recovery is fully priced.
What has to be true. The single most important thing that must go right: gross margin must step up from ~17.6% TTM to 20%+ by H2 2026 and sustain there. This is a margin trade, not a revenue trade — Q4 2025 revenue YoY was soft (+4.7%, far below the +9.1% FY25 rate) while gross margin trend is positive (Q4 2025 18.3% vs Q4 2024 17.0%, +130bps) on ~44% incremental gross margin. The bull mechanism: China utilization recovery from ~70% to 85%+ plus auto-grade revenue mix moving from ~10% to 20% over 2026–2028 drives operating margin expansion from ~7.5% TTM toward 15%+.
The tell. Walsin's 2026 pricing posture is the structural signal: while Murata pushed 15–35% high-end MLCC hikes effective April 1 and Yageo announced 10–20% on high-voltage/auto, Walsin is still negotiating case-by-case on loss-making SKUs only. Walsin is a fast-follower, not a price-setter — its 2026 gross-margin uplift comes almost entirely from utilization, not price.
Bull vs bear. Bull: auto-grade scales 10%→20% of revenue by 2028, AI-server X6S design-ins fill the high-cap line, operating leverage on existing capacity lifts ROIC from ~5–7% (below ~9% WACC — destroying capital at trough) toward the 12–15% it hit at the 2018 peak. Bear: Chinese domestics (Fenghua, Sunlord, Three Circle) compress commodity margins faster than the mix shifts. Layered on top is a real family-conglomerate governance discount (ISS QualityScore 10, worst decile) worth roughly 10–20% — minority shareholders are passengers on Chiao-group strategy. On a governance-adjusted basis, the higher-quality, better-diversified Yageo is the better core MLCC holding; Walsin earns a slot only on a 15–25% pullback that re-creates the discount. FundamentEdge gates failed: 3 of 5 (revenue-growth primacy fails, quality fails, second derivative mixed) — this is explicitly a cyclical-trade thesis, sized and priced accordingly.
Snapshot
One-liner. Taiwan's #2 MLCC pure-play behind Yageo — a high-volume, mostly-commodity multilayer-ceramic-capacitor maker rebuilding margins as the passive cycle turns up; share has tripled off the 2025 low while EBIT margins remain ~half their 2018 peak. The bull case is Auto + AI-server mix shift; the bear case is that commodity MLCC lives and dies by Chinese supply and smartphone units.
- Legal name: Walsin Technology Corporation (華新科技股份有限公司)
- Ticker / exchange: 2492.TW, Taiwan Stock Exchange (TWSE)
- Sector / industry (GICS): Information Technology / Electronic Components — Passive Components (MLCC)
- HQ: Taoyuan City, Taiwan · Founded: 1970 (incorporated)
- Website: passivecomponent.com
Valuation snapshot (as of 2026-05-20; price_at_writeup TWD 262.0):
| Metric | Value |
|---|---|
| Share price | TWD 262.0 (profile cites TWD 263.5, last close 266.0) |
| Market cap | TWD 127.0B (~USD 4.05B); profile cites TWD 129.0B (~USD 4.1B) |
| Enterprise value | TWD 224.3B |
| P/E (TTM) | 55.5x (profile: 56.4x) — trough earnings, meaningless except as evidence |
| Forward P/E | 15.1x (profile: 15.3x) |
| EV/EBITDA (TTM) | 36.9x |
| EV/Revenue | 6.0x |
| Price / book | 4.0x (profile: 4.08x) |
| FCF yield | ~4.0% (FCF TWD 5.1B / MCap TWD 127B) |
| Dividend yield | 0.99% (vs 5-yr avg 3.2% — payout cut as price ran) |
| 52-week range | TWD 76.30 – 266.00 (3.4–3.5x bottom-to-top in 12 months) |
| Beta | 1.21 |
Note: minor cross-fragment discrepancies in price/cap/multiples (profile vs deep-dive snapshot) reflect the TWD 262.0 vs 263.5/266.0 reference points on 2026-05-20 — both preserved above.
Key stats: ~484.8M ordinary shares (1M treasury), flat for 5 years; float 471.7M (Yahoo definition); insider 41.7% (family-controlled); institutional 4.8% (unusually low). Read: forward 15x assumes the cycle inflects hard; trailing 55x is evidence 2024–2025 earnings were trough.
Business
What they do, plainly. Walsin makes the tiny passive components — multilayer ceramic capacitors (MLCC), chip resistors, inductors, ferrite beads, RF filters, varistors and chip fuses — that sit by the hundreds inside every smartphone, PC, base-station, EV and AI server. A single AI accelerator board uses 5,000–10,000 MLCCs; an EV uses 8,000–12,000 (vs ~2,000 for an ICE car); a flagship smartphone carries ~1,200 (vs ~200 in 2005); an Nvidia HGX board uses 5,000–10,000. Walsin sells these as a commodity at fractions of a US cent per piece — economics turn entirely on capacity utilization × product mix × price (auto-grade, high-cap, high-voltage parts price 5–10x general-purpose).
Segments (no public segment disclosure; estimates triangulated from industry coverage)
- MLCC ~60–65% — general-purpose (Y5V, X7R), high-cap, high-voltage, auto-grade (AEC-Q200), RF, and the new X6S high-cap parts for AI server. ASP from sub-USD-cent (general-purpose 0402) to USD 20–50+ cents (ultra-small high-cap).
- Chip resistors ~20–25% — thick/thin-film chip resistors, arrays, networks, current sensors, anti-surge. Yageo is global #1; Walsin top-5. Structurally lower margin than MLCC, even more cyclical. Walsin announced up to 20% chip-R price hikes effective Feb 1 2026 (following Yageo's lead) — the only meaningful announced Walsin hike this cycle.
- Inductors & magnetics ~5–8% — power inductors, ferrite beads, transformers; not a thesis driver (TDK/Murata lead).
- Protection / RF / other (residual) — varistors, chip fuses, EMI/RF filters, ceramic dielectric powders. RF filter line is a long-tail option on 5G base-station / IoT.
Business model
High-fixed-cost, capacity-cycle manufacturing; no recurring revenue. Margin = utilization × mix × price. ~52% direct OEM, ~40% distributors, ~8% contract manufacturing (2019–2020 disclosure — stale, flag for refresh). Margin structure TTM: Gross 17.6%, EBIT 7.5%, net 6.8% — depressed; 2018 cycle peak saw gross 30%+. Operating leverage is the entire story.
Geographic mix (latest disclosed)
Greater China ~71% · Other Asia ~16% · US ~6% · Europe ~6%. Greater-China-heavy — both a customer-concentration and tariff-risk vector. Walsin is deliberately tilting new capacity outside mainland China.
Customers
600+ direct clients; top-10 = only 20–25% of revenue — unusually low concentration for a Tier-2 component maker (protective on the downside, dilutive of single-customer upside). No single customer >20%. Channels: distributors (Arrow, WT Microelectronics, Avnet, Future Electronics) ~40% combined; Chinese Android OEMs + Taiwan ODMs ~20%; auto Tier-1s ~15%; AI-server/networking ODMs (Quanta 2382.TW, Foxconn 2317.TW, Wistron 3231.TW, Inventec 2356.TW) ~10%; industrial/IoT/lighting ~10–15%. Walked-away test: losing the largest distributor costs ~10–15% revenue, recouped within 12–24 months — not existential.
Capacity footprint
- Taoyuan (HQ + R&D) — headquarters, R&D, pilot lines.
- Kaohsiung B-Building — auto-grade and high-end MLCC; Phase 1 commissioned Q1 2023 at 600B MLCC pcs/yr; target 1.5T pcs/yr by 2028.
- Dongguan (Walsin Technology Electronics), China — general-purpose MLCC, chip-R; ~70% utilization.
- Kaohsiung Luzhu plant — dedicated to auto-grade.
- New: Malaysia automated plant under construction — auto-grade focus (confirmed April 2026 coverage).
- Kamaya Electric (Japan, subsidiary) — resistors, magnetics; legal vehicle for the Matsuo move.
- Matsuo Electric (Japan, associate) — tantalum capacitors, polymer caps, micro-fuses; stake bumped Feb 2026 by JPY508M (~$3.3M), a direct response to Yageo's KEMET tantalum strength.
Asset-heavy. Cumulative capex 2022–2025 ~TWD 9.5B; capex stepped down sharply from TWD 4.8B (2022) to TWD 1.1B (2025) as Kaohsiung Phase 1 completed. By 2027 Walsin likely runs 50%+ of capacity outside China (vs 70% China-located two years prior).
Moat & competitive position
Narrow. Cost/scale at the commodity tier — moderate, eroding to Chinese domestics. Auto/industrial AEC-Q200 qualification — Walsin's primary defensive asset; 18–36-month design-in cycles create real switching cost. PSA-group dielectric powder access via Prosperity Dielectrics (PDC, 6173.TWO) — moderate cost advantage on input (Walsin holds equity in PDC, is its largest single powder customer). Brand/OEM relationships — strong in Greater China, weaker in US/EU auto. IP/patents — modest, not differentiating vs Murata/Samsung who hold the dielectric-formulation IP. Porter read: Buyer power HIGH, Supplier power MEDIUM, New entrants MEDIUM-HIGH (Chinese capacity), Substitutes LOW (tantalum/polymer don't displace MLCC at scale), Rivalry HIGH and cyclical (pricing collapsed 2018–2020 and 2022–2024). 3-test: marginal 5-year lock-up yes; engine = PSA-allied volume manufacturing + auto mix shift (not unique vs Yageo); blank-check disruptor = yes, Chinese domestics already trying. Quality verdict: cyclical with a mix-shift option, not a high-quality compounder.
Financials
All figures TWD millions unless noted; FY = calendar year; source yfinance (cross-check TWSE MOPS before high-stakes use). Cross-fragment discrepancy flag: the fragments disagree on the revenue series — preserved both below.
Income statement & margins
| TWD M | FY22 | FY23 | FY24 | FY25 |
|---|---|---|---|---|
| Revenue (profile) | 32,800 | 32,800* | 34,760 | 37,261 |
| Revenue (deep-dive) | 35,297 | 32,800 | 34,760 | 36,463 |
| Gross profit | ~7,803 / 6,300 | ~6,236 | ~6,485 | ~6,545 / 6,300 |
| Gross margin | ~23.8% / 17.9% | ~19.0% | ~18.7% | ~17.6% / 17.3% |
| EBIT | 3,802 | 3,877 | 5,429 | 4,278 |
| EBIT margin | 11.6% / 10.8% | 11.8% | 15.6% | 11.5% / 11.7% |
| EBITDA | 9,375 | 8,990 | 10,430 | 8,493 |
| Net income | 1,651 | 1,985 | 2,983 | 2,298 |
| Net margin | 5.0% / 4.7% | 6.1% | 8.6% | 6.2% / 6.3% |
| EPS (NT$) | 3.40 | 4.10 | 6.15 | 4.74 |
*Approximate (third-party aggregator). Revenue-growth discrepancy: profile reports FY25 revenue +9.1% YoY (to NT$37.3B); deep-dive/checklist report +4.9% YoY (to NT$36.463B). Both flagged; the difference traces to which revenue base each fragment used. FY22 gross-margin figures also conflict (profile ~23.8% vs deep-dive 17.9%) — the deep-dive series is the more internally consistent and is used for the cyclical narrative; both retained for the record. 5-year revenue CAGR is ~+1.5% (sub-inflation) — a cyclical commodity manufacturer, not a durable compounder.
Second-derivative / quarterly revenue
| Q3 2024 | Q4 2024 | Q1 2025 | Q2 2025 | Q3 2025 | Q4 2025 | |
|---|---|---|---|---|---|---|
| Revenue (TWD M) | 9,351 | 8,362 | 8,740 | 9,554 | n/d | 8,755 |
| Revenue YoY % | – | – | – | – | – | +4.7% |
| Revenue QoQ % | – | -10.6% | +4.5% | +9.3% | – | – |
Q3 2025 not in the yfinance feed (defer to TWSE primary). Q4 2025 revenue +4.7% YoY is soft; Q4 2025 gross margin 18.3% (vs Q4 2024 17.0%) is the cyclical-inflection signal. The market is betting margin recovery overwhelms revenue-growth deceleration.
Incremental margin (Q4 2025 vs Q4 2024)
| Q4 2024 | Q1 2025 | Q2 2025 | Q4 2025 | |
|---|---|---|---|---|
| Revenue (TWD M) | 8,362 | 8,740 | 9,554 | 8,755 |
| Gross profit (TWD M) | 1,426 | 1,500 | 1,623 | 1,599 |
| YoY ΔRevenue | – | – | – | +393 |
| YoY ΔGross profit | – | – | – | +174 |
| Incremental gross margin | – | – | – | ~44% |
| Incremental EBIT margin | – | – | – | n/m (negative) |
Q4 2025 shows ~44% incremental gross margin — well above the ~17.6% reported book; the cyclical operating-leverage signal. But EBIT is roughly flat YoY: opex (R&D, Kaohsiung depreciation) is eating the gross-margin gain. FY26 bull case requires 40%+ incremental GM to persist AND opex discipline as depreciation flattens.
FY26E (derived, not consensus-pulled)
Deep-dive FY26E: Revenue ~40,000 · Gross margin ~21% · EBIT ~7,500 (margin ~18.8%) · Net income ~8,400 · EPS ~NT$17.38 (market consensus implied). Yahoo analyst-estimate feed returned empty for this ticker (coverage thin / Taiwan-domestic). Forward P/E 15.1x × TWD 262 = NT$17.38 forward EPS = ~NT$8.4B net income — implies net margin expanding from ~6.3% to ~21% in 12 months (vs peak FY24 8.6%); gross margin would have to step from ~17.3% to 25%+. This is the cyclical-peak-extrapolation risk priced into the stock.
Cash flow & balance sheet
| TWD M | FY22 | FY23 | FY24 | FY25 |
|---|---|---|---|---|
| Operating cash flow | n/d | 8,494 | 6,141 | 6,146 |
| Capex | -4,771 | -2,376 | -1,281 | -1,087 |
| Free cash flow | 4,162 | 6,118 | 4,860 | 5,059 |
| FCF margin | 11.8–12.7% | 18.7% | 14.0% | 13.6% |
| Net debt | 8,688 | 20,097 | 22,075 | 21,818 |
| Net debt / EBITDA | 0.93x | 2.24x | 2.12x | 2.57x |
| ROE | n/d | 4.4% | 6.3% | 4.6% / 5.2% |
| ROIC (rough) | n/d | ~5% | ~7% | ~5% |
FCF is steady NT$4–6B/yr through the trough. Capex has rolled off (Kaohsiung Phase 1 complete) — FCF should expand if revenue grows without capex re-up. Net debt stepped up in 2023 (Kaohsiung build), stabilized since; leverage moderate (~2.6x EBITDA) but rising relative to depressed EBITDA — fine if the cycle turns, sticky if it doesn't. ROIC ~5–7% is below the implied ~9–10% WACC — Walsin is destroying capital at this point in the cycle. Capex intensity 2.9% of revenue FY25 (down from 14.5–14.6% FY22). Capex/revenue: 14.5% FY22, 7.2% FY23, 3.7% FY24, 3.0% FY25. Incremental revenue per dollar of capex over 2022–2025 ~0.39–0.45x — mediocre, but most Kaohsiung Phase-1 capacity is still ramping (auto-grade qualification lags capex 18–36 months).
Industry landscape
MLCCs are the most-consumed electronic component on earth by unit count — billions per day, hundreds of trillions per year — and per-device content keeps climbing (phones ~200 in 2005 → ~1,200 today; EVs 8,000–12,000 vs ~2,000 ICE; Nvidia HGX board 5,000–10,000), with no scalable substitute (tantalum/polymer caps fill niches at 5–20x cost). Global MLCC market ~USD 13–15B in 2025, ~7–8% CAGR through 2030 (Mordor Intelligence, directional); auto MLCC sub-segment ~USD 3B growing 12–15% CAGR; AI-server passives sub-segment small (~USD 1–1.5B) but fastest-growing per-unit. Structure: consolidated at the top (top-5 ≈ 68% of revenue — Murata ~26%, Samsung Electro-Mechanics ~18%, Taiyo Yuden ~12%, TDK ~7%, Kyocera ~5%), fragmented at the bottom. Walsin sits Tier-2 with Yageo: unit-volume share ~5–7%, revenue share ~3–5% (mix-light on highest-ASP). The 2026 market is bifurcating ("polarization"): high-end (AI, auto, high-voltage) at >80% utilization with 15–35% price hikes pushed through; commodity (general-purpose consumer) sluggish, flat pricing, China utilization mid-70s. Cycle position: mid-cycle recovery (trough Q3 2024); MLCC up-cycles run 18–30 months, so ~6–18 months of tailwind may remain. Secular tailwinds: EV electrification (4–6x MLCC content/car), AI server build-out, 5G/RF, localization. Structural threat: Chinese domestics (Fenghua 000636.SZ, Sunlord 002138.SZ, Three Circle 300408.SZ, EYANG) scaling fast in commodity. See sector page: passives-mlcc
Management
Overall management grade: 🟡 Yellow (B-) — disciplined family-conglomerate passenger profile; not value-destructive, not value-creating.
Leadership
- Yu-Heng Chiao (焦佑衡) — Chairman. Chiao-family principal; 34-year board tenure; 2.67% stake (~NT$3.1B). Simultaneously Chairman of Winbond Electronics (2344.TW), Director of Walsin Lihwa (1605.TW) and Nuvoton (4919.TW) — multi-entity executive. Strategic direction sits with him, not the President. Educational/operational background not verifiable in English; career entirely Chiao-group.
- Ming-Tsan Tseng (曾明燦) — President & R&D Officer (no separately-titled CEO). ~2.8 yrs tenure; Director of Dongguan Walsin Technology Electronics since Jun 2024; former Director at Apaq Technology (5312.TWO, tantalum/polymer, 2020–2023) — plausibly connected to Walsin's tantalum/Matsuo strategy; Bloomberg also links to Inpaq Technology. Technical/R&D background, not finance/M&A. Stake and compensation not disclosed in English — data gap.
- Tse-Kuang Yeh — Head of Financial & Accounting, Asst. VP (acting CFO); internal career staff.
- Yeh-Jeng Chen — Asst. VP, Chief Legal Officer, Head of Corporate Governance (5.5 yrs); 0.0062% (~NT$7.3M).
- Jui-Chang Huang — Chief Internal Auditor (3.5 yrs).
- Jui-Jung Chang — VP; 0.012% (~NT$13.8M). Chun-Hsiung Peng — VP; 0.017% (~NT$19.4M). All internal-promotion, token equity (Taiwan cash-bonus + small-share culture).
Ownership & alignment
Aggregate insider 41.7% (family-controlled bloc — Chiao family + Walsin Lihwa cross-holding, ~38–39% ex-named-individuals). Institutional 4.8% — unusually low, virtually no external accountability vector. Walsin Lihwa is the largest single shareholder (exact % only in Chinese-language MOPS). No disclosed insider buying or selling in the last 12 months (Simply Wall St table empty; Taiwan pre-announcement disclosure runs through TWSE/MOPS in Chinese) — absence of selling into the tripling is mildly bullish, absence of buying at the bottom mildly bearish/neutral. The Chairman's wealth is spread across the Chiao conglomerate (Walsin Lihwa likely largest, Winbond material, Walsin Tech 2.67%, Nuvoton + others) — Walsin Tech is plausibly 10–20% of his personal public-equity net worth, not dominant. Minority shareholders of Walsin Tech are passengers on Chiao-group strategy.
Capital allocation (grade B-)
Zero dilution over 5 years (share count flat 484.8M) — positive. Disciplined dividend: cut from FY19 peak NT$16.33 to FY24 NT$2.15 / FY25 NT$2.40 (payout ~51% of EPS), held rather than borrowed to maintain — appropriate; yield 0.99% has not kept pace with the re-rating. Buybacks: light/opportunistic — ~NT$351M in 2021 when depressed; no program since; did NOT buy at the 2023–2024 trough (shares NT$70–100 vs current NT$262) — a missed value-aware opportunity. M&A: light, no transformational deal (contrast Yageo's KEMET/Chilisin/Nexensos stack); only strategic stakes (PDC since 2005, Kamaya, Matsuo). Mediocre capex ROI to-date but Kaohsiung still ramping (deferred verdict). SBC trivial: TWD 12.2M / ~TWD 36.5–37.3B revenue = 0.03% — no comp dilution; no US-style PSU/PRSU hurdle structures (Taiwan structural — cannot run performance-grant forensics).
Governance flags
Board 7 directors (term Jun 2025–Jun 2028), 3 independent (43%) — meets Taiwan TWSE minimum (1/3) but doesn't exceed. ISS Governance QualityScore 10 (worst decile, April 2026), anchored by Board pillar score 10; Audit pillar 2 (best decile — well-structured audit committee), Compensation pillar 8 (weak), Shareholder Rights 5 (mid). Comp committee (per Jun 17 2025 announcement): Fan Po-Kang (independent), Ng Chien-Chun (independent), Chang Pi-Lan — 2 of 3 independent. Limited refresh (only 1 new director in 3 years — Ker-Hsin Chang, <1 yr). No dual-class/poison pill (Taiwan law has no equivalent; control via Chiao/Walsin Lihwa ~38–39% bloc). No activist filings, no proxy contests.
Chiao group & related-party scan
Walsin Tech is one node in a Chiao-family conglomerate (Walsin/Hua Hsin group, founder Ting Piao Chiao 焦廷標, 1966): Walsin Lihwa (1605, cables + stainless, Chairman Yu-Lon Chiao, older brother), Winbond (2344, DRAM/flash), Nuvoton (4919, MCU), HannStar Board (5469, PCB), HannStar Display (LCDs). Not a fraud red flag — normal Taiwan 財團 governance, same template as Formosa Plastics/Fubon/Far Eastern; openly disclosed (all major entities publicly listed), no shell entities, no asset migration, no revenue circularity (sells to 600+ external customers). The single material related-party flow: Walsin Tech ← PDC (6173) dielectric powder under the Passive System Alliance (PSA, Walsin-led since 2005); Walsin holds equity in PDC and is its largest powder customer (estimated NT$1–3B/yr purchases); arm's-length per Taiwan code but pricing requires Chinese-language MOPS verification — yellow, flagged. No red flags identified. Credibility moderate-to-high: Kaohsiung Phase 1 delivered on schedule; candid pricing posture (admits Tier-2 fast-follower rather than claiming leadership); no broken commitments in English-language sources (follow-through ~80% on assessable items, small sample).
Data gaps deferred to TWSE MOPS / Chinese-language follow-up: Walsin Lihwa parent stake %; President Tseng holdings & comp; Chairman detailed comp; PDC powder RPT terms; 12-month insider-transaction tape; cross-entity Chiao-group flows; detailed segment revenue mix; Apaq connection of President Tseng.
Catalysts & risks
Catalysts
Near-term (0–12 months):
- Q1 2026 earnings (~May 2026) — the key catalyst window; first read on cyclical recovery; consensus looking for >15% YoY revenue growth and gross-margin step-up to ~20%+. Beat/miss on GM is the read.
- Monthly Taiwan revenue prints (next ~June 10 for May data) — second derivative is the cyclical signal.
- AI-server platform design-in announcements via ODM channel; per April 2026 Digitimes, Walsin "successfully entered the supply chains of major AI server manufacturers, focusing on high-margin X6S and medium-to-high voltage MLCCs."
- Auto-grade Kaohsiung Luzhu plant ramp; Malaysia plant commissioning timeline.
- Possible move from case-by-case to broad MLCC/chip-R price hikes (following Yageo/Murata) — a positive signal if it happens.
Medium-term (1–3 years): auto revenue share 10%→20% (mgmt target); Kaohsiung B-Building Phase 2 toward 1.5T MLCC pcs/yr by 2028; high-cap AI X6S volume scaling; continued PSA / Matsuo / Kamaya / Japan consolidation.
Risks (bull/bear)
Top-3 to the position: (1) Cyclical recovery stalls in H2 2026 — Chinese commodity capacity online, smartphone units decline, AI-server capex digests; likelihood medium; impact GM stuck 18–19%, EPS misses the implied NT$17.38, multiple compresses to ~12x, price → TWD 130–170 (-35–50%). (2) Walsin lags Tier-1 on price recovery — Murata/Samsung capture high-end margin, Yageo captures auto specialty, Walsin gets leftover commodity; likelihood medium-high (its own pricing posture supports this); impact GM stuck ~19%, ROIC stays below WACC. (3) PSA/Chiao-group strategic decision overrides Walsin Tech minority interest — cross-entity reallocation, RPT pricing changes, dilutive M&A; likelihood low-medium; impact governance discount widens.
Structural risk register: Commodity-MLCC pricing cycle (HIGH, structural, not closable — only managed via auto/high-cap mix). Chinese domestic competition Fenghua/Sunlord/Three Circle (HIGH, partly closable by exiting commodity into spec'd parts). China demand / customer concentration 71% (MEDIUM-HIGH, slowly closable, geographic mix shifts a few hundred bps/yr). Late-cycle pricing posture (MEDIUM, closeable if cycle persists). Tariff / export-control sideswipe (MEDIUM, mitigatable not closable). Capital-allocation passenger to Walsin Lihwa group (LOW-MEDIUM, structural, not closable). PSA related-party powder pricing (LOW, disclosed, not closable). Dilution risk LOW (484.8M shares flat 5 yrs; no ATM/shelf/convertible; self-funded). Key-person risk MODERATE (Chairman Chiao the strategic anchor, no public succession plan; President short-tenured but replaceable from internal R&D bench; continuity high, transparency low).
What invalidates the thesis: broad cycle stall + Walsin gross margin staying sub-19% through 2026; Q1/Q2 2026 GM failing to exceed 19%; cycle-peak signal from Murata/Samsung; Chinese domestic capacity announcement >30% incremental supply; insider selling disclosed to MOPS (would not surprise at current price). Behavioral traps operative (3 of 6): FOMO (stock tripled off the low; swarm itself is a "look at MLCC peers" exercise), narrative seduction (auto+AI mix shift real but unfinished; mgmt's own case-by-case pricing admits no pricing power yet), recency bias (one good quarter of 18.3% GM extrapolated to 20%+ sustained).
Valuation / DCF
Multiples vs peers (forward P/E · EV/EBITDA TTM · P/B)
| Company | Ticker | Fwd P/E | EV/EBITDA | P/B | Comment |
|---|---|---|---|---|---|
| Murata | 6981/6981 | 6981.JP | ~25x | ~13x | 2.5x |
| Samsung Electro-Mechanics | 009150/009150 | 009150.KS | ~15x | ~7x | 1.2x |
| Taiyo Yuden | 6976/6976 | 6976.JP | ~17x | ~10x | 1.4x |
| TDK | 6762/6762 | 6762.JP | ~22x | ~9x | 2.0x |
| Yageo | 2327/2327 | 2327.TW | ~14x | ~12x | 3.0x |
| Walsin | 2492.TW | 15.1x | 36.9x | 4.0x | Cyclical recovery priced |
| Fenghua | 000636.SZ | ~30x | ~15x | 3.0x | China growth premium |
Walsin's forward P/E is in line with Yageo and similar to Samsung Electro-Mechanics — reasonable on the surface. But EV/EBITDA 36.9x is well above peers (depressed TTM EBITDA; would compress to ~12–14x on the bull-case forward EBITDA, in line with Yageo); P/B 4.0x is the high end (only Fenghua higher, on a China premium Walsin shouldn't get); EV/Revenue 6.0x is priciest. Trailing 55x is meaningless except as trough evidence.
DCF / cyclical-fair frame
Cycle-average earnings power ~NT$11–13/share (trough NT$5–6, peak NT$15–18; FY18 hit NT$26, exceptional). At a cycle-average 14–18x multiple, implied cycle-fair value TWD 175–235. At TWD 262 the stock prices above the cycle-average frame and roughly at the cyclical-peak frame (NT$15 × 17x = TWD 255).
- Bull case (peak EPS NT$17 × 17–19x): TWD 290–320
- Base case (NT$12–14 × 16–18x): TWD 200–245
- Bear case (NT$8–10 × 14–17x): TWD 130–170
At TWD 262: above the base-case midpoint, below the bull-case midpoint. Risk/reward skewed to the downside — upside ~+12% to bull, downside ~-35–50% to bear. Implied expectations at TWD 262: FY26 revenue ~+10% to ~TWD 40B, gross margin to 21%+ (from 17.6%), EBIT margin to 18%+ (from 7.5%, ~2.5x), net income ~3.7x trailing. Plausible but steep. Would NOT buy 10–15% higher (TWD 290–300) — that prices the bull case in full at fwd ~17x, zero margin of safety. Target price (cyclical-fair, 12-mo): TWD 245–290; net, at the bottom of fair. R&D ~2.7% of revenue (TWD 986M FY25) vs Murata 8%+ / Samsung 5%+ caps frontier leadership; watch the layer-count frontier (~500 → 800+ over 2027–2028 would unlock GM upside).
Decision log
- 2026-05-20 — Profile, deep-dive, mgmt-dd, buy-checklist (mlcc-peer-comparison swarm, price TWD 262.0, mcap TWD 127.0B / ~USD 4.05B). Cross-checked vault peers Yageo (2327), PDC (6173), Taiyo Yuden (6976), Murata (6981). SA mirror searched — no Walsin / MLCC sub-sector coverage; no SA contradiction to flag.
- 2026-05-20 — Deep-dive verdict: cautious participate, Conviction Medium-Low. Cyclical recovery fully priced at TWD 262 (fwd 15x implies EPS ~3.7x trailing); buy on a 15–25% pullback to TWD 200–225. Target price TWD 245–290.
- 2026-05-20 — Mgmt-DD verdict: 🟡 Yellow (B-). Disciplined family-conglomerate passenger profile; ISS QualityScore 10 (worst decile); governance discount ~10–20% is real and should be priced. No red flags identified. "Would I trust the Chiao family with my capital? Conditionally yes — at the right price." On a governance-adjusted basis, Yageo is the better core MLCC holding; Walsin only at a meaningful pullback.
- 2026-05-20 — Buy-checklist verdict: PASS at TWD 262 / WATCH for entry TWD 200–225. FundamentEdge gates failed 3 of 5 (revenue-growth primacy fail, quality fail, second derivative mixed) — cyclical trade not compounder. Behavioral traps operative 3 of 6 (FOMO, narrative seduction, recency bias). Pre-buy scorecard: thesis clear yes; business understood yes; valuation reasonable NO at 262 / YES at 200–225; technicals do not support buying now (stock at 52-week high, RSI likely >70).
- Exit criteria upfront: take profit above TWD 320 (bull fair value) or fwd P/E >22x → trim; stop/re-evaluate below TWD 180 (also cited TWD 150 as -25% stop from a TWD 200 entry) or if Q1/Q2 2026 GM fails to exceed 19%; time stop if no QoQ acceleration / GM stuck <19% by H2 2026. Max loss tolerance -25% from entry.
- Position-sizing schedule (by entry): TWD 262 → 0% skip or 0.5–1.0% starter tracker; TWD 200–225 → 1.5–2.5%; TWD 175 → 2.5–3.5%; ≤TWD 150 → 3.5–5.0%. Scale in via GTC limits at 225 / 200 / 175; do not chase. Expected holding period 12–24 months.
- MLCC peer ranking for new capital at today's prices: PDC > Yageo > Murata > Taiyo Yuden > Walsin > Samsung Electro-Mechanics > Fenghua. On a Walsin pullback to TWD 200–225: PDC > Yageo > Walsin ≈ Murata > Taiyo Yuden > Samsung E-M > Fenghua.
- Next earnings: Q1 2026 reporting expected ~May 2026 (Taiwan cadence).
Sources
Fragments folded in (consolidated 2026-05-30)
2492-tw-profile.md— company profile (2026-05-20, mlcc-peer-comparison swarm)2492-tw-mgmt-dd.md— management due diligence (2026-05-20)2492-tw-buy-checklist.md— pre-buy checklist (2026-05-20, price TWD 262.0)2492-tw-deep-dive.md— investment deep-dive (2026-05-20)
All four fragments are about Walsin Technology Corporation (2492.TW); no wrong-entity stubs. Any *-filings.md placeholder in the folder was an empty stub and was not merged.
External sources cited across fragments
- Yahoo Finance — 2492.TW quote / profile / insider-transactions feed (yfinance pull 2026-05-20; analyst-estimate feed empty, targetMeanPrice/numberOfAnalystOpinions null)
- Walsin Technology IR (passivecomponent.com/investor-relations — TLS cert-fail from environment; board-members and quarterly-reports pages referenced only)
- Walsin Lihwa corporate site — subsidiaries/affiliates, founder (Ting Piao Chiao) pages
- Digitimes — MLCC demand 2025; Walsin/Kamaya/Matsuo tantalum stake (Feb 2026); passive-components demand splits 2026; PSA group / PDC ties (2005, 2007, 2009 archives)
- TrendForce — chip-resistor hikes (Walsin follows Yageo, up to 20% from Feb 1 2026); Murata April 1 price hike
- BigGo Finance — AI servers drive MLCC pricing polarization
- Passive Components EU — Walsin non-China lines at full capacity; MLCC price-increase coverage
- Simply Wall St — Walsin management page (director stakes, ISS QualityScore 10)
- The Org — Yu-Heng Chiao cross-entity directorships; Bloomberg — Ming-Tsan Tseng profile (Apaq/Inpaq); CommonWealth Magazine — Walsin Lihwa / Chiao family history
- MarketScreener — shareholders (403); compensation-committee changes Jun 17 2025 (cert-fail, headline only)
- Mordor Intelligence — automotive MLCC market
- ISS — Governance QualityScore (10) and pillar breakdown
- TWSE MOPS (mops.twse.com.tw, Traditional Chinese) — source-of-truth for ownership, segment mix, related-party (PDC powder) pricing, executive comp, insider-transaction tape; deferred for primary verification
- Vault peers: 2327-tw-profile (Yageo), 6173-two-profile (PDC/PSA), 6976-t-profile (Taiyo Yuden), 6981-t-profile (Murata)
Consolidation queue (merged 2026-05-30)
The four research fragments below were folded into this canonical page on 2026-05-30. They remain live pending Pink's archive confirmation.
- [ ]
2492-tw-profile.md - [ ]
2492-tw-mgmt-dd.md - [ ]
2492-tw-buy-checklist.md - [ ]
2492-tw-deep-dive.md
Source updates (auto-maintained)
Intake (May 23, 26) - mops-diligence-2026-05-23
Mandarin MOPS diligence (2026-05-23) confirms the PSA-group alliance is structural and equity-linked (Walsin holds a stake in PDC), validates the ISS QualityScore 10 governance discount via confirmed Chiao-family cross-board control across Walsin Lihwa, Winbond, and Nuvoton, and flags that English-language disclosure is materially thinner than US norms — exact RPT pricing, stake percentages, and Walsin Lihwa's stake in Walsin Tech remain Chinese-only gaps requiring TEJ or Mandarin analyst work before sizing above satellite tier.
Relevant to your thesis: Reinforces the governance discount (10–20%) and the satellite sizing cap (max 3–4%) already embedded in the bear case, while confirming the PSA/PDC supply-chain moat framing is directionally correct but not yet verifiable as cost-advantaged.
Source: intakefile://mops-diligence-2026-05-23.md
Drop/Bottleneck (May 24, 26) - 东莞证券_MLCC行业深度报告：供需矛盾加剧，高阶MLCC价格有望上扬_260325 (1)
Dongguan Securities' March 2026 MLCC industry report shows Walsin's Q1 delivery times and prices trending up alongside peers (Figure 34), with Murata/Samsung capacity-constrained on high-end SKUs and domestic players like Fenghua and Sanhuan — not Walsin — named as primary domestic-substitution beneficiaries.
Relevant to your thesis: Confirms the fast-follower positioning: Walsin appears in price/lead-time trend charts but is absent from the named beneficiaries list, consistent with the wiki's finding that Walsin is negotiating case-by-case rather than driving hikes.
Source: dropfile://Bottleneck/MLCC/东莞证券_MLCC行业深度报告：供需矛盾加剧，高阶MLCC价格有望上扬_260325 (1).pdf
Drop/Bottleneck (May 21, 26) - global_passives_basket_comparison
Walsin scores 16/30 (Pass) in a seven-name passives basket dated 21 May 2026, dragged by Crowding (1) and Valuation (2) after trading limit-up; consensus price targets (~TWD 600) were flagged as stale relative to prior reference price of TWD 147.
Relevant to your thesis: The scoring directly corroborates the wiki's "fully priced" verdict — multiple expansion outpacing revisions is exactly the valuation/crowding risk already flagged in the thesis.
Source: dropfile://Bottleneck/MLCC/global_passives_basket_comparison.pdf