← back to picks
HQ Taoyuan · Taiwan Reporting TWD Credit synth Aaa/AAA · net cash Valuation 2026-06-01 Damodaran FCFF · 1000-iter MC · three-regime FCFF · Dark v3

Quality lab at a modest discount once the rate is normalized · blend +19.3% / high bookend −4.4% margin of safety

Cheap at the 6.86% blended WACC, fair at the high bookend

At the base-blend TWD risk-free of 2.65% (WACC 6.86%) intrinsic is NT$278/share (after a 5% governance haircut; pre-haircut NT$293) vs market NT$233 — a +19.3% margin of safety, with the Monte Carlo p5 (NT$234) sitting just above today's price and P(intrinsic<market)≈5%. Show all three regimes: the low bookend (rf 1.13%, WACC 5.34%) flatters it to NT$348 (+49.5%), the high bookend (rf 4.30%, WACC 8.51%) cuts it to NT$223 (−4.4%, roughly fair). About 63% of value is the terminal block; Sporton trades at ~21.6× P/E on a record NT$12 dividend (~5.2% yield). Cheap at the blend but not at the high bookend — a margin-of-safety BUY to size moderately, contingent on Taiwan’s structurally-low rate persisting and the FCC catalyst printing.

p5 NT$234 p25 NT$256 p50 NT$273 p75 NT$290 p95 NT$316 low bookend NT$348 MARKET NT$233 DCF NT$278 (blend) high NT$223
SectorEMC / wireless test & certificationCountry mixTW 75% · CN 12% · US 6%β / MC σ0.85 levered · ±NT$25/sh (1000 runs)GovernanceFounder Chairman+GM · 5% haircutQuality~30% EBIT margin · net cash · synth AAAIncomeNT$12 dividend · ~5.2% yield · >100% payout
Intrinsic / share
NT$277.96
blend NT$278 · low NT$348 · high NT$223 · floor ~NT$109
Market / share
NT$233.00
2026-05-29 close · TPEx · P/E ~21.6×
Margin of safety
+19.3%
vs intrinsic
Enterprise value
NT$26.25B
62.9% terminal
Cost of equity / debt
6.86% / 2.44%
β 0.85 · CRP 0.72%
Terminal ROIC / g
13.00% / 2.00%
spread ~614bp (ROIC 13.00% vs WACC 6.86%)

What it sells, where it sells

Operating segments

NT$4.5B FY25 revenue
Testing & certificationEMC / RF / safety lab hours + accreditation grants; sole Taiwanese-owned FCC TCB, world's largest 5G NR lab — mandatory, recurring, accreditation-gated~90%
Components / magnetic-parts tradingDistribution book run out of the same Taiwan footprint; lower-margin, folded into the franchise — not split for the model~10%

This is effectively a single-segment recurring-services franchise: ~90% is mandatory EMC/wireless/safety testing and certification billed by the lab hour, so the FY2025 ~30% EBIT margin (down from a ~33% 2022 peak) is a post-5G volume digestion off a fixed chamber base, not a structural margin break — Q4-2025 gross margin had already recovered to ~48%.

Country mix (revenue-weighted CRP input)

🇹🇼Taiwan75%
🇨🇳China12%
🇺🇸United States6%
🇩🇪Germany5%
🇰🇷South Korea2%

The mix is weighted by where the lab hours are billed, not where the end-customer sits — Taiwan ~75% (the rising, higher-margin engine) and declining China ~12% — so the country-risk premium is a Taiwan-anchored ~0.72%. End-customer demand is global (US/EU/Korea/Japan brand & chip makers), but that drives growth cyclicality, handled in the story, not the discount rate.

Background — five things to know before the case

Context the share price doesn't carry on its face. Skim this once and the bull / bear bullets below stop reading as inside baseball.

Where we diverge from the sector

Each row asks: if we'd used the sector median for this one assumption instead of our override, how much would the share price change? Negative = our override is more conservative (worth less); positive = more aggressive (worth more). Skim down the rightmost column to see which bets are doing the work.

Assumption
Our input
Sector / engine default
Value impact
01Revenue growthY1-5 CAGR vs TIC-market growth
~6% Y1-5Market (~4.6%) + dated FCC share-shift; NOT the 2022 boom (+20%) nor the 2023-25 bust (−4 to −6%); fading to ~2.0% terminal
~−5%Sell-side consensus extrapolates the down-slope
large +
02Operating marginYear-10 target vs FY2025 trough
32%Audited FY2021/FY2024 mid-cycle normalization off the 29.9% FY2025 trough; top of the TIC peer band — not the 2022 peak
29.9%FY2025 trough held flat
mod +
03Sales-to-capitalReinvestment efficiency on the chamber base
1.6×Through-cycle audited asset-turn (~4.5bn rev / ~2.6bn operating capital); capex flat through digestion, no step-change build
~1.5×Business-services sector standard
small +
04Terminal ROCFaded steady-state return vs engine auto-resolve
13%Durable accreditation moat (sole FCC TCB, standards-body leadership, world's largest 5G lab) — above the 0.10 industrial-Taiwan fade, far below the engine default
~41%Engine auto-resolve (32% × 1.6 × 0.80) — an indefensible perpetual excess return
large −
05Cost of capitalBase-blend WACC vs the three rate regimes
6.86%β 0.85 (sector-aware, anti-suppression over the 0.43 regression) · CRP 0.72% · TWD risk-free 2.65% blend · net cash → WACC ≈ Ke
5.34% / 8.51%Low bookend (rf 1.13%) → 5.34% · high bookend (rf 4.30%) → 8.51% — the verdict follows the high bookend
−NT$55
Net effect — and the load-bearing input
At the base-blend 2.65% TWD risk-free (WACC 6.86%) the growth and margin overrides lift intrinsic to NT$278 (+19.3% vs NT$233), but ~63% of that is the terminal block and the single load-bearing input is the rate regime. Drop to the low bookend (rf 1.13%, WACC 5.34%) and intrinsic flatters to NT$348 (+49.5%); re-strike at the high bookend (rf 4.30%, WACC 8.51%) and it falls to NT$223 (−4.4%). Cheap at the blend, fair at the high bookend — the verdict follows the high bookend for downside discipline.
−NT$55
Our base-blend input Bookend / sector reference Adds value Subtracts value

The story & the five claims

The 10-year story this DCF is built on, plus the five anchor claims that translate the story into engine inputs.

The 10-year story

Sporton is Taiwan's largest EMC and wireless testing-and-certification lab, the only Taiwanese-owned holder of US FCC certification-body authority, and the world's largest 5G testing lab — a net-cash, recurring-services franchise that the market has soured on after three years of earnings digestion off its 2022 5G-boom peak. The central question is not whether the franchise is good (it is) but whether it is cheap. In the base case revenue grows about 6% a year through year five — market growth plus a dated FCC share-shift catalyst as labs in China and Hong Kong lose certification standing — then fades to roughly 2.0% (Taiwan's long-run inflation), while the operating margin normalizes back to a mid-cycle 32% from the 30% trough. Because the risk-free rate of a low-yield currency like the New Taiwan dollar is itself a modelling choice, we discount at a blended 2.65% rate (WACC 6.86%) and report the two bookends around it: that produces an intrinsic value of NT$278 a share against a NT$233 price, a modest +19.3% margin of safety. At the flattering low bookend (the suppressed 1.13% government-bond yield, WACC 5.34%) the same cash flows are worth NT$348; at the high bookend (a normalized 4.30% rate, WACC 8.51%) they are worth only NT$223, essentially today's price. The whole valuation is taken after a 5% haircut for the founder's combined chairman-and-general-manager control, and the verdict deliberately follows the high bookend, not the flattering low one.

Two debates worth pressure-testing

At the blended decision rate, is the margin of safety enough to act?
Our view: It is a normal call, not a robust one. At the base-blend 2.65% TWD risk-free / 6.86% WACC intrinsic is NT$278 — a +19.3% margin of safety — and with ~63% terminal weight a 100bp rate move still swings value materially. The robustness rule decides it: cheap at the base blend but NOT at the high bookend (rf 4.30% / WACC 8.51% → NT$223, −4.4%) is a margin-of-safety call to size accordingly, not a robust buy. A compound worst-corner (28% terminal margin × the high-bookend 8.51% WACC) lands ~NT$200 (−14%). Same rate-regime-sensitive pattern as SINBON.
How much downside protection is there if both the catalyst and the rate regime go against us?
Our view: A hard floor of net cash per share plus the NPV of ~10 years of dividends discounted at the high-bookend WACC is roughly NT$109/share — about 47% of today's price. So ~47% of the market cap is floor and ~53% is faith in the terminal block and the FCC catalyst. That is a meaningful but not deep cushion — and the >100% payout off a declining base means even the dividend leg of the floor is soft.
CLAIM 01Revenue compounds ~6% Y1-5 as the FCC share-shift ramps, fading to ~2.0%.growth_high: 6.0% · terminal: 2.0% · NT$4.5B → ~NT$7.0BMarket (~4.6% TIC) plus a dated FCC non-MRA share-shift to the sole TW FCC TCB; Q1-2026 +6.3% and April +9.6% corroborate. Deliberately NOT the 2022 boom (+20%) nor the consensus bust (−4 to −6%). Terminal growth tracks TWD long-run inflation, not the suppressed bond.
CLAIM 02Operating margin normalizes to a mid-cycle 32% by Year 5.target_op_margin: 32% from Y5Audited FY2021/FY2024 mid-cycle off the 29.9% FY2025 trough — ~2 points of operating leverage given back in digestion, recovered. Top of the TIC peer band, justified by sole-FCC-TCB status and >50% automation; NOT the 2022 peak.
CLAIM 03Reinvestment stays moderate on the existing chamber base.S2C: 1.6× Y1-5 and Y6-10Through-cycle audited asset-turn; across the boom revenue rose +67% while net PP&E rose only +24%. The next product cycle (Wi-Fi 7/8, physical-AI, 6G) re-uses and modestly extends existing chambers — no step-change build.
CLAIM 04Margin and growth converge by Year 5 (2030).year_of_convergence: 5The catalyst and new mandatory-testing legs are near-term (Q4-2026 into 2027), so the margin recovery and outsized growth are front-loaded; growth then fades linearly to the terminal rate. Sporton is a mature franchise, not a decade-long ramp.
CLAIM 05No failure risk; terminal ROC faded to a durable-moat 13%, terminal g 2.0%.terminal_g: 2.0% · ROC: 13% · failure: 0% · gov haircut: 5%Net cash, zero financial debt, zero goodwill, record cash return — failure prob 0%. Terminal growth set to TWD long-run inflation (~2.0%, ≤ the 2.65% rf ceiling), not the suppressed bond. Terminal ROC faded from the engine's indefensible ~41% to 13% (above the 0.10 industrial fade for the accreditation moat). 5% governance haircut for founder Chairman/GM key-man + succession.
Where we diverge from sell-side
  • We report three rate regimes and let the verdict follow the high bookend. Sell-side extrapolates a −4 to −6%/yr decline; our base case prices a ~6% catalyst-driven re-acceleration. We headline the blended decision rate (NT$278, +19.3%) and publish both bookends — low NT$348 (+49.5%), high NT$223 (−4.4%) — rather than leaning on the flattering low-rate number.
  • We treat the digestion as a trough, not structural decay. The audited FY2018–FY2024 series shows the floor landing at/above the pre-5G-boom plateau, and Q1-2026 (+6.3%) and Q4-2025 gross margin (~48%) corroborate the inflection — the opposite of the melting-ice-cube consensus.
  • β set to 0.85 by sector anti-suppression, not the 0.43 regression. The raw Taiwan small-cap regression β (~0.43) is suppressed by thin float and a ~30% controlling block, implying an incredible ~3.2% cost of equity. Pinning 0.85 (consistent with global TIC pure-plays and at-market realized vol) raises WACC honestly rather than flattering the valuation.
  • Terminal growth tracks inflation (2.0%), terminal ROC faded to 13%, governance haircut 5%. Terminal growth is pinned to TWD long-run inflation, not the suppressed bond yield; the engine's auto-resolved ~41% perpetual return is faded to a durable-moat 13%. Founder control with a clean minority record earns a light 5% haircut.
  • We name a hard floor and the floor% vs terminal-faith% split. Net cash + 10y dividend NPV at the high-bookend WACC ≈ NT$109/share (~47% of price); we publish that ~53% of the price is faith in the terminal block and the FCC catalyst, rather than burying it inside a single point estimate.
Two-sided case — bear anchors
  • The cheapness is rate-regime-sensitive. ~63% of base-blend value is the terminal block at a 6.86% TWD WACC; the +19.3% margin of safety holds at the blend but evaporates at the high bookend (rf 4.30%, WACC 8.51% → NT$223, −4.4%). It is cheap at the base but not robustly cheap — a margin-of-safety call to size, not a conviction buy.
  • The FCC catalyst is partly priced and partly still an NPRM. The stock is +29% in 3 months; the PAG-review change is in force, but the outright non-MRA lab ban is a proposed rule (final ~Q4-2026) that can slip or soften. You are paying post-news for a catalyst that hasn't fully printed.
  • 32% margin off a 29.9% trough may be peak-chasing. The bear reads the record margins as a mid-cycle peak the market correctly prices; absorbing rushed displaced-competitor overflow at higher margins is an assumption on an assumption, and a public mandated surge pulls in competing accredited capacity.
  • The dividend floor is soft. A >100% payout funded from declining earnings is a cut waiting to happen — so even the net-cash-plus-dividend hard floor (~NT$109, ~47% of price) should not be over-leaned on.
  • Customer/capex concentration. Cert-lab revenue tracks a few OEMs' launch cadence, so FCC overflow may be a one-time pull-forward bulge rather than a plateau, and a physical lab needs chambers, accreditation and technicians to absorb it — capping how fast revenue converts.

Risks to thesis (tail, not bear case)

Rate normalizationHigh

~63% of base-blend value is the terminal block at a 6.86% TWD WACC. The +19.3% margin of safety at the blend turns to −4.4% (NT$223) at the high bookend (rf 4.30% / WACC 8.51%) — the single load-bearing input. The verdict follows the high bookend.

FCC catalyst slips or is already pricedMed

The lab-ban is still an NPRM (final ~Q4-2026) and the stock is +29% in 3 months. If the final rule slips or softens, the share-shift thesis weakens — monitor H2-2026 monthly-revenue acceleration and the Federal-Register date.

Margin recovery fails / peak-chasingMed

If 32% is a mid-cycle peak rather than a normalization off the 29.9% trough, the cheap multiple is fair. A 28% terminal margin × the high-bookend 8.51% WACC compound worst-corner lands ~NT$200 (−14%).

Dividend cut off a declining baseMed

The record NT$12 dividend is a >100% payout funded from falling earnings — a cut is plausible, which softens the net-cash-plus-dividend hard floor (~NT$109, ~47% of price).

Moat half-life / competitive responseLow

A public mandated surge pulls in accredited capacity — peers accrediting, China labs re-routing via Taiwan subsidiaries — which can compete the 32-38% margin away over time. Mitigated by the multi-year cost to replicate FCC TCB authority.

Founder key-man / successionLow

Huang Wen-Liang (~30%) is both Chairman and General Manager (26.7 yrs, no visible succession plan, thin independent-director majority). Clean toward minorities — no extraction, record cash return — so a long-tail risk already priced via the 5% haircut.

10-year forecast

Revenue NT$4.76B → NT$7.00B over 10y (~6% Y1-5 fading to ~2.0% TWD long-run inflation); operating margin normalizes to 32% from Y5 onward off the 29.9% FY2025 trough — a post-5G mid-cycle recovery, not the 2022 peak.

Y1 Y2 Y3 Y4 Y5 Y6 Y7 Y8 Y9 Y10 0 7.0B rev (TWD) 0% 35% op margin revenue FCFF op margin

Monte Carlo distribution

At the base-blend 6.86% WACC even the 5th-percentile outcome (NT$234/sh) sits just above today's NT$233 price and only ~5% of 1,000 correlated draws land below the market — a thin but positive cushion. That distribution is the un-anchored perpetual-DCF spread at the blended rate; the honest downside lives in the high-bookend regime (NT$223, −4.4%) and the ~NT$109 hard floor, not in this band.

p5 p25 p50 p75 p95 market 233.00 200.1 272.9 346.2 freq equity / share (TWD)

Mean NT$273.67 ± NT$24.65/sh, 1000 iterations (0 failed). P(intrinsic < market NT$233.00) = 4.6%.

Cost of capital build
Risk-free rate 2.65%
Mature-market ERP 4.23%
Levered β 0.85
Weighted CRP 0.72%
Cost of equity 6.86%
Pre-tax cost of debt (synth Aaa/AAA) 3.05%
D / V ~0%
WACC 6.86%
Full year-by-year DCF
Year Revenue Op mgn EBIT EBIT(1−t) Reinvest FCFF PV
1 NT$4.76B 30.30% NT$1.44B NT$1.21B NT$169M NT$1.04B NT$974M
2 NT$5.05B 30.73% NT$1.55B NT$1.30B NT$179M NT$1.12B NT$982M
3 NT$5.35B 31.15% NT$1.67B NT$1.40B NT$189M NT$1.21B NT$989M
4 NT$5.67B 31.58% NT$1.79B NT$1.50B NT$201M NT$1.30B NT$997M
5 NT$6.02B 32.00% NT$1.92B NT$1.61B NT$213M NT$1.40B NT$1.00B
6 NT$6.32B 32.00% NT$2.02B NT$1.68B NT$189M NT$1.49B NT$1.00B
7 NT$6.57B 32.00% NT$2.10B NT$1.73B NT$160M NT$1.57B NT$987M
8 NT$6.78B 32.00% NT$2.17B NT$1.77B NT$126M NT$1.64B NT$965M
9 NT$6.92B 32.00% NT$2.21B NT$1.79B NT$89M NT$1.70B NT$936M
10 NT$7.00B 32.00% NT$2.24B NT$1.79B NT$49M NT$1.74B NT$898M
Methodology & flags

Damodaran FCFF DCF, 10y explicit + perpetuity, in TWD. Risk-free 2.65% (local-currency government bond). Synthetic credit Aaa/AAA. CRP 0.72% from revenue-weighted country mix × Damodaran 2026 CRPs. Terminal ROIC faded to 13.00%; 5% governance haircut applied post-DCF. Monte Carlo: 1000 iterations. Engine v1.0.0 · result: /Users/valentin/Documents/notes_jiliac_labs/Finance/Damodaran/valuations/sporton/output/2026-06-01-result.json

  • R&D cap: OFF · Lease cap: OFF · Failure: OFF · ESO: OFF
  • Governance haircut: 5% applied post-DCF (NT$292.58 > NT$277.96)
  • Sensitivity tornado: not run