← back to picks

HQ Taipei (Neihu) · Taiwan Reporting TWD Credit synth Aaa/AAA · net cash Founded 1990 · IPO 2003 Valuation 2026-05-31 Damodaran FCFF · 1000-iter MC FCFF · Dark v3

At the methodology-correct TWD rate, a high-ROIC distributor sitting ≈ on fair value (+1%)

Market ≈ DCF at the 2.65% blended rate — the +34% was the suppressed bond rate

Base intrinsic value NT$456/share (after a 13% governance haircut; pre-haircut NT$524) vs market NT$453 — essentially fair value at the methodology-blended TWD risk-free of 2.65%. The eye-catching +34% only appears at Taiwan's structurally-suppressed 1.13% government-bond rate (low bookend NT$605); let the rate normalise toward ~4.3% and intrinsic falls to NT$337 (−26%). The Monte Carlo straddles the price: P(intrinsic < market) = 47%, a coin flip. Genuinely high-quality — ROIC ~20%, net cash, 29 straight years of dividends — but no discount to intrinsic to capture today.

SectorRetail (Distributors) — semiconductor materialsCountry mixTW 46% · CN 46% · US 5% · JP 3%β / MC σ0.85 pinned · ±NT$54/sh (1000 runs)GovernanceKuo-family control · 13% haircutQualityROIC ~20% · net cash · 29 yrs of dividendsRate regimebase 2.65% blend · low 1.13% · high 4.3%

Intrinsic / share

NT$455.83

post 13% gov · pre NT$523.94

Market / share

NT$453.00

May-29-2026 close · TWSE (volatile: ranged ~NT$394–453 in late May)

Margin of safety

+0.6%

vs intrinsic

Enterprise value

NT$97.81B

81.7% terminal

Cost of equity / debt

6.94% / 2.44%

β 0.85 · CRP 0.82%

Terminal ROIC / g

12.00% / 2.00%

spread ~524bp (ROIC 12.00% vs WACC 6.76%)

What it sells, where it sells

Operating segments

Semiconductor & electronic materialsSilicon wafers, photoresist, quartz, specialty chemicals/gases, parts & equipment — channel to TSMC and the fab ecosystem; Shin-Etsu/Fujimi franchises84.6%

Environmental engineeringWastewater / cleanroom / green-energy turnkey plant work; backlog >NT$10B, targeting ~20%/yr growth9.0%

Optoelectronics, green energy & otherLED/LCD materials, solar, plus a small Taiwan-domestic health/aquaculture arm (Anyong)~6.4%

Topco is not a chipmaker — it is the leading Taiwan-based distribution + engineering channel for the semiconductor supply chain. ~85% of revenue is materials (silicon wafers, photoresist, quartz, chemicals) resold and integrated for TSMC and the broader Taiwan/China fab ecosystem; the rest is environmental-engineering turnkey work plus a small optoelectronics/green-energy/health tail. The financial signature follows from that: thin (~6.6%) operating margin, very high asset turnover, ~20% ROIC. Apply distributor economics, not chipmaker economics.

Country mix (revenue-weighted CRP input)

🇹🇼Taiwan46%

🇨🇳China46%

🇺🇸United States5%

🇯🇵Japan3%

Weighted by destination revenue. The non-obvious finding: China is co-equal with Taiwan (FY2024 disclosed China 46.5% / Taiwan 46.0%), because Topco's mainland distribution of Shin-Etsu consumables (~60% share of that channel) is a genuine separately-served end market, not re-exported Taiwan demand. Taiwan (0.78% CRP) and China (0.91% CRP) carry near-identical low country risk, so the blended CRP lands at just ~0.82% — and the ~8% US/Japan tail barely moves it.

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.

At the methodology-correct rate, Topco is fair value — not a bargain. The decision rate is the TWD risk-free built per the canonical low-yield-currency method: 35% local bond (1.13%) + 65% inflation-parity (3.49%) = 2.65%. At that rate intrinsic is NT$456 vs a NT$453 price — essentially on top of the market. The +34% headline only exists at the raw 1.13% Taiwan government-bond yield, which capital controls and a domestic savings glut hold structurally below the currency's fair long-run hurdle.
The rate is the whole valuation — three regimes, one decision. Low bookend (1.13% bond): NT$605, +34%. Base (2.65% blend): NT$456, +1%. High bookend (4.3% normalized): NT$337, −26%. A robust buy is cheap at the base and the high bookend; Topco is cheap at neither, so the margin of safety is a bet on Taiwan's rate suppression persisting, not on mispricing.
The Monte Carlo is a coin flip. 1,000 correlated draws around the 2.65% base spread p5 NT$378 / p50 NT$457 / p95 NT$555, with P(intrinsic < market) = 47%. That near-symmetry around the price is the quantitative signature of fair value — there is no asymmetric edge to capture at NT$453.
The business quality is genuine, which is why it is a HOLD and not a sell. ROIC 19.5–22.3% every year 2021–2025, ROE 20–25%, net cash, 29 unbroken years of dividends, record FY2025 (revenue +18.6%, EPS NT$21.81). For a "distributor" this is an unusually durable, capital-light compounder — low margin × very high working-capital turnover — riding the AI / advanced-process materials wave through the TSMC channel.
Founder-family control is the standing overhang. Despite a float-heavy register (~61% public), the Kuo family controls the largest director bloc via private vehicles (Gao Yi ~99.99% founder-owned; Jia Pin, the founder's wife's vehicle, holds three board seats), the founder's son is a director, board independence is ~1/3, and investigative press has documented related-party ties to the loss-making Anyong food/health group — a 13% governance haircut (NT$524 → NT$456).

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 the Street / Simply Wall St

12% Y1-5Above SWS ~10%, in line with the Street's FY26-27E; fades to 2.0% terminal. NT$67.6B → NT$159B

~10%Analyst consensus revenue CAGR

+NT$35

02Operating marginYear-10 target vs the distributor ceiling

7.0%A modest drift up from the stable 6.6% base on mix toward own-products / advanced packaging — NOT the 9% sector ceiling

~9%Damodaran Retail (Distributors) pretax EBIT margin

~NT$0

03Sales-to-capitalReinvestment efficiency vs sector median

2.5×Below Topco's net-cash-inflated ~3.9 realized, above the sector 1.83 — captures the capital-light working-capital model conservatively

1.83×Retail (Distributors) global median

+NT$25

04Terminal ROCPerpetual return vs today's ~20%

12.0%Faded from today's ~20% — distributor moats erode as principals can disintermediate; still a premium to the 6.76% base WACC

~13.4%Engine-resolved (7% margin × 2.5 S2C × 0.77)

−NT$12

05Risk-free rateMethodology blend vs the raw local bond

2.65%65/35 inflation-parity / local-bond blend → WACC 6.76%. The single load-bearing input; this sets fair value at NT$456

1.13%Raw Taiwan 10Y bond − spread → WACC 5.25%, the suppressed low bookend

−NT$149

Net effect of the discount-rate choice

The business story (rows 01-04) nets to roughly the base intrinsic; the swing input is the risk-free (row 05). Moving from the raw suppressed bond (1.13%, NT$605) to the methodology blend (2.65%, NT$456) erases the entire headline discount — and a further move to the normalized 4.3% takes it to NT$337. At the correct rate the stock is fair value (+1%), which is why this is a HOLD, not a buy.

+NT$3

Our input Alt / bookend 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

Topco Scientific is the leading Taiwan-based distribution and engineering channel for the semiconductor supply chain — silicon wafers, photoresist, quartz, specialty chemicals and gases resold and integrated for TSMC and the broader Taiwan/China fab ecosystem, plus an environmental-engineering arm. It is a genuinely high-quality compounder: ROIC has printed ~20% every year for five years, the balance sheet is net cash, and it has paid dividends for 29 consecutive years while riding the AI / advanced-process materials wave to record FY2025 revenue (+18.6%). The valuation question is almost entirely about one number: the TWD risk-free. At Taiwan's raw 1.13% government-bond yield the DCF says NT$605, a tempting +34%. But that bond yield is structurally suppressed by capital controls and a domestic savings glut, so it understates the currency's fair long-run equity hurdle. Built the canonical way — 35% local bond plus 65% inflation-parity — the rate is 2.65%, and at that rate intrinsic lands at NT$456, within one percent of the NT$453 price. The Monte Carlo confirms it: across 1,000 correlated draws the median is NT$457 and the probability that intrinsic sits below the market is 47%, a coin flip. Push the rate further toward a fully-normalized 4.3% and intrinsic falls to NT$337, a 26% loss. So the business is fine and the quality is real, but at today's price there is no discount to intrinsic to capture — the apparent bargain was the suppressed bond rate doing the work. That is a HOLD: a high-quality name to own on a pullback, not at fair value with downside if Taiwan's yield suppression unwinds.

Two debates worth pressure-testing

Which risk-free is right — the 1.13% local bond or the 2.65% blend?

Our view: The blend, per the canonical low-yield-currency method. Taiwan's bond yield is held below the currency's fair long-run hurdle by capital controls and a savings glut, so Method 1 (1.13%) imports that distortion straight into WACC and over-values everything — the whole Taiwan batch printed +45–250% on it. The inflation-parity build (3.49%) strips the distortion but leans on foreign anchors; blending 65/35 toward parity gives 2.65%. At that rate Topco is fair value, and the +34% reveals itself as a rate artifact.

If the quality is this good, why not own it here?

Our view: Because quality is already in the price. ROIC ~20%, net cash and a 29-year dividend record are real, but at NT$453 the DCF gives you no margin of safety (+1%) and the Monte Carlo is a 47% coin flip. You are paid fairly for a fine business, not cheaply for a mispriced one. Wait for a pullback toward the p25 (~NT$422) or below, where the rate-normalization downside is already partly in the price.

CLAIM 01Revenue compounds ~12% Y1-5, fading to 2.0% terminal.growth_high: 12% · terminal: 2.0% · NT$67.6B → NT$159BAbove Simply Wall St's ~10% and roughly in line with the Street's FY26-27E, on advanced-process consumables (EUV photoresist, wafers, quartz), advanced-packaging materials, overseas expansion and ~20%/yr environmental engineering — below the recent 18% spike to respect semi-capex cyclicality. Terminal growth tracks TWD long-run inflation.

CLAIM 02Operating margin drifts only to 7.0% — distributor economics hold.target_op_margin: 7.0% (from 6.6%)A modest mix-shift up toward higher-margin own-products / advanced packaging. We deliberately do NOT reach the Damodaran distributor-sector 9% pretax ceiling: the principal captures the fat materials margin, Topco earns the channel spread.

CLAIM 03Reinvestment at a conservative 2.5× sales-to-capital.S2C: 2.5× Y1-5 and Y6-10Below Topco's net-cash-inflated ~3.9 reported but above the 1.83 sector median — captures the genuinely capital-light, working-capital-efficient model (asset turnover 1.74) without straining it.

CLAIM 04Terminal ROC faded to 12% — not today's ~20%.override_roc: 12% · terminal_g: 2.0%A channel distributor cannot defend a 20% return forever — the Shin-Etsu agency is renegotiable and principals can disintermediate. 12% keeps a healthy premium to the 6.76% base WACC for switching costs while refusing a perpetual-moat assumption.

CLAIM 05Risk-free 2.65% — the blended TWD rate is the whole call.riskfree: 2.65% (35% bond / 65% parity) · gov haircut: 13%Net cash, synth Aaa/AAA, 29 straight years of dividends. The decision rests on the rate: at 2.65% intrinsic NT$456 ≈ price; at the raw 1.13% bond NT$605; at a normalized 4.3% NT$337. The governance haircut (Kuo-family control, related-party Anyong ties) trims NT$524 → NT$456.

Where we diverge from sell-side

Risk-free at the methodology blend (2.65%), not the raw local bond (1.13%). This is the entire difference between the +34% headline and fair value. Sell-side and momentum screens implicitly discount at the suppressed bond rate; we use the inflation-parity-blended rate, which strips Taiwan's structural yield distortion. Worth −NT$149/share vs the low bookend.
Year-10 margin capped at 7.0%, not the 9% distributor-sector ceiling. Topco earns a channel spread, not pricing power; the principal keeps the fat materials margin. We refuse the sector-aspirational margin.
Terminal ROC faded to 12%, not today's ~20%. A renegotiable Shin-Etsu agency cannot defend a 20% perpetual return; the fade costs ~NT$12.
Governance haircut 13%, well above the 5% public-company baseline. Founder-family control via private vehicles (3 board seats), weak board independence and related-party Anyong exposure — real minority-agency overhang, partly offset by clean operating capital allocation and net cash.
We report all three rate regimes, not one headline. Low 1.13% (NT$605), base 2.65% (NT$456), high 4.3% (NT$337). Cheap only at the low bookend = rate-regime bet; fair at the base and exposed at the high bookend = a HOLD, sized accordingly.

Two-sided case — bear anchors

Taiwan's yield suppression unwinds. If the risk-free normalizes toward ~4.3%, WACC is ~8.4% not 6.76%, intrinsic is NT$337, and the stock is ~26% overvalued at NT$453 — with 82% of value in the rate-sensitive terminal block.
No margin of safety at the base rate. Even at the methodology-correct 2.65% the DCF is +1% and the Monte Carlo is a 47% coin flip. You are paying fair value for quality, with no discount to absorb a disappointment.
The Shin-Etsu agency gets disintermediated. The fat materials margin sits with the principal; Topco earns a spread the supplier could take back or compress. The whole ~20% ROIC, thin-margin model rests on the durability of distribution rights that are, in principle, renegotiable.
AI / advanced-process cyclicality at a record-high base. FY2025 and Q1-2026 are all-time highs; a wafer-fab capex pause hits the entire thin-margin P&L, and a market that has walked targets up to NT$480 leaves little room for an in-line quarter.
Founder-family control + non-core sprawl. The Kuo family runs the controlling director bloc through private vehicles; the group bleeds attention and (at the private level) cash into loss-making food/health/sports ventures, and the 2026 AGM lifted directors' non-compete restrictions — worth monitoring.

Risks to thesis (tail, not bear case)

Rate normalization (yield-suppression unwind)High

The single load-bearing input. The base uses a 2.65% blended TWD risk-free; if it normalizes toward ~4.3%, intrinsic falls from NT$456 to NT$337 (−26%), because 82% of value is terminal. The whole call is the rate.

No margin of safety at fair valueHigh

At the correct 2.65% rate the DCF is +1% and the Monte Carlo straddles the price (P(intrinsic<market)=47%). There is no discount to intrinsic to capture today — you are paid fairly, not cheaply.

Shin-Etsu agency disintermediationHigh

The principal captures the fat materials margin; Topco earns a renegotiable channel spread. If the agency is compressed or taken in-house, the ~20% ROIC, thin-margin model erodes — the moat protects share, not pricing.

Terminal-value concentrationMed

82% of EV is the perpetuity even at the higher 2.65% rate; the explicit 10-year FCFF is only ~18% and net cash adds just ~NT$15/share. The valuation is almost entirely an article of faith about the steady state and its discount rate.

Semi-capex cyclicality at a record baseMed

FY2025/Q1-2026 are all-time highs; a wafer-fab capex down-leg hits the whole thin-margin P&L, and a market that has walked targets to NT$480 leaves little room for an in-line quarter.

Kuo-family control + Anyong related partiesLow

Founder-family controls the director bloc via private vehicles (3 board seats), board is ~1/3 independent, and press has documented related-party ties to the loss-making Anyong food/health group. Priced via the 13% governance haircut; a step-up is a tail.

10-year forecast

Revenue NT$67.6B → NT$159B over 10y (12% Y1-5 on advanced-process consumables + advanced packaging + overseas + ~20%/yr environmental engineering, fading to 2.0% TWD long-run inflation). Operating margin drifts modestly from 6.6% to 7.0% by Y5 and holds — distributor economics cap the upside well below the 9% sector ceiling.

Monte Carlo distribution

The Monte Carlo straddles the market: 1,000 correlated draws around the 2.65% base give p5 NT$378 / p50 NT$457 / p95 NT$555, with P(intrinsic < market) = 47% — a coin flip. That near-symmetry around the NT$453 price is the quantitative signature of fair value. (The distribution samples around the base rate; the low-rate NT$605 and normalized NT$337 bookends sit outside it and frame the rate risk.)

Mean NT$460.28 ± NT$54.43/sh, 1000 iterations (0 failed). P(intrinsic < market NT$453.00) = 47.3%.

Cost of capital build

Risk-free rate	2.65%
Mature-market ERP	4.23%
Levered β	0.85
Weighted CRP	0.82%
Cost of equity	6.94%
Pre-tax cost of debt (synth Aaa/AAA)	3.05%
D / V	~4%
WACC	6.76%

Full year-by-year DCF

Year	Revenue	Op mgn	EBIT	EBIT(1−t)	Reinvest	FCFF	PV
1	NT$75.69B	6.68%	NT$5.06B	NT$3.88B	NT$3.24B	NT$635M	NT$595M
2	NT$84.78B	6.76%	NT$5.73B	NT$4.40B	NT$3.63B	NT$764M	NT$670M
3	NT$94.95B	6.84%	NT$6.49B	NT$4.98B	NT$4.07B	NT$914M	NT$751M
4	NT$106.34B	6.92%	NT$7.36B	NT$5.65B	NT$4.56B	NT$1.09B	NT$838M
5	NT$119.11B	7.00%	NT$8.34B	NT$6.40B	NT$5.10B	NT$1.29B	NT$932M
6	NT$131.02B	7.00%	NT$9.17B	NT$7.10B	NT$4.76B	NT$2.33B	NT$1.58B
7	NT$141.50B	7.00%	NT$9.90B	NT$7.73B	NT$4.19B	NT$3.54B	NT$2.24B
8	NT$149.99B	7.00%	NT$10.50B	NT$8.26B	NT$3.40B	NT$4.87B	NT$2.88B
9	NT$155.99B	7.00%	NT$10.92B	NT$8.66B	NT$2.40B	NT$6.26B	NT$3.47B
10	NT$159.11B	7.00%	NT$11.14B	NT$8.91B	NT$1.25B	NT$7.66B	NT$3.97B

Methodology & flags

Damodaran FCFF DCF, 10y explicit + perpetuity, in TWD. Risk-free 2.65% (local-currency government bond). Synthetic credit Aaa/AAA. CRP 0.82% from revenue-weighted country mix × Damodaran 2026 CRPs. Terminal ROIC faded to 12.00%; 13% governance haircut applied post-DCF. Monte Carlo: 1000 iterations. Engine v1.0.0 · result: /Users/valentin/Documents/notes_jiliac_labs/Finance/Damodaran/valuations/topco/output/2026-05-31-result.json

R&D cap: OFF · Lease cap: OFF · Failure: OFF · ESO: OFF
Governance haircut: 13% applied post-DCF (NT$523.94 > NT$455.83)
Sensitivity tornado: not run

FCFF DCF · dark theme v3. Generated 2026-06-01T07:59:32 from /Users/valentin/Documents/notes_jiliac_labs/Finance/Damodaran/valuations/topco/output/2026-05-31-result.json.