← back to picks

Founded 1980 · Taiwan · ~1,120 employees Reporting TWD Credit net cash · synth Aaa/AAA Valuation 2026-06-02 FCFF · 1000-iter MC · 5-axis FCFF · Dark v3

Cheap only on a suppressed discount rate — at a normalized TWD rate, −3% margin of safety

Rate-mirage SELL · slightly rich once the discount rate normalizes

Published on the normalized-rate stress regime (TWD risk-free 4.3%, WACC 7.96%): intrinsic NT$116.49/share (post-15% governance haircut; pre-haircut NT$137.05) vs market NT$119.50 — a ~3% shortfall, and Monte Carlo on this regime puts P(intrinsic < market) = 79%. The +31% “bargain” at the suppressed 50/50-blend rate is a discount-rate mirage on a governance-trapped, six-years-shrinking CDMO; NT$157 is the bull bookend, not the decision number. Avoid / sell into strength.

SectorSpecialty CDMO · supplements / beautyCountry mixUS 31% · China 23% · NL 13%β / MC σ0.90 levered · ±NT$10/sh (1000 runs)Governance~99%-pledged holdco · 15% haircutQualityROIC ~14% · net cash · synth Aaa/AAAVerdictStress-regime SELL · avoid

Intrinsic / share

NT$116.49

post 15% gov · pre NT$137.05

Market / share

NT$119.50

2026-06-01 close · TPEx

Margin of safety

-2.5%

vs intrinsic

Enterprise value

NT$14.60B

54.8% terminal

Cost of equity / debt

8.96% / 3.76%

β 0.90 · CRP 0.95%

Terminal ROIC / g

14.00% / 2.00%

spread ~604bp (ROIC 14.00% vs WACC 7.96%)

What it sells, where it sells

Operating segments

North AmericaNow #1 region — re-based up from 26% (2024) as MNC brand clients replaced China volume~31%

ChinaDeliberately wound down 73% (2020) → 23% on regulatory / livestream-channel risk — the lost volume~23%

Asia-Pacific ex-ChinaTaiwan home (~9%) + India + SE-Asia city-factory build-out~31%

EuropeNetherlands base, Barcelona plant ~2027 — stable ~13%~13%

One CDMO segment, split by geography: the entire thesis is whether the US (31%) + Europe (13%) + Asia-Pacific gains durably replace a China book (23%) that management is still actively shrinking — a mix-shift, not a scale-growth, story.

Country mix (revenue-weighted CRP input)

🇺🇸United States31%

🇨🇳China23%

🇳🇱Netherlands13%

🇹🇼Taiwan9%

🇮🇳India7%

🇮🇩Indonesia5%

🇲🇾Malaysia4%

🇻🇳Vietnam4%

🇹🇭Thailand2%

US + Netherlands + Taiwan ≈ 53% of revenue sit at near-zero country risk; the blended CRP (~0.95%) is carried by China (23%) and the small SE-Asia / India tail — so the geo mix is a margin / FX-translation story, not a sovereign-risk one.

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.

This is a post-peak business, not a growth stock. Revenue peaked at NT$9.57B in 2019 (~80% China) and has fallen or stalled for six straight years to a ~NT$7.0B base (NT$6.99B FY2025, −3.5% YoY) after China's 2020+ micro-commerce / livestream crackdown. The screener headline — “earnings declined ~17–23%/yr over 5 years” — is measured off the 2021 EPS peak of NT$13.17.
Profit is at a recent high on lower sales — the value driver is margin, not volume. FY2025 net income NT$1.108B (+13.8%), EPS NT$8.68 (+14%), full-year gross margin 44.1% (Q4 46.7%), operating margin recovered 13.2% (2023) → 16.6% (2025) on AI/automation, expense discipline and a higher-value mix (PDRN, GLP-1 support, longevity, functional jelly).
The controlling holdco has ~99% of its TCI shares pledged — the load-bearing red flag. 詠江投資 (Yong Jiang) pledged 6,738 of 6,789k shares, and the board-wide pledge ratio is rising (39% Sep-25 → 44% Jan-26). The Chairman and General Manager are the same person (Lin Yung-Hsiang), an independent director resigned (Dec-2024), and there was a 2021 insider-trading probe.
The dividend is not covered by cash earnings. FY24 dividend NT$10/sh ≈ 117–131% payout; SimplyWallSt flags a “cash payout ratio 321%.” With ~NT$2.27B of cash FX-trapped in China and a heavily-pledged founder who needs the payout to service margin loans, the ~6% yield is partly funded off the balance sheet — a potential forced drain, not a clean return.
Net cash and a genuine buyback-and-cancellation pass the value-trap test. ~NT$5.5B cash, total-liability ratio ~38%, no material financial debt; share count fell 118M → 110M over five years including a 532k-share treasury cancellation (May 2025). That real cash return reaching minorities is why this is a contingent watchlist, not an outright Pass.

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 median

Value impact

01Revenue growth10-year revenue CAGR vs sell-side consensus

~3% Y1-5fading to 2% terminal — six years of flat-to-down evidence; a decade to merely re-touch the 2019 peak

~8%serially-cut analyst consensus; management's optimistic +14%/NT$8bn ambition

−NT$18

02Operating marginYear-10 normalized margin vs sector median EBIT margin

17.0%above the 14% trough / 16.6% FY2025, below the un-repeatable 20% China-scale peak — the swing driver

~12%Healthcare-Products contract-manufacture median operating margin

+NT$34

03Sales-to-capitalReinvestment efficiency vs sector median

1.4×asset-light City Factory harvest model; PP&E grew only 4.84 → 5.70bn over 4y

~1.2×Healthcare-Products industry standard

+NT$3

04Terminal growthYear 10+ steady state vs TWD risk-free ceiling

2.80%TWD long-run inflation (upper), held ≤ the 4.30% normalized risk-free ceiling

4.30%normalized TWD risk-free (suppression-unwind regime)

−NT$3

05Cost of capital10y WACC vs WACC implied by the sector-median β

7.96%β 0.90 (council-hardened from 0.75) · CRP 0.95% · TWD rf 4.30% (normalized) · net cash

~8.3%Healthcare-Products global β_u 1.12 re-levered → β_lev 1.33; the 0.38 5Y regression is float-suppressed

+NT$3

Net effect of overrides

At the normalized stress rate the WACC override no longer adds value (7.96% ≈ the sector WACC), so the net edge over all-defaults collapses to ~+NT$19/share — resting almost entirely on the high 17% margin, offset by conservative ~3% growth. This IS the council's warning made visible: the +31% base case needed the suppressed TWD rate AND the 17% margin (the same bet made twice); remove the rate and the cushion is gone — stress intrinsic NT$116 sits below the NT$120 price.

+NT$19

Our override Sector median Override adds value Override 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 year-10 DCF inputs.

The 10-year story

TCI is a post-peak specialty contract manufacturer of supplements, functional beverages and beauty products whose top line was made by the 2015-19 China micro-commerce boom and un-made by China's 2020+ crackdown — six straight years of flat-to-down revenue into a ~NT$7.0B base, with the customer mix re-based toward diversified US, European and SE-Asian brand clients. This is not a growth story, and the base is not inflated. The value driver is margin durability, not a top-line re-acceleration: gross margin held 40-46% through the entire China collapse, and operating margin has climbed back from 13.2% to 16.6%. The base case assumes revenue compounds only ~3%/yr (well below the serially-cut ~8% consensus), reaching ~NT$9.1B by year 10 — a decade just to re-touch the 2019 peak — with a normalized 17% operating margin, mid-teens returns on capital, and 2% steady-state growth past year 10. Governance drag — a ~99%-pledged founder holdco, a Chairman who is also GM, and China-trapped cash — is modeled in the cash flows plus a 15% residual haircut, not as a punitive bolt-on.

Two debates worth pressure-testing

Is the +31% a real edge or a low-WACC artifact?

Our view: Largely rate-contingent, and the report says so out loud. Three of five council advisors independently flagged that (WACC − g) is only ~4 points, so a 200bp rate move erases most of the edge. At the stress TWD risk-free (WACC 7.96%) intrinsic is NT$116.49 (−3%); in the full bear ~NT$97 (−19%). The base is the optimistic, not the central, read.

Does a ~99%-pledged founder make this a value trap?

Our view: It is the single biggest risk, hence the governance haircut was hardened 0.12 → 0.15 and β 0.75 → 0.90 post-council. The pledge-dividend doom loop — a founder who needs an uncovered dividend (partly trapped in China) to service margin loans — means the “net cash pays you to wait” leg double-counts. It clears the trap test only on a genuine buyback-and-cancel reaching minorities.

CLAIM 01Revenue compounds only ~3% Y1-5, decaying to 2% terminal.growth_high: 3.0% · terminal: 2.0% · NT$7.0B → ~NT$9.1BWell below the serially-cut ~8% consensus. Six years of flat-to-down revenue is the evidence; a decade just to re-touch the 2019 peak.

CLAIM 02Operating margin normalizes to 17% — the swing value driver.target_op_margin: 17.0% by Y5Above the 14% trough and 16.6% FY2025, below the un-repeatable 20% (2021 China-scale) peak. Gross margin held 40-46% through the entire China collapse.

CLAIM 03Asset-light City Factory harvest model.sales_to_capital: 1.4× both windowsPP&E grew only NT$4.84 → 5.70bn over four years; net cash, FCF-positive, returning capital rather than ploughing it into organic growth.

CLAIM 04Margin converges by Y5; growth fades by Y10.year_of_convergence: 5The footprint build (France/Vietnam/India/Utah) is essentially complete and margin is already at 16.6% — a short convergence window is the honest treatment.

CLAIM 05Mid-teens terminal ROC, zero failure, 15% governance haircut.ROC: 14% · failure: 0% · gov haircut: 15%Net cash NT$3-3.8bn, harvest-mode returns (26% 2021 → ~14% now). The 0.15 residual is the methodology leakage ceiling for a ~99%-pledged controlled name.

Where we diverge from sell-side

Growth pinned at ~3%, not ~8%. Sell-side anchors on a consensus that has been cut every year through six straight down/flat years; we extrapolate the actually-observed pattern, not a China-economics restoration.
Margin durability is the whole thesis, capped at 17%. We refuse to credit the un-repeatable 20% China-scale peak; the bull case (operating-leverage coil → 20% margin + EPS NT$14-15) lives in the upside tail, not the base.
β hardened to 0.90 — WACC 6.05% at the base rate, 7.96% at the normalized rate we publish on. The 0.38 5Y regression is float-suppressed by the pledged holdco — laundering illiquidity that should raise the discount into a low β. We lifted β toward the controlled-microcap illiquidity premium rather than trust either pole.
Governance modeled in the flows, then 15% residual. Mid-teens ROC (not 20%+), a ~20% repatriation haircut on NT$2.27B China-trapped cash baked into the bridge, and effective tax normalized from an unsustainable 9.75% to 20% — before any bolt-on discount.
Three risk-free regimes reported, not one. Per the TWD low-rate-currency discipline the decision is rate-contingent — we publish on the stress (−3%) bookend — the base (+31%) and low (+55%) are the suppressed-rate bull bookends, and a full bear lands −19%. Headlining the base alone would have been the mistake the council caught.

Two-sided case — bear anchors

The edge is a discount-rate bet. (WACC − g) is only ~4 points; at a generous 7% WACC the terminal value deflates and the margin of safety vaporizes. The stress regime (WACC 7.96%) already prints NT$116.49 — below today's NT$119.50.
Margin reverts to the 14% trough. The whole bull case rests on a 17% margin holding on a business with no pricing power and MNC clients that can dual-source. Stress rate + 14% margin + 0.15 governance ≈ NT$97/share, a −19% loss.
The pledge-dividend doom loop fires. A price drop forces founder margin-call sales → more drop. The uncovered dividend (cash payout ~321%, partly China-trapped) is a forced drain, not a return — the “pays you to wait” leg is double-counted.
Revenue keeps melting. The Outsider's read: you can't cost-cut to growth, and the +3%/yr turnaround for a decade is a wish wearing a number's clothing. A flat ~NT$7B cash cow at trough margin is the genuine downside tail.
Net dividend after frictions is thin. The 6-8% gross yield faces 21% non-recoverable Taiwan withholding (≈4.7-6.3% net) before FX — the income leg the bull case leans on is ~⅓ smaller than it looks to a foreign minority.

Risks to thesis (tail, not bear case)

Founder margin-call / forced-sale overhang High

~99% of the controlling holdco's shares are pledged and the board pledge ratio is rising (39% → 44%). A price drop can trigger forced sales → reflexive further drops. Pledged shares are ~5.7% of total — bounded, but the fuse exists.

Rate-regime reversal High

The base intrinsic depends on a suppressed 2.31% TWD risk-free. A move to the 4.30% stress regime (WACC 7.96%) drops intrinsic to NT$116.49 — below today's price. This is the dominant non-operating risk.

Margin reversion Med

If the 17% operating margin reverts toward the 14% trough as MNC clients pressure pricing, intrinsic falls ~NT$30+/share. Margin durability, not growth, is the entire value case.

China-trapped cash & uncovered dividend Med

~NT$2.27B cash is FX-trapped in China and the dividend exceeds cash earnings. A 20% repatriation haircut is already in the bridge, but a tightening of capital controls or a dividend cut would re-rate the income leg.

Empire-building M&A Med

Serial capex-heavy expansion (Utah, Barcelona, a NT$700M long-term-care diversification) has compressed ROIC 26% → ~14%. A large non-core deal that doesn't earn its cost of capital is the way the in-flow governance drag worsens.

Tradeability & liquidity friction Low

TPEx trades on IBKR but needs FINI/Tax-ID + local custody; ~US$2M/day liquidity is fine for a small ticket via marketable limits. Verify a live limit order before sizing — if the ticket isn't live, the thesis is moot.

10-year forecast

Revenue NT$7.20B → NT$9.12B over 10y (~3% Y1-5 fading to 2% — a decade just to re-touch the 2019 peak); operating margin 16.6% → 17.0% held. The flows are fine — but discounted at a normalized 7.96% WACC they are worth only NT$116, below the NT$119.50 price. The value was never in the business; it was in the suppressed-rate discount.

Monte Carlo distribution

On the published normalized-rate regime (rf 4.3%, WACC 7.96%) the 5-axis Monte Carlo (1,000 runs) centres at NT$111 (p5 NT$96 / p95 NT$128) with P(intrinsic < market) = 79%. Once the suppressed TWD discount rate is removed, the operating story — 3% growth, 17% held margin — no longer carries a margin of safety. (At the base 50/50-blend rate the same model prints NT$157; the gap between the two IS the thesis risk the council flagged.)

Mean NT$111.14 ± NT$9.54/sh, 1000 iterations (0 failed). P(intrinsic < market NT$119.50) = 79.7%.

Three risk-free regimes & the council downgrade

This is a TWD low-rate-currency valuation, so the decision is rate-contingent. We publish on the normalized-rate stress regime (the conservative, defensible reading) — the +31% base case is the bull bookend that requires the suppressed TWD bond yield to persist. The LLM Council pressure-test downgraded TCI from a “robust BUY” to a contingent watchlist; β and the governance haircut were hardened in response (β 0.75 → 0.90, governance 0.12 → 0.15).

Regime	Risk-free	WACC	Intrinsic (post-gov)	vs market NT$119.50
Low bookend	1.13%	4.91%	NT$185.41	+55%
Base — 50/50 blend (bull bookend)	2.31%	6.05%	NT$157.05	+31%
Stress bookend — PUBLISHED decision (margin held 17%)	4.30%	7.96%	NT$116.49	−3%
Bear (stress rate + margin 14% + gov 0.15)	4.30%	7.33%	~NT$97	−19%

The two honest decision numbers are the stress (−3%) and the bear (−19%) — not the base alone. A real, sized-small speculative bet for someone who wants the margin-durability + cash-return story and can stomach the controlled-company tail; not the unconditional bargain the +31% implies. Highest-confidence council signal (3 of 5 independent): the margin of safety is heavily a function of the discount rate, and the 17% margin and the low WACC are “the same bet made twice.”

Cost of capital build

Risk-free rate	4.30%
Mature-market ERP	4.12%
Levered β	0.90
Weighted CRP	0.95%
Cost of equity	8.96%
Pre-tax cost of debt (synth Aaa/AAA)	4.70%
D / V	~19%
WACC	7.96%

Full year-by-year DCF

Year	Revenue	Op mgn	EBIT	EBIT(1−t)	Reinvest	FCFF	PV
1	NT$7.20B	16.64%	NT$1.20B	NT$1.02B	NT$150M	NT$869M	NT$805M
2	NT$7.41B	16.73%	NT$1.24B	NT$1.05B	NT$154M	NT$900M	NT$772M
3	NT$7.64B	16.82%	NT$1.28B	NT$1.09B	NT$159M	NT$933M	NT$742M
4	NT$7.87B	16.91%	NT$1.33B	NT$1.13B	NT$164M	NT$967M	NT$712M
5	NT$8.10B	17.00%	NT$1.38B	NT$1.17B	NT$169M	NT$1.00B	NT$683M
6	NT$8.33B	17.00%	NT$1.42B	NT$1.19B	NT$162M	NT$1.03B	NT$648M
7	NT$8.55B	17.00%	NT$1.45B	NT$1.21B	NT$155M	NT$1.05B	NT$613M
8	NT$8.75B	17.00%	NT$1.49B	NT$1.22B	NT$146M	NT$1.07B	NT$577M
9	NT$8.94B	17.00%	NT$1.52B	NT$1.23B	NT$138M	NT$1.09B	NT$542M
10	NT$9.12B	17.00%	NT$1.55B	NT$1.24B	NT$128M	NT$1.11B	NT$508M

Methodology & flags

Damodaran FCFF DCF, 10y explicit + perpetuity, in TWD. Risk-free 4.30% (local-currency government bond). Synthetic credit Aaa/AAA. CRP 0.95% from revenue-weighted country mix × Damodaran 2026 CRPs. Terminal ROIC faded to 14.00%; 15% governance haircut applied post-DCF. Monte Carlo: 1000 iterations. Engine v1.0.0 · result: /Users/valentin/Documents/notes_jiliac_labs/Finance/Damodaran/valuations/tci/output/2026-06-02-result.json

R&D cap: OFF · Lease cap: OFF · Failure: OFF · ESO: OFF
Governance haircut: 15% applied post-DCF (NT$137.05 > NT$116.49)
Sensitivity tornado: not run

FCFF DCF · dark theme v3. Generated 2026-06-02T07:22:25 from /Users/valentin/Documents/notes_jiliac_labs/Finance/Damodaran/valuations/tci/output/2026-06-02-result.json.