← back to picks

HQ Taichung · Taiwan Reporting TWD Credit synth Aaa/AAA · net cash Listed TPEx 1998 Valuation 2026-06-01 Damodaran FCFF · 1000-iter MC FCFF · Dark v3

Backlog at par, not a bargain · roughly fair after the rate re-rate — hard ~NT$50 floor, fair value ~NT$70–90

Net-cash, real backlog — but priced ~at fair value once future tenders are valued honestly

We do not underwrite the NT$192 perpetual-FCFF print (it was NT$251 before the risk-free was re-rated to the methodology 2.65% blend): ~73% of it is a residual perpetuity feeding on order replenishment that is decelerating, and future public-tender work is competitively bid — so its ROIC trends to WACC and its marginal NPV is ~zero. The one robust anchor is a hard floor of ~NT$50/share (contracted backlog run-off + net cash) — the NT$71.6 price sits ~1.4× it. Value the firm honestly (backlog at par + a thin franchise premium) and fair value is ~NT$70–90: roughly fair, a WATCH, not the +47-56% BUY the capped band advertised.

SectorEngineering / ConstructionCountry mixTaiwan 100%β / MC σ0.80 levered · ±NT$28/sh (1000 runs)GovernanceFamily-run · succession risk · 10% haircutQuality~17% ROE · net cash · synth AAAIncomeNT$3.6 cash dividend · ~6% yield

Intrinsic / share

NT$80.00

fair-value band NT$50–90 · roughly fair

Market / share

NT$71.60

1 Jun 2026 · TPEx

Margin of safety

+12.0%

vs intrinsic

Enterprise value

NT$22.46B

72.6% terminal

Cost of equity / debt

6.66% / 2.44%

β 0.80 · CRP 0.78%

Terminal ROIC / g

10.00% / 2.00%

spread ~359bp (ROIC 10.00% vs WACC 6.41%)

What it sells, where it sells

Operating segments

Civil / building constructionSingle-segment turnkey contractor — MRT depots, hospitals, high-tech fabs, public works; parent 德昌 + wholly-owned MEP sub 德鎮盛100%

Contracted backlogNT$58.9bn official work-in-hand (Nov-2025 investor day) — >5x annual revenue; the retail NT$80bn figure includes uncontracted pipeline and is not the base5.3x rev

Te Chang is a single-line civil and building contractor; the donut is whole-company revenue, not a segment split. The number that matters is the NT$58.9bn contracted backlog — the official consolidated work-in-hand disclosed at the Nov-2025 investor day, over five times annual revenue. That is the floor's load-bearing input; the widely-quoted NT$80bn includes awarded-but-uncontracted pipeline and is deliberately excluded.

Country mix (revenue-weighted CRP input)

🇹🇼Taiwan100%

Entirely a Taiwan civil-works story — revenue is domestic public and high-tech construction. That concentrates the thesis on the Taiwan property/construction cycle, currently under the central bank's seventh round of credit tightening, and means there is no geographic diversification to cushion a domestic order-intake slowdown.

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.

The NT$192 perpetual-FCFF print is not underwritten (it was NT$251 before the 2.65% rate re-rate). The engine grows revenue ~10% fading to a 2.0% terminal rate and capitalizes the stream forever, putting ~73% of value in a terminal block. A backlog run-off rebuild showed the contracted backlog the bull thesis points to is only ~18% (NT$34/sh) of that figure, while a residual going-concern perpetuity carries 73% (NT$137/sh). The headline is a perpetuity wearing a run-off costume — we report it but do not anchor on it.
The defensible anchor is a hard floor. The contracted NT$58.9bn backlog run-off NPV (NT$34/sh) plus net cash (NT$17/sh) is ~NT$50/share of near-certain, low-assumption value — the NT$71.6 price is ~1.4× it with zero credit for the firm winning another contract. Apply the rebuild's own 10x-after-tax-EBIT cap to the residual and the base lands at ~NT$105; but the council showed even that over-credits a decelerating order book, so the honest fair value is ~NT$70–90 — roughly fair vs the NT$71.6 price.
Net cash, ~17% ROE, ~6% dividend yield. Cash 2,426 + ST investments 139 − debt 516 − minority interest 121 = NT$1,928M net cash. Screen D/E ~15.5% versus 117-162% for the pure Taiwan property developers the market lumps it with. Pays a NT$3.6 cash dividend (plus NT$1.99 stock) at ~6% yield, with a synthetic AAA balance sheet supporting a 0% failure assumption.
FY25 earnings are a cyclical peak, not a run-rate. 2025 revenue NT$11.08bn (+18% YoY); FY25 EBIT margin ~13% is the highest in the observable 5-year record (5y mean ~10.6%, through-cycle sector norm ~9-11% and broad construction ~5%). The model fades the margin hard to 10%; holding 13% forever is explicitly forbidden. EPS and book value are POC-lumpy — single monthly revenue prints swing −57% to +44%.
Construction DCFs are structurally fragile and the order book is rolling over. A project-based contractor does not compound a growing revenue base in perpetuity, so any perpetual-FCFF value is suspect. The residual that carries 73% of the headline assumes ~NT$20bn/yr order replenishment forever, yet cited new-contract intake is decelerating sharply: NT$30.9bn (FY23) → NT$20.3bn (FY24) → ~NT$10–13bn (FY25). That deceleration is the residual's key swing factor and the reason we stand on the floor.

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 Engineering/Construction median

10% Y1-5Backlog conversion + double-digit recent momentum, fading to 2.0% terminal — but project-based, not a true compounder

~4.5% CAGREngineering/Construction, global cross-sector median

+NT$60

02Operating marginYear-5 target vs sector median EBIT margin

10.0%FY25 peak ~13% faded hard to the through-cycle norm — explicitly not held at peak

~5%Broad Engineering/Construction sector median EBIT margin

+NT$55

03Sales-to-capitalReinvestment efficiency vs sector median

2.0×Asset-light contractor (D/E ~0.10-0.17); progress billings largely self-fund working capital

~4.36×Engineering/Construction industry standard

−NT$18

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

2.0%TWD long-run inflation; held ≤ the 2.65% risk-free (Damodaran ceiling)

2.65%risk-free ceiling (65/35 blend); terminal g held below it

NT$0

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

6.41%β 0.80 · rf 2.65% (65/35 blend) · CRP 0.78% · net-cash, D/V low single digits

~5.3%Engineering/Construction sector β re-levered → higher WACC

+NT$15

Why we discount this ladder

Every positive override here feeds the perpetual-FCFF NT$192 we explicitly disown (NT$251 before the rate re-rate). The growth and margin rows are what inflate a residual perpetuity to 73% of value on an order book that is decelerating (NT$30.9bn → NT$20.3bn → ~10–13bn). The honest read is not this ladder but the floor: contracted backlog (NT$34) + net cash (NT$17) ≈ NT$51/sh at 71% of price, with the residual capped at the rebuild's own 10x-EBIT to reach the ~NT$70–90 fair-value band.

fair ~NT$80

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 engine inputs.

The 10-year story

Te Chang is a net-cash Taichung civil and building contractor carrying a contracted NT$58.9bn work-in-hand backlog — over five times annual revenue. After re-rating the TWD risk-free from a suppressed 1.13% local bond to the methodology-correct 2.65% blend (WACC 6.41%), the engine still prints NT$192, but we do not underwrite it. A backlog run-off shows only ~18% of that is the visible contracted backlog; the rest is a residual perpetuity, and a project-based contractor that must re-win work in competitive public tenders earns roughly its cost of capital on that future work — so its marginal value is close to zero, not a capitalisable perpetuity. Strip that out and the firm is worth its near-bond backlog run-off plus net cash — a hard floor of about NT$50 a share — plus a thin premium for relationships and bonding capacity. That lands fair value around NT$70–90, against a NT$71.6 price: backlog at par, roughly fair. An independent council confirmed the downgrade from BUY to WATCH — the cheapness the floored band advertised (+47-56%) was itself over-crediting an order book that is decelerating (intake NT$30.9bn → 20.3bn → ~13bn) while the dividend was cut 24%.

Two debates worth pressure-testing

If the backlog is real and net cash is real, why only a WATCH?

Our view: Because the floor (~NT$50: backlog run-off + net cash) is below the price, and the value above the floor is a bet on winning future tenders — which are competitively bid, so they earn ~WACC and add little NPV. Fair value is backlog-at-par ~NT$70–90; at NT$71.6 you are paying roughly fair, with a real floor but only modest residual upside. Not a trap, not a bargain.

Didn’t the prior write-up call this a BUY at NT$107–116?

Our view: Yes, and the council corrected it. The capped NT$105 band silently assumed ~NT$10bn/yr of normalised revenue replenishment forever; the decelerating intake implies ~NT$4–6bn/yr is sustainable, which roughly halves the residual and pulls fair value to ~NT$70–90. The re-rate to 2.65% also lifted WACC (4.90% → 6.41%). Two things to watch flip it back to BUY: a year of intake above ~NT$20bn, or a reversal of the dividend cut.

CLAIM 01The robust anchor is a hard ~NT$50 floor; fair value is ~NT$70–90.backlog run-off (~NT$33) + net cash (~NT$17) ≈ NT$50/shThe price (NT$71.6) is ~1.4× the floor. Value above it is future-tender work that earns ~WACC (zero-NPV), so fair value is backlog-at-par ~NT$70–90 — roughly fair, a WATCH.

CLAIM 02The NT$192 perpetual-FCFF print is a category error — not underwritten.perpetual DCF NT$192 (was NT$251 at the old 1.13% rate)~73% is a residual perpetuity on competitively-bid future work whose economic profit is competed away. A project contractor is not a perpetual compounder; we report it as the optimistic bound only.

CLAIM 03Order intake is decelerating — the core reason for the downgrade.intake: NT$30.9bn → 20.3bn → ~10–13bn (FY23→FY24→FY25)The capped NT$105 band assumed ~NT$10bn/yr revenue replenishment forever; sustainable is nearer NT$4–6bn. That halves the residual and pulls fair value to ~NT$70–90.

CLAIM 04The dividend was cut 24% — the income leg of the old BUY is gone.dividend NT$5.0 → NT$3.8 · ~5–6% yield · payout ~50%Possibly working-capital funding for the backlog ramp (bullish-neutral) rather than distress — but a 24% cut while booking record backlog is a watch-item, not a BUY signal.

CLAIM 05Net cash, ~17% ROE, synth AAA — a floor, not a trap.net cash NT$1,928M · terminal_g 2.0% · gov haircut 10%Cash 2,426 + ST inv 139 − debt 516 − minority 121. Family-run, chairman+president both >70 with no named successor; 10% governance haircut. The downside is cushioned; the upside is just modest.

Where we diverge from sell-side

We downgraded our own pick. A council pressure-test (2026-06-01) moved this from BUY to WATCH: the capped NT$105 floored band still over-credited a decelerating order book, and the re-rate to a 2.65% risk-free lifted WACC to 6.41%. Fair value is backlog-at-par ~NT$70–90, not +47-56%.
Future tenders are ~zero-NPV. Competitive public-works bidding drives ROIC toward WACC, so the residual perpetuity that carries ~73% of the headline should not be capitalised at all. The anchor is the run-off backlog (~NT$33) + net cash (~NT$17) = a hard ~NT$50 floor.
The perpetual NT$192 is reported, not underwritten. A project-based contractor does not compound a growing revenue base forever; read NT$192 as the optimistic bound, ~NT$105 (10x-EBIT-capped) as still-generous, and ~NT$70–90 as fair.
Margin faded to 10% from the FY25 ~13% peak; terminal growth 2.0% (≤ the 2.65% rf). Even faded, the perpetuity is fragile — which is exactly why we stand on the floor.
Governance haircut 10%, heavier than the typical 5%. Family-run, chairman and president both over 70 with no named successor, a related-party web including a loss-making tyre subsidiary, and a dividend just cut 24%.

Two-sided case — bear anchors

The residual leans on order replenishment that is rolling over. The going-concern perpetuity (73% of the headline) assumes ~NT$20bn/yr intake at near-peak economics, but cited intake fell NT$30.9bn → NT$20.3bn → ~10–13bn. If steady-state intake is nearer the recent run-rate, the residual is materially overstated — and the headline collapses toward the floor.
Construction-DCF fragility is structural, not a modeling choice. Percentage-of-completion earnings are lumpy and cyclical (monthly revenue swings −57% to +44%); a turnkey contractor is not a perpetual compounder. Any value above the backlog-plus-cash floor is an assumption about winning future work, not a contracted cash flow.
Succession risk is live. Chairman (黃政勇) and president are both over 70 with no named successor; the company is family-run with limited English disclosure and a related-party web including a loss-making tyre subsidiary. A disorderly transition is the realistic governance tail the 10% haircut only partly captures.
Taiwan property/construction cycle and credit tightening. 100% domestic exposure to a market under the central bank's seventh round of credit controls ('量縮'). A domestic capex slowdown hits order intake directly with no geographic offset — compressing the residual that carries most of the headline.
Low-rate fragility cuts both ways. The 6.41% base WACC (after the 2.65% re-rate) and the WACC−g spread still inflate the perpetuity; a move toward normalized rates compresses it sharply. The 4.3%-rate stress (~NT$101 capped) is informative precisely because it strips that flattery — and even it implies a residual multiple above the model's 10x cap.

Risks to thesis (tail, not bear case)

Order-intake decelerationHigh

New-contract intake fell NT$30.9bn → 20.3bn → 11.4bn. The residual perpetuity (73% of the headline) assumes ~NT$20bn/yr forever; if intake stays near the recent run-rate, fair value collapses toward the ~NT$50 floor.

Construction-DCF fragilityHigh

POC-lumpy, cyclical, project-based earnings (monthly revenue −57% to +44%). A turnkey contractor is not a perpetual compounder — value above the backlog-plus-cash floor is a bet on future order wins, not contracted cash.

Succession / family controlMed

Chairman and president both >70, no named successor; family-run with a related-party web including a loss-making tyre sub. A disorderly transition is the realistic governance tail — priced via the 10% haircut.

Margin reverts below 10%Med

FY25 ~13% EBIT is a cyclical peak (5y mean ~10.6%, sector ~5%). Fixed-price contracts plus cost inflation could push the through-cycle margin to the 8-9% low band — though the rebuild shows this dents the headline only modestly.

Taiwan credit tighteningMed

100% domestic; the central bank's seventh round of credit controls pressures the property/construction cycle. No geographic offset if domestic capex slows.

Low-rate regime unwindLow

~73% of the headline is a terminal block at the 6.41% WACC. Rate normalization compresses the perpetuity — but the 4.3%-rate stress (~NT$101 capped) and the bare floor (~NT$50; price ~1.4× it) do not depend on the low-rate regime persisting.

10-year forecast

Revenue NT$11.08B → NT$22.40B over 10y in the engine’s perpetual-FCFF path (~10% Y1-5 fading to 2.0% terminal); operating margin faded from the FY25 ~13% peak to 10%. This is the path behind the NT$192 print we do not underwrite — read it as the optimistic bound, not the anchor.

Monte Carlo distribution

The Monte Carlo spans NT$160 (p5) to NT$234 (p95) and every draw clears the NT$71.6 price — but that band only stress-tests the perpetual-FCFF model we disown, not the thesis. The thesis stands on the floor: backlog run-off + net cash ≈ NT$50/sh, with honest fair value ~NT$70–90 once decelerating future tenders are valued at their ~zero economic profit.

Mean NT$194.18 ± NT$22.23/sh, 1000 iterations (0 failed). P(intrinsic < market NT$71.60) = 0.0%.

Cost of capital build

Risk-free rate	2.65%
Mature-market ERP	4.04%
Levered β	0.80
Weighted CRP	0.78%
Cost of equity	6.66%
Pre-tax cost of debt (synth Aaa/AAA)	3.05%
D / V	~6%
WACC	6.41%

Full year-by-year DCF

Year	Revenue	Op mgn	EBIT	EBIT(1−t)	Reinvest	FCFF	PV
1	NT$12.19B	12.54%	NT$1.53B	NT$1.20B	NT$554M	NT$646M	NT$607M
2	NT$13.41B	11.90%	NT$1.60B	NT$1.25B	NT$610M	NT$644M	NT$569M
3	NT$14.75B	11.27%	NT$1.66B	NT$1.31B	NT$671M	NT$635M	NT$527M
4	NT$16.23B	10.63%	NT$1.73B	NT$1.36B	NT$738M	NT$617M	NT$482M
5	NT$17.85B	10.00%	NT$1.79B	NT$1.40B	NT$811M	NT$590M	NT$433M
6	NT$19.35B	10.00%	NT$1.94B	NT$1.53B	NT$750M	NT$775M	NT$534M
7	NT$20.67B	10.00%	NT$2.07B	NT$1.63B	NT$658M	NT$977M	NT$631M
8	NT$21.74B	10.00%	NT$2.17B	NT$1.73B	NT$537M	NT$1.19B	NT$720M
9	NT$22.52B	10.00%	NT$2.25B	NT$1.80B	NT$391M	NT$1.40B	NT$796M
10	NT$22.97B	10.00%	NT$2.30B	NT$1.84B	NT$225M	NT$1.61B	NT$855M

Methodology & flags

Damodaran FCFF DCF, 10y explicit + perpetuity, in TWD. Risk-free 2.65% (local-currency government bond). Synthetic credit Aaa/AAA. CRP 0.78% from revenue-weighted country mix × Damodaran 2026 CRPs. Terminal ROIC faded to 10.00%; 10% governance haircut applied post-DCF. Monte Carlo: 1000 iterations. Engine v1.0.0 · result: /Users/valentin/Documents/notes_jiliac_labs/Finance/Damodaran/valuations/techang/output/2026-06-01-result.json

R&D cap: OFF · Lease cap: OFF · Failure: OFF · ESO: OFF
Governance haircut: 10% applied post-DCF (NT$213.78 > NT$192.40)
Sensitivity tornado: not run

FCFF DCF · dark theme v3. Generated 2026-06-01T10:16:33 from /Users/valentin/Documents/notes_jiliac_labs/Finance/Damodaran/valuations/techang/output/2026-06-01-result.json.