← back to picks

HQ Taoyuan · Taiwan Reporting TWD Control Su family ~43% · ~73% pledged Listed TWSE Valuation 2026-06-05 Damodaran FCFF · 1000-iter MC FCFF · Dark v3

Cheap on peak EPS, expensive on honest earnings · −43% margin of safety after the governance haircut

Market prices the completion peak; through-cycle value sits ~43% lower

The headline "~6× earnings" is a peak-EPS artifact: FY2025 revenue of NT$12,276M was a completion peak, not a run-rate. Normalize to a through-cycle NT$6,500M / 18% margin base and the FCFF intrinsic is NT$15.04 pre-governance, NT$12.03 after a 20% haircut for a Su-family-controlled, ~73%-pledged, dilution-prone developer — versus a NT$21.25 price. That is −43%. Monte Carlo puts P(intrinsic < market) = 90%, and the fair read only appears at the most flattering suppressed-bond risk-free. Overvalued — AVOID.

SectorReal Estate (Development)Country mixTaiwan 100%β / MC σ0.90 levered · ±NT$5.2/sh (1000 runs)GovernanceSu family ~43% · ~73% pledged · 20% haircutQuality~8-9% peak ROIC · D/E ~155% · lumpyIncomeNT$1.5 cash div (volatile · tilting to scrip)

Intrinsic / share

NT$12.03

post 20% gov · pre NT$15.04

Market / share

NT$21.25

5 Jun 2026 · TWSE

Margin of safety

-43.4%

vs intrinsic

Enterprise value

NT$16.44B

73.3% terminal

Cost of equity / debt

6.82% / 3.04%

β 0.90 · CRP 0.78%

Terminal ROIC / g

6.48% / 2.00%

spread ~149bp (ROIC 6.48% vs WACC 4.99%)

What it sells, where it sells

Operating segments

Residential / commercial developmentSingle-line completed-contract developer — projects across Taoyuan, Taichung, Kaohsiung, New Taipei; revenue recognized on handover100%

FY2025 completion peakReported NT$12,276M (皇普莊園 93億 + 皇普摩天100 26億 handed over) — a peak we normalize down, not the base1.9× base

Crowell is a single-line completed-contract developer; the donut is the normalized through-cycle revenue, not a segment split. The load-bearing number is the gap between the FY2025 reported NT$12,276M completion peak and the NT$6,500M base we model. Revenue swung NT$1,850M (FY23) → NT$5,331M (FY24) → NT$12,276M (FY25), then Q1-2026 collapsed to ~NT$10M — the empty gap between completion waves. Annualizing either end is wrong; the base is set ~1.3× the trailing-5yr average.

Country mix (revenue-weighted CRP input)

🇹🇼Taiwan100%

Entirely a Taiwan residential/commercial development story — every project sits in Taoyuan, Taichung, Kaohsiung or New Taipei. That concentrates the thesis on the domestic property cycle, currently under the central bank's selective credit controls (信用管制) with sector-wide transaction volumes shrinking, and leaves no geographic offset to cushion a demand 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 "~6× earnings" cheapness is a peak-EPS artifact. FY2025 revenue NT$12,276M was a completion peak (皇普莊園 + 皇普摩天100 handed over in the same year). Through-cycle, the 5yr cumulative revenue is ~NT$24,380M (~NT$4,900M/yr) at ~19.5% blended operating margin. We normalize the base to NT$6,500M / 18% — credit for the larger launched pipeline, but ~47% below the FY2025 peak the trailing P/E is computed on.
The asset base is not cheap either. The stock trades ~1.35× book, not below NAV. Inventory (land + construction-in-progress) of ~NT$20bn is 85-95% of assets and 3-4× equity, financed by debt: net debt ~NT$9.6bn against NT$7.2bn equity (D/E ~155%). This is a leveraged, capital-intensive developer, not a discount-to-assets value play.
Controlled company with a live pledge overhang. The Su family controls ~43% (和築投資-蘇孟光 37% + 遠安投資-蘇永平 5.4%); chairman is also president. Share-pledge ratios on the controlling stakes run ~70-73% — a major Taiwan red flag (margin-call risk, misaligned incentives) — alongside a related-party JV signed Jan-2026 (羅杰投資, 80/20 on a Zhongli plot) and related-party insider selling through 2025.
A persistent dilution machine. Shares are up ~7.9% YoY (Jul-2025 1.05 stock dividend + 2026 盈餘轉增資), on top of recurring private placements (2019/2020/2022) and two convertible bonds (皇普三 NT$600M, 皇普四 NT$500M). Management is explicitly tilting toward more stock dividend / cash retention for 2026 — per-share value leaks over time.
A real but volatile dividend, tilting to scrip. The cash dividend has been lumpy (0.46 → 0.50 → 0.50 → 2.0 → 1.5) and is increasingly paid as stock. On a normalized through-cycle payout (~NT$0.65/sh at 40% of through-cycle EPS ~NT$1.6) the cash that durably reaches a minority is only ~NT$13.5/sh — below the NT$21.25 price. That is why the FCFF intrinsic is treated as a ceiling, not a target.

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

01Base revenueNormalized through-cycle base vs FY2025 reported

NT$6,500M~1.3× trailing-5yr avg; crediting the NT$13.9bn launched pipeline — but ~47% below the FY2025 peak

NT$12,276MFY2025 reported completion peak (handover-year spike)

−peak

02Operating marginThrough-cycle target vs blended history

18.0%Below the 19.5% blended 5yr; far below the non-repeatable FY2024 31%

~19.5%Crowell's own through-cycle blended operating margin

−conservative

03Revenue growth10-year CAGR off the normalized base

4.0% Y1-5Scaled pipeline (NT$13.9bn launched + 2027-28 projects), not peak-chasing; fades to 2.0% terminal

~3-4%Mature developer in a credit-controlled market

NT$0

04Sales-to-capitalReinvestment efficiency vs sector median

0.45×Land-heavy developer; growth consumes inventory reinvestment

~0.41×Damodaran Real Estate (Development) sector

−reinvest drag

05Cost of capital10y WACC at the base TWD risk-free

4.99%β 0.90 · rf 2.31% (50/50 blend) · CRP 0.78% · D/V high (leveraged developer)

~5.3%Sector β re-levered at market leverage

+rate flatter

Why the ladder still lands below market

Even crediting the scaled pipeline, a normalized NT$6,500M / 18% base discounted at a 4.99% WACC prints NT$15.04 pre-governance — and the 20% controlled-company haircut takes it to NT$12.03, ~43% below the NT$21.25 price. The value is fair only at the most flattering 1.13% suppressed-bond risk-free (NT$20.39); at the methodology 2.31% blend it is overvalued, and at a normalized 4.3% rate it collapses to NT$2.36.

−43% MoS

Our input Sector / reported 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 DCF inputs.

The 10-year story

Crowell is a lumpy, Taoyuan-based completed-contract developer that the screen flags as cheap on a ~6× trailing multiple — but that multiple is computed on a FY2025 completion peak (revenue NT$12,276M, when 皇普莊園 and 皇普摩天100 were handed over in the same year), and Q1-2026 revenue already fell back to ~NT$10M in the empty gap between completion waves. Normalize the base to a through-cycle NT$6,500M at an 18% operating margin, grow it a modest 4% off the genuinely visible NT$13.9bn launched pipeline, and discount at the methodology-blended 2.31% TWD risk-free (WACC 4.99%), and the FCFF intrinsic is NT$15.04 a share before governance. Apply a 20% haircut for a Su-family-controlled company with ~73% of the controlling stake pledged, a fresh related-party JV, related-party selling, and a structural dilution habit, and the intrinsic is NT$12.03 — about 43% below the NT$21.25 price. A 1000-run Monte Carlo puts the probability that intrinsic is below the market at 90%. The number is fair only at the most flattering suppressed-bond risk-free (1.13% → NT$20.39); at a normalized 4.3% rate it falls to NT$2.36. This is the Te Chang pattern — cheap on the headline, expensive on honest normalization — so the verdict is AVOID, not a buy.

Two debates worth pressure-testing

It trades at ~6× earnings and pays a ~7% dividend — how is that not a buy?

Our view: The "6×" is on peak EPS — FY2025 was a completion-year revenue spike, and Q1-2026 already collapsed to a loss. On normalized through-cycle earnings the multiple is far higher, the intrinsic is NT$12.03 post-governance, and the dividend is volatile and tilting to scrip — a normalized cash payout (~NT$0.65/sh) is worth only ~NT$13.5/sh, below the price. You are paying a peak multiple for a leveraged, controlled developer.

The low-rate bookend shows it roughly fair — doesn't that rescue it?

Our view: Only at a 1.13% risk-free does it reach NT$20.39 (−4%), and that rate is the suppressed local-bond floor we explicitly de-weight as a thin-market artifact. At the methodology 50/50 blend (2.31%) it is −43%; at a normalized 4.3% rate it is −89% (NT$2.36). A thesis that survives only at the most flattering rate is not a margin of safety — it is rate fragility.

CLAIM 01The cheapness is a peak-EPS artifact; normalized value is ~43% below price.base NT$6,500M / 18% → intrinsic NT$12.03 post-gov vs NT$21.25FY2025 NT$12,276M was a completion peak; Q1-2026 fell to ~NT$10M. On a through-cycle base the FCFF intrinsic is NT$15.04 pre-governance, NT$12.03 after the haircut — overvalued.

CLAIM 02The value is fair only at the most flattering suppressed-bond rate.rf 1.13% → NT$20.39 · 2.31% → NT$12.03 · 4.3% → NT$2.36Three-regime robustness: roughly fair only at the local-bond floor (−4%), overvalued at the methodology blend (−43%), and −89% at a normalized rate. The thesis depends entirely on rate flattery.

CLAIM 03Monte Carlo says the downside is the base case, not the tail.P(intrinsic < market) = 90% · median NT$13.4 · mean NT$13.9 ± 5.2Across 1000 runs the market price sits above the 75th percentile of intrinsic. The distribution itself, not a worst case, prices the stock as overvalued.

CLAIM 04Controlled company with a ~73% pledge overhang and structural dilution.Su family ~43% · ~73% pledged · gov haircut 20%Chairman = president; related-party JV (Jan-2026) and related-party selling; recurring private placements + 2 CBs + a stock-dividend tilt. The 20% haircut covers the forward leakage/pledge/scrip gap.

CLAIM 05FCFF intrinsic is a ceiling — the cash a minority actually collects is lower.normalized payout ~NT$0.65/sh → DDM ~NT$13.5/sh < NT$21.25The dividend is volatile (0.46→2.0→1.5) and tilting to stock (pure dilution). On a through-cycle 40% payout the cash that durably reaches a minority is below the price — reinforcing AVOID.

Where we diverge from sell-side

We normalize away the peak the multiple is built on. The thin sell-side coverage anchors on a ~6× trailing P/E computed on FY2025's completion peak. We model a through-cycle NT$6,500M / 18% base (~47% below the peak), which is the entire reason our intrinsic (NT$12.03) sits below the NT$21.25 price.
We treat the FCFF print as a ceiling, not a target. For a controlled company where cash reaches minorities through a volatile, scrip-tilting dividend, the durable cash claim (~NT$13.5/sh on a normalized payout) is the binding number — and it too is below price.
We refuse to underwrite the 1.13% suppressed-bond rate. The only regime where Crowell looks fair (NT$20.39) is the thin-market local-bond floor. The methodology 50/50 blend (2.31%) is the decision rate, and it prints −43%.
We size the governance haircut at 20%, not the typical 5%. A ~73% pledge ratio, a fresh related-party JV, related-party selling, and a persistent dilution machine warrant the controlled-company tier — not a token discount.
We read the dividend as volatile and dilutive, not as yield support. The ~7% headline yield rests on a NT$1.5 payout that has swung wildly and is increasingly paid as stock; we do not capitalize it as a stable income stream.

Two-sided case — bear anchors

Rate fragility is the core risk. The intrinsic moves from NT$20.39 (rf 1.13%) to NT$12.03 (2.31%) to NT$2.36 (4.3%). Any move toward normalized rates — entirely plausible over a holding period — compresses the value violently, because a low WACC against a near-flat growth profile is doing most of the work.
Leverage amplifies the cycle. D/E ~155% with inventory 3-4× equity means a slow-selling launch or a land write-down hits equity hard. Debt is only ~14-21% covered by operating cash flow; the model's 0% failure assumption is generous for a developer this geared.
The pledge overhang is a live tail. ~73% of the controlling stake is pledged. A share-price decline can trigger margin calls that force selling into a thin float (~2% foreign, ~77-83% closely held), accelerating the very decline that triggered them.
Dilution is structural, not one-off. Private placements (2019/2020/2022), two convertible bonds, and a management tilt toward stock dividends mean the share count keeps rising. Per-share value leaks even if the enterprise performs.
Demand is under central-bank credit controls. 100% domestic exposure to a Taiwan property market with selective credit tightening (信用管制) and shrinking transaction volumes. A slower sell-through delays completed-contract recognition and stretches the debt-funded inventory.

Risks to thesis (tail, not bear case)

Rate-regime fragilityHigh

Intrinsic NT$20.39 → NT$12.03 → NT$2.36 across rf 1.13% / 2.31% / 4.3%. The fair read exists only at the suppressed-bond floor; a move toward normalized rates compresses the value by ~80%+.

Peak-EPS mispricingHigh

The ~6× P/E is computed on a FY2025 completion peak; Q1-2026 already fell to a loss. On a through-cycle base the multiple is far higher and the stock is overvalued — the market is annualizing a handover spike.

Leverage + inventory write-downMed

D/E ~155%, inventory 3-4× equity, debt ~14-21% covered by operating cash. A slow-selling launch or land write-down hits equity directly; the 0% failure assumption is generous for a developer this geared.

Pledge overhang / forced sellingMed

~73% of the Su-family controlling stake is pledged. A price decline can trigger margin calls into a thin float (~2% foreign), accelerating the decline — a self-reinforcing governance tail.

Structural dilutionMed

Recurring private placements, two convertible bonds, and a stock-dividend tilt keep raising the share count. Per-share value leaks even if the enterprise delivers — the 20% haircut covers the forward gap.

Taiwan credit tighteningLow

100% domestic; selective credit controls (信用管制) shrink transaction volumes. A slower sell-through delays completed-contract recognition and stretches debt-funded inventory — a headwind, not an immediate solvency threat.

10-year forecast

Revenue NT$6,500M → ~NT$9,078M over 10y off the normalized through-cycle base (4% Y1-5 fading to 2.0% terminal); operating margin held at 18%, below the 19.5% blended history. This deliberately excludes the FY2025 NT$12,276M completion peak the trailing multiple is computed on.

Monte Carlo distribution

The Monte Carlo spans NT$6.3 (p5) to NT$23.2 (p95) with a median of NT$13.4 — and the NT$21.25 price sits above the 75th percentile, so P(intrinsic < market) = 90%. The overvaluation is the base case of the distribution, not a tail outcome; only the most optimistic ~10% of draws clear the price.

Mean NT$13.87 ± NT$5.20/sh, 1000 iterations (0 failed). P(intrinsic < market NT$21.25) = 90.0%.

⚠ Active diagnostic: sales_to_capital_y1_5 (0.45) outside the typical [0.5, 5] band; verify the industry/business model sales_to_capital_y6_10 (0.45) outside the typical [0.5, 5] band; verify the industry/business model

Cost of capital build

Risk-free rate	2.31%
Mature-market ERP	4.14%
Levered β	0.90
Weighted CRP	0.78%
Cost of equity	6.82%
Pre-tax cost of debt (synth Aaa/AAA)	3.80%
D / V	~48%
WACC	4.99%

Full year-by-year DCF

Year	Revenue	Op mgn	EBIT	EBIT(1−t)	Reinvest	FCFF	PV
1	NT$6.76B	18.00%	NT$1.22B	NT$998M	NT$578M	NT$420M	NT$400M
2	NT$7.03B	18.00%	NT$1.27B	NT$1.04B	NT$601M	NT$437M	NT$396M
3	NT$7.31B	18.00%	NT$1.32B	NT$1.08B	NT$625M	NT$454M	NT$393M
4	NT$7.60B	18.00%	NT$1.37B	NT$1.12B	NT$650M	NT$472M	NT$389M
5	NT$7.91B	18.00%	NT$1.42B	NT$1.17B	NT$676M	NT$491M	NT$385M
6	NT$8.19B	18.00%	NT$1.47B	NT$1.20B	NT$633M	NT$571M	NT$425M
7	NT$8.46B	18.00%	NT$1.52B	NT$1.24B	NT$583M	NT$653M	NT$461M
8	NT$8.69B	18.00%	NT$1.56B	NT$1.26B	NT$526M	NT$738M	NT$491M
9	NT$8.90B	18.00%	NT$1.60B	NT$1.29B	NT$464M	NT$825M	NT$517M
10	NT$9.08B	18.00%	NT$1.63B	NT$1.31B	NT$396M	NT$912M	NT$536M

Methodology & flags

Damodaran FCFF DCF, 10y explicit + perpetuity, in TWD. Risk-free 2.31% (local-currency government bond). Synthetic credit Aaa/AAA. CRP 0.78% from revenue-weighted country mix × Damodaran 2026 CRPs. Terminal ROIC faded to 6.48%; 20% governance haircut applied post-DCF. Monte Carlo: 1000 iterations. Engine v1.0.0 · result: /Users/valentin/Documents/notes_jiliac_labs/Finance/Damodaran/valuations/crowell/output/2026-06-05-result.json

R&D cap: OFF · Lease cap: OFF · Failure: OFF · ESO: OFF
Governance haircut: 20% applied post-DCF (NT$15.04 > NT$12.03)
Sensitivity tornado: not run

FCFF DCF · dark theme v3. Generated 2026-06-05T01:47:03 from /Users/valentin/Documents/notes_jiliac_labs/Finance/Damodaran/valuations/crowell/output/2026-06-05-result.json.