← back to picks
KOSDAQ 011560 · cleanroom MEP/EPC Reporting KRW (₩ millions) Credit Aaa/AAA (net cash) Valuation 2026-05-31 Damodaran FCFF · 1000-iter MC FCFF · Dark v3

Backlog-floored cyclical, net-cash, deeply mispriced · +61.2% margin of safety

Market far below MC p5

The post-governance DCF lands at ₩58,318/sh against a ₩22,600 market price; even the Monte Carlo 5th percentile (₩46,872) sits ~100% above today. A thin-margin but #1-in-niche, net-cash cleanroom contractor riding ₩70tn of 2026 Samsung/SK Hynix fab capex, still priced as plain construction.

p5 p25 p50 p75 p95 MARKET ₩22,600 DCF ₩58,318
Sector Cleanroom MEP/EPCCountry Korea 97% / VN 3%MC σ ±₩5,062Governance Family ~48-52% · −10%Quality Net cash · Aaa/AAANiche #1 1st-stage piping
Intrinsic / share
₩58,318
post 10% gov (₩64,798 pre)
Market / share
₩22,600
KOSDAQ 011560
Margin of safety
+61.2%
vs intrinsic
Enterprise value
₩487.9B
84.4% terminal
Cost of equity / debt
7.58% / 3.26%
β 0.76 · CRP 0.74%
Terminal ROIC / g
10.00% / 3.00%
spread ~246bp

What it sells, where it sells

Operating segments

~6.0% op mgn
Hi-tech facility (semi + display)Cleanroom piping, duct, specialty-gas for Samsung/SK Hynix · #1 1st-stage piping, ~60% fluororesin duct share~85%
Plant & general buildingLNG, power, petrochemical, general facilities · lumpier, lower-mix~15%

~85% of revenue is the percentage-of-completion derivative of Korean memory-fab construction starts — so the margin and growth debate is really a Samsung/SK Hynix capex-cycle debate, with HBM specialty-gas piping the mix upside and structural EPC thinness the floor.

Country mix (revenue-weighted CRP input)

🇰🇷South Korea97.0%
🇻🇳Vietnam (proxy)3.0%

Near-pure Korea: assets, labour, financing and end-customer demand are all won-based, so the weighted CRP lands at 0.74% (essentially Korea's Aa2 0.64% + a 3% Vietnam-proxy nudge). True economic exposure to the Korean chip cycle is arguably closer to 100% than 97%.

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 median
Direction vs sector
01Revenue growthThrough-cycle Y1-5 CAGR
7.0% backlog ramp net of cyclical give-back
~5% global E&C industry standard
adds
02Operating marginYear-10 target
6.0% demonstrated mid-cycle + HBM mix
5.1% global E&C EBIT margin
adds
03Sales-to-capitalReinvestment efficiency (ramp)
1.5 council cut: AR-build WC drag
1.86 global E&C industry standard
subtracts
04Terminal growthYear 10+ steady state
3.0% < 3.78% KRW risk-free ceiling
~3.0% mature-economy GDP proxy
~flat
05Cost of capital10y WACC
7.54% β_u 0.759, regression 0.74 confirms
~7.5% sector β at this near-zero D/E
~flat
Net effect of overrides
The value-adding calls (above-sector growth and margin) are partly offset by the deliberately conservative 1.5 sales-to-capital the council imposed to charge for the FY2025 AR build; terminal growth and WACC sit at sector defaults. Net modestly value-adding, but the conservative reinvestment is what keeps the case defensible.
net adds
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

Sebo is a thin-margin, asset-light, net-cash cleanroom contractor — Korea's #1 in first-stage fab piping and ~60% of the specialty fluororesin-duct market — whose revenue is the one-to-two-year echo of Samsung and SK Hynix memory-fab construction starts. This is not a growth compounder; it is a cyclical riding, not expanding, its incumbent share. Off a three-year-trough ₩715.8bn base, a record ₩757bn backlog and ₩70tn of 2026 chip capex re-accelerate revenue back toward the prior-cycle ₩1tn peak by year five, before a normal downcycle gives some of it back — netting to ~7% annual growth that fades to a mature 3% by year ten. Operating margin holds the now-demonstrated ~6%, and because reinvestment per won of new revenue is deliberately set conservatively, the cash the model credits is defensible rather than balance-sheet-flattered.

Two debates worth pressure-testing

Is the FY2025 cash drain distress, or a growth signal?
Our view: a ~₩41bn accounts-receivable build during an order ramp under POC accounting, on a net-cash balance sheet — working-capital mechanics, not a value trap. We still charged for it: the council cut sales-to-capital 2.5→1.5, lowering FCFF.
Does the FY2026 +27% consensus become a permanent base?
Our view: no — the FY2022 +138% spike faded three straight years, so we model a ramp-then-give-back, not a step-change. The bull "₩910bn compounds forever" is plausible-not-probable and lives in the stress case, worth materially more per share if it holds.
CLAIM 01Revenue re-accelerates off the trough, then mean-reverts.Y1-5 CAGR 7.0% → terminal 3.0%Record ₩757bn backlog + ₩70tn 2026 capex floor FY2026-27; through-cycle growth is mid-single-digit, not the 2026 peak extrapolated.
CLAIM 02~6% is the demonstrated mid-cycle margin.Year-10 op margin 6.0%FY2025 printed a cycle-high 5.64% on the lowest revenue (op profit +41% on −9.3%); HBM specialty-gas mix is the upside, EPC thinness the floor.
CLAIM 03Asset-light POC, but charged for WC drag.Sales-to-capital 1.5 (ramp) / 1.8Above the 1.86-ish sector level on a negative-WC model, but cut by the council from 2.5 to charge for the FY2025 AR build rather than credit a cash-flattered optical.
CLAIM 04One cycle from mature, not one decade.Convergence by year 5Margin is already at steady-state today and the backlog ramp plays out over FY2026-27, so growth decays to mature by year 10.
CLAIM 05Net-cash tail; terminal ROC disciplined.Failure 0% · ROC 10% · g 3.0%Never a going-concern question; ROC pinned to 10% so the net-cash-flattered optical ~28% ROIC does not leak into perpetuity.
Where we diverge from sell-side
  • We charge for the AR build the bulls ignore. Sales-to-capital cut 2.5→1.5 (ramp) after the FY2025 ~₩41bn receivables swing — a deliberate FCFF haircut, not a cash-flattered optical.
  • Terminal ROC pinned to 10%, not the optical 28%. The net-cash pile flatters reported ROIC to ~28%; we let only a modest ~150-200bp spread over the ~7.5% Korean WACC into perpetuity.
  • Governance haircut sized to the track record, not the stereotype. 10% — above a clean 5% for the ~48-52% family control, below 15-20% given treasury cancellation and a decade of rising dividends.
  • We model a ramp-then-give-back, not a new permanent base. The FY2022 +138% spike faded three years; revenue peaks near ₩1tn at Y5 then mean-reverts, vs the bull "₩910bn compounds."
Two-sided case — bear anchors
  • A 10-15% chip-capex cut flows straight to revenue. ~85% of demand is two buyers; a Samsung/SK Hynix pullback the next year would push revenue back toward the FY2025 trough and compress the ramp.
  • The AR build is a value-trap tell, not WC timing. If 2H-2025 negative operating cash flow persists, the dividend really is funded from the cash pile and the FCFF case erodes.
  • Margin re-acceleration dilutes mix. A volume up-leg historically pulls in lower-margin first-stage piping; if the 6% slips back toward ~5%, terminal value falls materially.
  • Korea-discount overhang resolves badly. A founder-family controls ~48-52%; if minority treatment reverts to the regional norm, more than the 10% haircut leaks out of the per-share number.

Risks to thesis (tail, not bear case)

Two-customer concentrationHigh

~85% of revenue rides Samsung + SK Hynix fab capex; a synchronized memory-capex deferral hits revenue the following year with little offset.

Cash conversion / AR buildMed

FY2025 2H operating cash flow turned negative on a ~₩41bn receivables build; if it does not unwind, the earnings-quality flag chips at the FCFF thesis.

Margin dilution on volumeMed

A strong volume up-leg historically pulls in lower-mix first-stage piping; ~100bp of margin give-back meaningfully lowers terminal value.

KRW rate normalizationLow

The 3.78% risk-free is cyclically elevated by an oil shock; a fall to ~3.0-3.3% would lower WACC and raise value — a tail to the upside, modeled separately in the stress case.

Thin coverage / liquidityLow

0-1 sell-side analysts and KOSDAQ small-cap liquidity mean the re-rating, even if right, can stay un-arbitraged for an extended period.

10-year forecast

Revenue ramps ₩765.9B → ₩1.00T by Y5 (back near the prior-cycle ₩1tn peak on a record ₩757bn backlog), then mean-reverts to ₩1.26T by Y10. Operating margin holds the demonstrated mid-cycle ~6.0%; FCFF inflects sharply from Y6 as reinvestment intensity eases.

Y1 Y2 Y3 Y4 Y5 Y6 Y7 Y8 Y9 Y10 0 1.3T rev (KRW) 0% 10% op margin revenue FCFF op margin

Monte Carlo distribution

Even at the 5th-percentile outcome (₩46,872), intrinsic value exceeds today's ₩22,600 price by 52% — across 1,000 correlated draws not a single path lands below the market. The debate is the size of the gap, not its sign.

p5 p25 p50 p75 p95 market 22600.00 20704.1 55567.6 71894.7 freq equity / share (KRW)

1000 iterations (0 failed) drawing revenue growth, operating margin, terminal growth, ramp-year sales-to-capital and the governance discount. Mean ₩55,434 ± ₩5,062/sh. P(intrinsic < market ₩22,600) = 0.0%.

Cost of capital build
Risk-free rate 3.78%
Mature-market ERP 4.00%
Levered β 0.76
Weighted CRP 0.74%
Cost of equity 7.58%
Pre-tax cost of debt (synth Aaa/AAA) 3.26%
D / V ~0%
WACC 7.54%
Full year-by-year DCF
Year Revenue Op mgn EBIT EBIT(1−t) Reinvest FCFF PV
1 ₩765.9B 5.71% ₩43.7B ₩33.6B ₩33.4B ₩237M ₩220M
2 ₩819.5B 5.78% ₩47.4B ₩36.4B ₩35.7B ₩708M ₩612M
3 ₩876.9B 5.86% ₩51.3B ₩39.5B ₩38.2B ₩1.2B ₩1,000M
4 ₩938.3B 5.93% ₩55.6B ₩42.8B ₩40.9B ₩1.9B ₩1.4B
5 ₩1.00T 6.00% ₩60.2B ₩46.3B ₩43.8B ₩2.5B ₩1.8B
6 ₩1.07T 6.00% ₩64.0B ₩49.3B ₩34.6B ₩14.8B ₩9.5B
7 ₩1.12T 6.00% ₩67.4B ₩52.1B ₩32.0B ₩20.2B ₩12.1B
8 ₩1.18T 6.00% ₩70.5B ₩54.7B ₩28.7B ₩26.0B ₩14.5B
9 ₩1.22T 6.00% ₩73.2B ₩56.9B ₩24.8B ₩32.1B ₩16.6B
10 ₩1.26T 6.00% ₩75.4B ₩58.8B ₩20.3B ₩38.5B ₩18.4B
Methodology & flags

Damodaran FCFF DCF, 10-year explicit forecast plus perpetuity, in KRW (₩ millions). No R&D or lease capitalisation (asset-light EPC/MEP contractor); net-cash balance sheet drives a synthetic Aaa/AAA credit and near-zero debt weight. Country risk premium is revenue-weighted across South Korea (97%) and a Vietnam overseas proxy (3%). Monte Carlo: 1000 iterations over five correlated inputs. Engine v1.0.0 · result: /Users/valentin/Documents/notes_jiliac_labs/Finance/Damodaran/valuations/sebo/output/2026-05-31-result.json

  • R&D cap: OFF · Lease cap: OFF · Failure: OFF · ESO: OFF
  • Governance haircut: 10% applied post-DCF (₩64,798 > ₩58,318)
  • Sensitivity tornado: not run · uncertainty captured via 5-axis Monte Carlo