← back to picks

HQ Hsinchu · Taiwan Reporting TWD Credit synth AAA · net cash Structure holdco · 55% of listed AES-KY (6781) Valuation 2026-06-04 Damodaran FCFF · 1000-iter MC FCFF · Dark v3

A sum-of-the-parts holdco, not a cheap assembler · +54% base, but it’s a leveraged AES-KY bet

Base MoS says BUY (+54%) — but read it as a ceiling: a 6% dividend + a no-expiry NAV option behind a holdco discount

Base intrinsic NT$631/share (after a 5% governance haircut; pre-haircut NT$665) vs market NT$410. But the screener’s “cheap P/E 12.9” is a consolidation artifact: Simplo owns ~55% of separately-listed AES-KY (6781) (~NT$99bn, ~30x P/E, AI-server-BBU growth) and consolidates 100% of it. Value = core IT-battery DCF (NT$332/sh) + AES stake at a normalized 20x (NT$194/sh) + parent net cash (NT$139/sh). At NT$410 the market gives the declining core away for ~free. Strong downside (net cash + ~6% dividend; compound bear only −3%), but the upside is trapped behind a persistent Taiwan holdco discount with no hard catalyst.

StructureHoldco · 55% of listed AES-KY (6781)Country mixUS 46% · DE 18% · CN 14% · TW 12%β / MC σ0.95 pinned · ±NT$46/sh (1000 runs)Cash return~6% div · 71% payout · 15+ yrs (~4.7% net of WHT)QualityNet cash ~NT$26bn · synth AAACatchNAV trapped behind a permanent holdco discount

Intrinsic / share

NT$631.55

post 5% gov · pre NT$664.79

Market / share

NT$410.00

2026-06-04 close · TPEx

Margin of safety

+54.0%

vs intrinsic

Enterprise value

NT$61.33B

63.0% terminal

Cost of equity / debt

6.71% / 2.17%

β 0.95 · CRP 0.40%

Terminal ROIC / g

10.00% / 2.00%

spread ~329bp (ROIC 10.00% vs WACC 6.71%)

What it sells, where it sells

Operating segments

Core IT battery DCFWorld-#1 notebook/IT pack assembler, ex-AES. Declining ~6% margin; low-rate WACC lifts it to ~15x EV/EBIT — a ceiling.50%

55% AES-KY (6781) stakeListed AI-server-BBU + light-EV growth co. Carried at a normalized ~20x (vs ~30x market). The thesis swing factor.29%

Parent net cash~NT$26bn, no real debt — funds the ~6% dividend; a hard floor.21%

Not a product split — a value composition. ~half the equity value is the core IT-battery DCF (itself low-rate-WACC-inflated to ~15x EV/EBIT on a declining business — a ceiling), ~29% is the 55% AES-KY stake carried at a normalized 20x (vs ~30x market), and ~21% is parent net cash. The thesis lives in the AES mark and the core multiple, not in the assembler’s operations.

Country mix (revenue-weighted CRP input)

🇺🇸United States46%

🇩🇪Germany18%

🇨🇳China14%

🇹🇼Taiwan12%

🇯🇵Japan5%

🇰🇷South Korea5%

Weighted by end-customer / revenue geography of the CORE IT business (where notebook/IT demand sits), not production-site location (China/Vietnam/Taiwan plants). US 46% = Apple/HP/Dell; Germany proxies W. Europe; China 14% = Lenovo + local demand; Taiwan 12% = Asus/Acer + home listing. Blended CRP lands low (~0.40%); production-country asset risk sits in the governance discount, not the CRP.

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 verdict is a BUY on the base MoS (+54%) — but read it as a ceiling, not a target. Simplo is a sum-of-the-parts holding company. Its “cheap” consolidated P/E is an artifact of consolidating 100% of fast-growing, 30x-P/E AES-KY (6781) into a low-multiple parent. The honest read: a net-cash-backed ~6% dividend plus a no-expiry option on a NAV gap that a persistent Taiwan holdco discount may never close.
At NT$410 the market gives the core IT-battery business away for ~free. Market cap NT$75.8bn ≈ 55% AES-KY stake at market (~NT$54bn) + parent net cash (~NT$26bn). So you are paying ~nothing for the world’s #1 notebook battery-pack maker (NT$66bn revenue, ~6% margin, cash-generative) — a real margin of safety, but one that rests on AES-KY’s multiple holding.
It is, functionally, a leveraged bet on AES-KY. AES-KY makes AI-server battery-backup units (BBU) + light-EV (e-bike) modules: FY25 revenue ~NT$16bn (+60%), ~23% op margin, ~NT$3.26bn net income, trading at ~30x. If you want that growth, owning 6781 directly is the cleaner expression; Simplo wraps it in a declining core and a holdco discount.
The base DCF over-values the core — a known low-rate-currency distortion. The TWD blended risk-free (2.31%) and β 0.95 give a 6.71% WACC, which lifts a structurally declining, ~6%-margin commodity assembler to ~15.3x EV/EBIT (terminal = 119% of core EV). A melting ice cube merits ~6–9x. Reconcile the core to a sane multiple and the AES stake to a realizable ~20x and NAV ≈ NT$480–510 (+18–24%).
Downside is genuinely protected; upside is genuinely trapped. Compound bear (rf 4.3% + AES@12x + core −1.5%/5.5% margin + 10% gov) lands NT$399 ≈ −3%: net cash + the covered dividend catch the fall. But no catalyst forces the discount shut — the unlock list (buyback-and-cancel, AES dividend up-streaming, activist/holdco action) is a wish list.
Clean cash-returner; structure modeled in the flows, not bolted on. ~6% dividend, ~71% payout, 15+ years, no dilution, net cash, Deloitte-audited, 4/7 independent board. The one real flag — family control across the Simplo↔AES-KY listed pair — is captured via the AES haircut + a 5% governance discount, per the Korea-controlled-company methodology (model it in the flows, never a flat “discount”).

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

Naive / market

Value impact

01AES-KY mark55% stake multiple vs market 30x

~20x P/ENormalized below the 30x AI-BBU market — froth/realizability haircut. NT$36bn

~30xMarket price taken at face — NT$54bn

−NT$99

02Core revenue growthY1-5 vs the IT melt

+0.5%~Flat: unit erosion ↔ ASP cell-cost pass-through + drone/Vietnam adds

−4 to −5%Trailing 5y core decline

+NT$60

03Core op marginYear-5 target vs 6.1% base

6.5%Slight mix/scale lift; commodity assembler ceiling

~6%Held flat at base

+NT$25

04Cost of capital (β)10y WACC vs suppressed regression

6.71% (β 0.95)Sector-aware compromise over the 0.41 suppressed Taiwan regression

5.5–8.7%β 0.41 → WACC ~5.5% (low) / sector 1.57 → 8.7% (high)

±NT$80–160

05GovernanceHaircut vs clean baseline

5%Founder-GM dual role + cross-entity (AES) related-party; structure modeled in flows

0%No haircut

−NT$33

Net effect

The base +54% rests on two soft pins the council flagged: the AES mark (20x vs 30x) and the low-TWD WACC that lifts a declining core to ~15x EV/EBIT. Reconciled (core ~8x, AES realizable ~20x) NAV ≈ NT$480–510 — a ceiling. The bankable return is the ~4.7%-net dividend + a no-expiry NAV option.

+NT$221

Our override Naive/market 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

Simplo is the world’s #1 notebook battery-pack assembler, and on a consolidated P/E of ~13 it screens cheap. That signal is an artifact. Simplo owns ~55% of separately-listed AES-KY (6781) — an AI-server battery-backup and e-bike battery company growing 60% a year at ~23% margins and trading at ~30x — and consolidates all of its earnings into a low-multiple parent. Decompose it and Simplo is a holding company: a 55% AES stake worth ~NT$54bn at market, parent net cash of ~NT$26bn, and a declining ~6%-margin core IT-battery business that the market, at NT$410, is effectively giving away for free. The engine’s +54% base is real on its inputs, but it is a ceiling: the low Taiwan WACC inflates a melting-ice-cube core to ~15x EV/EBIT (119% of its value is the perpetuity), and the AES stake is carried below the 30x market that itself looks hype-adjacent. Reconciled honestly — core at ~8x, AES at a realizable ~20x, net cash at face — NAV is closer to NT$480–510, still above the NT$410 price. The asymmetry is favorable: a compound bear stacking a rate shock, an AES de-rate, a shrinking core and a stiffer governance haircut only reaches −3%, because net cash and a covered ~6% dividend (about 4.7% net of Taiwan withholding) catch the fall. So the real question is not whether NT$410 is cheap — it probably is — but whether you will ever collect the gap. Taiwanese holdcos trade at persistent discounts and nothing on the calendar forces this one shut. Treat it as a yield-plus-optionality position, sized as the leveraged AES-KY bet it actually is — or, if it’s the growth you want, buy 6781 directly.

Two debates worth pressure-testing

Is AES-KY’s 30x a durable AI-infrastructure compounder or a cyclical/hype peak?

Our view: The whole valuation hinges on this. AES-KY’s AI-server BBU demand is real and content-per-rack rises, but 30x on +60% growth bakes in a lot. We carry the stake at a normalized 20x — above a mature 12–15x battery multiple, below the hype mark. Bull (Expansionist) says 20x leaves money on the table; the skeptics say a control block isn’t realizable at full market anyway. Either way it is ~70% of the stake value and the dominant Monte-Carlo axis.

Why buy the holdco at all instead of AES-KY (6781) directly?

Our view: Honest answer — for the 6% dividend and a cheap declining core, not for the growth. If the thesis is “own the AI-battery story,” 6781 direct is cleaner (no discount drag, no declining-core dilution). Simplo is the trade only if you want a net-cash-backed yield with free NAV optionality and are willing to wait without a catalyst.

CLAIM 01Simplo is a holdco: ~55% of listed AES-KY (6781) + net cash ≈ its entire market cap.AES stake ~NT$54B (mkt) + net cash ~NT$26B ≈ NT$75.8B capThe screener’s P/E 12.9 “cheap” is a consolidation artifact — 100% of high-multiple AES-KY earnings sit in a low-multiple parent. At NT$410 the market gives the core away for ~free (AES@30x).

CLAIM 02Core IT battery is a flat-to-declining, ~6%-margin commodity annuity.core rev +0.5% · op margin 6.5% · NT$66B baseWorld-#1 notebook/IT pack share, sticky OEM ties (Apple/HP/Dell), but price-competitive vs Dynapack and facing a 2026 memory-shortage headwind. Cash-generative; funds the dividend. Not a growth story.

CLAIM 03AES-KY stake carried at a normalized ~20x, not the 30x AI-BBU market.cross_holdings NT$35.9B · 0.55 × 20 × NT$3.26BAES-KY (AI-server BBU + e-bike) grows +60% at ~23% op margin — a genuine growth asset — but 30x is hype-adjacent and a control block isn’t realizable at full market. 20x is the fair, realizable mark. The single biggest swing (12x–30x → ±NT$78–120/sh).

CLAIM 04Net cash + a covered ~6% dividend put a hard floor under the price.net cash ~NT$26B · DPS NT$21.8 · 71% payout · 15+ yrsUnlike a trapped Korean holdco, cash actually reaches holders. ~21% Taiwan withholding trims 6% gross to ~4.7% net for a foreign retail holder — still the bankable leg while you wait for the NAV gap to close.

CLAIM 05The +54% is a ceiling, not a target — no catalyst forces the discount shut.β 0.95 · WACC 6.71% · terminal 119% of core EVThe low-TWD WACC lifts a declining core to ~15x EV/EBIT. Reconciled NAV ~NT$480–510. Taiwanese holdcos trade at persistent discounts; the unlock list (buyback+cancel, AES up-streaming, activist) is a wish list, not a calendar.

Where we diverge from sell-side

De-consolidated sum-of-the-parts, not a consolidated DCF. AES-KY (6781) is stripped from the base year and re-added as a market-valued cross-holding; minority interest set to 0. A naive consolidated DCF minus book NCI (NT$7.6bn) would massively overstate equity — AES’s minority is worth ~NT$45bn at market.
AES stake at a normalized ~20x (NT$36bn), not the 30x market (NT$54bn). Froth + control-block realizability haircut. The dominant swing: 12x–30x moves value ±NT$78–120/sh. Worth ~NT$99/sh between the bear and bull marks.
β pinned 0.95 over a 0.41 suppressed regression. Taiwan thin-float/family-control beta trap; sector (Electronics General) re-levers to ~1.57. 0.95 is a compromise; in this low-rate, terminal-heavy DCF it is the load-bearing discount-rate call.
Core terminal ROC faded to 10%, terminal growth 2%. A commodity assembler cannot hold the engine’s ~15% perpetual ROC against a ~6.7% WACC. Even so the low WACC lifts the core to ~15x EV/EBIT — read the base as a ceiling.
Governance 5%, modeled in the flows. The Simplo↔AES-KY family-control structure is captured via the AES haircut + dividend cross-check, not a flat “Korea/Taiwan discount,” per workspace methodology.

Two-sided case — bear anchors

AES-KY de-rates from 30x. The single most likely failure mode. AES is ~70% of stake value on AI-BBU enthusiasm; 30x→15x alone is ~−NT$78–145/sh before the core moves. The MC samples this (down to 12x) and still lands +36% at p5 — but only because the over-valued 15x core props it up.
The core is worth less than 15x EV/EBIT. A declining, concentrated, price-warred assembler probably merits 6–9x. At 8x the core is ~NT$174/sh not ~NT$332 — NAV falls toward NT$480–510, and the “cheap” shrinks.
The discount never closes. No catalyst, 15+ years of holdco discount. A +54% “ceiling” you never collect is, in practice, a ~4.7%-net dividend plus a lottery ticket. This is the First-Principles/Outsider verdict: WATCH.
Withholding + frictions erode the wait-and-collect leg. ~21% Taiwan dividend withholding (→ ~4.7% net), FX, and thin TPEx spreads for a foreign retail holder.
Core decline accelerates. 2026 memory-shortage cost inflation + soft notebook demand could push the core below the flat assumption, turning the “free” core into a slow value leak.

Risks to thesis (tail, not bear case)

AES-KY multiple de-rateHigh

~70% of stake value at ~30x on AI-BBU hype. A de-rate to a mature 12–15x lops NT$78–145/sh off the holdco before the core does anything. The dominant Monte-Carlo axis and the most likely failure mode.

Permanent holdco discount / no catalystHigh

15+ years of discount; nothing forces it shut. The +54% is a NAV ceiling you may never collect — in practice a ~4.7%-net dividend + a no-expiry option. Verdict pivots on this from BUY to WATCH.

Over-valued core (low-rate WACC)Med

The 6.71% TWD WACC lifts a declining 6%-margin assembler to ~15x EV/EBIT (terminal = 119% of core EV). At a defensible 6–9x the core is worth ~half the base figure.

β is the suppressed 0.41 or the sector 1.57Med

Pinned 0.95. At the sector 1.57 (WACC ~8.7%) the base falls to +32%; at 0.41 it balloons. The load-bearing discount-rate call in a terminal-heavy DCF.

Core IT decline + customer concentrationLow

Notebook/phone packs are mature, price-competitive (Dynapack), Apple/HP/Dell-concentrated; 2026 memory shortage is a near-term headwind. Cash-generative and net-cash-backed, so a value leak, not a solvency risk.

Withholding tax / FX / liquidityLow

~21% Taiwan dividend WHT cuts 6% gross to ~4.7% net; FX and thin TPEx spreads for a foreign retail holder. Erodes the “paid to wait” leg.

10-year forecast

This is the CORE IT-battery business only (AES-KY de-consolidated and valued separately as a cross-holding). Core revenue ~flat (+0.5% Y1-5: unit erosion offset by ASP cell-cost pass-through + drone/Vietnam adds); operating margin ~6.5%. A cash-generative annuity that funds the dividend — the growth lives in the separately-valued AES-KY stake, not here.

Monte Carlo distribution

The Monte Carlo’s dominant axis is the AES-KY stake (sampled 12x–30x), plus core growth/margin, terminal growth and governance. P(intrinsic<market)=0% across 1,000 draws — but note this is propped up by the low-rate-inflated 15x core; it confirms the floor (net cash + AES + a real dividend) is robust, not that the +54% upside is. The upside is gated by a holdco discount the MC cannot model.

Mean NT$632.11 ± NT$45.92/sh, 1000 iterations (0 failed). P(intrinsic < market NT$410.00) = 0.0%.

Cost of capital build

Risk-free rate	2.31%
Mature-market ERP	4.23%
Levered β	0.95
Weighted CRP	0.40%
Cost of equity	6.71%
Pre-tax cost of debt (synth Aaa/AAA)	2.71%
D / V	~0%
WACC	6.71%

Full year-by-year DCF

Year	Revenue	Op mgn	EBIT	EBIT(1−t)	Reinvest	FCFF	PV
1	NT$66.48B	6.16%	NT$4.09B	NT$3.12B	NT$110M	NT$3.01B	NT$2.82B
2	NT$66.81B	6.24%	NT$4.17B	NT$3.18B	NT$111M	NT$3.07B	NT$2.70B
3	NT$67.15B	6.33%	NT$4.25B	NT$3.24B	NT$111M	NT$3.13B	NT$2.58B
4	NT$67.48B	6.41%	NT$4.33B	NT$3.30B	NT$112M	NT$3.19B	NT$2.46B
5	NT$67.82B	6.50%	NT$4.41B	NT$3.36B	NT$112M	NT$3.25B	NT$2.35B
6	NT$68.36B	6.50%	NT$4.44B	NT$3.42B	NT$181M	NT$3.24B	NT$2.20B
7	NT$69.11B	6.50%	NT$4.49B	NT$3.49B	NT$251M	NT$3.24B	NT$2.06B
8	NT$70.08B	6.50%	NT$4.56B	NT$3.58B	NT$323M	NT$3.25B	NT$1.95B
9	NT$71.27B	6.50%	NT$4.63B	NT$3.67B	NT$397M	NT$3.27B	NT$1.84B
10	NT$72.70B	6.50%	NT$4.73B	NT$3.78B	NT$475M	NT$3.31B	NT$1.75B

Methodology & flags

Damodaran FCFF DCF, 10y explicit + perpetuity, in TWD. Risk-free 2.31% (local-currency government bond). Synthetic credit Aaa/AAA. CRP 0.40% from revenue-weighted country mix × Damodaran 2026 CRPs. Terminal ROIC faded to 10.00%; 5% governance haircut applied post-DCF. Monte Carlo: 1000 iterations. Engine v1.0.0 · result: /Users/valentin/Documents/notes_jiliac_labs/Finance/Damodaran/valuations/simplo/output/2026-06-04-result.json

R&D cap: OFF · Lease cap: OFF · Failure: OFF · ESO: OFF
Governance haircut: 5% applied post-DCF (NT$664.79 > NT$631.55)
Sensitivity tornado: not run

FCFF DCF · dark theme v3. Generated 2026-06-04T15:08:40 from /Users/valentin/Documents/notes_jiliac_labs/Finance/Damodaran/valuations/simplo/output/2026-06-04-result.json.