Full Report

Industry — Understand the Playing Field

Figures converted from HKD at historical FX rates — see data/company.json.fx_rates. Ratios, margins, and multiples are unitless and unchanged.

1. Industry in One Page

Ten Pao sits in the external power-supply and smart-controller corner of Electrical Equipment & Parts: the boxes, bricks, adapters, PCBA controllers and chargers that convert grid AC into the specific DC that a router, a power tool, a TV, a data-center server, an e-bike or a home battery actually consumes. It is a contract-manufacturer business — the end consumer almost never sees the maker's name. Branded OEMs (Cisco, TTI, Bosch, Samsung, plus increasingly Chinese hyperscalers and AI-server builders) hand over a spec and a price target, and ~10 large Asian ODMs compete to win the multi-year platform.

Money is made by hitting two thresholds at once: low enough unit cost to clear the OEM's bill-of-materials target, and high enough technical bar — efficiency, density, safety certification, modular firmware — to be one of the few suppliers qualified for the platform. The cycle hits first through end-product demand (smartphones, routers, EV charging piles, energy-storage cabinets) and copper, semiconductor and freight inputs, not through pricing power. Gross margins above 20% almost always mean a premium product mix (industrial PSU, AI server PSU, smart controllers) rather than a strong brand.

The thing newcomers usually miss: this is a portfolio of sub-segments stacked on the same factory, not one homogeneous "charger" market. Mobile-phone adapters earn 12–14% segment margin; AI-server and HPC power supplies above 24%; energy-storage and EV chargers currently 8–11% as Chinese capacity floods the market. Where a maker's revenue mix sits across these tiers explains margin more than any cost discipline narrative.

Takeaway: the same factory can earn very different margins depending on which row of this table it sells into. Tracking the mix is the single most useful lens for the rest of the report.

2. How This Industry Makes Money

The revenue model is contract-manufactured units shipped against a customer purchase order, repriced quarterly as commodity costs move. A "design win" — an OEM qualifying a supplier into a multi-year product platform — is the asset that earns the gross profit; once won, the supplier ships against forecasts, books revenue at delivery and absorbs raw-material moves with a quarter or two of lag.

The cost stack is heavy on commodities (copper for transformers and cabling, ferrite cores, aluminum housings, engineering plastics), semiconductors (MOSFETs, controllers, increasingly GaN and SiC switches), and labor / overhead to run high-mix SMT and PCBA lines. Capital intensity sits between contract electronics (Flex, Foxconn at 2–3% of sales) and specialty industrial (Delta at ~6%). Ten Pao's run-rate capex sits around 3–5% of sales as it digests the new Huizhou Intelligent Manufacturing Park and Mexico base.

Pricing unit and bargaining power. Adapters and chargers are priced per unit at a wattage/efficiency spec (e.g. a 65W GaN USB-PD adapter, a 6.6 kW EV onboard charger, a 5 kW AI-server PSU). The OEM is almost always larger than the ODM and dictates price; the supplier's leverage comes from being one of the few qualified for the specific platform, from holding inventory through a shortage, and from owning unique geographic capacity (Mexico for US-bound product to dodge tariffs, Hungary for EU). Switching cost is real but not infinite: re-qualifying a charger platform takes 6–12 months, so OEMs do it when a supplier slips on quality, delivery, or price by enough margin to justify the move.

Where the margin sits. The same kilogram of copper and silicon earns 12% gross at a router adapter and 28% at an AI-server PSU. The premium is paid for density (W per cm³), efficiency (e.g. 96%+ 80-Plus Titanium), thermal management, firmware/controls, and certifications — none of which the consumer ever sees, but all of which gate the OEM's spec sheet. Above $400/W of finished value, suppliers begin to look more like Delta Electronics (32%+ GP) than like a commoditised adapter shop (15% GP).

3. Demand, Supply, and the Cycle

End-demand is derived from the device that needs the power, so the cycle is a weighted average of: smartphones and routers (slow growth, refresh-driven), power tools and outdoor equipment (housing- and consumer-discretionary-sensitive), EV charging (subsidy- and infrastructure-driven), home energy storage (electricity-price- and policy-driven) and AI servers (capex-driven and currently roaring).

Supply behaves the way you expect from Chinese export manufacturing: capacity adds in waves, gets absorbed within two years, and then competes on price. The 2022–2023 mobile-phone weakness and 2024–2025 EV-charging-pile oversupply in China are the two most recent reminders.

Where it shows up first. Volume and gross margin move before reported revenue: copper rallies compress GP within one quarter; a customer-platform loss shows up first as a forecast cut and an inventory build, then a revenue drop two quarters later; an AI-server PSU win shows up first as a backlog comment in the chairman's statement and as raw-material commitments before any revenue is booked. Ten Pao's own FY2022 → FY2023 revenue contraction (US$703M → US$617M, -12%) with a gross-margin rise (16.7% → 18.8%) is a textbook example of "lost low-margin volume, kept high-margin mix" — the readable footprint of an industry-wide reset.

Read across: in 2021 the industry over-shipped on COVID-era consumer electronics and gross margin sagged even as revenue surged. In 2022–23, the cycle reversed: revenue fell, but mix moved toward industrial PSU and smart controllers and gross margin expanded by 200 bps. 2025 is a milder version of the same pattern — revenue up 3%, gross margin down 130 bps as new-energy pricing competes harder.

4. Competitive Structure

The industry is structurally fragmented at the bottom and concentrated at the top. Below roughly US$500M of revenue, hundreds of mostly-Chinese ODMs fight on price for short-cycle adapter business. Above that scale, eight or so listed players hold the multinational OEM accounts that demand global manufacturing, certifications, and IP. Delta Electronics is in a separate league as a vertically integrated power-and-thermal platform; the rest cluster as specialists by end market.

How to read this map. Delta is alone in the top-right (largest revenue, highest margin) — it has graduated from ODM into a vertically integrated power platform with proprietary technology. Lingyi iTech is large but commoditised (low GP, Salcomp-acquired mobile-charger business). Lite-On is the closest "scale plus margin" benchmark for what a successful PSU specialist can become. The middle band (AcBel, Chicony, Ten Pao) competes for the same multinational accounts on a different cost structure; Ten Pao currently earns the highest operating margin in that middle band, which is the only competitive claim worth making at this scale. Phihong is a cautionary tale: a once-respected adapter specialist now in operating losses as it pivots to EV chargers without enough scale.

5. Regulation, Technology, and Rules of the Game

External power supplies are one of the most heavily regulated product categories in consumer and industrial electronics — every adapter must clear safety (UL/IEC), efficiency (DoE Level VI in the US, CoC Tier 2 in the EU), EMC, and now interoperability standards. Add EV-charger safety codes (NEC 625 in the US, IEC 61851 globally), battery-storage codes (UL 9540A), tariff regimes (US Section 301, EU CBAM), and platform mandates (USB-C common charger directive in the EU since December 2024), and the regulatory thicket starts to look like its own competitive moat — for any maker scaled enough to keep up.

6. The Metrics Professionals Watch

Most ratios from a generic equity screen miss what actually moves a power-supply ODM. The right metrics tell you whether the mix is shifting up, whether utilization is high enough to leverage fixed cost, whether the customer base is concentrating risk, and whether geographic re-shoring is paying off.

Reading the heatmap: the share migration from consumer-power into industrial / smart-controller is the underlying story of Ten Pao's 2022–25 margin expansion. Whether this continues — and whether AI-server PSUs become a separately disclosed line — is the single most important industry-level question for the rest of the report.

7. Where Ten Pao Group Holdings Limited Fits

Ten Pao is a mid-scale Asian ODM in transition: still derives almost half its revenue from consumer adapters (router/STB, lighting, telecom) where it competes with hundreds of small PRC shops, but two-thirds of its profit pool now comes from industrial / smart-controller PSU where the customer set is multinational and the margin band is structurally higher. It is the smallest of the credibly global power-supply ODMs — roughly half the size of Chicony Power and AcBel, one-twentieth the size of Delta — but earns above-peer operating margins for its size band and runs a disciplined balance sheet.

8. What to Watch First

Signals that show whether the industry backdrop is improving or deteriorating for Ten Pao — each observable in filings, transcripts, market data, regulation, or industry sources.

Know the Business

Figures converted from HKD at historical FX rates — see data/company.json.fx_rates. Ratios, margins, and multiples are unitless and unchanged.

Ten Pao is a Chinese contract power-supply ODM earning very different gross margins from the same factories: ~24% on smart industrial chargers and AI/HPC server PSUs, ~13% on consumer-electronics adapters, ~8% on EV chargers and energy-storage. At HK$2.91 (US$0.37) it trades at ~7.9× earnings and ~1.4× book against listed peers ranging from 18× to 87× — a discount that compresses three separable issues: HK small-cap with mainland operations, 60% top-5 customer concentration, and a margin mix the market doubts migrates fast enough into industrial/AI-server. The May 2026 announcement that the Huizhou Electronic subsidiary — roughly the smart-charger + new-energy half — is proceeding toward an A-share listing is the capital-structure news that could force a parts-based valuation.

1. How This Business Actually Works

Ten Pao sells the same kind of object — a power-conversion box — into six different markets at six different gross margins. The unit of competition is the design win: an OEM (TTI, Bosch, an unnamed Fortune-500 server builder, an African branded lighting customer, a Chinese EV-charger integrator) hands over a wattage / efficiency / safety spec, multiple Asian ODMs bid, and the winner ships against forecast for 3–7 years. Once a platform is won, revenue is largely volume × commodity cost pass-through; pricing power lives in the qualification asset, not in negotiation room mid-cycle.

Where the real economic differentiation sits in FY2025:

One sentence to remember: the same Dongguan / Huizhou factory earns ~13¢ on every dollar of router adapter, ~24¢ on a smart industrial controller, and ~8¢ on an EV charger — so the only operating metric that genuinely moves group margin is what fraction of revenue is smart chargers + lighting + M&E versus telecom + new energy. Smart chargers stepped from 25.4% margin in FY2023 to 26.7% in FY2024 to 24.1% in FY2025 — that compression, together with new-energy margin falling from 9.3% to 11.3% to 7.7%, is most of why group gross margin dropped 130 bps in FY2025.

The cost stack is commodity-heavy (copper, ferrite, MOSFETs, GaN/SiC for AI-server PSUs) so a quarter of margin moves with the input cycle and a quarter with mix — only the residual reflects management execution. R&D spend lives inside SG&A and is what funded the FY2025 launch of 3,500–10,000W AI/HPC PSUs; those products are how Ten Pao tries to push more revenue into the 24%-margin column.

2. The Playing Field

Ten Pao is the cheapest, smallest, and most under-the-radar of seven listed peers that all make power supplies for some combination of consumer, industrial, EV, and data-center end-markets. The peer set tells you what "good" looks like (Delta Electronics, which has migrated decisively into AI-server PSU) and what the floor is (Phihong, the only one losing money).

What the panel reveals. Among peers with positive ROE, Ten Pao has the second-highest return on equity (17.5%, narrowly ahead of Lite-On at 17.3% and behind Delta at 24.3%) and by far the lowest multiple (7.9× vs 19–88× for the rest). Chicony Power (6412.TW) is the most directly comparable economic substitute — adapter + server PSU, US$1.07B revenue vs Ten Pao's US$0.71B, 16.5% gross margin vs Ten Pao's 18.2%, 14.8% ROE vs 17.5% — and trades at 2.4× the P/E and ~2× the P/B. Delta Electronics is the aspirational peer: it earns 35% gross margin because two-thirds of its sales are industrial / data-center / EV power, not consumer adapters, and the market pays 87× for that mix. The investment question is not whether Ten Pao is cheap on metrics — it visibly is — but whether the mix migrates enough toward Delta and away from Phihong to justify a rerate without the discount-closing event of the Huizhou A-share spin.

3. Is This Business Cyclical?

Moderately, and the cycle hits gross margin first, revenue second. Net margin has swung from 1.7% (FY2018) to 7.5% (FY2016) and back to 6.8% (FY2025) on revenue that grew almost monotonically — meaning the operating picture is far more cyclical than the top line suggests.

Where the cycle actually hits.

• Inputs first. Roughly 45% of cost of sales is copper, ferrite, plastics, PCBs; another 25% is semiconductors. The FY2018 net-margin collapse to 1.7% is the canonical case — copper +27% YoY, MLCC/MOSFET allocation tight, fixed-price OEM contracts. Repeated in milder form in FY2022 (China-shore semis + RMB labour). Mid-cycle gross margin sits 17–19%; trough gross margin is 14% (FY2018).
• End-market mix second. Telecom (router / WiFi) is reasonably steady. Smart chargers / AI-server PSU is hooked to the data-center capex cycle and currently a tailwind. New energy (EV chargers + storage) is the most volatile — Chinese capacity glut pushed segment margin from 11.3% (FY2024) to 7.7% (FY2025), and management is explicitly walking away from lowest-margin orders ("strategic adjustments to optimise customer and product mix … targeting customers and products with lower gross profit margins").
• Working capital third. Receivables $212M and inventory $123M are 47% of revenue combined; that ratio expands in slowdowns and is the channel through which the cycle hits free cash flow. FY2021 saw FCF go negative (–$44M) on capex-plus-WC build into the post-COVID surge.

For a value buyer, the relevant question is whether mid-cycle net margin lives at 6–7% (where it has spent most of the last 6 years) or at 4–5% (FY2017–FY2019 average) once smart-charger margins normalise. The five-year capex digestion at the new Huizhou park (running 5–6% of sales versus a steady-state more like 3%) should support 200–300 bps of operating leverage if revenue grows.

4. The Metrics That Actually Matter

The metrics most investors look at — revenue growth, P/E, dividend yield — are the least useful here. Revenue growth is 3–4% and largely commodity pass-through; P/E is statically cheap because the cycle is real; dividend (4.4% trailing yield) is generous but a result of the spin-off-pending posture, not policy. The single best leading indicator is segment gross margin on Smart Chargers & Controllers — when that line prints above 26%, the next quarter's group net margin rises by 50–100 bps. It dropped from 26.7% to 24.1% in FY2025, which is why the stock dropped 15% on the FY2025 results despite revenue growth.

5. What Is This Business Worth?

The right lens changed in May 2026. Until the May 14 announcement, Ten Pao was best valued as a single Chinese power-supply ODM on EV/EBITDA through the cycle (~5–6× current versus 11–20× peer median = clear discount to peers but rational given HKEX small-cap + customer concentration + cyclicality). After the announcement that the Huizhou Electronic subsidiary will spin off and list on an A-share exchange, a sum-of-the-parts lens is now defensible because the two pieces have materially different multiples available to them: HKEX listed parents trade at 7–10× P/E and Chinese A-share charger / EV-power peers trade at 25–45× P/E.

The mechanical math. At a conservative 20× P/E on Huizhou's ~$28M share of FY2025 net income (scaled to its segment-profit share), the subsidiary alone values at ~$565M — already 1.5× the entire group market cap of $383M. Ten Pao's retained 90%+ stake (the share-award plan covers 9.2% of registered capital) would be worth ~$510M at that multiple. Even at half that valuation, the residual consumer-side listco is effectively free. The variable to track is the spin's actual terms: a successful A-share listing forces a re-mark of Huizhou's economics at A-share multiples; a delayed or failed listing leaves the consolidated HK discount intact.

Three things would reset the valuation conversation:

1. Confirmed A-share filing date and indicative valuation range from the sponsor would compress the discount immediately, irrespective of the eventual listing outcome.
2. Smart-chargers segment margin holding above 25% through FY2026 — this is the operational evidence that the AI/HPC PSU mix is sustainable, not a 2024 one-off.
3. Net-energy segment margin rebuilding to 10%+ — would validate management's "walk away from low-margin orders" strategy and remove the one piece of segment evidence that currently looks worst.

Without those, the stock sets up as a 7.9× P/E cyclical with a free option on the spin-off. With them, the setup is a sum-of-parts where the parts sum to materially more than the whole.

6. What I'd Tell a Young Analyst

Don't anchor to consolidated metrics. Group gross margin (18.2%) is a numerical average of an industrial business earning 24%, a consumer business at 13–25%, and a new-energy business at 7.7%. The investment thesis is entirely about which row you are looking at — when you hear "Ten Pao's margins are compressed," the right reflex is "in which segment, and is that segment the one being spun out?"

Watch the Huizhou Electronic spin-off paper trail. HKEX cleared the proposal under Practice Note 15 on May 14 2026 — first hurdle, not last. The remaining gates are (i) CSRC acceptance of the A-share application, (ii) sponsor pricing range, (iii) lock-up structure for the parent's retained stake. Track those in HKEX filings and Practice Note 15 disclosures. The catalyst that resets the multiple is not completion of the listing but the valuation range disclosed at the application stage — that is when the SOTP arithmetic becomes consensus.

Customer concentration is the easiest way to be wrong. A single customer is 15.1% of revenue and the top five are 59.8%. The FY2025 disclosure mentions Fortune-500 server customers as the new pillar; the FY2024 disclosure was about WiFi 6/7 router customers; the FY2023 about EV-charger OEMs. The customers rotate but the concentration doesn't drop. The risk is not a single customer leaving — it is a qualification slip (yield, delivery, or pricing) on a multi-year platform, which compounds across the top-5 because they all share certain shared suppliers and ODM panel notes. Read the going-concern items and the "key audit matters" in the next annual report for any sign of receivables provisioning rising sharply on a single customer line.

The cheapness is partly real and partly a discount that can close. HKEX-listed small-cap mainland-operations companies trade at a structural discount of 30–40% versus A-share peers; that piece won't close without the spin-off. The cyclical piece (input costs) and the customer-concentration piece are real and won't close without operational evidence. Distinguish which piece of the discount you are getting paid for.

What would change the thesis. (i) The Huizhou spin-off is withdrawn or rejected — kill the SOTP case and value the whole company at 8–10× P/E. (ii) Smart-charger segment margin falls below 22% — the AI/HPC mix shift isn't happening fast enough, value on through-cycle 5% net margin. (iii) Receivables / revenue rises above 35% (currently 30%) — channel stuffing or customer trouble, value on book value at 1.0× P/B. (iv) Chairman Hung Kwong Yee (67) succession — he's the founder and the strategic decisions that brought the company to this point (Mexico, Hungary, AI-PSU, the spin-off) all bear his signature.

---

Figures from data/financials/income.json, segment.json, ratios.json, balance_sheet.json, FY2025 annual report (data/extracted/annual_reports/FY2025_annual_report.md), FY2025 results announcement (PRNewswire, 20 March 2026), spin-off announcement (HKEX / BigGo, 14 May 2026), and peer set from data/competition/peer_valuations.json (Yahoo Finance, as of 21 May 2026).

Figures converted from Hong Kong dollars at historical FX rates — see data/company.json.fx_rates. Ratios, margins, and multiples are unitless and unchanged.

Long-Term Thesis — 5-to-10-Year View

1. Long-Term Thesis in One Page

The long-term thesis is that Ten Pao is a founder-disciplined Asian power-supply ODM in mid-migration from a 13%-margin consumer-adapter business into a 24%-margin industrial / AI-HPC PSU business, and that over a 5-to-10-year window the protected slice of revenue (smart chargers + lighting + media at ~55% of sales today) compounds faster than the unprotected slice (telecom + new-energy) erodes. The 5-to-10-year case works only if three independent things hold: (i) smart-chargers segment gross margin defends 23–25% as the AI/HPC 3.5–10 kW PSU program qualifies a Tier-1 hyperscaler, (ii) the Mexico/Vietnam/Hungary footprint converts from disclosed-but-unquantified to a measurable 15%+ ex-PRC revenue line that earns a tariff premium, and (iii) Chairman Hung's "walk away from low-margin orders" discipline survives the daughter-led succession that is already staged. This is not a long-duration compounder if any of those three breaks; it is a narrow-moat cyclical ODM with a free option on the Huizhou A-share spin-off. The investor's job over a multi-year horizon is to monitor the protected-slice gross-profit pool, the ex-PRC revenue disclosure, and the related-party leakage trend — not the next quarter's revenue print.

The compounding math is unusually transparent for a HK small-cap: a nine-year average ROE of 22.0% (still 17.5% in FY2025), revenue growth around 10–11% per year over 13 years, capex now digesting after the Huizhou park completion, and a founder who has not sold a share since 2015 sit on top of a 7.9× P/E and a 1.4× book multiple. The asymmetry is in the discount, not the business — but the discount only closes if the protected slice keeps earning its way and the founder family does not extract the upside through the spin-off.

2. The 5-to-10-Year Underwriting Map

Driver #1 — the mix migration into 24%-margin industrial PSU — matters most. It is the single observable variable that translates Ten Pao from a 7.9× cyclical OEM into a low-teens compounder. The other five drivers are necessary conditions; this one is sufficient. If smart-chargers segment GM holds 23–25% for five consecutive years and segment share crosses 45% of revenue, group gross margin clears 20% and the case for a structural multiple rerate writes itself, independent of the Huizhou spin or the succession.

3. Compounding Path

Revenue has compounded at roughly 11% per year over the past 12 years, and ROE has averaged 22% over the last nine. Through the 5-to-10-year window, the path to owner value runs through three mechanical steps: keep growing revenue at the high-single-digit level the industrial mix supports, lift group gross margin from 18% to 20% as the protected slice expands, and let capex digestion convert that into 50%+ FCF conversion as the Huizhou park stops absorbing cash.

The picture matters because it shows the engine working through two completed cycles already. FY2018 was the trough — net income collapsed 65% to $7.0M on a copper-cost shock and the ROE bottomed at 9.5%; revenue still grew. FY2021 was the consumer-electronics peak — net income hit $48.6M on pandemic-era device demand that did not repeat; ROE 28.9%. FY2023 was the destock trough — revenue fell 12% YoY but gross margin expanded 200 bps as the company walked away from low-margin telecom volume; ROE held 19.9%. Through all three, revenue compounded; the variable was margin, and the variable that moved margin was which segment was selling. That is the rhythm to underwrite forward.

The compounding math is asymmetric. Even the base case — which assumes nothing about AI/HPC heroics or the spin-off — models $0.07 of FY2030 EPS at an 11× exit ($0.78 against $0.37, ~16%/yr before dividends, scenario-conditional). The bull case requires evidence currently absent (named AI customer, clean spin economics) but is structurally available if the protected slice keeps earning. The bear case is essentially the FY2018 trough re-run at scale, layered on succession risk. The asymmetry reflects that this is operationally a moat-narrow, finite-runway compounder but valuation-wise a forced-discount situation.

The single most important compounding variable is gross margin — not revenue growth. Adding 200 bps to group GM (18.2% → 20.0%) on FY2025 revenue alone adds roughly $0.013 of EPS at the current effective tax rate. That is more incremental EPS than three years of base-case revenue growth produces without margin expansion. Investors who model this name on "revenue compounds at 10%" miss the lever; the lever is the segment mix.

4. Durability and Moat Tests

A 5-to-10-year thesis only matters if the moat survives the stress tests it has not yet been put through. Two have already been passed — the FY2018 copper-cost shock and the FY2022 smartphone-driven revenue cliff — and the ROE rebuild in both cases is the evidence base for the thesis. Three remain open.

Two tests have been passed already. The FY2018 raw-material shock collapsed net margin to 1.7%; ROE rebuilt to 28.9% by FY2020. The FY2022 smartphone-driven cliff cut revenue 14%; gross margin expanded 200 bps as the company walked away from low-margin volume. Both required no balance-sheet rescue, no equity issuance, no covenant breach. That history is the strongest quantitative evidence the moat exists — a moat that bends, in the language of the Moat tab — and is the reason the thesis can underwrite a 5-to-10-year holding period despite the OEM cyclicality.

5. Management and Capital Allocation Over a Cycle

Founder Hung Kwong Yee built the business in 1979, took it public in 2015, has not sold a share since, and owns 66.78%. The 11-year listed track record — five geographies opened (Hungary 2017, Sichuan/Vietnam 2019, Mexico 2025, with Huizhou Park 2024–25), one major segment created from zero (new energy, 2020), one product platform launched in real time (AI/HPC PSU, H2 2024) — is operationally credible. Story tab gives credibility a 7/10: operational promises are met on roughly the timelines stated, and strategic pivots are telegraphed before being made. That is rare in a HK-listed family business.

What the capital-allocation file actually shows over a full cycle, and what it implies for the long term, is more nuanced:

The clean read on management. On operational execution, this is a B+ founder-operator: every plant said-and-built, every segment built that was promised, mix discipline visible through three cycles, balance sheet net cash, no equity issuance, no covenant trouble, no SFC inquiry, no auditor change. On capital allocation to outside shareholders, this is closer to a C: no buybacks at $0.26, related-party procurement scaling exactly as the spin is staged, daughter on the Nomination Committee that nominates her own appointments, INED tenure 10+ years until the listing-rule cap forces refresh in 2026, and a pre-IPO subsidiary share-award scheme that pre-dates the announced spin by two years. The long-term thesis depends on the second part not deteriorating further; the Golden Ocean Copper cap trajectory ($6.4M → $18.0M in two years) and the Huizhou spin terms are the two specific things that signal whether outside shareholders' compounding is captured or transferred.

6. Failure Modes

The 5-to-10-year thesis has four distinct failure modes. Each is independent — any one triggering would force a re-rate, and any two simultaneously would invalidate the durable-compounder frame entirely.

The Huizhou spin and the mix migration are the two failure modes that matter most. They are not independent — a clean spin that lists Huizhou at A-share multiples is partly conditional on the smart-charger margin holding — but they are observable on different timelines (12–18 months for the spin disclosure, 24–36 months for the segment margin trend), which means a long-term investor can underwrite the position from one signal to the next rather than waiting for both to land simultaneously.

7. What To Watch Over Years, Not Just Quarters

The first of these signals to print is the Huizhou A-share prospectus disclosure of post-IPO economic interest retained by 1979.HK — likely sometime between Q3 2026 and Q2 2027 depending on CSRC processing time. The most durable of the five is the smart-chargers segment GM holding 23–25% with AI/HPC PSU revenue separately disclosed — that is the multi-year evidence that the mix migration is a structural mechanic rather than a cyclical print.

The long-term thesis changes most if the Huizhou A-share prospectus discloses ≥85% economic interest retained by 1979.HK at a pricing range above 25× P/E and the FY2026 H1 interim prints smart-chargers segment GM at or above 25% with a named Tier-1 hyperscaler customer for the 3.5–10 kW AI/HPC PSU program — both conditions together convert this from a narrow-moat cyclical ODM at a forced discount into a structurally compounding industrial-PSU specialist whose 7.9× P/E is the cheapest exposure to AI-server power infrastructure on the HKEX board.

---

Sources: business-claude.md, moat-claude.md, industry-claude.md, numbers-claude.md, forensics-claude.md, people-claude.md, story-claude.md, competition-claude.md, bull-claude.md, bear-claude.md, verdict-claude.md, short-interest-claude.md, technicals-claude.md, research-claude.md. Underlying data from FY2025 annual report (Ten Pao Group, PRNewswire 2026-03-20, HKEX), spin-off announcement (HKEX / BigGo, 14 May 2026), peer financials (Yahoo Finance, 21 May 2026). All figures in US dollars unless noted; converted from HKD at historical FX rates.

Competition - Who Can Hurt Ten Pao, Who It Can Beat

Figures converted from HKD at historical FX rates - see data/company.json.fx_rates. Ratios, margins, and multiples are unitless and unchanged.

Competitive Bottom Line

Ten Pao competes in a fragmented external-power-supply industry with a clear premium tier (AI-server PSU, smart industrial controllers) and a commoditised tier (mobile and consumer adapters, undifferentiated EV chargers). Its moat is narrow but real: the highest operating margin in the under-US$2B revenue band, multi-region manufacturing footprint (Dongguan + Huizhou + Mexico + Vietnam + Hungary) that most size-matched peers lack, and a 38.8% industrial / smart-charger mix that has steadily compounded for three years. The most dangerous competitor is Delta Electronics (2308.TW) — not because Delta will compete for Ten Pao's router-adapter contracts, but because Delta defines the AI-server PSU benchmark Ten Pao is trying to reach with its 3,500-10,000W product series, and if Delta keeps moving down-stack into mid-power AI PSU before Ten Pao moves up, the whole "premium mix migration" thesis stalls. Phihong (2457.TW) is the cautionary mirror: same starting point ten years ago, pivoted hard into EV chargers, now losing money.

The Right Peer Set

Ten Pao competes in external power supplies and chargers sold to OEMs as a contract design and manufacturing service. The right comparators are listed Asian ODMs whose factories actually pour copper into transformers for the same end-markets (consumer adapters, industrial PSU, EV chargers, energy storage, server PSU). The peer panel below is not the broad "Electrical Equipment & Parts" Morningstar bucket — Schneider, Eaton, ABB and Siemens earn their margins on grid switchgear and process automation, not on adapter platforms — but the seven Asia-listed power-supply specialists that bid for the same multi-year OEM platforms.

Why these five-plus-one and not others. Salcomp is the most direct product substitute for Ten Pao's mobile adapter line but is private since its 2019 acquisition by Lingyi — that exposure is captured via 002600.SZ. Sungrow Power and GoodWe (Chinese inverter / energy-storage specialists) would distort the peer set because Ten Pao's new-energy segment is only 17% of revenue. Western mega-cap industrials (Schneider, Siemens, Eaton, ABB) appear in Morningstar's auto-generated peer panel for 1979.HK but are wrong economics: grid switchgear and process automation, not contract power-supply manufacturing. Anker is a consumer brand, not a contract ODM. Mean Well, FSP Group, TDK-Lambda are private. Lite-On is auxiliary because it is larger and broader than ideal (40% of revenue is opto and automotive) but is the closest "scale plus margin" benchmark for what a successful PSU specialist looks like at the top of the cycle.

Read the map. Delta sits alone in the top-right (the only peer that has graduated from ODM into a vertically integrated power-and-thermal platform). Lite-On has the second-best size-and-margin combination. Lingyi is large but commoditised on Salcomp economics. The lower band (AcBel, Chicony, Ten Pao, Phihong) compete for the same multinational OEM accounts: Ten Pao earns the highest operating margin in that lower band at the smallest revenue scale, and trades at the lowest multiple. The visible gap between Ten Pao (7.6% op margin, US$383M market cap) and Chicony (4.3% op margin, US$1.23B market cap) is the easiest-to-articulate piece of the mispricing thesis.

Where The Company Wins

Four advantages with evidence behind each.

One sentence. Ten Pao earns near-Lite-On operating margin (7.6% vs 9.4%) and Delta-class ROE (17.5% vs Delta's 24.3% — narrowly above Lite-On's 17.3%) at a multiple cheaper than Phihong (loss-making). The discount is partly structural (HK small-cap, mainland operations) and partly waiting on the Huizhou A-share spin to crystallise.

Where Competitors Are Better

Four areas where Ten Pao demonstrably lags, with the named beneficiary.

The Delta gap is the one that decides the rerate. Delta's FY2024 chairman letter describes 33 kW Open Rack v3 power shelves delivering 83% more output than the 2023 18 kW model and 97.5% efficiency — production-ready, hyperscaler-approved, shipping in volume. Ten Pao's analogous product (3,500-10,000W series) is at the "qualification" stage and has no disclosed customer name, no revenue split, and no order book commentary. If Delta locks down the 3-10 kW mid-power band as well as it locked the 30 kW+ rack-shelf band, Ten Pao's premium-mix migration thesis loses its single biggest accelerant.

The Phihong precedent is the warning. Phihong is the closest size-match peer (US$305M revenue vs Ten Pao US$714M) and the closest historical analogue: an established adapter specialist that pivoted into EV chargers. In the FY2025 results, Phihong's EV product mix climbed from 31.7% (Q1-Q3 2024) to 40.9% (Q1-Q3 2025) — exactly the kind of "new energy growth" Ten Pao narrates — yet gross margin collapsed from 28.0% to 21.0% and operating result swung from -0.9% to -5.9%. This is what happens when an ODM chases EV-charger volume without owning the integrator brand. Ten Pao's "walk away from lowest-margin EV orders" stance in the FY2025 MD&A reads as direct acknowledgment of this risk.

Threat Map

Six concrete threats, scored by severity and timing. The first two are the ones that decide whether the next 24 months are a rerate or a derate.

Moat Watchpoints

The five signals that tell an investor whether Ten Pao's competitive position is strengthening or weakening. Every one is observable in filings, investor conferences, or peer guidance — no proprietary access required.

---

Sources: peer financials from data/competition/peer_valuations.json (Yahoo Finance, snapshot 2026-05-21); peer annual reports in data/competitors/{ticker}/annual_report/; Phihong Dec 2025 investor conference in data/competitors/2457_TW/presentations/; Delta FY2024 chairman letter; Lingyi FY2025 annual report; Industry framework from industry-claude.md; Ten Pao segment data from Warren's business-claude.md and the FY2025 annual report.

Current Setup & Catalysts

Figures converted from Hong Kong dollars at historical FX rates — see data/company.json.fx_rates. Ratios, margins, multiples, share counts, dates, and names are unitless or non-currency and unchanged.

1. Current Setup in One Page

The stock is trading around $0.37 after a +78.5% 1-year run that culminated in a +21% breakout on 14 May 2026 — the day HKEX confirmed the company may proceed under Practice Note 15 with the proposed A-share spin-off of Ten Pao Electronics (Huizhou) — and a same-week all-time high of $0.43. The market is mostly watching three things over the next six months: (i) the disclosure that will come out of the CSRC A-share application (not yet filed; the 27 April 2026 filing was the PN15 application with HKEX) — specifically the listed parent's post-IPO economic interest and the sponsor's indicative pricing range; (ii) the FY2026 H1 interim results due around 22 August 2026, where smart-chargers segment gross margin (24.1% in FY2025, down from 26.7%) is the single mix-shift number that matters; and (iii) the 12 June 2026 AGM, where shareholders vote on a 10% buyback mandate that, if actually used at 7.9× P/E, would be the cleanest possible refutation of the bear's value-leak thesis. The setup is constructively bullish but stretched — 30-day realised volatility sits in the 80th-percentile stressed band and the calendar between mid-June and late-August is dense with thesis-defining disclosures.

2. What Changed in the Last 3-6 Months

The narrative arc over the last 3-6 months. In March the market got the FY2025 print — a margin disappointment that confirmed gross margin compresses even when revenue grows. The stock spent April in a corrective drift (RSI down to 39, MACD negative six straight weeks). Then on 14 May the spin-off filing reset the frame entirely: the question shifted from "is the mix migration real" to "will the listed parent retain the economic interest of the slice that is being repriced at A-share multiples." The stock cleared an 18-month base on real volume and the 50-day average turnover doubled. As of 21 May 2026 the market is paying for a re-rate that has not yet been earned at the listed-co level — that is the unresolved tension into the AGM and the FY2026 H1 interim.

3. What the Market Is Watching Now

The live debate has tightened around five concrete observables. Each shows up in a hard or soft disclosure inside the next six months — none of them is a vibes question.

1. Smart-chargers segment gross margin — the engine, currently 24.1%. The bear's primary trigger is a print below 22% in the FY2026 H1 interim; the bull's defence is the H1 2025 op-margin record of 8.6% combined with the AI/HPC product launch. Bear-confirming: segment GM at 22–23% in H1 2026 with no named AI customer. Bull-confirming: segment GM at 25%+ with a quantified AI/HPC PSU revenue line.

2. Huizhou A-share spin-off terms — the central catalyst. Watch the post-IPO economic interest retained by 1979.HK (target ≥85%), the sponsor's indicative pricing range (target ≥25× P/E to match A-share electrical-equipment comparables), and any related-party carve-outs into SpinCo. Bear-confirming: retained interest below 85% or pricing below 20× or material RPT carve-outs. Bull-confirming: retained interest ≥85% at 25× P/E or higher, mirroring A-share charger peers.

3. AI/HPC PSU customer naming — the slogan-to-revenue test. Single-customer concentration is already 15.1% of revenue; the name has never been disclosed. Bear-confirming: continued silence on the customer name through the FY2026 audited results in March 2027. Bull-confirming: a named Tier-1 hyperscaler or Fortune-500 server-OEM with a quantified FY2026 revenue contribution.

4. The 12 June 2026 AGM buyback mandate — and whether it is actually used — the capital-allocation tell. The mandate authorises repurchases of up to 10% of issued shares; whether anything is actually bought at 7.9× P/E is the question. Bear-confirming: mandate approved but never used through end-2026 while related-party caps stay elevated. Bull-confirming: any visible buyback execution while the spin-off is being staged.

5. New-energy segment trajectory — the Phihong analogue. Segment revenue fell 10% in FY2025 and segment GM halved from 11.3% to 7.7%. Management calls it a deliberate exit from low-margin projects; the bear calls it Phihong's trajectory (28% GM → 21% → operating losses). Bear-confirming: segment GM below 6% in FY2026 H1 with revenue still falling. Bull-confirming: segment GM stabilising at 8–10% with revenue flat or rising on the Thailand two-wheeled-vehicle and Southeast Asia projects management has flagged.

4. Ranked Catalyst Timeline

The ranking matters because two events — the FY2026 H1 interim (~22 Aug 2026) and the CSRC prospectus disclosure — together carry more decision weight than the other eight combined. The interim is the first observable margin test of the AI/HPC pivot since the breakout; the CSRC disclosure is the binary alignment test between listed-co shareholders and the controlling family. The AGM is procedural in form but the buyback-execution question is the single non-financial signal that would refute the bear's value-leak thesis at low cost. Everything else is either confirmatory or process-driven.

5. Impact Matrix

The four catalysts at the top of this matrix — the H1 interim, the CSRC prospectus disclosure, the buyback-mandate execution, and the Golden Ocean Copper purchase trajectory — collectively answer the only investment question that matters: whether outside shareholders capture the value the mix migration is creating, or watch it flow to the controlling family. Items five and six (the AI customer naming and the new-energy GM trajectory) are confirmatory rather than decisive on their own — but they are the inputs the H1 interim will reveal first.

6. Next 90 Days

The 90-day window from 21 May 2026 runs to roughly 19 August 2026. Three hard-dated events fall inside it, and a fourth — the FY2026 H1 interim — sits just past it (the prior-year interim was published 22 August 2025).

7. What Would Change the View

Three observable signals, taken in sequence, would most change the investment debate over the next six months. First, the FY2026 H1 interim segment-information note on 22 August 2026 — if smart-chargers segment gross margin prints at or above 25% with the AI/HPC PSU revenue separately broken out and a named customer disclosed, the long-term thesis Driver #1 (mix migration into 24%-margin industrial PSU) is validated quantitatively, the bear's "margin compression on +3% revenue" case from FY2025 is reframed as a single-year comp problem, and the Bull's $0.64 SOTP target moves from theoretical to underwritable. Second, the CSRC prospectus disclosure of the post-IPO economic interest retained by 1979.HK — anything ≥85% at a sponsor pricing range above 25× P/E converts the holding-company discount into an arithmetic problem with a known answer, and is the single piece of evidence that would force consensus to value the moat-rich Huizhou half of the business at A-share rather than HKEX multiples. Third, parent-level buyback execution under the 10% mandate authorised at the 12 June AGM — even a small repurchase at the current 7.9× P/E refutes the bear's structural concern that capital is being routed to the family supplier chain (Golden Ocean Copper FY2026 cap $18M) and the subsidiary share-award scheme ($3.3M FY2025 value) rather than back to listed-co shareholders. The reverse of any of these three would crystallise the bear case faster than the next earnings cycle, and the simultaneous failure of two would force the position back to the FY2018-style trough multiple from the financials tab. This is the event path that updates the long-term thesis — Stan's verdict adjudicates which path the evidence is on; this tab simply names the disclosures that get the evidence onto the table.

---

Sources: catalysts-spin-off announcement PDF (HKEX inside information, 14 May 2026, Stock Code 1979); FY2025 annual report (PRNewswire 302719944, 20 Mar 2026); FY2025 chairman's statement transcript; TipRanks 2026 AGM filing summary (12 June 2026 convened, buyback mandate, board refresh); aastocks.com NOW.1524666; futunn.com ex-dividend notice (22 June 2026, $0.0084, paid 17 August 2026); H1 2025 interim release (PRNewswire, 22 August 2025); HK Electronics Fair release (PRNewswire 302584209, 15 October 2025); upstream specialist files (long-term-thesis-claude.md, bull-claude.md, bear-claude.md, short-interest-claude.md, technicals-claude.md, numbers-claude.md, forensics-claude.md, people-claude.md, research-claude.md, story-claude.md, business-claude.md, industry-claude.md, competition-claude.md, moat-claude.md). Figures converted from Hong Kong dollars at historical FX rates per data/company.json.fx_rates; price reference 21 May 2026 close $0.37.

Figures converted from HKD at historical FX rates — see data/company.json.fx_rates. Ratios, margins, and multiples are unitless and unchanged.

Bull and Bear

Verdict: Watchlist — the decisive variable is the post-IPO economic interest the listed parent retains in Huizhou, and that number does not yet exist in any public document. Bull's sum-of-parts arithmetic and Bear's value-leak channel both read the same Huizhou A-share spin-off, the same 24.1% smart-charger margin, and the same $13M of related-party flow — but draw opposite conclusions because the prospectus has not yet shown how the value transfers. Bull's quality-to-price gap is real and the regulatory clock is running. Bear's broken FY2025 cash conversion (FCF $2.4M against $16M dividends paid) and the Golden Ocean Copper cap expansion to $18M are equally real. The single piece of evidence that resolves this debate is the CSRC prospectus disclosure of Huizhou's post-IPO economic interest retained by 1979.HK plus the FY2026 H1 smart-charger segment gross margin. Until either prints, this is a position to size at zero and re-underwrite on disclosure.

Bull Case

Bull scenario fair value: $0.64 over 12-18 months (~+72% implied; scenario, not a forecast) on sum-of-parts conditional on the Huizhou A-share listing — Huizhou at 20× FY2025 NI share (~$28M × 20 = $570M; parent's ~90% stake worth $510M) plus residual consumer parent at 8× consumer earnings, on 1,030.4M shares. Inside Morningstar's published 5-Star band of $0.71. Primary catalyst: CSRC acceptance and sponsor indicative pricing range — the valuation range disclosed at application stage is what would re-anchor consensus to A-share multiples versus the parent's 7.9×. Disconfirming signal: withdrawal, rejection, or 12-month delay of the Huizhou A-share application combined with smart-chargers segment GM printing below 25% at FY2026 H1; either alone weakens, both together invalidate the SOTP thesis.

Bear Case

Bear scenario fair value: $0.23 over 12-18 months (~-38% implied; scenario, not a forecast) on multiple compression to 5.5-6.0× P/E (cyclical OEM trough, consistent with FY2018 multiple-low) applied to normalised FY2026E EPS of $0.039 — strips ETR step-down by re-applying a 15% effective rate (vs reported 10.75%) and 50 bps further smart-charger GM compression. Cross-checked at 1.0× P/B on $0.27/share book value. Primary trigger: FY2026 H1 interim prints (i) group GM below 17% and/or (ii) smart-chargers segment GM below 22% and (iii) continued silence on the AI/HPC PSU customer line; layered on Huizhou prospectus disclosed minority valuation coming in below 25× P/E or listco retaining less than 85% economic interest. Cover signal: named Tier-1 hyperscaler customer disclosed for the 3.5-10kW AI/HPC PSU program with quantified FY2026 revenue contribution and smart-charger segment GM holding ≥25% at FY2026 H1 — both together; either alone is insufficient.

The Real Debate

Three real tensions, each anchored to a specific shared fact both advocates accept.

Verdict

Watchlist. Neither side wins outright because the decisive variable in tension #1 — the post-IPO economic interest the listed parent retains in Huizhou — is not disclosed in any document that yet exists. Bull's quality-to-price arithmetic is correct and the regulatory clock is genuinely running; Bear's value-leak channel (Golden Ocean Copper cap to $18M, Huizhou Share Award Scheme adopted before the May 2026 spin announcement) is equally concrete and material. The single most important tension is tension #1: a clean CSRC prospectus showing ≥85% economic interest retained and ≥25× indicative pricing converts the 7.9× P/E into a sum-of-parts gift; a prospectus that routes value to the subsidiary share scheme or carves out RPT-heavy cash flows turns Ten Pao into a textbook controlling-shareholder extraction story. Bear could still be right even on a clean prospectus if FY2026 H1 prints smart-charger GM below 22% and AI/HPC stays a slogan — that is a durable thesis breaker independent of the spin. The condition that flips this to Lean Long: CSRC prospectus disclosure of ≥85% post-IPO economic interest in Huizhou and FY2026 H1 smart-chargers segment GM ≥25% — the durable variable is the segment margin; the spin disclosure is the near-term evidence marker that re-rates the multiple.

Moat — What Protects This Business, If Anything

Figures converted from Hong Kong dollars at historical FX rates — see data/company.json.fx_rates. Ratios, margins, and multiples are unitless and unchanged.

1. Moat in One Page

Conclusion: narrow moat. Ten Pao earns durably above-peer returns inside one specific size band — Asian power-supply ODMs with under $2B of revenue — but the protection is shallow. The advantage is not a brand, a patent thicket, or network effects. It is a multi-decade qualification asset across roughly 40 OEM platforms combined with a five-country manufacturing footprint (Dongguan + Huizhou + Sichuan + Vietnam + Hungary + Mexico) that competitors at the same scale simply do not have. Both work. Neither is permanent. A moat — defined as a durable economic advantage that lets a company protect returns above its cost of capital better than competitors — is real here, but narrow and conditional on management not chasing low-margin EV-charger volume the way Phihong did.

A beginner investor should leave this page knowing three things. First, the evidence the moat exists: 17.5% return on equity (the share of profit earned on each dollar of book equity) versus a sub-$2B peer median around 9%, and a 7.6% operating margin that is the second-highest in the seven-name Asian PSU peer set, beaten only by Delta Electronics at 5× the revenue. Second, the evidence the moat is narrow: the customer-side pricing is set by Fortune Global 500 OEMs (top-5 customers are 59.8% of revenue), gross margin slipped 130 bps in FY2025 with no announced customer loss, and the new-energy segment compressed from 11.3% to 7.7% segment gross margin in a single year. Third, what the moat does not contain: there is no brand the end customer ever sees, no network effect, no software platform, no patent monopoly, no regulatory licence — none of the classic high-quality moats. What remains is qualification, geography, and cost discipline. Lose any one and the moat closes.

2. Sources of Advantage

A moat is not an adjective. It is a specific economic mechanism by which the company makes more money than competitors trying to do the same thing. Below are the five candidates the evidence supports — and three that often get claimed but are not evidenced here.

A few terms beginners should anchor. Switching costs — the time, money, engineering effort, and risk a customer faces in moving to a different supplier. For Ten Pao's OEM customers, switching costs are real but bounded: re-qualifying a new power-supply ODM takes 6–12 months of design, sampling, safety certification, and pilot production. Qualification asset — the contractual and reputational footprint that comes from having already passed those tests for a specific platform. It is the closest thing this business has to inventory of trust.

3. Evidence the Moat Works

The test of a moat is not what management says — it is what shows up in the numbers. Below are seven evidence items, four that support the moat and three that refute or qualify it.

Net read. Four of seven evidence items support a narrow moat with high or medium confidence; two refute the width of the moat (it does not extend into new-energy and Ten Pao does not have pricing power against OEMs); one is mixed. That balance is exactly what a narrow moat looks like in evidence: real, but bounded.

4. Where the Moat Is Weak or Unproven

The moat case has four fragilities. Each is independent — any one breaking would force a re-rating to a "no moat" frame.

Three claims that often get made for this company and that the evidence does not yet support. First, that the Mexico/Hungary/Vietnam footprint is a structural tariff advantage — it might be, but the FY2025 report does not disclose ex-PRC revenue or capacity share, so the claim is unverified. Second, that AI/HPC PSUs are a near-term moat extension — they are not yet, because no customer name or revenue contribution is disclosed. Third, that the Huizhou A-share spin-off proves the parts are worth more than the whole — the spin is a valuation event, not a moat event; the underlying economics are unchanged by the listing structure.

5. Moat vs Competitors

The peer set is six listed Asian power-supply ODMs plus Ten Pao. The right comparison is not "Ten Pao vs Delta" — Delta has graduated into a vertically integrated power-plus-thermal platform with a distinct moat (AI-server PSU + Open Compute Project leadership) that is unattainable at Ten Pao's revenue scale. The right comparison is Ten Pao vs its size-matched peer set (Chicony, AcBel, Phihong) plus the aspirational Lite-On benchmark.

Read the map. Delta sits alone in the upper-right with a moat the rest of the panel cannot reach. Ten Pao occupies the second-best (op margin, ROE) corner of the sub-$1B band — visibly better than Chicony, AcBel, and Phihong despite being smaller than each. That gap is the case for a narrow moat. The fact that the gap exists at sub-$1B scale, where structural advantages are usually harder to defend, is the positive evidence; the fact that Ten Pao still cannot reach Delta's quadrant is the negative evidence.

6. Durability Under Stress

A moat is only worth pricing if it survives stress. The table below tests six stress scenarios against Ten Pao's actual or predicted response. Two of these have already happened (FY2018 raw-material trough, FY2022 smartphone slowdown) — the others are speculative.

The two stress cases that have already been tested (raw-material shock 2018, smartphone-driven revenue cliff 2022) both showed the same pattern: ROE bottomed in single digits, revenue recovered within 2 years, mix shifted toward higher-margin segments, and the company kept building factories instead of cutting capex. That is a moat that bends. The two cases that have not been tested (AI-PSU competitive squeeze, founder succession) are the open questions.

7. Where Ten Pao Group Holdings Limited Fits

The moat is not evenly distributed across the company. This is the single most important nuance for sizing the position correctly.

A clean read. Roughly 55% of FY2025 revenue (Smart chargers + Lighting + M&E) sits in segments where Ten Pao has a narrow moat — combined gross profit ~$90M of $130M group GP, or 69% of group profit pool from 55% of revenue. The remaining 45% of revenue (Telecom + New Energy + Others) is commodity or commoditising. That distribution means the Huizhou A-share spin-off — which carves out roughly the smart-charger + new-energy half of the group — bundles the strongest-moat segment (smart chargers) with the weakest (new energy). The spin valuation depends almost entirely on which of those two gets weighted more heavily.

The geographic moat — multi-country footprint — is real but unverified at the revenue line. Mexico operational from 2025; Vietnam and Hungary in service. The FY2025 AR does not disclose ex-PRC capacity share or ex-PRC revenue split, so the tariff-resilience claim is directionally credible but unproven in numbers. First-time disclosure of an ex-PRC revenue share above 15% in the FY2026 AR would be the single most important moat-widening data point on the watchlist.

8. What to Watch

The first moat signal to watch is the smart-chargers & controllers segment gross margin in the FY2026 H1 interim — if it prints above 25% the narrow moat is widening; if it prints below 22% the AI/HPC PSU program is not converting and the moat case narrows toward "no moat."

---

Sources: business-claude.md (segment economics, peer panel, valuation framework); industry-claude.md (cost stack, cycle drivers, regulation); competition-claude.md (peer threat map, advantage scorecard); numbers-claude.md (returns, margins, cash-flow conversion); forensics-claude.md (governance lattice, restricted-cash mechanism); people-claude.md (founder discipline, succession question); story-claude.md (promise delivery track record); FY2025 annual report and results announcement (Ten Pao Group, PRNewswire 2026-03-20, HKEX); peer financials from data/competition/peer_valuations.json (Yahoo Finance, snapshot 2026-05-21); spin-off announcement (HKEX / BigGo, 14 May 2026); web-research evidence in data/moat/web-research/agent-research.json. All figures in US dollars unless noted.

Figures converted from Hong Kong dollars at historical FX rates — see data/company.json.fx_rates for the rate table. Ratios, margins, and multiples are unitless and unchanged.

Financial Shenanigans

1. The Forensic Verdict

Ten Pao's accounting reads as fundamentally honest but structurally exposed. The income statement, audit opinion, and accrual quality are clean; the worry sits one ring out — controlling-family scaffolding around the operating company, a restricted-cash and bank-acceptance-bill mechanism that flatters reported leverage, a pre-IPO equity scheme inside the Huizhou subsidiary that is now being teed up for an A-share spin-off, and a step-down in the effective tax rate from 22.8% to 10.8% that may not repeat. Forensic Risk Score: 38 / 100 — Watch. The single data point that would push this to Elevated is unfavourable disclosure in the Ten Pao Electronics (Huizhou) A-share prospectus — minority valuation, related-party carve-outs, or expansion of the connected-supplier cap.

Two findings dominate. First, the Chairman-and-family lattice: Chairman Hung owns 64.88% via two holding chains, his daughter sits on the executive board as of January 2024, his spouse owns the related-party copper-wire supplier ($12.2M of FY2025 purchases), and the Huizhou production base is leased from a Chairman-controlled company ($0.9M of FY2025 rent). Second, the bank-acceptance-bill / restricted-cash structure: $72.7M of "borrowings" are simply bills payable to suppliers, fully cash-collateralized by $75.0M of restricted bank deposits in the same banks. That mechanism keeps trade payables off the operating-cash-flow drag and parks 71% of "cash" as off-limits security.

2. Breeding Ground

The governance setup tilts toward founder dominance, but the technical controls — auditor, audit committee, related-party disclosure — are properly in place. PwC has signed clean opinions, the audit committee chair is a Hong Kong CPA who chairs Halcyon Capital's IPO advisory work, and the four INEDs collectively hold the audit, remuneration, and nomination committees. The standout governance friction is concentration of power: Chairman Hung is also CEO, chairs the Nomination Committee, sits on the Remuneration Committee, controls 64.88% of shares directly and via family trust, and his daughter has been an Executive Director since January 2024.

3. Earnings Quality

Reported earnings are durable on the income statement, but the income mix is leaning more on tax savings and other gains. Gross margin expanded from 16.7% in FY2021 to 19.5% in FY2024 before slipping to 18.2% in FY2025. The effective tax rate, however, stepped down sharply — from 22.8% in FY2020 to 10.8% in FY2025 — driven by recognition of previously unrecognised Hong Kong subsidiary tax losses and Vietnam preferential rates. Each percentage point matters: a return to the FY2020 effective rate would reduce FY2025 net income by approximately $6.6M, or 13% of reported earnings.

The other-gains line is a recurring tailwind. FY2023-FY2025 each booked $4.5-6.3M of "other gains — net" (mainly fair-value movements, foreign exchange, and government grants). FY2024 included $3.0M of government grants in other gains, falling to $1.5M in FY2025. Combined other income plus other gains was $7.5M in FY2025, or 13.7% of operating income — non-negligible. None of these are accounted for as exceptional; they sit inside operating income.

The receivables story is more nuanced than the headline. Total "trade and other receivables" grew from $203.1M to $212.0M (+4.4% in USD; +4.6% in native), but the trade-receivables portion actually shrank from $187.8M to $179.5M (-4.4%). The growth came from VAT recoverable ($8.3M to $20.6M, +149%) and a new $5.1M receivable from government. Both balances are normal artefacts of PRC manufacturing operations; they do not reflect deteriorating customer credit.

Trade-receivables impairment allowance is thin. PwC flagged ECL on trade receivables as the single key audit matter — sensible because $179.5M of gross trade receivables is 25% of total assets. Loss allowance is $0.8M, just 0.46% of the gross balance. That is low for an industrials manufacturer selling on 60-90 day terms to PRC OEM customers, but it is consistent with management's reported zero history of material customer defaults, and PwC accepted it after testing.

4. Cash Flow Quality

Cash flow is reliable in aggregate but lumpy, and the underlying mechanics depend more on supplier-finance plumbing than the headlines suggest. Five-year cumulative CFO of $320M against five-year net income of $227M gives a 1.41x ratio — strong on the surface. But the lumpiness matters: FY2024 CFO of $104.8M was bolstered by $114.5M of payables expansion, and FY2025 CFO of $45.4M (0.93x net income) reflects the company "accelerating payments to suppliers in order to acquire a more competitive pricing for the raw materials," per the MD&A. Said differently, FY2024's working-capital lifeline became FY2025's working-capital drag, and the real run-rate operating cash flow is somewhere between the two.

Free cash flow is the more honest read. Five-year FCF of $87M is 38% of net income — heavy reinvestment, not strong cash conversion. Capex grew from $22M in FY2020 to $75M in FY2021 (the Huizhou Industrial Park build) before settling around $39-50M per year. FY2025 capex of $43M against depreciation of approximately $21M means capex remains roughly 2x depreciation, consistent with ongoing capacity investment.

The bank-acceptance-bills mechanism deserves separate examination. The company issues bank acceptance bills payable to suppliers; the supplier receives cash from the issuing bank; the company posts an equivalent cash deposit in the same bank as security; the bill matures within months and is rolled. Bills payable sit in short-term borrowings ($72.7M at year-end FY2025) and the security deposits sit in restricted bank deposits ($90.2M, of which $75.0M directly secures borrowings). The net effect: a substantial chunk of payables is laundered through the financing line, while operating cash flow appears uncluttered.

The structure is fully disclosed and unremarkable for a PRC manufacturer of this scale, but it changes the read on two reported metrics. First, gearing: management discloses the headline gearing ratio at 36.6% but immediately tells the reader the ex-bank-acceptance-bills gearing is 10.6%. The economic answer sits closer to 10.6%, because the $72.7M of bank acceptance bills is back-to-back with restricted cash and cannot fund operations. Second, cash: $126.5M of total cash and restricted deposits is the headline, but only $36.3M is free to spend — the rest is locked against bank acceptance bills.

5. Metric Hygiene

The company runs the metrics quite cleanly. HKEX-listed manufacturers do not generally publish non-GAAP earnings, adjusted EBITDA, or "cash earnings" definitions, and Ten Pao does not deviate from that norm. There is no parallel earnings construct, no contract-asset reclassification, and no recurring restructuring charge being labelled non-recurring. The headlines investors should monitor are the management-defined gearing ratio, the segment splits (which were restructured for FY2023 disclosure), and the use of "cash generated from operations" (before interest and tax) versus the proper CFO line.

The segment restructuring in FY2023 is the one item that meaningfully reduces transparency. "Electrical home appliances", which contributed $318M (39% of group revenue) in FY2021, was folded into "others" from FY2023 onward without a continuous comparable series. This makes it harder to test whether the consumer-power-supply franchise has actually grown or whether revenue migrated between buckets.

6. What to Underwrite Next

The accounting risk is not a thesis-breaker. It is a position-sizing limiter and a valuation haircut — the multiple investors are willing to assign should reflect the controlling-family lattice, the pre-IPO subsidiary structure, and the dependence of reported earnings on a tax-rate step-down that may or may not persist.

Five concrete diligence items, in order of materiality:

1. Ten Pao Electronics (Huizhou) A-share prospectus. When filed, read the related-party transactions disclosure, the post-IPO economic interest of the listed parent, the share-award scheme vesting valuation, and any covenants on dividend upstreaming. This single document will resolve more accounting risk than two years of interim reports.

2. Golden Ocean Copper purchase trend versus the $18M FY2026 cap. FY2025 actual purchases were $12.2M against a $16.7M cap. If FY2026 purchases push toward the $18M cap with no corresponding revenue scaling, the related-party purchase is being used to channel margin away from the listed entity. The watch line is purchases as a percentage of group cost of sales — FY2025 was 2.1%.

3. Effective tax rate sustainability. The drop from 22.8% to 10.8% over five years adds approximately $7.7M per year of net income versus the historical baseline. Track whether previously unrecognised tax losses continue to be recognized, whether Vietnam profitability scales, and whether PRC subsidiaries lose their "New and High Technology Enterprise" status (15% versus 25% statutory rate).

4. Restricted cash and bank acceptance bills ratio. A move above 80% restricted-to-total cash, or a sharp expansion of bank acceptance bills payable without matching restricted deposits, would signal the working-capital plumbing is straining. FY2025 ended at 71% restricted, with bank acceptance bills $72.7M against restricted deposits $90.2M.

5. Trade ECL allowance versus revenue growth. Current 0.46% allowance on gross trade receivables is thin. If revenue accelerates into new-energy and Belt-and-Road markets without a corresponding step-up in the loss allowance, that gap is a future P&L surprise.

What would upgrade the grade to Clean (under 20): a Ten Pao Electronics (Huizhou) prospectus that fully discloses related-party terms at market and preserves the listed parent's economic interest, plus an unwinding of the bank-acceptance-bills structure (a multi-year reduction in restricted cash). What would downgrade to Elevated (above 40): expansion of the connected-supplier cap above $26M, an audit-committee composition change at the listed parent during the SpinCo IPO window, or evidence that the Huizhou share-award scheme grants were priced materially below A-share comparables.

For position sizing, treat this as a name where mid-single-digit cap weighting is sensible if the spin-off is delayed or fails, and a name where larger weighting requires the SpinCo prospectus disclosure as a precondition. The reported margins are real, the cash conversion is real-but-lumpy, and the auditor opinion is solid — but the family scaffolding around the operating business is real too, and accelerating events (the A-share spin-off filing, the share-award scheme grants, the cap raise on the connected-supplier purchases) are happening at the same time, in the same direction, and outside the listed entity. That is the dynamic to watch.

Figures converted from HKD at historical FX rates — see data/company.json.fx_rates. Ratios, margins, and multiples are unitless and unchanged.

Governance grade is C — a controlling founder with 66.8% of the float owns the upside, but he also chairs the board, runs day-to-day, sits on his own pay committee, employs his daughter, leases factories and offices from companies he owns, and buys copper wire from his wife's company. Skin in the game is real; checks on it are not.

1. The People Running This Company

The operating company is effectively three people: the 67-year-old founder who built it, the CFO who closes the books, and the founder's 32-year-old daughter. Everyone else is non-executive.

Mr. Hung Kwong Yee (Chairman & CEO, age 67). Founder of Ten Pao in December 1979; built the Huizhou plant in 1988; took the company public in 2015. He performs both chairman and CEO roles — the company concedes this is a deviation from CG Code provision C.2.1, justified on the basis of "strong and consistent leadership." His service contract was renewed for three more years from 11 December 2024. He sits on the Remuneration Committee (which votes on his own pay) and chairs the Nomination Committee (which nominates his own re-election). He owns 66.78% of the shares.

Mr. Tse Chung Shing (Executive Director, CFO, Company Secretary, age 55). A former Ernst & Young senior manager (1992–2003) and ACCA fellow who joined the Group in December 2010, became Company Secretary in June 2015 and CFO in November 2015. He was elevated to executive director on 1 January 2024. He is the most credible piece of management infrastructure — a 30-year accounting professional running finance, IR, audit liaison and board-secretariat work simultaneously. He owns no disclosed shares.

Ms. Hung Sui Lam (Executive Director, age 32). Chairman Hung's daughter. Appointed to the board on 1 January 2024, added to the Nomination Committee in August 2025. Her FY2025 base salary nearly doubled year-on-year ($58K → $109K). The annual report carries no biographical disclosure of her qualifications, prior experience, or current operating role. She is on the path to succession by family rather than by demonstrated competence.

2. What They Get Paid

Total director emoluments were $2.16M in FY2025 (FY2024: $2.28M). Chairman Hung took $1.50M — 70% of the total — split roughly 50/50 between fixed salary and "discretionary bonus." There are no share-option grants for any director, no equity-based pay, and no clawback disclosure.

Is it earned? The chairman's $1.50M package is modest on absolute terms — about 0.4% of last close market cap and less than 0.25% of FY2025 revenue. Pay fell ~9% year-on-year, broadly tracking the deterioration in earnings. But it is not at-risk in any meaningful sense. Bonus is discretionary, not formula-linked. There are no options, no PSUs, no clawback. The Remuneration Committee includes Chairman Hung himself (alongside four INEDs, three of whom have served beside him for a decade). For a controlling-shareholder CEO who already owns 66.78% of the equity, fixed cash compensation is the least efficient form of pay — and that is exactly what he gets.

The $2.67M share-award liability in the "five highest-paid individuals" disclosure is not a parent-level dilution. It is the Ten Pao Electronic (Huizhou) Share Award Scheme — a subsidiary-level equity scheme adopted February 2024 that becomes meaningful only if the planned Huizhou A-share spin-off proceeds. Outside shareholders of 1979.HK do not participate in that pool; subsidiary employees do. That is a real alignment question, not a cosmetic one — see §3.

3. Are They Aligned?

Ownership and control

Chairman Hung controls 66.78% of the company through three vehicles: a wholly-owned BVI company (Even Joy, 34.44%), a family discretionary trust (TinYing, 30.44%, trustee Vistra), and a personal stake (1.90%). The only other 5%+ holder is Fidelity China Special Situations PLC at 7.76%. Public float is exactly 25.0% — the regulatory minimum, with zero buffer.

Insider buying vs selling

There is no reported director dealing in either direction during FY2025 or through the date of the report. Hong Kong does not require Form 4-style real-time disclosure, but the SFO Section 352 register and the Model Code confirmation in the AR show no transactions.

Dilution and capital allocation

Share count is essentially stable at 1,030.4M. The 2015 share-option scheme expired on 23 November 2025 with only 456,000 options ever exercised over a decade. A new 2025 share-option scheme (10% mandate) was adopted at the June 2025 AGM but no grants have been made. Dilution at the parent level is currently negligible.

The capital-allocation flags are elsewhere:

1. Scrip dividend. The FY2025 final ($0.0085, up from $0.0077) carries a scrip dividend option, which has historically grown share count modestly each year — a soft form of dilution that conserves cash but trades off pro-rata ownership for non-electing holders.
2. Subsidiary share scheme. The Huizhou-level share award scheme directs equity value of $2.67M (FY2025) and $1.46M (FY2024) to subsidiary employees ahead of a possible A-share spin-off. Whether parent shareholders see a clean economic carry depends on the terms of the spin (announced 2026-05-14, not yet detailed).
3. No buyback execution. A buyback mandate is on the 2026 AGM agenda, but no parent-level repurchases have been disclosed.

Related-party behaviour — the alignment problem

This is the section that drags the grade down. The chairman is on both sides of three categories of recurring transaction with the listed entity:

Aggregate annual "leakage" to chairman-controlled entities is ~$13.2M, of which ~$1.0M is rent and ~$12.2M is procurement. The board's own Audit Committee meeting on the lease topic excluded Chairman Hung and Ms. Hung Sui Lam (Note 1 of the attendance table), which is procedurally correct, but the structural arrangement is what it is.

Skin-in-the-game scorecard

A 66.78% economic stake is unambiguously high; Chairman Hung's personal wealth tracks the share price almost 1:1. He has held since IPO and has not sold a share. That is worth a strong score on raw alignment.

We knock it down to 7/10 because:

• He extracts ~$13M/year through related-party transactions with family-owned entities.
• His pay is unindexed cash; the bonus is discretionary.
• His daughter is on the executive payroll with a sharply rising salary and no disclosed credentials.
• A scrip-dividend habit slowly transfers economics to non-electing holders.

The chairman wins big if 1979.HK goes up. The asymmetric concern is what happens to outside holders if the planned Huizhou spin-off transfers value to the subsidiary (and to its share-award scheme participants) rather than back to the parent.

4. Board Quality

Seven directors. Four are nominally independent — but three of those four had served continuously since the November 2015 IPO. Hong Kong's tightened tenure rule (effective 2026-07-01: 9-year cap on INED tenure) is what is forcing the cleanup.

Committee composition — independence in name vs. substance

What's missing

No director with disclosed experience in: EV charging, power electronics, North American or European customer markets (where the bulk of OBC and energy-storage growth must come from), or large-cap industrial M&A. The 2026 refresh adds gender diversity (Ms. Mahomed) but neither incoming INED is publicly profiled as a power-electronics expert. Capability gaps line up with the business's stated growth bets.

Compliance lapses

The annual report itself flags one Code deviation (C.2.1, combined Chair/CEO). No restatements, no qualified audit opinions, no SFC disciplinary matters surfaced in the data. Attendance was high (Chairman Hung missed one of six board meetings — the one approving leases from his own private companies, correctly recused).

5. The Verdict

The good. A founder-owner with 66.78% of the float and 47 years in the industry is rare on the Hong Kong main board. He has not sold a share since IPO. The CFO is a credentialed audit-industry professional. PwC has been continuous auditor. Audit Committee composition is clean. The 2026 board refresh is happening — albeit because the listing rule forces it.

The real concerns.

1. Related-party procurement at scale. $12M/year of copper wire flowing to the chairman's wife's company, with annual caps drifting up to ~$18M by 2026, is the largest single governance-economic risk.
2. Captive committee structure. The three INEDs who chair Audit, Remuneration and Nomination respectively all sit on a board where the chairman is also the CEO and the chairman's daughter is an executive director. Two of those three are leaving in June 2026; one (Mr. Lam) is staying. The refresh is partial.
3. Daughter succession optics. A 32-year-old executive director with no published credentials and a base salary that doubled year-on-year, on the very committee that nominates directors, is succession-by-bloodline staging.
4. Spin-off design risk. The proposed A-share spin of Ten Pao Electronic (Huizhou) — combined with a $2.67M subsidiary-level share award pool that is not shared with parent holders — is the single biggest pending alignment test. If parent shareholders are diluted at the subsidiary level via the share-award scheme before the spin, value transfer happens without their direct vote.

What would change the grade.

• Upgrade to B if (i) the Golden Ocean Copper relationship is competitively benchmarked or wound down, (ii) Mr. Lam Cheung Chuen also retires at or before the 2027 AGM, and (iii) the Huizhou spin-off is structured so that 1979.HK shareholders receive a clear pro-rata economic interest.
• Downgrade to D if the Huizhou spin-off proceeds with material economic value flowing to the chairman's family or subsidiary management at the expense of listed-entity shareholders, or if related-party transaction caps are again raised mid-year.

Figures converted from HKD at historical FX rates — see data/company.json.fx_rates. Ratios, margins, and multiples are unitless and unchanged.

The Narrative Arc

Ten Pao listed in December 2015 as a "switching power supply unit manufacturer" for smartphones and power tools, and its founder-CEO Hung Kwong Yee has never let go of the wheel. The story since has been a slow, deliberate widening of the lens: from chargers, to chargers-plus-factories-abroad, to chargers-plus-EV-and-energy-storage, and most recently to AI/data-centre power. What has not changed is the cadence of the language — measured, slogan-heavy, and almost entirely free of numerical guidance — which makes the company hard to catch out but also hard to credit with much beyond what the financials themselves prove. Credibility on operational promises (factories, R&D centres) is solid; credibility on whether the "new energy" growth engine is the durable second leg management has implied since 2020 is, on the 2025 evidence, deteriorating.

Chapter markers

Two anchors for every other tab's judgment:

• Current CEO start year: 2015 — Hung Kwong Yee has been Chairman and CEO of the listed entity since its IPO (June 2015 redesignation). He is also the 1979 founder of the underlying business.
• Current chapter start year: 2020 — the creation of the New Energy Business Division and the Huizhou Industrial Park groundbreaking mark the structural turn from a single-segment Chinese smartphone-and-tool charger maker to a four-pillar diversified power business. The 2024 second-generation board entry is best read as a continuation of that chapter, not a new one.

What Management Emphasized — and Then Stopped Emphasizing

The clearest signal in eleven years of Chairman's Statements is what got quietly dropped. Three themes were promoted hard for a window and then erased: electronic cigarettes (one mention, 2015, never again); the "Industrie 4.0 / Made in China 2025" slogan (heavy 2016–2018, gone by 2021); and the "Belt and Road" framing of overseas expansion (2017–2019, gone by 2022). Conversely, "new energy" went from absent (2015–2019) to dominant (2021–2024), and "AI / data centres / computing" appeared in 2024 and is already the lead noun of 2025. The 2025 statement is the first since IPO not to mention "switching power supply units" — the phrase that once was Ten Pao's identity.

Quietly dropped:

• Electronic cigarettes — flagged in 2015 ("electronic cigarettes markets will maintain their rapid growth momentum and are expected to continue as the driving force of the Group's development") and never mentioned again. A reasonable read: regulatory crackdowns made it impolitic and probably small.
• Industrie 4.0 / Made in China 2025 / Belt and Road — dropped together around 2021–2022, in line with the broader retreat of those slogans in PRC corporate communication. The substance (automation, overseas factories) lived on; the political branding did not.
• "Switching power supply units" as the lead identity — the 2025 Chairman's Statement positions Ten Pao as "an industry-leading supplier of intelligent power supply solutions" and never uses the original phrase.

Newly promoted:

• AI / data centres / computing — went from zero to lead theme in two years. The H2 2024 launch of "computing power supply equipment" is presented as the strategic anchor for 2025+.
• Green / "net-zero" / "fuel-to-electric" — emerged with COP29 attendance in November 2024 and is now consistently paired with the energy storage and EV charger narrative.

Risk Evolution

The disclosed-risk section is brief by design — HKEX issuers typically tuck principal risks into the MD&A rather than a separate item — but the language shifts trace the operating environment well. The standing risks (FX, credit, raw materials) never leave; what changes is the new risk added each year, and how it is framed.

The single concrete number that moved: Ten Pao's USD/HKD-denominated revenue (the FX exposure RMB-cost manufacturers care about) has steadily fallen as the business diversifies geographically and segments shift toward RMB-billing new-energy customers.

What entered the discussion newly in 2023–2025:

• Red Sea conflict (2023) — first non-China-centric supply-chain risk discussed in management commentary.
• AI / computing-power execution risk (2025, implicit) — the company has positioned the H2 2024 launch as a growth pillar without quantifying revenue or margin.
• New-energy slowdown (2025) — the segment fell 10% YoY in FY2025 after four years of being framed as the growth engine. The 2025 Chairman's Statement does not concede this directly; instead it pivots the narrative to AI/data centres.

How They Handled Bad News

Ten Pao has had two real disappointments since listing — the 2018 profit collapse (net income -65%) and the 2022 revenue collapse (-13.9%, with consumer-power-supply -18.8%). The pattern in both cases is the same: blame an external factor (raw materials in 2018; smartphone slowdown plus geopolitics in 2022), point to a structural offset already in motion (the power-tools segment in 2018; the new-energy build-out in 2022), and promise a "quick rebound when pent-up demand is released" — language used almost verbatim in 2022 and echoing 2018.

Guidance Track Record

Hong Kong-listed industrials of this scale rarely give numerical guidance, and Ten Pao gives none. What it does give are operational commitments — factories, product launches, R&D directions, dividend posture. Judged on those, the record is solid on plant build-out (Hungary, Vietnam, Sichuan, Mexico all opened broadly on the timelines promised, with the Huizhou Industrial Park the only meaningful slip) and weaker on growth-segment durability (new-energy "doubling" did not sustain). Strategic pivots have been telegraphed before being made, which is unusual for a controlled family business.

Credibility score: 7 / 10. Operational promises (factories, product launches, R&D centres) have been delivered roughly on stated timelines, with one ~2-year slip on the Huizhou Park. The story has remained internally coherent: every segment management said would be built has been built; every plant they said would open has opened. The two points off come from (a) the new-energy growth narrative breaking in 2025 without a candid acknowledgement, and (b) the absence of any measurable forward guidance, which lowers the ceiling — there is nothing to "beat or miss" in a quantitative sense, only directional signals.

What the Story Is Now

The story Ten Pao is telling in 2025 is simpler than the 2021–2024 story but more stretched. It has dropped the original switching-power-supply identity, downplayed the new-energy/EV-charger narrative that was the headline for four years, and replaced both with AI/data-centre power supplies, energy storage, smart controllers, and a state-of-the-art Huizhou green factory — a portfolio framing rather than a product pitch. The Mexico/Vietnam/Hungary/Huizhou multi-base manufacturing thesis is now real and de-risked (all four sites are operating). The new-energy thesis is half-real: the segment exists and is at scale, but growth has stalled. The AI/data-centre thesis is rhetorical: the product launched H2 2024 but no separately-disclosed revenue contribution has been quantified.

Figures converted from Hong Kong dollars at historical FX rates — see data/company.json.fx_rates. Ratios, margins, and multiples are unitless and unchanged.

Financials — What the Numbers Say

Ten Pao is a $714 million-revenue Chinese contract manufacturer of power adapters, smart industrial chargers, EV chargers, and battery-pack/energy-storage products. Revenue has nearly quadrupled since 2012, but the company sits in a low-margin (gross margin around 18%, net margin around 7%), capital-intensive OEM/ODM segment where pricing power belongs to a handful of Fortune 500 customers. Reported earnings have generally converted to cash over a multi-year cycle, but cash flow is lumpy because working capital and capex swing sharply with order books. The balance sheet shows a large gross debt position that is more than fully offset by cash plus pledged deposits, so net leverage is effectively zero. The current setup — share price $0.371, P/E roughly 7.9x, P/B about 1.4x, dividend yield 4.4% — prices the company as a slow-growth manufacturer despite double-digit returns on equity and an emerging AI-server / energy-storage tilt. The single financial metric that matters most right now is FY2026 gross-margin direction: FY2025 margin slipped 130bps to 18.2%, FY2025 free cash flow collapsed to $2 million, and the bull case rests on AI-server power and smart-controller mix lifting blended margin back toward 20%.

Financials in One Page

Revenue, Margins, and Earnings Power

Revenue is what the company sells customers; gross margin is what is left after raw materials, labour, and factory overhead; operating margin is what is left after selling, R&D and admin costs. Net margin is what is left after interest and tax. For an OEM like Ten Pao, gross margin is the most direct read on pricing power and mix.

A fourteen-year top-line view

Three things stand out. First, revenue has compounded at roughly 10–11% per year over 13 years, with a clear acceleration after 2019 as the company shifted into industrial chargers and new energy. Second, 2021 was a cyclical peak — pandemic-era device demand pulled revenue to $816 million that never repeated. Third, earnings power is variable: FY2018's $7 million net-income trough (margin 1.7%) and FY2021's $49 million peak (margin 6.0%) reveal an OEM whose profitability swings with mix and commodity costs, not with revenue alone.

Margin structure since detailed disclosure began (FY2020)

The margin path is telling. Gross margin rebuilt from 16.7% in FY2022 to 19.5% in FY2024 as the high-margin smart-chargers-and-controllers segment grew from 22.5% to 26.7% segment margin and rose to 37% of sales. FY2025 reverses some of that gain (gross margin 18.2%) on rising electronic-component costs and fiercer industry competition. Operating margin held up better than gross margin because admin expenses fell 5.2% year-on-year, evidence of cost discipline. Net margin stayed close to 6.9% thanks to a low 10.8% effective tax rate (driven by High-and-New Tech Enterprise status for five PRC subsidiaries and a 10% rate in Vietnam).

Recent trajectory — the first half pattern

The H1 2025 number matters because it is the most recent operating window. H1 2025 revenue jumped 19% year-on-year to $376 million, while operating margin reached an H1 record 8.6%. That sets up the read on the FY2025 full year: H2 2025 gross margin can be backed out at 18.6% (full-year 18.2% minus H1 17.9% on a weighted basis), so the second half actually firmed slightly even as competition intensified. The earnings power has not obviously deteriorated; the headline FY2025 gross-margin step-down is mostly comp against an unusually strong FY2024.

Cash Flow and Earnings Quality

Free cash flow (FCF) is cash generated by operations minus capital spending. It is the cash an owner could theoretically pull out without harming the business. For an OEM with heavy working capital and ongoing factory build-out, FCF is more volatile than reported earnings; what matters is whether the multi-year sum of FCF roughly equals or exceeds the multi-year sum of net income.

The chart reveals the OEM rhythm. Operating cash flow has outrun net income in five of the last six years — a sign that reported earnings are real cash, not accruals. But FCF whipsaws because capex is lumpy: FY2021 saw $75 million of capex (new factory build) that swamped operations; FY2024 was the inverse, with $105 million operating cash and $54 million FCF.

The FY2025 cash-flow problem

FY2025 free cash flow collapsed to $2 million — 5% of net income. Two drivers, both disclosed in management's commentary. First, the Group "accelerated payment to suppliers in order to acquire a more competitive pricing for raw materials," which drained working capital. Second, capex stayed elevated at $43 million as the Huizhou Intelligent Manufacturing Industrial Park ramped to full operation in October 2025. Both pressures look transitional rather than structural; if management is right that "future capital expenditure is expected to decline progressively," FY2026 FCF can rebuild.

Balance Sheet and Financial Resilience

The optics look highly geared: gearing ratio (borrowings / equity) was 36.6% at year-end 2025, with all $102 million of debt maturing within one year. That is the kind of profile that worries a credit committee on first read. But the company's own disclosure flags an important nuance: most of that "borrowing" is bank's acceptance bills — short-dated paper used to fund supplier purchases and secured by deposits held in the same bank. Adjusted for those secured bills, the underlying gearing ratio is just 10.6%.

Restricted cash sits inside the same banks that issued the acceptance bills. Functionally, the working-capital financing is a closed loop — the cash deposit collateralises the bill — so the balance sheet has effectively no net debt. Working capital is the real risk: receivables of $212 million plus inventory of $123 million represent roughly six months of sales, typical of a Chinese ODM with global customers but a real exposure if a major customer were to delay payment or cancel orders.

Returns, Reinvestment, and Capital Allocation

Return on equity (ROE) measures profit per dollar of book equity. Return on assets (ROA) does the same per dollar of total assets. An ODM with sustained ROE above 15% is creating value above its likely cost of equity.

ROE has averaged 22.0% over the last nine years and stayed above 17% even in FY2025. That is a strong reading for a thin-margin OEM — it works because the company turns assets briskly (FY2025 asset turnover roughly 1.0x) and uses minimal financial leverage net of secured working-capital bills. ROA has drifted lower from a 9% peak as the asset base bloated with the new Huizhou park, the kind of dilution one expects from a multi-year capex cycle that has not yet fully translated into revenue.

Where the cash is going

Cumulative six-year capex is $254 million; cumulative six-year dividends paid is $36 million. Roughly seven dollars of capex for every dollar of dividend. That ratio is the right answer in a growth phase, but it cannot persist — at the FY2025 FCF rate of $2 million, the dividend of $17 million ($0.016 per share, 1,030 million shares) is being funded out of the cash pile, not out of current free cash. The FY2025 payout ratio of 34.7% of net profit is on the upper side of management's "approximately 30%" benchmark.

Share count has been essentially flat at 1,030 million since 2022 — minimal dilution, no buyback program. Management has shown no appetite for repurchases at depressed prices, which is a missed opportunity if one believes the equity is materially undervalued. The capital allocation story is therefore: build the factory, pay a steady dividend, retain the rest. That is a coherent strategy if reinvestment returns stay attractive (FY2025 incremental ROIC will be the test) but a value-eroding one if growth fails to materialise.

Segment and Unit Economics

Segment disclosure was restated in FY2023: "new energy" was carved out of "others," and "electrical home appliances" was folded into the new "smart chargers and controllers" segment together with parts of "others." The figures shown below use the most recent restatement available.

Segment revenue mix and profitability — FY2025

Segment revenue trend FY2020–FY2025

The mix shift is the single most important fundamental story in this company. Smart chargers and controllers (the industrial / AI-server power segment) was 10% of revenue in FY2020 and is 39% today. At 24% segment gross margin, this segment now contributes roughly $67 million of the $130 million of group gross profit — more than half of gross profits from less than 40% of revenue. Telecom (mobile chargers) has shrunk from 35% of revenue in FY2020 to 24% in FY2025, and its margin sits in the low teens. The bull case is that this mix shift is permanent and that the new energy segment (currently a 7.7% gross-margin drag) eventually scales into profitability with the battery-swap and home energy-storage push.

Geographic mix — FY2025

Direct US exposure is 6.4% of revenue, but ultimate end-product destination is much broader because customers re-export. Africa grew to 5.0% of revenue on a single decade-long lighting partnership tied to solar-plus-storage lighting. The Mexico, Vietnam, and Hungary plants are positioned for tariff-driven nearshoring; the largest risk remains the 49% Chinese-mainland concentration in a tariff-volatile environment.

Valuation and Market Expectations

The stock closed at $0.371 on 21 May 2026 against a 52-week range of $0.192–$0.432. Market capitalisation is approximately $383 million (1,030 million shares).

What the market is paying for

The trailing P/E of 7.9x is well below the Hong Kong-market average (around 10x), and at roughly half the broader Industrials sector multiple of approximately 12x. P/B of 1.4x is in line with the broader market and modestly above the company's own historical 1.0–1.2x range, reflecting the ROE rebuild since 2017's profitability trough. EV/Revenue near 0.51x is consistent with a thin-margin contract manufacturer.

Multiple framework — bear / base / bull

The bear case assumes gross margin compresses further toward 16%, FY2026 net income drops to roughly $40 million, and the market pays a 6x multiple — consistent with the FY2018 trough. The base case (EPS $0.051, 8x) holds margin flat and credits a slow industrial-power mix shift. The bull case (EPS $0.064, 10x) requires gross margin near 20%, FCF normalising to over $40 million, and a re-rating toward Chicony Power's roughly 10x EBITDA multiple as AI-server power crystallises. The current price is roughly equal to the base case fair value, so the upside skews mildly positive only if mix continues to shift.

Peer Financial Comparison

All peer figures expressed in US dollars (peer trading currencies converted at 21 May 2026 spot rates) to allow apples-to-apples comparison across HKEX, TWSE, and SZSE listings. Ten Pao's row is the first.

The peer panel is sharply bifurcated. Delta Electronics (2308.TW) trades at 39x EV/EBITDA and 8.3x EV/Revenue — the data-centre / industrial power supremum. Chicony Power, the closest direct PSU comparable, sits at 11.8x EV/EBITDA on a 9% EBITDA margin. AcBel and Lite-On trade between 19x and 20x EV/EBITDA. Phihong, the closest size match, is loss-making at the EBITDA line.

Ten Pao's roughly 0.5x EV/Revenue and 7.9x P/E are at the bottom of this peer set despite an 18% gross margin that is better than four of the six peers and a 17% ROE that beats five of them. The discount has a credible explanation — HKEX small-cap illiquidity, China-revenue concentration, no buyback program, and a customer base disclosed only by segment — but the gap to Chicony Power and AcBel is the gap that fundamentally has to close (or be justified) for the stock to re-rate.

What to Watch in the Financials

What the financials confirm: Ten Pao is a real, cash-generating manufacturer with growing scale, an improving mix, double-digit ROE, and an effectively zero-net-debt balance sheet. What they contradict: the bear narrative that this is a stranded mobile-charger commodity OEM — industrial chargers now contribute the majority of gross profit, and the customer base has moved toward Fortune Global 500 names in AI-server power and battery swap. What to watch: whether FY2026 free cash flow rebuilds to $40+ million as capex normalises and supplier-payment timing reverses; that single line item determines whether the dividend stays funded and whether the equity can demand a higher multiple.

The first financial metric to watch is the FY2026 H1 gross margin — if it prints above the 18.6% implied for H2 2025, the mix-shift thesis is intact and the 1.4x P/B looks low for a 17%+ ROE business; if it drops below 17%, the bear case crystallises and 6x P/E becomes the right benchmark.

Web Research — What the Internet Knows

Figures converted from HKD at historical FX rates — see data/company.json.fx_rates for the rate table. Ratios, margins, multiples, percentages and share counts are unitless and unchanged.

The Bottom Line from the Web

The single most important fact the web reveals beyond the financial filings is Ten Pao's 14 May 2026 announcement that it will spin off its core operating subsidiary Ten Pao Electronics (Huizhou) Co., Ltd. for a separate A-share listing in mainland China — already cleared by HKEX under the spin-off rules, with terms and listing venue still being negotiated. The stock spiked +21% on 19.8x average volume that day, then printed a new 52-week high of $0.43 on 20 May 2026. The web confirms the operating subsidiary that holds the manufacturing/charger business is being prepared for an A-share re-rate, while the HK-listed parent stays consolidated; this is the catalyst that frames the entire next-12-months risk/reward and is not yet reflected in the audited financials.

What Matters Most

Recent News Timeline

The two events that move the thesis are the 14 May 2026 spin-off clearance (catalyst that can compress the HK holding-company discount once SpinCo pricing emerges) and the 20 March 2026 FY2025 results (the margin compression and payout-ratio lift, both visible without needing the prospectus).

What the Specialists Asked

Governance and People Signals

Three governance signals stand out in the public record. First, the combined Chairman + CEO role at the founder is durable but means no independent challenge at the top of the executive table. Second, the elevation of the Chairman's daughter Sui Lam Hung to Executive Director in January 2024 is a clear family-succession signal; the web has no controversy on the appointment but specialist queries flagged that her rising salary band and operating P&L responsibility deserve continued monitoring. Third, an INED 9-year cap may bind for Mr. Lam Cheung Chuen by July 2026, which would force a board refresh in time for the SpinCo prospectus — a date worth tracking ahead of the June 2026 AGM.

No insider-transaction data is published for the HK ADR Form-equivalent regime in any web source we accessed; HKEX disclosure of interests filings would need to be pulled directly. No major-shareholder block change has surfaced on the news wire.

Industry Context

The web confirms three structural dynamics that change how the next-12-months thesis should be priced.

(1) Delta Electronics is the aspirational peer trading 10-15x richer. Delta (2308.TW) closed 21 May 2026 at about $67.7 per share (+6.01% on the day), with 70.8x trailing P/E and a 1-year return of +455%. Delta's 2025 revenue grew 31.76% to approximately $18.5B with earnings up 70.62%. Delta's product mix is heavily weighted to AI-server and EV power systems — exactly the segments Ten Pao is moving its mix toward at the 3.5-10 kW band Delta does not aggressively contest. The valuation gap between the two is the variant-perception trade.

(2) China EV-charger / new-energy market is oversupplied. FY2025 management commentary acknowledges "competition in the new energy market remains intense, and some projects are still in the ramp-up stage with potentially limited short-term profit contribution." The deliberate decline in new-energy revenue from $135.0M to $121.1M (-10.1%) reflects an exit from low-margin projects rather than a customer loss. The pivot toward Southeast Asia (Thailand corridor in particular) and Mexico-to-USA shipping shows up clearly in the manufacturing footprint additions of 2023-2026.

(3) Geopolitics and the China+N hedge are pricing in. PRNewswire commentary: "As geopolitical dynamics continue to swing and the market accelerates its shift towards leading enterprises, this may also prompt companies to seek more secure supply chains." Ten Pao's six-country footprint (China, Vietnam, Hungary, Mexico, plus Japan/USA sales offices) is the structural answer to the tariff and reshoring narrative; it is a real cost-curve hedge and not just marketing.

Figures converted from HKD at historical FX rates — see data/company.json.fx_rates. Ratios, margins, and multiples are unitless and unchanged.

Web Watch in One Page

The Ten Pao long-term case rests on five concrete observables, and the five live monitors below are pointed at exactly those. The decisive variable across the bull and bear arguments is the post-IPO economic interest the listed parent retains in Ten Pao Electronics (Huizhou) — the A-share spin-off PN15 application was filed with HKEX on 27 April 2026; HKEX confirmed the company may proceed on 14 May 2026; the CSRC A-share application is the next gate, not yet filed. The single durable thesis variable is whether smart-chargers segment gross margin defends the 23-25% band as the 3,500-10,000W AI/HPC PSU program qualifies a named hyperscaler. The single cleanest refutation of the bear's value-leak read would be the company actually using the 10% buyback mandate approved at the 12 June 2026 AGM. The single non-financial signal the bear watches is the trajectory of the Golden Ocean Copper procurement cap, now US$17.9M after a 2.8x expansion in two years. The single external test of the Driver #1 margin moat is whether Delta Electronics moves down-stack from its 30kW+ Open Rack shelves into the 3-10kW band before Ten Pao gets a hyperscaler name on the door. The watch set below covers each in turn.

Active Monitors

Rank	Watch item	Cadence	Why it matters	What would be detected
1	Huizhou A-share spin-off prospectus and CSRC disclosure	Daily	The decisive variable — Verdict tension #1. Whether 1979.HK retains >=85% economic interest at >=25x sponsor pricing converts the holding-company discount into a SOTP gift, or routes the moat-rich half to subsidiary scheme participants and the family layer.	CSRC acceptance, feedback, suspension, registration approval, or rejection; sponsor and underwriter appointment; pre-IPO information memorandum or prospectus disclosing retained economic interest, indicative pricing range, listing venue, and any related-party transaction carve-outs; HKEX Practice Note 15 updates under Stock Code 01979.
2	Smart-chargers segment margin and AI/HPC PSU customer naming	Daily	Driver #1 of the 5-to-10-year thesis. The 3,500-10,000W AI/HPC PSU program was showcased in October 2025 with no named customer; smart-chargers segment GM fell from 26.7% (FY2024) to 24.1% (FY2025). A named Tier-1 hyperscaler with quantified FY2026 revenue would validate the mix-migration; continued silence past FY2026 results validates the Phihong-fade frame.	Disclosure or naming of a Tier-1 hyperscaler or Fortune-500 server OEM customer for the 3.5-10kW PSU; interim or annual results note disclosing smart-chargers segment GM or AI/HPC revenue line separately; HKEX inside-information major-customer-win announcements; trade-show showcase naming a customer; Huizhou Electronic press releases on shipped volumes or design wins.
3	Golden Ocean Copper and connected-supplier cap trajectory	Weekly	The bear's lead governance evidence. The chairman's spouse's company supplied US$12.2M of copper wire in FY2025 with the FY2026 annual cap raised to US$17.9M (US$6.4M two years earlier). Forensics rates this severity red; failure-mode #3 trips above US$19.2M.	New EGM circular or shareholder vote to raise or expand any connected-transaction cap; any newly added connected supplier or landlord; interim or annual report Note 6 cap utilisation disclosure; scope expansion beyond copper wire and HK office leases; HKEX listing-rule-compliance correspondence or audit-committee statement.
4	Parent-level buyback execution under the 10% mandate	Daily	The cleanest non-financial signal a controlling-family micro-cap can send. Zero parent buybacks executed in 11 years despite the stock trading at periods below US$0.26; if the 10% mandate approved on 12 June 2026 is actually used at 7.9x P/E it refutes the value-leak thesis, while a mandate-without-execution through end-2026 confirms it.	HKEX daily issuer's report on share repurchases under Stock Code 01979; inside-information announcement of off-market buyback or tender; next-day disclosure of interests filing showing company purchases; monthly Form F (return of equity issuer) reflecting share-count reduction; any IR statement of intention to execute or to let the mandate lapse.
5	Delta, Chicony, Lite-On moves into the 3-10kW AI server PSU band	Daily	The Driver #1 break-condition the bull case must outrun. Bear treats Delta down-stack into Ten Pao's wattage band as the structural way the 24% smart-charger margin compresses toward 18%. Phihong is the cited precedent (28% GM to 21% to operating losses).	Delta Electronics, Chicony Power, Lite-On, AcBel, or Phihong product launches, roadmap statements, or design wins in the 3.5-10kW AI/HPC PSU band; Delta moving down-stack from 30kW+ Open Rack v3 shelves; new hyperscaler or server-OEM customer wins for these peers in AI server power; pricing-action, capacity, or share-gain commentary in industrial PSU and smart-charger categories.

Why These Five

The report ends with a Watchlist verdict because two pieces of evidence have not yet printed: the CSRC prospectus disclosure of the listed parent's post-IPO economic interest in Huizhou (Monitor 1), and the FY2026 H1 interim print of smart-chargers segment GM with or without a named AI/HPC PSU customer (Monitor 2). Those two together resolve roughly 70% of the bull-bear gap. The other three monitors are the report's own validation-and-refutation set: Monitor 3 watches the bear's structural value-leak channel that scales as the spin is staged, Monitor 4 watches the single capital-allocation tell that would refute the value-leak read at lowest cost, and Monitor 5 watches the external competitive test that could kill Driver #1 independent of anything the company itself discloses. The five together are the smallest watch set that gives an investor sequential visibility into the events that decide whether Ten Pao compounds at 16% per year over five years (base case) or trades back to the FY2018 trough multiple (bear case).

Figures converted from HKD at historical FX rates — see data/company.json.fx_rates. Ratios, margins, and multiples are unitless and unchanged.

Where We Disagree With the Market

The market is still pricing 1979.HK as a stranded HK-listed consumer-adapter ODM with a free option on the spin-off, when the protected slice (smart chargers, lighting, media, others) already drives the majority of group gross profit and the spin's SOTP math is closer to arithmetic than to a coin flip. The 7.9× P/E versus a peer median around 25× is not a structural HK small-cap discount once the operating numbers are laid alongside Chicony Power — a peer 1.5× the size on objectively worse margins and lower ROE that trades at 18.9× P/E. We disagree with the consensus read on three things: the denominator the market is valuing (consolidated EPS at a HKEX consumer-OEM multiple), the duration the market is assigning to the FY2025 cash-conversion collapse (it is treated as a structural break when it is a working-capital reversal plus capex digestion), and the probability distribution the market is assigning to the Huizhou A-share spin terms (a +21% one-day spike on the announcement captured a fraction of the available arithmetic). None of these is a contrarian opinion — each survives the test of "what observable signal resolves it" and prints inside the next nine months.

The single resolving signal is the FY2026 H1 interim segment-information note around 22 August 2026: smart-chargers segment gross margin at or above 25% with AI/HPC PSU revenue separately disclosed validates the mix-migration variant; below 22% with continued AI/HPC silence kills it. The CSRC prospectus disclosure of the listed parent's post-IPO economic interest in Huizhou is the larger arithmetic-resolver but its hard date is window-dependent at 6 to 18 months.

Variant Perception Scorecard

The score reflects three things at once. Variant strength is medium-high because three independent disagreements pass the "what resolves it" test and the gap to consensus is material to valuation. Consensus clarity is only medium because sell-side coverage is empty — Investing.com and Reuters analyst tabs are blank, so "the market view" has to be inferred from price level, peer multiples, the 14 May 2026 reaction, and Morningstar's lone US$1.26 fair-value print with a self-contradictory star band. Evidence strength is medium-high on the operating side and lower on the spin side, because the decisive A-share retained-interest number has not yet been disclosed. Time to resolution is fastest for the H1 interim (about three months) and longest for CSRC sponsor pricing (6 to 18 months).

Consensus Map

The consensus is not a single coherent view — it is six separate implicit assumptions that survive because no sell-side reporting forces them to be defended. The two assumptions most worth disagreeing with are the cash-conversion read (because it confuses a working-capital reversal with structural quality) and the mix-migration read (because it ignores that the protected slice already produces around 60% of group gross profit at 19-25% segment margins).

The Disagreement Ledger

Disagreement #1 — mix migration is already mostly done. Consensus would say the smart-chargers segment GM compression from 26.7% to 24.1% in FY2025 is the leading edge of a Phihong fade and that telecom (24% of revenue at 13.3% GM) plus new-energy (17% at 7.7% GM) are still big enough to anchor the peer-set positioning. Our evidence disagrees because the gross-profit pool is already a different shape: smart-chargers contributes about US$66.8M of the US$130M group gross profit on under 40% of revenue, and protected-slice segments together produce around 60% of group GP at segment margins ranging from 19.8% to 25.1%. If we are right, the market has to concede that the relevant comp for 1979.HK is not Chicony Power's 16.5% gross-margin profile but a hybrid that earns the second-highest op margin in the listed Asian PSU peer set (7.6%, behind only Delta). The cleanest disconfirming signal is the FY2026 H1 segment-information note printing smart-chargers segment GM below 22% with the protected slice's share of gross profit declining; that would say the mix migration is reversing rather than complete.

Disagreement #2 — the +21% spin spike priced optionality, not arithmetic. Consensus would say the market needed exactly that kind of move to reflect spin uncertainty — a binary event with possible value-leak to the Huizhou Share Award Scheme (adopted Feb 2024, pre-dating the May 2026 spin announcement) and to Golden Ocean Copper's expanded US$17.9M FY2026 supply cap. Our evidence disagrees because the regulatory clock has materially advanced — PN15 application filed with HKEX 27 April 2026, HKEX confirmed proceed under PN15 on 14 May 2026; CSRC A-share application not yet filed — and because the conservative arithmetic at 20× retained-90% gives US$511M for Huizhou alone versus US$383M group market cap, even before applying any A-share charger-peer premium. If we are right, the market would have to concede that pricing this as a coin-flip lottery ticket understates a base-case multiple-arbitrage outcome that requires no AI/HPC heroics. The cleanest disconfirming signal is CSRC sponsor pricing below 20× or retained economic interest below 80% — either, alone, kills the SOTP arithmetic.

Disagreement #3 — "broken cash conversion" is the wrong duration. Consensus would point to the chart showing FY2025 FCF of US$2.4M against US$16.2M of dividends paid and conclude the dividend is funded from the cash pile, not current FCF. Our evidence disagrees because FY2024's CFO of US$104.7M was lifted by US$114.5M of payables expansion that mechanically reversed -US$45.6M in FY2025 — the two-year average CFO is around US$75M, the six-year average FCF is US$22.7M (US$0.022/share), and FY2025 capex of US$42.9M was the Huizhou industrial park's final ramp before October 2025 full-operation, after which management has explicitly guided capex "is expected to decline progressively." If we are right, the market would have to concede that the FCF/NI of 0.05× is a single-year reversal artifact, not a structural break, and that capital-return capacity is meaningfully higher than the trailing-yield framing suggests. The cleanest disconfirming signal is FY2026 H1 operating cash flow below US$19M with capex remaining above 6% of sales — both together would say the digestion isn't happening.

Disagreement #4 — coverage gap is bounded by the disclosure cycle, not by listing venue. This is the weakest of the four because it depends on behaviour by sell-side desks rather than on disclosure inside the company. We include it because it is the most under-rated mechanism by which the consensus itself changes: the absence of published price targets is not the same as a published price target of US$0.37. The variant claim is that the discount-from-coverage compresses faster post-CSRC-disclosure than post-FY2026 H1 print because the prospectus forces A-share benchmarking on every desk that has the SpinCo industry on its sheet.

Evidence That Changes the Odds

The two highest-leverage evidence items are #1 (op margin parity with Delta-class peers at sub-US$1B scale) and #4 (the FY2024-to-FY2025 working-capital reversal that explains the FCF collapse). The first refutes the "Phihong trough multiple" anchor and the second refutes the "dividend not funded by FCF" framing — together, they remove two of the three multiple-compressing assumptions in the consensus model and leave the spin terms as the binary still to resolve.

How This Gets Resolved

The signals are ranked by leverage on the variant view, not by date. Signals 1 and 2 land at the same H1 interim print (around 22 August 2026); they will be read together. Signal 3 is the largest single re-rater but has the widest timing band. Signals 4 and 5 are non-financial alignment tests that resolve the governance overlay on the variant view. Signal 6 is the slowest evidence to print and is more relevant to the long-term thesis than to the variant view this tab is making.

What Would Make Us Wrong

The single piece of evidence that would force a reset is the FY2026 H1 segment-information note printing smart-chargers segment gross margin below 22% combined with operating cash flow below US$19M. Either alone is recoverable; both together would confirm both the consensus "Phihong-style fade" frame and the consensus "structural cash break" frame in the same disclosure, and would also remove the operating-quality precondition for the SOTP thesis. The variant view does not require AI/HPC heroics — it requires the protected slice to keep earning at the segment margins it has demonstrated. If both numbers print at the upper bound of our refutation band, we are not just wrong on the disagreement — we have misread the underlying business as well.

The second way to be wrong is on the spin disclosure. If the CSRC prospectus shows post-IPO economic interest below 80% retained by 1979.HK, or sponsor indicative pricing below 20× P/E, or material RPT carve-outs of the smart-charger or new-energy P&L into the SpinCo, the arithmetic disagreement collapses. The bear's "engineered to transfer value" framing has real evidence behind it — Golden Ocean Copper cap raised 2.8× in two years on a commodity input, Huizhou Share Award Scheme adopted Feb 2024 (two years before the May 2026 spin announcement), US$3.3M of FY2025 KMP value already directed through the subsidiary scheme. We are betting that the regulatory disclosure resolves more accounting risk than two years of interim reports — but the disclosure has not yet printed and the family-control mechanism that could route value away from listed-co shareholders is structurally in place.

The third way to be wrong is more subtle and matters most for the operating disagreement. The effective tax rate stepped down from 22.79% in FY2020 to 10.75% in FY2025 — recognition of previously unrecognised HK subsidiary tax losses plus Vietnam preferential rate plus HNTE 15% status at five PRC subsidiaries. A return to a 15-16% mid-cycle ETR removes 8-16% of reported net income — about US$6.5-11.5M per year — which compresses the ROE-versus-peer-set comparison and tightens the operating-margin gap with Chicony Power. The variant view does not depend on the ETR but the consensus mathematical SOTP arithmetic in the bull case (US$28.1M Huizhou net-income share at 20× = US$562M) does. ETR normalisation alone would not refute the disagreement but it would compress the magnitude of the implied re-rate.

The fourth way is a single Fortune-500 platform loss at the 15.1% customer in a year when customer broadening has not yet absorbed it. That would erase the recent top-5 concentration improvement (66% → 59.8%) and reset the peer-set positioning by combining the consensus "concentration is durable" frame with measured revenue contraction.

The first thing to watch is the FY2026 H1 segment-information note around 22 August 2026 — smart-chargers segment gross margin and operating cash flow read together resolve more of the variant view than any other single disclosure inside the next nine months.

---

Sources: business-claude.md, long-term-thesis-claude.md, moat-claude.md, competition-claude.md, numbers-claude.md, forensics-claude.md, people-claude.md, story-claude.md, technicals-claude.md, research-claude.md, short-interest-claude.md, catalysts-claude.md, verdict-claude.md. Underlying data from FY2025 annual report (PRNewswire, 20 March 2026), spin-off announcement (HKEX, 14 May 2026), H1 2025 interim release (PRNewswire, 22 August 2025), peer financials (Yahoo Finance, 21 May 2026). All figures in US dollars (converted from HKD at historical FX rates); price reference 21 May 2026 close US$0.37.

Liquidity & Technical

Figures converted from HKD at historical FX rates — see data/company.json.fx_rates. The HKD is pegged to the USD at HK$7.80 = US$1, so conversion drift over the 10-year window is under 1%. Ratios, margins, percentages, share counts, RSI/MACD, and volatility measures are unitless and unchanged.

Ten Pao is institutionally tradable only at small absolute sizes — five-day capacity at 20% participation is roughly $1.4M, supporting a 5%-weight position in a fund of about $28M. The tape is constructively bullish on a 3–6 month view: price sits 24.6% above the 200-day SMA with a positive MACD and a 19.8× volume breakout on 14 May, but 30-day realized volatility at 64% sits above the 80th percentile of the 5-year range, signalling that risk premium is also being repriced.

1. Portfolio implementation verdict

2. Price snapshot

3. Ten-year price with 50 / 200-day moving averages

Price is above the 200-day SMA ($0.373 vs $0.299; +24.6%). A golden cross occurred on 26 June 2025 and remains intact. The 10-year chart shows two prior cycles — a 2017 spike to $0.33 that round-tripped, and a 2020–21 spike to $0.30 that faded to $0.13 — but the current move began from a clean $0.13 base in early 2024, more than 18 months of accumulation. This is the first time price has sustained above $0.26 in five years, and the current close of $0.373 sits within 14% of the all-time high ($0.432, set 14 May 2026).

4. Relative strength — three-year rebased trajectory

The benchmark series (broad-market or sector ETF) did not load for this run, so a strict relative-strength comparison is not available. In absolute terms the stock has returned +147% over three years (rebased 100 → 247) with the entire gain concentrated in the trailing 12 months (+78.5%) and most of the acceleration in the past 6 months. The shape — long base 2023, breakout April 2024, second leg from October 2025 — is consistent with a rerating story rather than a single news spike.

5. Momentum — RSI(14) and MACD histogram (last 18 months)

RSI sits at 66.3 — close to but not yet at the conventional overbought line (70), with room before the tape becomes mechanically overextended. The MACD histogram flipped positive on 14 May after a six-week negative run and has expanded for five straight sessions, the cleanest momentum re-acceleration of the past 12 months. Near-term (1–3 month) momentum is constructive but stretched; another 5–10% move would push RSI into overbought territory.

6. Volume, volatility, and sponsorship

The 14 May 2026 session is the meaningful event: 25.3M shares traded at 19.8× the 50-day average and price closed up 21%, taking out the prior 52-week high. The 50-day average volume has subsequently more than doubled (from ~0.7M to 2.1M shares), so this is not a one-day blip — sponsorship is building. The 2018 and 2024 spikes shown in the table happened at much smaller absolute share counts on a then-illiquid stock and are statistical artefacts rather than fundable signals.

Realised volatility has been re-anchored at the stressed end of the 5-year range: the current 64% sits above the 80th-percentile band (56%) and the prior six-month median ran 60–65%. This is not a calm tape that is drifting higher — the market is paying for the recent run by demanding a wider risk premium. Risk-budgeting note: stop distances and position sizing should be set off ATR ($0.023, ~6% of price), not off the 50-day average true range you'd expect on a normal equity.

7. Institutional liquidity panel

Reader note. The pipeline's auto-loader flagged "liquidity unknown" because the share count was not in the price feed. Using the 1,030.4M shares disclosed in the FY2025 Share Option Scheme (implied market cap $383M), I recompute the capacity tables below.

Note the gap between 20-day ADV (3.83M shares) and 60-day ADV (1.85M shares): the running 50-day average has more than doubled in the past month off the 14 May breakout. The 21.8% annual turnover figure is the smoother medium-term read — adequate for a Hong Kong microcap but not deep.

Fund-capacity table — what AUM does this stock support?

The implementable universe is small: at the aggressive 20% participation rate, a 5%-weight position is implementable for funds up to about $28M. A 5% weight in a $128M fund would take more than three weeks of accumulation. Even a 2% weight caps at $71M AUM at 20% participation. This is a name for boutique sleeves and special-situation books, not for diversified mid/large-cap funds.

Liquidation runway — exiting an issuer-level position

Price-range proxy. Median 60-day daily range is 3.4% of price — comfortably above the 2% institutional-friction threshold. This is not a name where a 1% slippage assumption holds for a 1% mcap unwind; impact costs are real and rise sharply once 5-day capacity is breached.

Bottom line on liquidity. The largest position that clears within 5 trading days at 20% ADV is ~0.5% of market cap ($1.9M). At the more conservative 10% participation, the 5-day clearing size drops to ~0.2% of mcap ($0.7M). Larger sizes are possible only with multi-week scaling or specialist block desks willing to accept the wide intraday range.

8. Technical scorecard + 3–6 month stance

Net score: +3 of 6. Stance: bullish on the 3-to-6-month horizon. The setup is a multi-month base breakout confirmed by a real volume signature, with both trend and momentum aligned. The bear-case reservation is the stressed volatility regime: the market is willing to pay for the move but is also pricing in chunky drawdowns, so position sizing needs to be calibrated to a wider stop distance than usual for a sub-$0.50 name. Decisive confirmation comes on a sustained close above $0.434 (clearing the 14 May all-time high and opening blue-sky territory). The bullish thesis fails on a close below $0.319 — that level loses the 50-day SMA ($0.321) and the pre-breakout consolidation base, and would convert the May 14 spike into a failed-breakout shake-out.

Liquidity is the constraint, not the tape. For sub-$65M AUM strategies the position is meaningfully implementable on weakness pullbacks. For larger mandates the correct action is build slowly over multiple weeks or watchlist only — either accept the multi-week accumulation timetable or wait for a higher-liquidity entry (a secondary placement, index inclusion, or sustained turnover step-up).

Short Interest & Thesis

Figures converted from HKD at historical FX rates — see data/company.json.fx_rates. Ratios, margins, multiples, share counts, dates, and names are unitless and unchanged.

Short interest is not a decision-useful variable for 1979.HK today. No per-stock reported short-interest position data is staged for this HKEX listing, no public short-seller report or activist campaign on Ten Pao was found in source-policy-filtered web search, and the share register makes meaningful crowding structurally hard: Chairman Hung controls 64.88%, Fidelity holds 7.76%, and the remaining public float is roughly 27.4% — about 282M shares — against a 20-day ADV of 3.83M shares (about US$1.4M). The institutionally relevant "thesis-risk" surface here is forensic, not short-side: a controlling-family lattice, a related-party copper supplier, a restricted-cash / bank-acceptance-bill structure that flatters headline gearing, and a pre-IPO equity scheme inside the Huizhou subsidiary that is about to be tested by the proposed A-share spin-off prospectus.

Source-class snapshot

The five short-side data classes that would normally drive a positioning page (outstanding short interest, daily short-sale volume, threshold disclosures, borrow indicators, peer context) are all empty or not applicable. The only decision-useful evidence comes from the share register, ADV, the forensic file, and the explicit absence of any public short campaign.

Reported short interest — unavailable, not zero

This is an evidence-availability problem, not a fundamental observation. HKEX disseminates aggregate short-sale turnover at the exchange level and publishes a list of designated securities eligible for short selling, but it does not publish a per-stock outstanding short-interest series in a form that the pipeline's v1 fetcher consumes deterministically. A retail short-interest aggregator (ORTEX) does carry an entry for SEHK:1979, but its Shorts panel records "No Data Available" — consistent with a sub-US$400M micro-cap where lendable supply and reporting coverage are both thin. Treat the absence as "we cannot measure it" rather than "it is zero."

Public short-thesis evidence — none identified

Multiple explicit searches across general web and major financial-press domains, plus the forensic specialist's own cross-check, returned nothing name-specific. There is no public short-seller report, accounting allegation set, activist campaign, regulatory probe, or class-action complaint against 1979.HK in the searched sources. That does not prove none exists in non-English Chinese-language venues that the search did not penetrate, but the institutionally relevant English-language and HK-listed coverage is empty.

Crowding vs liquidity — structural, not measured

The structural setup is squeeze-asymmetric, not squeeze-prone in the retail sense. Effective non-strategic float is around 282M shares — small, illiquid, and likely under-lent — so any meaningful short would face hard locates and a cost-of-borrow that the run did not measure. By the same token, no aggregate short can be inferred to exist at scale; the conditions for crowding don't really obtain because the borrow side is too thin to support it. The right framing for a PM is "this is a name where short-side data is unmeasured, but the borrow market would be hostile if it were measured."

Borrow pressure — not measured; inferred

No borrow-cost, utilisation, rebate, or locate data was staged. The inference from market structure is that borrow would be expensive and locate-constrained, not cheap — but that is structural reasoning, not a measurement, and should not be treated as evidence of an active borrow-cost spike.

May 14 volume spike — catalyst, not positioning

The most relevant recent print — May 14, 2026, 25.3M shares at +21.0% — is a catalyst-driven upmove tied to the same-day announcement of a proposed A-share spin-off and separate listing of Ten Pao Electronics (Huizhou). This is a positive catalyst with material positioning consequences, but it is not a short-cover signature and it should not be read as a positioning unwind.

Adjacent thesis risk — what a short would actually argue

If a credible short pitch on 1979.HK were ever written, this is the surface it would attack: the value of the listed parent's claim on the Huizhou operating company after the A-share spin-off, the durability of the tax-rate step-down, and the related-party copper supplier whose cap was raised 2.6x ahead of FY2025. None of these is a near-term, fraud-grade allegation; the audit opinion, accrual ratio, DSO trend, and capital structure controls are properly in place. A PM should treat these as position-sizing limiters and as the line items most likely to be re-priced by an unfavourable disclosure inside the Huizhou A-share prospectus — not as a current short trigger.

What would change this page

Evidence quality