Sungrow Power Supply: Q3 Profit Fell 67.67% as Margins Relapsed, and at CNY 65.29 the Shares Still Sit 48% Above Conservative Value

Buscando ações que quintuplicam em dez anos entre grandes empresas de crescimento — pressionando a questão do potencial: "Pode ficar muito maior?"

• PHow high is its market ceiling — is it growing a slice of an existing pie, or creating an entirely new market?6/10

  R

  Verdict: Sungrow is growing a large existing pie, not creating a new market. The ceiling is high in battery storage, flat to falling in PV inverters, and fenced off by politics in the US and parts of Europe.

  Storage is where the ceiling sits. Global lithium storage installs reached ≈ 140 GWh in H1 2026, up ≈ 30%, with overseas ≈ 80 GWh (≈ +40%) and Europe up 70–80% (IR record 20260828). Management expects installs to grow 30–50% a year for three years and Europe by more than 50% in 2027, pointing to the EU's 200 GW-by-2030 target (IR record 20260427). Sungrow is already a scaled incumbent here: it shipped 43 GWh in 2025 and ranked #2 global integrator behind Tesla, in a market where Chinese integrators held 76% (Wood Mackenzie, 2026-06-29).

  PV inverters are going the other way. Global PV installs fell ≈ 43% to ≈ 177 GWac in H1 2026. China dropped from 212 to 72 GWac after the Document 136 market reform, and overseas grew only ≈ 8%. Sungrow's inverter shipments fell from 76 GW to 66 GW (IR record 20260828).

  Two things cap the ceiling Sungrow can actually reach. First, value grows more slowly than volume: storage shipments rose 28% in H1 2026, yet storage revenue fell 13.18%, so revenue per GWh dropped ≈ 32%. Second, access is narrowing. The FCC added foreign-made inverters to its Covered List (FCC DA 26–786), and the EU's funding ban on high-risk inverters takes effect on 2026-11-01 (Euronews, 2026-09-23).

  The only new-market element is data-centre (AIDC) storage and power. Its backlog is ≈ 2 GWh, against 43 GWh of total 2025 storage shipments. Net: the pie is big and growing but commoditising and contested, and Sungrow is not creating a category.

  31 de outubro de 2026
• PCan its revenue at least double over the next five years? Is that growth driven mainly by volume, price, or new businesses?3/10

  R

  Verdict: A clean doubling in five years is unlikely in the base case. Storage volume is growing, but falling prices are eating it, and the 2021–25 surge owed much to mix shift and one-off mega-projects.

  Start with the base. FY2025 revenue peaked at CNY 89.18 bn, and the TTM to Q3 2026 is CNY 71.64 bn (−24.0%) (Q3 2026 report). Doubling would take ≈ CNY 178 bn from the peak or ≈ CNY 143 bn from TTM.

  The 2021–25 record (24.14 → 89.18 bn) is not clean organic growth. It was mix: storage rose from CNY 3.14 bn (13.0% of revenue) to 37.29 bn, while project development fell from 40.1% to 18.6% (Eastmoney F10 segments). Single projects swing the numbers: one Saudi 7.8 GWh project booked ≈ CNY 4 bn in Q1 2025 (IR record 20260427).

  Strip out price and cycle effects, and 2026 shows volume without value. H1 storage shipments rose 28% to 25 GWh, but storage revenue fell 13.18%, so revenue per GWh dropped from ≈ CNY 0.91 bn to ≈ 0.62 bn. Inverter shipments fell from 76 GW to 66 GW. The revenue drop came from China (−55%), the Middle East base effect (−91%) and development (8.4 → 1.26 bn) (IR record 20260828). The September list-price rise of 5–15% passes through copper and aluminium costs; it is not pricing power that adds growth (pv magazine, 2026-09-14).

  Looking ahead, storage volume at management's 30%+ market growth would more than double in three years if Sungrow holds share. Even so, the report's 2027E revenue is only CNY 82 bn in the base case and 95 bn in the bull case. Doubling needs volume to outrun price deflation for years, Europe and emerging markets to offset a shrinking US, and AIDC products to add revenue from 2028. Any doubling would be volume-led, and the evidence does not yet support it.

  31 de outubro de 2026
• PFive years out, what takes over as the next growth engine? Does that “second curve” exist today?4/10

  R

  Verdict: The last second curve, storage, is now the first curve. The next candidate, data-centre (AIDC) power and storage including solid-state transformers (SST), exists today only as a pilot-scale pipeline with no disclosed revenue.

  Sungrow has built a second curve before, which counts in its favour. Storage went from CNY 3.14 bn (13.0% of revenue, GM 14.11%) in 2021 to 37.29 bn (41.8%, GM 36.49%) in 2025 (Eastmoney F10 segments), and reached 50.0% of revenue in H1 2026. Storage is now the core, and its margin is under pressure: H1 2026 storage GM was 32.43%, down 7.49 pp.

  Candidates for the growth engine five years out:

  • AIDC storage: backlog ≈ 2 GWh and a pipeline of "a dozen-plus GWh" (IR record 20260828). The backlog is about 5% of the 43 GWh of storage shipped in 2025.
  • AIDC SST: management expects first orders from Q4 2026, small batches in 2027 and scale in 2028 (same IR record). No revenue, margin or order value has been disclosed.
  • European localisation: the Poland plant (20 GW inverters, 12.5 GWh storage, ≈ €230 m) (AAP/Cision, 2026-02-05). This is a geographic hedge against EU vendor rules, not a new product curve.
  • Project development is shrinking, not emerging: 4.1% of H1 2026 revenue.

  The optionality is real. Its in-house power electronics already earn a 42.72% inverter GM, and that skill set carries over to 800V data-centre power and grid-forming systems. But today the second curve is a roadmap and a pipeline, not a business with numbers. Nothing in the filings lets anyone size it against a CNY 70–90 bn revenue base. The first checkpoint is whether SST orders actually land in Q4 2026.

  31 de outubro de 2026
• PWhat is its core competitive advantage? Will that moat widen or narrow over the next three to five years?5/10

  R

  Verdict: The moat is real but narrow. It rests on pricing power from in-house power electronics in inverters (holding or widening outside China), on bankability in utility storage (narrowing), and on a net-cash balance sheet. Over three to five years the overall moat is more likely to narrow than widen.

  Inverters are the strongest leg. In H1 2026 inverter GM rose to 42.72% (+6.98 pp) even as segment revenue fell ≈ 19% (Eastmoney F10 segments), and the company claims a ≈ 10% price premium over peers (IR record 20260427). A premium that survives a volume decline is genuine brand and technology pricing.

  Storage bankability is the weaker leg. Sungrow ranked #2 global integrator by 2025 shipments and #1 in Wood Mackenzie's first comprehensive ranking (ESS News, 2026-07-14). Yet the top three's share fell from 36% to 30% in 2025, and BYD climbed five places to #3 (Wood Mackenzie). Sungrow buys all its cells and cannot fully pass through lithium costs. CATL and BYD, by contrast, can bundle cells with their own power conversion systems (PCS) and compete on price.

  Supporting assets: 30+ overseas branches and 520 service points (H1 2026 summary), and net cash of ≈ CNY 24.39 bn (Q3 2026 report).

  The peer check is sobering. In 9M 2026, Deye grew revenue 94% with a 36.8% ROE, while Sungrow's revenue fell 26.4% and its ROE was 13.5% (Eastmoney F10). The moat protects margin in its niche but is not stopping share loss in the growth segments.

  On direction: grid-forming technology and data-centre power could widen the technical edge. Against that, storage is commoditising, there is no cell margin, and market access depends on US and EU politics. Those three narrow it. On balance the moat is narrowing.

  31 de outubro de 2026
• PIf its core business were disrupted, does it have the DNA to reinvent itself? How does it handle mistakes and bad news?5/10

  R

  Verdict: Sungrow has reinvented itself once, from inverters and project development into storage, and it adapts quickly in commercial terms. Its handling of bad news, however, is optimistic first and revised later rather than candid up front.

  The reinvention record is strong. Between 2021 and 2025 project development fell from 40.1% to 18.6% of revenue, while storage rose from 13.0% to 41.8% and its GM from 14.11% to 36.49% (Eastmoney F10 segments). The company chose an asset-light, cell-neutral model on purpose.

  It also adapts fast in 2026:

  • It let China revenue fall from CNY 18.1 bn to 8.2 bn rather than chase low-margin volume (IR record 20260828).
  • It raised list prices 5–15% from 2026-09-20, the second rise of the year (pv magazine).
  • It is localising in Poland (AAP/Cision).

  The test for disruption is different. If cell makers bundling cells with power conversion is what breaks storage economics, the cell-neutral model is the exposed part. Management's answer so far is only that it is "considering" cross-shareholdings with cell makers (IR record 20260427). That capability is unproven.

  On mistakes and bad news, the record is mixed:

  • In April management said storage GM would decline only gradually and that the US business "will not vanish overnight". By August it said US volume "will gradually shrink" and declined to build US plants. Q3 group GM then fell to ≈ 29.4%.
  • The two IR records give different Q1 storage GMs (≈ 30% in April, ≈ 32% in August).
  • There is no numeric guidance.
  • Buybacks of CNY 325 m were executed at CNY 97.81–114.99, with none at the September–October lows (buyback progress, 2026-10-09).

  The DNA for change is there commercially. Disclosure culture is the weaker part.

  31 de outubro de 2026
• PDoes management — the founders especially — hold a long-term view with interests deeply tied to the company? Are they willing to sacrifice current profit for the payoff five to ten years out?7/10

  R

  Verdict: Strongly aligned founder control. The founder is still running the company and holds about 30% of it directly. The evidence shows a margin-first, capital-disciplined owner rather than one visibly giving up current profit for a ten-year bet.

  Alignment is not in doubt. Cao Renxian founded Sungrow in Hefei in 1997 and is still chairman (Wikipedia). At 2026-09-30 he held 631.41 m shares, 30.46% of the company, in his own name as the largest shareholder. Only 20.61 m of those shares are pledged, 3.3% of his stake (Q3 2026 report). No holding-company layer or dual-class structure is involved; at CNY 65.29 the stake is worth ≈ CNY 41 bn.

  Long-term behaviour is mixed:

  • In 2021–25 he backed storage when it was a 13%-of-revenue, 14%-GM line, and it became the core (Eastmoney F10 segments). That was a genuine multi-year bet.
  • In 2026 he chose margin over volume. China revenue fell 55%, and management declined to build US plants even though that rules out the US 45X manufacturing credit (a US production tax credit) (IR record 20260828). That is capital discipline. It is also the opposite of giving up profit for a market position.
  • Spending on the future is real but moderate: capex of ≈ CNY 3 bn a year, including the ≈ €230 m Poland plant. R&D spend is not covered in the evidence base.
  • Shareholder returns are measured: an interim payout of 25% and a CNY 0.5–1.0 bn buyback, of which CNY 325 m was spent at prices above CNY 97 (buyback progress).
  • An H-share listing was re-filed in 2026-04 despite ≈ CNY 24 bn of net cash. Its size and dilution are unknown (cninfo, 2026-04-24).

  Willingness to sacrifice profit shows up in strategic pivots, not in deliberately depressed margins.

  31 de outubro de 2026
• PIf it disappeared tomorrow, how badly would customers miss it? Is the way it grows sustainable, without relying on harm to society or regulators?4/10

  R

  Verdict: Customers value Sungrow but could replace it, and its growth is good for society, since it is decarbonisation equipment. The binding constraint is national-security regulation in the US and the EU, which makes its growth heavily dependent on regulators.

  Customers would miss it, but they would not be stranded. Sungrow ranked #1 in Wood Mackenzie's comprehensive integrator ranking (ESS News), claims a ≈ 10% inverter price premium (IR record 20260427), and runs 520 service points. Its installed fleet relies on its firmware and service. Substitutes are plentiful, though: 8 of the top 10 integrators are Chinese, and Tesla, BYD and CATL all bid for the same tenders (Wood Mackenzie). New projects would re-source within a tender cycle; the pain would fall on servicing existing assets.

  Access is the real issue, tightening on three fronts:

  • US: Foreign-made inverters went on the FCC Covered List on 2026-07-28, so new models cannot be authorised without Conditional Approval (FCC DA 26–786). Sungrow has none so far. Existing authorisations and firmware updates remain allowed until at least 2029-01-01 (Davis Wright Tremaine). EO 14420 lets the US Department of Energy (DOE) bar Chinese-linked bulk-power equipment connected at ≥ 69 kV, including equipment installed before the order, with DOE rules due ≈ late December 2026 (Trade Compliance Resource Hub). Separately, the US tax credit for storage projects (48E) now requires ≥ 55% of project cost to come from non-Chinese-linked (non-FEOC) suppliers, rising to 75% by 2030. Sungrow has no US plant.
  • EU: The funding ban on high-risk inverters starts 2026-11-01 (Euronews). The company estimates it affects 10–20% of European projects. A Cybersecurity Act revision could turn it into a binding market restriction (ESS News, 2026-05-04).
  • China: The battery export VAT rebate falls to zero from 2027-01-01 (MoF/STA notice).

  None of this stems from social harm, but growth now hinges on how regulators read Chinese ownership.

  31 de outubro de 2026
• PWhat are the unit economics of this business (gross margin, incremental returns)? Do they get better or worse at scale? Where does the money it earns go?5/10

  R

  Verdict: Unit economics were excellent and improved with scale in 2021–25. In 2026 they are deteriorating: storage price deflation is compressing margins, and cash is being absorbed by working capital rather than returned.

  The scale years were strong:

  • Gross margin rose from 22.25% to 31.83% and operating margin from 7.9% to 18.3% between 2021 and 2025. Weighted ROE was 40.96%, 33.99% and 31.26% in 2023–25 (FY2025 summary; Eastmoney F10).
  • Incremental returns were high. From 2023 to 2025 revenue grew CNY 16.93 bn and attributable NI CNY 4.02 bn, a ≈ 24% incremental net margin.
  • The business needs little fixed capital: 2025 capex was CNY 3.01 bn on 89.18 bn of revenue, and FCF rose from 4.24 to 9.28 to 13.91 bn.

  The 2026 reversal shows the limit of scale in storage. H1 storage GM fell to 32.43% (−7.49 pp), and revenue per GWh fell ≈ 32% even as shipments rose 28% (IR record 20260828). Q3 group GM was ≈ 29.4%. Because Sungrow buys its cells, scale benefits accrue in power electronics (inverter GM rose to 42.72%), while storage margin follows cell prices and competition.

  The cash went into working capital. Cumulative OCF/NI was 91% in 2021–25 but only 55% in 9M 2026, and Q3 OCF was ≈ −CNY 0.09 bn. Inventory reached CNY 40.97 bn (+50.33% YTD, ≈ 299 days of COGS) and prepayments rose 305% (Q3 2026 report). The rest went to:

  • capex of ≈ CNY 3 bn a year (Poland and Egypt plants);
  • dividends of CNY 1.33 per share, a 2.0% yield (Eastmoney dividends);
  • CNY 325 m of buybacks.

  Net cash slipped from CNY 27.84 bn to 24.39 bn. Contract liabilities rising 37.64% to CNY 14.67 bn is a partial offset. The model is asset-light but working-capital heavy, and economics now worsen with volume unless storage pricing stabilises.

  31 de outubro de 2026
• PFor it to rise fivefold in ten years, what conditions must all hold at once? Are they realistic? What expectations does today's share price already imply?3/10

  R

  Verdict: A 5x in ten years needs attributable profit 2.5–3x above the 2025 peak, plus open Western markets and stable storage margins. Several of those conditions are moving the wrong way, so a 5x is low-probability. Today's price embeds roughly flat TTM earnings.

  The arithmetic: 5x the CNY 135.36 bn market cap is ≈ CNY 677 bn. At 20x, near the 5-year median P/E of 21.0x (Eastmoney valuation history), that needs NI of ≈ CNY 33.8 bn. At 17x it needs ≈ 39.8 bn. That is 2.5–3.0x the 2025 peak of CNY 13.46 bn and 4.1–4.9x TTM NI of 8.18 bn (Q3 2026 report). At a 12–14% net margin, it implies revenue of ≈ CNY 240–330 bn, versus 89.18 bn at the peak.

  All of these would have to hold:

  1. Storage volume compounds for a decade at the market's 30%+ rate with Sungrow holding its #2 share; yet the top-three share fell from 36% to 30% in 2025 (Wood Mackenzie).
  2. Storage GM settles at or above ≈ 30% instead of converging toward CATL's 22.7%. H1 2026 was 32.43% and falling.
  3. US and EU access stays broadly open; EO 14420 and EU high-risk-vendor rules point the other way (Trade Compliance Resource Hub).
  4. AIDC power becomes a scaled second business after 2028.
  5. Cash conversion returns to ≈ 90%, from 55% in 9M 2026.

  Is the P/E distorted? Only mildly. Q3's −CNY 0.49 bn fair-value loss put Q3 recurring NI (1.69 bn) above reported (1.34 bn), but 9M recurring NI of 5.97 bn is ≈ 10% below reported 6.60 bn, so the 16.55x TTM P/E (Tencent Finance) is, if anything, slightly understated. The bigger distortion is stale consensus: 13.90 bn for 2026 implies Q4 NI of ≈ 7.30 bn (10jqka).

  Embedded expectation: at 17x, CNY 65.29 capitalises ≈ CNY 7.9 bn of NI, roughly TTM held flat. The price assumes neither recovery nor further erosion. A 5x needs a rebuild today's data do not show.

  31 de outubro de 2026
• PWhy hasn't the market grasped all this yet — does it not understand, not respect it, or not see far enough? What would become the “narrative inflection point”?3/10

  R

  Verdict: The market has not missed Sungrow. It sees the margin and policy risks clearly and is discounting them. If anything is mispriced, it is the time horizon: the market cannot see past the 2026–27 politics. Near term, consensus may even be too optimistic.

  The market is not failing to understand or looking down on the stock. It has sold it hard:

  • The share price is down 61.5% YTD from CNY 169.76 (NBD, 2026-10-30).
  • P/B of 2.74x is the lowest in five years (≈ 0.1st percentile), and the TTM P/E of 16.55x sits at ≈ the 33rd percentile (Eastmoney valuation history).
  • Northbound (HKSCC) holdings fell from 8.08% to 4.59% of the company in one quarter, a 72.38 m-share exit by foreign investors (Q3 2026 report).

  About a third of the fall since August is company-specific: the stock dropped 12.24% on the EO 14420 day and 6.34% on the H1 results (21jingji). The other two-thirds tracked the ChiNext index's 23.02% October slump (Sina Finance). The Q3 miss is not yet in consensus, which still implies Q4 NI of ≈ CNY 7.30 bn against Q3's 1.34 bn.

  Where the market may be short-sighted is in treating Sungrow as a China solar cyclical. Overseas markets supplied 73.4% of H1 2026 revenue, inverter GM is 42.72%, global storage demand is still growing ≈ 30%, and net cash is ≈ 18% of market cap.

  Inflection points:

  1. The DOE rules under EO 14420, due ≈ late December 2026. A rule that spares distribution-level and storage equipment below 69 kV would lift the main overhang (IR record 20260828).
  2. FY2026 storage GM ≥ 28–30% and Q4 OCF > CNY 3 bn, showing the CNY 40.97 bn inventory turning into cash.
  3. The first AIDC SST orders, expected from Q4 2026.
  4. EU Cybersecurity Act decisions and the Poland plant clearing EU rules.

  Consensus cuts and a harsh DOE rule are equally live negatives.

  31 de outubro de 2026