Fair-Value Report — Aluminum Corporation of China (2600.HK)
Verdict
The signal is real but weak. A large active fund manager bought on-exchange, and Chinalco (the parent) separately announced a stake increase. The stock has fallen 55% from its 52-week high even though H1 2026 profit rose 68%. The market is pricing a fade in aluminium margins, not a collapse. At HK$6.805 the shares sit roughly at our base-case value, with an asymmetric range (+53% / −32%). That is not enough margin of safety for a Buy. Rating: Hold.
Rating-caps box
- Consistency / +25% rule: base-case FV is −1% vs the live price, so no Buy of any kind. This alone fixes the rating at Hold.
- NON_INSIDER_BUYER (fund manager): would cap at Speculative Buy even if the base case cleared +25%.
- FALLING_KNIFE (55% below the 52-week high): blocks Moderate Buy / Buy. The next-print test is the protection; see Catalysts.
- TOKEN_BUY: fired on total market cap (0.19%). Against the H-share class the trade is larger (see signal box). It is not a Hold cap.
- Not fired: PRICE_MOVED (live is −8.3% vs Lazard's price), RUN_UP, EXTENDED, MICRO_ILLIQUID, RECENT_IPO. No going-concern, placing, or lock-up issue found (see red-flag checks, with limits stated there).
Insider-signal box
- Filer: Lazard Asset Management LLC (HKEX DI filing CS20261021E00372, report date 22 Oct 2026, event date 15 Oct 2026).
- Reason 1101: on-exchange purchase of 31,480,000 H shares at an average HK$7.424 (≈ HK$233.7m).
- Holding: 234,106,000 (5.94% of the H-share class) → 265,586,000 (6.73%), a +0.79-point step. Long only, no short.
- Context: the 6.73% is a percentage of the H-share class (about 3.95bn shares implied), not of all 17.15bn shares. Against H-share market value (~HK$26.9bn) the purchase is ~0.9%. Lazard's whole stake is worth ≈ HK$1.8bn at the live price.
- Interpretation: a discretionary asset manager adding on a large decline. This is allocation buying with weak timing information, not management conviction. Lazard bought at HK$7.424 and the shares have since fallen 8.3%.
- Related, more relevant: on 20 Jul 2026 Chinalco (≈33.55% holder with affiliates) announced it would buy RMB1.0–2.0bn of A and H shares (≤2% of share capital, 12 months) and not reduce its holding. This is real insider support, but it is a ceiling-limited, price-insensitive programme, not a signal on timing. I could not verify how much has been executed.
- I could not open the HKEX DI form directly in this run; the figures are from the scanner's parsed copy of the filing.
Verified snapshot
| Item (30 Jun 2026, RMB) | Value | Note |
|---|---|---|
| H1 revenue | 125.4bn | +7.7% YoY |
| H1 net profit to owners | 11.87bn | +67.9% YoY; basic EPS RMB0.693 (H1 2025: 0.413) |
| Primary aluminium segment profit | 26.6bn | +228% YoY; alumina segment lost RMB0.82bn |
| Operating cash flow | 26.2bn | +84% YoY |
| Equity to owners / non-controlling | 84.5bn / 55.5bn | Large minority interests: the profit to owners is not the whole group |
| Debt / cash | 59.8bn / 43.1bn | Net debt ≈ RMB16.7bn; asset-liability ratio 42.76%, down 3.25 pts |
| Interim dividend | RMB0.276 | 39.9% payout; payable by 23 Oct 2026 |
FX: I use ~1.11 HKD per RMB, implied by the source's HK$5.47 NAV vs RMB84.5bn ÷ 17.15bn shares. Verify before relying on HKD figures at the margin. FY2025 profit of ~RMB15.4bn is derived (trailing-12-month profit of RMB20.16bn per StockAnalysis less the H1 increase of ~RMB4.8bn), not read from the annual report.
Business
Chalco is China's largest integrated alumina and primary-aluminium producer, with energy and trading segments. H1 2026 output: primary aluminium 4.02mt (+1.3%), metallurgical alumina 7.38mt (−14.2%), refined alumina 2.34mt (+12.5%). About 70% of alumina needs are reportedly self-supplied. The earnings driver is the aluminium-to-cost spread: primary-aluminium profit more than tripled while alumina turned to a loss as alumina prices fell. Shanghai A00 spot was RMB23,710/t on 22 Oct 2026 and the SHFE Dec-26 contract closed at RMB23,300/t (10jqka). So the spread that produced H1 profit still exists today.
Valuation
Method: cyclical, so a normalised-earnings P/E cross-checked with P/B, not a long-horizon DCF. I did not run the stock-valuation DCF scripts: the detail needed for a defensible cost curve (unit costs, power mix, 5–7-year margins) was not available from verifiable sources in this run. Earnings are anchored on verifiable inputs: H1 2026 actual, the derived FY2025 level, and Goldman's published −50% 2027 view.
| Scenario | Net profit (RMB) | EPS (HK$) | Multiple | FV (HK$) | vs HK$6.805 |
|---|---|---|---|---|---|
| Bull: H1 run-rate holds (spot ~RMB23,300/t) | 23.0bn | 1.49 | 7.0x | 10.4 | +53% |
| Base: half the H1 gain reverses to ~FY2025 level (consistent with BofA's RMB21,000/t 2027 assumption) | 16.0bn | 1.04 | 6.5x | 6.7 | −1% |
| Bear: Goldman case, 2027 recurring profit −50% vs H1 run-rate; trough multiple | 11.9bn | 0.77 | 6.0x | 4.6 | −32% |
Bear cross-check: 0.85x NAV (HK$5.47) = HK$4.65. NAV is not a floor here. At trough, ROE would fall toward ~14%, and a third of group equity belongs to minorities.
Annualising H1 alone gives EPS RMB1.39 (HK$1.54), or 4.4x P/E at the live price. That is the bull case's raw material, and it is exactly what the market refuses to capitalise.
Reverse read. HK$6.805 is RMB6.13 per share. That equals 6.6x our base-case EPS, or 8.8x the Goldman-bear EPS of RMB0.69. The market is therefore discounting roughly a halving of H1 profit, consistent with the Goldman downgrade (Sell, TP HK$7.5) and the "capacity above 48mt by end-2026" argument. The price does not require aluminium to fall below RMB20,000/t, but BofA (Buy, TP HK$10.5, 14 Jul) argues exactly that the stock discounts sub-RMB20,000/t against ~RMB23,000/t spot.
Peer table (secondary sources, undated forward P/Es; treat as indicative)
| Company | P/E | Source note |
|---|---|---|
| China Hongqiao (1378) | ~10x trailing; brokers 8–16x 2026e | Simply Wall St; broker notes |
| Yunnan Aluminium (000807) | ~12x forward | aggregator |
| Shandong Nanshan Aluminium (600219) | ~7.7x forward | aggregator |
| Nanshan Aluminium Intl (HK) | 11x / 7x 2026e/27e | broker note |
| Peer median | ≈10–11x | Chalco at 6.6x base EPS / 4.4x annualised H1 is a clear discount |
Caveat: only four peers, with mixed dates and bases; the aggregator figures for Chalco itself (7.4x trailing, 8.6x 2026e) look stale against H1 actuals. Our 6.5x target is below the peer median because earnings are at a spread peak. Part of the discount reflects Chalco's alumina exposure and minority leakage.
Red-flag checks (Step 1b)
- Filings sweep: I reviewed the H1 2026 results through secondary sources, since the HKEX titleSearch could not be queried directly in this run. No going-concern, material-uncertainty, overdue-filing, default, qualified-opinion, or suspension language appeared in any source. Balance sheet is improving (gearing −3.25 pts, short-term borrowings down). This is "none found", not a verified clean sweep.
- Dilution / supply: no placing, subscription, rights issue or convertible found in ±30 days. The only capital-markets items found were the Chinalco buy-in (supportive) and an EGM notice (content not checked). JPMorgan was reported to have raised its H-share stake to ~6.04% earlier in the year, so Lazard is not the only large holder adding.
- Catalysts / news: Goldman's June downgrade is the main overhang. The company was also re-admitted to the Hang Seng Index after 18 years and included in the SSE 50. No major M&A or licensing event found in the last 60 days.
- Earnings quality: the H1 non-recurring items were small (non-GAAP profit RMB11.65bn vs 11.87bn). The profit swing is spread-driven (cyclical), so the base case uses a reversion level, not H1.
- Next-print test: the base case (RMB16bn) is below the annualised H1 run-rate, so it does not need a >30% improvement; the test passes. The exposure is the other way: a Q3 print that shows a narrowing spread would hit sentiment.
Conclusion
The signal is real (a genuine on-exchange purchase of HK$234m), but its information content is low: an asset manager buying a stock that is down 55%, with its price already 8% below its entry. The valuation says the market already discounts about half of the H1 profit gain. Base case ≈ price; the upside needs the spread to hold above ~RMB23,000/t, and the downside needs the Goldman capacity story to play out. With the base case at −1%, no Buy rating is allowed. Hold; revisit after Q3 results and if the SHFE price breaks below RMB20,000/t (re-rate the bear case) or holds above RMB23,000/t with a stable Q3 spread (the bull case gains weight).
Catalyst calendar
- 23 Oct 2026: last date for the interim dividend payment (RMB0.276/share), per the company.
- By 31 Oct 2026: Q3 results are due under SSE quarterly-report rules. I did not find a confirmed board-meeting date.
- Through 20 Jul 2027: Chinalco buy-in window (≤2% of share capital).
- Ongoing: SHFE/LME aluminium and alumina prices; Chinese capacity additions toward the 48mt Goldman expects; Indonesian and Middle East supply.
Key risks — and what would prove us wrong
- Aluminium price falls as new supply arrives (Indonesia, Middle East restarts, China capacity cap at ~45mt+ being exceeded); Chalco's profit is highly geared to the spread.
- Alumina oversupply: the alumina segment is already loss-making and China has flipped back to a net importer of alumina.
- Power, carbon and policy costs; Guinea bauxite policy; export and tariff measures.
- Minority leakage: 40% of group equity belongs to non-controlling interests.
- Data limits: several figures here come from aggregators with some internally inconsistent items (unit errors, undated multiples). I used only figures that tie out across at least two sources, or marked them as derived.
What would prove us wrong (too cautious):
- Q3 2026 net profit at or above H1's quarterly run-rate (≈RMB5.9bn) and SHFE aluminium holding above RMB23,000/t into the print. That would put the bull case in play and the stock on ~4–5x earnings.
- Chinalco disclosing material open-market buying (a meaningful part of the RMB1–2bn) at or below the current price.
What would prove us wrong (too optimistic): SHFE aluminium below RMB20,000/t, or Q3 profit well below RMB5bn.
Sources
- HKEX Disclosure of Interests, filing CS20261021E00372 (Lazard, 2600), via scanner log of 22 Oct 2026.
- FilingReader and iTiger summaries of Chalco's 1H26 results (17 Sep 2026); mrjjxw, 10jqka; etnet balance sheet (30 Jun 2026).
- StockAnalysis (price/share data, 23 Oct 2026); 10jqka spot/SHFE prices (22 Oct 2026); China International Aluminium Week summary (Sina, 22 Oct 2026).
- FilingReader (20 Jul 2026) on the Chinalco stake increase; iTiger on the Goldman downgrade and BofA note (14 Jul 2026); Tencent News (13 Jul 2026).
- Peer figures: Simply Wall St, broker notes via iTiger, aggregator pages.
No redactions applied.
Published via wiis. Originals are encrypted at rest; this page reflects the redacted version. report abuse.
Discussion
No comments yet.