DATANG INTERNATIONAL POWER GENERATION CO. (DAT.L) — Research Note

Executive summary

Datang International Power Generation is one of China's "Big Five" state-owned electricity generators, with a largely coal-fired thermal fleet supplemented by a growing hydro, wind and solar portfolio, selling power into the regulated/quasi-regulated Chinese grid. Across 2021–2025 the filings cadence is dominated by routine annual/interim results, dividend declarations and connected-party transactions (e.g. the July 2025 entrusted-loan disclosure 2025-07-14 connected transaction), with no profit warnings, no equity issuance and no strategic pivot — i.e. a stable, slow-moving SOE utility. The single most important point for valuation today is that this is a leveraged, state-controlled, low-ROE, capital-intensive thermal utility whose earnings are driven by Chinese coal prices and government-set tariffs — none of the investor's three preferred attributes (AI-receiver, operating leverage with optionality, valuation discipline with quality balance sheet) are well-fitted, so the question is whether the price is low enough to compensate.

Fair value estimate

Caveat: the filings supplied are only the investegate announcement headers — the actual financial PDFs are linked but not in the text, so I cannot read line-item revenue, EBITDA, net debt, capacity additions or dividend per share figures directly. The fair-value range below is therefore anchored on (a) the disclosed market cap and (b) sector-typical multiples for Chinese SOE thermal generators dual-listed in HK/LSE; conviction is correspondingly low.

Methodology: sector multiple cross-check (P/B 0.5–0.7x and EV/EBITDA 6–8x are typical for Chinese Big-Five thermal SOEs; a normalised dividend-yield anchor of 4–6% on a payout consistent with the recurring final-dividend filings throughout 2023–2026).
Indicative fair-value range: broadly in line with the current market cap. I estimate a fair market cap of £1,300m – £1,900m, equivalent to roughly −17% to +21% vs the £1,569m disclosed level.
I have deliberately not quoted a per-share pence range, because the LSE listing is a thinly-traded GDR/secondary line whose pence price requires the underlying H-share price and the GDR ratio to convert — neither is in the filings supplied. Setting per-share fields to null is the honest call.
Conclusion: stock looks roughly fair at current levels, with upside only if Chinese coal prices stay benign, tariff reforms hold, and the renewables build-out is value-accretive — none of which can be tested from these headers.

Sector context

Classification confirmed: Utilities / Electricity (Conventional + Renewable mix).
Quality / growth / leverage profile vs. listed utility peers: leverage materially above typical Western utility peers (Chinese SOE thermals have historically run net debt > 4x EBITDA); growth in line; quality (ROE, FCF conversion, disclosure transparency) below Western regulated-utility peers.
Peers: Huaneng Power International (HNP, 0902.HK), Huadian Power International (1071.HK), and China Resources Power (0836.HK) — same Big-Five cohort. For UK-listed comparators on the LSE main market the read-across is weaker (SSE, National Grid, Drax) given different regulatory regimes.

Investment thesis (3 bullets)

Stable cash-generative state asset paying a recurring final dividend. A final dividend has been declared every year from 2022 through 2025 2024-03-22 dividend, 2025-03-25 dividend, 2026-03-27 dividend, with multiple "updated" announcements suggesting the payment is taken seriously by the board — a yield anchor for income buyers.
Operational leverage to falling Chinese thermal coal prices. Chinese power tariffs are partially indexed but with a lag; a fixed installed-capacity base means EBITDA per unit improves materially when coal costs fall 2025-08-28 interim, 2025-03-25 annual — the kind of inflection that has driven the sector in past cycles.
Indirect read-through to Chinese power demand including data-centre buildout. Datang sits in the generation layer of one of the world's largest power systems; AI/data-centre demand in China is incremental to the grid, and large generators are eventually called on to supply that load 2026-03-27 annual results announcement — but this is a generic sector tailwind, not company-specific AI exposure.

Key risks (3 bullets)

Coal-price and tariff regulation risk. Earnings can swing materially with thermal coal input cost vs. an administered tariff cap — the only reason 2021–2022 results were weak across the Big Five 2022-08-30 interim, 2023-03-28 annual; underlying figures not in supplied text but the issue is well-documented sector-wide; inferred.
High leverage and continued capex on coal + renewables transition. Chinese SOE thermals carry net debt well above sector-typical Western utility levels and continue to invest in both legacy thermal upgrades and renewables build-out (not disclosed in the headers but inferred from the asset base).
Related-party / governance risk. The July 2025 "supplemental announcement of connected transaction — provision of entrusted loan" 2025-07-14 is a textbook reminder that the company sits inside China Datang Corporation's web of related parties; minority shareholders are price-takers on these transactions.

Operating leverage

A coal-fired generator has a very high fixed cost base — depreciation of plant, interest on project debt and minimum staffing dominate, while fuel is the dominant variable cost. In principle this gives the business operating leverage to either (i) higher utilisation hours, (ii) higher tariffs, or (iii) lower coal prices. However, that leverage is bounded by tariff regulation on the upside and amplified by leverage on the downside: when coal spikes and tariffs lag, EBIT can disappear and the high debt service crushes net income (the dynamic that hit the sector in 2021–2022 2022-08-30 interim, 2023-03-28 annual). It is therefore the wrong kind of operating leverage for this investor profile: outcomes are wide, but the upside is capped by regulation and the downside is amplified by debt — not the "long-tail to multiples of profit" leverage the strategy seeks. A 10–20% revenue upside surprise would not translate into a multiple of profit; it would be partially given back through tariff adjustment.

Value-trap signals

Persistently low P/B and high yield characteristic of Chinese SOEs that rarely re-rate to international utility multiples — structural, not temporary.
Repeated related-party transactions (entrusted loans within the China Datang group) 2025-07-14 limit minority-shareholder upside capture.
Coal-to-renewables transition consumes capex with uncertain returns and is a multi-decade structural overhang on the legacy fleet.
Thin LSE liquidity (the Main Market listing is a secondary line vs. the primary HK/SH listings) — the discount may be permanent.

Earnings vs. expectations

The filings supplied are announcement headers only; they do not include the actual results figures, prior management guidance or consensus references. I cannot reconstruct a beat/met/miss pattern from this text. Filing cadence is regular and there are no profit-warning RNSs or sudden trading-update style releases in the set — consistent with an SOE that does not give Western-style quantitative guidance and where outcomes track Chinese coal prices and tariff policy rather than management commentary. Pattern: not enough disclosure in the supplied headers to score, so I treat surprise track record as neutral.

Conviction

Conviction: 2 (low). Anchors: (i) the business model is well-understood at the sector level — a Chinese SOE thermal-plus-renewables generator — so directional judgements are reasonable; (ii) the consistent dividend cadence over 2022–2025 multiple dividend filings indicates stable enough cash generation to anchor a yield-based sense-check. Limits: (i) the supplied filings are headers without the underlying financials, so I cannot verify revenue, EBITDA, net debt, capacity mix or dividend per share; (ii) the LSE pence price requires the H-share ↔ GDR ratio that is not in the text, so a per-share fair value cannot be defended.