Verification Report on the Investment Logic of Thermal Power Industry Transition from Strong Cyclical to Dividend Asset

I have verified the investment logic you proposed: ‘Can the thermal power industry transition from strong cyclical to dividend asset as its capital expenditure cycle declines?’ Subject to current tool and data availability constraints, the following conclusions are strictly based on obtained information and tool limitations, without any speculation or views unsupported by data:

Core Conclusion (Evaluation under current data and tool limitations):

• Whether there is a logic for transitioning to dividend asset: Insufficient evidence. Currently, there is a lack of systematic evidence such as verifiable capital expenditure trends, free cash flow and dividend rate/dividend sustainability, dividend yield and valuation (PB/PE, relative to industry/market), financial health and capital structure. Therefore, it is impossible to give a definitive ‘yes/no’ judgment based on existing information.
• Comparability with the electrolytic aluminum industry: Cannot be confirmed. No quantitative cases or publicly authoritative conclusions on capital expenditure decline, cash flow improvement and dividend increase in the electrolytic aluminum industry have been retrieved, so it is impossible to judge whether the two have comparable transformation paths.

Key Information Missing and Evidence Gaps:

1. Target company level (e.g., Jingneng Power, Guodian Power, Wanneng Power):
   • Real-time market quotes and company overview: Unavailable (tools did not return valid quotes or company information).
   • Financial and health assessment: Uncompleted (financial analysis tools failed due to insufficient data).
2. Industry level:
   • Capital expenditure trends: Currently unable to provide data on structural changes and inflection points of capital expenditure in the past 5 or 3 years.
   • Cash flow and dividends: Lack of historical comparison of quality indicators such as ‘operating cash flow/net profit’ and time-series evidence of dividend rate/dividend yield.
   • Coal cost and elasticity: No authoritative, structured data on the central level and volatility of thermal coal prices has been obtained, so it is impossible to quantify the boundaries of ‘cost increase limiting downside space’ and profit elasticity.
   • Policy and electricity price mechanism: Lack of systematic evidence on the latest policies and implementation details of coal-electricity linkage, capacity compensation, electricity price pass-through and utilization hours (especially the downward step of coal-fired power utilization central level).
3. Comparative industry (electrolytic aluminum):
   • No publicly available research or data support on verifiable capital expenditure decline and dividend asset transformation has been retrieved; comparability cannot be judged.
4. Market and valuation:
   • Lack of historical quantile evidence of valuation multiples (PE/PB), dividend yield, free cash flow yield and relative valuation of the industry and targets.
   • No information on sector capital flow, institutional holdings or transaction structure has been obtained.

Methodological main risks (to be noted when analyzing after subsequent data supplementation):

• Difficulty in separating cycle and structure: Thermal power performance is affected by coal prices, electricity prices, utilization hours and policies; it is necessary to distinguish between ‘cyclical fluctuations’ and ‘structural improvements’ in the model.
• Cash flow quality: Need to examine depreciation and amortization, working capital changes and interest coverage to avoid misinterpreting ‘high book cash flow’ as distributable free cash flow.
• Dividend sustainability: Dividend rate does not equal dividend yield sustainability; it needs to be comprehensively evaluated in combination with capital expenditure, debt servicing capacity and regulatory requirements.
• Nature of capital expenditure decline: Need to distinguish between ‘natural decline’ due to capacity cycle peaking and ‘active return to shareholders’; the latter is more conducive to the establishment of the dividend logic.

Currently feasible next steps (need your choice and authorization):

• A. Enable deep investment research mode (priority recommendation): It is recommended to provide a list of target companies (e.g., Jingneng Power, Guodian Power, Wanneng Power) and time window (2019-2024), and I will conduct the following based on available A-share data:
  1. Trend and inflection point verification of capital expenditure and free cash flow;
  2. Sequence comparison of cash flow quality (OCF/NI, FCF/OCF, etc.) and dividend rate/dividend yield;
  3. Thermal coal cost central level and price elasticity (if feasible), electricity price pass-through and utilization hour changes;
  4. Relative quantile of valuation and dividend yield, safety margin and dividend sustainability stress test;
  5. Comparability analysis with ‘dividendized’ industries such as electrolytic aluminum (if comparative industry data can be obtained).
• B. If target company data cannot be obtained, please provide stock codes/markets (A-share/H-share/US-share) with available data, or specify alternative targets to complete the comparison under the same framework.
• C. If you are more concerned about macro logic and industry structure, I can continue to retrieve more specific policies (coal-electricity linkage, capacity compensation, electricity price pass-through, new energy substitution rhythm, etc.), but this does not replace the quantitative verification at the target level.

At the current stage of lacking systematic evidence, I will not give structured labels such as ‘recommend/not recommend’ or ‘strong cyclical/dividend asset’. It is recommended to prioritize path A and conduct rigorous quantitative evaluation and conclusion output after obtaining complete data. If you agree, I will conduct retrospective and forward-looking analysis of the target companies in accordance with the above framework in the deep investment research mode.