The official explanation of renewable electricity pricing reform by China's National Development and Reform Commission and National Energy Administration states that renewable electricity supplied to the grid should, in principle, enter the market and receive a market-determined price. It also introduces a sustainable-development pricing settlement mechanism and distinguishes existing from new projects.[1] Project revenue must therefore be decomposed under the applicable rules. This does not mean every project has the same mechanism coverage, settlement price or trading conditions. Assessment must return to the effective documents in the project's jurisdiction.

ZHICE starts by treating a project as an asset that must be operated over time. Resources determine potential generation; grid connection and operating constraints determine deliverable electricity; markets and contracts determine pricing; settlement and collection arrangements determine when cash arrives. These are distinct stages. Hiding any one behind an annual average can make an investment case appear more stable than the underlying business.

Development materials should include a revenue-evidence register. For each revenue stream, identify the legal or contractual basis, effective period, corresponding electricity volume, settlement counterparty, deductions and unresolved matters. Conditions without implementing rules or verifiable support may represent opportunities, but should not automatically enter the base cash flow supporting financing. Gaps in evidence should become decision constraints, rather than numbers inserted at the modeller's discretion.

This change also requires a broader definition of the investment team's role. Investment cannot finish with a calculation and a handover to trading; trading cannot wait until commissioning to become involved in revenue optimisation. At appraisal, the teams should jointly confirm the deliverable generation profile, available contract choices and required operating capabilities. If modelled revenue depends on a capability that does not yet exist, its development time, cost and owner must enter the investment case as well.

Begin by aligning physical and commercial data. Theoretical generation, plant generation, grid exports, traded volumes and settled volumes answer different questions; each needs a defined metering point and adjustment rules. Retain time granularity and version information so prices match tradable volumes in the same period. Assign maintenance, curtailment, equipment availability and forecast errors to distinct places in the model to avoid omissions or double deductions.

Revenue can be reconciled sequentially through electricity production, contract coverage, market settlement, mechanism adjustments, fee deductions and collection arrangements. Define how the layers connect, especially how volumes covered by an applicable mechanism correspond to electricity traded in the market. Two apparently different settlement entries must not be mistaken for two complete payments for the same electricity. Calculations that cannot be tested against actual settlement statements should retain a human review requirement.

Contract analysis should distinguish fixing a price from fixing an outcome. Even where a price is agreed for part of the volume, generation deviations, profile matching, performance obligations and settlement adjustments can affect final cash flow. Trading teams should explain whether the contract covers aggregate volume, time periods or delivery obligations; production teams should establish feasible operating boundaries. Finance must translate accounting revenue into funds available for debt service and reinvestment.

A model built this way need not be complex at the outset. First verify the revenues, costs and constraints most likely to affect the decision, then add detail. A model that explains billing differences and supports accountability is more useful for investment decisions than one with many parameters but no reliable reconciliation. Record the reasons and effects of every model change so assumptions are not quietly adjusted to make a project fit a predetermined conclusion.

Reducing electricity prices or generation in isolation is often insufficient to test resilience. Concentrated renewable output may be related to price behaviour, while transmission constraints, maintenance, demand changes and forms of market participation can affect revenue together. ZHICE recommends describing the causal chain of each scenario before selecting parameters. Mechanically stacking contradictory pessimistic assumptions creates an extreme result that the business cannot explain.

The base scenario should reflect operating conditions the company can currently substantiate. Stress scenarios identify limits of tolerance, while improvement scenarios explain the actions on which additional returns depend. All three should use the same calculation structure and vary only drivers with a defensible basis. Alongside project returns, outputs should show when cash shortfalls emerge, what triggers them, how long they last and which adjustments are available.

Reviews should separate inherent uncertainty from execution gaps. Future weather and market prices contain uncertainty that cannot be eliminated. Unfinished contract negotiations, unresolved grid connection and unconfirmed settlement definitions are matters that can still be investigated. The former require risk capacity and portfolio arrangements; the latter require completion conditions and deadlines. Putting both into a discount rate conceals problems that could otherwise be resolved.

Scenario inputs should identify their source, date, applicable region and approver. External studies may inform the assessment but cannot replace actual project conditions; historical averages are not future commitments. Parameters with the greatest effect on results and the weakest evidence deserve priority for further research or staged decisions. If a small change reverses investment rankings, make that instability explicit to decision-makers.

Portfolio management identifies exposures shared across projects. Projects in different regions may still have similar generation profiles, contract structures or policy assumptions. Different resources in one region may complement each other through operating times or service capabilities. Geography is therefore only one dimension: examine concentrations in revenue sources, trading approaches, technologies, financing maturities and management capacity as well.

The IEA's study of electricity grids highlights the need to consider grid infrastructure, planning and management capacity together in the energy transition.[3] For companies, this is a prompt for local verification: strong resource quality should not be assessed separately from evacuation and system constraints. This report does not infer the utilisation or returns of any specific region from that research. It recommends including the relevant conditions in project due diligence and portfolio review.

Project selection can use two dimensions: confidence in business value and control over critical conditions. Where the revenue logic is clear and the company can execute, a project can proceed through the standard process. Where value is clear but external conditions are unresolved, preserve the option while gathering evidence. Where the business model remains unproven, limit pilot investment. These groups guide research and capital allocation; they are not permanent labels.

Portfolio assessment must also include organisational capacity. Individually economic projects may simultaneously draw on the same trading, engineering, maintenance and financial resources. Expansion beyond team capacity can turn manageable issues into operating risks. Headquarters should decide how scarce capabilities are allocated and when to pause new development to improve existing operations, rather than assessing performance solely through development scale.

The official questions and answers on China's Basic Rules for Electricity Market Operations describe arrangements covering market participants, transaction types and operating requirements.[2] These provide a foundation for understanding the market; specific eligibility, product design and settlement details still require local verification. A service or revenue model available in one market should not be copied into every project's appraisal without checking that it applies.

Contract reviews should work backwards from critical investment assumptions. Before signing, establish which price or volume assumptions the contract supports, which remain exposed to the market, who bears forecasting and performance obligations, and how exceptions can be renegotiated. Decompose a partner's proposed overall return into base service fees, variable sharing, guarantee conditions and exit costs, paying particular attention to provisions that may offset each other.

Investment, trading, production and finance can jointly maintain an assumptions register. Investment defines the opportunity boundary; trading interprets markets and contracts; production confirms operating capability; finance aligns settlement and cash definitions. For disputed parameters, record the disagreement and evidence before deciding whether external verification is needed. One department's system data should not automatically override other legitimate definitions.

Incentives should reflect lifecycle value. If development is accountable only for commissioned capacity, trading for a selected revenue category and production for generation, local optimisation can reduce overall cash flow. Use an explainable net operating contribution to connect departmental objectives, while retaining process measures each team can control. This also reduces the tendency to assign all market-driven changes retrospectively to one team.

An investment threshold should contain more than a return rate. Appraisal must ask whether adverse conditions create an intolerable cash shortfall, what flexibility remains if financing terms change, and whether expansion can be delayed, partnerships changed or further investment stopped when a critical assumption is disproved. Buffers should address specific failure scenarios rather than replace discussion with a uniform adjustment factor.

For storage, flexibility retrofits or other supporting investment, first establish the constraint being addressed. The National Energy Administration's explanation of the power-system flexibility action plan considers resource development, dispatch arrangements and market mechanisms together.[4] This points to the need to verify technical capability and commercial conditions at the same time. Installing equipment does not itself create realisable revenue. Additional investment needs a separate explanation of operating strategy, effects on asset life and allocation of costs.

Staged decisions align research spending with evidence maturity. Early work should establish resource, connection, market and partnership boundaries, followed by detailed investigation of high-value unresolved questions. Contracts and stress scenarios should be reviewed before major capital expenditure. Staging is not a synonym for delay: explicit completion conditions reduce irreversible commitments and the pressure to continue simply because early costs have already been incurred.

Management should require each material project to present the strongest evidence both for and against investment. Record at approval which new facts would change the judgement. Reassess when those triggers arise, rather than waiting for the annual budget process. This discipline protects reasonable experimentation and helps distinguish external change from shortcomings in the original due diligence.

After commissioning, retain a frozen version of the original investment model and establish an operating reconciliation consistent with actual bills. Compare actual and expected electricity volumes first, then attribute differences to prices, contracts, fees, collections and operating events. Classify each variance as external change, an incorrect assumption, execution variance or a definition adjustment. The same effect should not be counted at multiple stages.

Measures should be explainable. Price capture can help interpret the relationship between generation timing and market prices, but requires a defined benchmark, period and electricity-volume basis. Settlement variance measures should exclude periods awaiting adjustment, while cash-collection measures should reflect contractual due dates. Without consistent definitions, an apparent improvement may be a calculation change that offers no basis for action.

Review meetings should focus on the few matters that change decisions: which assumptions have been supported, which require revision, whether common portfolio-level variances have emerged, and whether new projects should still use the original inputs. For a successful operating improvement, explain whether it can be replicated and under what conditions. A favourable market episode is not, by itself, evidence of enduring team capability.

Post-investment learning should return to development and budgeting. Preserve settlement rules, contract experience, operating constraints and partner performance as versioned evidence for the next due-diligence exercise. Failed assumptions and their causes are equally valuable. Keeping only the final conclusions from successful projects gradually understates risk and forces the company to pay repeatedly for the same learning.

Companies need not wait for every model, system and organisational change to be completed. Select one proposed investment and one operating project, and use the same definitions to assemble revenue evidence, time profiles, stress scenarios and an accountability register. Comparing the two reveals the missing links between appraisal information and operating reality before deciding whether to extend the approach company-wide.

The pilot should produce more than a revised report template. It should deliver verifiable calculation workpapers, explicit unresolved questions, jointly accepted metric definitions, and the timing and triggers for the next review. Mark unavailable data and their effect on the decision. Market opportunities without local evidence can remain subjects for research, but should not be presented as certain income.

The management habit to establish is to explain how value is created before discussing how scale should grow, and how the downside will be borne before pursuing the upside. Renewable projects continue to require engineering and resource capabilities, supported by trading, organisation and capital management. This report offers investment-governance recommendations; it does not replace project-specific engineering, legal, tax or financial due diligence.