For banks financing renewables in Southeast Europe, debt sizing increasingly needs to account for a range of possible electricity-market revenues. The widening gap between wholesale electricity prices and renewable captured prices means lenders can no longer rely on a single central revenue case. The traditional framework remains necessary, but it is no longer sufficient on its own.
Financing renewable energy projects has traditionally involved converting a relatively small set of assumptions into a debt-capacity estimate. Expected generation was multiplied by forecast electricity prices to calculate revenue, operating costs were deducted, and the resulting cash flow was used to determine how much debt a project could service.
A more complete sequence is production → captured price → curtailment → imbalance and operating costs → cash flow → DSCR → debt capacity. Each step can materially affect the next, so financing ratios should be tested only after the relevant variables have been brought together.
Project revenues depend on hourly market conditions
The first requirement is an hourly production profile, because annual P50 output alone does not show when electricity will be generated. For photovoltaic projects in particular, generation timing determines exposure to periods of system surplus. A baseload forecast should not automatically be treated as the price available to a solar or wind project.
Instead, hourly production needs to be matched with hourly market prices to estimate the project’s realised price and develop a technology-specific captured-price curve. AleaSoft argues that average market prices are no longer sufficient for photovoltaic valuation. It says captured prices, hourly production and curtailment should all be included in the assessment.
Availability and curtailment form another layer of the revenue calculation. Electricity that cannot reach the market cannot be valued in the same way as delivered power. Technical availability, grid restrictions, connection conditions and contractual compensation therefore need to be reflected in cash-flow assumptions.
Balancing costs also matter: forecasting errors and profile deviations can impose costs even when the gross electricity price remains attractive. Contract terms add a further layer, including PPAs, CfDs, floors, caps and merchant exposure. Where appropriate, these arrangements should be reflected hour by hour rather than represented as a single annual average selling price.
Financing tests should reflect scenarios and correlated risks
AleaSoft highlights the difference between an asset retaining enough cash to meet scheduled principal and interest payments and its DSCR falling below the thresholds required by financing agreements. That distinction is especially relevant when revenues deteriorate gradually instead of collapsing outright. A project may remain economically viable even as its financial structure becomes inappropriate.
Possible responses may include extending tenor, reshaping amortisation, resizing debt, changing reserve requirements or refinancing rather than abandoning the asset. Probabilistic modelling should be central to this process. AleaSoft says a single long-term forecast curve cannot capture the range of possible market outcomes and advocates assessing multiple revenue scenarios with associated probabilities.
For lenders, a broader framework can combine P50/P75/P90 generation cases with central and downside captured-price scenarios, curtailment stresses and balancing-cost stresses. It can also include PPA counterparty and expiry scenarios, merchant-tail assumptions, interest-rate and refinancing sensitivities, and DSCR and LLCR outcomes.
Correlation between variables matters. A project should not combine pessimistic assumptions that cannot logically occur together, but it should also avoid treating adverse factors as independent. High renewable output may coincide with depressed captured prices and increased curtailment, creating substantially worse cash flows than separate small discounts to production and price would suggest.
The methodology therefore needs to move away from arbitrary conservatism and towards internally consistent downside scenarios. Technical, market and financial advisers can no longer work in isolated workstreams. The engineer’s production model affects the market consultant’s captured-price calculation, while the grid review affects available volumes; those findings feed into the financial model, and the PPA changes how the resulting exposure is allocated.
For Southeast European banks, renewable finance is becoming less like lending against a predictable infrastructure yield and more like lending against actively managed electricity-market exposure. The key figure in a credit paper may therefore no longer be the project’s average expected electricity price. It may instead be the debt that remains serviceable when several plausible adverse market conditions occur at the same time.