What actually happens to electricity prices when renewable generation increases by 1 GWh?
New research from Imperial College London and the National Energy System Operator (NESO) answers this question using causal machine learning, moving beyond traditional correlation-based analyses.
Most econometric models struggle to properly isolate the effect of renewables because of:
Non-linear supply-curve dynamics.
Confounding factors (load, gas prices, carbon permits, seasonality).
Time-varying market conditions.
High-dimensional interactions between weather and market variables.
A new machine learning framework is used to isolate the true causal impact of wind and solar generation on UK day-ahead electricity prices (2018–2024), while controlling for dozens of covariates.
Key Findings
1️⃣ Wind power has a U-shaped causal effect
A 1 GWh increase in predicted wind generation reduces day-ahead prices by:
Up to £7/MWh at low penetration.
A smaller impact at mid-level penetration (~20–30%).
A stronger price-reducing effect again at higher penetration.
This U-shape reflects how wind displaces different parts of the supply curve as penetration increases.
2️⃣ Solar power consistently reduces prices
While raw observed data shows a misleading “price bump” at 4–7% solar penetration, the causal analysis reveals that solar always lowers prices, with reductions up to £9/MWh per additional GWh at low penetration. The “bump” is driven by confounders (load, seasonality), not by solar itself.
3️⃣ The effects are strengthening over time
Between 2018 and 2024:
Wind’s price impact has become significantly stronger.
Solar’s influence is increasing as installed capacity grows
Renewables are playing a progressively larger role in UK price formation.
Figure: How the price impact of wind power has changed over time
This chart shows how the effect of wind power on electricity prices has evolved over the past few years. Each line represents a different two-year period (for example, 2018–2020 or 2022–2024). The further a line moves to the left, the more wind power pushes prices down. What you can see is that in the most recent years, wind power has a much stronger price-reducing effect than it did in the earlier years. As wind penetration increases, the impact becomes even larger. The shaded areas around each line show the uncertainty in the estimates. Overall, the message is simple: as the UK has deployed more wind capacity, the ability of wind to lower electricity prices has grown significantly.
Why This Matters for the Energy Transition
These results carry important implications for the future of electricity markets. As price formation becomes increasingly driven by renewable output, market design must account for the growing influence of wind and solar generation. The non-linear nature of these effects also suggests that static support mechanisms may be suboptimal, highlighting the need for more adaptive subsidy structures. At the system level, the strengthening causal impact of renewables points to rising market sensitivity, underscoring the importance of robust capacity planning. Finally, incorporating causal insights into forecasting and risk management can significantly enhance hedging, valuation, and flexibility modelling. By distinguishing causation from correlation, this framework offers a far more reliable foundation for policymaking, market design, and strategic investment as renewable penetration continues to accelerate.
The full paper is available at: https://ieeexplore.ieee.org/document/11231232