How increased cycling costs due to PV and wind integration impact the merit order: Learning from an application to France

The large-scale integration of variable renewable energy (VRE), particularly wind and solar, requires greater operational flexibility from conventional generators. Yet, the associated cycling costs such as start-ups, shutdowns, and ramping remain insufficiently accounted for in long-term planning models. This paper develops a tractable methodology to quantify cycling costs with minimal data and computational needs by extending the e4clim investment–dispatch framework. Dispatchable producers are represented through a synthetic merit order mix, allowing cycling costs to be diagnosed ex post as additional variable costs proportional to start-up frequency and capacity. The approach is applied to mainland France under different levels of VRE penetration. Results show that while VRE integration reduces total system costs overall, start-up costs increase non-linearly with penetration and can reach up to 90% of VRE system value in extreme cases. Costs are unevenly distributed, with base load producers increasingly assuming flexibility needs at high VRE levels, while peaking units are less affected. Solar PV tends to induce roughly twice as many start-ups as onshore wind due to its diurnal profile. The analysis relies on simplifying assumptions and limited data, reflecting the interdisciplinary challenge of linking climate variability, electricity markets, and plant operation. As such, the method is best viewed as a qualitative tool to explore merit order disruptions and investment signals rather than a precise forecast for the French system.