Chile's curtailment is a contract problem before it is a grid problem
Between 2023 and 2024 Chile added 2.2 TWh of solar generation. Over the same period curtailment rose by 2.7 TWh, more than the entire increase in output, on a 2024 total of 5,909 GWh curtailed, per Acera data drawn from Coordinador Eléctrico Nacional dispatch records. The system built more and delivered less.
Curtailment was treated for years as a rounding error, a small deduction applied to a P50 case and forgotten. In Chile that assumption stopped holding somewhere around 2024, when the deduction began growing faster than the capacity it was deducted from. Full-year 2025 curtailment reached roughly 6,084 GWh, up 7.8% year over year, per Acera.
Most of the analysis that follows a number like that goes straight to transmission. Northern Chile generates, central Chile consumes, and the lines between them are congested. All true, and all slow to fix. We think the more immediate question for anyone holding these assets sits somewhere else entirely: in the contracts.
Where the exposure actually sits
Most bilateral PPAs signed in Chile over the last five years were priced against irradiance. A P50 case, a P90 downside, an availability assumption, and a resulting energy volume the plant is expected to deliver. That volume describes what the panels can physically produce.
It does not describe what the system operator will accept. When CEN instructs a plant to back down at midday because the network cannot move the energy, the panels are working perfectly and the revenue does not arrive. Whoever holds volume risk in that contract absorbs the difference.
Debt sized against generated GWh, rather than settled GWh, is debt sized against a number the grid increasingly declines to deliver.
That distinction is not academic once leverage is involved. A DSCR built on a P90 exceedance curve assumes the downside case is a bad resource year. It is not built for a structural dispatch constraint that recurs every clear-sky midday and worsens as more capacity connects behind the same congested corridor. The resource risk the model priced and the delivery risk the asset actually faces are different animals, and only one of them was underwritten.
Storage helps, and it is not keeping pace
Battery storage is doing real work here. Discharge from the Chilean storage fleet offset roughly 2 TWh of curtailment in 2025, cutting potential curtailment by close to 24%, per figures linked to CEN. That is a material contribution by any standard, and it arrived faster than most people expected.
It still did not stop the annual total from rising. Storage is being added to a system whose congestion is growing along a different curve, and co-location decisions made today take years to reach commercial operation. An asset that reaches Ready-to-Build in 2028 is being designed now for a grid whose constraints in 2028 nobody has modelled honestly.
What we think follows
Volume risk belongs on the term sheet, explicitly. That means curtailment-sharing language negotiated into the PPA, or storage co-located behind the same interconnection point, or both. What it should not mean is an assumption buried inside an exceedance curve where nobody revisits it until a refinancing forces the conversation.
There is a reasonable counterargument, and it deserves stating. Offtakers have their own exposures and limited appetite to absorb a risk the developer is closer to. Push too hard on curtailment sharing and the contract does not get signed at all. That tension is real, which is precisely why it belongs in the negotiation rather than in a footnote.
Our own posture is to treat the delivery path as part of the asset rather than an externality attached to it. Olmos Wind, the most advanced project in the ZFC Energy Group portfolio at approximately 135 MW in Lambayeque, carries battery storage in the project design rather than as a retrofit after Ready-to-Build, which is targeted for the first quarter of 2028.
The open question
Chile now has to choose between two paths, and the market has not settled which one it is on. Either a formal compensation mechanism emerges for curtailed energy, closer to what parts of Spain and the United States have attempted, or the system keeps rationing through falling capture prices until new solar stops penciling without a hedge attached.
The second path is not a failure of the energy transition. It is a repricing of it, and the projects underwritten before the repricing are the ones that will feel it.
See the full active portfolio or read more Insights.