Ecuador keeps underwriting hydro as baseload. The reservoirs disagree
The Mazar reservoir, the regulating head of Ecuador's Paute complex, was sitting just 22 meters above its operational minimum by early April 2026. By the end of June, flows across the Mazar, Paute, Agoyán and Minas San Francisco basins had fallen near the critical 10th percentile of historical record. This is the same basin that produced 91 days of loss-of-load in 2024. Two years on, the system is running the identical experiment and waiting to see if it gets a different result.
Ecuador draws roughly 77% of its electricity from hydro, most of it from a single river system. That concentration cost the country an estimated $2 billion in 2024, close to 2% of GDP, and outages running as long as 14 hours a day at their worst. No material new generating capacity had been added since Coca Codo Sinclair came online in 2016, while demand had grown by roughly 400 MW, about 24%, over the same stretch.
None of that is in dispute. What is in dispute is whether the response since then changed the risk carried into 2026, or just bought time until the next dry season proved the same point again.
What actually got built
The state's answer to the 2024 crisis was overwhelmingly thermal: three rented power barges adding roughly 300 MW, the 100 MW Salitral project, and the 254 MW Toachi Pilatón plant, finally commissioned six years behind schedule. North of 540 MW of new firm capacity, almost none of it renewable.
On the private side, the October 2024 Organic Law to Promote Private Initiative in Energy Generation raised the self-authorization threshold for a private generation project from 10 MW to 100 MW and attached automatic incentives and state-backed PPA guarantees to qualifying projects. By August 2025 the ministry had authorized 643 MW of private renewable and distributed generation under that framework, mostly solar, with an 18-month window to complete permitting and reach an enabling title.
A 100 MW cap and an 18-month permitting clock do not describe a system correcting its hydro concentration. They describe a system buying itself a second dry season to see if it gets lucky.
Where the mechanism actually bites
Two structural choices in that response matter more than the megawatt count. First, the 100 MW ceiling per project is small against what a bankable utility-scale asset looks like elsewhere in the region, where 300 MW solar platforms are routine. A developer capped at a third of that scale carries the same fixed permitting, interconnection and legal cost across a smaller revenue base, and that shows up directly in the levered return a fund can underwrite.
Second, the state-backed PPA guarantee attached to the new law is a sovereign credit instrument before it is anything else. A guarantee is only as good as the balance sheet behind it, and Ecuador's fiscal position, the same one that left thermal maintenance underfunded for eight years, is precisely why the guarantee needs pricing rather than taking at face value. Lenders sizing debt against that structure are underwriting the sovereign twice: once on the offtake, once on the promise standing behind it if the offtake fails.
Neither of those is a reason to stay out of Ecuador. It is a reason to treat the 2024-2025 policy response as a partial fix aimed at surviving the next dry season, not a redesign of a grid that still assigns firm, dispatchable value to a resource the last two years have shown is neither.
What we think follows
A grid planner can call hydro baseload on paper. A reservoir 22 meters from its floor does not care what the plan says. Until Ecuador's capacity accounting treats large hydro the way Colombia has been forced to treat its own reservoirs after 2024's near-miss, as a resource with a real derating in a dry year, every dispatchable-capacity number the system publishes overstates what it can actually deliver when the rain does not come.
There is a fair counterpoint. Ecuador is a smaller, dollarized market with limited fiscal room to subsidize a faster renewable build, and removing the sovereign guarantee would likely slow investment rather than accelerate it. That tension is real. It is also exactly why the guarantee's credit quality, not just its existence, belongs in every term sheet priced against it.
Our own pipeline sits in Peru, not Ecuador, but the discipline transfers directly. Olmos Wind, our most advanced asset at approximately 135 MW in Lambayeque with battery storage integrated into the design, was structured against an independent system operator's dispatch rules, not a single utility's balance sheet standing behind a single hydro basin.
The open question
INAMHI's own outlook for the Paute basin through May 2026 called for normal to below-normal rainfall under La Niña-neutral conditions, and flagged that a Q3 2026 rationing event becomes materially more likely if Mazar fails to recover by mid-year. We do not know whether it will. Nobody publishing a forecast right now does either. What we do know is that a private renewable framework capped at 100 MW a project and running on an 18-month permitting clock will not have meaningfully changed Ecuador's generation mix before that question gets answered, one way or another, by the weather.
See the full active portfolio or read more Insights.