Costa Rica's cap on private power is closer than ICE admits

Private generators supply roughly 18.1% of Costa Rica's electricity against the Instituto Costarricense de Electricidad's 73.2%, well inside the ceiling Law 7200 sets for private participation, per figures cited in local reporting on the country's single-buyer model. ICE reads that gap as proof the market still has room. Developers reading the same statute read a wall sitting closer than the production numbers suggest.

Law 7200, passed in 1990, lets ICE buy from private hydro plants up to 20 MW as long as that parallel generation never exceeds 15% of national system capacity. A later chapter opened another 15% for wind, geothermal, biomass and other non-conventional sources, putting the practical ceiling near 30% once both baskets are counted. Almost every renewable developer in the country is financing against one of those two baskets.

How a statutory limit behaves differently

Most of the constraints we write about in this market are physical and incremental. An interconnection queue clogs one substation at a time. A transmission line gets permitted section by section. Both are slow, but both move, and a specific project can usually trace its own position within them.

A statutory cap behaves differently. It applies to the whole country at once. It doesn't care which plant is closer to completion or which has the better resource, and once the national total brushes the ceiling, ICE has no legal basis to sign another private PPA, regardless of how bankable the project in front of it looks on a standalone basis.

Where the mechanism actually bites

That matters because Costa Rica has exactly one legal buyer for private power: ICE. The single-buyer model means a private generator's revenue case depends entirely on ICE's legal capacity to contract, a different kind of exposure than competing for load in a market with multiple offtakers or a bilateral contract system.

A plant with a closed interconnection study and a signed term sheet can still have no legal buyer left to sell to.

For debt sizing, that turns the diligence question upside down. A lender needs comfort on the national aggregate as much as on the plant's own permits and resource study, a number that moves with every other developer's contract in the country rather than with anything the borrower controls.

Two readings of the same number

ICE's own position, as reported by Semanario Universidad, is that the ceiling is nowhere near exhausted and will not bind even once recently awarded private plants reach commercial operation around 2028. Critics cited in local coverage push back on the 73.2%/18.1% split itself, arguing it already shows a market concentrated far past where competition policy would put it.

The legislature has a bill in front of it, the Ley de Armonización del Sistema Eléctrico Nacional, Expediente 23.414, that would repeal Law 7200 outright roughly eighteen months after passage and build a national electricity market in its place, per La Nación's coverage. The Universidad de Costa Rica's governing council recommended rejecting that bill. Its own alternative was narrower: raise the private cap from 15% to 20%, not tear up the single-buyer model.

What I think follows

I would underwrite Costa Rica against the statutory ceiling before the grid. ICE's 2028 timeline reads to me like a political estimate wearing engineering clothes, and a reform bill that has already drawn opposition from the country's own public university is a thin thing to build a financing calendar around. The more likely outcome, if anything moves at all, looks closer to UCR's narrower fix than to full liberalization. That's a guess on my part, and any developer with a plant already inside the 15% basket has every incentive to agree with ICE that there's room left.

Our own pipeline sits in Peru, under a different structure entirely. Peru runs competitive auctions rather than a single legal buyer, and the live constraint there is that COES has not opened a new renewable auction round since 2024, a different failure mode of the same underlying problem: offtake access that depends on something outside the project itself. ZFC's roughly 870 MW of pipeline, including Olmos Wind at approximately 135 MW with integrated storage, is financed against that auction reality rather than a statutory headroom calculation.

The open question

Does Costa Rica's legislature actually move the number, even to UCR's proposed 20%, or does Expediente 23.414 stall the way earlier attempts at this reform have, leaving the single-buyer model and its 15% basket exactly where they are while 2028 gets closer?

See the full active portfolio or read more Insights.