The cap that arrives next week

Since 2023, prequalification for FCR-D has been split into a static and a dynamic variant, and until now that split has had no consequence in procurement. Both bid, both cleared, nobody cared which was which.

That ends in September. Energinet, Fingrid, Statnett and Svenska kraftnät will apply a limit on how much static FCR-D upward capacity can be procured, enforced during bid selection rather than at prequalification. Static bids still compete normally up to the limit, and accepted static bids still receive the same marginal price as dynamic ones. It is pure volume rationing, with no price discrimination attached.

Why the TSOs are doing it

Static FCR-D exists for assets that cannot meet the full dynamic requirements, specifically activation and deactivation performance and dynamic stability. A static provider gets a grace period of fifteen minutes in which it is not required to deactivate or to deliver a second activation.

That grace period is tolerable in small quantities. It stops being tolerable when the system has little rotational energy to ride through a disturbance, because the reserve that is supposed to arrest the frequency may not be able to act twice. Hence a limit with two components: a fixed ceiling of 50 percent of Nordic FCR-D upward demand, and a variable component tied to system inertia that can only push the admissible share lower. Never higher.

The fixed 50 percent ceiling is confirmed. The variable component is not: the TSOs said details would be published before implementation and, at the time of writing, they have not been. The curve here is illustrative of the stated direction only, that lower inertia means a tighter static limit. Do not use the numbers, use the shape.

What it does to the price

Nothing much, and this is the part worth getting right.

The cap does not reduce how much FCR-D upward is bought. The Nordic requirement is still dimensioned against Oskarshamn 3 at 1 450 MW, Sweden's share is still roughly 550 MW, and every one of those megawatts still gets procured. All the cap does is decide who is allowed to supply them.

Displacing static bidders hands the freed volume to dynamic bidders. Battery storage holds 2 710 MW of prequalified FCR-D upward capacity in Sweden alone, against a national requirement near 550 MW. Giving a five-times oversupplied group a larger share of a fixed procurement does not create scarcity. The marginal bid still comes from an asset with almost no opportunity cost in offering it, which is exactly why the clearing price sits where it does.

The cap changes who is in the queue. It does not shorten the queue.

Expect this to be reported as a tightening, and expect at least one revenue model to be marked up on the back of it. That is confusing access with scarcity. Access improved for dynamic providers. Scarcity did not appear anywhere.

Who actually loses

Whoever is static. The TSOs are explicit that FCR-D upward goes first precisely because it carries the larger volume of prequalified static capacity, so the bind is real rather than theoretical.

I would treat the technology mapping carefully. Svenska kraftnät's prequalification register reports capacity by technology, not by static or dynamic variant, so I cannot tell you from published data exactly how the 410 MW of prequalified demand response or the 1 250 MW of hydro splits across the two. What is defensible is narrower and still useful: assets that struggle with fast deactivation and repeat activation sit on the static side, and those characteristics describe some demand-response configurations and some older plant far better than they describe a grid-scale battery.

The structural point does not depend on that mapping. The variable component tightens as inertia falls, and hours of low inertia become more common as the generation mix shifts. So the static allocation ratchets downward over time by design. Any revenue case built on static FCR-D participation is being written against a shrinking entitlement, and the shrinkage is a policy instrument rather than a market outcome.

The open item

Implementation is days away and two things are still unpublished: how the total Nordic limitation divides between the four TSOs, and what the variable component actually looks like as a function of inertia.

If you hold static capacity, you are being asked to plan a bidding strategy for September against a constraint whose size you have not been told. That is a small scandal in a market that otherwise publishes well, and it is the thing I would be chasing this week rather than the headline percentage.