Croatia’s energy regulator reported on 23 July that the number of negative-price hours on CROPEX increased from 196 in 2024 to 257 in 2025, an expansion of approximately 31 per cent.
The total represents about 2.9 per cent of the year, equivalent to almost 10.7 full days of continuously negative pricing. The events were concentrated in spring and on sunny Sundays, when demand was weak, photovoltaic production was strong and some subsidised generators continued producing under mandatory-offtake or legacy support arrangements.
The commercial effect extends beyond the individual negative intervals. Solar developers must also absorb the much larger number of hours in which prices remain positive but fall below operating budgets. A plant selling at €5–€20/MWh during its highest-output periods can experience severe capture-price erosion without technically producing during a negative-price event.
Newer Croatian market-premium contracts can suspend support when prices remain negative under specified conditions, transferring part of the market signal to generators. That improves dispatch discipline but increases revenue volatility for lenders and equity investors unless projects add storage, curtailment rights or more sophisticated offtake arrangements.
Croatia had only approximately 11 MW of battery capacity in March 2026, although the first large-scale system entered trial operation early in the year. The regulatory framework for wider deployment is still being completed, with more meaningful additions expected from 2027.
The mismatch is therefore acute: negative-price exposure has already become structural, while storage capable of responding at scale remains embryonic. Early batteries can benefit from less crowded balancing and arbitrage markets, but they also face unresolved questions over network tariffs, charging status, licensing and the treatment of double grid fees.
For solar investors, the Croatian data argues against valuing new projects solely from annual average CROPEX prices. Capture-rate forecasts need to incorporate at least 257 negative hours, additional near-zero periods and the likelihood that continuing photovoltaic construction deepens the central-day discount before sufficient storage is commissioned.
