Curtailment and negative electricity prices explained

In grids with high solar penetration, it is now routine for wholesale electricity prices to fall to zero or below during sunny midday hours, and for grid operators to instruct some solar plants to reduce output even though demand for electricity has not disappeared. Both phenomena - negative pricing and curtailment - follow directly from basic supply-and-demand mechanics once enough inflexible, near-zero-marginal-cost generation is added to a system that was not designed around it.

This article is for anyone evaluating a new solar investment in a market where these effects are already visible, or trying to understand why an existing system's export revenue has become less predictable than the original feasibility study assumed.

We separate the market mechanism (negative pricing) from the physical/operational mechanism (curtailment), since they are related but distinct, and finish with the practical implications for sizing, self-consumption strategy, and storage decisions.

Why prices go negative

Wholesale electricity markets in most of Europe clear at the price where supply meets demand for each settlement period. Solar and wind generation have near-zero marginal operating cost - once built, running them costs almost nothing per additional kWh - so they bid into the market at or near zero. When total generation from these sources plus inflexible baseload (certain nuclear and combined heat and power plants that are costly to ramp down) exceeds demand at a given hour, the market-clearing price falls to zero or negative to induce someone to reduce output or increase consumption.

Negative prices occur specifically when demand is low relative to available generation - typically a mild, sunny weekday around midday, or a windy public holiday - and when some generators are willing to pay to keep running rather than shut down, because shutting down and restarting costs more than paying the negative price for a few hours (relevant for thermal plants with high startup costs, and in some markets, for generators still receiving a fixed subsidy per kWh that outweighs a moderately negative wholesale price).

Curtailment: the operational response

Curtailment is the deliberate reduction of generation output, either by market mechanism (a plant is priced out of the merit order and simply does not run, or reduces output because the price it would receive is unattractive) or by direct grid operator instruction (a transmission or distribution system operator orders a specific plant to reduce output because of a local network constraint - insufficient line capacity, voltage limits, or a stability concern - independent of the overall market price).

These two forms matter differently to an asset owner. Market-based curtailment (choosing not to sell into a negative-price hour) is usually compensated by the absence of a loss - the plant simply does not sell that hour's output, or a PPA/subsidy structure may still pay it under certain contract terms. Grid-instructed curtailment for network reasons often comes with a compensation mechanism in some jurisdictions (particularly where curtailment stems from delayed network reinforcement that is the DSO/TSO's own responsibility) but not in others, where the generator simply absorbs the lost revenue - the specific rule depends entirely on national regulation and the connection agreement signed at commissioning.

Comparing the two mechanisms

Worked example: revenue impact on an uncontracted export

Consider a 500 kWp ground-mount system exporting to the grid on a pure wholesale-price basis (no subsidy floor), located in a bidding zone where day-ahead prices average EUR 0.06/kWh across all hours but fall to an average of -EUR 0.02/kWh for the 500 hours per year when prices are negative, based on recent historical data for that zone.

Annual generation: 500 kWp x 1,100 kWh/kWp = 550,000 kWh. Suppose 12% of annual generation (66,000 kWh) falls within negative-price hours. Revenue from that portion at -EUR 0.02/kWh = -EUR 1,320 (a cost, not income). Revenue from the remaining 484,000 kWh at the average positive price of EUR 0.075/kWh (the average across non-negative hours in this example) = EUR 36,300. Net annual revenue = EUR 36,300 - EUR 1,320 = EUR 34,980, against a naive estimate of 550,000 kWh x EUR 0.06/kWh = EUR 33,000 flat-average that ignores the timing mismatch entirely - in this particular numeric example the timing-adjusted figure happens to come out close to the flat estimate, but the two diverge substantially as the negative-price hour count grows, which is the trend observed across most high-solar European bidding zones over the past several years. A system owner should always model against actual hourly price data for the target zone (available from ENTSO-E Transparency Platform or national market operators) rather than a single annual average price.

Implications for new system sizing

The economic value of an additional kWp of solar capacity declines as more solar is added to a grid, because each additional unit of capacity increasingly generates during the hours when prices are already lowest - a phenomenon sometimes called "value deflation" or the solar cannibalization effect. This does not mean solar stops being viable; it means:

• Oversizing a system purely to maximize kWh output, on the assumption that all exported kWh earn a flat average price, increasingly overstates returns in markets with high existing solar penetration.
• Self-consumption becomes proportionally more valuable relative to export as wholesale export prices compress, strengthening the case for sizing to the building's own load rather than to roof area (see /blog/commercial-rooftop-solar-business-case).
• Pairing with battery storage to shift midday generation into evening peak-price hours becomes more economically attractive as the midday-evening price spread widens (see /calculators/battery-sizing), though the storage capex still needs to clear its own payback threshold independently.
• Fixed feed-in tariff contracts that pre-date market liberalization insulate the owner from this effect for the remaining contract term, but new contracts increasingly reference market price directly, exposing new entrants to negative-price hours from day one.

Common mistakes

• Modeling export revenue using a single flat average wholesale price instead of hour-matched price and generation data.
• Assuming curtailment losses are always compensated; this depends entirely on jurisdiction and the connection agreement.
• Oversizing an export-heavy system without checking the target bidding zone's historical negative-price hour count and trend.
• Ignoring that negative-price hour frequency is increasing year on year in most high-solar-penetration European markets, making historical price data from 5+ years ago an unreliable baseline.
• Treating curtailment purely as a loss to be minimized, without recognizing that in some cases (deeply negative prices) not generating is the financially correct outcome for an uncontracted merchant asset.
• Failing to distinguish, in a feasibility study, between a subsidized fixed-price contract (unaffected by market price swings) and a merchant/PPA structure that is directly exposed to them.

Frequently asked questions

Why don't grid operators just add more storage to absorb the surplus instead of curtailing?

Storage capacity is being added rapidly in many markets, but it currently remains a fraction of the surplus generation seen on the sunniest days, and network-constraint curtailment specifically requires storage or reinforcement at the exact congested location, not just anywhere on the grid.

Does curtailment mean the panels are switched off entirely?

Not necessarily; inverters can typically reduce output to a specified percentage rather than shutting down completely, allowing partial generation to continue during a curtailment instruction.

Can a homeowner's small rooftop system be curtailed?

Direct grid-operator curtailment instructions are far more common for larger, directly connected plants; small residential systems are more likely to be affected indirectly through low or negative feed-in compensation during oversupply hours rather than a direct dispatch order, though some markets are beginning to include smart-inverter curtailment capability for aggregated small systems.

Will battery storage eliminate the negative-price problem for a new system?

It reduces exposure by shifting self-generated energy away from low-price export hours, but it does not eliminate system-wide oversupply; if enough systems add storage and shift to the same evening hours, that period's price can itself compress over time, a dynamic already visible in some mature markets.

How can I check negative-price hour frequency for my region before investing?

The ENTSO-E Transparency Platform publishes day-ahead price data by bidding zone, and several national transmission system operators publish historical price and curtailment statistics directly.

Standards and references

Relevant standards: EN 50549 series (requirements for generators to be connected in parallel with distribution networks, including remote curtailment capability), IEC 61850 for the communication protocols often used in DSO curtailment dispatch, and IEC 61724-1 for performance monitoring needed to reconcile actual versus curtailed generation.

Sources: IEA "Renewables" market report series covering negative pricing trends across European markets; ENTSO-E Transparency Platform day-ahead price and generation data; Fraunhofer ISE analyses of negative price hour frequency in the German market; IRENA reports on system flexibility and curtailment in high-VRE grids; JRC reports on solar value deflation in EU electricity markets.

Related reading: /blog/commercial-rooftop-solar-business-case for sizing systems against load rather than roof area, /calculators/battery-sizing for evaluating storage as a mitigation, /calculators/self-consumption, and /glossary for definitions of merit order and bidding zone.