Hjartar
01 / 07

Below Zero

Friday 1 May 2026. Nine continental markets cleared at the algorithm's floor for the same fifteen minutes.

NINE EUROPEAN MARKETS · MAY 1 HOLIDAY 2026 · €/MWhSDAC ALGORITHM FLOOR · −€5000−€5004 ZONES AT FLOOR · 10:45+€369HUNGARY · 18:30DE · FR · NL · BE · AT · PL · CZ · SK · HU

The continental core crashed at the algorithmic floor. Surrounding regions held by separate mechanisms.

Day-ahead clearing across every European bidding zone at 11:15 UTC, 1 May 2026.

−3−2−10−498−4990†−382−500*−500−3−30−90−498+920†−500*−311+21500−499+32+28+32+3+95−490−20†−471−500*0000000BRITISH ISLESSubmarine cables full both ways.Could not absorb continental surplus.NORWAYHydro reservoirs absorb cheap continental power.Grid bottlenecks keep zones uneven: NO5 €95, NO4 €3.NORDICSSweden, Finland near zero (hydro + nuclear).Denmark cushioned by Nordic hydro.IBERIAPyrenees crossing at maximum capacity.Spain only mildly negative.BALKANSStayed positive. Held by a gas-fired floor.Effectively an energy island.ITALYSubsidies pay nothing below zero.Plants stop bidding lower by contract.EUROPEAN BIDDING ZONES · 1 MAY 2026 · 11:15 UTCDay-ahead clearing price, €/MWh≤ −480−100< 00 floorpositive* CZ, SK, HU forward-filled from their 10:45 UTC −500 print through a publication gap; mid-gap recovery ruled out by market coupling with the western core.† BG, RO, GR clearings 08–14 UTC missing from ENTSO-E A44 feed. Map value 0 is a conservative placeholder; the structural argument suggests €60–€110. Pending verification from local exchanges.

Five mechanisms held the crash inside the core: saturated submarine cables to the British Isles, contract floors in Italy, hydro and nuclear in the Nordics, gas-fired floors in the Balkans, capacity-capped Pyrenees crossing into Iberia.

Surplus has four physical exits.

Cables, curtailment, demand response, storage. On 1 May, three were closed and the fourth was small.

01
Export

Send the surplus to neighbours over interconnectors. On 1 May every corridor out of the continental core was full or capped — the cables had no headroom to take more.

02
Switch off

Stop generating until prices recover. TSOs ordered no curtailment on 1 May; operators pulled back voluntarily to dodge the negative-price clawback on their subsidies.

03
Boost demand

Flexible loads soak up the surplus. Most demand isn't flexible. Industry was off for the holiday; households on fixed retail tariffs had no signal to use more.

04
Store

Cycle cheap midday power into expensive evening hours. The fleet that exists today is small; the one being built across Europe is roughly three times that size by 2030.

Cables are slow to build, curtailment is waste with a clawback bill, demand response needs household-level price signals that don't yet exist. Storage is the only exit being scaled at speed.

Negative pricing across European power markets is increasing in both duration and depth.

Cumulative depth × duration across major European markets. By 5 May, 2026 had already accumulated more negative-price exposure than the whole of 2024.

025K50K75K74K202482K202575K2026 YTDJANFEBMARAPRMAYJUNJULAUGSEPOCTNOVDECCUMULATIVE NEGATIVE-PRICE EXPOSURE · ENTSO-E + GBSum of |hourly mean| × 1 h, across major European bidding zones, when the day-ahead price cleared below zero.

Each hour where the day-ahead clearing price went below zero contributes |price| to a running total — depth × duration in one number. 2026's first four months have already accumulated more severity than all of 2024.

The negative-price signal pays for storage. Three groups absorb its cost.

Wind and solar without a contract earn nothing. Always-on power plants pay to keep running for days. Customers on fixed prices cover the grid's cleanup bill.

LOSER 01
Wind and solar without a contract

These plants sell into the day-ahead market without a long-term buyer or subsidy. When the price drops below the cost of running, they either generate at a loss or switch off and earn nothing. New rules in Germany, France, Italy and Spain (Solarspitzengesetz, Décret 22 Dec 2025, FER X, RD 997/2025) are clawing back subsidies during negative hours, so even the protected fleet is shrinking.

LOSER 02
Always-on power plants

Gas, lignite (brown coal), and biomass plants are built to run for days at a time. Switching them off and back on during a multi-day negative-price stretch costs millions, so operators dial them down to the lowest output the plant can sustain and absorb the loss. Cheaper than restarting.

LOSER 03
Fixed-price customers

Households and businesses on fixed retail contracts. They never see the cheap wholesale prices, but they do pay the grid's cleanup bill. The cost of rerouting power around bottlenecks, balancing supply with demand, and compensating wind and solar farms told to switch off all flows back through network charges (Netzentgelte in Germany, TURPE in France).

Negative prices reward whoever can react fast. The bill falls on whoever can't.

Six levers can compress negative pricing. None closes the gap alone.

01
Batteries
Fast
+Benefit

Buy cheap solar at midday, sell it back at evening peak. Already commercial; the market pays for the buildout on its own. About 30 GW of operating batteries in Europe today, with central forecasts near 95 GW by 2030 (BNEF, Aurora).

Challenge

Most installed batteries hold about four hours of energy. Fine for the daily cycle, useless for a still and cloudy week. And the more get built, the smaller the price gap they earn from.

02
Hydro
Patchy
+Benefit

Decades-old, large-scale, proven. Pump water uphill when power is cheap, run it back through the turbine when prices return. Existing plants cycle daily; new reservoirs would stretch storage from hours to weeks.

Challenge

Geography decides everything. The best sites are already built. New ones need a mountain, a river, and a permitting process that runs a decade. Norway, Switzerland and Austria are expanding; the rest of Europe is mostly stuck.

03
Long duration storage
Pre-commercial
+Benefit

Stores energy for days or weeks, where lithium-ion batteries only stretch to a few hours. Several technologies are in the running, including iron-air batteries, compressed-air storage, and hydrogen. None of them yet at scale.

Challenge

Still mostly demonstration projects. The costs, and how much energy you actually get back after storage, are unproven at the volumes Europe would need.

04
Dynamic pricing
Deployed, uneven
+Benefit

Let household electricity bills follow the wholesale market hour by hour. Heat pumps, EV chargers, dishwashers and cold stores can then run when power is cheap or free. Smart meters and the regulatory plumbing are mostly already in place.

Challenge

The hourly contracts exist; people are not signing up. Over 10% of households are on them in Estonia, Denmark, the Netherlands and Norway. Fewer than 2% in Germany, France, Italy and Spain. (ACER market monitoring)

05
Contract redesign
Early
+Benefit

Stop paying the subsidy when wholesale prices go negative. A producer's lowest-acceptable bid then moves from roughly −€80 toward €0. Four markets brought in the rule in 2025 (Germany, France, Italy, Spain). Settling subsidy contracts hour-by-hour rather than annually would attack the −€500 algorithmic floor directly; that reform is still on paper.

Challenge

Negative prices are how the market tells solar developers to add storage instead of more panels. Cap the floor and that signal goes quiet. The price gap that batteries earn from also narrows. This lever is in active tension with the one next to it.

06
Grid buildout
Slow
+Benefit

Move cheap power from where it is produced to where it is needed. On 1 May, every cable out of the continental core was full and the surplus had nowhere to go. More cables fixes that.

Challenge

New transmission lines take 8 to 12 years from plan to power flow. Permitting is brutal, public opposition is constant, costs run into the tens of billions. Several major projects are under construction (Eastern Green Link, NeuConnect, Celtic, SuedLink). None will materially close the gap before 2030.

Six levers, three rates of change. Storage and dynamic pricing are scaling now. Hydro and contract redesign are constrained by geography and policy. Long-duration storage and grid buildout remain pre-commercial or slow.

END · PART 1

Thanks for reading

One Friday, one regime, one deck. Same regime, two sides of the ledger.

Part 2 looks at what 1 May means for capital — where the storage builds, who signs the cables, which contracts re-write themselves to price negative hours correctly.

Questions, pushback, or a topic you want covered? hello@hjartar.com
Sources & methodology
Below Zero
ENTSO-E A44 day-ahead prices, PT15M resolution, 2026-05-01. Nine continental zones (DE, FR, NL, BE, AT, PL, CZ, SK, HU). 87–96 reported slots per zone; CZ/SK/HU forward-filled across the SDAC publication gap (~7 slots each). Floor reference: SDAC algorithmic minimum of −€500/MWh.
The continental core crashed at the algorithmic floor. Surrounding regions held by separate mechanisms.
ENTSO-E A44 day-ahead prices, PT15M resolution, 11:15 UTC slot. CZ/SK/HU forward-filled from the 10:45 UTC −€500 print across the SDAC publication gap. BG/RO/GR clearings 08–14 UTC missing from the A44 feed; map values are conservative placeholders pending verification from local exchanges. Bidding-zone geometry from ENTSO-E zone shapefile, simplified for SVG rendering.
Surplus has four physical exits.
Interconnector capacity utilisation: ENTSO-E Transparency Platform cross-border physical flows, 1 May 2026. Curtailment instructions: TSO daily reports (50Hertz, RTE, REE, Terna) — no compulsory curtailment ordered. Demand response: ACER market monitoring 2025; industrial load was off for the May 1 holiday. European BESS fleet ~30 GW operating, 95 GW central forecast for 2030 (BNEF, Aurora).
Negative pricing across European power markets is increasing in both duration and depth.
ENTSO-E A44 day-ahead prices for the European bidding zones (40 zones, excluding Ukraine) plus Elexon BMRS for GB. Method: each hour where the volume-weighted hourly mean of the day-ahead price clears below zero contributes |hourly mean| to a running total. Mid-2025 transition from PT60M to PT15M MTU handled by aggregating 15-min slots to hourly volume-weighted means. 2024 final: ~74K. 2025 final: ~82K. 2026 YTD (through 5 May): ~75K.
The negative-price signal pays for storage. Three groups absorb its cost.
Subsidy clawback rules: Solarspitzengesetz (DE, 2025), Décret 22 Dec 2025 (FR), FER X (IT, 2025), RD 997/2025 (ES). Always-on plant economics: lignite/gas minimum-load technical constraints from operator capacity reports. Network charges: Netzentgelte breakdown (BNetzA), TURPE 6 framework (CRE). Editorial framing observational.
Six levers can compress negative pricing. None closes the gap alone.
Battery deployment: BNEF 2025 ESS outlook, Aurora EU storage forecast (~95 GW by 2030 vs ~30 GW today). Hydro: Eurelectric pumped-storage capacity database. LDES: LDES Council pipeline tracker. Dynamic-tariff penetration: ACER Market Monitoring Report 2025 (>10% in EE/DK/NL/NO; <2% in DE/FR/IT/ES). Subsidy clawback: same legislation list as the losers beat. Major cross-border projects under construction: Eastern Green Link (UK), NeuConnect (DE-UK), Celtic (FR-IE), SuedLink (DE).