Aller au contenu principal
Climat & Énergie

Nuclear power: where do Belgian parties stand in 2026?

A neutral comparison of Belgian parties' positions on nuclear power in 2026: repeal of the phase-out law, extension of Doel 4 and Tihange 3, the 8 GW target, the state buy-out of Engie's fleet. Who voted what, pros and cons, public sources.

ByCamille11 min read

Where does Belgian nuclear power actually stand in 2026?

Belgium is no longer required to phase out nuclear power. On 15 May 2025, the Chamber repealed by a large majority the law of 31 January 2003, which scheduled the gradual closure of the reactors and banned any new construction. Two decades of exit policy thus ended with a vote.

On the ground, the picture is more modest than the symbol. Two reactors are still producing: Doel 4 and Tihange 3, whose operation was extended by ten years to 2035 under a deal negotiated in 2022 with Engie and overseen by the Federal Agency for Nuclear Control. The other units are shut down, and Tihange 1 is approaching the point of no return in its dismantling — once crossed, a restart becomes technically illusory.

The repeal lifts a legal lock; it builds nothing. No new reactor is under construction, no site has been chosen, no funding is secured. It is precisely this gap between the announcement and the concrete that shapes the 2026 debate: the question is no longer "should we exit nuclear power?" but "who takes the industrial and financial risk of going back in?".

How can you read the parties' positions without taking sides?

Each party gets a sign here, not a grade: a green + when it clearly defends the approach, an amber ~ for an intermediate or conditional position, a red − when it opposes it. This system deliberately replaces stars or marks out of five, which would suggest a moral ranking.

Two columns, two distinct questions. The first is about extending existing reactors — an area where the consensus has widened considerably, including on the left. The second is about building new capacity and committing public money to buying the fleet: there, the lines harden. A party can accept extension without wanting to build; the reverse barely exists.

Neutral illustration of the Belgian nuclear debate: an abstract scale weighing a stylised cooling tower against the public cost of the buy-out, dismantling and waste
The 2026 debate has shifted: less 'for or against the atom' than 'who pays, and over how many decades'.
PartyExtend existing reactorsBuild new capacity (8 GW target)
MR++
N-VA++
CD&V++
Vlaams Belang++
Les Engagés++
Open VLD++
Vooruit+~
PS~
PTB·PVDA~
Ecolo
Groen

This table is not a ranking: a + is not worth "better", it indicates a direction. The vote of 15 May 2025 confirms it — the MR, N-VA, CD&V, Vooruit and Vlaams Belang voted for the repeal, Ecolo and Groen voted against, while the PS and PTB·PVDA abstained. The abstention is not a detail: it reflects a left that no longer intends to close reactors on principle, but refuses to sign a blank cheque.

What does the Arizona government actually want with 8 GW?

The Arizona coalition — N-VA, MR, Vooruit, CD&V and Les Engagés — targets 8 gigawatts of nuclear capacity, against roughly 2 GW today. The announced path combines two blocks: about 2 GW of further extensions on existing reactors, and 4 GW of new capacity, conventional reactors or small modular reactors.

Energy Minister Mathieu Bihet (MR) told Parliament he was negotiating with Engie on restarting units already shut down, while keeping "all options open, including bringing in other nuclear operators". In July 2026, eleven million euros were allocated to projects in the sector: SMRs, fusion, advanced materials, robotics, research infrastructure. That is a signal, not a construction programme.

The order of magnitude is worth stating. Going from 2 to 8 GW means quadrupling a capacity the country spent fifteen years dismantling politically, in a country that has not built a reactor since the 1980s and where the industrial supply chain has largely dispersed. Supporters see a necessary catch-up; their opponents see a target whose timetable and bill are equally undocumented.

Why is the Engie buy-out so divisive?

Because it shifts a private risk onto the public budget. Engie refuses to extend reactors other than Doel 4 and Tihange 3: without a willing operator, the 8 GW target remains an intention. Hence the strategic pivot announced on 30 April 2026 — a letter of intent between the state and Electrabel opening exclusive negotiations on buying the group's entire nuclear business in Belgium, with a memorandum of understanding targeted for October 2026.

The scope is wide: the seven reactors, their operation, the staff, the subsidiaries, maintenance — but also the liabilities. In other words, the obligation to dismantle and manage waste for decades. Some €25 billion is held in the Synatom and Hedera funds to cover this, and a dispute of more than €2 billion still pits the state against Engie over the level of those provisions. That figure is not an accounting abstraction: if it is under-estimated, the gap lands on the taxpayer.

The readings clash head-on. Bart De Wever defends "an investment that pays off", arguing the state would capture the revenues of a written-down fleet rather than leave them to a foreign shareholder. Mathieu Bihet promises not to "buy a pig in a poke" and says he can "backtrack". Conversely, voices from across the spectrum — the greens, but also François De Smet (DéFI), who welcomes the principle while flagging "grey areas that could prove abyssal for the taxpayer" — demand costed guarantees before signing. The disagreement is not about the atom: it is about the price.

What do Ecolo and Groen hold against this revival?

The greens contest the promise more than the physics. Samuel Cogolati, co-president of Ecolo, called the repeal "a communication operation that will change strictly nothing": lifting the ban does not make a reactor appear, and the build time for a new plant — ten to fifteen years in the best recent European cases — makes it useless for the 2030 climate targets.

Their case stacks three grievances. Cost first: every public euro committed to nuclear is not committed to wind, solar, grids or insulation, which deliver faster per euro invested. Waste next: no final storage solution is operational in Belgium, and the 2003 repeal does not settle that point. Dependence last: uranium is imported, and the safety of an ageing fleet carries a recurring price.

The counter-argument exists and deserves a hearing. Nuclear power supplies dispatchable, very low-carbon electricity, available when there is neither wind nor sun — something renewables alone cannot guarantee without massive storage or back-up gas plants, which emit CO₂. It is this dilemma, not a simple ideological clash, that the two camps resolve differently.

Why did the PS and the PTB abstain rather than vote against?

Because their electorate looks first at the electricity bill. The PTB·PVDA supports extending the current reactors, provided safety is guaranteed and the extension translates into lower prices for households — its criticism targets Engie's profits, not the reactor itself. In parallel it defends massive investment in renewables and a public energy pole.

The PS followed a comparable path: long anchored to the exit timetable, it ended up distinguishing extension (acceptable if it secures supply and prices) from building new reactors (deemed costly and slow). The abstention captures exactly that middle ground: not voting for a text from the right, not locking itself back into a closure promise that has become hard to keep.

This shift is one of the defining facts of the legislature. In 2003, the nuclear phase-out was a left-wing consensus; in 2026, only Ecolo and Groen still defend it as such. The divide has moved from principle to wallet: how much, for whom, and over what horizon.

What this comparison does not settle

It does not say whether the bet will pay off. An 8 GW target announced in 2026 depends on a deal with Engie or another operator, on public funding of unknown size, on safety approvals, and on an industrial supply chain that has to be rebuilt. No party controls those variables alone, and political announcements on nuclear power have a long history of slipping timetables — in both directions.

Nor does it factor in your situation. Depending on whether you look at your bill over the next five years, the country's carbon footprint in 2050, the public debt, or security of supply on a windless winter evening, the same measure changes colour. The useful reflex is not to crown a winning side, but to link each position to the lever it pulls — dispatchable, low-carbon electricity on one side, long-term financial commitment and waste on the other.

How can you check these positions yourself?

The sources are public. The vote of 15 May 2025 can be consulted on the Chamber's website; the Federal Agency for Nuclear Control (FANC) documents the long-term operation of Doel 4 and Tihange 3; the federal government agreement sets the coalition's energy path; the RTBF, VRT NWS, Le Soir and La Libre have followed the Engie buy-out and its grey areas step by step. Each party's 2024 programme gives the direction it announced before the compromises.

To go faster, the comparator puts two parties side by side on climate and energy, the ranking summarises positions theme by theme, and the quiz starts from your priorities rather than a programme. Our climate and energy party comparison places nuclear power within the wider transition debate. The methodology explains how these positions are gathered and remains open to challenge.

Climat & Énergie comparator

Compare the parties' positions on climat & énergie.

Compare now →

Frequently asked questions

No. The law of 31 January 2003, which scheduled the gradual closure of the reactors and banned any new construction, was repealed by the Chamber on 15 May 2025. Nothing now requires reactors to close on a fixed date, and new plants are legally possible again. The repeal builds nothing by itself, however: it lifts a lock, it does not fund or plan a reactor.

The MR, N-VA, CD&V, Vooruit and Vlaams Belang voted in favour. Ecolo and Groen voted against. The PS and PTB·PVDA abstained. The fault line therefore does not exactly match the majority/opposition boundary: the Vlaams Belang, in opposition, backed the text, while the left split between head-on opposition and abstention.

Doel 4 and Tihange 3 have had their operation extended by ten years, to 2035, under a deal negotiated in 2022 with Engie and framed by the Federal Agency for Nuclear Control. The other units have been shut down, and Tihange 1 is approaching the point of no return in its dismantling. The government says it wants to 'switch back on' as many reactors as possible, but no restart is secured at this stage.

It is the goal set by the N-VA, MR, Vooruit, CD&V and Les Engagés coalition: to raise Belgian nuclear capacity to 8 gigawatts, against roughly 2 GW today. The maths combines about 2 GW of further extensions (existing reactors) and 4 GW of new capacity, conventional reactors or small modular reactors (SMRs). It is a political ambition, not a construction site: neither the funding, nor the operator, nor the locations are settled.

Because Engie refuses to extend reactors other than Doel 4 and Tihange 3. Without a willing operator, the 8 GW target stays theoretical. On 30 April 2026, the state and Electrabel signed a letter of intent opening exclusive negotiations on buying the entire nuclear business: the seven reactors, staff, maintenance, but also the liabilities — waste and dismantling. A memorandum of understanding is targeted for October 2026.

Nobody can put a firm figure on it today, and that is the heart of the controversy. You have to add the price of the deal, the investment needed to extend the plants, then decades of dismantling and waste-management costs. Some €25 billion is held in the Synatom and Hedera funds, and a dispute of more than €2 billion still pits the state against Engie over nuclear provisions. Bart De Wever calls it 'an investment that pays off'; his critics warn of a potential money pit.

No. Meilleur Parti Politique is not affiliated with any party and recommends no vote. Reviving nuclear power promises low-carbon, dispatchable electricity but commits public money for decades and leaves the waste question open; giving it up avoids those commitments but means covering peak demand some other way. The right trade-off depends on what you are willing to pay, and when.

Camille est politologue, diplômée en sciences politiques de l'UCLouvain. Elle a suivi trois campagnes électorales belges comme analyste et décortique depuis dix ans les programmes des partis, vote par vote. Sur Meilleur Parti Politique, elle traduit le jargon politique en comparaisons concrètes — sans jamais dire pour qui voter.