Skip to main content
EU_ECONOMICS05 / 18 · story of the day3 min · 520 words · 36 sources

Wallonia clears CETA as France holds out

Written by AIto brief AI · 18 June 2026, 03:50
How it was written

A legal architecture of paper must hold the immense weight of global commerce.

Image composition · tobrief
the text · 3 min read

In Namur, the capital of French-speaking Wallonia in southern Belgium, a regional parliament representing about 3.6 million people voted on 18 June to ratify CETA, the EU-Canada trade agreement. This is the same legislature that blocked the deal in 2016, briefly turning a local vote into a global symbol of democratic friction. The constitutional paradox it exposed hasn't gone away: Brussels can spend years negotiating sweeping trade deals, but a single regional parliament can freeze the whole machine.

The deal that already works

Wallonia's yes doesn't suddenly open trade with Canada. Most of CETA has been provisionally applied since September 2017. The agreement covers tariffs on goods, public procurement (government contracts open to foreign bidders), services, and regulatory cooperation (EUR-Lex). Firms have been using these trade rules for nearly nine years. The European Commission describes the agreement as largely operational.

What remains stuck is the Investment Court System, a proposed tribunal where foreign investors could challenge government decisions under the treaty. These provisions require every member state to ratify individually, because CETA is a "mixed agreement." Some powers belong to Brussels alone — setting tariffs, for example — but investment law stays with national capitals. The EU's Court of Justice confirmed that split, and separately ruled the tribunal compatible with EU law. Legal compatibility hasn't settled the politics.

So the commercial half works. The legal architecture for investor protection waits.

Who gains, who reaches for the brake

Export-heavy economies treat trade agreements as job protection. About a third of Dutch employment depends on exports, generating €375.9 billion in 2023, according to Rijksoverheid, the Dutch government's public information service. For the Netherlands, Ireland, Denmark, and Sweden, deals like CETA aren't abstract diplomacy. They protect jobs. Germany's ifo Institute estimated CETA could triple German exports to Canada, raising real income per person by 0.19% (consulting.de).

The same treaty looks different from Warsaw or Vienna. In Poland, debate centres on whether cheaper imports undercut domestic farmers and whether safeguard clauses actually protect them (Top Agrar Polska). Austrian and Italian coverage focuses on food standards and GMOs (Attac Austria, Adnkronos). Who gets to pull the emergency brake when local farmers take a hit?

France is the real bottleneck now

Wallonia's vote doesn't end the delay. France's Senate rejected ratification in March 2024, even though the French Constitutional Council had found the treaty acceptable back in 2017. French resistance centres on agriculture, food safety rules, and sovereignty. As Vie publique, a government explainer, puts it: most of CETA already applies, but investment provisions still need parliamentary approval Paris hasn't given.

Germany ratified in 2022, though only after the Federal Constitutional Court required that provisional application remain reversible. Even pro-trade governments wanted a way out.

The bigger question this saga leaves unanswered: who actually benefited from nine years of provisional application? Total trade volumes rose. But did CETA's tariff preferences mostly help large exporters or small firms? Industrial regions or peripheral ones? That evidence barely exists. And the structural tension persists. The EU keeps designing trade agreements that reach beyond tariffs into investment law and regulatory standards. The further these deals reach, the more national parliaments want a vote. Wallonia's belated yes won't change that. It will surface again with every ambitious deal Brussels tries to close.

How was this article?

Help us get better

Details about this article
Model:
claude-opus-4-6
Generated:
6/18/2026, 3:14:27 AM
Pipeline run:
eu_pipeline_20260618_015006
Watermark:
SynthID (Google's invisible watermark)
Human review:
None before publication
Learn more about our methodology