The AfD won 39 of the Landtag’s 83 seats in Saxony-Anhalt on 6 September. The five other parties in the chamber hold 44 between them, and cannot assemble a cabinet out of any of them.

The arithmetic is worth stating slowly, because it is the entire event. A majority is 42. The CDU has 15, down 19.9 points from 2021.

The SPD has 8, the Greens 8, Die Linke 8, the BSW 5. Any combination reaching 42 must contain Die Linke, and the CDU refuses Die Linke as firmly as it refuses the AfD. Without Die Linke’s votes, or at minimum its abstention, Sven Schulze cannot be returned as Ministerpräsident even with every other non-AfD party behind him.

The new parliament must convene by 6 October. Turnout was 77.8 percent, seventeen and a half points up on the last election, which is the detail that should worry Berlin more than the seat count: these were not absent voters staying absent.

Friedrich Merz called it the worst result his party has had in decades and said it shook the CDU to its foundations, and then ruled out cooperation with the AfD. Tino Chrupalla said the firewall had been definitively rejected by the voters. Both men are describing the same object from opposite ends, and both are describing it as a wall around the AfD.

It has quietly become something else. The cordon has stopped being a wall around the AfD and become a wall around the government.

Vienna already ran this experiment, and ran it in the other direction.

In February 2000 the ÖVP did what the CDU refuses to do: it took Jörg Haider’s Freedom Party into federal government. The other fourteen EU member states froze bilateral contacts, the first time the Union had moved against one of its own. Martti Ahtisaari, Jochen Frowein and Marcelino Oreja were appointed to review it, reported that the measures had become counter-productive, and the fourteen lifted them on 12 September 2000, roughly seven months in.

Haider called it a victory all the way down the line, and on the narrow question of the sanctions he was right.

What happened next is the part worth carrying. The FPÖ had taken 26.9 percent and 52 seats in 1999. It governed.

It fought with itself, lost several of its leading figures at Knittelfeld, forced an early election, and on 24 November 2002 came back with 10.01 percent and 18 seats. The ÖVP, its partner, went to 42.30 percent. Three years of office cost the FPÖ two thirds of its vote.

The external punishment campaign lasted seven months and strengthened it. Governing lasted three years and nearly finished it.

A protest vote is a claim on a future the party has never had to deliver, and office is the only instrument that prices it.

That asymmetry is the deep variable, and it has nothing to do with Germany or Austria specifically. A protest vote is a claim on a future the party has never had to deliver, and office is the only instrument that prices it. Exclusion suspends the pricing.

Every year the AfD is kept out of a state chancellery is a year in which its promises remain untested against a budget, a police force, a hospital rota and a Bundesrat vote, and in which the exclusion itself supplies the grievance that renews the claim. The cordon is doing to the AfD what the EU-14 did for the FPÖ, at greater length and with better manners.

The cordon has stopped being a wall around the AfD and become a wall around the government.

This is the same structure the column has read at Hormuz, in the rare-earth controls, and at the dollar-clearing rail: leverage whose power lies in its universality, spent by the act of using it. A cordon is cheap to observe while the excluded party is small enough to outvote. At 39 of 83 it stops being cheap, and the cost lands on the excluder’s capacity to govern rather than on the excluded party’s support.

Saxony-Anhalt is where the price first shows up on the invoice.

For Southeast Asia the transmission is dull and real. Germany is the heaviest single weight in the Council on the instruments that reach exporters in Jakarta, Kuala Lumpur and Ho Chi Minh City, and the carbon border levy and the deforestation rules are enforced by ministries whose room for compromise in Brussels narrows every time a governing party has to defend its flank at home. The EPC reads exactly that constraint into this result, ahead of 2027 votes in France, Poland, Spain and Italy.

The number a trade ministry in the region should track is the width of the margin a German chancellor still has to trade away in Brussels, and that margin moves on state elections.

The call. By 31 May 2027, Saxony-Anhalt will be governed by a cabinet containing no AfD minister and holding fewer than 42 of the Landtag’s 83 seats among its formal coalition parties: the firewall standing, and a minority government behind it, dependent on Die Linke ballot by ballot. Confidence: likely.

The load-bearing assumption is the CDU’s second refusal rather than its first, because a formal coalition with Die Linke is the only arithmetic route to 42 that excludes the AfD; should Merz permit Schulze to sign one, or should the CDU accept a governing pact it declines to call a coalition, the call breaks and the cordon turns out to have been cheaper than it looks.

Perhaps somebody in Magdeburg has done the other sum. Three years in office at 43.8 percent is the one thing nobody has yet made the AfD pay for.