It was two other considerations that tipped the scales against Grexit. The first was the threat of contagion. If Greece left, the myth that there is no way out of the euro would be instantly exploded, bringing the single currency closer to a fixed exchange-rate regime. The markets might fret that Portugal or even Italy could follow, presaging the currency’s eventual collapse. Yet Greece accounts for only 2% of the EU’s total GDP, so if the EU fears that the departure of such an economic tiddler could destroy the euro, it has alarmingly low confidence in its own creation. Besides, institutional changes have provided the euro with far stronger defences than it had before.

The second objection is the potential cost of Grexit, not only in support for Greece’s banks and people but through “TARGET” balances at the ECB. These reflect inflows and outflows of euros in national banking systems, which usually attract little attention. But the numbers have recently risen, a sign of renewed market nerves. Greece and Italy at the end of January had negative balances of over €70bn and over €360bn respectively, whereas Germany had a positive one of almost €800bn, its all-time high (see chart). Were a country to leave or the euro to break up, these balances would probably crystallise into genuine (and surely unpayable) claims.

Some economists have suggested that Germany, not Greece, should temporarily leave the euro, and rejoin later at a higher rate. The argument is that the underlying causes of the euro’s problems are Germany’s strong competitiveness and its huge current-account surplus. Yet a German exit seems politically implausible: the issue for markets is Greece’s membership, not Germany’s. Political or economic events could restart talk about Grexit at any time. It would be prudent to prepare for the worst—and seek to minimise the collateral damage.