What the aggregate shows

Between February and April 2026 the euro area consumer confidence index fell by around 12 index points, and nominal consumption growth slowed from 3 per cent to 4 per cent to about 2.5 per cent year on year by April. The authors put a number on the link: a 10-point fall in confidence is associated with a 0.4 percentage point reduction in consumption growth, against about 0.2 points in historical patterns, a response they describe as comparable in size to the one after the 2022 invasion of Ukraine.[1]

An aggregate that size hides the only thing worth knowing about it. The contraction was concentrated in discretionary categories while housing and food held up, and energy spending rose because transport costs went up. So the average household in this data did not cut its spending evenly; it dropped one kind of purchase and kept paying for another kind, and the bill it could not drop went up.[1]

Who cut, and who could not

The decomposition is in the box itself. Higher-income households adjusted spending more, postponing luxury purchases and non-essential services. Lower-income households cut durable goods and restaurant spending. The authors conclude that the slowdown reflects postponed discretionary spending rather than binding income constraints. Read from the top of the distribution that is a reassuring finding, because postponement is reversible. Read from the bottom it says something narrower: the households with the least room contributed the smallest share of the cut, which is what you would expect from households that had already spent their flexibility.[1]

That distinction matters because a postponement and a forced cut are the same line in a consumption series and different events in a household. A family that delays a kitchen appliance still holds the money. A family that drops a durable good while its energy bill rises because transport costs rose has moved the money, not saved it. Both show up as weaker discretionary spending. One is a decision that can be reversed when confidence recovers; the other is a budget that has already been rebalanced. There is a limit on how far this can be pushed: the series is nominal, so part of what looks like resilience in housing and food is price rather than volume, and part of the fall in discretionary categories could be the same effect running the other way.[1]

The test that separates them

Fortunately the two stories make different predictions, and the same survey can tell them apart. If the weakness really is postponement, then a recovery in the confidence index towards its February level should be followed by discretionary spending recovering without any change in income, and the recovery should be led by the higher-income households that did most of the cutting. If part of it was a constraint, discretionary spending in the lower half stays down while confidence recovers, and durable goods are the category that fails to come back. By the end of the first quarter of 2027 the survey rounds will have enough observations after the confidence trough to show which pattern held.[1]