Two kinds of price rise, and only one of them is in the forecast
Serguey Asael Shinder2 min readTwo items crossed the desk today that both raise European prices and are not the same kind of thing, and the difference is the whole of the analysis.
The first is a shock. Philip Lane told Le Temps that a second wave of energy price increases is running — oil and gas — which is why the ECB signalled on 10 September that the shock will last longer than assumed in March. Inflation will be higher for longer, returning towards target from mid-2027. He reports no pass-through into other goods and services between February and now, while expecting upward pressure ahead on food, on energy including electricity, and on goods, with services contained.
The second is a decision. Bruegel's report, reported the same day, costs EU local-content requirements at battery cells rising from €50 to €85 per kilowatt-hour — roughly €2,100 on a typical electric car — plus €200 for a low-carbon steel requirement, against €61 per car of administrative saving from simplified approval rules.
A central bank projection handles the first well and the second badly, for a structural reason rather than an oversight.
A shock has a curve behind it. Lane says the baseline reflects the market view captured in oil and gas prices, and the futures curve points to some resolution later this year. That is what makes "higher for longer, then back to target" a sentence a forecaster can write: the mechanism that raised prices is expected to reverse, and the projection inherits the reversal from the market.
An administered cost has no curve. If EU cells cost €85 rather than €50 because a rule requires it, nothing in a futures market brings that back to €50. It enters the price level as a step and stays. In inflation-rate terms it shows up once and then vanishes from the year-on-year comparison — which is precisely why it is easy to treat as irrelevant and precisely why it is not. The rate normalises; the level does not.
Three practical consequences.
Look-through logic does not transfer. A central bank can reasonably look through an energy spike it expects to unwind. Looking through a permanent level shift means accepting it, which may be the right call but is a different decision and should be made as one.
The distributional shape is opposite. Energy shocks hit everyone and fade. Bruegel's point is that the local-content cost falls hardest on cheaper models and less wealthy buyers — a regressive, permanent change sitting underneath a temporary, broad one.
Watch what the two do together. Lane's benign observation is that goods pass-through has not happened yet. Administered goods costs arriving while an energy shock is still running is exactly the combination that turns a relative-price story into something a household experiences as general inflation — regardless of how the deflator eventually attributes it.
⚠️ One is a central banker's published interview, the other a think tank's modelling reported second-hand. Neither is a forecast we are making, and nothing here is investment advice.