Europe's inflation problem has become more expensive to finance. On September 10, the European Central Bank raised its three policy rates by a quarter percentage point, taking the deposit rate to 2.50% from September 16. The move arrives while an important measure of underlying price pressure is easing. That apparent contradiction is the useful starting point for understanding the decision, rather than treating every higher inflation print as an automatic instruction to raise rates. The ECB's decision is explicit about the uncertainty.
The central question is how much of an energy shock will survive inside other prices after the initial disruption passes. Monetary policy cannot produce fuel. It can influence the conditions under which businesses pass costs on and households sustain spending. That makes this a judgment about propagation, with a real cost if policymakers overestimate it.
The inflation split behind the vote
Eurostat's August flash estimate puts annual inflation at 3.3%, up from 2.9% in July. Energy inflation accelerated to 14.3% from 10.3%. Yet inflation excluding energy, food, alcohol and tobacco eased to 2.4% from 2.5%, and services inflation fell to 3.0% from 3.3%.
Those figures describe a split, not a uniform acceleration. A business facing higher fuel bills may raise selling prices, accept a lower margin, improve efficiency or reduce activity. Which response dominates depends on contracts, competition and demand. The energy component alone cannot establish that the wider economy has entered a self-reinforcing inflation cycle.
It also cannot establish that the shock is harmless. Repeated increases in operating costs can become harder to absorb, particularly when contracts renew. The relevant distinction is between today's direct fuel contribution and tomorrow's broader repricing. A falling core rate is evidence against immediate broad acceleration; it is not proof that delayed effects will never arrive.
Persistence is the disputed part of the forecast
The new policy baseline projects headline inflation averaging 3.0% in 2026, 2.5% in 2027 and 2.1% in 2028. It projects inflation excluding energy and food at 2.5%, 2.6% and 2.3%, respectively. These are staff projections, not observed outcomes or promises. Their significance is the expected persistence beyond the immediate energy surge.
The strongest counterargument comes from the ECB's own press conference statement: wages do not yet show a material response to the energy shock. Compensation per employee slowed to annual growth of 3.3% in the second quarter from 3.5% in the first. There is therefore a gap between a risk policymakers want to contain and a wage response already demonstrated by the data.
A credible analysis must keep that gap visible. Waiting for unmistakable propagation could leave policy acting too late. Acting before it is clear could restrain demand unnecessarily. Neither position eliminates uncertainty; each places the burden of error on a different part of the economy. The decision reflects the ECB's preferred balance of those risks.
The squeeze reaches borrowers unevenly
The Associated Press report identifies greater economic resilience as part of the rationale for tightening. Resilience matters because it can make additional restraint easier to absorb, but an aggregate description does not tell us which firms or households have room in their budgets.
According to the ECB's financial-conditions assessment, bank lending rates for companies reached 3.8% in June and July, against 3.6% in May. Annual bank lending growth nevertheless increased to 4.4% in July. Higher loan prices and growing loan volumes can coexist; the volume figure is not evidence that policy has failed to transmit.
For an individual borrower, the relevant variables are the next reset date, debt maturity and need for fresh working capital. A firm with fixed borrowing costs may initially feel more pressure from energy than from interest. A firm refinancing now can face both. This is a transmission mechanism, not a claim that all borrowers have already experienced the September increase.
For lenders, higher asset yields are only one side of the calculation. Funding costs and credit losses can offset them. For equity holders, an apparently resilient revenue line may conceal pressure on financing costs and margins. The announcement alone is insufficient to calculate either industry's net earnings effect.
A reversible judgment about propagation
The case for restraint would strengthen if broader price-setting and wages began to reflect persistent energy costs. It would weaken if core disinflation continued, wages stayed contained and credit-sensitive activity deteriorated as supply pressures eased. These are alternative evidentiary paths, not predictions about the next meeting.
The ECB has committed to neither a sequence of increases nor a fixed destination for rates. Its September decision is best evaluated by whether the propagation it fears appears, and by the cost of preventing it. Reading the energy headline alongside wage behavior and borrower repricing gives a more useful account of that trade-off than the rate increase alone.