ECB Survey Lifts Three-Year Inflation Expectations to 2.9%

Median three-year inflation expectations in the euro area rose to 2.9% in August, according to the European Central Bank’s household survey. That is the figure the bank uses to judge whether consumers still see prices settling near target, and it now sits at the top of the band the ECB itself treats as on-target for this group.

The 12-month view only moved to 3.0% from 2.9%. Euro area annual inflation was 3.2% in August, up from 2.9% in July, with energy at 14.3%. Households answered between 6 and 24 August, before the Governing Council’s 10 September rate rise.

What do consumers in the euro area expect will happen to inflation, their incomes, their spending, and the value of their homes 12 months from now?

We’ve published the latest results of our Consumer Expectations Survey.

Press release https://t.co/AUO2lBsOtO pic.twitter.com/liRjiCO2v5

— European Central Bank (@ecb) September 18, 2026

Three-Year Inflation Expectations Rise to 2.9%

The ECB published the August Consumer Expectations Survey results on 18 September 2026. Median expectations increased across every horizon it tracks, while the median of what people think prices already did over the past year held at 3.5%.

The three-year median is the move that stands out. It rose 0.2 points, from 2.7% in July, after falling from 2.9% in May to 2.8% in June and 2.7% in July. One month erased that two-month decline. Five-year expectations rose 0.1 points to 2.5%, after sitting at 2.4% from May through July.

THE AUGUST INFLATION PRINTS

• Past-year view: Households still put perceived inflation over the previous 12 months at 3.5%, unchanged from July and 0.3 points above the 3.2% HICP rate.
• Next 12 months: The median expectation rose to 3.0% from 2.9%, a smaller step than the 0.3-point rise in actual annual inflation from July to August.
• Three years ahead: The median jumped to 2.9% from 2.7%, the largest move of the three forward horizons and a return to the May level.
• Five years ahead: The median ticked up to 2.5% from 2.4%, the first change in that series since April’s data cycle left it at 2.4%.

Uncertainty around the 12-month forecast fell, the ECB said, but it stayed higher than the level that prevailed before the conflict in the Middle East. Younger respondents aged 18-34 kept reporting lower inflation perceptions and expectations than those aged 35-54 and 55-70. The survey does not represent the views of the ECB’s decision-making bodies or staff.

What Counts as Anchored in the ECB’s Own Tests

In Issue 5/2026 of the Economic Bulletin, ECB economists Colm Bates, Aidan Meyler, Giovanni Trebbi and Zivile Zekaite set out how the bank scores whether longer-term inflation expectations are still tied to the 2% target. For consumers, they defined a 1-3% target-consistent range for three-year-ahead expected inflation, a wider window than the 1.8-2.2% band they use for professional forecasters, because household answers are noisier and biased high.

At present, level anchoring of longer-term inflation expectations broadly holds.

Colm Bates, Aidan Meyler, Giovanni Trebbi and Zivile Zekaite, ECB Economic Bulletin, Issue 5/2026

That box used Consumer Expectations Survey data through June 2026. In June the three-year median was 2.8%. August’s 2.9% is still inside the 1-3% window, so the test they wrote down has not been failed. It is now at the ceiling of that window, after a 0.2-point rise in a single month, and the five-year median has left the 2.4% line it held through the summer.

The same box said consumers had reacted most to the recent energy-price shock, while the term structure stayed downward sloping, with longer-term views revised up only slightly. It also said close monitoring was still warranted, because some agents may be more sensitive to new price jumps after the 2022-2023 surge. A three-year median at 2.9% is the sort of print that test was built to catch: short-term fuel pain is one thing, a pay-round and shopping-list number three years out is another.

Professional forecasters in the third-quarter Survey of Professional Forecasters still put headline inflation at 2.0% in 2028 and in 2031. They put indirect effects of the Middle East conflict at about 0.2 percentage points in 2026 and second-round effects at less than 0.1 points. Households, answering later in August, now see 2.9% three years ahead.

Spending Plans Run Ahead of Income Growth

The rest of the survey barely moved, which is what makes the inflation shift harder to write off as a general mood swing. Nominal income growth expected over the next 12 months stayed at 1.0%. Expected nominal spending growth stayed at 3.6%. That is a 2.6-point gap between what households think they will earn and what they think they will spend.

THE HOUSEHOLD LEDGER FOR THE NEXT YEAR

• Income: Expected nominal income growth held at 1.0%, with the bottom income quintile at 1.7% and the top quintile at 0.7%.
• Spending already done: Perceived nominal spending growth over the previous 12 months rose to 5.2% from 5.1%.
• Spending ahead: Expected nominal spending growth over the next 12 months stayed at 3.6%.
• Growth: Expected economic growth over the next 12 months stayed at -1.2%.
• Jobs: The expected unemployment rate 12 months ahead fell to 11.0% from 11.2%, against a perceived current rate of 10.5%.

Households still expect the euro area economy to shrink, and they still expect a jobless rate far above anything professional forecasters put on 2026. They also still plan to raise spending by 3.6% on 1.0% more income, after already seeing spending run at 5.2% over the past year. That mix is how a fuel shock gets into the shopping basket even when people think the wider economy is contracting.

The labour-market answers point to a stable, grim picture rather than a sudden layoff scare. Consumers continued to expect future unemployment to be only slightly higher than the 10.5% they think they see now. Bottom-quintile households put the rate 12 months ahead at 13.4%; the top quintile put it at 9.4%. Those CES jobless figures run well above the 6.3% unemployment rate professional forecasters expect for 2026, a gap the survey has shown for years because respondents overstate the jobless rate they think is already in place.

The Income Split Inside the August Survey

Lower-income respondents kept reporting higher inflation perceptions and expectations than higher-income respondents, a pattern the ECB has flagged in successive waves. The same split shows up in housing and credit, where the households with the least room to absorb another price rise are also the ones pencilling in the highest mortgage rate.

BOTTOM QUINTILE VERSUS TOP QUINTILE

The people who already expect the weakest income growth are not the ones expecting the highest pay rise. They are the ones expecting the highest jobless rate and the highest mortgage rate. They also expect their homes to rise 4.0% in price, faster than the 3.2% expected at the top of the income scale, so the same group is marking up both the cost of borrowing and the value of the asset that borrowing buys.

That is a distributional problem inside a single survey wave, not a side note. A 2.9% three-year inflation view concentrated among lower-income households is the group most likely to press for wages and to bring purchases forward. The net share of households reporting tighter credit over the past 12 months did fall in August, as did the net share expecting tighter credit over the next 12 months, so the squeeze they describe is more about prices and rates than about doors slamming at the bank.

Professional Forecasters Still See 2% by 2028

The gap between what households told the CES in August and what professional forecasters told the third-quarter SPF is the cleanest way to see why the three-year print matters for policy. Forecasters still have inflation back at 2.0% in 2028. ECB staff projections published with the 10 September decision have it at 2.1% in 2028. Consumers now have it at 2.9% three years out.

THREE SETS OF INFLATION NUMBERS

Staff also see inflation excluding energy and food at 2.5% in 2026, 2.6% in 2027 and 2.3% in 2028. The SPF has that core measure at 2.4%, 2.2% and 2.1% across the same years, then 2.0% in 2031. Actual core inflation in August was 2.4%, with services easing to 3.0% from 3.3%. The HICP jump to 3.2% was an energy story: 14.3%, up from 10.3% in July. Food, alcohol and tobacco ran at 1.1%. Non-energy industrial goods ran at 1.2%.

Forecasters in the SPF put real GDP growth at 0.6% in 2026. Staff put it at 0.9%. Households put the next 12 months at -1.2%. The consumer recession number has been stuck there, and it did not worsen in August. What worsened is the price path they attach to that weak activity outlook.

Fieldwork Dates Left the September Hike Out

The Governing Council raised the three key rates by 25 basis points on 10 September 2026, taking the deposit facility rate at 2.50%, the main refinancing rate to 2.65% and the marginal lending facility to 2.90%, with effect from 16 September. The previous change, on 17 June 2026, had put the deposit rate at 2.25%. That makes September the second increase of 2026.

Staff projections published with that decision have headline inflation averaging 3.0% in 2026, 2.5% in 2027 and 2.1% in 2028, with growth at 0.9%, 1.4% and 1.5%. The Council said the conflict in the Middle East continues to generate inflation pressures, and that inflation is set to remain well above target for an extended period. It also said it was not pre-committing to a particular rate path and would decide meeting by meeting. The next scheduled decision is 29 October 2026.

THE DATES THAT FRAME THE PRINT

1. 6-24 August 2026: Fieldwork for the latest CES wave, about 19,000 adults in Belgium, Germany, Ireland, Greece, Spain, France, Italy, the Netherlands, Austria, Portugal and Finland.
2. 10 September 2026: Governing Council raises the three key rates by 25 basis points, 17 days after the last survey answers came in.
3. 16 September 2026: New official rates take effect, including the 2.50% deposit facility.
4. 18 September 2026: ECB releases the August CES, two days after the new rates apply and eight days after the decision.
5. 23 October 2026: Scheduled release of the September CES, the first wave that can include the September hike in household answers.
6. 29 October 2026: Next Governing Council monetary-policy meeting.

August’s CES cannot be read as a verdict on the September rise, because nobody in the sample had seen it. It can be read as the starting point the Council will have on 29 October, together with whatever the 23 October CES wave shows. Energy at 14.3% was already in the August HICP. The three-year median at 2.9% is now in the household survey that sits beside it.

Mortgage Rate Forecasts Hold at 4.9%

Housing answers were as still as the income and spending answers. Consumers expected the price of their home to increase by 3.4% over the next 12 months, unchanged from June. Expected mortgage interest rates 12 months ahead stayed at 4.9%. Those two numbers have not followed the inflation medians up, even though the deposit facility is now 2.50% and the Council has already raised rates twice this year.

A 4.9% mortgage expectation against a 3.4% home-price expectation is a real-rate squeeze for anyone who has to refinance. It is also a reason the three-year inflation view can feed through without a boom: households plan to keep spending 3.6% more in nominal terms, on 1.0% more income, while they still see the economy contracting 1.2% and they still see mortgage rates at 4.9%. The September CES, due on 23 October, is the first chance to see whether those housing and rate answers finally move after the 16 September increase, or whether the 2.9% three-year inflation median is the number that moves first.