Search Options
Home Media Explainers Research & Publications Statistics Monetary Policy The €uro Payments & Markets Careers
Suggestions
Sort by

1 Overview

Trade tariffs and related uncertainty contributed to strong fluctuations in economic activity during the first half of 2025, with frontloading of activity, especially in Ireland. The unwinding of these factors in the second half of the year is expected to entail further volatility, blurring signals of the underlying momentum of the euro area economy. In fact, looking through the volatility caused by the fluctuations in Irish data, economic growth in the rest of the euro area was more stable, and it is expected to remain so in the second half of the year. Although the new US-EU trade agreement implies higher tariffs on euro area exports to the United States, it has helped to reduce trade policy uncertainty. Later in the horizon economic growth in the euro area is projected to strengthen, supported by several factors. Rising real wages and employment, together with new government spending on infrastructure and defence, mainly in Germany, should bolster euro area domestic demand. Furthermore, less restrictive financing conditions – mainly reflecting recent monetary policy decisions – and a rebound in foreign demand in 2027 are also seen to support the growth outlook. Annual average real GDP growth is projected to be 1.2% in 2025, 1.0% in 2026 and 1.3% in 2027. Compared with the June 2025 Eurosystem staff macroeconomic projections, the outlook for GDP growth has been revised up by 0.3 percentage points for 2025, reflecting better than expected incoming data and a carry-over effect from revisions to historical data. As not all of the data surprises relate to stronger than previously assumed frontloading of activity, they are only seen to be partly offset in the second half of the year. The appreciation of the euro and weaker foreign demand (in part related to somewhat higher tariffs than assumed in the June projections) have resulted in a 0.1 percentage point downward revision for 2026. The projection for 2027 remains unchanged.[1]

The inflation outlook is consistent with inflation stabilising around the medium-term target of 2%. Headline inflation, as measured by the Harmonised Index of Consumer Prices (HICP), is projected to move sideways, at around 2%, for the rest of 2025, and then to drop to 1.7% in 2026 before recovering to 1.9% in 2027. The drop in 2026 reflects a further gradual easing in the non-energy components, while energy inflation is expected to remain volatile, but to rise over the projection horizon, in part because of the start of the EU Emissions Trading System 2 (ETS2) in 2027. Food inflation is expected to remain elevated initially, as lagged effects from past price increases in international food commodities feed through, but to moderate to rates somewhat above 2% in 2026 and 2027. HICP inflation excluding energy and food (HICPX) is expected to decline as wage pressures recede and services inflation moderates, and as the appreciation of the euro feeds through the pricing chain and curbs goods inflation. Lower wage growth, as past real wage losses have been recouped, coupled with a recovery in productivity growth, is expected to lead to significantly slower unit labour cost growth. Compared with the June 2025 projections, the outlook for headline HICP inflation has been revised up by 0.1 percentage points for both 2025 and 2026 on account of higher energy commodity price outcomes and assumptions, as well as lagged effects from higher international food commodity prices, which more than offset the appreciation of the euro. For 2027, the lagged effects of the appreciation of the euro are seen to predominate, resulting in a 0.1 percentage point downward revision.

Table 1

Growth and inflation projections for the euro area

(annual percentage changes, revisions in percentage points)

September 2025

Revisions vs June 2025

2024

2025

2026

2027

2024

2025

2026

2027

Real GDP

0.8

1.2

1.0

1.3

0.0

0.3

-0.1

0.0

HICP

2.4

2.1

1.7

1.9

0.0

0.1

0.1

-0.1

HICP excluding energy and food

2.8

2.4

1.9

1.8

0.0

0.0

0.0

-0.1

Notes: Real GDP figures refer to annual averages of seasonally and working day-adjusted data. Historical data may differ from the latest Eurostat publications owing to data releases after the cut-off date for the projections. Revisions are calculated from rounded figures. Data are available for downloading, also at quarterly frequency, from the Macroeconomic Projection Database on the ECB’s website.

2 The international environment

While US tariffs have continued to increase since the cut-off date for the June 2025 projections, the announcement of preliminary agreements on trade deals and the extension of the tariff truce between the United States and China have mitigated trade policy uncertainty, which nonetheless remains elevated. The baseline assumptions for tariffs and trade policy uncertainty are presented in more detail in Box 1. Section 232 investigations by the US Administration regarding sectoral tariffs for pharmaceuticals, semiconductors and lumber are still ongoing, thus posing an upside risk to the tariff assumptions.

The outlook for global growth is expected to weaken, although less sharply than envisaged in the June 2025 projections.[2] Global real GDP growth is projected to be 3.3% in 2025, down from 3.6% last year, and then to decrease further, to 3.1% in 2026, before recovering to 3.3% in 2027 (Table 2). Real GDP growth surprised on the upside in the second quarter in major economies (including the United States, China and the United Kingdom), but recent economic news – such as weakening labour demand in the United States and subdued retail sales and investment in China – point to a slowdown in activity in the second half of 2025. However, the outlook has been revised up compared with the June 2025 projections, partly reflecting positive data surprises, while fiscal expansion in the United States, receding trade policy uncertainty, and improving financial conditions are projected to cushion the impact of new tariffs.

Global trade growth is projected to decline significantly, to 2.8% in 2025 and 1.5% in 2026, before recovering to 3.1% in 2027. Despite the apparent resilience of trade in the first half of 2025, owing to the frontloading of US imports ahead of the imposition of tariffs, US trade policies are weighing on trade dynamics. The projected sharp slowdown reflects the unwinding of frontloading, as well as the impact of tariffs and elevated trade policy uncertainty – with the latter dampening investment and leading to a less trade-intensive composition of global demand. Global trade is expected to recover in 2027 – notably as US imports rebound after being seen to contract in 2026 – although at a pace somewhat below global real GDP growth. Compared with the June projections, world import growth has been revised down for both 2025 and 2026, mostly reflecting the newly implemented tariffs as well as a downward revision to the assessment of the import-intensity of growth in China which more than offset the impact of upward revisions to real GDP growth.

The outlook for euro area foreign demand remains subdued and has been revised down slightly compared with the June projections. Growth in euro area foreign demand is expected to weaken to 2.8% in 2025, from 3.6% in 2024, and then to slow further to 1.4% in 2026, before recovering to 3.1% in 2027 (Table 2). Compared with the June 2025 projections, euro area foreign demand has been revised down for 2026. The cumulative revisions to euro area foreign demand over 2025-27 are, however, smaller than revisions to global trade, reflecting geographical composition effects – namely higher than expected imports in the United Kingdom in the second quarter and upward revisions for central and eastern European countries on account of the German fiscal stimulus and defence-related spending.

Table 2

The international environment

(annual percentage changes, revisions in percentage points)

 

September 2025

Revisions vs June 2025

2024

2025

2026

2027

2024

2025

2026

2027

World real GDP (excluding the euro area)

3.6

3.3

3.1

3.3

0.0

0.2

0.2

0.1

Global trade (excluding the euro area)1)

4.2

2.8

1.5

3.1

0.0

-0.3

-0.2

0.0

Euro area foreign demand2)

3.6

2.8

1.4

3.1

0.1

0.0

-0.3

0.0

World CPI (excluding the euro area)

4.0

3.2

2.9

2.5

0.0

-0.1

0.1

0.0

Export prices of competitors in national currency3)

2.3

1.6

2.1

2.3

-0.2

-0.2

-0.5

0.0

Note: Revisions are calculated from rounded figures.
1) Calculated as a weighted average of imports.
2) Calculated as a weighted average of imports of euro area trading partners.
3) Calculated as a weighted average of the export deflators of euro area trading partners.

Global headline inflation is projected to decline gradually over the projection horizon, notwithstanding inflationary pressures in the United States.[3] Global headline consumer price index (CPI) inflation is projected to moderate from 4.0% in 2024 to 3.2% in 2025, and then to fall further to 2.9% in 2026 and 2.5% in 2027. Compared with the June 2025 projections, global inflation has been revised down slightly for 2025 owing to negative data surprises for the second quarter in large economies (including the United States, China and India), but has been revised up for 2026. The latter is mainly on account of developments in the United States where higher tariffs and the fiscal expansion are projected to intensify inflationary pressures. These developments are partly offset by downward revisions across emerging market economies, reflecting lower than expected consumer price inflation momentum in China and the impact of lower food prices.

Euro area competitors’ export prices are projected to increase at a pace below the historical average, driven by persistently weak export prices in China. Euro area competitors’ export prices (in national currencies) are expected to increase by 1.6% in 2025, 2.1% in 2026 and 2.3% in 2027. This is below the historical average (2.6% over 2000-19) as Chinese export price inflation is projected to remain in negative territory until late 2026. Compared with the June 2025 projections, downward revisions for 2025 and 2026 reflect the impact of lower export prices in China, partly on account of the entrenched deflation of producer prices and lower food prices, which more than compensate for the effect of slightly higher assumptions for oil prices.

Box 1
Assumptions about US tariffs and trade policy uncertainty, and technical assumptions

Assumptions about US tariffs and trade policy uncertainty

Compared with the June 2025 projections, the US effective tariff rates on euro area imports and those of all its trading partners are higher following the US-EU trade deal and other announcements. The staff projections take into account those tariff rates, pauses and exemptions in effect at the time of the cut-off date for the projections (28 August) and assume that these will remain at the same level over the entire projection horizon.

  • The effective tariff rate for euro area exports of goods to the United States is assumed to be 13.1%, up from the 10.2% assumed in the June projections (Chart A, panel a). This reflects the trade deal of 27 July 2025 and the subsequent US-EU joint statement, which foresees that the tariff rate applied is the higher of 15% and the most-favoured nation (MFN) rate for all goods that are not affected by sectoral tariffs. This compares with a flat 10% tariff rate assumed in the June 2025 projections. Sectoral tariffs include higher tariffs on steel and aluminium (increasing from 25% in the June projections to 50%) and a new 50% tariff on the aluminium and steel content of 407 products, as well as on copper more broadly. Cars and car components are subject to a 15% tariff (down from 25% in the June projections), while for aircraft and aircraft parts now only the MFN rate applies.[4] For products currently under Section 232 investigation, namely pharmaceutical products, semiconductors and lumber, the tariff is assumed to be at the rate applicable before the Trump Administration took office, which would mean a rate close to zero. When considering also trade in services, the effective tariff rate has increased by 1.8 percentage points to 8.1% from its level in the June projections.
  • The overall US effective tariff rate on goods imports from all trading partners has increased to 21.0%, from 16.8% in the June projections (Chart A, panel b). This reflects the doubling of universal tariffs on steel and aluminium (from 25% in the June 2025 projections to 50%) as well as the addition of 407 steel and aluminium derivatives to the list of tariff-affected products, new universal tariffs on copper (50%), bilateral trade deals signed with key trading partners (e.g. the European Union, Japan, Korea and Vietnam), higher reciprocal rates implemented on 7 August in the case of countries with which no trade deal had been reached (e.g. India and Switzerland), tariff hikes on Brazilian exports (to 50%) and Canadian exports (an increase to 35% in duties on goods that do not comply with the United States-Mexico-Canada Agreement), and secondary tariffs on India (an additional 25 percentage points) to penalise its purchases of Russian oil.

Chart A

Assumed US effective tariff rates for goods in the September 2025 projections versus the June 2025 projections

a) Imports from the euro area (goods only)

b) Imports from all trading partners (goods only)

(percentages and percentage point contributions)

(percentages and percentage point contributions)

Sources: Conteduca, Mancini, and Borin (2025), CEPII’s ”Base pour l’Analyse du Commerce International” (BACI), World Integrated Trade Solution, Trade Data Monitor and ECB staff calculations.
Notes: Panel a) shows average tariff rates weighted at the product level using 2024 trade data, accounting for permanent and temporary exemptions. “Metals” includes the additional 25% tariff on steel and aluminium and the 50% tariff on copper together with the 50% tariffs on the aluminium and steel content of an additional 407 products. “US-EU deal” includes the 15% tariffs on cars and car components, the MFN rate for aircraft and aircraft parts and the higher of 15% and the MFN rate for other goods. Panel b) shows average tariff rates weighted at the product level using 2023 trade data, accounting for permanent and temporary exemptions. “Metals” covers the doubling of tariffs on steel and aluminium (from 25% in the June 2025 projections to 50% in the September 2025 projections), the 50% universal US tariffs on copper and the 50% tariffs on the steel and aluminium content of an additional 407 products. “Others” includes tariff increases affecting Brazil and Canada, secondary tariffs on India, higher reciprocal rates announced on 1 August 2025 and preliminary trade agreements between the United States and key trading partners (e.g. Vietnam, Japan and Korea).

Trade policy uncertainty has declined faster than expected in the June projections, although it remains elevated by historical standards. In July trade policy uncertainty stood roughly 55% below the level assumed in the June projections, mainly on account of the announcement of trade deals.[5] As a result, trade policy uncertainty is assumed to be lower – albeit still elevated –over the projection horizon compared with the previous projections, while following a similar declining path (Chart B, panel a).

Higher US tariffs and trade policy uncertainty since the March projections have reduced the outlook for real GDP growth in the euro area by 0.7 percentage points cumulatively over the period 2025-27 (Chart B, panel b), while the impact on inflation is negligible. The somewhat higher tariff assumptions since June are assessed to imply an additional cumulated impact of -0.1 percentage points compared with the June 2025 projections, concentrated in 2025 and 2026. The limited impact, which excludes exchange rate effects, relates both to the small changes in the effective tariff rates and to the fact that the increases are relatively smaller than for other competitors trading with the United States (Chart A). As regards trade policy uncertainty, the reduction in uncertainty since the June projections has led to a small upward impact on real GDP growth. As a result of these two partly offsetting factors, the overall impact of tariffs and uncertainty on GDP growth is projected to be slightly more negative than in the June projections and amounts to -0.7 percentage points cumulated over 2025-27. As in the June projections, the impact of US tariffs and uncertainty on euro area HICP inflation is seen as very limited, with a cumulated impact over the period 2025-27 of less than 0.2 percentage points.

Chart B

Staff assumptions on trade policy uncertainty and the impact of tariffs and uncertainty on growth

a) Evolution of trade policy uncertainty

b) Impact of tariffs and uncertainty on euro area real GDP growth (cumulative impact 2025-27)

(index, three-month moving average)

(percentage points)

Sources: Caldara et al., “Does Trade Policy Uncertainty Affect Global Economic Activity?”, September 4 2019, and ECB staff calculations.
Notes: In panel a) the figures refer to a “cleaned” measure of trade policy uncertainty (see footnote 5). The latest observation is for July 2025. For panel b), the estimates exclude the impact of tariffs on the technical assumptions (such as the exchange rate).

Technical assumptions

Compared with the June 2025 projections, the main changes to the technical assumptions are higher oil prices, lower gas and commodity prices and a stronger euro exchange rate, but the financial assumptions are broadly unchanged. Oil price assumptions have been revised up by around 3% on average for 2025-27, and electricity price assumptions by around 2%, while the assumptions for wholesale gas prices have been revised somewhat down. The price for EU Emissions Trading System 1 (ETS1) allowances is also assumed to be somewhat lower, while for EU Emissions Trading System 2 (ETS2) the price was kept unchanged. Non-energy commodity prices have been revised slightly down. The euro has appreciated by 3.3% against the US dollar and by 2.3% in nominal effective terms since the June projections. The assumptions for interest rates are broadly unchanged.

Table

Technical assumptions

 

September 2025

Revisions vs June 2025

2024

2025

2026

2027

2025

2026

2027

Commodities:

Oil price (USD/barrel)

81.2

69.7

65.1

65.1

4.5

3.7

1.4

Natural gas prices (EUR/MWh)

34.4

37.3

32.9

29.7

-1.8

-1.2

1.4

Wholesale electricity prices (EUR/MWh)

77.7

84.6

77.8

73.2

2.8

0.4

1.9

EU Emissions Trading System 1 (ETS1) allowances (EUR/tonne)

65.2

70.7

71.9

73.8

-0.9

-1.8

-2.2

EU Emissions Trading System 2 (ETS2) allowances (EUR/tonne)

-

-

-

59.0

-

-

0.0

Non-energy commodity prices, in USD (annual percentage change)

9.2

4.8

-1.0

0.9

-2.0

-0.6

0.3

Exchange rates:

USD/EUR exchange rate

1.08

1.13

1.16

1.16

1.9

3.3

3.3

Euro nominal effective exchange rate (EER41) (Q1 1999 = 100)

124.1

127.7

129.9

129.9

1.2

2.3

2.3

Financial assumptions:

Three-month EURIBOR (percentage per annum)

3.6

2.2

1.9

2.1

0.0

0.0

-0.1

Ten-year government bond yields (percentage per annum)

2.9

3.1

3.4

3.6

0.0

0.0

0.0

Notes: Revisions are expressed as percentages for levels and as percentage points for growth rates and percentages per annum. Revisions for growth rates and interest rates are calculated on figures rounded to one decimal place, while revisions reported as percentage changes are calculated on unrounded figures. The technical assumptions about euro area interest rates and commodity prices are based on market expectations, with a cut-off date of 15 August 2025. Oil prices refer to Brent crude oil spot and futures prices. Gas prices refer to the Dutch TTF gas spot and futures prices. Electricity prices refer to the average wholesale spot and futures price for the five largest euro area countries. The “synthetic” future price for ETS1 allowances (EUA) is derived as the end-of-month linearly interpolated value of the two nearest European Energy Exchange EUA futures. Monthly EUA futures prices are then averaged to produce an equivalent to annual frequency. In the absence of meaningful trading of ETS2 allowances, the price assumptions were set by staff at the threshold price above which additional allowances will be released, with the price updated to 2027 prices (for further details, see the box entitled “Assessing the impact of climate change transition policies on growth and inflation” in the December 2024 Eurosystem staff projections). The paths of commodity prices are implied by futures markets in the ten working days ending on the cut-off date. Bilateral exchange rates are assumed to remain unchanged over the projection horizon at the average levels prevailing in the ten working days ending on the cut-off date. The assumptions for euro area ten-year nominal government bond yields are based on the average of countries’ ten-year bond yields, weighted by annual GDP figures. Where the necessary data exist, the country-specific ten-year nominal government bond yields are defined as the ten-year benchmark bond yields prolonged using the forward par yields derived, on the cut-off date, from the corresponding country-specific yield curves. For the other countries, the country-specific ten-year government bond yields are defined as the ten-year benchmark bond yield prolonged using a constant spread (observed on the cut-off date) over the technical euro area risk-free long-term interest rate assumption.

3 Real economy

Euro area economic activity rose by 0.1% in the second quarter of 2025, which was slightly below the June projections, after much stronger than estimated growth in the first quarter (Chart 1). As a result of the outcome for the first quarter of 2025 and an upward revision to data for the fourth quarter of 2024, the level of GDP in the second quarter was still about 0.5% higher than foreseen in the June projections. In the first quarter of 2025 growth was strongly supported by the frontloading of exports in anticipation of higher US tariffs. This effect started to unwind in the second quarter, primarily – but not exclusively – driven by developments in the Irish pharmaceutical sector. Across sectors, industrial activity was subdued in the second quarter, amid the unwinding of frontloading effects. Meanwhile the underlying recovery in demand for goods continued, fostered by improving real incomes, while activity in the services sector continued to increase.

Economic activity is expected to stagnate in the third quarter owing to an expected further unwinding of the frontloading of exports, particularly in Ireland, and owing to higher tariffs, but it should recover in the fourth quarter. The composite output Purchasing Managers’ Index (PMI) increased to 51.0 in August, suggesting positive underlying growth momentum. The industrial sector showed a more pronounced recovery, with a manufacturing output PMI of 52.5, while the services sector PMI continued to signal only moderate growth (50.5). The European Commission’s Economic Sentiment Indicator fell back in August after improving in July and remains at subdued levels. Contrary to survey indicators, the industrial production data for June highlight ongoing challenges in the industrial sector, which are partly related to the unwinding of frontloading effects but also related to persistent competitiveness issues. The baseline projection assumes that over the projection horizon US tariffs vis-à-vis the EU remain at the level announced in the context of the US-EU trade deal agreed on 27 July, and that trade policy uncertainty will gradually decline but remain elevated with a fading dampening impact on activity (see Box 1). An expected sharp contraction in growth in Ireland in the third quarter of 2025, related to the unwinding of earlier frontloading effects, is seen to offset growth of 0.2% in the rest of the euro area, leading to a stagnation of euro area activity (Chart 1, panel c). Moderate underlying growth momentum, in the context of slightly higher tariffs than previously assumed, but lower uncertainty, is seen to result in a return to growth in the fourth quarter, at 0.2%. The growth outturns and revisions to historical data over recent quarters account for almost all of the projected annual growth rate for 2025 of 1.2% (Chart 1, panel d).

Chart 1

Euro area real GDP

a) Real GDP growth

b) Real GDP level

(quarter-on-quarter percentage changes, seasonally and working day-adjusted quarterly data)

(chain-linked volumes (2020); EUR billions)

c) Euro area real GDP growth – impact of Ireland

d) Euro area real GDP growth – impact of within-year growth and carry-over effects

(quarter-on-quarter percentage changes and percentage point contributions, seasonally and working day-adjusted quarterly data)

(annual percentage changes and percentage point contributions)

Notes: Historical data may differ from the latest Eurostat publications owing to data releases after the cut-off date for the projections. The vertical line indicates the start of the current projection horizon. In panel a) the ranges shown around the central projections provide a measure of the degree of uncertainty and are symmetric by construction. They are based on past projection errors, after adjustment for outliers. The bands, from darkest to lightest, depict the 30%, 60% and 90% probabilities that the outcome of real GDP growth will fall within the respective intervals. For more information, see the box entitled “Illustrating the uncertainty surrounding the projections” in the March 2023 ECB staff macroeconomic projections for the euro area. In panel c) the red bars refer to the impact of Irish real GDP growth on the euro area aggregate and the green bars refer to growth in the other euro area countries. In panel d) carry-over effects indicate the impact on the annual average growth rate for year t coming from growth in year t-1. The striped, red bar shows the impact of the outturns for the first and the second quarters of 2025 on within-year growth for 2025.

Quarterly real GDP growth is projected to increase to 0.3% over the medium term, supported by rising disposable income, reduced uncertainty, stronger foreign demand, and fiscal stimulus related to defence and infrastructure. Private consumption growth is expected to be sustained by increasing purchasing power of households, driven by moderate wage and employment growth – despite some cooling in the labour market – together with easing inflation and a slight decline in the household saving rate. Investment is projected to gradually strengthen over the projection horizon, reflecting a reduction in uncertainty, increased defence and infrastructure spending, support from Next Generation EU (NGEU) programme funds and improving demand conditions in line with the cyclical recovery. However, weakness in foreign demand continuing into early 2026, persistent euro area competitiveness challenges, higher tariffs and a stronger exchange rate are likely to lead to continued losses in export market shares, which are also seen to dampen the recovery of business investment. Fiscal measures announced in the first half of the year related to defence and infrastructure should provide a limited boost to growth in the medium term amounting to 0.25 percentage points cumulatively over the projection horizon. Overall, quarterly GDP growth is expected to remain at 0.3% from the start of 2026. A dip in annual average real GDP growth, from 1.2% in 2025 to 1.0% in 2026, is seen to be a consequence of estimated lower carry-over effects in 2026 following the expected weak growth in the second half of 2025 (Chart 1, panel d).

The adverse effects on growth stemming from past monetary policy tightening are likely to have already materialised, while the policy rate cuts since June 2024 should now start supporting growth. Following the policy rate cuts since June 2024 and based on market expectations regarding the future path of interest rates at the time of the cut-off date for the projections (Box 1), little is seen to remain of the negative impact of the previous monetary policy tightening cycle on economic growth throughout the remainder of 2025. However, there is considerable uncertainty surrounding the magnitude and timing of the past and future impacts.

Real GDP growth should continue to be driven by private consumption. From an expenditure perspective (Chart 2, panel a), private consumption is projected to grow robustly throughout the horizon, providing the largest contribution to real GDP growth. Moreover, government consumption and total investment should increase every year from 2025 to 2027, with private investment compensating the expected strong cliff effect in public investment in 2027 resulting from the expiry of NGEU grant funding. By contrast, net exports are projected to make negative contributions to real GDP growth in 2025 and 2026 – with the contribution of imports (-1.2 percentage points and -0.9 percentage points respectively) outstripping that of exports (0.6 percentage points and 0.5 percentage points respectively) – but this is partly offset by a positive contribution from changes in inventories in 2025 (Chart 2, panel b). The highly volatile inventory changes over 2025 are related to stockbuilding and destocking in the context of frontloading effects and their unwinding.

Chart 2

Euro area real GDP growth – decomposition into the main expenditure components

a) September 2025 projections

b) Revisions vs the June 2025 projections

(annual percentage changes and percentage point contributions)

(percentage points and percentage point contributions)

Notes: Data are seasonally and working day-adjusted. Historical data may differ from the latest Eurostat publications owing to data releases after the cut-off date for the projections. The vertical line indicates the start of the projection horizon. Revisions are calculated based on unrounded figures.

Compared with the June 2025 projections, the projection for real GDP growth has been revised up by 0.3 percentage points for 2025, revised down by 0.1 percentage points for 2026, and remains unchanged for 2027. The upward revision to the projection for 2025 reflects positive carry-over effects stemming from revisions to historical data, as well as the significantly better than expected data outturn for the first quarter of 2025 and the latest indicators suggesting a slight upward revision to growth in the third quarter (Chart 3). Somewhat higher tariffs, weaker foreign demand and a weakening of price competitiveness related to the stronger euro more than offset the impact of improving confidence on activity, leading to a downward revision of 0.1 percentage points to the growth outlook for 2026. The outlook for 2027 remains unchanged. In terms of expenditure components (Chart 2, panel b), the largest revisions for 2025 refer to investment (with more than half of them relating to volatile investment in intellectual property products in Ireland) and to net exports.

Chart 3

Revisions to real GDP growth projections compared with the June 2025 projections – decomposition into carry-over and within-year effects

(percentage points and percentage point contributions)

Notes: The vertical line indicates the start of the projection horizon. Revisions are based on unrounded figures. The striped, red bar shows the impact of the outturns for the first and second quarters of 2025 on within-year growth for 2025.

Table 3

Real GDP, trade and labour market projections

(annual percentage changes, unless otherwise indicated, revisions in percentage points)

 

September 2025

Revisions vs June 2025

2024

2025

2026

2027

2024

2025

2026

2027

Real GDP

0.8

1.2

1.0

1.3

0.0

0.3

-0.1

0.0

Private consumption

1.1

1.3

1.3

1.3

0.1

0.1

0.1

0.1

Government consumption

2.3

1.6

1.1

1.1

-0.3

0.0

-0.1

0.1

Investment

-2.1

2.1

1.7

1.8

-0.3

1.4

0.0

-0.1

Exports1)

0.7

1.3

1.1

2.5

-0.4

0.8

-0.5

-0.1

Imports1)

-0.2

2.8

2.0

2.7

-0.4

0.9

0.0

0.0

Contribution to GDP from:

Domestic demand

0.6

1.5

1.2

1.3

-0.1

0.3

-0.1

0.0

Net exports

0.4

-0.6

-0.3

0.0

0.0

0.0

-0.2

-0.1

Inventory changes

-0.2

0.4

0.1

0.0

0.1

0.0

0.1

0.0

Real disposable income

2.4

0.9

0.9

0.7

0.2

0.1

-0.1

-0.1

Household saving ratio (% of disposable income)

15.0

14.8

14.5

14.0

0.0

0.1

0.0

-0.1

Employment2)

1.0

0.6

0.5

0.5

0.0

0.0

0.0

-0.1

Unemployment rate

6.4

6.4

6.3

6.1

0.0

0.1

0.0

0.1

Current account (% of GDP)

2.6

2.4

2.5

2.5

-0.1

-0.3

-0.1

-0.1

Notes: Real GDP and components refer to seasonally and working day-adjusted data. Historical data may differ from the latest Eurostat publications owing to data releases after the cut-off date for the projections. Revisions are calculated from rounded data. Data are available for downloading, also at quarterly frequency, from the Macroeconomic Projection Database on the ECB’s website.
1) This includes intra-euro area trade.
2) Persons employed.

Private consumption is expected to strengthen, supported by increases in labour compensation and a gradual decline in the saving ratio. Private consumption continued to expand in the first quarter of 2025 and likely also in the second quarter, despite some moderation in the growth rates compared with the end of 2024 on the back of declining confidence amid trade tensions. Household spending should strengthen from an annual rate of increase of around 0.8% in 2023-24 to 1.3% in 2025-27. This recovery largely reflects rising real disposable income, supported mainly by wage income and, to a lesser extent, by non-wage income, particularly from self-employment. In addition, private consumption growth should benefit from a moderate decline in the saving ratio as consumer spending and savings gradually normalise. An expected gradual recovery in consumer confidence towards its historical norm over the medium term, amid overall resilient labour market prospects, may contribute to a gradual decline in the saving ratio (Chart 4). However, the household saving ratio is expected to remain elevated overall, reflecting the fact that interest rates remain somewhat above historical average levels and access to credit is still tight.

Chart 4

Household consumption, income and savings

(annual percentage changes and percentage point contributions)

Notes: Data are seasonally and working day-adjusted. An increase in the household saving ratio implies a negative contribution of savings to consumption growth. Historical data may differ from the latest Eurostat publications owing to data releases after the cut-off date for the projections. Statistical discrepancies between the national and sectoral accounts and their different publication schedules result in minor differences between the figures for private consumption growth and the respective contributions from income and the (inverted) change in the household saving ratio. The vertical line indicates the start of the current projection horizon.

Housing investment is estimated to have contracted in the second quarter of 2025, but should recover over the projection horizon on the back of favourable financing conditions and rising household real incomes. After a strong outturn for the first quarter of 2025, housing investment is estimated to have declined in the second quarter, but it should grow at an increasing pace in the coming quarters. The recovery in housing investment reflects improved housing demand – as also evidenced by the recent upturn in the number of residential building permits issued – and is attributable to the steady decline in mortgage rates since the end of 2023 as well as support from continued growth in real household income.

Although distorted by tariff-related factors in the near term, business investment is expected to gradually recover over the medium term, driven by improving economic activity, reduced uncertainty and the impact of national policies. Following a small boost in the first quarter – driven by frontloading effects in anticipation of higher tariffs, even when volatility in investment in Irish intellectual property products is excluded – euro area business investment is estimated to have stagnated in the second quarter. It is projected to contract slightly in the third quarter of 2025 owing to waning anticipatory effects related to frontloading, as well as elevated policy uncertainty, higher tariffs and the appreciation of the euro. Investment is expected to gradually rise from the fourth quarter of 2025, as domestic and foreign demand picks up, uncertainty declines, and the implementation of national policies incentivises additional private investment spending related to defence and infrastructure.

The euro area’s export prospects are still being dampened by higher US tariffs, the appreciation of the euro, and persistent competitive pressures. The imposition of US tariffs as outlined in the US-EU trade deal (see Box 1) has further constrained the outlook for euro area exports. Euro area firms are likely to face challenges in competing against domestic producers in the US market, while the tariffs imposed on products from the euro area are generally lower than those on products from other exporting economies. At the same time, the appreciation of the euro is also expected to reduce the competitiveness of euro area goods in global markets over the medium term. This is seen to result in overall limited competitiveness gains for the euro area relative to other trading partners of the United States, while price competitiveness relative to US producers has clearly deteriorated (Chart 5).

Chart 5

Changes in effective tariff rates and exchange rates in the US market since the current US Administration took office

(percentage points)

Sources: World Bank World Integrated Trade Solution, Trade Data Monitor, CEPII’s “Base pour l’Analyse du Commerce International” (BACI), ECB and ECB staff calculations.
Notes: The latest effective tariff rate is calculated for goods only and includes new tariffs on the aluminium and steel content of products on the new list of 407 product categories, applicable to all countries, and the 50% tariff rate applicable to India. The exchange rate changes are calculated based on the ten-day window up to 15 August relative to the ten-day window up to 17 January 2025. The aggregate “Excl. EA” is calculated based on the 23 main trading partners of the United States, which, together with the euro area, account for 89% of total US trade in goods. The effective tariff rate for the euro area includes the tariff regime detailed in the US-EU joint statement of 21 August, with tariffs on cars and on car parts set at 15%.

While a decline in trade policy uncertainty has provided some relief, uncertainty levels remain elevated compared with historical norms. In particular, some uncertainty still surrounds the US-EU trade agreement for pharmaceuticals and semiconductors (currently their tariff rates are still the same as before the current US Administration took office, however, they may increase up to the 15% ceiling after the US Section 232 investigation), which could have a further impact on euro area exports. Although euro area exports surged in the first quarter, driven by firms frontloading orders ahead of higher US tariffs, this momentum has faded as the effects of frontloading unwind. Looking ahead, euro area exports are expected to struggle to keep pace with global demand due to competitiveness issues, likely resulting in further losses of export market shares. Compared with the June 2025 projections, euro area exports have been revised up for 2025, with growth expected to reach 1.3%, reflecting the temporary boost in the first quarter owing to frontloading, and have been revised down to 1.1% for 2026, but are then projected to recover to 2.5% in 2027.

On the import side, growth is projected to be robust with euro area imports expected to grow by 2.8% in 2025, 2.0% in 2026 and 2.7% in 2027. This is seen to be driven partly by the frontloading of exports at the start of 2025, a considerable increase in sourcing from China, and an increase in domestic investment. Overall, net trade is expected to make a negative contribution to euro area growth, with a -0.6 percentage point impact in 2025 and a -0.3 percentage point impact in 2026. In 2027 net trade is expected to make a neutral contribution to growth.

The labour market is set to remain resilient overall, with the unemployment rate projected to decline in 2026 and 2027. The unemployment rate is expected to increase in 2025, peaking in the second half of the year (Chart 6, panel a). It is then projected to decline over the medium term to reach a historical low level of 6.0% in the fourth quarter of 2027. The growth of the labour force is projected to decline over the projection horizon (in part owing to demographic developments), which also leads to a decline in the number of unemployed workers in the medium term (in part owing to the lower unemployment rate of older workers). Compared with the June staff projections, the unemployment rate is projected to be broadly unchanged and to move in a narrow band between 6.0% and 6.4% over the projection horizon. For 2027, it has been revised up by 0.1 percentage points.

Chart 6

Unemployment rate and employment

a) Unemployment rate

b) Employment

(percentage of labour force)

(index, Q1 2022 = 100)

Note: The vertical line indicates the start of the current projection horizon.

Employment growth is projected to slow over the projection horizon. Overall, employment is projected to grow at 0.6% in 2025 (Chart 6, panel b), suggesting robust employment growth in the short term. This persistent employment growth in 2025 is in part driven by robust labour force growth. However, labour force growth is projected to decline over the projection horizon, limiting the growth in employment. While employment growth in the first quarter of 2025 was slightly below the June projections, the outcome for the second quarter was slightly above. Employment is projected to continue to grow at a higher rate than previously expected throughout the remainder of the year.

Labour productivity growth has been stronger in 2025 than previously projected and is expected to continue its cyclical recovery. Over the full projection horizon labour productivity growth is seen to remain positive but far below its pre-pandemic trend (Chart 7). The speed of the recovery continues to be limited by structural factors, such as the gradual reallocation of economic activity towards the services sector, transition costs related to the greening of the economy, a lasting adverse impact of the energy price shock, the slow adoption of highly innovative artificial intelligence technologies, and the ageing population. The higher labour productivity growth in 2025 compared with the June projections is mostly due to the upward revision to GDP growth, in large part owing to developments in the Irish data which introduced some volatility into the series. As a result of this revision of past data, labour productivity in the first quarter of 2025 has been revised up by 0.4 percentage points. For 2025 as a whole, labour productivity growth has been revised up by 0.2 percentage points compared with the June projection. For 2026, productivity growth has been revised down. This was mostly driven by a downward revision to GDP growth, while employment growth remained broadly unchanged.

Chart 7

Labour productivity per person employed

a) Labour productivity level

(index, Q1 2022 = 100)

b) Labour productivity growth rate

(year-on-year growth)

Note: The vertical line indicates the start of the current projection horizon.

4 Fiscal outlook

The euro area fiscal stance is projected to tighten only slightly in 2025, to loosen in 2026 and to tighten again, somewhat more strongly, in 2027 (Table 4).[6] After tightening significantly in 2024, on account of both non-discretionary factors and fiscal policy measures, the fiscal stance is projected to tighten only slightly in 2025 owing mostly to discretionary revenue measures. These include increases in social security contributions and, to a lesser extent, increases in direct and indirect taxes.[7] These tax increases are partly offset by continued growth in public spending and fiscal loosening relating to non-discretionary factors.[8] In 2026 the fiscal stance is projected to loosen, mainly on account of higher public investment. This reflects the higher defence and infrastructure spending, particularly in Germany (over 2026-27), incorporated into the baseline as of the June 2025 projections, as well as high NGEU-funded investment growth in Italy, Spain and other countries. In 2027 the tightening of the NGEU-adjusted fiscal stance and discretionary measures primarily reflects lower assumed government spending related to the expiry of the NGEU grant financing and some tightening of non-discretionary factors.[9]

Compared with the June 2025 projections, the discretionary fiscal policy measures incorporated into the baseline point to only a slight additional fiscal tightening over the projection horizon, particularly in 2026. This tightening in discretionary measures amounts to less than 0.1 percentage points of GDP cumulatively over the projection horizon, which stems mostly from 2026 and is partly reversed in 2027. Together with slightly revised non-discretionary factors, the fiscal stance is seen to tighten by close to 0.1 percentage points more than foreseen in the June projections over the period 2025-27. The tightening is mostly on account of upward revisions to net indirect taxes. Other sources of revisions include lower government consumption growth and higher direct taxes. In 2027 a marginally less tight fiscal stance than foreseen in June mainly reflects upward revisions to government consumption and fiscal transfers in several countries.

Table 4

Fiscal outlook for the euro area

(percentage of GDP; revisions in percentage points)

 

September 2025

Revisions vs June 2025

2024

2025

2026

2027

2024

2025

2026

2027

Fiscal stance1)

1.0

0.1

-0.2

0.4

0.1

0.1

0.1

-0.1

General government budget balance

-3.1

-2.9

-3.2

-3.4

0.0

0.1

0.2

0.1

Structural budget balance2)

-3.0

-2.9

-3.1

-3.3

0.1

0.1

0.2

0.2

General government gross debt

87.4

88.0

89.1

89.8

-0.1

-0.5

-0.5

-0.4

Notes: Revisions are based on unrounded figures.
1) The fiscal stance is measured as the change in the cyclically adjusted primary balance net of government support to the financial sector. The figures shown are also adjusted for grants under the Next Generation EU (NGEU) programme, which do not have an impact on the economy on the revenue side. A negative (positive) figure implies a loosening (tightening) of the fiscal stance.
2) The structural budget balance is calculated as the government balance net of transitory effects of the economic cycle (or the cyclically adjusted primary balance, referenced above, plus interest payments) and net of measures classified under the European System of Central Banks definition as temporary.

Regarding the euro area fiscal outlook, the budget deficit and debt ratios are projected to continue to rise over the horizon, albeit less than foreseen in the June projections (Table 4). After a decline expected in 2025, the euro area budget deficit is projected to increase and to reach 3.4% of GDP by the end of the projection horizon. This is mainly on account of interest payments, followed by a slight deterioration in the cyclically adjusted primary balance and the cyclical component of the budget balance. Compared with the June projections, the budget balance has been revised up over the entire projection horizon, but mostly as of 2026, primarily on account of the additional tightening of discretionary fiscal policy measures mentioned above. The euro area debt-to-GDP ratio is seen as being on an increasing path as the continuous primary deficits and positive deficit-debt adjustments more than offset the favourable, although rising, interest rate-growth differentials.

Euro area fiscal policy assumptions and projections continue to be surrounded by high uncertainty. This is mainly related to not yet specified defence spending plans, following the June 2025 NATO summit, as well as to euro area governments’ concrete budget plans for 2026 and beyond in the context of the EU fiscal framework. More information on such plans should become available this autumn. Finally, uncertainty relates to political risks in several jurisdictions.

5 Prices and costs

Headline inflation is expected to average 2.1% in 2025 and 1.7% in 2026, before edging up to 1.9% in 2027 (Chart 8). Headline inflation is expected to stay close to 2% for the remainder of 2025 and to move below 2% and remain there throughout 2026. The pattern in the first half of 2026 is strongly influenced by base effects in the energy component, which explain an initial drop to 1.6% in the first quarter followed by a rebound in the second quarter. More fundamentally, the lower rate for headline inflation in 2026 reflects lower HICPX inflation – particularly in services – and lower food inflation, accompanied by slightly negative energy inflation (Chart 9). The increase in headline inflation to 1.9% in 2027 reflects an upward impact from energy inflation coming from climate change transition-related fiscal measures, in particular the introduction of a new Emissions Trading System (ETS2).[10] HICPX inflation is anticipated to moderate further into the first half of 2026 and to level off at around 1.8% over the remainder of the projection horizon.

Chart 8

Euro area HICP inflation

(annual percentage changes)

Notes: The vertical line indicates the start of the current projection horizon. The ranges shown around the central projections provide a measure of the degree of uncertainty and are symmetric by construction. They are based on past projection errors, after adjustment for outliers. The bands, from darkest to lightest, depict the 30%, 60% and 90% probabilities that the outcome of HICP inflation will fall within the respective intervals. For more information, see the box entitled “Illustrating the uncertainty surrounding the projections” in the March 2023 ECB staff macroeconomic projections for the euro area..

Chart 9

Euro area HICP inflation – decomposition into the main components

(annual percentage changes, percentage points)

Note: The vertical line indicates the start of the current projection horizon.

Energy inflation is projected to remain negative until late 2026, before rebounding in 2027 owing to climate change-related fiscal measures (Chart 10, panel a). Sizeable negative rates are expected into 2026, predominantly stemming from developments in the transport fuels component, which reflect the broadly flat profile for the assumption on oil prices (see Box 1) as well as a downward base effect for the first quarter of 2026. At the beginning of 2026, there is also a small downward impact from an expected cut in electricity prices in Germany owing to lower levies and network fees. The increase in energy inflation from -1.1% in 2026 to 2.4% in 2027 largely reflects an upward impact from the implementation of the EU “Fit for 55” package, specifically the new ETS2 for the heating of buildings and for transport fuels.

Chart 10

Outlook for HICP energy and food inflation

a) HICP energy

(annual percentage changes)

b) HICP food

(annual percentage changes)

Notes: The vertical line indicates the start of the current projection horizon.

Following the recent surge, food inflation is expected to moderate as of the last quarter of 2025, declining to close to 2.0% in the second half of 2026 before rising slightly thereafter (Chart 10, panel b). Food inflation is expected to increase further to 3.2% in the third quarter of 2025, reflecting a pass-through of higher food commodity prices (especially for cocoa and coffee) and unfavourable weather conditions (for example the recent heatwaves). The subsequent moderation until mid-2026 reflects an easing in the assumptions for international and euro area food commodity prices accompanied by a downward base effect from the strong increases in mid-2025. Food inflation is projected to average 2.3% for both 2026 and 2027, below its long-term average, consistent with the assumption of benign energy and food commodity prices and easing labour cost pressures.

Table 5

Price and cost developments for the euro area

(annual percentage changes, revisions in percentage points)

 

September 2025

Revisions vs June 2025

2024

2025

2026

2027

2024

2025

2026

2027

HICP

2.4

2.1

1.7

1.9

0.0

0.1

0.1

-0.1

HICP excluding energy

2.9

2.5

2.0

1.9

0.0

0.0

0.0

-0.1

HICP excluding energy and food

2.8

2.4

1.9

1.8

0.0

0.0

0.0

-0.1

HICP excluding energy, food and changes in indirect taxes

2.8

2.3

1.9

1.8

0.0

-0.1

0.0

-0.1

HICP non-energy industrial goods

0.8

0.6

0.4

0.8

0.0

0.0

-0.3

-0.1

HICP services

4.0

3.4

2.7

2.3

0.0

0.0

0.1

-0.1

HICP energy

-2.2

-1.6

-1.1

2.4

0.0

0.7

1.0

-0.3

HICP food

2.9

2.9

2.3

2.3

0.0

0.2

-0.1

0.0

GDP deflator

3.0

2.3

2.1

2.0

0.1

0.0

0.1

-0.1

Import deflator

-0.4

0.1

0.6

1.7

0.1

-0.3

-0.6

-0.3

Compensation per employee

4.5

3.4

2.7

2.7

0.0

0.2

-0.1

-0.1

Productivity per employee

-0.2

0.6

0.5

0.8

-0.1

0.2

-0.2

0.0

Unit labour costs

4.7

2.8

2.2

1.9

0.0

0.0

0.1

-0.1

Unit profits1)

-1.4

0.8

1.6

2.0

0.0

-0.1

-0.4

-0.2

Notes: Revisions are calculated using figures rounded to one decimal place. The figures for the GDP and import deflators, unit labour costs, compensation per employee and productivity per employee refer to seasonally and working day-adjusted data. Historical data may differ from the latest Eurostat publications owing to data releases after the cut-off date for the projections. Data are available for downloading, also at quarterly frequency, from the Macroeconomic Projection Database on the ECB’s website.
1) Unit profits are defined as gross operating surplus and mixed income (adjusted for the income of the self-employed) per unit of real GDP.

HICPX inflation is expected to decline from 2.4% in 2025 to 1.9% in 2026 and 1.8% in 2027 (Chart 11) as the effects of past large shocks on services inflation continue to fade. The slight easing in HICPX inflation observed in the first half of 2025 is expected to continue into the first half 2026, with the rate averaging 1.8% over the rest of the projection horizon. The easing is mainly driven by services inflation and largely reflects receding labour cost pressures, while the impact from past supply chain shocks and energy price shocks has broadly unwound. Non-energy industrial goods inflation is expected to be initially dampened by the appreciation of the euro before picking up slightly in 2027 towards its historical average of around 1%, counterbalancing the slight moderation in services inflation and implying a broad sideways movement in HICPX inflation throughout the latter part of the projection horizon.

Chart 11

Euro area HICP inflation excluding energy and food

(annual percentage changes)

Notes: The vertical line indicates the start of the current projection horizon. The ranges shown around the central projections provide a measure of the degree of uncertainty and are symmetric by construction. They are based on past projection errors, after adjustment for outliers. The bands, from darkest to lightest, depict the 30%, 60% and 90% probabilities that the outcome of HICPX inflation will fall within the respective intervals. For more information, see the box entitled “Illustrating the uncertainty surrounding the projections” in the March 2023 ECB staff macroeconomic projections for the euro area.

Compared with the June 2025 projections, the outlook for headline HICP inflation has been revised up by 0.1 percentage points for both 2025 and 2026, but down by 0.1 percentage points for 2027 (Chart 12). The upward revision for 2025 reflects higher rates of energy and food inflation, largely due to stronger than expected data and higher oil and electricity price assumptions, as well as lagged impacts from past increases in international food commodity prices. For 2026, energy inflation is also slightly higher due to a softer dampening impact from the administered electricity price-related measure in Germany. HICPX inflation is unrevised for 2025 and 2026, with small offsetting revisions to the services and non-energy industrial goods components, but it has been revised down by 0.1 percentage points for 2027, reflecting mainly the impact of a stronger euro. Combined with slightly lower energy inflation, these revisions imply a lower headline inflation projection for 2027.

Chart 12

Revisions to the inflation projection compared with the June 2025 projections

(percentage points)

Notes: The revisions are calculated based on unrounded figures.

Nominal wage growth is projected to decline in 2025 and 2026 as pressure from inflation compensation fades, and to move sideways in 2027, at above average levels, supported by a tight labour market. Growth in compensation per employee is estimated to have declined in the second quarter of 2025, but to have been above the level foreseen in the June projections. It is projected to decline further, moderating from an average of 3.4% in 2025 to 2.7% in both 2026 and 2027 (this compares with a long-term average of 2.5%) (Chart 13). This profile reflects an initially lower contribution from social security contributions and wage drift, while there is seen to be only a very slight further decrease in negotiated wage growth (corroborated by the ECB wage tracker). Real wages are estimated to have returned to levels observed before the inflation surge, but growth in real wages is expected to remain somewhat above productivity growth for most of the projection horizon in the context of a still tight labour market. The wage share is expected to move broadly sideways over the horizon, remaining slightly above its long-term average. Compared with the June 2025 projections, growth in compensation per employee has been revised up by 0.2 percentage points for 2025, owing mainly to data surprises for the first half of 2025, whereas for the rest of the horizon the outlook is 0.1 percentage points lower and in line with a slight easing in labour market tightness.

Chart 13

Outlook for wage developments

(annual percentage changes)

Notes: The vertical line indicates the start of the current projection horizon. The data shown refer to compensation per employee.

Growth in unit labour costs is projected to decline further over the horizon on account of falling wage growth and rising productivity growth. Unit labour cost growth is estimated to have continued to ease in the first half of 2025. A drop in growth in compensation per employee implies a further decline in unit labour cost growth in the second half of 2025, which is seen to average 2.2% in 2026 and 1.9% in 2027 (broadly in line with the historical average). Compared with the June 2025 projections, growth in unit labour costs is unchanged for 2025, but for 2026 it has been revised up by 0.1 percentage points owing to lower productivity growth. For 2027, the downward revision of 0.1 percentage points is due to lower wage growth.

Domestic price pressures, as measured by the growth of the GDP deflator, are projected to continue decreasing in 2025 and to level off as of 2026, when declining unit labour cost pressures allow profit margins to increase (Chart 14). Amid some volatility in the course of 2025, the growth rate of the GDP deflator is projected to decline from an average of 2.3% in 2025 to 2.1% in 2026 and 2.0% in 2027. As growth in unit labour costs moderates, unit profit growth is expected to continue to expand over the projection horizon, aided by the economic recovery, strengthening productivity growth and a temporary accounting boost in 2027 related to the statistical treatment of ETS2.[11] Compared with the June 2025 projections, GDP deflator growth is unrevised for 2025 but has been revised up by 0.1 percentage points for 2026 and down by 0.1 percentage points for 2027.

Following declines in import prices until the end of 2025 owing mainly to the appreciation of the euro, import price inflation is expected to pick up as the dampening effects of the past appreciation fade. In annual terms, growth in the import deflator is expected to increase from 0.1% in 2025 to 0.6% in 2026 and more strongly to 1.7% in 2027 (Chart 14). The downward revisions to import price inflation over the entire projection horizon compared with the June projections initially reflect the stronger euro and lower assumptions for international food commodity prices and euro area farm gate prices. However, later in the horizon they relate mainly to the flatter profile for energy commodity price assumptions. The outlook for import price growth remains subject to elevated uncertainty on both the downside (e.g. lower Chinese export prices) and the upside (e.g. supply shortages). See Section 6 for sensitivity analyses on these topics.

Chart 14

Domestic price pressures

(annual percentage changes, percentage point contributions)

Note: The vertical line indicates the start of the current projection horizon.

6 Sensitivity analyses

6.1 Alternative energy price paths

Alternative paths for oil and gas commodity prices suggest some upside risks for inflation in 2026 and 2027. The staff projections are based on the technical assumptions outlined in Box 1. In this sensitivity analysis, alternative downside and upside paths are derived from the 25th and 75th percentiles of the options-implied neutral densities for both oil and gas prices.[12] According to these densities, risks to oil prices are broadly symmetrical around the baseline. Upside risks, such as additional sanctions on Russian and Iranian oil or a potential escalation of tensions in the Middle East, are balanced by downside risks stemming from the potentially persistent economic impact of the initial US tariff announcements or from larger than expected supply increases by OPEC+. By contrast, the gas price distributions indicate slight upside risks to the technical assumptions (Chart 15), likely reflecting supply uncertainties. These uncertainties are associated with disruptions in the global liquefied natural gas (LNG) market, potentially linked to strong demand from Asia for LNG, and with risks related to the European market. Notably, doubts have been raised about the feasibility of the European Commission’s plan to phase out all Russian gas imports by 2027, contributing to concerns that could put upward pressure on gas prices. Still, downside risks are present, particularly if trade-related uncertainty has a lasting impact, or if there is meaningful progress towards resolving Russia’s war against Ukraine, which could in turn lead to a relaxation of US sanctions on Russian LNG facilities. A constant price sensitivity analysis is also carried out for both oil and gas prices. In each case, a synthetic energy price index (a weighted average of the oil and gas price paths) is computed, and the impacts are assessed using ECB and Eurosystem macroeconomic models. The average results across these models are shown in Table 6.

Chart 15

Alternative paths for energy price assumptions

a) Oil price assumption

(USD/barrel)

b) Gas price assumption

(EUR/MWh)

Sources: Morningstar and ECB calculations.
Note: The option-implied densities for gas and oil prices are extracted from 15 August 2025 market quotes for options on ICE Brent crude oil and Dutch TTF natural gas futures with fixed quarterly expiry dates.

Table 6

Alternative energy price paths and their impact on real GDP growth and HICP inflation

Path 1: 25th percentile

Path 2: 75th percentile

Path 3: constant prices

2025

2026

2027

2025

2026

2027

2025

2026

2027

(deviations from baseline levels, percentages)

Oil prices

-4.6

-16.6

-18.2

4.3

16.9

19.0

1.3

5.8

5.8

Gas prices

-6.3

-19.3

-23.9

7.0

24.3

30.5

-0.9

-0.7

9.8

Synthetic energy price index

-4.8

-17.1

-20.1

5.1

18.1

21.6

0.3

3.0

7.1

(deviations from baseline growth rates, percentage points)

Real GDP growth

0.0

0.1

0.1

0.0

-0.1

-0.1

0.0

0.0

0.0

HICP inflation

-0.1

-0.6

-0.4

0.1

0.6

0.4

0.0

0.1

0.2

Notes: In this sensitivity analysis, a synthetic energy price index that combines oil and gas futures prices is used. The 25th and 75th percentiles refer to the option-implied neutral densities for the oil and gas prices on 15 August 2025. The constant oil and gas prices take the respective value as at the same date. The macroeconomic impacts are reported as averages of a number of ECB and Eurosystem staff macroeconomic models.

6.2 Alternative exchange rate paths

This sensitivity analysis assesses implications of alternative paths for the exchange rate, which suggest the possibility of further upward pressure on the euro and hence indicate downside risks to growth and inflation. The technical assumptions for exchange rates in the baseline projections are held constant over the projection horizon. Alternative downside and upside paths are derived from the 25th and 75th percentiles of option-implied neutral densities for the USD/EUR exchange rate on 15 August 2025, which was tilted towards an appreciation of the euro (Chart 16). The impacts of these alternative paths are assessed using ECB and Eurosystem staff macroeconomic models. The average impact on output growth and inflation across these models is shown in Table 7.

Chart 16

Alternative paths for the USD/EUR exchange rate

Sources: Bloomberg and ECB staff calculations.
Notes: An increase implies an appreciation of the euro. The 25th and 75th percentiles refer to the option-implied neutral densities for the USD/EUR exchange rate on 15 August 2025. The macroeconomic impacts are reported as averages of a number of ECB and Eurosystem staff macroeconomic models.

Table 7

Impact on real GDP growth and HICP inflation

Path 1: 25th percentile

Path 2: 75th percentile

2025

2026

2027

2025

2026

2027

USD/EUR exchange rate

1.12

1.14

1.13

1.14

1.24

1.28

USD/EUR exchange rate (% deviation from baseline)

-0.3

-2.0

-2.8

1.1

6.5

9.8

(deviations from baseline growth rates, percentage points)

Real GDP growth

0.0

0.1

0.1

0.0

-0.2

-0.2

HICP inflation

0.0

0.1

0.1

0.0

-0.2

-0.3

Sources: Bloomberg and ECB staff calculations.
Notes: An increase implies an appreciation of the euro. The 25th and 75th percentiles refer to the option-implied neutral densities for the USD/EUR exchange rate on 15 August 2025. The macroeconomic impacts are reported as averages of a number of ECB and Eurosystem staff macroeconomic models.

6.3 Sensitivity analyses on the impact of US tariff policies

Amid ongoing uncertainty in the global trade environment, this section examines selected risks associated with US tariff policies. Despite the recent trade deals agreed between the United States and the European Union, and between the United States and some of its other trading partners, there is still considerable uncertainty about how sectors and companies will adjust to this new global trade environment. This section focuses on three potential risks. First, there could be stronger trade diversion owing to the reallocation of trade as US tariffs differ across countries. Second, Chinese export prices could be lowered as a strategic response to the tariffs. Third, there could be more pronounced bottlenecks in global supply chains as a result of the tariffs. The baseline projections only include limited effects related to these factors.

Trade reallocation could be stronger than in the baseline and Chinese export prices could be lower. The ECB-Global model implies that, under the tariff assumptions entailed in the baseline (as described in Box 1), trade diversion away from the United States and towards the euro area is limited.[13] This means that there is only weak reallocation of trade by US importers away from countries subject to higher tariffs, such as China, towards other countries, and that there is a limited reduction in China’s export prices to offset the tariffs. Two alternative scenarios around this baseline are analysed. In the first, US importers shift demand away from China and towards the euro area and the rest of the world. In the second, Chinese exporters respond to higher US trade barriers by reducing export prices towards all their trading partners to fully offset the effects of such barriers on their total exports.

Chart 17

Impact on exports of stronger trade diversion and lower Chinese export prices

a) Stronger trade diversion

b) Lower Chinese export prices

(percentage deviation from baseline in the fourth quarter of 2027)

(percentage deviation from baseline in the fourth quarter of 2027)

Sources: ECB-Global model and ECB staff calculations.
Notes: The chart shows changes in bilateral trade in the fourth quarter of 2027 (unweighted). China has a partially managed exchange rate. The rest of the world is computed as the simple average of deviations of exports from the baseline for all regions in the ECB-Global model excluding the United States, China and the euro area.

Stronger global trade diversion, with greater substitution by US importers between source countries, significantly affects the US economy but has only a small impact on the euro area (Chart 17, panel a).[14] Simulations using the ECB-Global model show that this shift would induce higher US import prices, which are passed through to domestic consumer and producer prices, and, at the peak, US inflation would be 0.2 percentage points higher in 2026 compared with the baseline. Interest rates in the United States would be higher, which would have a negative effect on consumption and investment and, as a result, US GDP growth would be 0.2 percentage points lower in 2026 compared with the baseline. The impact on China would be smaller. For the euro area, the simulations using the ECB-BASE model suggest that higher euro area exports to the United States would be offset by weaker exports to China and the rest of the world. As a result, euro area GDP growth and HICP inflation would be around 0.05 percentage points lower in 2026 and 2027.

A reduction in Chinese export prices, aimed at offsetting the impact of US tariffs, would significantly affect the euro area.[15] By strategically reducing export prices, China could completely offset the decline in its overall exports triggered by the US tariffs. This scenario would imply an initial reduction in Chinese export prices of nearly 8%. The reduction in export prices would boost its competitiveness and increase its exports to other regions, thereby offsetting part of the decline in trade with the United States. Chinese exports to the United States would then recover and increase by around 10% compared with the baseline by the fourth quarter of 2027 (Chart 17, panel b). This would imply that, overall, Chinese exports to the United States would be 13% lower cumulatively by 2027, compared with a 23% reduction in the baseline simulation. As China’s competitiveness increased, the euro area would face trade losses, with a decline in its exports to the rest of the world. However, euro area exports to the United States would rise owing to an effective appreciation of the US dollar.[16] This appreciation of the US dollar would also reduce oil demand, leading to a 1.4% drop in oil prices by 2027. An increase in imports from China would contribute to a reduction in euro area GDP growth of around 0.2 percentage points in 2026, with an accumulated reduction in HICP inflation of almost 0.3 percentage points over the projection horizon (Chart 18).

Chart 18

Macroeconomic implications for the euro area of stronger trade diversion and lower Chinese export prices

a) Real GDP growth

b) HICP inflation

(deviation from the baseline in percentage points)

(deviation from the baseline in percentage points)

Sources: ECB-Global and ECB-BASE models, and ECB staff calculations.
Note: In these sensitivity analyses the inputs on the international environment are compiled with the ECB-Global model and are run through the ECB-BASE model to compute the impact on the euro area.

Supply bottlenecks may arise as new tariffs, trade distortions and heightened uncertainty may prompt companies to restructure their supply chains, leading to some reduction in production effectiveness – at least temporarily. There are currently no signs of any significant bottlenecks in global supply chains, and the US-EU trade deal agreed recently has reduced the likelihood of the euro area being affected by such disruptions. As a result, the baseline projections do not assume any significant supply bottlenecks. However, given the scale of the US tariffs applied to some countries that play a key role in global supply chains, as well as the possibility of a further escalation of the trade conflict, the emergence of such disruptions cannot be ruled out. To assess this risk, we construct two alternatives paths around the baseline path of the Global Supply Chain Pressure Index (GSCPI)[17], using a large structural Bayesian vector autoregression (BVAR) model conditional on the September 2025 projections (Chart 19).[18] With the baseline path, the GSCPI would be around its current level over the projection horizon, which implies limited effects on euro area activity. A first alternative path is created by applying global supply chain shocks during the first Trump Administration (yellow bars in Chart 19), as identified by the BVAR model, to the baseline path. A second alternative path models severe supply bottlenecks, mirroring shocks observed during the COVID-19 pandemic (red bars in Chart 19). This is an extreme assumption as the likelihood of such a shock is currently assessed as very low.

Chart 19

GSCPI shock decomposition, baseline path and alternative paths

(index, contributions)

Sources: Bańbura et al. (2023) and ECB staff calculations.
Notes: The green bars represent the combined impact of other shocks identified in the model and variable-specific idiosyncratic components, including energy shocks which played an important role in explaining the increase in the Global Supply Chain Pressure Index (GSCPI) during the pandemic period.

The macroeconomic effects on the euro area are minimal under the first alternative path, but sizeable under the second alternative path with more extreme disruptions to supply chains, as during the pandemic. Focusing first on the results with an economy with fully functioning production networks, and using the ECB-BASE model, under the path with supply bottlenecks similar to those seen during the first Trump Administration, only minimal inflationary pressures would arise, with HICP inflation being less than 0.1 percentage points higher in 2027 (Chart 20, panel b, blue bars). The scenario with shocks similar to those that occurred during the pandemic period would lead to more pronounced inflation effects, with HICP inflation being up to 0.15 percentage points higher in 2027 (Chart 20, panel b, blue bars). In both sensitivity analyses, the impact on real GDP is negligible or slightly positive, reflecting the assumption of unchanged interest rates which, given higher inflation, implies a reduction in real interest rates that would boost activity (Chart 20, panel a, blue bars). The analysis is extended using a multi-country multi-sector model which takes account of input-output linkages, with production networks in other regions being hampered in addition to the supply chain blockages. This assumption reduces substitutability within production chains and would result in more pronounced effects (Chart 20, yellow bars). In this case, compared with the baseline, inflation in 2027 could be up to 0.1 percentage points higher in the first Trump Administration scenario and almost 0.25 percentage points higher in the more extreme pandemic scenario, and GDP growth would be slightly dampened.[19]

Chart 20

Impact of alternative paths of supply chain bottlenecks on output and inflation and the role of network effects

a) Real GDP growth

(percentage points)

b) HICP inflation

(percentage points)

Sources: The ECB-BASE model and the multi-country multi-sector model by Aguilar, P. et al. (2025), and ECB staff calculations.
Notes: The yellow areas show the overall impact of trade bottlenecks including the effects of hampered production networks. These effects are estimated using the multi-country multi-sector model by comparing results from a model without intermediate goods input-output substitution in the production of goods with those from a model that incorporates them. The blue bars show the ECB-BASE model results where the trade network effects are implicitly captured by the model.

Box 2
Comparison with forecasts by other institutions and the private sector

The September 2025 ECB staff projections are within the range of other forecasts in 2025, but at the bottom or below the range for both growth and inflation in 2026 and 2027. The staff projection for growth lies at the top of the range of forecasts from other institutions and surveys of private sector forecasters for 2025 and below the range for 2026 and 2027. For inflation, the ECB staff projection is in line with other forecasts for 2025, lies at the bottom of the range for 2026 and is below the other available forecasts for 2027. The ECB staff projections for HICPX inflation are within the range of other available forecasts for 2025-26 and below the other available forecast for 2027.

Table

Comparison of recent forecasts for euro area real GDP growth, HICP inflation and HICP inflation excluding energy and food

(annual percentage changes)

 

Date of release

Real GDP growth

HICP inflation

HICP inflation excl. energy and food

2025

2026

2027

2025

2026

2027

2025

2026

2027

ECB staff projections

September 2025

1.2

1.0

1.3

2.1

1.7

1.9

2.4

1.9

1.8

Consensus Economics

August 2025

1.2

1.1

1.5

2.1

1.8

2.0

2.3

1.9

-

Survey of Professional Forecasters

July 2025

1.1

1.1

1.4

2.0

1.8

2.0

2.3

2.0

2.0

International Monetary Fund

July 2025

1.0

1.2

1.4

2.0

1.8

2.1

-

-

-

OECD

June 2025

1.0

1.2

-

2.2

2.0

-

2.2

2.0

-

European Commission

May 2025

0.9

1.4

-

2.1

1.7

-

2.4

1.9

-

Sources: Consensus Economics Forecasts, 14 August 2025 (data for 2027 are taken from the July 2025 survey); ECB Survey of Professional Forecasters, 25 July 2025; IMF World Economic Outlook, 23 July 2025; OECD Economic Outlook, 3 June 2025; and European Commission Spring 2025 Economic Forecast, 19 May 2025.
Notes: These forecasts are not directly comparable with one another or with the ECB staff macroeconomic projections, as they were finalised at different points in time. Additionally, they use different methods to derive assumptions for fiscal, financial and external variables, including oil, gas and other commodity prices. The ECB staff macroeconomic projections report working day-adjusted annual growth rates for real GDP, whereas the European Commission and the International Monetary Fund report annual growth rates that are not adjusted for the number of working days per annum. Other forecasts do not specify whether they report working day-adjusted or non-working day-adjusted data.

© European Central Bank, 2025

Postal address 60640 Frankfurt am Main, Germany
Telephone +49 69 1344 0
Website www.ecb.europa.eu

All rights reserved. Reproduction for educational and non-commercial purposes is permitted provided that the source is acknowledged.

For specific terminology please refer to the ECB glossary (available in English only).

PDF ISBN 978-92-899-7313-7, ISSN 2529-4466, doi:10.2866/9477283, QB-01-25-148-EN-N
HTML ISBN 978-92-899-7312-0, ISSN 2529-4466, doi:10.2866/7028423, QB-01-25-148-EN-Q


  1. The cut-off date for the technical assumptions was 15 August 2025. The macroeconomic projections for the international environment and the euro area were finalised on 28 August 2025.

  2. Unless explicitly mentioned, references to world and/or global aggregates of economic indicators throughout this section exclude the euro area.

  3. Global headline CPI inflation is computed as the weighted average of inflation rates across 24 countries: 15 advanced economies (United States, United Kingdom, Japan, Switzerland, Canada, Australia, New Zealand, Sweden, Denmark, Norway, Czech Republic, Hungary, Poland, Romania and Bulgaria) and nine emerging market economies (China, Russia, Brazil, India, Türkiye, Korea, Mexico, Singapore and Hong Kong).

  4. On 28 August the European Commission initiated the legislative process to introduce necessary legislation to remove all tariffs on US industrial goods and to provide preferential market access for some US seafood and agricultural food. According to the US-EU joint statement, this condition is sufficient to ensure a tariff rate of 15% on cars and car components.

  5. This is based on the “unadjusted” measure of trade policy uncertainty that is “cleaned” to remove the estimated impact of media attention to uncertainty unrelated to trade policy, effective tariff rates, financial conditions and supply chain pressures. The assumptions shown in Chart B of Box 2, entitled “US tariffs and trade policy uncertainty”, in the June 2025 projections referred to the unadjusted measure of trade policy uncertainty, whereas Chart B, panel a) in Box 1 shows the “cleaned” measure.

  6. The fiscal stance of the euro area is defined as the change in the cyclically adjusted primary balance, further adjusted for the NGEU grants on the revenue side. While the fiscal stance is a top-down measure of the orientation of fiscal policy, discretionary fiscal measures are gauged using a bottom-up approach. These measures capture changes in tax rates, fiscal entitlements and other government spending that have been passed or are likely to be passed by the national parliaments of euro area countries.

  7. One source of tightening in 2025, particularly in respect of indirect taxes, is the further scaling-down of the remaining energy support measures since 2022.

  8. All main spending categories continue to grow at robust rates, with some slowdown in government consumption after the high growth rates in 2024. The loosening of non-discretionary factors in 2025 is mainly due to revenue shortfalls and other tax residuals, while composition effects, which were strongly positive in 2024 (tax bases growing faster than nominal GDP with a tightening effect on the fiscal stance), are estimated to be broadly neutral both in 2025 and over the rest of the projection horizon.

  9. The NGEU grants (with no macro impact on the revenue side of the budget) are estimated at a level close to 0.5% of GDP for 2026 and only slightly above zero for 2027. The projected tightening of NGEU-funded government spending for 2027 in the case of the large NGEU beneficiaries is partly offset by high projected public investment growth in Germany.

  10. The impact of the new ETS2 on HICP inflation is uncertain and depends on the price of ETS2 permits, the speed and extent of pass-through to consumer prices, and the modalities of the transition to ETS2 in countries with existing national schemes. For a more detailed discussion on these uncertainties, see the box entitled “Assessing the impact of climate change transition policies on growth and inflation” in the December 2024 Eurosystem staff projections report.

  11. The ETS2 scheme is expected to be recorded in the national accounts as a tax on production when the emission permits are surrendered, i.e. the year after the auctioning. The ETS2 revenues for 2027 will thus only be recorded in the national accounts in 2028, while final prices are expected to be increased already in 2027. The gross operating surplus is therefore expected to be temporarily boosted in 2027.

  12. The market prices used are those prevailing on 15 August 2025 (the cut-off date for the technical assumptions).

  13. In the baseline simulation of US tariffs, US imports from China and the euro area decline cumulatively by 23.5% and 3.9% respectively by the end of 2027. Moreover, US exports to China and the euro area drop cumulatively by 4.9% and 5.4% respectively over the same period, whereas China’s exports to the euro area increase by 0.7%.

  14. In the ECB-Global model, trade diversion operates under the assumption that the prices of international competitors have an outsized influence on import patterns, beyond their proportional weight in the domestic consumer price index. As a result, exports from country A to country B are shaped not only by the relative prices of country A’s goods compared with country B’s overall consumer price level, but also by how these prices compare with the import prices of other international competitors supplying country B.

  15. In the analysis, the euro exchange rate is assumed to remain constant against other currencies. This contributes to fuelling imported deflationary pressures from China to the euro area.

  16. While the nominal effective exchange rate of the euro is assumed to remain constant in line with the technical assumptions, the euro depreciates against the dollar, compensated by appreciations against other currencies. This gives a competitive edge to euro area exporters in the US market.

  17. For details, see https://www.newyorkfed.org/research/policy/gscpi#/overview.

  18. The large structural BVAR model is conditional on the technical assumptions for oil prices in euro, gas prices, real GDP, the exchange rate and euro area farm gate prices, and on the projections of HICP, HICPX, HICP services, compensation per employee, real GDP and world GDP. The model also includes oil production, producer prices and their breakdown into energy and intermediate goods, the output PMI, PMI delivery times, and the GSCPI. The model identifies eight shocks related to global supply chains, oil supply, oil-specific demand, gas commodity price, domestic supply, the labour market, domestic demand and foreign demand. See Bańbura, M., Bobeica, E. and Martínez Hernández, C., “What drives core inflation? The role of supply shocks”, ECB Working Paper Series, No 2875, 2023.

  19. See Aguilar, P., Domiguez-Díaz, R., Gallegos, J. and Quintana J. (forthcoming, 2025), “The Transmission of Foreign Shocks in a Networked Economy”.

Annexes
11 September 2025