International Edition

Thursday, 8 October 2026

Private Trade News

Global markets, trading & world business — for professional traders

Policy

Meeting of 9-10 September 2026

Private Trade News venue-risk note (2026-10-08): Account of the monetary policy meeting of the Governing Council of the European Central Bank held in Berlin on Wednesday and Thursday, 9-10 September 2026 1.… Primary source: original at ECB (ecb.europa.eu).

· ECB

Account of the monetary policy meeting of the Governing Council of the European Central Bank held in Berlin on Wednesday and Thursday, 9-10 September 2026

1. Review of financial, economic and monetary developments and policy options

Financial market developments

Ms Schnabel started her presentation with the observation that, since the Governing Council’s monetary policy meeting on 22-23 July 2026, euro area financial markets had continued to be driven by energy price developments, with market attention turning to the global rise in long-term yields amid a resilient euro area macroeconomy.

Persistently high energy prices, reinforced by concerns about higher food price inflation, had pushed market expectations regarding the outlook for inflation and policy rates higher. Energy prices and futures curves had moved up owing to renewed US-Iran hostilities. In Europe, the focus had shifted from oil to refined products and gas. The price and futures curve of gas oil – a refined, diesel-type product – had moved up significantly, reflecting concerns about limited refining capacity. Similarly, European gas prices had risen to the highest level recorded since early 2023, and low European gas storage levels could put further upward pressure on gas prices. The recent heatwave in Europe and what could likely be a strong “El Niño” event had also added to concerns about upward pressure on prices, with international food prices rising significantly over recent months.

The market-based inflation outlook continued to be driven primarily by energy price developments. Over time, markets had reassessed the persistence of the inflation shock. Following the outbreak of the conflict in the Middle East, markets had quickly priced in a substantial near-term inflation shock, with the December 2026 inflation fixings jumping sharply. The June 2027 inflation fixings had responded less strongly initially but had risen steadily since mid-April as market participants increasingly viewed the conflict as a persistent source of disruption. As a result, markets currently expected inflationary pressures to extend well beyond the initial energy price impulse. Looking further ahead, the December 2027 inflation fixings had so far increased only modestly, suggesting a decline in inflation during 2027, in line with the staff projections, while remaining above 2%.

This reassessment had been reflected in expectations for the ECB’s terminal rate, which had risen above 3% for the first time during this hiking cycle. Markets were fully pricing in a rate hike at the present meeting and had priced in 84 basis points of hikes overall by the end of 2027, compared with 64 basis points at the time of the July meeting. Respondents to the Survey of Monetary Analysts took a more benign view, with the median expectation suggesting only a final rate hike at the September meeting. In the United States, monetary policy expectations also pointed to a somewhat higher terminal rate, with markets currently pricing in between one and two rate hikes in 2026.

Market discussion had focused on the global rise in long-term yields, which had originated in the United States where high private and public issuance, concerns about the fiscal trajectory and uncertainty about inflation had pushed yields higher. Since the Governing Council’s July monetary policy meeting, the nominal overnight index swap (OIS) yield curve in the euro area had shifted higher in a broadly parallel manner. The move in the front end had been primarily driven by higher inflation compensation and higher expected rates, reflecting the expected monetary policy response to the inflation shock. The longer end – the five-year rate in five years’ time – had been driven by higher real term premia, which likely reflected more duration to be absorbed by markets due to increased debt issuance, as well as heightened uncertainty about the fiscal outlook.

Taking a longer-term perspective of the euro area ten-year OIS rate and its components showed that long-term nominal rates, real rates, rate expectations and term premia had all traded in a relatively narrow range since the fourth quarter of 2022 and were not far above their four-year averages. However, since a trough in late 2024, all components had gradually increased. The bulk of the increase had come after two distinct events: the announcement of the German fiscal package in March 2025 and the start of the Middle East conflict in February 2026. A decomposition of the euro area ten-year nominal OIS rates into structural drivers showed that domestic factors – an improving euro area macroeconomic environment and a reassessment of ECB monetary policy expectations – had been the dominant drivers. US factors seemed to have played some role recently, but spillovers from US to euro area yields had generally been more modest over the past two years. Whereas US term premium shocks had previously explained the largest part of the variation in euro area term premia, this share was estimated to have fallen to 20%. One possible explanation for the structurally weaker spillovers was that upward movements in US term premia could increasingly reflect US-specific factors.

Euro area sovereign bond markets had remained orderly. Sovereign bond spreads over OIS rates had been broadly stable since the Governing Council's previous monetary policy meeting. This suggested that the recent increase in euro area yields had not triggered a broader reassessment of sovereign credit risk, while markets continued to differentiate among issuers.

Regarding exchange rates, the coordinated intervention by the US and Japanese authorities to support the yen had led to an initial sharp appreciation of the Japanese currency against the US dollar, while the euro had appreciated against the US dollar.

On the back of continued positive data surprises on the macroeconomy, strong corporate earnings growth expectations and sustained optimism about artificial intelligence (AI), euro area equity markets had risen further. At the same time, concerns were rising about the increased leverage of AI companies, a significant part of which was off-balance sheet. While direct euro area exposures to AI-related leverage remained limited, investors were increasingly exposed to AI-related valuations and earnings expectations through equity markets. An important question was whether the AI investment boom, and the increase in debt in particular, could generate crowding-out effects in euro area bond markets. So far, euro area markets had absorbed the greater issuance by US hyperscalers without there being a noticeable deterioration in financing conditions for other issuers.

Ms Schnabel concluded by pointing out that euro area financial conditions had tightened modestly since the Governing Council's previous monetary policy meeting. This had been driven by somewhat higher long-term nominal and real rates, as well as an appreciation of the euro, which had been only partly offset by the easing stemming from risk assets.

The global environment and economic and monetary developments in the euro area

Mr Lane then went through the latest economic, monetary and financial developments in the global economy and the euro area. He noted that, six months on, the energy shock triggered by the conflict in the Middle East was generating persistent inflation pressures. After some volatility during the summer, there was renewed upward pressure on oil prices, compounded by an upward shift in refining margins. In addition, spot and forward prices for gas had surged. Oil prices stood at USD 97 per barrel – up 3% since the Governing Council’s July monetary policy meeting – and European diesel crack spreads (i.e. the differences between wholesale petroleum product prices and crude oil prices, often used to estimate refining margins) were above USD 70 per barrel – more than three times their pre-war level. Gas prices stood at €73 per MWh – 17% higher than at the time of the July meeting.

Headline inflation in the euro area, as measured by the Harmonised Index of Consumer Prices (HICP), had increased to 3.3% in August, from 2.9% in July, according to Eurostat’s flash estimate. Similarly, energy price inflation had risen to 14.3%, after 10.3% in July. This increase was likely to have reflected a strong contribution from refining margins on liquid fuels, as well as higher energy commodity prices. According to the flash estimate, food price inflation had remained unchanged at 1.2%.

So far, indirect effects remained contained, with no material signs of second-round effects. Compensation per employee had grown at an annual rate of 3.3% in the second quarter, down from 3.5% in the first quarter. Rising labour productivity had also helped contain growth in unit labour costs, which had slowed to 2.6%, from 3.5% in the first quarter. At the same time, the more limited growth in unit labour costs had been offset by stronger growth in unit profits, which had risen from 0.3% to 2.2%. Profit margin indicators based on Purchasing Managers’ Indices (PMIs) had remained squeezed in the first two months of the third quarter, although less so than in the second quarter. Negotiated wage growth had continued to ease in the second quarter, to 2.4%. This outcome was broadly in line with the ECB’s wage tracker. This suggested a broadly unchanged outlook for wages since the Governing Council’s previous meeting and pointed to a modest uptick, to 2.7%, in negotiated wage growth in the first half of 2027. Most measures of underlying inflation had been broadly stable in July. At the same time, the lagged effects of the energy shock were expected to continue to exert some pressure on non-energy inflation over the coming year.

The conflict in the Middle East and recent developments in Russia’s unjustified war against Ukraine had pushed the path of energy prices up further. This was likely to keep headline inflation well above target into the first half of 2027. Thereafter, energy inflation should decline and turn negative up to mid-2028, bringing headline inflation down. The baseline of the September ECB staff projections saw headline inflation averaging 3.0% in 2026, 2.5% in 2027 and 2.1% in 2028. Compared with the June staff projections, headline inflation had been revised up by 0.2 percentage points for 2027 and by 0.1 percentage points for 2028. The projected headline inflation rate for 2028 would have been 1.9% without the impact of the EU Emissions Trading System 2. Higher energy prices were expected to feed through gradually to core and food price inflation. The improved economic outlook should also contribute to slightly higher core inflation, which was expected to keep rising until early 2027 and stay elevated for the rest of the year, before moderating in 2028. For inflation excluding energy and food, the baseline foresaw 2.5% in 2026, 2.6% in 2027 and 2.3% in 2028.

Inflation expectations over shorter horizons remained at elevated levels, but most measures of longer-term inflation expectations stood at around 2%, supporting the stabilisation of inflation around target in the medium term.

The global economy was expected to strengthen in the near term following a deceleration in the second quarter. Survey indicators suggested a pick-up in global growth momentum in the third quarter, with AI-related investment continuing to be a key driver. Strong trade dynamics in technology products would also continue to support growth in global trade in the near term. Energy supply shortages had eased in recent months, although the PMI scores indicated increasing global supply shortages for semiconductors. Since the previous meeting of the Governing Council, the euro had appreciated by 1.9% against the US dollar and by 0.7% in nominal effective terms.

The euro area economy had proved to be resilient in the second quarter, despite headwinds from the energy shock. Concerns at the start of the Middle East conflict regarding shortages of fertilisers and supply bottlenecks had largely not materialised. Growth was broad-based across countries and sectors. This pattern was likely to have continued into the third quarter. Survey indicators had improved further in the third quarter, with the August PMIs for flash composite output and new orders having edged up compared with July. Manufacturing continued to perform solidly as governments spent more on defence and infrastructure. Consumer confidence had rebounded from low levels, helping services recover from the initial energy shock. Increased AI-related activity was visible in digital services, business investment and exports.

The labour market had remained robust, with the unemployment rate unchanged in July at 6.4%. Growth in employment and the labour force continued to slow, while productivity had gradually picked up. Labour demand had cooled further, as the job vacancy rate had edged down by 0.1 percentage points to 2.1% in the second quarter, and Indeed job postings had weakened, albeit less markedly than in previous months.

With respect to fiscal policies, the September staff projections were broadly unchanged from the June projections, foreseeing the euro area budget deficit to increase from 3.0% of GDP in 2025 to 3.6% in 2026 and 3.7% in 2027, before moderating again slightly to 3.6% in 2028. The projection for the fiscal stance, defined as the change in the cyclically adjusted primary balance, indicated a substantial fiscal loosening by 0.5 percentage points in 2026, which would support the economy. This loosening would, however, be reversed in 2027 and 2028, constituting a headwind to growth in those years.

Looking ahead, the near-term growth outlook had improved compared with the last round of staff projections, reflecting, in particular, the resilience of private consumption and public spending. Over the medium term, consumption should be supported by gradually falling energy prices and a strong labour market. Growth would increasingly be bolstered by business and housing investment. Export growth should benefit from rising foreign demand but was being held back by competitiveness challenges and uncertainty about global trade policies. The September ECB staff baseline projection for economic growth was 0.9% for 2026, 1.4% for 2027 and 1.5% for 2028. This was an upward revision for both 2026 and 2027, mainly reflecting the greater than expected resilience of the euro area economy.

The outlook remained highly uncertain and critically dependent on geopolitical developments. Risks were to the upside for inflation and to the downside for economic growth. In relation to the energy shock, the updated scenarios put together by ECB staff illustrated the broad range of outcomes for how growth and inflation would evolve under different assumptions about its intensity and duration, as well as its indirect and second-round effects. The September projections baseline was complemented with a milder scenario, an adverse scenario and a severe scenario, calibrated using market‑implied probability distributions for commodity prices from the fourth quarter of 2026 onwards. Under the adverse and severe scenarios, inflation would remain above target over the entire projection horizon, while growth would be lower than in the baseline. Under the milder scenario, inflation would moderate more quickly than in the baseline and fall below target in the medium term, while growth would be slightly above the baseline for 2027 and 2028.

Market interest rates had increased since the Governing Council’s previous meeting, reflecting similar moves in global markets. Financing costs were adjusting smoothly to the increase in ECB interest rates in June: bank lending rates for firms had risen, to stand at 3.8% in June and July, from 3.6% in May. The cost of market-based corporate debt had stood at 4.0% in July, which was similar to previous months and well above its level before the conflict in the Middle East. The annual growth rate of bank lending to firms, which usually responded to changes in monetary policy with a longer delay, had increased further to 4.4% in July, from 4.0% in May and June. The annual growth rate of corporate bond issuance had been 3.4%, after 3.6% in June and 3.3% in May. Mortgage rates had been unchanged in June and July, at 3.5%, while mortgage lending growth had softened to 3.0% in July, from 3.1% in May and June.

Monetary policy considerations and policy options

On the basis of the incoming information and a comprehensive assessment of the inflation outlook and the risks surrounding it, as well as the dynamics of underlying inflation and the strength of monetary policy transmission, Mr Lane proposed that the Governing Council increase the three key ECB interest rates by 25 basis points. The conflict in the Middle East continued to generate inflation pressures, and the updated staff projections indicated that inflation was set to remain well above target for an extended period. Increasing the deposit facility rate from 2.25% to 2.50% was a robust decision across a wide range of scenarios. While measures of underlying inflation had shown little change so far, the full inflationary impact of the energy shock had yet to play out. Financing costs were adjusting smoothly to the increase in ECB interest rates in June, while the economy had shown resilience. The absence of financial stress and the solidity of household, corporate and bank balance sheets meant that a hike should be transmitted through the financial system in an orderly manner. This hike would ensure that the Governing Council remained well positioned to navigate the uncertainty caused by the conflict.

Looking ahead, the Governing Council should retain its data-dependent, meeting-by-meeting approach without a pre-commitment to any particular rate path.

2. Governing Council’s discussion and monetary policy decisions

Economic, monetary and financial analyses

Regarding the economic analysis, members broadly agreed with the assessment provided by Mr Lane in his introduction. The global economy was robust, although the narrative consisted of two countervailing forces. On the one hand there was the global AI investment boom, while on the other hand there was the global energy shock and elevated geopolitical uncertainty, especially in relation to the conflict in the Middle East and Russia’s unjustified war against Ukraine. Although the AI boom was clearly dominating economic developments in the United States, this was less true of the euro area. Higher energy prices created a concern about a rise in inflationary pressures in the global economy. Moreover, partly on account of the disruption to major shipping routes, global shipping rates were rising in several parts of the world, and this could put more pressure on prices.

Turning to commodity markets, developments since the Governing Council’s previous monetary policy meeting confirmed that the conflict in the Middle East and the resulting increase in energy prices had become more persistent. Several concerns were expressed in this regard. Natural gas prices had increased substantially in recent weeks and concern was expressed about low storage levels, especially in Europe. This could lead to pressure on prices during the coming winter, especially if it were to be unusually cold. Given the important role of gas in the production of electricity, it was argued that this could have a significant impact on electricity prices and subsequently spread to other consumer prices as well. Gas was also important for industry and for heating. As for crude oil, reserves in China and the United States had been used to buffer the increase in prices during the first few months of the conflict. However, these reserves had fallen recently, putting upward pressure on oil prices. Moreover, rising refining margins owing to the destruction of refining capacity in the Middle East and Russia had led to additional price increases for fuels, in particular diesel. More generally, recent developments in the Middle East and Russia had demonstrated that pressure on energy markets could remain high. In this context, it was argued that the conflict in the Middle East could turn out to be prolonged, given the opposing political objectives of the parties involved. Indeed, it appeared that the second wave of the crisis in the Middle East was now unfolding. Relative to the material decline in energy commodity prices over the course of next year as implied by futures curves and embedded in the staff projections, all of these factors created upside risks to energy prices, thereby also creating simultaneous upside risks to the inflation outlook and downside risks to the growth outlook.

At the same time, it was pointed out that the energy shock could be less persistent than assumed, as it was essentially a politically driven shock that could disappear as quickly as it had appeared. In particular, it was suggested that, if the conflict in the Middle East had created an adverse long-term impact on supply, the increase in oil and gas prices should have persisted independently of developments in political negotiations between Iran and the United States. Instead, energy commodity prices had been very volatile, closely reflecting these political developments. Higher energy prices had also led to lower energy consumption. In addition, gas was progressively being replaced by renewable energy in the production of electricity. This substitution had recently been much faster than expected, with renewables accounting for around 50% of gross electricity consumption in the EU at the end of 2025, up from a share of just below 40% in 2021. Industrial gas usage had also declined since 2021 and many companies were well hedged against a rise in gas prices. Together, these factors meant that any increase in gas prices would probably have a smaller adverse effect on the euro area than had been the case following the gas price shock in 2022.

Rising energy prices could also transmit to food prices. In addition, higher fertiliser prices precipitated by the energy shock, together with the impact of El Niño and recent heatwaves, could put upward pressure on food commodity prices. Against this background, appreciation was expressed for the analytical rigour with which the economic effects of the unfolding climate and nature crises had been integrated into staff analysis. Such analytical rigour would remain important in the future, as the increasing frequency and severity of climate and nature-related events, including heatwaves, wildfires and droughts, would have increasingly material economic and financial implications.

With regard to economic activity, members concurred with the assessment presented by Mr Lane. The economy had proved resilient in the second quarter, performing better than expected despite headwinds from the energy shock and ongoing geopolitical instability. Incoming data had surprised to the upside, with the Citigroup Economic Surprise Index standing at its highest level since early 2023 owing to the strength of both soft and hard indicators. Better than expected private consumption, robust government spending – driven by the fiscal expansion in Germany and funding under the Next Generation EU programme – and strong exports had supported the economy in the second quarter, while investment had remained sluggish. Growth had been broad-based across countries and sectors. This pattern was likely to have continued into the third quarter. Manufacturing continued to perform solidly as governments spent more on defence and infrastructure. Consumer confidence had rebounded from low levels, helping services recover from the initial energy shock. Increased AI-related activity was visible in digital services, business investment and exports.

Although the incoming data seemed to warrant cautious optimism about the growth outlook, it was important to assess whether the strong outcome in the second quarter was temporary or the beginning of a more lasting improvement in demand. It was still too early to draw a firm conclusion about the degree of resilience of the economy on the basis of the latest growth figures. While AI investment was picking up in the euro area, it was not the main source of the recent resilience, as private investment had surprised to the downside in the second quarter. The stronger than expected growth had instead been driven by private consumption. However, this was likely to have been mechanically supported by lower than expected inflation. Therefore, the recent worsening of the inflation outlook would probably imply a slowdown in consumption in the coming quarters owing to lower than expected purchasing power, unless wages were to adjust rapidly, which was unlikely given the current institutional approach to wage-setting in the euro area.

In this context, it was stressed that it was important to have a better understanding of whether the stronger than expected economic performance was due mainly to less negative supply or to positive demand factors. If it was due to stronger than expected demand, then this would generate additional inflationary risks. Conversely, positive supply developments would reduce inflationary risks. On the one hand, it was noted that the temporarily lower energy prices seen earlier in the summer constituted a partial unwinding of the previous negative supply shock, which could explain both higher than expected growth and lower than expected inflation. On the other hand, it was argued that it would be misleading to attribute recent economic fluctuations solely to supply shocks given that the economy was simultaneously experiencing positive demand shocks, especially from AI and fiscal policy, which also showed up in stock market prices. More generally, with respect to the multitude of shocks that would continue to affect the economy in the future, such as climate change, geopolitical shocks, changing trade restrictions, fiscal developments and AI, it was important to understand whether such shocks would ultimately be inflationary or deflationary.

Looking ahead, the near-term growth outlook had improved compared with the previous round of staff projections, mainly reflecting the greater than expected resilience of the euro area economy, in particular the resilience of private consumption and public spending. This also implied significant carry-over effects for 2027. Compared with June, the staff baseline projection for economic growth had been revised up for both 2026 and 2027, to 0.9% and 1.4% respectively. For 2028, the growth projection was unchanged at 1.5%. Over the medium term, consumption should be supported by gradually falling energy prices and a strong labour market. Growth would increasingly be bolstered by business and housing investment. Export growth should benefit from rising foreign demand but was being held back by competitiveness challenges and uncertainty about global trade policies.

Members exchanged views on the growth outlook in the staff projections. On the one hand, given that the better than expected performance of the economy followed a pattern seen after previous global shocks, including the April 2025 tariff shock, it was suggested that the staff projections might not sufficiently account for the flexibility and adaptability of the euro area economy. It was also argued that ongoing global growth momentum, in particular the AI investment boom spreading across the globe, together with greater defence and infrastructure spending, could have a significant positive impact on the euro area economy. From this perspective, the view was held that the recent resilience of the economy and the upward revision to the growth outlook suggested that the overall risk assessment for growth might place too much weight on the global energy crisis, and that risks to the economic outlook were arguably more balanced than previously, even if still skewed to the downside. On the other hand it was argued that, even without an abrupt correction in financial markets, a mere loss of momentum in the global AI boom would be sufficient to weigh on the euro area growth outlook via effects unfolding through global demand and net exports. In addition, a further rise in long-term interest rates could adversely affect growth. Finally, the escalating trade dispute between the United States and Canada highlighted the fact that higher tariffs could still pose a downside risk to the growth outlook.

The labour market had remained robust. Unemployment remained low, standing at 6.4% in July. Although employment was still expanding, growth in employment and the labour force continued to slow, while productivity had gradually picked up. Indicators of labour shortages and the employment expectations of households pointed towards a cooling labour market. At the same time, the employment PMIs for both services and manufacturing had increased measurably since the spring and now stood in expansionary territory. It was also observed that the staff projection for unemployment had broadly returned to the path projected before the start of the conflict in the Middle East, with the unemployment rate expected to fall to new historical lows. In this context, in-depth analysis of the labour market was seen as a key input for assessing indirect and second-round effects of the energy shock through higher wages. However, given the typical timing of wage negotiations and the lags with which wages typically reacted to higher inflation, most new insight regarding wage developments was only to be expected early next year.

Turning to the fiscal outlook, it was suggested that the recent increase in government spending and rise in long-term government bond yields could add to vulnerabilities over time. It was therefore important to maintain sound public finances. Fiscal responses to the energy shock should be temporary, targeted and tailored.

Structural reform remained necessary to support higher potential growth, although it was argued that both the continued weak momentum of reform efforts and the time lags involved limited the extent to which potential reforms should be seen as an upside risk to the growth outlook over the projection horizon. Nonetheless, simplifying and harmonising rules across the EU’s Single Market, accelerating the energy transition and completing the savings and investments union were key building blocks. As the process for agreeing on the legal framework for the digital euro moved into its final stage, members reiterated the importance of reaching agreement on the Single Currency Package as quickly as possible.

Against this background, members assessed that the risks to the growth outlook were to the downside. This was due, in particular, to the Middle East conflict and developments in Russia’s unjustified war against Ukraine. Renewed disruption of energy supplies could cause energy prices to rise further and for longer than currently expected. This would weigh on real incomes, spending and investment. A worsening of global financial market sentiment or spillovers in global bond markets could tighten credit conditions and thereby dampen demand. A resurgence of trade tensions between major economies could also further disrupt supply chains, reduce exports and weaken consumption and investment. By contrast, growth could turn out to be higher if the economy and energy markets were to adapt more quickly than expected to the disruption caused by the ongoing conflicts or if these were resolved sustainably. Moreover, the adoption of new technologies by euro area firms and spending on defence and infrastructure, as well as reforms to enhance productivity and complete the EU’s Single Market, might drive up growth by more than expected.

With regard to price developments, members concurred with the assessment presented by Mr Lane in his introduction. According to Eurostat’s flash estimate, inflation had increased to 3.3% in August, from 2.9% in July. Energy price inflation had also risen. This increase was likely to have reflected, in particular, a strong contribution from refining margins on liquid fuels, as well as higher energy commodity prices. However, the outturns for headline inflation in both July and August had been slightly lower than expected. This was partly due to lower than expected food inflation but also to a small downward surprise in core inflation in August, which might indicate some overestimation of the strength of indirect effects. In particular, services inflation had fallen from 3.3% in July to 3.0% in August. It was suggested that this deserved close attention because, if the trend were to continue, it could have a meaningful impact on the outlook for core inflation.

The main drivers of inflation since the start of the conflict in the Middle East had been rising oil and gas prices as well as refining margins. While uncertainty regarding the wider impact of the energy shock on inflation remained high, it was suggested that the transmission of energy price rises to non-energy components had so far been more benign than anticipated, with no significant broadening of higher inflation across consumption items to date. Consistent with this, and in clear contrast to the 2021-22 inflation surge when non-energy inflation had risen substantially, the percentage of goods and services for which prices were growing by more than 3% or 4% on an annual basis had not increased by much. More generally, lower core and services inflation, moderating wage growth and broadly stable longer-term inflation expectations argued against describing the transmission of the shock as broad-based, while monetary aggregates provided little evidence of demand-driven inflation. At the same time, it was pointed out that price pressures were rising in some areas. So far, the pass-through of higher costs had been seen most clearly in non-energy goods prices. Higher energy costs globally were causing pipeline pressures to rise. These pressures were evident in considerably higher inflation for imported and intermediate goods, which might increasingly spill over to the broader consumption basket, as reflected in the gradual increase in non-energy industrial goods inflation. Another inflationary factor was the increase in the prices of AI-related products.

The inflation outlook as reflected in the September staff projections had deteriorated compared with June. The surge in the prices of oil, refined products and gas and the upward shift in their futures curves suggested that inflationary pressures would be more persistent, with inflation likely to remain above target for longer than previously expected. Although the staff baseline projection for inflation in 2026 was unchanged from June, the projections for 2027 and 2028 had been revised up. The new baseline saw headline inflation averaging 3.0% in 2026, 2.5% in 2027 and 2.1% in 2028. While headline inflation was projected to come down quickly in 2027, this was mainly due to energy price inflation, which was projected to fall from almost 15% at the end of 2026 to around -4% a year later. Overall, headline inflation was expected to fall from 3.6% in the final quarter of 2026 to 1.9% in the final quarter of 2027, but it was projected to rise to 2.2% in the final quarter of 2028, partly on account of the introduction of the EU Emissions Trading System 2. Inflation excluding energy and food was projected to be 2.5% in 2026, 2.6% in 2027 and 2.3% in 2028. This implied that core inflation was expected to exceed 2% throughout the projection horizon, which was partly due to indirect effects and some second-round effects.

Members exchanged views on the inflation outlook in the staff projections. The upward move in energy prices since the cut-off date for the technical assumptions created upside risks to the inflation projections. It was argued that these upside risks did not stand in contradiction to the lower than expected inflation prints over the past months, since it seemed likely that these had largely been driven by backward-looking factors. For example, the exceptionally low food price inflation over recent months was unlikely to persist. Instead, given the recent heatwaves, droughts and wildfires, as well as the intensifying El Niño, it was argued that there were upside risks to the projection for food inflation, especially early next year as long-term contracts expired. Moreover, the pass-through of higher energy and fertiliser prices to food prices was likely to occur with a lag. However, it was pointed out that the recent downside surprises in food prices had been due to better than expected weather earlier in the year and lower than expected fertiliser and European agricultural prices, and that the new projection for food inflation reflected these lower than expected outcomes. Moreover, while it was true that most indicators suggested a notable pick-up in food price inflation in the period ahead, reflecting indirect effects from energy prices on food prices and the likely substantive impact of El Niño on international food prices, the updated baseline projection already contained a substantial rise in future food price inflation. Therefore, it was not clear that there were upside risks to food inflation relative to the baseline projection; in fact, relative to that projection, the risks could even be to the downside.

It was also observed that the inflation projection was influenced by the size of the output gap and that potential output may have been higher in view of stronger than expected productivity growth recently. More generally, it was suggested that the actual data since June – encompassing better than expected outturns for both growth and inflation – indicated that the impact of the energy price shock so far had been different from what had previously been expected, which should be considered when interpreting the latest projections.

Regarding wage growth, recent developments had been broadly in line with expectations. Wage growth was still moderating and recent labour market analysis contained the steady message that, in contrast to the pre-pandemic period, the nominal wage outlook seemed consistent with the inflation target in an environment of low unemployment. In this context, it was highlighted that the staff projections assumed that some second-round effects would materialise over the coming years, but so far there was little evidence of this in terms of wage growth or hiring data. While wage growth was projected to be marginally higher, unit labour costs, which were a critical variable driving inflation, were expected to grow more slowly owing to stronger productivity growth, and the number for the second quarter of 2026 had also surprised notably to the downside. At the same time, it was argued that it was important not to be complacent about wage developments, since they were likely to reflect second-round effects only gradually over time. Moreover, there remained an open question about the extent to which the headline inflation rates currently being projected could prompt a material adjustment in wage negotiation trends. Still, the profile in the latest projections incorporated a partial wage response to higher inflation that was in line with historical patterns.

Turning to inflation expectations, while most measures of longer-term inflation expectations had been relatively stable and continued to stand at around 2%, inflation expectations for shorter horizons were significantly higher, as reflected in surveys of both experts and households. Inflation expectations captured in the Consumer Expectations Survey had edged up again, in line with higher energy prices. Five-year-ahead inflation expectations in the survey had increased to 2.5% in August, which was the highest level recorded since the data started in late 2022. However, it was also pointed out that surveys of household inflation expectations typically contained some bias in terms of the level of expectations. The readings from the Consumer Expectations Survey should therefore not be compared mechanically with the 2% inflation target and it was instead better to focus on the changes in the numbers over time.

Against this background, members assessed that the risks to the inflation outlook were to the upside. This was due, in particular, to the Middle East conflict and developments in Russia’s unjustified war against Ukraine. The energy shock could intensify further and its effects on other prices and wages could be stronger than currently expected. Gas prices, in particular, could increase in the event of further supply disruptions or an unusually cold winter coinciding with low storage levels. The longer energy prices stayed high, the more likely they were to drive up broader inflation through indirect and second-round effects. Renewed trade tensions could give rise to more fragmented global supply chains, curtail the supply of critical raw materials and worsen capacity constraints in the euro area economy. Extreme weather events, potentially reinforced by intensifying El Niño conditions, and the unfolding climate and nature crises more broadly, could drive up food prices by more than expected. By contrast, inflation could turn out to be lower if ongoing geopolitical conflicts were resolved sustainably or if indirect or second-round effects from the recent energy price shock proved less pronounced than anticipated. More volatile and risk-averse financial markets could weigh on demand and thereby lower inflation as well.

Given the use of the forward energy price curves in the baseline projections and in view of the ongoing highly uncertain environment, members considered that supplementing the staff baseline projections and the regular risk assessment with updated alternative scenarios was an informative way to illustrate the broad range of possible outcomes for inflation and growth. While the milder scenario remained a possibility that should not be excluded, it was suggested that the adverse and severe scenarios were particularly helpful in assessing risks surrounding the baseline, as each of them could be mapped to scenarios for the quantity of barrels of oil or megawatt-hours of gas not available to the market. Overall, while members generally judged that the baseline inflation projections, albeit with upside risks, were more plausible than any of the alternative scenarios, the view was also held that the adverse scenario was more likely than the baseline.

Turning to the monetary and financial analysis, members largely concurred with the assessment provided by Ms Schnabel and Mr Lane in their introductions. Financial markets were now fully pricing in an ECB interest rate hike at the September meeting, with a further two to three hikes priced in by the end of 2027. However, survey-based rate expectations – from the Survey of Monetary Analysts and surveys conducted by Bloomberg and Reuters – were quite different and suggested that the deposit facility rate would plateau at 2.50% after a rate hike in September. This gap could be explained partly by risk premia embedded in the market curve and partly by the earlier timing of the surveys, although the reasons behind the difference warranted further examination.

Long-term yields in the euro area had increased since the Governing Council’s July meeting and the overall rise since the start of the conflict in the Middle East had been substantial. Recent moves at the longer end of the curve appeared to be driven primarily by higher real term premia, which could be explained by greater public and private debt issuance and uncertainty about the fiscal outlook.

The increase in market interest rates since the July meeting reflected similar moves in global markets. Against this backdrop, it was important to analyse both developments in long-term yields globally and their drivers, understand the effects of the gradual rise in yields – including for financial conditions, public finances and the macroeconomy – and assess further developments.

Wider financial conditions had remained benign. Euro area sovereign bond markets had functioned in an orderly manner, showing some differentiation among Member States with different fundamentals. Stock markets had performed strongly over the preceding month, reflecting positive demand impetus – especially from AI and fiscal policy – and expectations of productivity gains from AI technologies. At the same time, the dependence of stock prices on AI and the seeming decoupling of valuations from the wider economy, particularly in the United States, could also be a concern, as it created the risk of an abrupt correction in financial markets. A gap between the expected and realised benefits of AI could generate stock market volatility. Moreover, even without a sharp correction, a loss of momentum in the AI boom could adversely affect financial conditions in the euro area. In addition, greater bond issuance in Europe by US AI-related companies represented another potential source of contagion risk to the euro area from any material deterioration in the AI sector.

Regarding financing conditions for firms, bank lending rates had risen following the interest rate increase in June, to stand at 3.8% in June and July, after 3.6% in May. The cost of market-based corporate debt had stood at 4.0% in July, which was similar to previous months and well above its level before the conflict in the Middle East. Overall growth in lending to firms had remained robust and was continuing to increase further. The annual growth rate of bank lending to firms, which usually responded to changes in monetary policy with a longer delay, had increased further to 4.4% in July, from 4.0% in May and June, while the annual growth rate of corporate bond issuance had been 3.4%, after 3.6% in June and 3.3% in May. However, the corporate credit expansion was uneven and sector-specific, and it was partly dependent on AI investment, fiscal support and temporary financing needs created by the energy shock. In particular, although credit for smaller firms had, in contrast to several recent years, ticked up and was growing this year, credit for larger firms had picked up by more and remained stronger. This was especially the case when considering both bank credit and private credit, which had become increasingly important since the pandemic, and also the possibility that larger firms could substitute between these types of credit. By contrast, smaller firms could not normally access private credit and were much more reliant on bank credit. It was also observed that AI-related companies were receiving more credit than other companies and high-risk companies were receiving less credit than other companies. Together, these factors suggested that credit to small and medium-sized enterprises (SMEs) and the self-employed could be constrained. This would be a concern, given the importance of SMEs to many European economies, and further analysis on this topic was therefore warranted.

Mortgage rates had been unchanged in June and July, at 3.5%, while mortgage lending growth had softened to 3.0% in July, from 3.1% in May and June. It was suggested that overall household lending growth had remained robust. At the same time, it was argued that credit dynamics appeared to have lost steam recently, with mortgage activity weakening.

Monetary policy stance and policy considerations

Turning to the monetary policy stance, members assessed the data that had become available since the last monetary policy meeting in accordance with the three main elements that the Governing Council had communicated in 2023, and updated in July 2025, as shaping its reaction function, namely: (i) the implications of the incoming economic and financial data for the inflation outlook and the risks surrounding it; (ii) the dynamics of underlying inflation; and (iii) the strength of monetary policy transmission.

Members largely agreed that the inflation outlook had deteriorated, as the conflict in the Middle East and recent developments in Russia’s unjustified war against Ukraine had pushed the path of energy prices up further. According to the September staff projections, inflation would remain above target for longer than previously expected. Compared with June, the baseline projection for inflation was unchanged for 2026 but had been revised up for 2027 and 2028, even if the view was also held that the projections painted a picture that was not significantly different from the previous two exercises, including the projections produced just after the start of the Middle East conflict. The more persistent energy shock was likely to keep headline inflation well above target into the first half of 2027, which was seen as uncomfortable but clearly not as dramatic as during the 2021-22 inflation surge. Thereafter, energy inflation was expected to decline and turn negative up to mid-2028, bringing headline inflation down. Headline inflation was expected to return to around target towards the end of 2027, supported by the effects of higher interest rates.

Higher energy prices were expected to feed through gradually to core and food price inflation. The improved economic outlook was also expected to contribute to slightly higher core inflation, which was expected to keep rising until early 2027 and stay elevated for the rest of the year, before moderating in 2028. However, the decline from the peak was expected to be gradual and core inflation was still projected to stand at 2.3% at the end of the projection horizon, implying that it would exceed 2% for the entire projection horizon. This persistence partly reflected the gradual build-up of indirect and, to a lesser extent, second-round effects from the energy shock.

The outlook – and therefore the baseline of the September staff projections – remained highly uncertain. In relation to the energy shock, the updated scenarios put together by staff illustrated the broad range of outcomes for how growth and inflation would evolve under different assumptions about its intensity and duration, as well as its indirect and second-round effects. The longer energy prices stayed high, the more likely they were to drive up broader inflation through indirect and second-round effects. By contrast, inflation could turn out to be lower under a milder profile for energy prices than in the baseline, or if indirect or second-round effects from the energy price shock proved less pronounced than anticipated. In this context, it was argued that, given the persistence of the energy shock, accumulating price pressures and better demand conditions, the risk of indirect and second-round effects had become more relevant and needed to be carefully monitored. So far, however, indirect effects had remained contained and second-round effects had not been seen. It was also suggested that a cooling labour market and moderating wage growth could limit the risk of second-round effects. In addition, while inflation expectations over shorter horizons remained at elevated levels, the credibility of the ECB’s commitment to bringing inflation back to target over the medium term remained solid, with medium and longer-term inflation expectations remaining anchored. Most measures of longer-term inflation expectations stood at around 2%, supporting the stabilisation of inflation around target in the medium term.

All members viewed the risks surrounding the inflation outlook as being to the upside relative to the staff baseline projections, with the evolution of the conflict in the Middle East and developments in Russia’s unjustified war against Ukraine being the key sources of risk. The substantial upward movement in energy prices – particularly gas prices – since the cut-off date for the projections had pushed up the paths of futures prices and created an upside risk to projected inflation. With energy reserves lower than at the start of the conflict in the Middle East, the energy shock could intensify further. In particular, gas prices could continue to increase in the event of further supply disruptions or an unusually cold winter coinciding with the currently low storage levels. Refining margins could remain at elevated levels for longer than currently anticipated and could rise even further if refining capacity fell further. In this context it was observed that, if there was a materialisation of these major upside risks to inflation linked mainly to geopolitical events, there would also be an adverse effect on real incomes, which would weigh on economic growth. The assessment that risks to the growth outlook were to the downside was therefore part of an integrated assessment that saw upside risks to inflation.

Inflation could also turn out higher if the effects of the energy shock on other prices and wages were stronger than currently expected. It was argued that, if sustained, the more resilient economy, together with the demand impulse coming from the AI boom and fiscal spending, could imply that firms might be in a stronger position to pass higher input costs on to consumer prices. This could increase the risks of inflation persistence and second-round effects. Extreme weather events, potentially reinforced by intensifying El Niño conditions, and the unfolding climate and nature crises more broadly, could drive up food prices by more than expected. Finally, there remained a risk that renewed trade tensions could give rise to more fragmented global supply chains, curtail the supply of critical raw materials and worsen capacity constraints in the euro area economy.

At the same time, risks did not all point in the same direction. Inflation could turn out to be lower if ongoing geopolitical conflicts were resolved sustainably or if indirect or second-round effects from the recent energy price shock proved less pronounced than anticipated. In this context, it was observed that there was a possibility of the Middle East conflict being settled this autumn. This made the milder scenario a possibility, especially given that high volatility in energy prices had closely reflected the evolution of political tensions, which questioned the narrative of persistent supply disruptions in energy markets. There was also a risk that either a worsening of global financial market sentiment associated with more volatile and risk-averse financial markets or spillovers in global bond markets could tighten credit conditions and dampen demand, thereby lowering both growth and inflation.

Turning to underlying inflation, most measures had been broadly stable in July. Although all measures continued to stand above 2%, underlying inflation was judged to have remained contained so far. There had not yet been any significant broadening of inflation following the energy shock, with core inflation edging down to 2.4% in August from 2.5% in July, little evidence of effects on domestic inflation and broad stability in the trimmed-mean measures of underlying inflation. Wages did not show a material response to the energy shock at this stage. Wage growth was still moderating, with compensation per employee growing at an annual rate of 3.3% in the second quarter, down from 3.5% in the first quarter. Rising labour productivity had also helped contain growth in unit labour costs. At the same time, growth in unit profits had risen. Looking ahead, the ECB’s wage tracker pointed to a modest uptick, to 2.7%, in negotiated wage growth in the first half of 2027. All of this suggested that there had been no second-round effects from the energy shock on wages so far, and it was also argued that future wage claims could be moderated by a cooling labour market. Therefore, the risk of second-round effects via wages could be lower than previously thought. At the same time, it was argued that it was important not to be complacent about wage developments, since they were likely to only gradually reflect second-round effects over time. In addition, the more resilient economy could mean that workers might find it easier to ask for higher wages, or at least for a recovery of real wage losses, especially in sectors where labour was scarce.

Finally, the transmission of monetary policy had been smooth. The effects of higher interest rates would support the expected return of inflation to around target towards the end of 2027. The repricing at the long end of the yield curve, provided it remained orderly, also supported the intended monetary policy stance and could have implications for appropriate policy rates in the future. Model estimates suggested that the effects of higher long-term interest rates on growth and inflation could be material. In this context, it was observed that the September staff projections incorporated the direct impact of the recent significant increase in long-term interest rates until the cut-off date.

It was argued that robust growth in lending to households and firms suggested that financing conditions were still not restrictive. However, it was noted that firms were looking for credit partly for liquidity reasons, to cover higher input costs and working capital needs following the energy shock. It was therefore argued that high overall corporate credit growth currently did not imply an investment boom. In addition, household credit growth was being supported by a significant increase in applications for credit from constrained households who were looking for loans to cope with the higher cost of living. As such, this positive contributor to household credit growth was not a good sign but rather signalled that some people were under pressure. More generally, it was suggested that recent developments, including higher market rates, the appreciation of the euro, weaker mortgage activity and pressure on lower-income households, indicated that monetary and financial conditions were already restraining parts of demand.

Financing conditions for both firms and households were also expected to tighten, and this would probably act as a drag on growth in the period ahead. In this context, it was observed that AI investment in the euro area was heavily dependent on credit and sensitive to interest rates.

Monetary policy decisions and communication

Against this background, all members supported the proposal made by Mr Lane to raise the three key ECB interest rates by 25 basis points. This decision underscored the Governing Council’s commitment to setting monetary policy to ensure that inflation stabilised at the 2% target in the medium term. The conflict in the Middle East was continuing to generate inflation pressures. The energy shock was more persistent than had previously been anticipated and the inflation outlook had deteriorated, with the staff baseline projection for inflation revised up for 2027 and 2028. Inflation was set to remain well above target for an extended period, with the return to target now expected to be later than previously anticipated. Core inflation was expected to remain above 2% for the entire projection horizon. Risks to the inflation outlook remained to the upside. The economy also continued to be resilient and could therefore absorb an increase in interest rates. While it was noted that the response should remain proportionate, it was also pointed out that a deposit facility rate of 2.50% remained in the range of neutral interest rates estimated by staff. Overall, another 25 basis point rate hike was therefore warranted by the incoming data, the September staff projections and the evolving inflation outlook, and necessary to secure a sufficiently timely and sustained return of inflation to target. Such action should support the credibility of the Governing Council’s price stability commitment and its reaction function, thereby helping to ensure that medium and longer-term inflation expectations remained well anchored. The decision to increase rates by 25 basis points was also robust across all three scenarios – milder, adverse and severe – produced by staff, reflecting the expected deviation of inflation from target over the medium term in the absence of appropriate monetary policy action. The decision also left the Governing Council well positioned to navigate the high uncertainty caused by the conflict in the Middle East.

With regard to communication, members reiterated that the Governing Council’s future interest rate decisions would continue to be based on its assessment of the inflation outlook and the risks surrounding it, in light of the incoming economic and financial data, as well as the dynamics of underlying inflation and the strength of monetary policy transmission. The Governing Council would also continue to follow a data-dependent and meeting-by-meeting approach to determining the appropriate monetary policy stance, without pre-committing to a particular rate path.

In view of the volatile environment and pervasive uncertainty, which meant that the situation could change rapidly, it was particularly important to refrain from giving any guidance regarding the future interest rate path and retain full discretion for all future meetings, based on data-dependence and informed by a comprehensive assessment. In this context, it was sensible to simply acknowledge the high uncertainty, the risks surrounding the inflation outlook and the different possible scenarios, given that the Governing Council’s reaction function already indicated how it would respond in different circumstances. Against this backdrop, communication should remain neutral, neither suggesting that the current decision was another step in a predetermined tightening cycle nor that it was the last rate hike. At the same time, continued vigilance was vital. Therefore, it was important to communicate that the Governing Council was firmly committed to delivering 2% inflation over the medium term, was closely monitoring all incoming information, and remained agile and flexible to respond to changes in the inflation outlook in either direction.

Looking ahead, the Governing Council would set monetary policy as appropriate to ensure that inflation stabilised sustainably at the medium-term target. It would continue to monitor closely the size and persistence of the energy price increase and how it was feeding through to price and wage-setting, inflation expectations and overall economic dynamics. Over the coming months, it would be particularly important to gauge the strength of indirect effects and watch out carefully for second-round effects, as these dynamics could make the direct inflationary impact of the energy shock more persistent.

Taking into account the foregoing discussion among the members, upon a proposal by the President, the Governing Council took the monetary policy decisions as set out in the monetary policy press release. The members of the Governing Council subsequently finalised the monetary policy statement, which the President and the Vice-President would, as usual, deliver at the press conference following the Governing Council meeting.

Monetary policy statement

Monetary policy statement for the press conference of 10 September 2026

Press release

Monetary policy decisions

Meeting of the ECB’s Governing Council, 9-10 September 2026

Members

  • Ms Lagarde, President
  • Mr Vujčić, Vice-President
  • Mr Cipollone
  • Mr Demarco
  • Mr Dolenc
  • Mr Elderson
  • Mr Escrivá*
  • Mr Kaasik
  • Mr Kazāks*
  • Mr Kažimír
  • Mr Kocher
  • Mr Lane
  • Mr Makhlouf
  • Mr Moulin
  • Mr Nagel
  • Mr Panetta
  • Mr Patsalides*
  • Mr Pereira
  • Mr Radev
  • Mr Rehn
  • Mr Reinesch
  • Ms Schnabel
  • Mr Šimkus*
  • Mr Sleijpen
  • Mr Stournaras*
  • Mr Wunsch
  • Mr Žigman*

* Members not holding a voting right in September 2026 under Article 10.2 of the ESCB Statute.

Other attendees

  • Ms Senkovic, Secretary, Director General Secretariat
  • Mr Straub, Secretary for monetary policy, Director General Monetary Policy
  • Mr Kapadia, Head of Division, Directorate General Monetary Policy

Accompanying persons

  • Ms Bénassy-Quéré
  • Ms Brezigar
  • Mr Dechaene
  • Mr Horváth
  • Mr Koukoularides
  • Mr López
  • Mr Lünnemann
  • Mr Madouros
  • Ms Mauderer
  • Mr Nicoletti Altimari
  • Ms Radeva
  • Mr Randveer
  • Ms Raposo
  • Mr Reichenbachas
  • Mr Rutkaste
  • Ms Schembri
  • Mr Šošić
  • Ms Stiftinger
  • Mr Tavlas
  • Mr Välimäki
  • Mr ter Weel

Other ECB staff

  • Mr Proissl, Director General Communications
  • Ms Vansteenkiste, Counsellor to the President
  • Ms Rahmouni-Rousseau, Director General Market Operations
  • Mr Arce, Director General Economics
  • Ms Nickel, Deputy Director General Economics

Release of the next monetary policy account foreseen on 26 November 2026.

European Central Bank

Directorate General Communications

Reproduction is permitted provided that the source is acknowledged.

Media contacts