“Navigating an uncertain world: steering monetary policy in the right direction to hold our course”

Ladies and Gentlemen,

Thank you for welcoming me to this institution, where I unfortunately did not study, but where I had the privilege of teaching economic policy for nearly ten years. This evening, I’d like to talk to you about monetary policy: the subject may seem technical – I’ll do my best not to be dry and to explain it clearly – but it is undeniably essential. And since we set this date two months ago, monetary policy has once again become a burning issue: at the start of 2026, it was firmly in a good position” of “2+2” – 2% inflation and a 2% interest rate – and was almost boring. But today, the role of a central banker is to make decisions amid uncertainty, to distinguish information from noise, and to remain credible. And it is no small matter to discuss this at Sciences Po, where public policy needs to be both a technical matter and a question of judgement, trust and institutions.

With the war in the Middle East, we are directly faced with an apparently simple question: are we seeing a repeat of 2022, with an energy price shock in the wake of a military attack, followed by a surge in inflation not seen for decades? (I) Then I will turn to an even less simple question: what should we do now? (II) 

I. Why we are not witnessing a repeat of 2022

Are we going through 2022 once again? The answer at this stage is no. But this “no” needs to be explained, as it shapes both our assessment and our course of action. Admittedly, the shock appears to be significant, particularly as regards oil, since the price of gas was lower before the shock of February 2026. 

But we are not in 2022, first and foremost because the macroeconomic situation is different. The euro area economy, like that of France, is not facing post-pandemic imbalances: strong demand for durable goods, particularly in the United States thanks to fiscal support, coupled with supply chain pressures, was fuelling inflationary pressures at the time. In February 2022, inflation in the euro area was 5.9% (4.2% in France); in February 2026, it had fallen to 1.9% (1.1% in France).

The monetary policy stance is also different today. We are not subject to any constraints relating to asset purchases, forward guidance or the proximity of the lower bound on interest rates. And the normalisation of monetary policy has led to a gradual and predictable reduction in the ECB’s balance sheet. 

This strong starting position is largely due to the episode we went through between 2022 and 2025. At the time, there was strong inflationary pressure and many feared that a recession would be necessary to reduce it. Yet disinflation was achieved without a recession: economic activity slowed without collapsing, labour markets held up well, and, above all, inflation expectations remained firmly anchored. This episode taught us two things. 

First, not all supply shocks are the same.ⁱ Shocks to energy prices, for example, feed through to the prices of other consumer goods, such as food. But for energy-importing countries, this amounts to what is sometimes referred to as an implicit tax imposed on consumers from abroad, which dampens demand for other goods.ⁱⁱ  While tensions in goods transport have temporary effects, a shortage of intermediate goods has medium-term repercussions.ⁱⁱⁱ The lesson from 2022 is, first and foremost, that a supply shock is not necessarily inflationary in the long term: it all depends on its propagation. The problem is not just the initial impact on energy prices or supplies, but its gradual spillover into the domestic components of inflation: wages, service prices, core inflation – i.e. excluding energy and food – and inflation expectations. 

The second lesson concerns the credibility of the central bank. When expectations remain anchored, there is no need to overreact by aggressively raising interest rates. Once credibility has been established, and expectations are therefore firmly anchored, monetary policy can respond in an unhurried manner, guided by analysis and judgement, rather than applying a mechanical response such as the Taylor rule, which recommends setting key interest rates above inflation (in Taylor’s original rule, key rates had to be raised by 1.5 percentage points for every 1 percentage point rise in inflation).

As I have already pointed out,^iv this could create a virtuous circle: because market participants believe in our determination, expectations remain more favourable; and because they remain more favourable, disinflation is less difficult and therefore less costly in terms of economic activity and employment. 

Credibility is therefore not merely an additional source of comfort; it is a key factor in the effectiveness of monetary policy. And this channel of expectations does more than just make our task easier: it profoundly alters the dynamics of disinflation.^v But the still vivid memory of the recent shock may have a mixed effect here. On the one hand, this memory can make price and wage-setting behaviour more responsive: economic agents are mindful of the recent bout of inflation, and their expectations may be more sensitive to a new energy shock. On the other hand, this same memory also serves as a reminder of the success of monetary policy: we have demonstrated our ability to bring inflation back towards the target.

That is why 2026 is, at this stage, different from 2022. But we must remain vigilant. 

II. What should we do now?

In these (highly) uncertain times, our good position has never been a comfortable one.^vi But we now find ourselves in a more uncomfortable situation, for two reasons: we are faced with a high degree of uncertainty, and we have to deal with a negative supply shock. 

Dealing with uncertainty

As early as 1921,^vii the economist Frank Knight became famous for distinguishing between risk – which can be quantified – and uncertainty – which cannot. The Iranian crisis illustrates a situation where probabilities are ill-defined, where non-linear reactions are possible, and where market reactions can be swift and sometimes excessive. In an uncertain environment, the use of multiple scenarios is an important tool.  This is exactly what we stated in our monetary policy review last June: our decisions must take into account “not only the most likely path, but also the risks and uncertainties surrounding it”, in particular through “scenario and sensitivity analyses”.^viii This is where we are today, thence the three scenarios published by the ECB and the Banque de France on 19 March and 25 March, respectively.

The “baseline” scenario would be a significant but temporary shock, with limited spillover and firmly anchored expectations. A more “adverse” scenario would witness much greater spillover with second-round effects, whereby oil prices would reach USD 119 per barrel in the second quarter of 2026. Lastly, a “more severe” but less likely scenario would combine a larger and more persistent shock with increased financial market volatility; oil prices would reach USD 145 per barrel and remain at higher levels for a sustained period. Both current and future oil prices now exceed USD 100 per barrel, bringing us closer to the intermediate “adverse” scenario.

To identify the most probable scenario, monitoring indicators becomes crucial. As Christine Lagarde stressed at the press conference on 19 March, we are closely tracking developments in the commodities and energy markets, as well as the possible re-emergence of supply bottlenecks. 

This uncertainty also calls for heightened vigilance concerning financial stability. Since early March, we have already witnessed tighter financial conditions in the markets. Investors have revised their forecasts for key interest rates upwards, with two to three rate increases expected by the end of the year. Market rates have risen for short term maturities; for example, the day-to-day interest rate benchmark for the euro area, the 2-year €STR, has risen by around 70 basis points since the beginning of March, as has the 10-year composite bond yield. 

Some of this financial tightening is therefore already underway. It is more pronounced in countries with greater fiscal vulnerabilities: spreads on 10-year bonds have widened further – in France, unfortunately, and even more so in Italy and the United Kingdom. Unlike in 2022, this very fragility requires greater restraint in terms of fiscal support measures.

But this trend should not be confused with the current central bank action: it results from both a combination of higher inflation expectations, higher anticipated interest rates and rising risk premia. If this shock were to persist, there would be a risk that this pressure could spread to other segments: the equities market, the debt of the most vulnerable companies, certain sovereign markets, or non-bank financial institutions, including private credit markets. This could lead to financial instability. It could also raise another crucial issue, namely the potential transmission to the financing of the economy through a tightening of the credit conditions offered to households and businesses. 

Dealing with a negative supply shock

A second challenge relates to the course of action: such a negative supply shock pushes up inflation whilst slowing economic activity, as it also squeezes real income and demand. 

The problem is to not let inflation get out of hand, whilst at the same time avoiding overreacting to a shock that is, in any case, already slowing down the economy. Obviously, our responsibility does not concern the price of oil or gas taken in isolation. Our responsibility lies more in how this shock will – or will not – feed through to core inflation, expectations, financial conditions, and ultimately to price stability over the medium term. Our policy response continues to be based around the three criteria we have outlined since 2023 – the outlook for inflation including inflation expectations –, core inflation and monetary policy transmission. In the face of the current supply shock, our focus is especially on the first two components. Aside from the initial shock to energy prices, the key issue is the pass-through to the domestic components of inflation: wages, service prices, inflation excluding energy and food, and their future trajectory. This is where persistence comes into play. We are therefore paying very close attention to firms’ price expectations, particularly through our Banque de France monthly business surveys, which are excellent indicators of the frequency and magnitude of price adjustments. We are also tracking the inflation expectations of households as well as wage indicators. 

Monetary policy therefore works in two different stages: in the first stage, assessing the intensity and duration of the shock as accurately as possible and, crucially, limiting its contagion through second-round and indirect effects. 

So, what can we say about the very latest developments, aside from the considerable volatility of statements and information? 

• First, that the prolongation of the conflict is obviously a negative factor. As of today, 2 April, we are closer to the intermediate adverse scenario than to the baseline scenario adopted by the ECB on 19 March.
• Second, March inflation figures for the euro area and for France, published the day before yesterday, confirm a strong first-round effect: energy prices rose by 4.9% in the euro area, and consequently, headline inflation climbed by 2.5% in the euro area and by 1.9% in France. However, core inflation (excluding energy and food, which accounts for 73% of the price index in France) has so far remained firmly under control in France and the euro area, at 1.3% and 2.3%, respectively.

• Nevertheless, the utmost vigilance is required: market inflation expectations have risen sharply; we do not yet have those of businesses and households. And our macroeconomic models may underestimate more negative microeconomic consequences: disruption to certain supply chains for plastics or other related products; and the temptation for companies – even in other sectors not directly affected by the initial shock – to ‘pre-empt’ price increases.
• It is far too early to predict a timetable for ECB interest rate rises but it is clear that we have the capacity to act when and in whatever way necessary. Obviously, the next change in key interest rates is highly likely to be upwards.

Finally, allow me to dispel a misconception: “monetary policy is not suited to France, which has the lowest inflation in the euro area.” Let us remember that monetary policy targets the entire euro area, not a single country. In France, the recent low level of inflation is an advantage for competitiveness and purchasing power. But this is attributable to specific and partly temporary factors: changes in electricity prices since February 2025 and French characteristics that result from the previous shock — less exposure to gas, the energy price shield, and more frequent wage negotiations, leading to faster disinflation in service prices. Moreover, France was less exposed than other euro area economies to the 2021–22 energy crisis: HICP inflation peaked at 10.6% in the euro area, compared with 7.3% in France. Above all, comparing a 10-year rate with current monthly inflation is not the correct economic calculation. To gauge the real relevant rate, one must subtract expected inflation over the coming years. However, over a 5- or 10-year horizon, French inflation expectations are anchored to the same benchmark as those of the euro area. 

I shall conclude by returning to the title of my speech: we must navigate in an increasingly uncertain world. As Edgar Morin so nicely put it, “We try to surround ourselves with as many certainties as possible, but living means navigating a sea of uncertainty, through small islands and archipelagos of certainties where we replenish our supplies.” ^ix Clearly, we cannot command the wind or the sea. We do not choose geopolitical conflicts, the conditions for navigation through the Strait of Hormuz or their impact on oil prices, but we do know our starting and finishing points. We started out from a good position, facing significantly less inflationary pressure than in 2022. At the finish, we have an unwavering determination to stabilise inflation at 2% in the medium term. Our course remains clear: price stability, because that is our mandate, and because that is what our fellow citizens expect, in the service of purchasing power, confidence in the currency and the strength of our economy. Households and businesses can have confidence in the central bank: between 2023 and 2025, we were able to bring inflation back towards our target without causing a recession. And mark my words, this time even more so, we will act without haste but without any hesitation if this proves necessary. Thank you for your attention.

ⁱ Villeroy de Galhau (F.) (2024), Monetary Policy in Perspective (II): Three landmarks for a future of “Great Volatility”, Speech, London School of Economics, London, 30 October.
ⁱⁱ Kharroubi (E.), Smets (F.) (2024), “Energy shocks as Keynesian supply shocks: Implications for fiscal policy”, European Economic Review, 170, 104875.
ⁱⁱⁱ Elsayed (M.), Marx (M.), Grosse-Steffen (C.) (2025), Global sectoral supply shocks, inflation, and monetary policy, Banque de France Working Paper Series No. 1026, 22 December.
^iv Villeroy de Galhau (F.) (2024), A Monetary Policy Perspective: Three lessons from the recent inflation surge, Speech, University of New York, New York, 22 October.
^v Dupraz (S.) and Marx (M.) (2025)The benefits of well-anchored inflation expectations, Blog post 396, Eco Notepad, Banque de France, 20 March.
^vi Villeroy de Galhau (F.) (2025), “About our monetary policy: A good position but not a comfortable, nor a fixed one",  CEPR Paris Symposium 2025, Paris, 5 December.
^vii  Knight (F.H) (1921) Risk, Uncertainty, and Profit, Boston, Hart, Schaffner & Marx; Houghton Mifflin Company, 1921, pp. 19-20.
^viii ECB (2025) The ECB’s monetary policy strategy statement, June § 9:   “The Governing Council   […] takes into account not only the most likely path for inflation and the economy but also surrounding risks and uncertainty, including through the appropriate use of scenario and sensitivity analyses”.
^ix Morin (E.) (2020), “Nous devons vivre avec l’incertitude" target="_blank", interview, CNRS Le Journal, 6 April