Source: European Central Bank
Interview with Isabel Schnabel, Member of the Executive Board of the ECB, conducted by Olaf Storbeck on 14 February 2025
19 February 2025
How relevant is the natural rate – R* – for day-to-day policymaking from your point of view?
The natural rate of interest is an important theoretical concept. But it’s not well-suited to determine the appropriate monetary policy stance. The ECB staff analysis that was published recently had one main message: we know that we know very little. Model and estimation uncertainty result in confidence bands that are so wide that they include any reasonable interest rate that the ECB may set at this point. Moreover, R* is a steady-state concept for a world without shocks. That’s certainly not the world that we are in today. Just look at what’s happening with the evolving trade conflict on which we are getting news on a daily basis. So for all those reasons, I think R* cannot be any reliable guide for monetary policy in real time.
Has your view on this changed?
The point I have always emphasised is how R* is evolving over the longer term. People have focused too much on the narrow range for R* that was given in the staff note. This is misleading for several reasons. The narrow range only includes the models for which estimates were already available for the fourth quarter of 2024. If you look at the R* estimates for the third quarter, you see that the range actually goes up all the way to 3%. This is even above the current deposit facility rate of 2.75%. And that range still only includes the uncertainty stemming from using different models. If you add the parameter and filtering uncertainty, you get even wider bands. The one thing that you do see is that the overall range seems to have moved up over recent years. For me, that is the key point.
But the most recent ECB estimates of R* also suggest that the current level is still lower than it was before the global financial crisis and the European sovereign debt crisis.
That remains to be seen. There has been a clear upward trend. I expect this trend to continue for a number of reasons, including high and rising public debt and the huge investment needs for the digital and green transitions. Another factor is increasing global fragmentation. It leads to a partial reversal of the global savings glut, due to shrinking current account surpluses of some major economies, which was one of the main factors that had pushed R* down. So for me, the main message from the R* analysis is: maintaining price stability over the medium term is likely to require higher real rates in the future than before the pandemic. We cannot pin down the level of R* with any degree of confidence, but we can get an impression about the direction. For me, that direction for R* now is upwards again.
The Euro zone economy suffers from a lack of economic dynamism and economic growth. Doesn’t this put downward pressure on the natural interest rate?
Yes, there have been secular factors that have pushed R* down. But we are currently in a situation of transformation that may actually reverse that trend. That’s the whole point.
When you say that R* is not very helpful for short-term monetary policymaking, why have you stressed it so much in your speeches and interviews?
It’s important that we understand general macroeconomic trends. Also in the pre-pandemic period, it was very important to understand the underlying natural real rate environment. It can never be precise, but it helps us understand the broader picture. It has no impact on any individual rate decision.
But would you say that it is relevant for the medium-term trajectory of monetary policy, let’s say for the next year or two? Or does it only matter over the next ten or 20 years?
I think it has an impact on our medium-term thinking.
Medium-term thinking would mean: it matters over the next two to three years, right?
Well, it’s hard to pin down precisely.
Some ECB observers have suggested that the natural rate was used by more hawkish voices as an argument in favour of being more careful and not lowering interest rates too fast. Would you agree?
If you believe that R* has moved up, this argues for a more cautious approach. But this cannot just depend on R*. We need to look at the incoming data in order to understand how restrictive our monetary policy is. And the more evidence we have that monetary policy is no longer restrictive, the more cautious we have to become because further rate cuts may no longer be appropriate.
So how restrictive is the ECB’s monetary policy at the moment?
The data are showing that the degree of restriction has come down significantly, up to a point where we can no longer say with confidence that our monetary policy is still restrictive. One of the important data sources in this context is the bank lending survey.
We’re looking at that very carefully. For corporate loans, 90% of banks said in the most recent round that the general level of interest rates has no impact on loan demand, while 8% said it has lifted credit demand. A year ago, a third of banks said that interest rates were weighing on loan demand. It’s even clearer when you look at mortgages. Almost half of banks said in the most recent round that the general level of interest rates is supporting loan demand. A year ago, more than 40% said that it was constraining loan demand. This is also reflected in a historically strong increase in mortgage demand in that same survey, which is gradually transmitting into the hard data on loan growth. Corporate loans were growing by 1.5% in December, mortgages by 1.1%.
The easing is also being transmitted to the real economy. Consumption picked up in the third quarter by more than we had expected. And the savings rate has started to come down from its very high level. But of course, there are transmission lags, and part of the easing is still in the pipeline.
You said that you can’t say with confidence anymore if monetary policy is still restrictive. The last ECB policy statement clearly stated that it still is. Do you have a different view than the ECB stated in its latest policy statement?
No, I fully agreed with the statement last time. But we are now a step further, right? The January monetary policy statement referred to the interest rate of 3% and the level of restrictiveness before the latest monetary policy decision. The further we go down, the lower my conviction in such a statement will be. And note that I’m not saying our monetary policy is no longer restrictive. What I’m saying is I’m no longer sure whether it is still restrictive. But we should not overstate a difference of 25 basis points.
Should the ECB drop the reference to restrictiveness in March?
That is a discussion we should have in the next meeting.
In an FT survey of Euro zone economists just before Christmas, half of them said they think that the ECB is behind the curve. What is your view on this?
I’m firmly in the camp of the other half who think that we are right on track. The data that we’ve seen have confirmed that our gradual and cautious approach has been appropriate. Domestic inflation is still high, wage growth is still elevated, and we’ve seen new shocks to energy prices. We’ve also seen that inflation expectations are very sensitive to such shocks. So I think our approach is just right.
Some economists argue that the big uncertainty and all those shocks could justify insurance cuts. Do you have any view on that?
I don’t see any argument for that at this point, especially as we are getting closer to no longer being restrictive. If anything, we are getting closer to the point where we may have to pause or halt our rate cuts.
Pause or halt… but not increase?
No. That I would exclude.
How close do you think we are to the point where the ECB should pause its easing?
I will leave that to your interpretation. I don’t know what’s going to happen in the next meetings, so let’s see. But we need to start that discussion.
That’s not what markets take as the base case scenario right now. Do you think that markets are ahead of themselves?
Well, markets have been jumping around a bit in response to what is happening in the world. But an April rate cut is no longer fully priced in. So markets are not entirely sure either.
How well is monetary transmission working at the moment? We saw quite an uptick in yields in December although there wasn’t any change in monetary policy. All other things being equal, this slows down monetary policy transmission, doesn’t it?
We have lowered the deposit facility rate by 125 basis points over the past eight months, and this has been transmitted smoothly to short-term market rates. We’ve also seen that bank lending rates have come down quite a bit – corporate loan rates by 92 basis points and mortgage rates by 64 basis points by December. This is significant. It tells you that transmission is working. When it comes to government bond yields, it’s important to look through the short-term volatility and take a somewhat longer perspective. And what you see then is that sovereign bond yields have remained rather stable. We had a strong repricing in 2022, when the ten-year Bund moved from negative territory at the end of 2021 to around 2.4% in October 2022. That is very close to the number that we’re seeing today. So we’ve been seeing a return of long-term sovereign bond yields to their new normal. We shouldn’t overstate the short-term volatility that we’ve experienced over the past weeks.
There’s another aspect that is quite important. One of the most interesting features of this tightening cycle is that it has not led to a comparable tightening of broader financial conditions. The exceptionally strong risk appetite of financial investors has even boosted equity prices and compressed credit spreads, and that has weakened monetary policy transmission. And part of that is due to the fact that we are still holding a very large monetary policy bond portfolio.
But overall, also taking into account the lags, monetary policy transmission is working fine.
Is the ECB’s “meeting-by-meeting” communication really credible? The ECB now says that the direction of travel is clear. Isn’t this a pre-commitment to further rate cuts?
I firmly believe in the meeting-by-meeting approach. The current time of high volatility is certainly not the time to tie our hands through forward guidance. And this is also what we stress in our monetary policy statements: we are not pre-committing to any particular rate path. At the time when it was still relatively clear that monetary policy was restrictive, one could infer the direction of travel from that. But this is no longer the case. And therefore, for me, the direction of travel is not so clear anymore.
Is this view shared by the majority of the Executive Board or the Governing Council?
It’s not for me to comment on that. It’s going back to the point that we now have to start the discussion on how far we should go. I’m not saying that we’re there yet. But we have to start the discussion.
If we take the meeting-by-meeting approach and data dependency as a given, does the type of data that has to be assessed need to change over time?
There are broadly two sets of data that we need to focus on. The first one refers to the inflation outlook: inflation itself, inflation expectations, wages, productivity, exchange rates. We use incoming data to cross-check the assumptions underlying our projections. This is why I never saw data dependence as a backward-looking concept. It was always forward-looking because we use incoming data to learn more about the credibility of our inflation outlook. The second set of data relates to the level of restrictiveness of monetary policy: interest rates, broader financial conditions, lending markets, the housing market as well as domestic demand, that is consumption, savings and investment. Of course, when we have a monetary policy meeting, we always look at all available data.
Can I challenge you on your claim that it was always forward-looking? At the time of high inflation, the ECB put a lot of emphasis on the actual inflation data from the previous month, which by definition is backward-looking. GDP numbers are by definition also very backward-looking.
I don’t agree. What do we learn from the current inflation data? We learn whether the transmission of our policy or of shocks is working as expected. High services inflation tells us something about its stickiness. If we spot deviations, we will eventually adjust our models but we also have to change our view about the medium-term outlook. So, in my view it was never backward-looking.
Data dependence is all the more important in today’s world. Some people say that the projections have become more credible. But who knows what’s going to happen as regards the trade conflict, the war in Ukraine and so on. We are faced with an unusual number of shocks, and that requires us to be always able to react. I don’t have a fixed mindset about what to do. Quite the opposite. I think we need to be able to adjust to whatever data or shock is coming in and what’s happening in the world and in the euro area economy.
What are the current data telling us about the inflation outlook?
Both services inflation and wage growth are still at an uncomfortably high level. Our projections foresee a deceleration of both. But this still needs to materialise. Services inflation has been stuck at around 4% since November 2023, and it still has to come down. For me, this is actually quite important. And therefore, the incoming data will be very relevant because our projections foresee a relatively quick deceleration of services inflation over this year.
How quickly do you want to see service inflation coming down?
It should start to come down in February. That’s what we expect. Over time, it does not necessarily have to come down to 2% but to a level that is consistent with our medium-term 2% target. Wage growth is also still high, but we have many indications that it is going to decelerate. For example, our wage tracker shows that wage growth is expected to drop steeply in the second half of the year. Part of that is due to a base effect from one-off payments. Hence, wage growth is expected to stay relatively elevated over the first half of the year. So we still need to see this deceleration. This is something that I pay a lot of attention to.
How concerned are you about recent swings in energy prices?
Energy and food prices can always offer surprises. We have seen some relatively strong moves in energy prices recently. Gas prices moved up a lot. That was mainly driven by cold temperatures. Very recently, gas prices dropped sharply. This seems to be driven partly by uncertainty about whether countries will fill up their gas storages as quickly as originally intended. A second reason is the debate about a potential ceasefire in Ukraine. This can cause a lot of volatility, which can have a strong impact on headline inflation and also on underlying inflation because energy serves as an input. We have to monitor this carefully.
What are the implications for monetary policy from energy price volatility? Is this deflationary or inflationary?
Recent volatility has been extreme. Before the recent fall in gas prices it was clearly inflationary. But now we have to see how that is going to play out. In general, I see risks to our inflation outlook as somewhat skewed to the upside. So I would not exclude that inflation comes back to 2% later than we had anticipated. But that remains to be seen.
The ECB this year will review its monetary strategy. President Lagarde has excluded the current inflation target from that review. Do you think that’s the right call?
Our symmetric, medium-term inflation target of 2% has served us very well in the high inflation period. So I really don’t see any reason to question it. And I believe there is strong support for this view in the Governing Council. What we have seen, however, is how quickly the inflation environment can change. And we have also learned how much people dislike inflation. But for me, that has implications primarily for the reaction function and not for the target. I think these two should be kept apart.
What are the potential implications for the reaction function?
The reaction function should be part of the debate. Back in 2021 during the previous strategy review, the discussion was very much under the impression of the low-for-long period. The main concern at the time was that our monetary policy was constrained by the effective lower bound on interest rates. When you read the monetary policy strategy statement today, you would think it comes from a different world. It focused on the risk of inflation being too low, and stated that we should be particularly forceful or persistent in such a scenario. But we have shifted to a new world. The past few years have shown that there are also risks of a de-anchoring of inflation expectations to the upside and that upside inflation risks can materialise quickly and become more persistent due to second-round effects. And therefore, I believe that the new reaction function should be symmetric in order to take into account the risks in both directions. This is especially true given that we are likely to face more adverse supply-side shocks going forward.
So effectively you are arguing in favour of a more hawkish reaction function?
I don’t like these notions of hawks and doves, and I don’t think that they are relevant here. My point is that our reaction function should acknowledge the fundamental shift of the macroeconomic environment. Up to 2021, we paid very little attention to upside risks to inflation. There was the perception that central banks would know precisely how to deal with a surge in inflation. But we’ve experienced that it has been quite difficult. Inflation has been above target now for almost four years. Looking forward, we should be putting equal weight on risks in both directions. And I wouldn’t call that a hawkish assertion.
Should the ECB toolkit be changed?
We’ve gained a lot of experience with the different tools. I do believe that all the tools we have should remain in our toolkit. But we’ve learned how important it is to carefully weigh the benefits and costs of our instruments – especially when it comes to asset purchases. They have proven very effective in stabilising markets. But as a monetary policy stance instrument, they have been less beneficial and costlier than we thought. This should be taken into account. The same applies to forward guidance. Many people believe that forward guidance led to a delayed response to the inflation surge. So forward guidance is another tool that we need to look at very carefully.
Are you implicitly saying that ECB should not have done as much quantitative easing as it did in the years up to 2021?
My point is that once we are back to a more normal world – a situation where inflation expectations are well anchored, and services inflation and unit labour cost growth have come down – and we are confident that we are sustainably back at our target, then we could become more tolerant of moderate deviations from our target. We should stop fine-tuning and responding to single data points. We should instead focus on large persistent shocks that give rise to a risk of a de-anchoring of inflation expectations in either direction.
So is your point that the ECB should be more willing to tolerate downward deviations to the 2% target in a steady state?
We should be more willing to tolerate both moderate downward and upward deviations, and act when there is a threat of de-anchoring.
But that’s an implicit change to the inflation target, is it not?
No, not at all. My point is that we should be less activist and rather take the time to assess whether shocks pose a serious risk to inflation expectations. Of course, we should keep in mind that the vulnerability of inflation expectations may have changed after the recent inflation experience. People have learned that inflation can increase sharply and that this is very harmful. Firms have learned that they can reprice relatively quickly, and we have to take this into account.
Finally, we need to think about how to deal with the uncertainty around our economic and inflation outlook. For me, the most useful way to deal with that is to make greater use of scenario analysis – and in a different way than we’ve done over the past years. Back then we were looking at tail risks, which was very useful. But in the future, we should also look at plausible alternative scenarios in order to get away from the illusion of precision that we create by just focusing on the baseline point estimate. We all know there is a lot of uncertainty around it. So I think it would be important to also look at plausible alternative scenarios to illustrate this uncertainty.
