Is it just a phase? The monetary policy response to supply shocks

Thank you. Let me start with something a bit more personal. I don’t know how many of you have children, but I do. Three of them. They’re 16, 18, and 21. And if there’s one thing being a parent teaches you, it’s this: letting go. Again and again.

It’s a bit like sitting on your hands sometimes. You watch, you wait, you let them figure things out for themselves. You give them trust, and you hope for the best. Most of the time, that works.

A few weeks ago, we picked up my 18-year-old son from the airport. He had just come back from a holiday with friends, right after his exams. And as you can probably imagine, it had been… that kind of holiday. He looked completely exhausted.

Now, I didn’t ask too many questions. As a parent, you learn not to. You trust that they’re old enough to make sensible choices. Most of the time, at least.

But every now and then, you must step in. And that’s the tricky part: knowing when to let go, and when to intervene.

To me, that judgment call is one of the hardest things in parenting. And in a way, it’s not so different from what central banks do. Because in monetary policy, we face a similar question: do we look through what’s happening or do we step in?

And today, I’d like to talk about that judgment. About how central banks should respond to supply shocks, and when doing nothing is, in fact, not an option.

As you know, the ECB has a single mandate, which is to maintain price stability. Nevertheless, our strategy recognizes that achieving this objective over the medium term helps to avoid unnecessary volatility in output and employment, also in view of the lags with which monetary policy operates. This is particularly relevant in the case of short-lived supply shocks, that push output and inflation in opposite directions. A forceful response to such a temporary shock may only affect inflation after the shock has faded, risking an inflation undershoot while adding to an unnecessary slowdown in activity.

For this reason, the textbook prescription for monetary policy is to look through small and temporary supply shocks. There are, however, limits to looking through. The recent inflation surge in the euro area illustrates that when a shock becomes more persistent and broad-based, a swift pivot of the policy stance may be necessary. The important question, therefore, is when is looking through no longer appropriate? When do we have to intervene?

In the latest strategy review, the Eurosystem concluded that the appropriate response to a deviation of inflation from its target is context-specific.

In my speech, I want to pick up on three factors that shape the optimal policy response to supply shocks, and which I think are relevant for our current situation. In particular, I will discuss how the policy response depends on the persistence of the shock, how it feeds into inflation, and whether it is amplified by the state of the economy.

Do you recognize that moment when your kids suddenly start to change in ways you’re not entirely comfortable with? Take something simple. One day your son is happily building little worlds in Lego… and the next, he’s completely absorbed in a shooting game like Call of Duty. And you find yourself thinking: what happened to Lego?

At first, you shrug. You think: fine, it’s just a phase. He’ll get bored soon enough. But then you notice something. He starts spending more and more time behind that screen. What seemed like a passing whim begins to stick. And you catch yourself thinking: hmm… this is becoming a bit more persistent than I’d hoped.

In the economy, it’s the same. Supply shocks can be persistent. How should monetary policy react? I will start with the extreme case: a very persistent, or even structural, supply shock, like demographic change.

According to Eurostat’s population projections, the EU’s working age population is expected to decline by about 4 percentage points in the coming decade as baby boomers gradually retire.

My colleagues at DNB modeled the macro-economic impact of such a structural supply shock depending on different responses by the central bank. The shock is calibrated to deliver a 4 percentage point decline of the labor force by 2035.

In the model this leads to a fall in GDP, reflecting lower labor income as well as weaker investment. Inflation initially declines due to an excess capital stock relative to labour, but soon picks up when wages increase. 

In the baseline scenario, here depicted in blue, the central bank follows a standard policy rule. It responds to lower inflation and output by temporarily lowering the policy rate. But in the longer term it increases policy rates when inflationary pressures start to mount.

In the alternative scenario, the central bank decides to look through by leaving rates unchanged. As you can see from the chart, this comes at the cost of higher inflation volatility, while the benefits in terms of output are negligible.

What’s important here is that the demographic shock is structural: potential output declines. This, in turn, implies that the contraction in activity is efficient. Consequently, monetary policy cannot offset this decline without generating excessive inflation.

For monetary policy, this means that the case for looking through to protect output weakens as shocks become more persistent. In other words, if it’s Call of Duty all day, perhaps it’s time to switch-off the Wifi.

The problem is, of course, that the persistence of a shock is typically unknown beforehand. The question is then: what information does the central bank have to assess the risk that a shock is not short-lived?  

Sometimes, unfortunately, there is little we can do. Luckily, in the case of the current energy shock, financial markets provide us with an estimate. And while it’s true that energy prices are extremely difficult to forecast, option prices may provide us with some information about the market-perceived risks.

As you can see on the slide, even though oil prices have recently come down in response to the prospect of further negotiations, and are expected to decline towards the end of the year, the risks around this expectation are clearly tilted to the upside. This tells us that scenarios where energy prices stay higher for longer should not be discounted.

Practically, the Eurosystem deals with this uncertainty by using scenarios. Our most recent projections include a severe and adverse scenario regarding the development of energy prices, as well as a milder one. These scenarios, together with the Governing Council’s data-dependent approach and readiness to act, serve to make our policy choices more robust against uncertainty. 

But understanding the nature of the shock itself is only part of the story. What ultimately matters for inflation, and thus for monetary policy, is how such shocks propagate through the economy.

Suppose your daughter suddenly becomes close friends with someone who’s really into the tattoo scene. Of course, there’s nothing wrong with that. Friendships at that age come and go. But if she comes home one day with a beautifully done sleeve, the result is rather permanent.

Translated to our world of economics: even temporary shocks can generate persistent effects. This depends on how they feed through the production chain.

In a recently published analysis, my staff analyses exactly this. Using the New York Fed’s Global Supply Chain Pressure Index, they estimate the unexpected changes to the index, here plotted in yellow. Looking at the very last observations, you can see a sizeable and clearly positive shock.

The study then shows that these global supply chain shocks have a positive and persistent effect on core inflation. When we look at the response of the different sub-components of HICP on the right-hand side, we find that HICP energy responds most strongly. This reflects the high reliance of euro area energy consumption on imports and global supply chains. At the other end of the spectrum, we find prices of services to respond relatively mildly to the shock, reflecting that services are less exposed to frictions in global supply chains. Importantly, as you can see on the left-hand side, in all cases, the response is hump shaped with the peak effect only coming months after the initial shock.

While these results are based on broader supply chain disruptions, the underlying mechanisms are relevant for energy price shocks as well: even in a benign scenario where energy prices return to normal relatively quickly, their impact on broader inflation may last considerably longer. Such a delayed, but more persistent response may provide another limit to the extent to which monetary policy can safely look through such shocks.

In fact, last week’s decision by the Governing Council to raise the ECB’s key interest rates by 25 basis points was also underpinned by the assessment that higher energy prices are to some extent feeding into broader inflation and the risk that these spill-overs could be higher than expected.

I have so far focused on the persistence of shocks and their transmission to inflation. I will now turn to a third dimension, namely the extent to which these effects are amplified in a state-dependent manner.

From about the age of twelve, children start to change enormously. I have seen that happening with my own kids. They move to a new school, their interests begin to change, their bodies change, and so do their brains. It’s a period full of uncertainty. They’re still figuring out who they are and where they belong, and that can bring quite a bit of stress.

Now imagine something dramatic happens during that already vulnerable phase. That stress can deepen, turning into anxiety or even depression, conditions that can have a long-lasting impact.

That’s exactly why young people were hit particularly hard during the Covid pandemic. It wasn’t just the shock itself, but when it hit.

Just as with stress, when inflation is already high, the impact of supply shocks can be significantly stronger.

One key reason is that pricing behaviour changes in such environments. When inflation is low and stable, firms typically don’t adjust their prices very often. In contrast, when inflation rises, price adjustments become more frequent.

The magnitude of the shock plays an important role in this context, too. While a small shock might be absorbed by profits and wages and passed on only gradually, larger shocks are more likely to trigger immediate and widespread price adjustments.

This is consistent with evidence from the euro area. Eurosystem research documented substantial shifts in the frequency and nature of price adjustments in the euro area during the inflation surge of 2021–2023.

Now, you might know that the frequency of price adjustments plays an important role in modern macroeconomic models. Usually, however, it is treated as being constant over time. Together with our colleagues from the OeNB, we have embedded this state-dependent pricing into an otherwise standard New Keynesian model and estimated the model on data until the fourth quarter of 2024. The results, which you can see on my slide, show that for a given supply shock, the effects on inflation and output are much higher, when inflation is already high. In other words, nonlinearities in the price setting amplify the initial shock.

Even though this is a stylized example, the message for policymakers is that the linear models that we often use may underestimate the amplification of such shocks. This is particularly relevant in the current environment, where inflation has only recently declined from elevated levels.

But these nonlinearities in the price setting are a boon and bane for monetary policy. On the one hand, a higher frequency of price adjustments implies a faster passthrough from higher input costs on to consumers prices, leading inflation to accelerate much more rapidly.

On the other hand, more flexible prices also lead to a steepening of the Phillips Curve. This implies a lower sacrifice ratio, as you can see in the slide, making it less costly in terms of output to reduce inflation.

In other words: the risks of an acceleration of inflation are higher, but the tools to push back are also more powerful.   

I am approaching the end of my remarks. As I have tried to argue, central banks are sometimes a bit like parents. When supply shocks hit, we have to make that judgement call: is it just a phase, or is it something more worrying and do I have to step in? Today, I have discussed three factors that shape, and ultimately limit, the extent to which central banks can look through supply shocks: the persistence of the shock itself, the extent to which it feeds into broader inflation dynamics, and the risk that these effects become amplified in a state-dependent manner.

What does that mean for policy in the current situation?

As you know, last week, the Governing Council decided to raise the three key ECB interest rates by 25 basis points.

This decision reflects the assessment that upside risks to medium-term inflation have increased. Specifically, incoming information suggests that the recent energy shock may prove more persistent than previously expected and that its effects are increasingly feeding through to broader inflation.

Compared with the March projections, energy prices, and oil prices in particular, have been revised upwards across the projection horizon. This implies that energy-related price pressures are likely to remain present for longer.

At the same time, core inflation has also been revised upwards and is expected to continue increasing gradually over the coming quarters. Its peak is only expected for early next year. This suggests that the effects of the shock are increasingly transmitting through the broader economy.

Finally, also the recent past matters. Research shows that lived experiences of inflation can have lasting effects on how people form expectations, and how the pass through to inflation is higher when inflation is already elevated.

Taken together, these developments underpinned the case for a measured adjustment of the monetary policy stance. At the same time, the Governing Council remains data dependent and stands ready to adjust the policy rate to deliver 2% inflation over the medium term.

And with that, let me wrap up. But let me reassure you. I wouldn’t want to leave you with the impression that the ECB’s Governing Council resembles a kind of parenting support group for worried mums and dads. There are, of course, some important differences between central bankers and parents… if only because children generally don’t listen to speeches. At least, my kids certainly don’t.

So all the more reason for me to appreciate the fact that you have stayed with me. Thank you very much for your attention.