伊莎贝尔·施纳贝尔：在风暴中航行：货币政策反思

Navigating stormy waters: reflections on monetary policy

Speech by Isabel Schnabel, Member of the Executive Board of the ECB, at the ECB Conference on Money Markets 2026

Frankfurt am Main, 5 November 2026

The past seven years have been an extraordinary journey.[1]

We have moved from a world in which inflation was persistently too low to one marked by recurrent inflationary shocks; from a historically accommodative monetary policy stance to the fastest tightening cycle in the history of the ECB; and from a deeply integrated global economy to one increasingly shaped by fragmentation and geopolitical tensions.

As I reach the end of my journey here at the ECB, I would like to reflect on the lessons these experiences have taught me about the conduct of monetary policy in a world that has become less predictable.

My remarks today are personal reflections on how the experience of recent years has shaped my thinking about monetary policy and what it may teach us about navigating the challenges ahead.

There are four key takeaways.

Takeaway one: preserve room for manoeuvre by avoiding commitments

The first takeaway is that central banks should prioritise instruments that allow them to adjust monetary policy quickly as conditions change.

We entered the pandemic after a long period in which inflation had been below our target (Slide 2, left-hand side). Globalisation, and the integration of China in particular, had structurally lowered the global cost of producing goods, putting downward pressure on inflation.

At the same time, the global financial crisis and the sovereign debt crisis had weakened public and private balance sheets and left lasting scars on demand, weighing on economic activity and inflation.

Over time, this experience gradually fostered the belief that these forces were permanent, leaving monetary policy primarily focused on bringing inflation back up to target.

As a result, after chronically overpredicting inflation before the pandemic, we were slow to recognise that we had entered a new regime in which we would repeatedly underpredict it (Slide 2, right-hand side).

Indeed, even as evidence of a more substantial and persistent rise in inflation was mounting, the initial reaction was one of relief: the economy finally seemed to be escaping the low-inflation environment.[2]

But the rapid increase in the share of goods and services experiencing inflation above 2% signalled that price pressures were becoming increasingly broad-based (Slide 3).

This exposed two key features of the inflation process that our analytical frameworks had not sufficiently accounted for.

One was the steepening of the Phillips curve, which was in part a result of more frequent price changes as firms sought to protect their profit margins in the face of sharp cost increases (Slide 4).

The other, related feature was the role of firms’ profits in inflation. Unlike the high-inflation period of the 1970s, when inflation was almost entirely driven by growth in unit labour costs, the post-pandemic inflation surge was to a significant extent driven by rising unit profits (Slide 5).[3]

So, the pandemic was a structural break in many of the trends we had witnessed before.

This included a shift in the global balance between savings and investment.

Rising investment needs related to the green and digital transitions, alongside a changing geopolitical environment, have turned the savings glut of the past into a bond glut, as both private and public issuance have increased sharply.[4]

This shift puts upward pressure not only on inflation but also on the natural rate of interest, r* (Slide 6, left-hand side).[5]

None of these shocks or their consequences could have been predicted easily. The lesson is therefore that we should expect the unexpected – and avoid becoming too attached to a particular view of the world or a particular class of models.

This has implications for the design and conduct of monetary policy.[6]

We should be wary of commitments that depend too heavily on a particular economic environment or class of models, leaving little room to adjust when conditions change.

Charles Goodhart – a distinguished scholar of monetary history who has seen the world change more than once – put it this way: “The world changes, and if the world changes you need sufficient flexibility to change the way that you operate within that.”[7]

In retrospect, forward guidance during the pandemic was too Odyssean. It tied our hands too closely to a world of persistently low inflation.

The same applies to the choice of other policy instruments.

Some instruments are inherently more agile than others. Policy rates can be adjusted swiftly at each meeting, whereas large-scale asset purchases continue to provide monetary accommodation long after the purchases themselves have ended (Slide 6, right-hand side).

An instrument that is difficult to reverse is itself a form of commitment.

That is also why we emphasise in our monetary policy strategy that our primary monetary policy instrument is the set of key policy rates.[8] This does not mean that asset purchases are never useful.[9] They are highly effective in stabilising financial markets that suddenly turn illiquid. And they remain a tool to counter deflationary pressures at the effective lower bound.

One way to preserve flexibility is to separate instruments used to stabilise markets from those used to calibrate the monetary policy stance.[10] This can limit the scale of asset purchases and make them easier to unwind.

The Transmission Protection Instrument (TPI) embodies this principle. It provides a dedicated tool to address unwarranted, disorderly market dynamics that threaten the transmission of monetary policy, subject to clearly defined criteria.

Takeaway two: what matters is how shocks affect the inflation outlook

The second takeaway is that the appropriate monetary policy response to a shock primarily depends not on its nature, but on its implications for the projected inflation path.[11]

Traditionally, the nature of the shock guided our thinking about how monetary policy should respond to price pressures. The standard prescription was that demand shocks called for an immediate response, while supply shocks could often be looked through, as the shock would either fade on its own or weaker growth would dampen emerging price pressures, reducing the need for monetary tightening.

This distinction can be misleading, for two reasons.

First, central banks rarely face one shock at a time. During the pandemic, supply chain disruptions, pent-up demand and unprecedented fiscal and monetary policy support interacted in ways that were difficult to disentangle (Slide 7, left-hand side).[12]

Russia’s unjustified invasion of Ukraine added a major energy and food price shock to an already challenging environment. Such overlapping shocks can prolong the period of elevated inflation and make the “last mile” of disinflation particularly challenging.[13]

Services inflation was still running above 3% before the start of the conflict in the Middle East, reflecting the drawn-out adjustment to the earlier inflation surge, including through staggered wage negotiations (Slide 7, right-hand side).

Today, we again face overlapping shocks, from geopolitical tensions and rising energy prices to the rapid expansion of AI-related investment and large fiscal spending needs, including for defence. These developments affect demand, productive capacity, prices and financial conditions through different channels and over different horizons.

Second, a sufficiently large and persistent adverse supply shock can push projected inflation above target, requiring a timely monetary policy response. What begins as a relative price adjustment can turn into a broader inflation problem through indirect price effects, wage dynamics and changes in inflation expectations.

This is why we are closely monitoring these developments. In response to the latest ene