Christine Lagarde: Building confidence in the path ahead

Since the pandemic, monetary policymakers have been facing an exceptionally complex environment. As inflation rose, we were confronted with profound uncertainty about how far it would go and how widely it would spread across the economy. And even as inflation has eased, uncertainty about its persistence has remained.

The potential costs of mis-calibrating policy have been high, which is why we had to employ a policy framework that minimises the risk of mistakes. And we have done so by building our reaction function around three criteria: the inflation outlook, the dynamics of underlying inflation and the strength of monetary transmission.

Though we conceived these criteria when we had low visibility of future inflation, they have also helped guide our decisions as inflation has fallen and forecasts have become more accurate.

As Marie Curie once said, to thrive through the ups and down of life, "we must have perseverance and above all confidence". And our framework has indeed encouraged us to persevere when necessary and to build up confidence when needed.

It has served as a reliable compass for calibrating policy through three phases of our current policy cycle.

First, it helped create robustness during our tightening phase when we were devising how far we needed to go to rein in inflation.

Second, it has helped us practice patience during the holding phase until the signals from our inflation projections and underlying inflation are more consistent.

Third, it will support us in building up sufficient confidence to begin the dialling-back phase in which we make policy less restrictive.

The tightening phase

In the early phase of our tightening cycle, our main priority amid surging inflation rates was to exit our accommodative policy stance as quickly as possible. While the policy challenge was immense, the policy path was relatively simple to calibrate.

But as rates rose and approached restrictive territory, calibrating our policy stance became more complex. We first had to assess how much rates needed to rise until they were sufficiently restrictive, and then for how long they needed to stay at that level. But our assessment was blurred by much lower than normal visibility of the future.

Our forecasts repeatedly underpredicted inflation by large margins, even at shorter horizons. From 2021 to 2022 for example, the absolute inflation forecast errors in the staff macroeconomic projections, one quarter ahead, more than doubled, largely owing to volatile energy prices.¹

At the same time, the mix of shocks that emerged from the pandemic and its aftermath – rotations in spending, energy spikes, "bullwhip" cycles in manufacturing, supply bottlenecks, tight labour markets, fiscal expansion and reopening effects – heightened the risk of inflation becoming more persistent.

We faced a highly unusual conjuncture of high inflation and declining real wages, but also rising employment. This combination essentially implied a multi-year catch-up process to make up for real wage losses. In turn, this process could have triggered what I referred to at last year's conference as a "tit-for-tat" inflation dynamic.²

And we faced uncertainty as to how quickly and forcefully our monetary policy response would succeed in bringing down inflation. The ECB had not been through a tightening cycle for more than a decade, and there were reasons to believe that the transmission of monetary policy to firms and households might have changed.³

So, to calibrate policy accurately, we needed a framework for policy decisions that would work when we had low visibility and would mitigate heightened uncertainty. This is why we built our monetary policy response around the three criteria I referred to earlier: the inflation outlook, the dynamics of underlying inflation and the strength of monetary transmission.

This approach made our decisions more robust, as the inflation path we foresaw in our projections had to be validated by data we could observe in real time and extrapolate into the medium term. That, in turn, enabled us to take forward-looking decisions with a higher degree of confidence.

And it served us well in practice.

The three criteria helped us to map out the remaining climb, allowing us to bring rates to sufficiently restrictive levels to break the persistence of inflation.⁴ But also, by guiding us to carefully evaluate the strength of policy transmission, they acted as a cross-check against overtightening. This helped us reach the decision to stop rate hikes after last September.

The holding phase

We then entered the current phase of our policy cycle – the holding phase – during which we committed to keep rates at restrictive levels for as long as necessary.

Since the start of this phase, inflation has been declining consistently and our projections have been showing inflation returning to our target over the medium term. We now project inflation to average 2.3% in 2024, which is 0.4 percentage points less than projected in December and 0.9 percentage points less than September. We then expect inflation to decline to 2.0% in 2025 and 1.9% in 2026.

And unlike in the earlier phases of our policy cycle, there are reasons to believe that the expected disinflationary path will continue.

First, for some time now inflation outturns have been broadly in line with our expectations. In 2023 we saw a reduction of about 70% in the average absolute error in our staff projections relative to 2022, one quarter ahead.

Second, we now see inflation returning to 2% earlier in our projection horizon than before, in mid-2025, and not exceeding our target for the remainder of the horizon.

Third, the composition of inflation is improving, as we now expect lower core inflation in the medium term. This suggests that the convergence to 2% is likely to be more durable and less beholden to assumptions about commodity prices, although the latter can always prove hazardous.

The other criteria are also becoming more consistent with this improved inflation outlook.

The transmission of our monetary policy is unfolding in the right direction. Financing conditions have reacted strongly to higher rates, loan demand has weakened and, in turn, activity has slowed notably in the most interest-sensitive sectors of the economy.

And underlying inflation is generally easing. Nearly all the measures that we track are declining, and the range of readings between the different measures has narrowed from 4.1 percentage points at its peak to 2.4 percentage points today. Some of the measures of underlying inflation with the best leading indicator properties for future inflation have dropped steeply.⁵

But, at the same time, domestic price pressures remain strong.

Services inflation is still stubborn and hovering around 4%, while momentum increased somewhat in February. And our indicator of domestic inflation, which measures items with a low import content, stands at 4.5%, at the top of the range of underlying inflation measures that we monitor. This measure has also been found to have good leading indicator properties.⁶

These pressures largely reflect robust wage growth as the catch-up process continues, as well as a tight labour market that has so far been resilient to a slowing economy. Employment grew by two million cumulatively during 2023, even as the economy stagnated, while firms continue to hoard labour. This pattern is mechanically lowering labour productivity and pushing up unit labour costs.

At this stage, it is difficult to assess whether these price pressures simply reflect the lag in wages and services prices and the procyclical nature of productivity, or whether they signal persistent inflationary pressures.

So, although we have made significant progress in all three of our framework criteria, we are not yet sufficiently confident that we are on a sustainable path towards our inflation target.

Building sufficient confidence to dial back policy

So the essential question is: what do we need to see to become sufficiently confident to start dialling back our restrictive policy stance?

Put simply, we need to move further along the disinflationary path. And there are three domestic factors that will be decisive to ensuring that the inflation path evolves as we project.

The first of these is wage growth.

Our forecast sees nominal wages slowing to 3% over the next three years, allowing real wages to fully catch up to pre-pandemic levels over the projection horizon, also including productivity gains.⁷ But with the unemployment rate expected to remain very low at 6.6%, this wage path cannot be taken for granted. Sensitivity analysis by ECB staff shows that if there were an earlier full catch-up by the end of this year, inflation would rise to 3% in 2025 and only fall to 2.5% in 2026.⁸

The second is profit margins.

The compression of profit margins has allowed wages to catch up without further accelerating inflation. Unit profits accounted for more than 50% of the GDP deflator in the last quarter of 2022 but this figure fell to just 20% a year later. But our sensitivity analysis shows that, if firms were to regain pricing power as the economy recovers and profit margins were to rise by an accumulated 1 percentage point more than we project until the end of 2026, inflation would be 2.7% in 2025 and 2.4% in 2026.

The third factor is productivity growth.

We expect that a pick-up in demand, if accommodated by fully utilising hoarded labour, will lead to rising productivity growth and falling unit labour costs. We project labour productivity growth of 0.1% this year before it rises to 1.2% in 2025 and 2026. But the path of inflation could be different if, in a new geopolitical environment, productivity losses for European firms turn out to be partly structural.

Given the delays with which these data become available, we cannot wait until we have all the relevant information. To do so could risk being too late in adjusting policy. But in the coming months, we expect to have two important pieces of evidence that could raise our confidence level sufficiently for a first policy move.

First, we will have more data to confirm whether wages are indeed growing in a way that is compatible with inflation reaching our target sustainably by mid-2025.

The latest data point in this direction. Growth in compensation per employee edged down to 4.6% in the fourth quarter of last year – slightly below our March projection – from 5.1% in the third quarter. Negotiated wage growth, which accounts for the lion's share of compensation per employee growth, also decreased from 4.7% to 4.5% in the fourth quarter.

Similarly, the ECB's forward-looking wage tracker, which anticipates the development of negotiated wage growth in the euro area, is showing early signs that pressure is easing. Average wage growth in 2024 for all existing wage contracts⁹ fell from 4.4% at the time of our January Governing Council meeting to 4.2% at the time of our meeting in March.

The coming months will help us form an even clearer picture.

We will receive data on negotiated wage growth in the first quarter of this year at the end of May. And many wage negotiations are currently taking place in large sectors, the outcome of which will be entered into our wage tracker as soon as the negotiations are concluded. Employees whose contracts ran out last year and have not been renewed, or will run out by March 2024, account for around one-third of those in our wage tracker.

Second, by June we will have a new set of projections that will confirm whether the inflation path we foresaw in our March forecast remains valid. These projections will also implicitly give us more insight into the path of underlying inflation. We will have more visibility on the strength of the recovery and the likely direction of the labour market, and therefore on the consequences for wages, profits and productivity.

In addition, we will have had a longer window to assess whether inflation data continue to fall broadly in line with our projections. If they do, we can be more confident that our models are now better accurately capturing inflation dynamics. And this confirmation will be particularly important for the more persistent components, such as services, so that we can trust these components will continue to decline in keeping with their typical lagging pattern.

If these data reveal a sufficient degree of alignment between the path of underlying inflation and our projections, and assuming transmission remains strong, we will be able to move into the dialling back phase of our policy cycle and make policy less restrictive.

But thereafter, domestic price pressures will still be visible. We expect services inflation, for example, to remain elevated for most of this year. So, there will be a period ahead where we need to confirm on an ongoing basis that the incoming data supports our inflation outlook.

This has two important implications for the policy path ahead.

First, our decisions will have to remain data dependent and meeting-by-meeting, responding to new information as it comes in. This implies that, even after the first rate cut, we cannot pre-commit to a particular rate path.

Second, our policy framework will remain important to process the incoming data and calibrate the appropriate policy stance. At the same time, the relative weights assigned to the three criteria will have to be regularly examined.

Conclusion

Let me conclude.

I said after our last Governing Council meeting that, when it comes to the data that is relevant for our policy decisions, we will know a bit more by April and a lot more by June. I hope that my remarks today help you to better understand our analysis and logic.

In the coming months, we will receive more data, which will help us to assess whether we are sufficiently confident in the path ahead to move to the next phase of our policy cycle.

---

¹ Chahad, M., Hofmann-Drahonsky, A.-C., Meunier, B., Page, A. and Tirpák, M. (2022), "What explains recent errors in the inflation projections of Eurosystem and ECB staff?", Economic Bulletin, Issue 3, ECB; Chahad, M., Hofmann-Drahonsky, A.-C., Page, A. and Tirpák, M. (2023), "An updated assessment of short-term inflation projections by Eurosystem and ECB staff", Economic Bulletin, Issue 1, ECB; Chahad, M., Hofmann-Drahonsky, A.-C., Martínez Hernández, C. and Page, A.(2024), "An update on the accuracy of recent Eurosystem/ECB staff projections for short-term inflation", Economic Bulletin, Issue 2, ECB, forthcoming (21 March).

² Lagarde, C. (2023), "The path ahead", speech at "The ECB and Its Watchers XXIII" conference, Frankfurt am Main, 22 March.

³ ibid.

⁴ Lagarde, C. (2023), "Breaking the persistence of inflation", speech at the ECB Forum on Central Banking 2023 on "Macroeconomic stabilisation in a volatile inflation environment" in Sintra, Portugal, 27 June.

⁵ Bańbura, M., Bobeica, E., Bodnár, K., Fagandini, B., Healy, P. and Paredes, J. (2023), "Underlying inflation measures: an analytical guide for the euro area", Economic Bulletin, Issue 5, ECB.

⁶ ibid.

⁷ This exercise is based on compensation per employee deflated by HICP and productivity growth using a four-quarter moving average. Conducting the same exercise without considering productivity, the catch-up would take place in mid-2025.

⁸ The analysis is based on Arce, O., Ciccarelli, M., Kornprobst, A. and Montes-Galdón, C. (2024), "What caused the euro area post-pandemic inflation? An application of Bernanke and Blanchard (2023)", Occasional Paper Series, No 343, ECB.

⁹ Including one-offs.