![Logo](https://i0.wp.com/agenparl.eu/wp-content/plugins/Plugin%20Logo/Agenparl.png?w=788&ssl=1)
(AGENPARL) – mer 26 giugno 2024 Monetary policy after a perfect storm: festina lente
Speech by Fabio Panetta
Governor of the Bank of Italy
3rd Bank of Finland International Monetary Policy Conference on
‘Monetary Policy in Low and High Inflation Environments’1
Helsinki, 26 June 2024
Good afternoon ladies and gentlemen. I would like to thank the Bank of Finland for
organizing this conference, and Governor Rehn for inviting me. It is a great pleasure to
be here. Finland has been crowned the happiest country in the world for the seventh
consecutive year in the recently published World Happiness Report:2 I am really happy to
be part of that happiness.
We are at a turning point in the European monetary policy cycle. In June, the Governing
Council decided to cut the key ECB interest rates by 25 basis points after holding them
at high levels for nine months. Turning points inevitably attract attention, and the policy
debate is now focused on the next steps: ECB watchers are eagerly looking for clues
about the pace of the normalization path.
I will briefly review what has happened to inflation over the past three years. I will then
discuss some of the remaining concerns about the disinflation process, namely price
developments in the services sector and the role of profits, input costs and labour costs.
After that, I will examine the outlook for inflation and growth from a forward-looking
perspective. I will outline my thoughts on the evolution of monetary policy before
concluding.
My main message is that the disinflation process requires two ingredients. We must
keep adjusting policy rates to complete the process and smoothly reach our 2 per cent
inflation target in a scenario where our expectations turn out to be broadly correct. But
I would like to thank Piergiorgio Alessandri, Marco Bernardini, Francesco Corsello, Alessandro Notarpietro,
Marianna Riggi, Alessandro Secchi, Fabrizio Venditti and Eliana Viviano for their contributions and
comments. Festina lente – “hasten slowly” – is the Latin motto that epitomised a decisive but calm policy
action for Augustus, the first emperor of Rome. The motto was also used in XVI-century Florence by
Cosimo I De’ Medici, whose fleet aptly adopted as a symbol a turtle endowed with sails.
Helliwell J.F., R. Layard, J.D. Sachs, J.-E. De Neve, L.B. Aknin and S. Wang (eds.), 2024, ‘World Happiness
Report’, University of Oxford, Wellbeing Research Centre.
we also need to be prudent, and stand ready to adjust our policy in response to shocks
that may create upside or downside risks to the current outlook.
Genesis and unfolding of a perfect storm
After years of low inflation, the euro area experienced an exceptional rise in prices between
2021 and 2023 (Figure 1). Supply constraints, the rebound in consumption and soaring
commodity and energy prices created a perfect storm for inflation. The nature of the storm
has been debated in countless papers and opinion pieces, and I will not systematically
review it again today.3 There are just a few points I would like to emphasize.
Figure 1
Euro area headline inflation decomposition
(monthly data; per cent and percentage points)
Headline y−o−y
Energy
NEIGs
Services
Source: calculations based on Eurostat data.
Notes: the figure reports the year-on-year percentage changes of the euro-zone HICP, decomposed into contributions of its components.
The first one is that the conditions for the storm took shape in 2020, when inflation was
close to zero. The pandemic disrupted supply chains and caused a structural shift in
consumption patterns – including a reallocation from services to goods – that proved to
be highly consequential in the following two years.
The second point is that, paradoxically, the recovery in the US economy complicated
the picture in the euro area. It was fast and strong, underpinned by a generous fiscal
stimulus and the excess savings accumulated during the pandemic lockdowns, and it
initially involved more goods than services. This led to large swings in relative prices,
pushing up headline inflation;4 it also strained supply chains and increased commodity
and energy prices on international markets. In a twist of fate, a positive demand shock
in the US turned into an additional negative supply shock for Europe.
See e.g. Panetta F., 2023, ‘Everything everywhere all at once – responding to multiple global shocks’,
speech at ‘The ECB and its Watchers XXIII’ conference, Frankfurt am Main, 22 March.
Bernanke B., and O. Blanchard, 2023, ‘What Caused the U.S. Pandemic-Era Inflation?’, Hutchins Center
Working Paper 86, June.
The Russian invasion of Ukraine in February 2022 radically changed the inflation outlook.
On the eve of the invasion, inflation was hovering around 5 per cent, and Eurosystem
economists, analysts and investors broadly agreed that price pressures would recede
relatively quickly. In the months that followed, an extraordinary series of energy and
commodity price shocks doubled headline inflation, unsettling markets and central banks
alike.5 In short, we entered into a different world.6
2. Should we (still) worry about services and wages?
In response to the rise in inflation, in July 2022 the ECB started an unprecedented series of
rate hikes. This decisive response paid off: supply shocks, bottlenecks and soaring energy
prices have not triggered an inflationary spiral, as they had in the past, and inflation
expectations remained firmly anchored. The disinflation process is now well under way.
Inflation fell from a peak of 10.6 per cent in October 2022 to 2.6 per cent in May, and the
speed of the decline was unprecedented, as the previous increase.
However, inflation has not yet returned to target. Some commentators have pointed
to price dynamics in the services sector and the labour market as possible obstacles
in the ‘last mile’ of the price stabilization efforts. We should certainly remain vigilant
on developments in these areas, but a closer look at the data suggests that the ‘last
mile’ may require nothing more than patience and a careful monitoring of the ongoing
disinflation process.
2.1. The rise and fall of inflation in the services sector
The services sector has attracted attention for two related reasons. First, services inflation
seems to be more persistent than goods inflation. Second, services price growth remains
above the ECB target. Let me discuss these points in turn.
The persistence is only apparent. It reflects the fact that inflation in the services sector
started to rise later, peaked later and started to fall later. In turn, this delayed rise and
fall depends on the combination of two forces.
One is purely coincidental and specifically related to the pandemic episode. Initially,
the recovery in consumption was mainly driven by demand for goods, in particular
the durable goods that consumers were unable to buy during the first stage of the
The shocks had a particularly severe impact on Europe, a net importer of most commodities –
unfortunately including Russian gas. The exorbitant energy bills also left European households with
far less money to spend on other goods and services than their American counterparts.
As this rough summary suggests, and as I have noted on previous occasions, the euro area faced a
wave of particularly ‘bad’ inflation. The ‘good’ type is driven by buoyant domestic demand and wage
growth consistent with the target; monetary policy should nurture it or mitigate it (depending on the
initial level) until the target is reached. The ‘bad’ type reflects negative supply shocks that raise prices
and depress economic activity; monetary policy should look through it. There is also an ‘ugly’ type
of inflation driven by a de-anchoring of inflation expectations; monetary policy should immediately
stamp it out. See Panetta F., 2021, ‘Patient monetary policy amid a rocky recovery’, speech delivered at
Sciences Po, Paris, 24 November.
lockdowns. The consumption of goods recovered quickly in the euro area; it grew
even faster in the US, where it exceeded pre-pandemic levels. Services consumption
recovered later and more slowly, especially in the euro area (Figure 2).
Figure 2
Diverging consumption paths of goods and services
(quarterly data; index numbers, Q4 2019=100)
(a) United States
(b) euro area
Goods
Services
Source: calculations based on US Bureau of Economic Analysis and Eurostat data.
Notes: panel (b) is based on euro area countries for which data are available.
The second force is structural. Goods and services react differently to exogenous cost-push
shocks because they are characterized by different price adjustment mechanisms. In
particular, price adjustments are slower for services. One reason for this is that many
services are provided directly by the public sector or are subject to administrative and
contractual frictions.7
Consider energy price shocks. Estimates based on historical data and standard
econometric models show that, following an energy shock, non-energy industrial goods
(NEIGs) inflation peaks in about one year and returns to equilibrium within two years;
instead, services inflation peaks about six months later and normalizes after almost
three years (Figure 3a). In terms of relative magnitude and timing, these estimates
are virtually mirror images of the inflation patterns actually observed between 2022
and 2024 (Figure 3b). This remarkable coincidence corroborates the hypothesis that
the asymmetric propagation of energy shocks explains a good part of the ‘puzzling’
divergence between the two sectors.
The second concern is that services inflation remains higher than goods inflation and
above the ECB target. However, this is not surprising: historically, it has almost always
been the case. In the euro area, services inflation has been around 1 percentage point
higher than goods inflation since the late 1990s; before the great financial crisis, it
was also systematically above the ECB target (Figure 4). Whatever the explanation for
The a-cyclical, backward-looking ‘latecomer’ items, which include rents and insurances as well as
services directly provided by the public sector (health, education, post), played an important role in
boosting inflation over the past 12 months (Figure A1).
this difference, the data suggest that there is no reason to expect goods and services
inflation to coincide in the long run.8
Figure 3
Inflation persistence, services versus non-energy industrial goods
(a) Responses to energy shocks in a VAR model
(b) Annual inflation rates, 2021-2024
(percentage points)
(monthly data; per cent)
−0.2
−0.2
NEIGs
Services
Source: calculations based on Eurostat data.
Notes: panel (a) reports impulse-response functions based on a quarterly VAR model over the sample period 1996-2023; the x-axis indicates the number of
quarters after the energy shock. Model specification and identification of energy shocks are similar to those presented in Neri S., F. Busetti, C. Conflitti, F. Corsello,
D. Delle Monache and A. Tagliabracci, 2023, ‘Energy price shocks and inflation in the euro area’, Banca d’Italia, Questioni di Economia e Finanza (Occasional
Papers), n. 792. Panel (b) reports the year-on-year percentage changes of the euro-zone HICP of services and non-energy industrial goods (NEIGs) over the
sample period 2021-2024.
Figure 4
Inflation levels, services versus non-energy industrial goods
(monthly data; per cent)
NEIGs
Services
Source: calculations based on Eurostat data.
Notes: the figure reports the year-on-year percentage changes of the euro-zone HICP of services and non-energy industrial goods (NEIGs) over the sample period
1998-2024.
Since services are less tradable than goods, the gap may reflect for instance a Balassa-Samuelson
effect. According to this theory, if productivity grows faster in the tradable sector than in the nontradable sector (for instance because of international competition), while wages are equalized across
sectors in equilibrium, then the relative price of tradables and non-tradables will diverge and the
non-tradable sector will experience a higher inflation rate. Productivity growth has traditionally been
higher in the goods sector, lending credibility to this interpretation of the data. See Balassa B., 1964,
‘The purchasing-power parity doctrine: a reappraisal’, Journal of Political Economy, 72 (6) 584-596 and
Samuelson P.A., 1964, ‘Theoretical notes on trade problems’, Review of Economics and Statistics, 46 (2)
145-154.
What matters is the convergence of overall headline inflation towards our target, not the
behavior of individual components of the price index. The relatively high level of services
inflation will be closely monitored in the future, but we have reasons to believe that the
stickiness in this sector is not abnormal in any way.
2.2. Wages, costs and profits
Wages have been rising gradually to make up for previous losses of purchasing power,
and in some countries they may rise further. Concerns that this will reignite inflation are
mitigated by three factors.
First, intermediate costs have fallen significantly, offsetting the impact of wage increases
on the dynamics of total costs. This mechanism is already at play, and it has contributed
to a weakening of domestic price pressures: the output deflator decelerated sharply since
the end of 2022 (Figure 5).
Figure 5
Contribution of costs and profits to year-on-year changes in the output deflator
(quarterly data; per cent and percentage points)
Output Deflator
Unit Cost of Intermediate Inputs
Unit Labour Cost
Unit Profit
Source: calculations based on Destatis, Insee, Istat and Eurostat data.
Notes: the figure is based on data for Germany, France and Italy. The three components of the deflator (UCIP, ULC and UP) are obtained dividing,
respectively, the cost of intermediate goods, the cost of labour and the gross operating surplus by the real value of production
In addition, in recent months firms have not fully passed on the fall in energy and intermediate
input costs to final prices. As a result, they have accumulated substantial profits, especially
in some sectors; and it is precisely in these profitable sectors that workers are demanding
a rapid recovery of past real wage losses.9 This means that, going forward, profits can play
a buffering role, limiting the pass-through from wage increases to consumer prices.
Finally, since 2021 productivity has been held back by a very intensive use of the labour
factor, driven by increases in the costs of other production inputs and the decline in
Consider for instance the well-known case of the retail sector in Germany. Wages are expected to rise
on average by almost 7 per cent in 2024, but profits have increased by over 40 per cent since 2019, from
110 to 155 billion euros, creating significant shock-absorption capacity. At a more general level, Bank of
Italy calculations show that aggregate markups (defined as the ratio between the value of production
and total variable costs) are well above their long-term trends in Germany, France and Italy (Figure A2).
the real cost of labour. Productivity growth is likely to resume in the coming months,
as labour becomes relatively more expensive, leading to an increase in the capital-labour
ratio. This will compress growth in unit labour costs and hence consumer prices.
3. The outlook for inflation and growth
A careful examination of disaggregated, high-frequency price movements is essential
to understand inflation dynamics, but there is a limit to the amount of information that
can be gleaned from raw data alone. Monetary policy is forward-looking; as such, it
must also rely on an estimate of how inflation is likely to evolve in the medium term.
One way of informing this estimate is to use indicators of ‘underlying inflation’ that
filter out the noise and cross-sectional heterogeneity in the data and provide a clearer
signal on common, medium-term inflation trends. Many available indicators of this
kind show that inflation has been firmly on a downward trend since 2023 (Figure 6). The
simplest indicator, core inflation, has been on a downward trend too; we should not
forget that relatively high services inflation has been accompanied by very low inflation
for non-energy industrial goods.
Figure 6
Rise and fall of underlying inflation
(monthly data; per cent)
HICPXX
PCCI Core
UCI Core
Source: calculations based on Eurostat and ECB data.
Notes: core (HICPX) is the y-o-y inflation rate of the overall HICP excluding energy and food. HICPXX is the y-o-y rate of change of the index that excludes energy,
food and also components related to travel and clothing. UCI is a model-based indicator of underlying inflation created by Bank of Italy staff: Aprigliano V. and
F. Corsello, ’Underlying Composite Inflation (UCI): a novel indicator to track inflation developments’, Banca d’Italia, Questioni di Economia e Finanza (Occasional
Papers), forthcoming. PCCI is the ECB indicator of persistent and common pressures on inflation; Bańbura M. and E. Bobeica, 2020, ‘PCCI – a data-rich measure
of underlying inflation in the euro area’, ECB Statistics Paper Series, n. 38.
Another strategy is to use model-based projections combined with experts’ judgement.
The Eurosystem’s latest projections show inflation gradually returning to target over
the medium term, and are thus fully consistent with the underlying inflation indicators.
In 2022-2023 inflation forecasts took a battering because they were volatile and
(unsurprisingly) subject to large revisions. However, we are no longer in that situation:
the projections for 2025 have fluctuated in a narrow range – between 2.0 and 2.2 per
cent – for more than a year, and those for 2026 have not changed since last December,
standing at 1.9 per cent.
While our confidence in the disinflation process has grown steadily over time, the
outlook for economic activity remains fragile. The euro-area economy has been mired in
stagnation since the end of 2022. The uptick in GDP growth in the first quarter of this year
may be less reassuring than it appears, given the disappointing figures for consumption
and business investment. Moreover, GDP is projected to grow by an average of
1.5 per cent over the next two years – about half a percentage point less than in the last
expansionary phase.10
4. Monetary policy implications
So where does all this leave us in terms of monetary policy?
The recent cut in interest rates reflects the observed and expected decline in inflation
that I have just described.
Going forward, the pace of monetary policy normalization will continue to be based on
the assessment of the inflation outlook, the dynamics of underlying inflation, and the
strength of monetary policy transmission. There may be sporadic unexpected data points
along the way, but the overall course of monetary policy will not change as long as the
big picture confirms that we continue to advance towards the target.
Two elements need to be borne in mind when assessing this baseline.
First, the transmission of the restrictive monetary policy stance adopted so far is still
unfolding. Our estimates suggest that the interest rate hikes implemented in 2022-2023
are still working their way through the economy, and that their impact on inflation may
be stronger in 2024 than in 2023.11
Second, the effect of any rate cuts is likely to be mitigated by the ongoing contraction
of the Eurosystem’s balance sheet. This could make the normalization phase different
from the past, for instance by affecting bank funding costs and credit supply. Indeed,
the data collected through the Bank Lending Survey (BLS) indicate that the reduction in
the monetary policy asset portfolio is already having a tightening effect on bank lending
conditions (Figure 7).
Most importantly, the baseline – no matter how carefully constructed – is only one of
many possible outcomes. The key lesson of the past four years is that policymakers
also need to consider alternative scenarios. Allow me therefore to qualify the picture by
highlighting the main risks that the euro area may face over the medium term.
Between 2013 and 2019, GDP in the euro area grew on average by 1.9 percentage points per year.
Weak growth may partly reflect headwinds coming from the credit market: bank funding costs remain
onerous, lending rates on new loans are still very high, and loan growth continues to be subdued for
both firms and households.
Panetta F., 2024, ‘Challenges for monetary policy transmission in a shifting landscape’, speech delivered
at the Inaugural Conference of the Research Network on ‘Challenges for Monetary Policy Transmission
in a Changing World’, European Central Bank, Frankfurt am Main, 25 April.
Figure 7
Impact of the ECB’s monetary policy asset portfolio on credit supply
(half-yearly data; net percentage shares)
(a) Credit standards
(b) Terms and conditions
−10
Enterprises
Households for house purchase
−10
Consumer credit and other lending
Source: ECB Bank Lending Survey (BLS).
Notes: the figure shows banks’ replies to the BLS question on the impact of the ECB’s monetary policy asset portfolio on their credit standards (panel a) and terms
and conditions (panel b) for loans to enterprises and households. A positive net percentage share indicates a tightening impact in the preceding six-month period.
The last observation refers to the April 2024 survey and incorporates banks’ expectations for the next six-month period (dashed line).
One possible risk is another wave of geopolitical shocks. It is now clear that diplomatic
and military confrontations between countries can have dramatic effects on trade, capital
flows, growth and prices.12 Given the current state of international relations, we must
hope, but certainly not assume, that the global landscape will be more stable in the
future than it has been over the past two years.
Another risk is an increase in political uncertainty within countries. Some of the world’s
largest economies have elected or will elect their leaders in 2024, and political turnover
physiologically translates into policy uncertainty: households and investors need to form
a view on how incoming governments will handle many critical economic and political
decisions. It is hard to imagine how this might affect inflation. Uncertainty can disappear
without consequences. It can trigger capital outflows and currency depreciations, creating
upward price pressures. But it could also shake confidence and weaken demand, halting
or even reversing the fragile recovery we have seen so far. In short, we know that we don’t
know.
Central banks should be prepared to deal with the consequences of such shocks if and
when they materialize. This implies a readiness to use the full range of tools at their
disposal to adjust the monetary stance, addressing any threats to price stability, and
protect the transmission mechanism of monetary policy.
Panetta F., 2024, ‘The future of Europe’s economy amid geopolitical risks and global fragmentation’,
Lectio Magistralis delivered at the University of Roma Tre, 23 April. Lagarde C., 2024, ‘Policymaking
in a new risk environment’, Speech at the 30th Dubrovnik Economic Conference, Dubrovnik, 14 June.
Reichlin L. and J. Zettelmeyer, 2024, ‘The European Central Bank must adapt to an environment of
inflation volatility’, Bruegel Policy Brief, 12 June.
5. Conclusions
Let me conclude.
The euro area has emerged from a ‘perfect storm’, a violent bout of inflation driven by
a series of large and unpredictable supply-side shocks. Inflation is now on a downward
path, and many of the issues that we have debated so vigorously in recent months may
become less important going forward.
The current macroeconomic picture is consistent with a normalization of the monetary
stance. The ECB duly started this process a few weeks ago and, in the baseline scenario,
it will pursue it gradually and smoothly.
In calibrating the normalization we should be data-dependent, taking into account
incoming information on the macroeconomic outlook without overweighting temporary
blips in the data. We should also be cautious in our communications, avoiding the ‘casual’
forward guidance that can arise from (implicit or explicit) predictions on the timing and
sequence of future interest rate cuts. Erratic communication would distract attention
from the key determinant of our monetary policy decisions, namely our reaction function.
Finally, monetary policy requires a management of risks and tail scenarios, not only
baselines. Political and geopolitical risks remain high, and call for awareness, flexibility
and state-contingent action plans. Like other central banks, the ECB has been thoroughly
tested along these dimensions more than once in the last 15 years. The tests were passed
successfully. They were difficult, but also instructive, and they hold promise for how
monetary policy will be managed in the future.
Based on those experiences, I am confident that – barring new upheavals – we will soon
transition from the saga of the ‘last mile’ to the reality of the ‘first smile’ accompanying a
full restoration of price stability.
Thank you.
APPENDIX
Figure A1
Role of ‘latecomers’ in service inflation
(monthly data; per cent and percentage points)
Services y−o−y
Accomodation and Transport
Restaurants
Other Services
Latecomers
Source: calculations based on Eurostat data.
Notes: the figure reports the year-on-year percentage changes of the euro-zone services HICP decomposed into its contributions. Items in the latecomers group
include housing rentals, insurance and financial services, health-related services, education, cultural and social services.
Figure A2
Markups, recent evolution and long-term trend
(quarterly data)
1.290
1.290
1.285
1.285
1.280
1.280
1.275
1.275
Markup
Pre−pandemic trend
Source: calculations based on Destatis, INSEE, Istat and Eurostat data.
Note: estimates based on data for Germany, France and Italy. Markups are defined as the ratios of the value of production over total variable costs. The pre-pandemic
trend is calculated over the period 2000-2019.
Designed by the Printing and Publishing Division of the Bank of Italy