 I would like to remind all of our participants, both those of you in the room and those of you joining us online, that you have until 9am on Wednesday morning local time to vote for your favourite young economist's paper. The traditional wisdom in central banking has held that policymakers should look through volatility in prices caused by shocks emanating from the supply side. However, in situations where surges risk having second round effects, such wisdom looks unwise. Today's first session will explore under which conditions policymakers should respond to situations such as those in recent years when supply shocks have combined to produce inflation that has proven both more strong and more persistent than expected. Will then move straight on to a second session when we'll discuss who pays for inflation and by how much. The first session will be chaired by Fabio Panetta, executive board member at the European Central Bank, the second by his board colleague Frank Elderson. Just as a reminder, we'd welcome questions both from the audience here in the room and via Zoom. Mr Panetta, the floor is yours. Thank you, Claire. Good morning everybody. Welcome to this first session on monetary policy in the face of multiple supply shocks. This topic goes to the heart of the situation that central banks have been facing as of late. In just a few years we have seen a terrible sequence of shocks. We had a pandemic, severe supply chain disruptions, a war, an energy crisis, and let's hope this is the end of this sequence. Claire just reminded us that central banks would usually want to be patient as inflation typically folds once a supply shock is absorbed. In order to avoid exacerbating the risks that temporary supply shock mores into a negative demand shock. But the combination of several supply shocks may create an inflationary shock of such a size and persistence that inflation may risk getting entrenched. For central banks it is not easy to address such a situation for a number of reasons. The first reason is that monetary policy influences demand and with a lack. It cannot do much against the initial spike in inflation triggered by supply shock. Rather, monetary policy initially acts by lowering demand expectations, making it more difficult to assess how much restriction is necessary to bring inflation back to target. The second reason is the difficulty to assess how long it will take for supply shocks to unwind and for this unwinding to reach prices. Similarly, there is uncertainty on the speed and strength of the pass-through of the monetary tightening to the real economy. This increases the risks of policy errors on both sides. So today's session is of great relevance for the current monetary policy debate. We are first going to hear from Silvana Tenreiro, member of the Monetary Policy Committee of the Bank of England. Silvana will have 25 minutes to present a paper co-authored with colleagues from the Bank of England. She will address the topic of this session with particular emphasis on the role that should or should not be given to inflation expectations. Daniel Gross, director of the Institute of European Policymaking at Bocconi University, will then have 15 minutes to discuss the paper. There will be subsequently the opportunity for questions and answers. Participants wishing to comment or ask questions could prepare to raise their hands virtually if you are following online. So Silvana, thank you for accepting our invitation and for your very interesting paper. The floor is yours. Thank you. Good morning everyone and thank you for having me. I will talk about Monetary Policy in the face of supply shocks. This is joint work with my colleagues at the Bank of England, Nicola Banderra, Lauren Barnes, Matthew Chavar and Lucas von Denberg. The usual disclaimer applies. The paper represents our views and not those of the bank or any of its committees. In this paper we addressed three per any questions in macroeconomics. First, how should monetary policy respond to a supply shock? Second, how would that response change if supply shocks became more frequent? And third, what should inflation expectations play in the assessment and calibration of that response? Let me start with the preview of the main takeaways from our analysis and our reading of the literature. First, whether monetary policy should look through, tighten or loosen in response to a single supply shock depends on the nature and duration of the shock, on the strength of second round effects and on the shock's effects on real incomes and demand. Efficiency considerations also add nuance to the response. Second, if a sequence of negative supply shocks keeps inflation above target for a longer period, signs of drifting inflation expectations or stronger backward-looking inertia would call for a tighter policy response. But third, the mapping between measures of expectations and policy is not a straightforward one. Despite the prominence in economic models and policy thinking, our understanding of the formation and the impact of expectations on pricing behaviour and activity remains limited. It is a humbling exercise to read in particular the most recent literature on expectations. This suggests caution on opinion policy decisions too strongly on measures of expectations or in using inflation expectations as intermediate targets for policy. The soundest policy, as always, is to focus on returning inflation to targets in the medium term. In answering the first question, how should monetary policy respond to a single supply shock, the first observation is that supply shocks come in different shapes and sizes. We focus on a particular shock, a global increase in energy prices, as seen from the perspective of an energy-importing economy. The orthodox monetary policy response to a global shock to energy prices is to look through them. For instance, in 2011, UK inflation rose above 5% largely or went to a sharp increase in global energy prices. The Bank of England's Monetary Policy Committee did not raise interest rates in response, and when the shock faded, inflation returned to the 2% target. The rationale for looking through energy price shocks is that the main effects of monetary policy on the economy come through with a lag. Estimates of the speed of policy transmission vary, but the peak impact of policy on inflation typically comes sometime beyond the first year following a change in the policy rate. That makes responding to short-lived price-level impacts of energy shocks counterproductive since they drop out of the annual inflation calculation by the time the policy impact is at its peak. Trying to offset the shock with an increase in the policy rate will cause more inflation volatility rather than less, making it more difficult to meet the inflation target in the medium term. For concreteness in this chart, we illustrate some stylized paths for inflation in the case of an unexpected increase in global energy prices that raises measured inflation from 2% to 6%. After 12 months, assuming energy prices remain at their new higher level but do not experience another unexpected shock, the energy contribution to headline inflation will disappear and the headline inflation rate should fall back to target. The inflation path and the looking through scenario is the one illustrated here in the blue line. Now, rather than looking through the shock, the central bank could try to lean against the shock by tightening monetary policy as illustrated by the various dotted orange lines. However, as said, monetary policy works with the lag building towards peak effectiveness 12 to 18 months after the intervention. If, for example, the central bank wanted to achieve the inflation target six months after the shock, it could quickly and more aggressively tighten monetary policy when the shock hits. But because policy works with the lag, to hit the target at six months, the central bank would need to be willing to undershoot the inflation target at 12 months and subsequently bring inflation back to target from below by loosening monetary policy. This would be the dashed orange line path illustrated here. Many other paths are policy but possible but given the nature of the specific shock, any shortening of the period of above-target inflation will necessarily come at the cost of undershooting the inflation target for some period in the medium term. The various dotted orange lines here. For a central bank with a symmetric forward-looking inflation target, it's not obvious that shortening the period of above-target inflation at the cost of incurring a period of below-target inflation is the optimal thing to do. If, moreover, the central bank has a secondary objective to limit output or employment volatility, it may indeed be a suboptimal path to follow. This conclusion changes when there are important second round effects. As Olivier Blanchard and Jordi Galli point out in their now classical paper, rigidities can give rise to inertia or second round effects. For instance, wages, benefits or certain prices could be indexed to headline inflation delaying the return of inflation to target. From the perspective of an energy-importing economy, an increasing global energy prices is an adverse terms of trade shock. Real incomes and real wages fall on impact, but if everyone tries to resist a fall in the real income and firms try to defend the real profits by raising domestically set prices, real resistance can lead to nominal inertia and delay the return of inflation to target. This is akin to a cost-push shock in the simple new-cation model. Abstracting from questions of timing and policy lags for a moment, the Philips curve shifts inward and the central bank faces a trade-off between stabilizing inflation and stabilizing the welfare-relevant output gap, as illustrated here in this chart. In this chart we assume the central bank seeks to minimize the sum of inflation deviations from targets and output deviations from potential subject to the aggregate supply constraint of the economy given by the new-cation Philips curve. The resulting monetary policy response, MR here, depends on the relative weight the central bank places on stabilizing output vis-à-vis stabilizing inflation represented by the parameter lambda in the canonical specification. The central bank here will raise interest rates to reduce inflation, effectively leaning against the inertia that stems from real income resistance. But it will not try to return inflation to target immediately given the impact on employment or output. This is a different rationale for moderation or caution than the one discussed before. In there the central bank faces a trade-off between above-target inflation in the near term and below-target inflation in the medium term. Here instead the central bank faces a trade-off between stabilizing inflation and stabilizing output. The bottom line is that in the presence of second-round effects, looking through energy shocks may no longer be optimal. By tightening monetary policy the central bank can bring inflation back to target more quickly without necessarily pushing inflation below target further out, as illustrated by the dashed orange line in this chart. The policy prescription becomes a bit more nuanced in multi-sector models with downward nominal rigidities when sectors have differential exposure to the energy price shock. In that case, efficiency considerations may call for changes in relative prices, which can be facilitated by somewhat higher aggregate inflation. This is illustrated in this chart. Relative prices of more energy intensive goods should go up as the supply curve contracts as shown here in the left-hand side. Substitution will lead to an expansion in non-energy intensive sectors, the dashed line in the right-hand side chart here. With downward nominal rigidities the relative price adjustment calls for some accommodation of inflation. This is what Veronica Overieri, Michaela Markinson, Lucrecia Eichlin, and I discussed in depth in the Geneva report, building on Veronica's Jackson Hole paper. Here I want to focus on what happens when the real income effects from the adverse terms of trade shock is strong, as formalized in a recent paper by Adrienne O'Clair and his co-authors. The point they stress is that in a heterogeneous agent model with financial constraints, the real income loss due to an energy shock can lead to a reduction in aggregate demand, which endogenously reduces the persistent inflationary effects of the energy shock. This channel, illustrated by the dotted orange line in the right-hand side chart here, can push inflation below targets in the medium term. My colleagues, Jenny Chan and co-authors at the bank, make a similar point in a tractable two-agent newcation model tank. This chart shows the impulse responses of consumption, inflation, and the nominal policy rate to an energy price shock in both an open economy representative agent model, RUNC, and a two-agent newcation model tank. The central bank leans against inflation by raising interest rates, which creates a further reduction in aggregate consumption. But compared to the representative agent baseline, the two-agent model with some constraints households generates a weaker path for both consumption and inflation in response to the energy shock, which requires a looser path for monetary policy to return inflation to targets in the medium term. As Chan and co-authors highlight, and in line with Eau Claire and co-authors, for sufficiently severe financial constraints, the optimal monetary policy response could be an outright loosening of the monetary policy stance. In an economy that exhibits both second-round effects and household or sectoral heterogeneity, the appropriate monetary policy response to an energy price shock is not straightforward quantitatively or even qualitatively. It's not obvious that the central bank can improve on the path for inflation that incorporates both second-round effects from the inflation overshoot and the demand weakness from the terms of trade shock weighing on inflation in the medium term, the solid line, blue line here. Titanium policy helps reduce second-round effects, but risks pushing inflation below targets and output below potential in the medium term, the dashed orange line. Monetary policy could help the medium term outlook for both inflation and output, but would keep inflation higher in the near term and above targets for longer. That's illustrated in the dotted orange line. And so this is the colorful summary so far for a single shock. More research facilitating the quantification of all these channels would help address the challenges monetary policy could face with similar shocks in the future. Now rather than a single shock, economies around the world, including the UK and the Euro area have experienced an extraordinary succession of external shocks over the past two years, which have pushed inflation very far above targets. The question is what should monetary policy do when there are multiple inflationary shocks in succession? Some economies have argued that we should expect to experience more enlarged adverse supply hits in coming decades, but more frequent and systematically adverse supply shocks could be seen as akin to a downward shift in trend growth, with a likely increase in volatility around that trend. Lower potential trend growth will endogenously lead to lower trend demand, and the result of those two forces might not necessarily be inflationary, unless, of course, private consumption and investment cannot foresee the effectively lower trend pattern. In other words, lower potential growth would not require necessarily a tighter policy stance on average, unless households, firms and markets systematically overestimate supply over time, in which case policy would be needed to close that demand supply imbalance. But what happens when there is a sequence of inflationary shocks in succession without any change in the longer term patterns? In principle, monetary policy could operate as usual, responded to each shock as if it occurred in isolation, but responded to each shock individually trading off every time near-term against medium-term inflation and inflation deviations from targets without deviations from potential could result in a long period of above target inflation. The models discussed so far all implicitly assume that households' expectations remain well anchored at the target in the medium term. But the question is, can this assumption reasonably be maintained if inflation remains above target for multiple years? Once we depart from the rational benchmark, there is an infinite variety of ways in which people could form expectations and in which those expectations could affect behaviour. Even with the rational setting, as Yvonne Berning makes clear in his recent paper, deviations from the standard sticky price calvo model mean that inflation expectations would typically have a different, in his case, smaller effect on pricing than the nearly one-to-one relation implied by the calvo setting. Improving our understanding of the role of expectations on the economy is of course an important area of research in macroeconomics, but it also calls for humility. As economic researchers, we are on the edge of our area of expertise when it comes to people's expectations bearing into the fields of psychology and neuroscience. The models that deviate from rational expectations have different and interesting implications for the nuances of monetary policy strategy. But the overarching takeaway is that, much like in the case of real wage resistance, the more inflation expectations drift away from target following an inflationary shock, the more monetary policy would need to lean against inertia to return inflation to the target. In the second part of the paper, we turn to the literature on inflation expectations and focus on two sets of questions. First, what factors shape those expectations and in particular, how does monetary policy affect expectations? And second, how do expectations affect pricing and activity? On the first question, what factors shape expectations, the literature led by Corvion and Gorod Nichenco finds that expectations tend to track spot inflation very closely. Inflation expectations also show a very high sensitivity to some volatile components of the basket. Energy in the case of firms and both food and petrol for households. Regarding the role played by monetary policy in shaping those expectations, our analysis and the literature do no point to systematic evidence of a direct impact from innovations or news in interest rates on inflation expectations. And when there is an effect, perhaps contrary to the economist's expectations, households and firms more often than not respond that interest rate increases lead to higher price or cost inflation. The evidence instead is more supported of the standard indirect effect of monetary policy via actual inflation and demand current or future, which is a mechanism, the demand channel embedded in most new systems, turning to how expectations affect activity and pricing, the literature finds higher inflation expectations may increase or decrease consumption and investment. This reflects the tension between a standard real interest rate channel, which should boost activity, versus a real income effect which depresses activity. In other words, the impact seems to depend on whether households or firms perceive inflation as driven by positive demand or negative supply factors. On the impact of expectations on pricing, the evidence is also inconclusive, in part because there are challenging identification issues at play. On the theory front, there is new work, but as I mentioned by Ivan Berning, showing that the quantitative impact of expectations on pricing might depend on the modality of price setting, with calvo price setting, given the maximum impact. Ivan's work also points out that as inflation increases, prices should become more flexible and hence expectations become less relevant than sport or past inflation in setting prices or wages. Back to the main takeaways. First, the optimal monetary policy responds to a single supply shock, depends on the nature and duration of the shock, the strength of second round effects, and the impact of the shock on real incomes and demand, as well as efficiency considerations. The relative strength of these factors determine whether monetary policy should look through, tighten, or loosen, and by how much. Second, a sequence of inflationary supply shocks could result in a long period of above-target inflation if the central bank were to respond to each shock individually, trading off near-term against medium-term inflation and inflation deviations from targets with output gap. Drifting inflation expectations or backward looking inertia in that case would call for a tighter policy response. But there, despite the prominence in economic models and policy thinking, understanding of the formation and economic impact of expectations is still limited and a large gap remains between standard model assumptions on expectations and their actual patterns of behaviour. In particular, empirically, households confirm inflation expectations tend to move with actual inflation and are often highly sensitive to some volatile components of the basket, particularly energy. Identified monetary policy shocks appear to affect actual inflation, but do not seem to have a direct impact on households or firms' inflation expectations over and above their impact on inflation. Recent empirical and theoretical work has challenged existing priors and assumptions on how inflation expectations affect pricing, suggesting a weaker impact of expected inflation on prices than the nearly one-for-one link implied by standard carbon models. In turn, when inflation is driven by a supply shock, higher inflation expectations can be associated with weaker consumption and investment, while the opposite is true when inflation is viewed as demand-driven. Overall, first, the high sensitivity of households and firm measures of inflation expectations to volatile components of the basket. Second, their limited reaction to monetary policy over and above the effects of policy on actual inflation. Third, there are uncertain effects on the economy called for some caution when using households or firms measures of inflation expectations as intermediate targets to guide central banks' policy decisions. I will pause here, Fabio. Thank you very much, Silvana. Thank you. Thank you very much for your very interesting presentation. Before I give the floor to Daniel, let me remind participants to get ready to raise your hands, physical or virtual. Then you have the floor. Thank you for being here. To thank the organizers for having me. It's a pleasure and an honour to be here, especially at this juncture when we are facing all these supply shocks and expectations of inflation are doing things which we don't expect them to do. It's actually a theme which was already discussed last year at last year's intra-conference. But now, of course, we are one year on. We are perhaps a bit wiser and have a bit more data also. It's a particular pleasure to comment this paper which is really well rounded. It lays out first the theoretical background and then it provides a really thorough review of the literature of the nexus between... Oh, sorry, I forgot here. The nexus between expectations, monetary policy and how they work together to have an impact on the economy. Now, when you have such a well rounded paper then the question is what do you do as a discussant? I would like to make three small contributions. By first of all, looking at the evidence of some of the factors which Silvana mentioned as being decisive for the impact of supply shocks on the economy, namely real wage rigidity and the importance of financially constrained households. And then asking what shocks actually we're facing today. And there perhaps I will try to take at heart the dictum of Kyrgygad which President Lagarde mentioned yesterday to live in the future. So I will not just look in the past but also to try to see what actually should we be expecting. And then finally perhaps a small complement to the work which Silvana mentioned only on person, her own work on the UK, on the nexus between supply shocks and expectations. Just doing something similar for the Eur area because my contention would be that actually the Eur area and the UK in some respects are very comparable because they have a similar degree of energy dependency. And actually by chance the degree of openness in terms of good trade is exactly the same. 21% of GDP. So in that sense at least they are comparable of course there's a different size but not a difference in openness. So the first point is as Silvana mentioned from models and from common sense you can have a price, wage price spiral when you have real wage rigidity. I would submit that there's no evidence of real wage rigidity, at least in this case. The question of course is how do you measure real wage rigidity? On the left hand side I've given you both for the UK, sorry for the US and for the Eur area the real wage is measured by the CPI. And there you can see I mean they really have taken it on the chin workers. One cannot really fault them for not accepting the loss of purchasing power. On the right hand side I think that is too often forgotten I have deflated wages by the GDP deflator because that is the relevant price index if you look at it from the producer side. And maybe if there had been no movement in that indicator you could think about wage rigidity. But I would submit to you there has been by the way extraordinarily similar development on both sides of the Atlantic even GDP deflated wages have fallen. Of course the question is always where do they go in the future. We can discuss that in a bit. But I would therefore submit no evidence for wage rigidity, but that also means that the increase in energy prices is not really a good excuse or a reason for expecting high inflation, core inflation. The second point as Yvanna mentioned also which is important for gauging the impact on the stock on the economy is the degree to which we have financially constrained consumers. And we all know that there has been a huge increase in excess savings. Here I have taken just the amount of deposits by households over and above the trend line. For the euro area only I think for the US the picture is similar. And what's interesting I find is there has been a very large excess savings 40% of GDP almost if you look at the difference between the two lines which has now been almost fully reabsorbed. So we had a period there again the past might be different from the future. We had a period when there were few financially constrained households and now we might be getting back to the more normal situation. OK. The next point is what actually supply shocks are we facing? And here I would like to go into the development because I think it is not sufficiently recognised that these shocks have been actually exposed temporary. All the models and of course also the things that the charts which Yvanna presented assume a permanent shock for energy for example. Natural because we assume that energy prices follow random walk. Once they go up best prediction is they stay. But they have gone up and have come down. You know the best prediction should be that they stay down. So and therefore I think this has to be taken into account but of course the economic impact would be very different. The next point is Yvanna made depending on where the terms of trade went. And then we have the supply chain shock which I would argue that it was should have been expected ex ante to be temporary. Because it was basically a problem of temporary dislocation in supply chains. Here I have given you the one indicator of the Federal Reserve I believe where I draw attention to the scale because there it's in standard deviation so you have really something exceptional for standard deviations again from the past. But actually my title is wrong there because supply chain pressures have disappeared in that sense we are back to the pre-pandemic in normal. So that was something temporary and the echo effect of that should actually be positive for the future. Now for energy I would submit that you should look at two prices separately. One is oil price and of course the second one will be natural gas I come to that. And the oil price here I have given you both the prices in the US and in the Euro area in national currency because at times they were very different you see that in 2008 when the dollar price increased a lot the euro price whereas more recently the difference has been smaller. What I wanted to show with this picture is actually that of course you see that price has come up and down so the shock has been retraced the second point is that we are now at a level which we have had in the past several times we have had these ups and downs in the past even more extreme to some extent without inflation so I think the task for our models and theories is to explain why with a shock which is very similar to what we had in the past we have inflation which is not at all similar to the past and I think that is very important of course what you also see here is and that comes about in many discussions what is the baseline if you take the deepest the best point for us at the height of the corona crisis when oil prices went negative of course we have had a huge shock but maybe the central banks then should have seen that there was a temporary positive supply shock and it should have seen through it but that is the past right now I think we are back in a situation which we cannot claim to be exceptional on the price of natural gas it might be difficult to see you see two lines blue and red which are very closely correlated and one is the TTF price the famous Amsterdam spot price and the other one which is falling closely is the Japan Korea marker just to indicate that in Asia basically the spot price moved one to one with our European price and then you see a green line down there which doesn't budge at all and that's the happy US where basically there's too much gas natural gas which cannot be exported more quickly enough and therefore prices will stay down so that is the picture in terms of energy price I would claim not really totally exceptional and most of that already back to normal pre-pandemic now of course as Siva now mentioned from the point of view of a net energy importer you then have a negative terms of trade impact which impacts demand and then impacts the probability that you have excess demand in the non-tradable sector for the your area I've given you the calculations which I think representatives of the ECP have also made in the past so the loss of terms of trade was rather substantial whereas for the US let me perhaps go immediately here it was of a very different size here you see the red line that's the United States you see actually the US has had a major improvement in the terms of trade so you have if you want a temporary supply again but at the same time a boost to domestic demand which was an additional factor driving inflation whereas in our case the green line you see how much peak to strife of course temporary was Covid there was a deterioration of which I would say most of it has been traced back the latest data point is Q1 I bet you if you take Q2 2023 we'll be back to the one line which is the average so I would say in terms of the terms of trade we have actually come back in Europe so that brings me to my major issue how can we explain sticky inflation with shocks which at least exposed were temporary temporary supply shocks in Sivana's models and others should leave the price level unaffected in the long run first goes up and then goes down right and we talked about it already last year that supply shocks supposedly have a strong immediate impact on the way up but why not on the way down how can we explain that when these shocks are over probably have a price level which is 10 to 12 15 percentage points higher in two or three years then before the shocks with energy prices and the global value chains being back to normal there we must have some asymmetries and it's not real wage resistance it's not that nominal wages don't go down because very far from being even stable and then the question is could expectations be responsible and that is of course where Sivana did a lot of work I mean I show you here just to the extent to which this confirms what Sivana said which is that visible prices, energy prices this is the HICP energy drive a lot of the expectations short term expectations I think Ricardo Reis last year already mentioned that if you take the long term expectations they are basically uninformative because they are flat they take their data from the ECP and say that's it what will happen but if you look at the more the short run variations here as I mentioned we have done a compliment to what we have done for the EUR area and I won't go into the details some of them are available in the slides which are on the internet on the website sorry but basically we confirm her findings that inflation expectations follow to a large extent energy price shocks but also finding that I find interesting that energy price shocks seem to drive ECP policy in the past if you think about the famous 2011 mistake exposed that was one of the cases of a temporary supply shock ok let me stop here my contention is that the major supply shocks which we have seen turned out to have been temporary and we have to explain why they have had such a sticky impact and I think only if you feel comfortable with that explanation can be then also set policy and set our expectations because a priori the future the next 12 months should be a rerun of the last with the opposite sign right energy looks good no longer supply pressures and so on and it is with this conandrum that I would like to leave you and apologies for the 15 minutes overdraft thank you very much thank you Daniel for your presentation and I'm sure that now you are burning to ask questions so let me remind you that you are supposed to raise your hands but while we collect questions I would give the floor to Silvana for a comment in response to Daniel's presentation very good thank you thank you Daniel for the comments very helpful so I would briefly comment on a couple of things I think what this underscores is importance of understanding the whole input supply chain and how the energy shock transmits through the various sectors in the economy and through these input output languages because even core sectors are very independent on energy so we cannot completely separate and in this I want to highlight on the theoretical side the work that Elisa Rubo has done on highlighting these input output linkages and on the more empirical side the work that Shemlem Calemlioscan and co-authors and then in the Geneva report Veronica Guerriere and co-authors highlighting how a temporary shock by the time it permeates through the rest of the economy can have a slightly longer lasting but we haven't seen the end of the story so the question is how quickly inflation will come back and so that's an open question it took a year and a half to go up how long will it take to go down and I agree with you that there is more to that than just real resistance part of that is a natural process those prices, those high energy prices last year are still in the inventories of many companies and then there's an additional question of course that is this kind of lively in Europe with profit margins and so on and just to highlight Europe, the euro area and the UK obviously were exposed to the same shock in a different way than the US quantitatively however the size of the shock was very different the cumulative increase in energy prices retail energy prices in the UK was 115% at peak for Europe, for the euro area that reached peak at 80% but then unwind very quickly unwind very quickly in the UK because of the often a mechanism of price setting it takes a bit longer for that to unwind and we've seen the first tranche down but there's still more to go in the US again it's a very different story because it's not an important economy it's an exporter on the margin and energy prices went up only by 30% and again reverse very quickly so these are very very different shocks quantitatively and then qualitatively across the Atlantic I'll stop there Fabio and perhaps we can let me remind you that you should keep your questions within 60 seconds so I say Falker then Ricardo look at him and then please ask the floor also from home can you please identify yourself for Kavila and Goethe University thank you very much I try to be brief but I would like to submit a different narrative all the discussion was about a sequence of energy price shocks but I think you need to go back to the pandemic 2020 the policy response which lasted well into 2022 and then the Ukraine exogenous shock but if you look at models who combine macro and epidemic modeling just like Eichenbauer-Weller-Trabant a new Keynesian model and you simulate a pandemic or you study a pandemic you will find that you can have an 8% of GDP recession with a very mild inflation disinflation impact of 60 basis points that's because the gap between supply between potential and GDP is much much smaller than on top of this that's what from that perspective explains quite well what happened in 2020 namely a very mild disinflation effect but then you have a very expansionary policy monetary policy, fiscal policy unprecedented lasting well into 2022 all throughout 2021 so you have high demand side support and an endogenous response as the economy recovers and rises in energy prices so the foundation of inflation was laid in my view because of that very strong and lasting support and as luck has it inflation was at 5-6% at the end of 2021-22 then on top you get a temporary shock like the Ukraine war but if that story is right you have to remove that foundation of angry demand and you cannot wait just for this to go away by itself maybe we take another question then you will answer those Riccardo Thank you, Riccardo Reis Sylvan, you concluded and I think I'm reading it right, our understanding of the formation and economic impact of expectation remains limited suggests caution on policy decisions to rely strongly on expectations measures going back just an hour ago to Madame Lagarde's speech she did mention expectations in one paragraph she also mentioned and I ask for your patience productivity surely our understanding of the formation of productivity and how it evolves where the corridor is also very limited and how it transmits to your labor cost is very hard she mentioned profit margins boy there we barely even have a literature where profit margins markups sometimes go up sometimes go down it depends strongly on industry it depends strongly on sector and third wage formation I understand that you're asking for humility with regards to all economic indicators and I think I've covered all and if I went back yesterday to get a gopinath and employment financial stability we can also talk about our difficulties in their formation and impact and therefore to have a much stronger focus on inflation I guess or is there something about expectations that led you to highlight that in terms of your talk Thank you So let me start with Riccardo I agree with you that we should be humble in general and that's why we have big fund charts when we do our inflation projections and sometimes the numbers, the point estimates are taking literally and people don't understand the uncertainty surrounding those estimates or get too trapped in the exact point estimates without taking into account the probability distribution and the conditionality of many of these forecasts so I would agree with you The reason I focus mostly on expectations is because perhaps this is the most uncertain area because there is a projection forward by actors that perhaps don't have the tools to perform those projections and within the whole setting of if you think of profit margins which formation we're seeing it on the ground it's not a projection that we can observe more and more in real time but I agree with you that when we project those forward there is ample scope for uncertainty and same with productivity and obviously we have a very active field in economics right under some productivity and what drives it but my focus on expectations is because it comes mostly in the context of these discussions but I would fully agree with you that all these other aspects are hugely important Of course I am fully aware of what happened I really demand the responses of monetary and fiscal policy during the pandemic were extraordinary times it was once in a lifetime shock you might question when the support should have stopped or not but we haven't seen certainly in Europe and this is very different again from the US we haven't seen an expansion in aggregate demand in the UK in particular consumption is not back at the level of 2019 pre-COVID so we are not talking about an overheating demand we are talking about a severe supply shock that led to a substantial response by policy makers and then obviously there is a quantitative debate on the size of that response and when to end it but that is a separate one for the US the situation is very different because in the US as you know consumption recovered to pre-COVID levels early in 2021 overtook the pre-COVID trend by the middle of 2021 so we have seen really a boom following COVID very very different from what happened in Europe the euro area is still well below the pre-COVID trend and as I said the UK is below the pre-COVID level so it's a very different picture in the jurisdictions I think Sivana did a perfect summary and I think the key point is as you mentioned in the EU area as in the UK domestic demand doesn't show signs of overboarding and therefore being a driver of inflation what I would submit is that maybe the delayed effect of the monetary easing still going on in 2021 should have been felt in 2022 and should now be slowly fading out that could have been one of the additional elements last year I already have a long list I see nobody asking the floor from Zoom am I wrong or I don't see anything on my screen but anyway I have on my list I've certainly forgotten somebody Lucrezia, Jordi, Robert Christine and Tiff then we will see, okay, Lucrezia Thank you I just wanted to comment on two important points that Daniel made on oil price and I think you made two points we have similar shock in oil but in energy but a very different response from the past on inflation and the second point is it matters whether the energy shock is temporary or permanent so I'd like to offer some empirical evidence which is actually in the Geneva report that we quoted with Silvana that in the past, in the euro area and if you shock the economy with an exogenous oil shock and you look at the general equilibrium effect actually core inflation reacts with the lag and from the first shock until it returns to the steady state is about 60 months so I think that is more or less so we are still in the norm of what we have seen in the past and the second thing is that the core is a lagging indicator with respect to headline so this is a feature that maybe is different in the US economy but it is incredibly robust in the euro economy so I think it is very important to get this fact straight Jordi, where are you? Thank you Just a quick comment on Daniel's discussion the fact that real wages have declined somewhat in terms of the GDP deflator which is a relevant measure as you said is not a proof in my opinion that there are no real wage rigidity because we don't have a counterfactual we don't know how much real wages should have gone down in the absence of real wage rigidity and I would claim that the fact that inflation has gone up that much in particular GDP deflator inflation is indirect evidence that real wages didn't go down as much as a model with fully flexible wages would have implied I agree with you look right so it is part of this input-output diffusion of the shock and that's what we have in the Geneva report and in previous work that Lucrecia has done and with Jordi yes I agree I guess it was mostly a comment to Daniel but there has been certainly we've seen in the UK some degree of resistance and the fact that real wages fell somewhat is not proof that those rigidities are not in certain sectors and in certain parts of the economy we have three more questions if you are quick we can come with it all three and see other hands but let's start with Robert then Christine then Robert this gentleman here thanks a very simple question and answer to the question which Daniel Boester in the US sir, we had no terms of trade shock but we had a lot of fiscal expansion in Europe we had a lot of fiscal expansion ynglyn â'i bwysig, a'i ddweud o'r ffysgol ynglyn â'i ddweud, ychydig y roi yn unig o ddweud o'r ddweud. Mae'r cyfnodd rhai o'r cyfnodd yn yr EIS y byddai'r cyfrin o'r ddweud, a'r ddweud o'r ddweud o'r ddweud o'r ddweud o'r cyfrin o'r ddweud o'r ddweud o'r ddweud o'r ddweud o'r ddweud o'r ddweud o'r ddweud. Yn y pwysig yw Mike, Yn Ysgrifennu? Thank you. Kristen Forbes from MIT. So last night at the dinner speech, Gita raised an uncomfortable truth that Central Bankers needed to think more carefully about how they responded to supply shocks. This paper is an excellent response, very timely, very quick, but it is very nice how it lays out clearly what characteristics of the supply shock determine the optimal response. So I think this is a great resource for Central Bankers now. But with that in mind, I was wondering if you could talk about how your results would change if you extended the model along two characteristics of the recent supply shock. First, one characteristic of the recent energy shock is that countries have responded with large policy responses, energy price caps, and targeted subsidies, and that has mitigated the impact of the big price rise on demand and inflation. We'll hear more about that in Pierre Olivier's excellent paper tomorrow. But if you then take away some of this effect on consumption and inflation, how does that change what is the optimal monetary policy response? Second extension, the characteristic of this shock is that the supply shocks were very large, so you can get nonlinear effects, which we'll also hear about later. So if you get nonlinear effects on wage bargaining and price setting, how will that affect your implications for the optimal monetary policy response? Well, Sylvanna and Danielle, thank you both for excellent presentations. My question, Sylvanna, is, as you've looked at inflation expectations, what do you think the horizon is that matters the most for monetary policy? I mean, as you showed, longer run expectations have been very well anchored. That's been a great source of comfort to central bankers. But on the other hand, short run expectations have moved around a lot. And going back to President Lagarde's opening remarks, when you think of firms setting their prices, how wage bargainers setting their wages, if those short run expectations are more important for that, the situation might be more complicated. And certainly, at least in the Canadian case, if you look at survey expectations of inflation at the short run, they are above our own forecast. So to put it more provocatively, I mean, are we taking a bit of false comfort that longer run expectations are well anchored? Last question. Last year at John Wilbur. Last year I talked about the inadequacy of central bank models and the chart that Daniel Gross put up of how excess liquidity has evolved post-pandemic and how it's declined since is really important. Differences in household portfolios between the UK and the Eurozone have a huge implication for differences in monetary policy responses and also how the real economy is going to respond. And as Christine Lagarde pointed out yesterday, at the difference, actually this morning, the difference between a floating rate and fixed rate mortgages, those structural differences, another really important feature that distinguishes the UK from the Eurozone. Thank you. OK, let me start with that last one. John, yes, excess liquidity played a big role in the forecasting models of the bank. This is a reason why we had a very strong demand in the recovery from the pandemic in the projections. My colleague Ben Broadman has a nice description of the forecast at the time, given that excess savings, as it connects to what Daniel discussed, we were expecting a much faster recovery in demand from the COVID pandemic. But then the war hit. And then what we saw is actually we undershot massively because the war represented a big, or led to a big increase in the price of energy, very adverse in terms of trade shocks. So that played an important role early on. And as Daniel mentioned, most of those excess savings are now being absorbed. There was a question of what expectations matter, and this is a crucial question. I think when we think about the anchoring, typically we think about longer-term inflation expectations, and that's the focus, and for that we need to resort to markets, but those markets are not the ones involved in the actual pricing decisions of firms. When we think about what determines pricing decisions by firms, well, what's the relevant horizon for them? Our intuition comes from sticky price models. Typically, with an average duration of prices of one year, then you would think it's one year that matters. And that's a point that Ivan Berning makes very clear in his recent paper, short-time expectations matter more than those long-term expectations. But obviously, as inflation increases, price flexibility increases. And so prices become very flexible at the extreme. Those expectations shouldn't matter much. I mean, you're really setting prices based on spot inflation or past inflation. So this, I think, is an important aspect to work through more and build that connection between academic models and firm decision on how they switch to these higher levels of flexibility. I think I'm in a negative time, but that's a great point. Obviously, as you know, when we do monetary policy, we consider aggregate demand with all its influences and fiscal is a huge influence in there. Basically, what I'm talking here is the residual shock left after fiscal policy or energy policy have done their jobs. And this is what's left for monetary policy. Obviously, when you're thinking about an energy price shock, monetary policy is not the first best response to that. You need an energy policy that would take care. Exante, of course, of anticipating and mitigating the impact of those shocks through diversification of supply, diversification of trading partners, inventory policies, et cetera, et cetera. And then comes the remedies that you can put in place when a shock hits. And once you account for all those, then comes monetary policy and picks up the pieces left. In this case, it was full exposure to the shock, but in the future, hopefully, if we can address this in advance in different ways, maybe the shock will be less impactful and less work will be left to monetary policy. I don't know if you want me to continue. I guess that connects to Robert's question on fiscal. So I'll stop there. I just wanted to emphasize we are not facing at present the tailwind of a permanent shock, but actually we have had over the last six months a positive shock. So we should be living in the future. And if we had had nonlinear effects in the past on the way up, should they be also nonlinear on the way down? So I come back. I think in the face of temporary shocks, we should look at the question of asymmetry, and that is crucial for understanding of where the economy is going. Thank you very much, Daniel. I think we need now to bring this session to a close. If I had to summarize the main takeaways of this session in one sentence, I think that that sentence would be that, unfortunately, there is no shortcut, no intermediate target or single indicator that would allow to simplify decision making and the calibration of the monetary policy response to multiple supply shocks. And this is, in fact, reflected in the reaction function of the European Centre Bank. We have repeatedly emphasized that our monetary policy decisions are determined by our assessment of the medium-term inflation outlook in light of the incoming economic and financial data, the dynamics of underlying inflation and the strength of the monetary policy transmission. Please join me in thanking Silvana and Daniel for their contributions, and I will now give the floor to my dear friend, Frank Elderson, who will moderate the next session. Thank you very much. Thank you, Silvana.