 I will take advantage that you are traveling to the seats to introduce Pierre Olivier Gourichas. Pierre Olivier is economic councillor and director of the research department of the International Monetary Fund and currently on leave from the University of California, Berkeley. And, Giancarlo, take a seat, please. His peer, Wernher Chair at the Robert Schumann Center of Economics at the European University Institute. And, you know, today is going to be, you know, this session is going to be dedicated to the interconnection between fiscal and monetary. So, Pierre Olivier, you have the floor for the presentation of your paper and please adjust to the time that we have allocated 25 minutes. Thank you. Well, thank you very much for the invitation to present here. It's really a great pleasure. This is work that is co-authored with Maideau, Alan Bizzoli, Chris Jackson, and Daniel Lee. Daniel is in the room somewhere over there. Now, the usual disclaimer applies. These are not necessarily the views of the IMF Management Executive Board or IMF policy. Now, the first paper in this session was talking about unconventional monetary policy and asset purchases. Now, this paper will ask about another set of unconventional policies that have emerged since the energy crisis in 2022. And these are unconventional fiscal policies, and we are going to define them as the set of policies that are implemented or have been implemented by a number of countries to protect households and businesses from the energy crisis, but with a secondary objective of reducing headline and possibly core inflation along the way. And what our paper is going to attempt to do is assess the impact of these policies on inflation in the euro area. Now, the starting point of our analysis is a surge in worldwide inflation. I'm not going to dwell too much on this. Everyone in this room knows what I'm talking about. What I'm showing you here is that the composition of headline inflation is the black line on this figure into core inflation. And here, this is defined as the weighted median inflation. This is a measure of core that strips out the effects of unusually large changes in a number of prices. And this is in blue. The cylinder shown in red represents headline inflation shocks and reflects the importance of sharp shocks to a small number of prices, in particular energy. Now, what you can see on the left is that headline inflation shocks represent a significantly larger share of inflation in the euro area, especially in 2022, compared to the U.S. on the right, where in the U.S., since basically December of last year, headline accounts for more, than its value. Now, as also everyone knows here, this inflation surge has been followed by one of the most aggressive and synchronized tightening of monetary policy on record. That's something you can see on the figure here. Even though inflation has remained somewhat persistent, something that has been emphasized by Gitta Gopinath in our speech two days ago or by President Lagarde in their opening remarks yesterday. Now, so it's a very natural question to ask is there a role for fiscal policy in helping to reduce inflation? And the textbook answer here is an unambiguous yes. Fiscal policy helps compress aggregate demand, and it also helps maintain the credibility of the overall and high inflationary stance. In other words, it helps people thinking that there's no there's no gap between what fiscal and monetary policy are trying to achieve. But in fact, a number of European economies have chosen a different path. They've aimed to directly contain the surge in energy prices on households and firms' budget, but also in doing so to reduce headline and possibly core inflation. This was articulated recently in a Vox EU piece by Francesco Giavazzi, until recently, Economic Advisor to Mario Draghi. Now, this measures, what we call unconventional fiscal policy measures were economically significant. What you see on the left are some of the measures hired by our colleagues in the fiscal affairs department at the fund of the size of these measures broken down by different categories, whether they're targeted or untargeted, whether they're price suppressing or non-price suppressing across your area countries. And on average, they represent about 3.3% of GDP in 2022 and 2023. 1.3% in 2022 and an estimated 2% in households and small and medium enterprise. What you see on the right is that these measures were also larger in countries with more initial exposure to the energy price shock, which is perhaps not too surprising. So I'm going to present, now, there are a number of reasons why we should be a little bit skeptical about these unconventional fiscal policies and actually the views were quite skeptical in the economic community. There was a survey done in June of last year at the School of Business that expressed that sentiment, although there were a few dissenters, for instance, Olivier Blanchard, and I put the quote at the bottom, said, well, this is maybe one of those cases where a larger fiscal deficit can make the job of monetary policy easier. Now, let me list a few of the criticism that had been voiced at the time. First, of course, energy support measures that would, in principle, increase energy demand in the context where we have an energy crisis. Energy is limited in supply that could exacerbate the energy crisis and makes things worse both on the energy side, but also possibly on the inflation side. The second main concern was that fiscal measures were poorly targeted and it was costly. I just gave you some estimates of their budgetary cost and so that would delay much needed fiscal consolidation in the wake of all the measures that were implemented during the COVID-19 crisis and in the context in which debt-to-GDP levels in many countries were already quite elevated. The third concern was that if the energy shock turned out to be persistent then the measures would quickly become unsustainable from a budgetary standpoint. They would have to be removed and at that point the increase in price would happen no matter what so that it would only delay the inevitable and if the shock was transitory it would perhaps smooth inflation have a little bit less today, a little bit more tomorrow but maybe not change accumulated in a change in price level so you wouldn't achieve any significant gains. And finally there was a concern that the stimulus the positive stimulus component of these measures would fuel aggregate demand and through the usual channels would in fact increase inflation and make the job of central banks even more complicated. So here our focus in this paper is going to be narrowly on point 3 and 4. We're not going to address directly the point about the impact on energy price. In our model simulation we take that into account to some extent but we're not going to dwell too much into this and we're not going to look directly at issues of fiscal sustainability that could be raised by these measures. But we are going to be looking at the impact of these measures in terms of the inflation shock and we're going to be looking at the persistence of the energy shock here. Now I want to be crystal clear that the first two concerns are very valid in our view but we're not exploring them because we don't have really a framework that can address them. Doesn't mean that they are not relevant from a policy perspective. And I also want to state at the outset just to be clear that we agree that is the primary task of monetary policy here in particular the ECB to be addressing price pressures and inflation instability and so we are asking ourselves well in addition to what central banks are doing what ECB is doing is there an effect of these type of measures okay? So let me illustrate with a simple conceptual framework here that I think will be useful for thinking through the kind of results we're going to be obtaining. So I want to focus on a situation where we have here a Phillips curve that's what I'm representing on this graph. You have the output gap on the horizontal axis and you have the inflation rate on the vertical axis. It's a very standard Phillips curve if output is increasing that there's going to be more price pressure and we have that in many of our monetary policy frameworks. Okay? Now this Phillips curve has three ingredients the output gap which you see here on the axis it's going to have also supply shocks or what we call the headline inflation shocks really to energy I'm going to show them to you in a minute and there is in the background also something that is related to inflation expectation I'm not going to talk about this now but it's going to be part of our framework as well. Now there's a growing body that has some theoretical evidence that emphasizes the importance of non-linearities in Phillips curve in other words when the economy is running hot the relationship between output gap and price is different than when the economy is not as hot or is cool. Now this is illustrated on the figure that you have here by the slope of this relationship between the output gap and inflation so we have a blue part of the curve to the left and the bottom point like point A where the economy is not running hot and where there is a relatively modest relationship between output gap and inflation and then there is a steeper part of the curve that's in red where you have a point like point B where the economy might be running hot and where you have a sharper connection between price changes and economic activity and that steep part of the curve is also where stabilization policies whether monetary or fiscal are going to be most effective in sort of cooling off the economy and the blue part of the curve it's going to have less of an impact on inflation dynamics. Now let me bring in our unconventional fiscal policy and our headline inflation shocks in that picture so the way you can think about headline inflation shocks here is like a supply shock it's pushing up this curve for any level of the output gap you have a higher level of inflation you have an energy crisis it's increasing inflation across the board it will now become the point like point C which is vertically above it and a point like point B will become a point like point D which is also vertically above it where you have more inflation for the same level of the output gap okay now here in this context what would unconventional fiscal policy do well one way to think about it is that by neutralizing the increase in the energy prices it's in fact going to bring down this higher Phillips curve for one that we had initially before the shock in fact if it absorbs entirely the increase in energy prices let's say we would be back to the lower Phillips curve in this diagram but we would not end up back at point A or at point B because in fact what would happen in that case is that there is also a stimulative component of these unconventional fiscal measures they are adding to aggregate demand so we would also move along the curve to the right so we would end up for instance to the red to the black dot that is to the right of point A or we would end up to the black dot that is to the right of point B now notice the difference between these two scenarios in the first one where the economy is sort of cool to start with when these unconventional fiscal policy measures are implemented they reduce inflation there is a little bit of a stimulative effect but we're on the flat part of the Phillips curve and so we end up with but most importantly with a reduction in inflation by contrast if the economy is running hot you're starting from point D and you're implementing these fiscal policy measures you're going to bring down the economy back to the lower curve but you're going to move to the right by the same stimulative policy you're going to end up with a point that is up and to the right of point D in other words the stimulative effect will have overwhelmed your inflation reducing effect of more inflation so what our paper is really doing is asking empirically which of these two scenarios is the most relevant one given what European economies have been implementing in the past in the past year now let me spend a minute giving you a summary of the main findings and then I'll get into them the first result is that inflation in the euro area has been mostly driven by headline inflation shocks and not by overheating of the economy unlike the United States and I'll show you some decomposition here that make this point very sharply the second result is that the unconventional fiscal policy measures help reduce inflation in the euro area by between 1 and 2 percentage point in 2022 they kept inflation closer to target both in 2223 and also in 2024 there was a little bit less inflation in this first two years and under our projections there would be a little bit more inflation in 2024 but overall keeping inflation closer to target about a third of the reduction is coming from the direct effect you're reducing the size of the shock but two thirds of the effect come from the pass through of this headline inflation shocks into core inflation that is also muted once you reduce headline inflation shocks and we find relatively limited effects from the circulating part of the fiscal policy on aggregate demand and instead we find that there is some moderate stabilization of inflation expectations that comes from bringing down core and headline inflation now when the non-linearities in the Phillips curve which are really important in the diagram I just showed you are taken into account the average inflation in the euro area between 2021 and 2024 is lower by about 0.5 percentage point so there is a net reduction in inflation as well now we should not just stop there and think that somehow this is a miracle instrument and we should all just do this in fact we point out that there are a number of fortunate circumstances that help make this work in the European context the first one is that the energy shock was very temporary improved more temporary in fact than it was expected to be maybe a year ago so energy prices came down whether we're looking at gas or electricity or oil they came down quite substantially over the last six months or so the second one and that goes back to point number one is that there was very little overheating in the euro area to start with and so we were sort of on the lower part of this Phillips curve where it was helpful to have this measure to bring down inflation but by the same argument if the economy is overheated then the set of results would be very very different and I'll show you some counterfactual simulations where we apply similar policies to an overheated economy namely the US and we find very very different results now other factors are not explored here as I've already mentioned we're not looking at spillovers of these policies from one country to another we're not looking too much at the impact on overall energy markets we're not looking at fiscal sustainability all of these things are important but they are not the focus of what we do today and so the conclusion is going to be that this is not a blanket endorsement it's just the nature of the shock the state of the economy and the design of the local instruments all these things do matter in the end alright so let me now dig in into the way we established these results we're going to use both model based simulations and empirical estimates in a complementary way so first we'll use our own IMF flexible system of global models which is a semi-structural model that we have to evaluate the impact of this unconventional fiscal policy measures as they were implemented in the euro area so this is a semi-structural model with a lot of very interesting features it has commodity production there's consumption trade it's inter-temporal it has both liquidity constraint recording households it has all kinds of bells and whistles we use it all the time we love it this is within that model we can implement and calibrate a number of very finely detailed fiscal policy tools to capture the effect of the fiscal measures that were implemented in the euro area one drawback of that model approach is that it does not have the kind of non-linear Phillips curve I argued is actually quite important so you can think of this as showing you what would happen in a more complex model but that does have a linear Phillips curve in the background so we're going to supplement this with an empirical framework that builds directly on estimating a Phillips curve environment for the euro area and for the US that will allow for this non-linearities we'll find that they are important when we think about the impact of slack on inflation they're also important when we think about the pass-through form headline inflation shocks to actual inflation so there are two dimensions of non-linearities here and both of them are important in practice we'll also allow in that framework for endogenous impact of inflation experience on inflation expectations so if you have periods of inflation above target for a long time that's going to feed into inflation expectations leading to some amount of the anchoring and we're going to obtain some insights into the differences in inflation drivers for the US and euro area by doing that so let me start with the model results so I have one slide on the model results you have three different panels on this figure here the first one is showing you in a blue the path of inflation in 2022-2024 for the euro area we have separate results for France, Germany and Italy in the paper in the absence of these fiscal measures and everything is relative to a baseline where you have no energy shock in the first place and so you have more inflation in 2023 but then as the energy supply shock goes away you would get in 2024 less inflation compared to a scenario without the energy supply shock in the first place in red we show you what happens when we feed into the model the energy and conventional fiscal policy measures what do we get? We get a little bit less inflation in 2023 and of course as these measures are unwound we get more inflation in 2024 this is what you see in the middle graph the bars reflect the difference between the blue and the red lines for inflation on the left and for output on the right we get a little bit less inflation in 2022 0.9% a little bit less inflation in 2023 0.5% and a little bit more inflation in 2024 1.5% the cumulative is slightly positive but it's relatively small so really what the measures are doing is they're reallocating inflation over time they're smoothing the inflation path keeping it closer to inflation targets now what I'm showing you on the right remember these measures are expansionary here in the model we're assuming that they're financed by maintenance we're asking well what if we close the deficit by cutting government consumption by the same amount about 3.3% of GDP in the euro area one would be the impact on output and the impact on inflation and that's what you have on the third panel so you can think of this as the impact through the model of fiscal consolidation which is additive to the other results and what you see there is that there would be some impact on output growth it would be lower output growth both in 2022 and 2023 for around 0.9% of GDP but there would be also a little bit of moderation in inflation but it's very very modest about 0.1% in 2022 0.3% in 2023 overall the fiscal consolidation so if you want to think about conventional fiscal policy has some effect that goes in the right direction but the effects are quite modest so this is not a feature of our paper this is something that you have in most of the models that look at fiscal consolidation and when you're on this somewhat not very steep part of the Phillips curve you get some impact in terms of output but you don't get that much of an impact on inflation so now let me turn to the role of non-linearities and how our results might change so we're going to estimate a Phillips curve for the euro area and for the US we have the three usual drivers of inflation core inflation here measures of labor market tightness for the euro area we're going to be using the unemployment gap for the US we're using the ratio of vacancies to unemployment the reason we're using a different measure here of labor market tightness is there is very strong evidence of a shift in the beverage curve in the US there is no corresponding shift in the euro area there is no significant profit for the US with that we have a second driver which is the path through of this headline inflation trucks to core inflation and so we're going to there are a number of channels we can think about for this path through it can come from labor markets and wage negotiations so it can come from price to price increases from energy so we're going to be using longer term expectations of inflation for the US the survey of professional forecasters, tenure forecast and for the euro area the ECBs forecast for the five years ahead inflation so here what I'm showing you is our estimated Phillips curve if you are looking at two of the two drivers that are important here the unemployment rate on the left and side and the headline inflation trucks the headline inflation to core inflation on the right in blue is the euro area in red is the US and this is a way to illustrate the extent to which we do find non-linearities in the estimated relation between these two drivers and actual inflation and if you look on the left that inflation gap versus unemployment the thing that jumps at you is that for the euro area as a whole there isn't much evidence of non-linearities here maybe in part because the euro area has not been very overheated so we've not been to the extreme left part of the curve unlike the US where you see this very sharp increase in the relationship as unemployment rates fall below say four and a half percent by contrast when you look on the right there is evidence of non-linearities in the past rule of headline inflation to core inflation both for the US and for the euro area if anything maybe even more for the euro area as you can see from the strength of the blue the blue curve here so when we put all of these things together we're allowed to decompose the actual inflation experience between 2020 and 2023 through the drivers that we have in the Phillips curve and this is what you have both on the left for the euro area and the US here the solid line is actual core inflation the dashed line is our fitted inflation through the same period and the red line is what happens if we basically only assume that there is the output gap component so we remove the headline inflation shocks from the estimation so the red line is showing you what is the component that is coming from the economy is overheating and the gap between the red and the blue is what's coming from headline inflation shocks so this figure is quite striking in our view because you have almost mirror images between the US and the euro area in the euro area most of the increase in inflation is coming from headline inflation shocks there is almost no contribution from overheating in the US by contrast most of the increase in inflation is coming from overheating of the economy almost none of the increase in inflation is coming from the headline inflation shock accumulation there was some component in the middle but by 2023 it's mostly gone now if we look ahead and now I'm going to show you three counterfactuals I'm going to first say what would have happened now that we have this framework if we remove the unconventional fiscal policy and then I'm going to ask what would have happened if the shocks had been different or if we had been in a different environment now in order to do this counterfactual we need to feed in a different value for the price of energy that would have happened if these measures so we construct these counterfactuals using some national statistical data from France and we have some data from Spain for both gas and for electricity and so we have on the right here you have the counterfactual path for energy price relative to 2020 in the absence of the fiscal policy measures at the euro area level then we can take that and we can feed that sequence of headline inflation shocks into our Phillips curve and that's going to tell us what will happen to headline what we call the inflation gap which is the gap between core inflation and expected inflation and then the final step is to figure out what would be the impact on expected inflation itself which is also a contribution to the actual inflation experiment so we do this thing in three steps now that middle step when we feed in the counterfactual path for the headline inflation shock removing the fiscal policy measures we do this also taking into account the impact this will have on aggregate demand and through aggregate demand on the output gap and through that component of the Phillips curve now what do we find so this is our first decomposition here if I look at the actual inflation experience between 2020 and 2023 this is the blue line then I remove the fiscal policy measures and so I get a higher direct headline inflation shock that's the light blue line and then I get the pass through to core inflation that brings us to the orange and red lines and then there is the aggregate demand component and expectation drift but these are relatively small so what you see is that the inflation overall would have been 2.2 percent point higher in 2022 on average about one third of this would be the direct effect two third of this would be the pass through from headline inflation to core now I'm running a little bit low on time I have a minute to wrap up and present the two three more slides 20 seconds each so then we ask what happens if we look beyond 2023 what happens if we say the headline inflation shocks there won't be any more headline inflation shocks going ahead and we're going to let the system sort of run and what would happen to inflation both with the measures and without the measures so the blue line shows you what our estimation predicts for inflation in the euro area from 2023 onward and what you see is that inflation would be coming down and would land close to inflation target by 2024 okay in red we show you what would have happened if none of the measures had been implemented and that means higher inflation 2022-2023 but an undershoot in 2024 where inflation would have fallen below a target at about 1.6 percent the right hand side shows you the difference between the two curves on average the red line says there would have been inflation would have been lower throughout the period thanks to these measures by about 0.5 percentage point as I said there was a set of fault in its circumstances two reasons why these fault in its circumstances are there one is the energy shock was very temporary so what would have happened if the energy shock was much more persistent if it was permanent so let's assume that energy prices would have remained at their October 2022 peak and gradually the fiscal measures would have to be unwound throughout 2023 because they would become unsustainable what would have happened to inflation well you can see on this figure that it would have been more inflation both of course with and without the measures that's given but mostly the measures would have led to much more persistent inflation inflation would have remained higher for much much longer and in fact in this simulation inflation expectations would start drifting upwards more significantly and so inflation would remain higher for much much longer time so the transitoryness of the energy supply shock was a really important factor in helping why understand why these measures may have been quite effective final slide I want to show you what would have happened if the economy is hot to start with well in that case we'll kind of in point B, point D part of the Phillips curve here I'm doing the counterfactual on the US so I'm starting from the US I'm feeding in measures are comparable to what happened in the euro area in terms of the fiscal impulse and in terms of the reduction of energy prices and what you see starting from the blue lines that it would bring down inflation due to the direct effect pretty quickly by 2023 inflation starts drifting up again in fact the gap by April 2023 is about 1.6 percentage point in other words the aggregate demand effect overwhelms the direct reduction of inflation so conclusion in the euro area this helped reduce inflation by about 1 percentage point in 2022 the channels are direct and the path through from a headline to core inflation we would have had needed a much larger conventional fiscal tightening to obtain similar measures so here these measures were in sense much more effective and finally there was quite a bit of luck involved the economy was not overheating and most importantly the energy price shock was very temporary as it turns out to be of course a lot of research still needed on these questions, thank you so now Giancarlo it's your turn Olivia you have only exceeded four minutes well that's the norm this morning ok Giancarlo I am sure that you will stick to the plan much for that thank you for inviting me thank you for the ECB president Lagarde it's a pleasure and a challenge to be here I start saying that the view expressed in each slide does not necessarily reflect the view expressed in the previous slides I will have a little bit of a complex discussion to present to you so I am going to do three things a small discussion of the working of the conventional fiscal policy on the paper results in key policy implications and then a context for the exercise so unconventional fiscal policy is very conventional for a political point of view so parties or governments may be worried about distribution may be worried about firms may be worried about demand and this kind of measures help also firms in the sense that for example reduce the pressure for higher wages remuneration by keeping the impact of CPI inflation or large shock or low and it is good policy it's mooting potentially the stabilizing shocks and when the shock is large people may be creative and they may be able to use the same instruments and I think they also fit what we did in the last 10 years which is we sort of gamble on future recovery gamble on better times for many very large shocks and every time we consider value and risk in our policies price stability today basically always assessing adjustment now cost adjustment tomorrow with some probability so it is rational to absorb shocks in this kind of value and risk attitude we wrote a paper for the MF with somebody at the ECB so I wanted to start with a quick example of unconventional policy in action taken from the Italian fiscal council which is a very serious and well structured institution and here we show a calculation with a micro simulation model like very rich in data basically the impact on nominal spending on consumption of a number of unconventional fiscal policy so you see to the to the left is 2022 without this policies inflation would have been 9.6 of which 7 energy with this policy went down to 5 now in 2023 with the fall in energy price and the withdrawal of the measures stay basically at 5 which is a little bit what is an example of what the Pierre Olivier was showing and also if we do by the side of income you see that this is a way also to address the nasty regressive tax effect of taxation in the sense by the side of income you see that the inflation would hit lower income much more there's a beautiful paper by Lucas Nord this is done without the substitution of goods that comes with inflation but remember poor households have much less to substitute from so it would be not far what we see in this graph is not far from reality and you see that inflation stay much lower with this measure in 2022 and it's a little bit higher in 2023 also because the the shop propagates so it goes into goods and services so this is basically one example of what Pierre Olivier was showing us very clear and it's nice an example nice because it's micro, micro-base it's a large dataset so the paper presented today by the others is at the same time very ambitious and very humble in the sense that humble is not a grand jury out whether this could have been done better like cost, risk, debt to GDP ratio, incentive, efficiency it's more about given what was done what is the effect on output and inflation and the verdict is that with some caveat successful is moving inflation shocks in 2022-2023 which I think is a very fair assessment now the results at the glance repeats what I showed you before so there is a substantial smoothing impact in 2022 that was basically paid quote-unquote with a little bit more inflation in the future and the simulation is actually good because it avoided the undershooting of inflation so this is the picture that Pierre Olivier showed a few minutes ago one thing I should say suppose we are worried about debt to GDP ratio this is a quick and dirty estimate that I did with an Excel file on the contribution of the GDP the further inflation to the ratio in Europe in the data from the result of Pierre Olivier this measure will not change from the denominator much I mean the inflation rate would have been a little bit affected but not much more on the numerator because it's deficit the dynamic would have been affected by the deficit now let's come to the model the model basically it's a philips car model where core inflation net inflation expectation regress on unemployment gap for the EA and headline inflation shocks which is basically the memory or changes in the relative price out of the medium in the past so and what is interesting is that if you compare the example before COVID and after COVID the unemployment gap becomes little bit steeper negative but remain linear it was actually not linear before but with the wrong sign it has this kind of strange positive non-linearity is linear while the memory of the inflation the propagation of inflation shock becomes non-linear the distention between the slope of the philips curve that is linear and the memory the propagation which is non-linear which then becomes dominant in all exercises I do they do a cubic model for the US where they change the unemployment gap with the vacancy ratio I do the quadratic instead of the cubic they do and the results are actually that both are non-linear it's a little different from what they show because with the cubic there is more significance there but if you look at the graph on the relevant support it's basically almost the same prediction from a quadratic instead of a cubic regression okay so what are the implications you already show you that the composition Euro area versus the United States tell a very different story mostly overheated in the United States mostly propagation of shock in the Euro area what can we say here basically propagation of that line of inflation shocks take time and it's a tie spot for policy because leaning against the memory of the shock the sacrifice ratio is high so you fight with a linear weapon a non-linear wave from the past so I needed to think better about this thing because you know the message is uncomfortable both for fiscal monetary policy I mean you can lean against the past the propagation but it's going to be costly so let me go back to the Barcelona report which is in the hole if you want to have it Shabya Birch is ready to autograph it for you the we start with the narrative of the crisis using this graph United States Euro area where output is plot against core inflation so core inflation against output Cecinepine Philip Curve it's not a Philip's Curve it's just a description but it's a useful description because it gives you three faces the black basically the COVID the reopening and the inflation crisis okay in the two countries and it's a beautiful way to organize your thinking because there is a lot of timeline here right so we have the booming good prices the good prices are tradable so there is a global component of inflation the good prices are intensive in energy and commodity so there's a global component of inflation that propagates around the world and then there is the catch up of services that you see here now it is not a Philip's Curve it is not a Philip's Curve because to be a Philip's Curve you need to take a stack a stand on the slack so where is potential output on the X axis right and unfortunately all the indicators of slack started to move differently they were very well behaved before COVID after COVID they just move differently and if you look into the slack granularly is a mess because there is sectoral differences granular misalignment in the price dynamics and conditions in each market then you have the issue of expectations and then you have the issue of energy shocks and other shifter so to bring this picture into Philip's Curve is a no job for professionals you so let me so this is amazing I actually is not very often that we have such a big misalignment in prices right in the report you have the picture going back sometimes in the last one quarter but like this we haven't seen it right so I think it's important to start with this narrative because the energy shock does not fall into a linear steady state it falls into this situation actually the energy shock is in part also fed by this situation so how do we bring how do we think about our economies so let me go to the other report not the Barcelona but the Geneva report by Veronica the for women report actually as they say so let's think of the economy as a simple structure where you have manufacturing services and wages you can do it in a network so also we upstream and downstream nominally we think it's increasing when you go down right so the shock was at the beginning a big change in the markups in opposite directions right markups go out for goods down for and it starts perpetuating propagating in the economy then the shock goes into that big mess of propagation and you know you can choose in principle any average inflation there right but with different nominally this and the situation of the COVID it's your inflation with a big disaster for the economy you want to accommodate right it's reasonable, it's rational, it's not to be ashamed of of applying good principles from you know economics to go for to minimize the efficiency cost of price deviations when there are different degrees of rigidity nominally so the question that we are now so actually you can complicate the model because with energy clearly the energy is not upstream on the manufacturing a little bit less on the services but then it goes with the intermediate immediate there is also this intermediate good transmission and then there is the feedback effect on wages which is sometimes delayed and I think this is one of the big issue now now what is the how you set inflation to govern the feedback effect of wages that are the same time you know fair for workers we can you know we cannot expect workers to accept an enormous in real wages fair for workers and good for the economy as a whole you can complicate Elisa Rubo as a Phillips with all the input output Lorenzo Niverny, Berino Eggerson add the complication to this and our report is not technical but it's a narrative but we all like singing more or less the same song here and this is exactly what Pierre Olivier has and company in the model because remember their inflation shock is the change from the medium in each industry or in some industries which mean a situation like Covid in which you know 50% of the firm may not immediately change prices the other one raise prices what equals inflation shocks record the adjustment of the price of goods at the beginning and then the coming up from the back of the price of services so should I be critical of the paper yes a little bit the discussant role is to be critical do I believe the UGAP of the IMF it's a you know there are honorable people so I should believe them but I have a problem because if I look at the measure of UGAP relative to today labor market was much tighter before the great financial crisis in the 90s and we know that the game of choosing the R star U star E star is a difficult game so I went in to try to so I also look at the picture so the regression as the black is before the Covid red is after the Covid see when you add the red in principle you go from a strange no linear to a linear so the regression will try to look at the linearity despite the shifter not because remember they have the shifter for the past inflation and what is also interesting is that today in April 23 actually it's actually pretty good the core inflation metal expectation is one of the lowest so the question is where are we today I call shorty my former student Ricardo Trezi underlying inflation I guess president from the past day in the Fed and Ricardo and other people express this opinion that the actually maybe there are other measures of tightness for Europe and especially employment ok so we have in many countries employment hours is very very high and so when I run I could not really get the linearity but I didn't have the time to actually look at it so there is an issue of how tight the labor market is today I think it's a very important question and this is not in the gap just the employment rate again I it could be high because of the shifter or it could be high because of the employment tightness the vacancy ratio I have this joke vacancies everywhere except in the statistics in Europe but it doesn't really matter because the measurement also depends suppose you have two countries right one has very quick feeling the other one has not very quick feeling the vacancy may not we could have a tight labor market independently of the shift of the beverage curve so there are issues that need to be worked out more deeply so I run out of time but let me conclude with my view ok we could be in a much tighter labor market but remember the message of the paper is that nonlinearity versus linear ok if the memory is nonlinear and your instrument is linear there is an issue there so even in a tight labor market even a tight labor market you can have a sacrifice ratio that is worth considering carefully and if I look into the black box of the A private consumption is still below trend where does the market tightness come from comes from government spending and maybe the export market so will government spending be there would or if you want even if the sacrifice if you are a little bit more nonlinear would there be a better idea to get together monetary and fiscal policy to work against inflation in this situation and then you have other questions that we discussed like ongoing effect of tightening the effect on expectation and everything but this is to me is where the conundrum of the European policy making is the fact that we also need to shift G to C again I don't know the export dynamics in the future there are shadows there that is not clear how much it is but I want to conclude just to say that despite this complication that I see in the paper for the retrospective analysis I think is a very good paper it really captures what happened of course Pierre Rivier is always saying not to be repeated without adult supervision is not something that you can replicate in all circumstances but it was actually not as crazy as initial pessimistic view were saying taking the analysis forward I think is a little bit more involved there is a lot of issues and by the way the policy mix again this is what we see in the Barcelona report I think very soon we'll have a little bit of a changing game going from this cooling down and dealing with inflation to reshaping the interaction between monitoring fiscal policy in which there will be a larger role for fiscal policy to rebalance demand there will be that sustainability which according to our calculation is possible it's a narrow path but this narrow path can only be walked down with a sensible interaction of monitoring fiscal policy whereas the guiding principle of monetary policy the monetary policy is useless if it's not credible so this is a little bit the conundrum you need to have interaction with monetary policy but monetary policy cannot interact in a virtuous way if it's not credible the issue of monetary backstop and liquidity and convenience that we discussed this morning I think is going to be central thank you very much before opening the floor Piero Olivier do you want to react to the comments made by first thank you Giancarlo very insightful just two things, two comments I want to offer first I think it's important the separate covid and the aftermath of covid and how this impacts inflation dynamics and the energy crisis of 2022 and I think we can all agree that a lot of the inflation experience that we've had in one way or another is also related to economies reopening and supply demanding balances but we are looking more narrowly at what happens in the the consequence of the energy crisis the second point I want to make on the measures of the gap because of course it's critical and here I think you're right you can look at different indicators and they point in different directions when we look at the euro area if you look at an employment maybe it looks like it's not it's a little bit tight but maybe not too much if you look at output it suggests that output in the euro area is not back to the pre-2019 trend so there's some gap there if you look at employment it looks high I think we're sort of within the range of within the range of estimates I'm not feeling too bad about where we land on the measure of tightness for the euro area stop here thank you very much Peter Olivier Eric we have the floor thanks very much Eric Nielsen for unique credit thanks very much for this IMF paper really interesting I thought persuasive if you have done this paper what would be the result a year ago or there about seeing the differences in shocks differences in economic outcome would you have recommended what has de facto been virtually identical monetary policies in the two areas Harold Uli from the University of Chicago there's a very thought provoking paper I'm still skeptical but I was mostly struck by the precision of the by the precision of the numbers given out here so the whole analysis rests on the Phillips Curve which has been a super frustrating tool for all of us if you look at the Phillips Curve it's actually a curve it's a cloud and then people try to do all kinds of tricks to resurrect the curve from that so here we have a non-linear Phillips Curve I think there was a slide on there which was estimated with great precision for example for the EU there was a 95% confidence band I mean it was I have to think about where that comes from so if I think about what goes into the sausage let's say if I think about the shock decomposition that was done here it could start from a simpler model which is a linear VAR and then try to identify the shocks in the linear VAR and do decomposition based on that in order to do that I would have to estimate the VAR and then I would have to have identification schemes and usually there's not that kind of precision that I've seen here now if I if I'm absolutely confident if I have strong priors about the VAR coefficients and the identification scheme you know maybe I can get to precise results of that sort but that raises the question whether what we are seeing is a result of priors rather than data and that gets me very nervous even if I take a theoretical model let's take a DSG model the Smets-Vauders model which is very popular was developed at the European Central Bank has been a core analytical tool at the European Central Bank for many years I think still is at this day if you take that model the agate economy is driven by seven shocks there is a Phillips curve trade-off built into the model but if you look at the actual realization of the inflation unemployment trade-off in order to replicate the Phillips cloud that we are seeing it's different shocks that come to play in fact the actual Phillips curve that we see in the data shows that inflation is driven by some set of shocks and unemployment is driven by another set of shocks and they are unrelated they are uncorrelated in fact and the Phillips curve that we like to use a theoretical menu and have been using as a theoretical menu doesn't show up in that empirical trade-off so I'm just you know we all yearn for the precision of the numbers and it's lovely to see you know if we turn of that fiscal policy and so forth and I'm just wondering whether you can illuminate how we can be certain that the numbers are as precise as you have shown them thank you very much let's complete the first round of questions with another question later thank you very much I think this is a very nice paper to put some numbers on the impacts of unconventional fiscal policy and some of the trade-offs are two comments or suggestions first I hope you will extend this paper to include the UK whether it's for this conference or another paper but the UK would be a very useful example of a country that has the inflationary external inflationary shock with an unconventional energy response as in the Euro area combined with a very tight labour market is in the US so it could be a very useful country to make the example that you rushed through at the end of the paper of how the trade-offs could be different with a very tight labour market and I think that could be particularly important for the debate in the UK where the strong unconventional fiscal response was in the context of a tight labour market and there now has been a large inflation overshoot and part of that could be because of the interaction of these policies with these non-linearities you highlight which was not in most of the inflation models so to put a number on how much of the inflation overshoot this might explain the second comment more quickly is at the start of your paper you had a very nice graph where you showed that even countries that might have done a roughly similar magnitude of this unconventional fiscal policy did it in very different ways for example some were price suppressing some were not price suppressing some were targeted, some were not targeted it looks like in your model you could differentiate the effects of these different types of policy on inflation and the growth trade-off in many of these simulations where you could say that say the type of policy followed by France versus by Germany which were very different some were price suppressing some not price suppressing did some lead to a bigger shift down in the Phillips curve in terms of bringing inflation down with a less impact on growth so less concern if you have a tight labour market so the trade-off could be quite different based on the type of policy and I think your model could be very useful to understand that if we are in this scenario again thank you Olivier take into consideration that we are entering into the coffee break I'll be brief so first the first question on monetary policy I would flip it around I mean I would say these measures have been implemented if these measures had not been implemented inflation would have been much higher monetary policy would have had to do even more probably so it's certainly not our view that what has been implemented by the ECB for instance was too much or is too much Harold on the precision there are two parts to this your point was about precision which we can all agree the precision is you know we could put some standard confidence bands on any of these things I'm not going to vouch for the second decimal of some of the estimates but I think that's true in general there is a broader question of the Phillips curve and whether we think there is a Phillips curve now on this we may agree to disagree and I think there is a lot of evidence that we do have a Phillips curve that once we do the empirical work carefully of the founders it's there it shows up in the data it has the right slope and that slope has changed actually there is a very nice paper out there that is documenting that also for the US using city level data so I'm pretty confident on this now Kristin the UK that was part of the original plan we really wanted we got close but not close enough that we felt confident putting the results there but precisely the reason you mentioned and on the granularity of the measures we could do it in the model not in the Phillips curve estimation but let me offer something there is something that is standardizing here which is you'd like to find the kind of measures that would not be price distorting because you don't want to mess up with the energy market and you know I left that out but it's in the background and at the same time you'd like to have something that maybe helps on the inflation dynamics now it's an open question as to whether you know if you send households a check and it says I cover X% of last year's energy bill so you don't have as much of a cost of living crisis but I'm not changing the marginal price you're facing if you buy an extra kilowatt then whether that affects inflation dynamics I suspect it might because it reduces the pressure to go to your boss and ask for an increase in wages or you know if you're a firm pass it on prices because you face a higher cost so I think there is something to explore there about things that are not necessarily price depressing but might still have an impact on inflation dynamics Thank you very much I think we have time for a couple of questions more Oscar Orte and I see a lady let's diversify a little bit to take into consideration the question Thank you, Silvia Ardagna from Barclays hi thanks for this great paper back to what you said on the intuition responding to Kristin I think it's very important to highlight the channel of these measures and also on the more conventional policy and here what I have in mind is what the German government for example did it by providing tax incentives to one of bonuses de facto provided incentive not to increase wages of the public employees or private employees by a larger extent and I think it can give you different results also on your persistence of inflation afterwards once you remove these measures because negotiated wages would have gone up much more which means that inflation would have been more persistent afterwards it could be important for fiscal sustainability after all a one-off tax incentive is a temporary measure if you increase the wage bill forever it's permanent and you can look back at the literature that Alberto Cina has done with various co-authors or GITA on the fiscal devaluation that would have important supply side implications by shifting the supply side and so giving you less tightness in a sense from from your results thank you very much and finally Oscar thank you very much this is a very good contribution especially for those of us in the business of doing forecasting for the euro area I like very much the analysis but it came to my mind to my attention one important difference quantitative difference of the positive impact that the removal of the measures will have on inflation in 24 if I remember well your estimate is around 1.5 extra percentage points in inflation in 24 following the removal of the measures we have an alternative estimate in the euro system I will not claim ours is better because this is a very tricky issue it's very complex many hundreds of measures in many different jurisdictions we estimated this but our estimate is roughly half of that the positive impact on inflation in 24 will be roughly half of this 1.5% and I was trying to rationalize what is behind this difference I have two potential explanations and I would like to know your view on this first at the euro system we assume that only 70% of those measures in place in 23 will be removed in 24 and the remaining 30% being removed later in 25 so this may explain part of this difference and the other has to do with the way the methodology to calculate this marginal impact I mean in our case we think that some of these measures especially those implemented via caps on prices at the time of the removal they will not be binding or they will be binding only to a very small extent the removal of these price caps will not push inflation up precisely because the level of the energy prices at that time will be lower than when these measures were implemented in 22 and 23 so to some extent this may explain this difference as a matter of fact at the end of 25 we get a negative impact on the price level an accumulated negative impact on the price level of course when you take into account the demand channels this negative impact on the price level gets somewhat reduced because of this demand channel but my view on the demand channel is that this time perhaps it will not be very strong precisely because the measures were not well targeted they were all across the board and they were too many people that really didn't need this money so people with a very low margin of propensity to consume so this leads me to think that perhaps this general equilibrium kinesian demand channel this time will not be super powerful but I would like to know your view on this well thank you very much Oscar I'll be very brief so first I will give the floor to Yanca so first tell me yes I agree that again this is a call to explore how to design these kind of measures maybe in a more efficient way Oscar we can take that bilaterally but I think you hit the right thing to be clear when we talk about the 1.5 percentage point this is the extra inflation compared to a path where you didn't have the measures at all while I think maybe some of your calculation is relative to where you were the previous year so I think that is maybe part of the difference I'll stop here very briefly the paper is actually richer in many dimensions we haven't discussed the expectation and everything so it's actually one thing that perhaps what one could draw from this is like models could be brought a little bit more complex on the propagation so we need to reach models maybe not going completely micro but the different sectors the way in which their spontaneous shocks is a little bit the game at this point just to understand better what are the what are the tradeoff and exactly the cost of intervening reflecting shocks that create a lot of disaggregation a lot of the heterogeneity so that's I think one of the big conclusion of the thank you very much to both of you I am not going to be so pretentious as to try to summarize and to sum up perhaps only to comments first fiscal policy has a role to play not to be the only getting down and second if it does not have a positive impact on inflation perhaps fiscal policy should try to avoid entering into conflict with monetary policy because it would have consequences in terms of reaction of the markets and financial stability considerations and with that Claire