 Hello, welcome back. Global warming poses a grave threat to our way of life. Or, to put it another way, our way of life, if it goes on unchecked, poses a grave threat to the planet. Last year's forum shed light on the grave macroeconomic and financial implications of our efforts to cut carbon emissions to mitigate the impact of climate change. But today we're going to ask how do those challenges affect central bankers' decisions? For a topic of such magnitude, there's relatively little research on this. Today's second session seeks to rectify that. I'd like to now introduce the session chair, ECB Executive Board Member, Frank Elderson. Mr Elderson, the floor is yours. Thank you so much, Claire, for your kind introduction. And let me take this opportunity also to thank the staff involved in the organization of this year's ECB Forum. In spite of the circumstances requiring us to have this year's forum again as an online event, and staff has managed to put together an impressive program with highly topical papers and distinguished speakers. And I am honored to moderate one of these sessions. As indeed mentioned by Claire, last year's ECB Forum featured a discussion of macro financial implications of climate change and the carbon transition. Climate change and the carbon transition, the development that the president in her opening address yesterday referred to as the most important yet least explored trend. Today, in this session, we bring this crucial topic to the core of central banking, monetary policy. In July, the ECB announced as part of the outcome of its strategy review a climate-related action plan to incorporate climate change considerations in its monetary policy. A roadmap with several areas of work, including macroeconomic modeling and changes in the operational framework for monetary policy implementation. As always, in conducting this work, we will be facts and evidence-based. This is one reason why the ECB is a very active participant in the central banks and supervisors network for greening the financial system. Central banks and supervising and exchanging views, this is what we do in the NGFS, and share practices on how to incorporate climate-related and environmental considerations in the pursuit of our mandates. But as so often has been in the case of human history, to make real progress, policymakers need to stand on the shoulders of giants. And we are very much aware that those shoulders typically can be found in the research community. And therefore, I am without a doubt that the paper and discussion in this session will be supportive to the work that we are so actively engaging in. I'm very happy to introduce to you the presenter of this session, Professor Warwick McKibbin from Australia National University. And I actually recommend all participants and listeners to visit the forum website to take note of the impressive CV of Professor McKibbin. He has worked on a wide range of topics on applied macroeconomic analysis and public policy. And here I want to draw attention to the CEPR report that he contributed to, released in September, providing a very comprehensive and thorough and, might I add, timely review of the outcome of the ECB strategy review. Today, Professor McKibbin will present some of the analytical work that informed also the CEPR report, which is already a testament to the policy relevance of his work. The paper has been jointly produced with Professor Beatrice Veda Di Mauro and Maximilian Kondrat, both from the Geneva Graduate Institute. Warwick, we look forward to your presentation. And the floor is yours for 20 minutes. Brian, thank you so much for such a kind introduction. It's a pleasure to be here and thank you to the ECB for allowing me the opportunity to present this joint research with Beatrice Veda Di Mauro and Maximilian Kondrat. What I'll do very briefly, I only have 20 minutes, so I will be giving a quick overview of the paper, present some empirical results on carbon taxes and inflation and output in Europe. Very, very briefly discuss a global model that we've developed for analyzing climate policy and monetary policy interactions and present some results looking at a number of different monetary rules and how they respond to global climate shocks, which is physical risk. Climate policy changes in the euro area, which is transition risk and transition risk from the global level where we undertake global climate policy. I've already mentioned the structure of the paper, but the goal is to look at the first of the empirical evidence of what has impacted on output of inflation from countries within Europe that have implemented carbon taxes since 1985, and then to look forward using this global economic model to explore climate shocks, climate policy, and the importance of the central bank reaction function in responding to these particular outcomes. Starting with the empirical work, if you look at the carbon taxes, there's been a number of countries, as you would be aware, that have had carbon taxes since 1985, varied across countries, implementation of different points at different rates and different coverage in the economy. What we do in the first part of the paper is firstly do the econometric analysis. We followed the approach of Metcalf and Stock. It was in the AR very recently in 2020. What we're doing is estimating an equation of the form that's on the screen. The change in various price concepts, we're using the CPI, but we look at a whole range of different measures of inflation, and we regress this on tau, the real carbon tax, in each country i, the lags of the carbon tax, and the lags of the change in the CPI, and we allow for unobserved heterogeneity specific to countries and years. The idea here is to calculate what's significant and what's the sign of these different effects. Again, there's a lot of work has gone into this part of the paper. Let me very briefly summarise it. Here are one example for GDP and two different definitions of CPI, headline CPI and core CPI. What we find is if you just run country fixed effects, the impact in the first year of the carbon tax on GDP is negative, and then that eventually dissipates over time. If you incorporate the reaction function of the central bank and additional country and time fixed effects, you find the sign changes. The best estimate here is that carbon taxes across countries and across time have marginally raised GDP, although it's fairly insignificant. Also in terms of the CPI, the CPI with a core CPI or headline CPI, the carbon tax, when you adjust for the various factors in the regression, either has a small negative impact or a small positive impact, but eventually the inflation effects dissipate. Just to give you an idea, when you look at some of the impulse response functions, here we're perturbing the carbon tax 40 euros a ton, and we find that there's a slight increase in GDP with an enormous amount of uncertainty. If you look at the responses for CPI, you see it's varying around zero in terms of core CPI and again with very large confidence bands. That's the history. Carbon taxes do impact on the economy, on the real economy as well as inflation, but it varies across countries and it varies across time. Looking forward to carbon taxes that we're likely to observe and the climate shocks are likely to be larger than what was seen in the historical record. So we move to a simulation model. The model is called G-Cubed. It's well documented. It's appeared in the Handbook of CGE Modelling with my co-author Peter Warcoxon and it's widely used in the CGE, Dynamic CGE Literature, much less inside central banking until now. What is the G-Cubed model? Well, it's a hybrid of the DSGE, the Stochastic Dynamic General Equilibrium Models that we find in central banks and the much larger scale computable general equilibrium models that are largely used for trade and tax policy issues. What's important in G-Cubed is that it models the linkages across sectors of the economy, across countries, it models international capital flows, it models inter-temporal decision-making on consumption and investment and it's very dynamic. The dynamics are at the sectoral level. It's an annual model and there's a lot of frictions in the short run and this is very important. Long-term equilibrium models are useful for asking questions about long-run equilibrium. Integrated assessment models are very useful for answering questions about long-term technology issues. But to look at the macro environment in the face of climate shocks, you do need a timeframe of 10 to 30 years which has the real-world dynamics that come from labour markets, adjustment costs and capital accumulation and the lack of mobility of labour across countries. And very importantly, the G-Cubed model has very specific fiscal rules for government spending and taxes. These matter a lot, both for the impact of monetary policy and the impact of climate policy. They also have for every country specific monetary rules which also matter in the short term for the impact of climate shocks, climate policy and the impact of government spending changes. So the monetary rules matter in the short to medium term in the modelling that doesn't matter so much in the longer term. Now what we did for this paper was we created a new version of the model. For most of the analysis we've done for the Paris Accord and other international agreement analysis for different countries, we always had Europe as an aggregate group of countries, including the UK. What we've done for this specific paper pulled the UK out in the non-Eurozone countries and put them in the rest of the OECD or rest of the advanced economies and specifically just included the countries that are in the Euro area. The model itself has a lot of desegregation and this is very important for central bankers who are worried about modelling climate change policy because climate change policy and climate shocks are about relative price changes. In this model we have 20 sectors in each region so there's 200 different firms producing in different countries. We have the model, the region, defined up into energy sectors. So there's five primary energy sectors, one of which is electricity delivery and the electricity delivery sector is using inputs from all the different technologies, coal, oil, gas, nuclear. And then we have the rest of the economy divided up into another seven sectors which are quite conventional manufacturing, agriculture, services. It's important to capture not just the desegregation on the energy system and not just the desegregation of electricity technologies but also the interaction and the changing behaviour of firms in these different sectors in the economy because it's not just the incentives on the energy side that matter. It's the change in the relative prices of the energy as it's entering into other parts of the economy. It's very important to capture those relative price effects. For those who are a little bit more technical about the production structure that we use in GQ, each sector, this is just one particular sector that's called agriculture, it has inputs of energy, capital, labour and materials. The energy inputs are a composition of clean energy coming from different electricity generating technologies which refine petroleum and these sectors that emit the CO2 which we're focused on coal, oil and gas. Each sector in each country has different shares and each sector within a given country have different elasticities of substitution. It's very important to stress that it's not just the elasticity of substitution within energy that matters, it's the elasticity of substitution across all production in the economy. The other aspect which is very important is think of this as a consumption nesting for the household, same sort of nest. Households use energy, capital, labour and material. When you change carbon taxes, it changes the relative prices of bundles of goods that households consume and it changes their behaviour and that's important when designing climate policy. You need to change the behaviour of all the actors in the economy because the more parts of the economy that can react, the less costly the policy will be. So that's a very brief overview of the modelling framework. How do we do this? Well firstly, it's complicated. What we first do is generate a world starting from 2019 to 2100. We assume that the climate policies that were in place in 2018 don't change. So there's no international agreement. This is what a continuation of the world from 2018 would look like. What's important is the assumptions about population growth by country, productivity growth by country and sector, technology assumptions, policy rules. There's a lot of assumptions that have to be made in doing this sort of scenario analysis. Again, we're not trying to do a prediction here. We're trying to learn the complexity of the interactions when we're dealing with climate change and monitoring fiscal policies. So to be very clear, we assume that in 2021, people start to understand that these climate shocks are coming. Not only the climate shocks in 2021, but they understand what the distribution might look like in the future. They also, when we model climate policy, we're assuming that there's a credible agreement in 2021 and countries will stick to the policies that have been agreed to and implement them using carbon taxes. Very important, the revenue assumptions that you use with the carbon tax, and I'll touch on that shortly. Once we get into the future, agents know what's coming, although not everybody knows it completely. So in the model, households and firms are divided between those that look at the current economic activity, the 30% of firms, sorry, 70% of firms are looking at a current economic situation. 30% of firms and households are looking forward, informing expectations based on this future simulation results of the model. The two monetary rules we look at, and in fact, you could look at many, many different policy rules, normal income targeting, price level targeting. We chose two in particular because there's too many combinations and permutations of assumptions you can make. The two that we make is one is the Hartman-Smetz Rule, which is published in the Brookings Papers on Economic Activity, and this says that the ECB follows an interest rate rule where the change in the interest rate is equal to the coefficient on inflation expected one year ahead relative to desired inflation one year ahead, plus the coefficient on the growth rate expected next year relative to the desired growth rate next year. So that's deemed by the authors to be a good representation of ECB monetary policy up until 2018. What we then do in an experiment is say, well, that's a lot of forward-lookingness in that rule. What if we change the rule so that we have half of current information and half of future information in the rule? So the idea here was to see how important it was to use contemporaneous information of what's happening in the economy rather than expected future information. And so we call that the Modified Hartman-Smetz Rule. Now, there's a lot of analysis behind this paper and appears in other papers we've published this year. The first thing we do is come up with a developer set of climate shocks following Roshan Fernando, Larry Lowe, and my own paper earlier this year where we look at climate risk scenarios and we distinguish between chronic climate change, which other studies have looked at. That is the impact on productivity growth gradually over time from the changes in temperature trends and the impact of those temperature trends on essentially productivity. And then we actually implement this new approach of looking at extreme climate events. These are hurricanes, cyclones, floods, droughts, forest fires. And so we distinguish between these types of events and it turns out that it has a big impact. Now, in terms of, in that paper, we looked at all the various representative concentration pathways. These are different assumptions about what the CO2 concentrations in the atmosphere will be by 2100. What we're doing in this paper is we're picking just one of them, which was RCP 4.5. And this is the assumption that CO2 concentrations stabilise at 650 parts per million by 2100. That will give you about 2.4 degrees warming by 2100, which would be consistent with the fact that there was no policy implemented in the baseline. And so we're trying to incorporate that. Now, I won't go through the climate shocks in detail. There's a few interesting results here, but I'll talk about them shortly. The first thing to note is, in terms of Europe, again, all the graphs I'll present will be percent deviation from what otherwise would be the case. That is, you generate the baseline, you implement the shocks or the policies, and I'm showing you the difference between the two. And I'm showing you the difference between the two across the Modified Heartsman Spets Rule and the Heartman's Spets Rule. And you can see that the main point from this chart is that these climate shocks will have long-term effects on the GDP relative to what it would have been. This is not the GDP growth rate. It's the GDP level. It will tend to depreciate the European exchange rate, and it will tend to permanently lower real interest rates in Europe. Now, that's a very quick overview because there's a lot we can talk about in the discussion. What's the policy we follow? The policy is, suppose Europe alone implements a carbon tax in 2021 at 50 euros per tonne of CO2, which is currently actually below the European trading system price by 10 euros a tonne. And this tax rises by 3% per year in real terms. Now, what does it do to the economy? These are, again, deviations from what otherwise would be the case. On the left, these are the energy prices, and you can see the biggest impact is on coal. Coal price jumps up and then tends to rise at the rate of increasing the carbon tax. And the other energy prices rise as well. In this case, I'm just looking at the Modified Heartsman Spets Rule. I'll come back to what the Heartman Spets Rule does shortly. But what you can see is that some prices move up and actually some prices move down. This relative price fall is in construction. So there's a sufficient slowdown in construction in Europe that it causes the price of construction to fall. But these relative price changes across non-energy and across energy sectors are fairly significant. What happens then under the two rules? So here we have the Europe GDP. Again, the Modified Heartsman Spets is, I'm sorry, Heartman Spets, we get this big fall in GDP. What the central bank in that rule is looking at isn't the fall in GDP growth rate, it's looking at the increase in the growth rate a year later. And so it interprets that as a reason to tighten monetary policy because the growth rate is above desired. That causes a tightening of policy which causes a slight dip in inflation and then a recovery. When you do the Modified Heartsman Spets, you know that the economy is going through this year and contraction. So you don't tighten policy as much because you're using current information. So you can see in the short term the effects are quite different in those two types of rules. And inflation can be higher or lower depending on which rule is in place. Very consistent with the empirical results. What does it do about the short-term trade-off? This chart shows you the change in inflation relative to the GDP change in Heartsman Spets. Well, I call it HS, it's easier. It's down in this quadrant and Modified Heartsman Spets is in this quadrant. So that's what happens in Europe. Now, what if all countries implement the same policy as Europe? Except we cannot do a 50 euro as a tonne carbon tax in the Middle East because of the economic outcomes are so negative the model is difficult to solve in that case and keep the Middle East as a single-country unit. And so the sector is the same policies happening across the world. What's interesting is what happens when you compare what happens in the Europe tax. So this is the one I just showed you. So GDP falls gradually over time. In the world tax GDP in Europe actually rises initially. This is because the impact of the carbon tax although exactly the same values have very big differences across countries depending on economic structures. What happens is capital actually flows out of the most impacted countries very fossil fuel intensive countries like Australia and they flow into Europe. We tend to appreciate the exchange rate increase the capital stock temporarily in Europe and that means that the global tax has a less negative impact than the European alone tax. And you can see the inflation outcomes are very similar although the world tax is a slightly higher inflation rate. Here's the story in terms of what's driving this capital flows. Capital flows into Europe for the world tax which appreciates the real exchange rate. It flows out of Europe when Europe acts alone which depreciates the exchange rate. Capital inflow is seen in the green line here for the world tax. A trade deficit is an inflow of capital to Europe. A trade surplus is an outflow of capital from Europe. What's interesting too is when you dig down this is what happens with durable manufacturing. Now what's interesting here is with the world tax there is an investment slump in all the investment that would have gone into fossil fuel industries. As it turns out Europe particularly Germany exports a lot of investment goods that are made up of durable goods. And so the world tax has a big negative impact on investment and that drives down your durable manufacturing output but it actually doesn't drive down non-durable manufacturing output. So the sectoral shifts are quite significant. This depends crucially on the assumptions you make about fiscal policy and about the investment incentives to drive renewable energy investment and other non-fossil fuel investments. So just coming to the end here across regime so again we've used the modified hardsman smet rule here because we prefer that in terms of the inflation output trade-offs. And we look at what's the cumulative effect just like the impulse response functions I showed you. What are the effects after 10 years? And you can see over across the range of variables a key one is that the climate shock has a bigger negative impact on European GDP than the European transition policy the carbon tax. And actually globally coordinated carbon tax has a much better outcome for European GDP under the assumptions we've made about the policy reaction functions of fiscal policy. And again the price level effects you can see under a global carbon tax you get higher price level effects. You get trade surpluses coming from the European carbon tax deficits coming from the global carbon tax and very very different and negative impacts on investment under the closure that we've assumed in these scenarios. So let me conclude what we found is historically carbon taxes in Europe have tended to have a short-term positive effect on headline inflation but the central bank hasn't been able to manage that. The impact on core inflation tends to be negative and that's we find and my co-authors found in a separate paper it's really a story about changing relative prices rather than a story about the overall price level for a given monetary policy. We find that the outcomes in the short run depend very importantly on what the ECB is doing and so you can get inflation or deflation depending on the policy rule. Using current information as well as future information seems to be a much better way of handling these sorts of supply side shocks and they are very much supply side shocks and inflation in the euro area can be contained in all shocks and the magnitudes we come up with are very similar to the empirical evidence in the first part of the paper. Key point is the largest cumulative negative impacts on GDP is due to physical climate risk not so much due to transition risks from carbon taxation and you can design the carbon tax policy to give you very very different outcomes in terms of economic activity and inflation that's a question about fiscal policy. Climate shocks and climate policies tend to reduce the level of GDP and in the particular examples we've simulated here you get a slumping global investment which is front loaded because the investment that would have happened in fossil fuels is now going to be brought forward and reinvested in sectors that are very small and that very difficult to expand quickly enough to offset the investment slump. It requires additional fiscal support as was shown in the IMF World Economic Outlook in October last year and papers with Florence Jamar and Larry Lewin myself the fiscal response really matters particularly in terms of infrastructure investment to turn around this collapse in private investment. So what future research do we think is important? Well really I think the coordination of monetary, fiscal and climate policies within regions, within the eurozone but also globally is very very sensitive to the sign and the size of the shocks we're talking about and so really all actors the central banks need to worry about the fiscal and the climate policies that are being put in place and each of the other policymakers need to worry about that as well and that's the end of my presentation more detail on the GQ model can be found at this website. Thank you very much. Well thank you so much Warwick for this very interesting presentation and I must say I'm intrigued by the model analysis showing how monetary policy matters for the way that materializing climate risks transmit through the economy and I'm sure that we can dwell a bit more on this in the discussion. Now let me take this opportunity to remind all of you that during the plenary discussion participants wishing to comment or ask questions should raise their virtual hands and your raised hands will reach the display that I have in front of me so that I can give you the floor. However before moving to that I will first give the floor to the Deputy Governor of Svirgis Riksbank Anna Bremen who has kindly agreed to deliver a discussion of this sessions paper. So Anna thank you so much for joining me in this session and the floor is yours for around 10 minutes and I was almost afraid that Warwick would stick exactly to the 20 minutes but the one minute that he took extra was gladly given due to the super interesting presentation he gave so Anna please 10 for you but maybe if you take one more that's fine as well. Well thank you Frank and thank you very much for inviting me here today and a warm thank you to Professor McKibbin and to your co-authors B.H.I.S. wither de Maro to provide a quick presentation of the climate change and its complex effects on the climate system. Just a few months of work on climate change and monetary policy so climate change is ultimately a scientific question so to model the economic effects of climate change we have to start with the physical science I would therefore like to show you some results from the latest So just the past few months, citizens of your area countries have experienced extreme heat, flooding, and fire. And climate scientists, they call these kind of events climate impact drivers. When addition to the three events I just mentioned, the IPCC lists another 32 such climate impact drivers in this report. It looks like this. I don't actually expect you to see the details of this slide. I just want to stress the many different channels through which climate change affects households and firms and therefore our economies. This is the physical science. So the economic perspective on climate change is that it stems from a failure to put a price on a negative externality, the emission of greenhouse gases. So the first best policy response is therefore to put a price on carbon. For example, a tax and this tax should be global. So economists want a climate policy to cause a change in the relative price between goods and services that emit greenhouse gases and those that don't. So now let's combine the physical science with economics. Successful models should combine the complex changes in the climate system with an economic model to estimate the effects on standard macro variables such as GDP, employment, inflation. And the model should also include transition risks such as the effects of carbon taxes. Clearly that is no easy task, but that is what Professor McKibbin and his quarters have been asked to do. So let's look at the research presented today. The first empirical part of this paper looks at carbon taxes in the euro area. I think I went too far, right? Yeah, sorry about that. As I've already mentioned, the purpose of carbon taxes is to change relative prices. So I think it's encouraging that this paper finds that carbon taxes in the area has done exactly that. It's changed relative prices rather than the overall price level. However, we should be careful in interpreting this result as indicative of future climate policies and effect on inflation. So first of course, and Warwick mentioned this, these taxes are low compared to the tax levels needed to meet the commitments in the Paris Agreement. And also these taxes are local. So it means that consumers can easily substitute goods produced locally with imports from other countries with lower or no carbon taxes. So global tax and carbon would give less room for substitutions, but also spur more innovation into new green technologies. So the net effects on consumer prices are uncertain. So that is why we need a general equilibrium type of model with multiple sectors and multiple countries. But that is the second part of the paper presented here today. So I would like to stress and highlight why consider the most important contributions in this paper and some further research and I'll do it from a policy perspective, from the perspective as a policymaker. First, this model is novel in the respect that it estimates the effects on variables that are relevant to central banks, in particular inflation, and it compares different monetary policy responses to climate-related shocks. And it shows that the monetary policy response matters both for output and price dynamic. But in terms of further research, I would like to see models that include policy tools that are currently used by both the ECB and the Riggs Bank, such as large-scale asset purchases, targeted lending, and collateral policies. The other nice contribution of this model is that it simulates both physical risks and transition risks. And it takes the physical science very seriously. It does both the chronic risk from climate change as well as the extreme weather events. But it does the transition risks and the physical risks separately. And as a policymaker, we're already facing a near future where multiple shocks are likely to impact inflation at the same time. And this is, in particular, the case with climate change. So let me give you an example to illustrate the challenges that we face as monetary policy makers when multiple shocks happen at the same time. And I'm going to use Swedish inflation as an example. This is CPIF. That's our preferred measure of inflation. You see the developments between 2014 and 2019. And that's a pretty nice upward trend. And then in 2020, the pandemic hit. So we faced three shocks to inflation at the same time. We had a demand shock due to the pandemic. At the same time, we had supply shocks, which were also due to the pandemic. And in Sweden, we also faced a shock to energy prices. And there was particularly electricity prices in the beginning of 2020. We had a warm, wet, and windy winter. And we have a lot of hydropower and wind power. So electricity prices fell sharply. So this is what it looked like in the past year and a half. We've had more volatility and inflation than the six years prior to the pandemic. This is our current forecast. And this is the contribution from energy prices, which is mainly actually driven by electricity prices, not so much global oil prices or gas prices. My main point here is that in real time, it is difficult to distinguish between a relative price change and an overall increase in the price level. And as a result, it will be difficult when we face multiple shocks at the same time to accurately forecast the persistence of the shock, its second round effects, and its effect on inflation expectations. And this is important when we consider the type of shocks that we're likely to face due to climate change. So coming back to this paper, I agree with the conclusion in this paper. We should not tighten monetary policy if you face transitory shocks from either carbon prices or extreme weather events. And we should focus on core measures rather than headline inflation. But even if we do so, we're likely to face some difficult trade-offs in actually setting monetary policy. So some final comments on this paper from the perspective of as a policymaker again. One thing I found very interesting in the paper is that the research shows larger effects from physical risks as compared to transition risks. So let me give you some more research. And you see that here on the slide on the physical risks from extreme weather events. Empirical research shows that extreme weather events already have a significant effect on inflation. This is particularly true for developing countries, but there is a paper here by Kimman co-authors that actually show these effects also in the United States. It's a very recent paper. And the effects come through large increases in food and energy prices. And that causes a discrepancy between headline and core inflation, exactly like Professor McKibbin could found in his paper. So why is that a challenge to us? Well, I've already mentioned one thing. In real time, it's difficult to distinguish between a relative price change and an overall increase in the price level. But also, we're likely to face high volatility. And if we get high volatility in food and energy prices, it makes forecasting more difficult. And it increases the risk of policy mistakes. It extends the time before we accurately can adjust policy to meet an inflationary or deflationary trend. And we may sometimes act when the best response is actually to do nothing. My third point was just probably the most important one. We can face a credibility problem towards households and firms. So we know that households and firms tend to form inflation expectations from changes in prices of goods and services that are salient. And food and energy prices are typically such items. So if we see substantial volatility in food and energy prices, but we as policymakers, the central banks continue to stress core prices, it can cause a large discrepancy between households' inflation expectations and central bank communication. That can severely hurt credibility. And in the worst case scenario, we risk inflation expectations becoming unanchored. So let me conclude. Monetary policy and climate policies. What are the implications for Europe? Should central banks care about climate change? Well, empirical research shows that climate change is already a threat to price stability. So therefore, we need to better model, we need to better analyze, and we also have to prepare for difficult trade-offs in setting monetary policy. Because exactly like Professor McKibben showed us, the monetary policy response will matter. Central banks do have an obligation to consider the risk that climate change posed to our economies and act in accordance with our individual mandates, but exactly how we should do that. And whether we should also adjust our policy tools to combat climate change, I'm more than happy to discuss that, but I'll leave it for the discussion. Thank you. Thank you, Anna, for this very interesting discussion. And with that, let us now move to the plenary discussion. And again, please raise your virtual hands if you wish to comment or ask questions to either of the two speakers. And as you are raising your hands, let me just ask Warwick if he wants to maybe break the ice and briefly respond to Anna's discussion. Thank you very much, Frank. And thanks, Anna. Excellent comments. You raised a number of really important points, one on volatility of prices. One issue which we focused on in a paper that was published in Oxford Review of Economic Policy with Augustus Panton, Peter Wilcoxon and Adel Morris, who's just moved to the Fed to work on climate change inside the Fed. We argued that the projections of potential output will start to deteriorate. And one of the key aspects of inflation forecasting and inflation targeting framework is the ability to forecast potential output. And if potential output is difficult to forecast, then central banks will get the forecast on inflation wrong, and that will also undermine credibility. And we're seeing that in the data in the US. The Richmond Fed has a study which shows the forecasting errors of potential output have been much larger in the last decade than in previous decades. So I think that's an important point. The forecasting capacity of central banks, I think they can forecast nominal income better than they can forecast inflation in a climate changing world. And so we argue in that paper that central banks should really be making forecasts of nominal growth and targeting nominal growth. That enables them to decipher between relative price shocks and absolute price shocks. It doesn't really matter for the inflation forecast. So if you miss, if you get a negative climate shock and output is lower, then you thought it would be that inflation will be higher if you have achieved your nominal growth target. So I think there's a serious debate about shifting the monetary focus from purely inflation targeting to more of a nominal GDP or some other measure of nominal growth. And I think that's an important issue that I think central bankers should be and probably are debating. A final point if I could just mention, we do have a full vector of assets in this model. So the share prices of different companies are impacted and central banks could actually do share purchases or other sort of asset purchases. We don't model it in this paper, but in fact we could model that as part of our future research. It's already in the model. We just don't use quantitative easing or other non-conventional. Now it's conventional, but we're focusing here on interest rate policy. Well, thank you very much, Warwick. And, you know, I'm delighted to see that questions are coming in. So I will be calling first on Elga Barge, who is the head of macro research at BlackRock. And then maybe I can remind you that, you know, please try to stick to one minute, one and a half minutes maybe for your question. After you Elga, I will go to Diego Kensig. But there's room for more questions. So please don't be shy. Elga, the floor is yours. Thank you very much for this hugely interesting presentation and also the discussion of it. I just want to, we positioned the discussion a little bit to say that climate change is real. That means the inflation implications of climate change itself should be our baseline. And then I think from there on we need to think about how that changes as we are transitioning to a low carbon economy. And I'd be very interested in getting views on what difference it makes, whether we sort of commit to net zero and sort of much more aggressive in what kind of changes in relative prices and in the shadow price of carbon we need to achieve. And whether or not the main difference in policy reactions by monetary policy is not really whether the central bank is targeting headline inflation or core inflation. Everything else could just be of second order magnitude in terms of how central banks react to that. So I think we can no longer, also when we talk about inflation, ignore the fact that our baseline includes already pretty material climate damages and just changes to the inflation baseline. We're very interested of hearing the views from both speakers. Thank you. Thank you, Elgai. And thank you for these very, very good questions, but also for breaking the ice. Warwick, can I first turn to you and then maybe after you Ana would like to chip in as well. Yeah, thanks very much for the question. So we actually explored net zero in the IMF World Economic Outlook using this model in October last year and also in a new research paper that Florence Jamar and myself and Larry Lou put out as an IMF working paper a few months ago. Net zero is actually much deeper cuts than we explored in this particular example. And getting there isn't difficult as long as you're willing to put in place other policies to stabilise or assist with the adjustment. And that is really a story of fiscal policy primarily and infrastructure investment fixing the market failures on the pricing of carbon, but also fixing the market rate failures on the provision of infrastructure capital that enables the climate transition and monetary policies role is just to stabilise around that. And I think to be careful about the financial instability that can come from stranded assets and the sort of issues that the regulators are worried about. But I think it really is that's why I said it's fiscal and climate and monetary all have to be coordinated and the assumptions you make about the reactions of each of those policy makers makes a big impact on the other policy makers. Thank you, Warwick. Ana, would you like to also answer the questions? Yeah, but I agree with Warwick that fiscal policy is going to be the main driver here. But I think as central bankers we still have to consider how we fit in compared to all these policies that are enacted or will be enacted by the political side. And I think it's important that we're not seeing ourselves completely on the sidelines because there's a risk that we actually counteract the policies that are being enacted by other parts of the economic policymaking. So if we don't consider, for example, the purchases that we do in terms of large-scale asset purchases, we might actually slow down the transition. We don't want to be on the side of slowing down the transition. To put it simply, we know that our primary mandate is price stability. But we have an obligation to consider the risk that we put on our balance sheet. And that gives us again an obligation to consider the kinds of assets that we have. And then we also have to consider that we should not actively go against policies enacted by other policy makers. Thank you. Thank you, Ana. And now it's my pleasure to turn to Diego Cantig, who is a PhD student at London Business School. You are already on the screen, Diego. But I want to compliment you because I think the more students and PhD students who are here in the virtual audience who actually dare to speak up, the better and the more interesting this will be. So please go ahead and give us your thoughts and your questions. Great. Thanks so much, Frank. First of all, let me say a very interesting presentation and discussion. I really enjoyed it. My question concerns this finding that I found very intriguing that the monetary policy rule matters a lot for the transmission of climate policy. You looked at these two different rules that you also consider other rules. And could you also use your framework maybe to study the optimal monetary policy? And then my second question is related to fiscal policy that you also hinted at plays an important role. So there, can you elaborate maybe a bit more what the effect could be if the revenues are rebated to households, for instance, by lowering taxes or by direct transfers? Thank you very much. Thank you, Diego, for your questions. And maybe again, I will keep to the order of first Warwick and then Danaana. Yeah, thanks. Very good question. So yeah, there's a lot of different monetary rules that we could consider. My favorite when we've been doing these analyses is a nominal GDP rule because you don't need as much information as you do in an inflation targeting framework. You don't need to forecast potential output as well. On the fiscal rule, what's really important is what you do with the revenue. So in the paper, we made the assumption that the revenue is used to reduce the budget deficit permanently. And if we change that to lump sum transfers to households, it changes the consumption impacts. If you change it to infrastructure investment, it has a very big impact in the medium term productivity growth of the economy as long as that investment is productive. And that is all incorporated in the IMF World Economic Outlook chapter, as well as in a number of other papers that are on my website. But revenue recycling matters a lot on the outcomes. Thank you, Warwick. Anna, maybe you want to pitch in as well. Yeah, well, of course, I mean, when we have excellent graduate students being interested in this topic, I think I would just like to encourage them to look into all the different kinds of policy tools the central banks are using today to see if we could find different effects. And if it matters whether it's large-scale asset purchases or whether it's interest rates changes or if it's targeted lending, I would like to see much more research on this. So I encourage you to follow that road yourself. Very well, and a very, very good advice, Anna. Next question is going to be asked by Neville Hill, who is the Chief European Economist of Credit Suisse. And before you take the floor, Neville, let me just say that there is room for more questions. So if there are anyone still under the audience who wants to take this opportunity, please don't feel shy. This is the moment to really contribute. But please, Neville, please go ahead. Great. Thank you very much. And thank you for what's been an absolutely fascinating discussion. I guess I'd be interested in what you think the sort of practical consequences for monetary policy will be in terms of what a central bank like the ECB would practically do in the event of a substantial increase in carbon taxes and carbon prices. It strikes me that one of the key challenges here is to use a monetary policy rule that perhaps focuses on the downside risks to economic output rather than focusing on the overshoot of inflation. And so maybe one idea that I'd be interested in hearing your thoughts on is whether it would be sensible for central banks to articulate in advance what they think the impact of substantial increases in carbon taxes will be on growth and inflation, critically, and consequently articulate what their policy response will be in order to allow inflation expectations to remain anchored at the same time as potentially inflation overshoots for a number of years, their target, and monetary policy I guess remains in a more expansionary direction in order to mitigate the negative effects of carbon taxes on growth if we're unlucky enough to have a fiscal policy response that you've got to your model where policy is effectively tightened. Thank you. Wonderful. Thank you very much, Neil. Warwick, Anna. Well, very quickly, I think the key issue here, I think, is the design of the monetary framework. When there's rigidities in the economy, this is where the economic costs come from, and the rigidities that we're picking up are the fixity of physical capital in particular sectors. So when you're changing these relative prices, you stop investing in those sectors, but that capital can't move, it's there, it just loses value, and new investment will go to new sectors. The other issue is if you're targeting inflation and you have a big spike in one price, then you have to have a contraction in another price to get back to your inflation target. So some sectors are going to have to be compressed to get inflation to fall back into target. And so that's an important issue, because that combines the elasticities of substitution and the ease of moving capital across sectors matter a lot in that case. And so really, I think, again, if you have either the rule like we have, the modified Hartzmann smets, which trades off explicitly inflation deviations from target versus output deviation from target, that builds in an explicit trade off. But you could also do normal income targeting or normal income growth targeting, and that to me is much more attractive in terms of maintaining credibility and Anna made that important point that the key issue here is central banks don't lose credibility in their monetary frameworks, because that'll unhook inflation expectations. Thank you, Warwick. Anna, please, if you want. Yeah, but I was going to say that I do think that the most important factor here is actually to keep inflation expectations well anchored. But I do think it can be a challenge. And as I mentioned, that's why I stress that both Professor McKibbin and if you look at other empirical research, they will define that the effects from climate change on prices is likely to come mainly through food and electricity and energy prices in a wider sense. And that is going to be a challenge for us. Because if you also think about the discussions that we have today on income inequality, which households will be the most hit if we see much higher food and energy prices will be low income households. So we will have to focus on core prices, but we will have to communicate that we are aware of these changes. And hopefully, but that's just a hope, we will see the distribution of the taxes going to the right being used the revenues used in the right ways to spur innovation to spur infrastructure and to transfer those things back to the people that will be the most hurt by these taxes. But we have to be careful, because in the end, our mandate is to maintain price stability. So if we see a long period of inflation overshooting the target, we'll have to respond to that to keep inflation expectations anchored. So I think volatility is is is a big risk. And we have to consider that. And if you know it in advance, it's going to be easier in terms of communication. But it is it's a challenge to monitor policy. Thank you, Anna. Then I think the next question will go to to Glenn Rudebusch. Thank you. Thank you for a great discussion. I want to second something that Frank mentioned about the importance of research. Central banks have been important sources, important centers for economic research, and climate economics, climate finance, climate risk. These are important areas for further research. And I think it's it's a little hard for central banks, in part because even for the economics profession more broadly, because of course, climate implications cut across international labor, investment, consumption, financial markets. So there's that all in interdisciplinary nature of the work that's needed. So that I think is important. And and Warwick's work is is a great step in that regard. More subsequently, I also want to agree with Anna that we don't, you know, how should central banks care about climate change? It's not just a matter of being reactive to output or prices and to think about output dynamics and price dynamics, but also for many central banks who have a more direct motive, really to support climate policies as well that all in government approach. So it's not just a matter of of reacting, but also also contributing. And finally, a question for Warwick is you've talked a lot and this was really about monetary policy, but there must be a connection with financial policy. So we've talked a lot about there's been a lot of traction with regard to climate risk and thinking about financial policies to mitigate climate risk. But there might be some interconnection. There's a, you know, coordination between that monetary policy and and and, you know, supervisory and financial policies as well. Thank you. Thank you, Glenn. Maybe I can ask Warwick in order to be very short and then we can see whether we still have time for one more question or not. But but please go ahead. Good to see you, Glenn. Thanks for the question. Very briefly, carbon taxes aren't my actual favorite way of pricing carbon. We have an approach called the climate asset and liability mechanism, which focuses on stabilizing the balance sheets of firms and households in a way in which the revenue doesn't go to the government, but it goes through the asset markets to the balance sheets. No time to discuss it here, but it used to be called the hybrid approach, the McKibben-Wilcoxon hybrid. It's now called the climate asset and liability mechanism. Happy to send you some material on that. But I think that deals with the issue just address. Thank you, Warwick. Anna. Well, I very much agree with the point that this kind of research is is needed and it will help us make better policies when we face greater threats from climate change. But I would like to stress from a central bank perspective, the risk management perspective. We have large balance sheets in many countries now. And we have we have an obligation to consider the risks on our balance sheets. And climate change is a threat to different kinds of assets. And therefore we have to act in accordance with those risks that we see. One important thing and that has been done by central banks is to demand better disclosures from firms, because that will help us in in in addressing those risks and making decisions on on what to put on our balance sheets. Thank you, Anna. Oliver Kiesige is our next. We'll ask the next question. Oliver, please. Hi. Yes, I actually have a question regarding the model. My understanding is that you're marrying like a traditional economics model with a climate model. And this economic model is usually in a model with risk neutral agents. So climate risk really strikes me as having certainly like an idiosyncratic component, but also an aggregate risk component. So I suspect if you were to take risk very seriously in the model that you probably even get a sort of an amplification of your effects that you find through this risk channel and risk premium that it demands. Thank you, Oliver. Warwick. That's a that's a very good question. We actually do take it seriously. We have risk premium throughout the various arbitrage conditions, not derived from the utility function, but put into make the model replicate the database in particular year. And in the paper with Fernando and Lou, where we model climate shocks and climate change, we actually model changes in risk premium based on the impact, the historical data on the impact of climate events on stock market values in different sectors in different countries and build that changes in risk premium into the into the shocks in addition to the climate shocks. And they're actually, as you suspected, they actually can be much bigger, in fact, than the climate shocks themselves. Going forward, the historical we use the historical data to estimate that relationship and then use the forecast of future temperatures to then change the shocks in the future, which have these equity and asset asset market effects. But it's a very good question. And I referred to that paper that's it's on the karma of my website, the Australian National University. Thank you. Thank you, Warwick. I'm going to use my privilege as chair of this panel to ignore the time because I see that James Bollard is on the screen. So you are going to have the privilege of having the last question is great to see here. Please, go ahead. Thank you, Frank. And thanks to the forum. I thought this was a great session. I just want to encourage Warwick in his research and his co-authors. I think this is really great and it serves to focus the issue. There often is a lot of hyperbole around climate issues, but this is really getting down to brass tacks about how would this transition really work and how would the interaction of fiscal and monetary policy work. I love the emphasis on nominal GDP targeting. I myself have tried to emphasize nominal GDP targeting might be a better approach to monetary policy. And I see flexible average inflation targeting in the U.S. as being a step in that direction, probably not as recognizable as the economists would like, but I do think it is a step in that direction. And then my question would be, and you've been asked this in several ways, but our carbon tax is really the optimal policy here. So I guess the way I would ask it is you know, in the long run does this mitigate the climate risk or stabilize the climate risk or reverse it possibly in the very long run? And I realize you may not be able to say that in the big model, but maybe in some stripped down model with the right kinds of transfers back to households and businesses, is this the right fiscal policy mix in this model so that you really are talking about optimal monetary fiscal policy? Thanks a lot. Warwick, I will go to you, then I'll ask Anna and then I'll rip up and we will end up with Claire. Warwick, please. Thanks very much. Thanks very much, Jim. It's good to see you. So again, as I mentioned, I think if you take a broader view than just economic optimal, what you need is a policy that's politically sustainable. And what the climate national liability mechanism does is it gives you a credible future carbon price and it creates a political constituency because the balance sheets of firms and households have the long-term carbon assets built into them and therefore they're less likely to convince the government to drop the carbon pricing policy. So if you want political credibility as well as economic optimality, I'd refer to the climate asset liability mechanism as a way of achieving that. What you'd lose is the revenue that would otherwise be able to be used for infrastructure investment, but you can do a mix of the two. And I think that's important that we just fixate on carbon taxes. There are many different ways to price carbon and I think the climate asset liability mechanism in my view is the best one. Thank you, Warwick. Anna. Well, the key thing is to get a price on this negative externality that we have, which is the emission of greenhouse gases. The exact form has to be decided by politicians, preferably in international agreement. And I agree with Warwick that the key question is to find a solution that is politically sustainable in the longer run. That's it. Well, thank you again. Anna, Warwick, all those who have intervened, who have dared to intervene have brought their thoughts and their questions. Let me just throw a couple of things back to you that I heard. Climate change is real. Climate change is a hot topic. We don't want to be on the side of those who slow down the transition. Climate change is a threat to price stability. These were just some of the things that came to the fore in the last very interesting hour in which I think we have brought climate change. And if you allow me the climate crisis to the core, to the very heart of central banking and who would have thought that only three to four years ago and maybe in some time in the future when we look back, we will see that not only climate change but also broader biodiversity loss will come part of the debate of central bankers. And with that, I want to thank you a lot and over to you, Claire. Thank you, Mr. Eldersen. You did a really excellent job of moderating what was a very, very good session. I'd like to thank Warwick and Anna for doing a supremely good work in terms of building on what we saw last year. I think you saw from the questions, this is an issue that's very much on people's minds. Again, I don't envy the task of central bankers. It's yet another thing along with a very, you know, uncertain outlook for inflation in general that they will have to consider in the policymaking process. We're going to take a break right now. We'll return at 1630 central European time. We'll be looking at the impact of unemployment and inequality on the work of central banks. Join us again then.