 Well thanks very much for for organizing this conference and for including our our paper on the program. So this is joint work with with Kevin Novin and Will Peterman and we have the usual disclaimer that the views in this paper are not necessarily the views of the fact. So the US does not currently have a federal climate policy in place and by that I mean we don't have a nationwide tax on carbon emissions or a national cap and trade system but there is widespread awareness that we could introduce a climate policy at some point in the future. And so for example if we just think back over the past decade or so there have been several serious legislative attempts to get federal action on climate change in the US. Going back to 2009 the US passed the waxman market the house passed the waxman market though which of it had also passed in the senate would have established a cap and trade system for the US. Then there was the clean power plan under the Obama administration and where recently Biden has strongly signaled his interest in taking action on climate change by rejoining the the Paris Climate Agreement and announcing the goal to reduce US emissions by about 15% by 2030. So we're going to use this term climate policy risk to refer to the possibility that we could introduce a climate policy at some points in the future. And we think this climate policy risk could have important implications for the macro economy because it affects firms decisions to invest in long-lived capital assets especially those that are used in in conjunction with fossil fuels. So for example think about a firm that wants to invest in a new coal boiler. The return on that coal boiler is going to be a lot lower if the government introduces a carbon tax in the future than if the government doesn't introduce that tax. So we can think of this climate policy risk as creating this downside risk for investment in things like coal boilers. But parallel but opposite reasoning would suggest that it creates upside risk for investment in things like solar panels. It's a large share of the US capital stock about a fifth by our estimates is is kind of specialized to use fossil fuel. We think this policy risk could have important macro implications. And so that's what we're going to look at in this paper. Think about what are the effects of climate policy risk on the macro economy that's going to operate through these implications for investment. And then we'll look at the resulting impact on emissions output and welfare. So to do this we develop a general equilibrium model of climate policy risk in which entrepreneurs can invest in different types of capital. So think like a clean capital and then a fossil capital. Throughout I'm going to use a carbon tax as an example of that future climate policy. Should really do this though as kind of a stand-in for any sort of coordinated federal action on climate change or to be a carbon tax, a capital trade system or some kind of suite of regulations. And our general approach is going to be to study an equilibrium in which there is no carbon tax in place. But there's this chance that the government could introduce a carbon tax at some point in the future. And so what we find is that the risk of a future climate policy alone even without the actual carbon tax in place gives you a small reduction in emissions. And this much introduction occurs because that policy risk is put a shift investment towards cleaner capital and reduce the total level of the capital stock. So in essence we're reducing emissions both because we're just producing less and then while we are producing we're producing more cleanly. This also will affect how we think about the cost of an action on climate policy. In some sense there's this silver lining in that we're reducing emissions even though we don't have an actual policy in place. But on the other hand what we'll show is that this is a very costly way to reduce emissions through this policy risk relative to just using an actual tax. And then the last thing we'll do is compare our model to the literature on the green paradox. And the literature on the green paradox comes up with the opposite prediction in the sense that they argue or that literature argues that the possibility of climate policy in the future will actually increase emissions today. And that rationale is coming from how policy risk affects the extraction incentives for fossil fuel. Viewing fossil fuels is resourcing in finite supply. So for the majority of out paper we're going to take the view that the scarcity constraints on fossil fuels are not binding. And so we're going to treat fossil fuels as a resource in infinite supply so we won't have this green paradox mechanism. But then at the end we're going to come back and say what if we put the green paradox mechanisms into our model and run kind of a horse race between these two predictions. And what we'll find is that even if you allow for fossil fuel extraction to respond to the policy risk the effect the decrease in demand for fossil fuel coming from these investment chain channels is going to dominate. And so we'll still find that policy risk reduces emissions today. So this relates to both the environmental literature particularly that literature on the green paradox and a macro literature on policy uncertainty. So what I'm going to do is start with the with a simple model and the purpose of the simple model is to highlight the different channels through which climate policy risk affects the economy. So for this case the simple case will just focus on the production side of the economy. So we'll think about a world inhabited by infinitely lived entrepreneurs that are making investment decisions to maximize expected lifetime profits. And these entrepreneurs have the opportunity to invest in two types of capital. First there's a fossil capital. This is capital that produces fossil fuels so I think like an oil rig or requires fossil fuel to operate. So say a coal boiler. And using the the detailed data on capital assets from the the National Income and Product Council we find that about a fifth of the US capital stock would fall in this fossil fossil capital category. The opposite fossil capital we have clean capital. This is capital that's specialized either to replace fossil capital so think like a solar panel instead of a coal boiler or to replace fossil fuel. And by that what we really have in mind are improvements in energy efficiency. So for example if I put better insulation in my house then I can use less fossil fuel for heating it cooling. So it'll be a unique final good which is why produced from a clean intermediate XC, a fossil intermediate XF, and then labor which is just going to be fixed in an exogenous supply for our simple model. The clean intermediate is produced from clean capital and the fossil intermediate is produced from a Leontia function of fossil capital and fossil fuel. Now the Leontia implies that there's no substitutability between the fossil capital and the fossil fuel. If I have a coal boiler I have to shovel in a certain amount of coal in order to produce the electricity. But so in the model all substitution away from fossil fuel is going to come through increases in clean capital. But remember that clean capital includes both improvements in energy efficiency and renewable energy which empirically are the two ways that we can substitute away from fossil fuel. So we just model the investment required for that substitution directly. And then again for the majority of the paper we're going to think about fossil fuel being produced being a resource and an infinite supply and so it's just produced from units of final good at a constant marginal cost. So what we'll do is we're thinking about the U.S. economy as being in this equilibrium in which we don't have a carbon tax in place but there's this chance that we could introduce a carbon tax at some point in the future. So with probability one minus row we'll assume the government does not introduce the carbon tax and so we would then stay in this equilibrium in which there's no tax in place but there's risk of the tax in the future. Now with probability row the government does introduce the carbon tax. In that case the economy begins the dynamic transition to a new long run equilibrium with the carbon tax in place. Now since climate policies are at least designed to be permanent we'll model that introduction of the carbon tax as an absorbing state and so once the carbon tax is introduced all uncertainties resolve and that transition and the resulting policies they are entirely deterministic. Now of course we can see policies, climate policies are designed to be permanent but they don't always end up that way so we know Australia's carbon tax because repealed after being in place for two years this is kind of a case in point there and so what we do in the paper what we won't go through in the talk is show that that assumption really isn't critical for the quantitative results and if we think about policies ending after one, two, or three years we get relatively similar effects. So what I want to do now is solve analytically for this steady state in which there's no carbon tax but there's this chance that we could introduce the carbon tax in the future and look at how the allocation of capital affects or how that policy risk affects emissions through its effects on capital. So this is showing you the ratio of clean to fossil capital in that steady state. Now for a minute let's just pretend that the extraction cost of fossil fuel is zero and that there's no policy risks except this row tau term to zero. Then the ratio of clean to fossil capital just equals the ratio of the factor shares in that cost of production function. Now there is an operating cost so let's make this data guide positive in that case sorry extraction cost of fossil fuel. That extraction cost of fossil fuel raises the operating cost of fossil capital which shifts the economy more towards towards clean capital so increasing its ratio. Now the possibility of a future carbon tax this term here is going to do the same thing it raises the expected operating cost of fossil capital and so that also is going to shift the economy towards clear capital. So we see climate policy risk reducing emissions simply because it's changing the composition of capital it's giving us more clean capital and less fossil. The other channel is what we'll think of as a level effect and so this expression is showing you the total level of capital in the economy and the important part here is that the level of capital is decreasing in this policy risk term. To understand the intuition think about what an actual carbon tax does. An actual carbon tax is going to shift the economy away from the privately optimal not the socially optimal allocation of capital which is going to reduce the marginal product of capital which then in turn results in a lower total capital stock. Now the possibility of a carbon tax in the future does the same thing in expectation and that it reduces the expected marginal product capital leading to a lower capital stock. So in essence here we see that policy risk is reducing emissions both because we're just producing less this is the level effect and what we're all producing we're producing more clean money which is that composition. So what I do next is kind of highlight two pieces of the quantitative quantitative model so that's our simple model and now to get some numerical realism we'll make it a little more complicated. On the production side we'll add a non-energy sector that will offer labor to be allocated across the different sectors and then we'll solve for the full general equilibrium and rich risk of first households are making all the decisions. So on the production we'll now think about output being produced from a CES function of three intermediate inputs the clean the fossil and this new one which is a non-energy intermediate. The non-energy intermediate is going to be produced from non-energy capital as well as labor. Now that non-energy capital accounts for all capital that we wouldn't necessarily classify as fossil. It's not specialized to use or produce fossil fuel and we also wouldn't classify it as clean and that it's not specialized to kind of replace fossil fuel. So for example thinking about a sewing machine. A sewing machine requires electricity to operate but the sewing machine doesn't care if that electricity is produced by a coal boiler or by a solar panel. So the sewing machine will classify as what counts as this non-energy capital. The other change we've made aside from adding more parameter values is we're now allowing for the allocation of labor across these different intermediate inputs and this is important because it's giving entrepreneurs another dimension along which to respond to the policy risk and the kind of labor response is going to differ from the capital response because the timing of labor and capital decisions differ. In particular you have to decide next period's level of capital i.e. your investment at the end of the current period before you know if there's going to be a tax next period but labor is a much more flexible input. And so we assume that you choose labor for the current period after you know whether or not there's a tax this period. So this labor choice lets you kind of mitigate the effects of climate policy risks. For example if I hired had a lot of clean capital thinking there might be a tax I find out there's not a tax I might hire a little bit less labor but I would normally hire to go with that level of clean capital to adjust my production ex post. The other PCA is now the economy is going to be inhabited by a continuum of infinitely lived, identical households that are comprised of workers and entrepreneurs. The workers endogenously choose their labor supply, exploit any entrepreneur and they earn the market wage. So labor markets here are perfectly competitive and entrepreneurs like before producing intermediate inputs and everyone is making decisions to to maximize the expected lifetime welfare of the household. So we want to calibrate the model and our goal again is to take the view that the US economy is currently in this equilibrium in which we don't have a carbon tax in place but there's this chance that we could introduce a carbon tax in the future. So we're going to set the size of that potential tax equal to $45 a ton but in line with the current policy proposals and estimates of the social costs of carbon. The time period is a year and what I want to do in the interest of time is not go through the whole calibration but try to give you a sense of how we're going to pin down and best we can the probability row that firms place on that $45 tax in the future. So what we observe when we look at the data is that US firms are voluntarily reducing their emissions kind of beyond what would be required by any sort of local level policy and there are also different ways they do this but our general approach is to say well what if we can quantify those firms abatement efforts using a mechanism known as an internal carbon fee which is simply a tax that the firm places on its emissions and if we know how much firms are voluntarily reducing their emissions then we want to say okay we see firms are doing this amount of emissions reduction voluntarily what probability row would they have to place on that $45 tax so that that voluntary emissions reduction is the optimal response the climate policy risk comes in the context of our model and so what we find here is that row is about 10% and so there's kind of this 10% chance every year that we'll put on that $45 tax or put another way this would imply about a kind of 50% chance that the US introduces such a tax and within the next eight years. Okay so I want to turn next to to our results so goal is to use the the calibrated model to do three things first to show that climate policy risk gives you this small reduction in emissions even though we don't have an actual policy in place and that because it's a misdirected it can affect how we think about the cost of an action on climate policy and then finally we'll come back and say how does this compare to what if we bring in the green paradox and can we run kind of a simplified horse race between our model and the predictions of the green paradox so we're going to do is compare three study states first is going to be a steady state in which there's no tax and there's no risk of of a future tax second is going to be a steady state in which we have that $45 tax in place that's our policy studies comparing these two study states provides a familiar point of reference this is the exercise that's normally done in the in the literature then to understand the effects of climate policy risk will add in that third study state in which there's no tax but there is risk of a future tax specifically there's this 10% chance of a $45 tax comparing the first and the third study state let's us get at the effects of risk so let's start with that familiar reference point this says that we put on the carbon tax we reduce emissions by about 16 1516 percent relative to a world with no tax and no risk and as you can see here that emissions reduction is coming through that same level channel that we discussed before the capital stock is lower and the composition channels tax shifts the economy towards cleaner capital labor and intermediates just like we talked about in the context of the policy risk now what I want to do is is add to this graph what are the effects of just the risk of that $45 tax alone not the actual tax and so that's the blue bars and the first thing to notice about the blue bars is they all move in the same direction as the green bars and that there's the sense in which climate policy risk takes the economy partway to that long-run study state with the tax in place now if we look to your focus just on emissions what you see is that policy the policy risk gives you a small reduction in emissions and to put that in perspective kind of the height of the blue bar there is about eight percent the height of the green bar meaning that eight percent of the emissions reduction that we would get from actually putting on the tax we get just from the possibility of the tax alone now what I want to do next is zoom in on these on these blue bars because they're kind of hard to see with with the green bars on the same graph so these are the same blue bars but we've changed the the y-axis and what you can see here is that climate policy risk is operating through again the level effect so we have just less capital and through the composition effect to reduce to reduce emissions now notice here with the composition effect that the if that climate policy risk leads to a larger increase in the ratio of clean to fossil capital than it does in the ratio of clean to fossil labor and that's coming back to the different timing of the capital and labor decisions again I have to decide clean capital what I before I know whether or not the government has introduced that carbon tax after I learned the government didn't introduce the carbon tax I can hire a little bit less labor to kind of adjust my my production X post so you see clean to fossil labor rising by less than clean to fossil capital what this is going to imply is that climate policy risk distorts the capital labor ratio and that it creates this kind of misallocation of capital labor so what I think about next is how does reducing emissions through policy risk compare to reducing emissions through an actual tax and so what we do is we solve for the tax that would give you the exact same emissions reduction that we're getting under this policy risk and that tax equals three dollars and and twenty one cents a ton and then we can compare how the tax operates through those levels and composition channels differently from the risk and so those are the remaining bars and the key point here is that look at that level effect the tax is relying much less on the level effect to reduce emissions than policy risk another way that means that the tax is relying less on just shutting down production to reduce emissions and kind of more on shifting that economy towards cleaner capital which is going to in turn make the tax a much less costly way to reduce emissions now there are a couple reasons why the the level effect is higher under under policy risk or policy this leads to a larger reduction in capital but what I want to highlight here comes back to the to the kind of different responses and the ratios of clean to fossil capital and clean to fossil labor and in particular we see that clean to fossil um under the tax there's no uncertainty and so notice the height of of these purple bars is all the same the different timing of the capital labor decisions doesn't matter and so the tax doesn't distort the the capital labor ratio but the policy risk does distort the capital labor ratio and that distortion further reduces the marginal product capital which is part of what gives you that larger level effect under the tax so I think about what are the implications of this policy risk for the cost of inaction on climate change and again in some sense maybe there's kind of a silver lining here that the costs of inaction are smaller than we thought because we're getting small emissions reduction just from the chance of that policy even though we don't actually have the policy in place but what we can show is you know their cost of this risk too and using a consumption equivalent variation measure of welfare we find that climate policy risk is about twice as costly as that three dollar and twenty one cent tax there are a couple reasons for this higher cost first is that households are risk adverse they just don't like that cost and second is that larger level effect and that policy risk is relying more on just shutting down production and hence reducing consumption than the taxes to reduce emissions now when we first started this paper we had this notion that you know if if agents anticipate that we might introduce a carbon tax in the future maybe that would make the welfare cost over the transition lower than if agents don't anticipate that the kind of reasoning there being that because we we anticipate the the possibility of a carbon tax the economy has already done some of the some of the adjustment and what we find is that actually that's not true and that the welfare cost of introducing a carbon tax beginning from a world in which agents would anticipate that we could do this so in that steady state with risk is almost exactly the same as if we were to begin from a steady state with with no risk and the agents can don't anticipate and the reason is that there are these two offsetting channels first if you start in that steady state with risk it is true that the economy you know some of the adjustment has already taken place and so that itself is going to lower the the welfare costs but then offsetting that is if you start in that steady state with risk the level effect implies that you have a lower capital stock and this means you're less able to disable over that transition period and that's going to increase the welfare costs and those two effects almost perfectly offset and so the welfare costs of introducing a carbon tax are virtually the same whether you begin in that steady state in which agents understand that there's this risk of a future tax or if you begin in a steady state where there's where there's no risk so the last thing that I want to do is is think about our paper in the context of the of the literature on green the green parallax and this is important because we we have different or opposite really predictions for the effects of of policy risk on emissions and that our paper is finding that because of these investment responses policy risk reduces emissions today first the green parallax is arguing that policy risk increases emissions today and the papers are kind of beginning from different starting points and that the green parallax is is starting from this world where we're thinking about fossil fuels as being in finite supply and then the effect of of policy risk on extraction incentives for that finite resource is critical to this argument that policy risk is going to increase emissions today and so what I want to do is first just walk through the intuition from the the green parallax literature and then show how that first compares to our model and then how we can incorporate those green parallax mechanisms into our model to run kind of a horse race between these two opposite predictions so the backbone of the green parallax literature is a hoteling type model of of fossil fuel supply and so what I've plotted here is this D is just showing you demand for fossil fuel MEC is the marginal extraction cost which I've set to be to be constant and now the price of fossil fuel is is above that that marginal extraction cost because of the of the of the scarcity rights that you would get in a hoteling model then the picture on the right thank you picture on the right is the is the same graph but move forward one period in time and then you have the standard hoteling type prediction that the price of fossil fuel is rising at the rate of interest so now let's think about the effects of policy risk in this in this setting so policy risk was going to reduce expected demand for fossil fuels tomorrow so that's shifting that demand curve to the left which in turn then reduces the expected profitability of extracting that finite resource tomorrow now fossil fuel extractors are solving this dynamic optimization problem so if it's less profitable to extract tomorrow then they're going to choose to pull extraction forward and extract more today so we see fossil fuel extraction and hence emissions increasing today and the price falling and so this is this green the sense of the green paradox in which we're getting this result that possibility of future climate policy is actually going to to increase emissions now what is our paper say our paper says well if we think about the possibility of climate policy in the future that's actually going to reduce demand for fossil fuel today because it changes the composition of capital shifting towards the cleaner capital and it reduces the total level of the capital stock so we're arguing that we get this leftward shift in demand and all else equal that left forward shift in demand is going to reduce fossil fuel fossil fuel consumption today instead of increasing that consumption and so now comes down to this question of well which effect is bigger this is supply side response where fossil fuel extractors choose to to move extraction to today and extract more or this demand side response where the kind of changes in investment patterns and by that we demand less fossil fuel today because of the risk of policy in the future so to think about this or put this in our model what we're going to focus on is the effects of of policy risk on fossil fuel prices when we're not in this world where when we have this is hoteling type model of fossil fuel extraction so this graph on the right is showing our model as we've set it up and in our model we don't have the hoteling like dimensions and so the price of fossil fuel is always going to equal the marginal extraction cost whether we're in that steady state with risk or in the steady state without risk now we have to bring in the green paradox now the price of fossil fuels higher than that marginal extraction cost because of the scarcity rents and importantly it's going to differ between the steady state with risk and the steady state without risk so the red here has shown you the price of fossil fuel and the steady state without risk and we're arguing that that's going to be larger than in the steady state risk with risk for two reasons first is we move from the steady state without risk to the steady state with risk we see demand for fossil fuel fall that's our investment channel that decrease in demand is going to push down this price the other thing that's happening as we can as we move from the steady state with risk to steady state without risk is that our fossil fuel extractors are choosing to extract more today because it might be less profitable in the future that increase in supply is also pushing down this price and so to bring in the green paradox we're going to allow for the price of fossil fuel and the steady state with risk to be lower than the price of fossil fuel and the steady state without risk now the question becomes well how much should these prices differ and for that we'll draw on the empirical literature on the on the green paradox so there are two really nice papers on this the first one by Derek Ramoy is looking at how changes in the probability that congress passed the waxman marquis bill which would have established a cap and trade system for the us or not changes inside the level of congressional support for for that for the waxman marquis bill affects coal prices and then the second paper which is by Kyle Meng is taking that one step further and saying changes how do changes in the in the effect on changes in congressional support for the bill affect the probability that the bill is passed and so combining these these two we get this result that we can get how do changes in the probability of a cap and trade system for the us affect affect all prices and so drawing on their evidence we find that reducing the probability of which are moving the probability say from from zero which is what it would be in our steady state without policy risk to point one which is our steady state with policy risk would apply that the fossil fuel prices is two and a half percent lower in that steady state with policy risk and so that's what we'll do so we just bring in um from our same experiment as we did before but this time to see does policy risk reduce emissions for you to allow for the fossil fuel price in that steady state with policy risk to be two and a half percentage points lower than in the steady state without and when we do that we find that climate policy risk is still reducing emissions but by less than it was in our in our baseline case and this is you know what you would expect because the screen paradoxes is sort of pushing in the other direction so all this is to say that even we account for for these green paradox mechanisms we still see that this demand side response through investment dominates and climate policy risk reduces gives you a small reduction in emissions so just to conclude what we've done here is show that climate policy risk gives you a small reduction in emissions today affects how we think about the cost of an action and when we think about this in the context of green paradox we kind of get the opposite results now over time it's possible that pressure for for climate policy could increase in the future perhaps is um you know climate change becomes more salient and if this is the case we show in the paper that as as the effects of climate policy grow we see um larger consequences from from both effects of climate policy risk also also increase and becomes more important that's it thank you thank you very much our discussant is Lena Bonneva from the ECB Lena before with yours hello can you hear me yes we can hear you okay let me uh share my slides we can see your slides there's some strange oh the strange boxes disappear okay let's just make them full screen okay and perfect we can see them you have 10 minutes let me just redo the full screen because I think it cut off um some part okay here we go first of all let me thank the organizers for inviting me to this very interesting conference I very much enjoyed reading Steffi's paper and to those of you who have not done so yet I very much recommend you to read it it's really a pleasure to read in my view there are three key contributions of the paper so the authors develop a novel dynamic general equilibrium model and use that model in order to quantify the impact of climate policy risk on emissions welfare and investment a key ingredient of the model is firm's belief about the possibility of a future carbon tax and the authors use a novel data set on internal carbon fees in order to calibrate these parameters in the model and finally the model is then used to uncover the transmission mechanisms at work through which climate policy risk affects the macroeconomy here are the key findings first if you're a policy maker it might be enough to actually make firms believe that you're going to implement a carbon tax in the future that's going to be enough to reduce emissions so there is no green paradox and so you don't have to go all the way to actually implement the carbon tax if you are just after reducing emissions and that's because climate policy risk reduces the expected return of brown relative to green capital and also lowers the expected marginal product of capital so these are the level and composition effects Stephie explained very well however your policy maker it turns out that reducing emissions by making firms believe you're going to implement a carbon tax in the future it's not a good idea because the welfare costs of doing so are very large compared to actually implementing the carbon tax and that's because in the model with risk the capital stock is inefficiently low in addition to risk averse households as i already mentioned it's a great paper and i very much enjoyed reading it however i have three comments the first one is on how Stephie and her co-authors calibrate the probability of a future climate policy in the model which is really key to obtaining the results and because the paper is already in a very advanced stage so it's very clearly written it has lots of robustness tests my other two comments are offering some ideas on how Stephie and her co-authors could proceed on their research agenda yes and these are essentially if climate policy risk matters empirically and also one point about central banks and if they should respond to climate policy risk now coming to my first comment this probability of a future climate policy is really key key to the model it's the key novelty the authors develop and add on to an otherwise pretty standard dynamic channel like libya model and in order to discipline this parameter the authors use a novel data set on in internal carbon fix however i thought the sandal they use is pretty small so it's just six firms and also pretty unrepresentative uh the the carbon fee used in the model is ten us dollar and i thought that seemed rather low in comparison to some most recent data iphone so here's a table i took from a recent report by the carbon closure project and they put the median price price of an internal carbon fee on 18 us dollar which is almost twice the number used in Stephie's model and also that table reveals there is a huge dispersion of internal carbon fees among firms with the maximum price of over 500 us dollar so keeping that in mind i was wondering if you could essentially use this most recent data to recalibrate the probability of a future climate policy and ideally base your analysis on a larger number large assembly of fair firms given how important the parameter is in your model and now the table also revealed this huge dispersion in the internal carbon fees so given that i was wondering if you could essentially report ranges for your key results based on for example in the quartile range of the observed carbon fees and finally when reading the paper it was not entirely clear to me if the carbon fee data is covering score one two or three emissions so that would be useful to clarify and ideally reporting results for all different scopes and again data on that is available in the report cited on the previous slide now coming to my second point so i think Stephie did a great job in convincing us that climate policy risk does matter in her model but i was wondering if climate policy risk also matters empirically you know there's a huge literature on investigating the macro effects of different types of uncertainty using empirical methods starting from the seminal paper of bloom but lots of recent papers including some cited by Stephie in her paper and Dresden-Tasch and i was wondering if you could essentially cooperate your model-based results with some empirical estimates where you essentially assess the effect of climate policy risk on the macro economy on capital on investment on emissions using methods like a structure vector to questions or a local projections now clearly measuring climate policy risk in such an empirical approach is a very difficult task but i do think this challenge can be overcome and i do offer some ideas so for example you could use a textual approach by searching news papers for keywords like climate policy risk future carbon tax and the like and in addition you could also use survey data for example there are polls on on climate policy carried out by the carbon tax center and you could use this version in a survey respondents view on future climate policy to essentially gorge climate policy risk empirically or you could take a narrative approach essentially narratively identify events associated with expected unexpected changes in climate policy risk and ideally normalizing those with movements in Korean enterprises and finally some countries and regions are already running different carbon pricing schemes and you could also use observable activity in in these carbon prices over the short or long run and of course there are also lots of different approaches including approaches followed by some of the papers cited in instead this presentation now coming to my final comment i thought with the ECB co-hosting this conference i should say something about monetary policy as well clearly stressing here that the views here in this presentation are my own personal views and should not be taken to represent the ECB or the euro system first of all so here my comment is it it would be very interesting to essentially assess in a model-based framework how central banks should respond to climate policy risk i think this is a highly relevant question also given the ongoing ECB strategy of view and it can be answered in a model fairly closely related to to the world you you did present and for example one approach to tackle this question would be to add some decantion features to your model and ideally also model climate policy risk as a stochastic shock as done by previous work alternatively you could also set up a model for optimal policy where you assume that there's imperfect knowledge about climate policy shocks so essentially in a model firms would pay own limited attention to what politicians say the path you know regardless which approach is taking it would be interesting to study optimal monetary policy in response to to climate policy shocks both assuming the policymaker acts with discretion or commitment and obviously from a practitioner's perspective it would be very interesting to see how close optimal policy is to a policymaker following a similar tailor rule for example now with that let me conclude i think it's a really great paper addressing a highly relevant policy question i do think there is some scope to improve the probability to prove how staffing causes calibrated probability of future climate policy given that this parameter is really key to the results and i'm very much looking forward to the next paper on this research agenda and i did offer two ideas which the authors might want to consider in the future thank you very much and i'm looking forward to defy's response thank you very much Lena we also have one question from Eric Puik what if the global energy demand grows faster than green energy supply how would that impact your results so Steffi you have one minute to react to Lena's comments and to reply to Eric please choose what you would like to do you can always use the chat to reply to the question i think what i want to do is just say a lot of thank yous to Lena and those are those are just fantastic ideas and so we to ask you for your slides but those are definitely things that we should look at and i'm really excited that there's this new new data on on internal fees that maybe we can we can use and i think using a range is a fantastic idea so i i really appreciate those thoughts and you know how how how you know central banks should should start thinking about responding to this policy risk is is also a good a fascinating question and you know what sense is it different from from other types of risk and what sense is it similar and so those are those are also really great great suggestions so thank you for for reading it so carefully and and coming up with ideas on the question on what if and i forgot which was bigger green i'm going to answer the question in the chat when i can read it again but um but thank you all very much and thanks thanks for coming