 Hi. Thank you, everyone. I think we can reconvene and move into the report back. We're just, we're just going to ask each of our breakout groups to, to for one person to kind of give a brief three minutes summary of the key takeaways from your breakout group so we can take down these slides actually since we're not going to be sharing slides for this one. I might first ask the, we can start the virtual breakouts and we can do the first room the climate macro modeling group I think that was Bob sorry, was that your group. Oh, Adele I'm sorry Adele. Sure yeah. And thanks to all my virtual participants to talk about climate and macroeconomic modeling. We didn't necessarily follow the full script so apologies if we deviate from the prescribed outline, but a few, a few things in, in not any real particular order. Some of the questions that came up for our group participants related to macroeconomic forecasts and projections is that for those that use econometric models for their macroeconomic projections. How should they incorporate climate related factors, and this might be in the context of other countries or applications. And what would be the appropriate methods to do that. And what do you do if you don't necessarily have a long time series for the relevant data to do those more standard econometric projections. So those, those question arose. Another question was, what could be the roller example of model inner comparison and improvement projects. So there are a number of these groups. One of the early ones was the Stanford Energy Modeling Forum that convenes modelers to look at stylized scenarios of policy and technological futures that help develop these models and identify strengths and weaknesses and and fruitful solutions for growth and development of those models. There have arisen similar inner comparison projects in climate science, and in agricultural models. And these, these meetings of modeling teams I think can be really fruitful. I think there's a question there for the macro modelers. And just to give an example, we talked about agricultural models because that was one of the areas of expertise of our participants and looking at, you know, how you model crop productivity in in a context of climate, and how you convert that into economic productivity, and how and whether that gets translated into total factor productivity in in your macro economic projections. And then how do you deal with the, the wide variations in crop production and what in figure out what drives those also in the context of climate variation. Adding a whole, you know, whole new levers of variation and trying to figure out how how to insert all that one one kind of proposed assertion that came up in our group and I'm going to just try this out. It's not necessarily a consensus view, but here it goes. It's the perception that the macro economic projections the GDP forecasts in particular that we were talking about yesterday on the factors that go into determining output, very importantly, spatially and sectorally within the United States. And this is going to be especially true when those factors are driven by climate related considerations, and whether it's climate damages or, or transition factors. So all those things that go into those, those inputs to macro economics whether it's labor supply, or productivity or capital, there are complicated things behind there. And accordingly, I guess the question we're asking is whether these traditional person monies macro models are really going to be well suited to a low carbon and climate disrupted future. And, and whether or not it makes sense to try to, you know, retrofit those with climate factors, or whether it makes sense to consider new classes of models for macro economic projections. And I'm not enough of a macro economist to know just how inflammatory that possible suggestion is, but I, I'm just the reporter here so and then in the category of, of limits and opportunities. Again, we talked about agriculture, and in one example in ag models is just like, these are really important for, for farm credit oversight, regulators and, and project projecting productivity and an important sector, but when you start adding climate in there, everything gets even more crazy complicated. And if you just take livestock. So, you might have a relationship of heat and and livestock production but with climate. It's not just heat, it's generally it's heat in the nighttime, and how often, and how, and how many nights sequence, you've got hot nights. So just like adding these new factors like how do you even do that and how do you get the right data and so on. And then one final thought is just looking at the example of the creation of shared scenarios. So the first convene and create like these representative concentration pathways, and the shared socio economic pathways that the IPCC is developed. And these illustrate narratives of future outcomes. Perhaps there's a question of whether stylized macroeconomic now narratives and scenarios might be useful. I realize this is kind of at odds with the idea of forecasting, and they are really different exercises. Perhaps there, there, I guess the question arises, is there any model in this convening of macro modelers to talk about illustrative narratives that might be incorporated in these projections. So I'll stop there. Thank you. Thank you, Del. Can I now ask the whoever was the self appointed rapid tour from the in person breakout on the same topic that climate macro modeling to to give your summary brief summary. It is I were okay. I'll wait for the camera to find me. Yes, there I am. Okay, I'm going to be a bit more brief. So, let's see what did we talk about one thing we talked about is that some, you know, an existing in the administrations macro modeling framework supply side factors matter a lot. The economy in general and so migration is very important, and it's not explicitly represented any of these climate macro modeling connections yet. And probably probably won't be for political reasons but that's something valuable to think about. We had a long discussion about what type of climate data is most useful for kind of macro economic modeling. And in particular we discussed how correlated events and extremes or estimates of the duration of exposure is very valuable. Those are like moments of the distribution of weather exposure that are really challenging to model. But those are probably moments of the distribution of other exposure that have the most influence on the economy. So, to the extent we can improve our modeling of those moments we'd improve our modeling of the economy. And we, we talked a little bit about some some other things that aren't included in macro models right now like effects on human capital accumulation, which would arguably be very important for economic growth. There's, there's very little research on that. The research that exists is, is like pretty good in my precipitation is. That was, that was a very, that was a very Noah moment. Is good like in my evaluation but yeah I feel like we just need more need more research on human capital effects of climate that's probably very important. And then we had talked about how distributional consequences of the energy transition and of physical climate damages are really interesting the policymakers both on like on a spatial and a sectoral basis. It doesn't feel like existing macro models. I might just hazard making a similar suggestion that we just throw throw them out but you know I don't think that's, I don't think that's going to happen so maybe it doesn't make sense to have one model, maybe instead we should think about having multiple models one kind of an administration model for the net effects forecasts, and then one that can get more at sectoral or spatial details that could be housed some more else it would it would be, because it's talking about distributions it probably has more political weight and so it might make sense to house it, you know somewhere else in the government. That's just me. That's my personal glass on that. Those are the main things we talked about. Neil and Rita, Peter, if there's any if there's anything I missed, feel free to to mention that. Great, thank you. And then moving back to the virtual breakouts for this breakout to on modeling economic damages, or I think Bob is the repertoire there. So I think you're going to start to see some commonalities with the previous two groups. In terms of key questions. We had a bunch of questions that I think can broadly be put under the category of cascading risk. If we have, you know, does the idea that aggregate national damages in the US are small aligned with the fact that we have regional damages that can be quite substantial to regional damages in fact interact with under on another so they add up in a super linear way. Our compound extremes. are economically significant at a national scale. Our physical temping points potentially economically significant so if you, for example, have brought shift in the North Atlantic Sempolar gyre, which would have concert climatic and sea level consequences for Europe, North America. Africa with that. Would that be economically significant. But that the distributional lens was really an important lens to bring to all of this much and maybe it's much more than the aggregate perspective and therefore you need models capable of addressing distributional questions, and yet has both of the previous approaches already noted current macro models often lack that needed regional resolution. And also both macro models and climate damage models often lacked both adequate sexual resolution and treatment of intersectoral interaction which are another pathways cascades might happen. So we also talked about the importance of uncertainty characterization. And again we mentioned the need to think about higher order moments not just mean invariance and the need to think about ambiguity and misspecification as well as quantifiable uncertainty, and also to think about how uncertainty is communicated and not just how it is characterized and technical documents. And finally, I think the main new thing we have to add to the discussion so far as we talked a little bit about computing opportunities. So opportunities from open science through sort of modular open systems like Freddie that can bring in modules, other from a variety of sources. Cloud computing opportunities bring together large climate and socioeconomic data sets to make it easier to do new climate econometric studies. And potentially using of AI to help with sort of scraping and cleaning new socioeconomic data sets to look at. I think that's super thanks. And then the in person self appointed repertoire for the second group on modeling damages. Hello. Yes, my name is Daniel from SF. So our group had a very rich discussion on this topic. And there, I'm going to report on just a few key takeaways I think. But we recognize that a lot of models that we currently use don't have important feedbacks. So that's really, it's difficult to capture the changes in human behaviors in response to either the single disaster events or to long term climate change. Also it doesn't capture the reaction from the different sectors to the impact of the climate change. So as a result is very difficult to derive the actionable policies or, or programs based on those very aggregated models so one possible solution or opportunity is to to develop more flexible models that can really link to each other's or perhaps some new causes of models that are better suited to answer some of the questions that we're interested in. I think we have mentioned that there's some new research on embedded models or agent based models that definitely has a potential to to play a role in this space. The other takeaway is that when you look at the, especially the impact from the extreme events like hurricanes or flooding or wildfires. And we just don't have enough historical record that we can use to train a lot of empirical models. So projecting their frequency and and intensity into the future. We feel that there's additional research required to look at the impact of climate change. Those events end up happening at the same time concurring concurrence or cascading facts is one disaster lead to to another and end up creating a much bigger effect on the US economy and abroad. And also, there's also some interest in look at how the those extreme events become more clustered either in the temple or spatial dimension so they happen to, you know impact on certain part of the United States or the space between those events, you know we have seen the that got hit by hurricanes back and back and you know that is definitely something creating additional challenges to those committee while they were recovering from the previous one. The last thing I want to mention is that we also recognize there's already a lot of models out there. In addition to what we discussed in the past two days, but they are very specific to either individual sectors like energy or agriculture or, or food systems so perhaps there's opportunity to, you know, bring those kind of model models together to compare and see what we can learn from each other and also related to that we also already have some models really have a more granular spatially so there's models for regional economy or even for certain watershed so but within that region there's a lot of different industries or different features being a model so so we're hoping that we can really find some kind of synergy between those different models in order to move the conversation forward. Thank you. Thank you. So for our virtual rooms we combined breakouts break rooms three and four so Emmy will will be the repertoire for both of those. Okay, our discussion was a fairly small group and it was very interesting discussion but it's sort of straight away a little bit from the main topics that we've been talking about here I thought it was very interesting, but different perspective on some of these issues. So the conversation really focused on the role of community in some of the questions that we're discussing I mean we're discussing them in a very quantitative way. Thinking about aggregated statistics either the national level or at the regional level but you know of course these are real people who are involved. And you know real people who are who are proud of communities and those communities are being affected by climate change and by the energy transition so that was sort of one focal point. There was an emphasis on the connection that people have to places and to the possibility of solutions to climate change and the energy transition that help support communities. So in addition on the idea that some of these solutions should focus not just on disaster recovery but also on forms of proactive resilience building. And there was also an emphasis on on the idea that many of the issues we're discussing have have broader impacts beyond the economic impacts that that were focused on measuring. And then for the, the third in person group. What, what was that the energy transition group. So I am presenting. We had a great discussion I have to say I only learned at the very end that I am the one presenting. And I wasn't, I wasn't able to connect with everybody so hopefully I characterize folks contributions correctly. So let's start with a basic premise that under current law, what if we project that over the next decade we go from 60% of electric electricity generation from non renewables to 20%. So how do we incorporate that transition into economic projections, an important input, so maybe the basic way to do it is that an important input renewable electricity falls in price. And then becomes relatively cheaper to another input that is substitute non renewables. So we have basic models that the, you know, tell us what to do with that news. I think a big part of this is that the assumption is that the, that the decrease in prices big enough that investment increases to take advantage of that change in price so think of electric car plants. You want to think about just incorporating that price correctly and having everybody respond to that price. We have lots of off model ways of thinking about how to how to project productivity growth. So maybe thinking about the effect of total factor productivity just in the energy sector, but aggregating up from the renewable and non renewable. And how that then affects overall productivity. And a couple of other ways that you want to think about the distribution, you know, both to think about the distributional effects, but then outside the model but then also think about how to get them correctly into the model in aggregate, I should say, you were, you were invoked many times in our conversation of what would, what would we, what would we suggest to Steve. So thinking about the labor share. We have lots of experience as modelers thinking about the labor share in different sectors thinking about the labor share in the manufacturing sector and the service sector different parts of the manufacturing sector and how those have changed over time and then what we think about then the labor share going forward so you might similarly do that thinking about the labor share in the energy sector, and in other sectors and how labor reallocation will matter for the labor share, and then making sure you have all of the effects of trade in terms of trade to help to help incorporate it into the model. There are. So that was one big way that we thought about how to incorporate these changes into an aggregate model. So I think it's wonderful about the distributional effects along the way. There are past experiences that can help us to calibrate, you know, or at least how to inform how to model this. So, we talked about, like, what you might think of as the China shocked, which I would characterize as a reduction in production of imported final goods and how that affected manufacturing in the US, but then sitting here typing that up. I thought maybe it was, it's better to think about also because of trade, a reduction in cost of inputs and there I had in mind like semiconductor chips. So maybe think about how the reduction in the cost of semiconductor chips. And you know you got to do some quality adjustment there because it's not just that. I mean, chips today are not the same as chips from 20 years ago, how that created just a massive change in the way we produce things in the United States. So you can look at these big past transitions and decide whether or not. There was a problem, just looking at aggregates or whether whether or not you had to get under the hood. Okay, then I'm going to come to the risk word in the part of transition risks. And the way I thought about this and people sort of seem to buy is that there are both intended and unintended consequences of this transition. So, unintended consequence would be like, we fully intend through this transition for fossil fossil fuel assets to become less valuable. Like that's just an intended consequence. And unintended consequence what we don't intend but will happen, unless policy mitigates it or undoes it. I mean, actually the one I'm about to explain is just an unintended consequence that would happen is that workers in the fossil fuel industry would face a need to switch industries. So you need to think about what the intended consequences are what the unintended consequences are and then how policy wants to lean into those mitigate those prevent those. Somewhat similarly expected and unexpected consequences. So unexpected consequence would be we fully expect that we're going to need a significant amount of construction and non renewable infrastructure. And that will create land use issues like they're all sorts of things that we fully expect are going to be consequences of this transition that we need to think about in the models. And they're all and we also fully expect, I think that this transition is going to change geopolitical power dynamics. But then they're unexpected consequences. And just because things may go wrong, there may be shortages, there may be price spikes, you know here I was thinking about like intraday price spikes in particular, like, it turns out not everybody wants to use, you know, energy when this ends up. And of course they're unknown unknowns. And so you want to incorporate this is the true idea of transition risks you want to be thoughtful of what these risks are. And then beyond all of this I feel like everything that I just talked about started with current law. And if you're not, if you're not, you know, the administration trying to create its forecast you want to think about policy risks. And that policy risk can also add to the transition both. I mean, I, when I was saying that I meant domestic but then it occurred to me that we as well think about global policy risk as well. Thank you, Wendy. And then we have one more group. This last one, the public and financial sector risk and response who I think is rapid, rapid toward by Eric, and if I can ask you to try to be brief so we can get to our last panel. I'm going to be super brief, especially since the previous group encroached on our space. Yeah, policy risk is very important. Okay, there we go. And one thing I just want to start with was something that came up at the end, which is that at least in our sample of four researchers we couldn't identify scholarly work that focuses on how uncertainty about future climate provides fiscal risk at federal agency and state level, there's literature on fiscal risk. There's literature on climate, but there isn't a literature on that ties those two together with that we are aware of, and a couple of folks in the group had been looking rather intently. So this is, this is something to pay attention to. And as a result of that so this is this is just the kind of a large area. Moreover, there's the issue of transmission of risk from one level of government to the other. And that raise came back to a couple of our key questions who is paying now, since these are perhaps, you know, uncosted risks and who should be. And a number of different topics, but I'm just going to jump to the risk between of the strong divergence between audiences and actors. So this is, this is a part of the policy public risk is that communicating these is vulnerable to very different reactions to what we're communicating and that is in particularly in particular true as Rachel Cletus emphasized earlier on talking about risk. There's a risk and talking about risk because people don't understand it. And one other that I'm going to mention is that there is a risk that modest GDP deltas translates and this is what we've been talking about the whole time translates into serious local and regional impacts. So this is just for interpreting models. We have a lot more though my gosh it's just beautiful there's lots of stuff. Thank you. All right thank you everyone I will now turn it over to Bob and the, the we have three panelists for this final discussion to just kind of have an open discussion trying to trying to synthesize some of the discussions that came up in the past two days.