Yeah, let me… let me start. I want to thank everybody for joining us today. I want to have a couple of housekeeping items. So first of all, everybody's microphone will be muted.

We want you to put questions into the chat as we're going along. If something comes up that.

We want to interrupt Dr. Lemoyne is willing to interrupt and take the question. But putting it in the chat, I think will make it maybe a little bit smoother.

Um, we are recording the presentation and. We'll get it linked onto the Tucson CCL website.

Yep, that's right, we will.

After the presentation. And let me introduce let me introduce two people today. The first person is our guest, Derek Lemoine, Dr. Derek Lemoine received his PhD in energy and resources from the University of California at Berkeley.

Prior to joining the University of Arizona Eller College of Management in 2011.

In addition to his teaching at Eller, he's a research associate for the National Bureau of Economic Research and Associate Fellow for the Center for Economic Policy Research in the Climate Change Research and Policy Network.

Dr. Lemoyne recently published. Research on the economic impact of climate change that was highlighted in an article in the Arizona Daily Star earlier this year, and he's graciously agreed to talk to us about the economics of climate change.

The second person that I want to introduce is Dr. Michael.

Jones, who will be our moderator. Michael is a CCL member from long standing about 9 years. He's a retired academic economist, research specialty of international economics.

He belongs to Tucson CCL and the CCL Midcoast Main.

His emphasis has been on the carbon neutral tax, or carbon fee and dividend, and he was part of the economics policy network at Cco for a while.

And that action team was publishing papers to help people like us that aren't PhDs to understand the economics of climate change and carbon pricing. So Dr. Lemoine, it's all yours.

Great. Thank you so much, and thank you, Bill and Ed, for organizing this and for having me. And thank you, Michael, for taking the time to moderate and sort of guide this thing along. I'm going to… let me share my screen, because I don't think it's up right now.

before I forget, but I really do appreciate you all having me see. You should be able to see that now. Does that work?

Yep.

Okay, great. So again, thank you for having me. I'm going to cover. I was trying to decide what to do, and I figured that I would do a combination of overview versus like semi dives into things that might be of particular interest. So I'm going to focus on.

What we know about the relationship between climate and the economy, and then I'm going to move into what might be an interesting alternate way to regulate climate change that that might have some appeal, especially in light of our inability to figure out what even the right emission tax should be, even if we could enact it. But I'll kind of build up to that.

So I'll start with obvious things that we know. I'll start with a few less obvious things that we know, just at a high level, some things we know about climate and the economy.

I'll then talk a little bit about why I think we'll never really know what climate damages would be ahead of time, which is pretty salient for, like, picking what the emission tax should be. This builds on a review piece that I've been working on for the last couple of years.

Then I'll talk about damages and climate change to date, and then I'll be motivated by what I've said before, and that's the new paper that was mentioned in the intro. And then I'll talk about an idea for how to use markets to project damages, which is a working paper that I've had out for a couple of years now.

Okay, so obvious things that we know. So obvious things. So how does climate change matter? Oh, we have forecast of weather days, weeks, months, sometimes even a few years ahead of time, whether then affects the economy through all sorts of channels. I'll touch on a couple of them.

There's something important called adaptation that economists really focus on, and that's one of the big questions in climate change is how much adaptation will there be in the longer run in the short run?

So when we say adaptation, we think about actions that are taken as in particular as a response to the weather or anticipation of the weather in order to offset the effects of the weather or take advantage of weather. So there's two kinds we think about. One is ex ante adaptation, so that you know weather is coming ahead of time. So you do something ahead of time.

An example would be you might have a forecast of rain, so you change your schedule for the day, or you have a forecast of extreme heat, so you change your workout to being in the early morning. So that would be all ex-ante adaptation.

There's also ex post adaptation where you walk outside and it's raining, and so you go back in and you grab a rain jacket, or you walk outside and it's hot, and so, you know, you go back into the air conditioning. But these are just reactions to what the weather is, not based on information in advance of the weather happening.

You can illustrate that with air conditioning. So this is a paper with data from Mexico, and there are similar results elsewhere. So what we're looking at on the vertical axis is electricity consumption, and left to right, we're going from cold days to hot days, and this is all in Mexico, like I said. So you see basically no change in electricity consumption from cold days in Mexico.

And, like, again, they might not have… a lot of the heating might not be done through electricity, so they might be using heating, but it's just going through gas or something. Then on the far right, though, on hotter days, there's a lot more electricity use, and that's ex post adaptation. You're using your air conditioning more intensively on hotter days, all sequel.

Exanti adaptation would be how much air conditioning do we have installed in the country. And so this plot is looking at the colder municipalities in Mexico, and we're plotting how many of these households, what share of them have air conditioning by income as they go left from less income to right more income.

Where you see the comparison of exanti adaptations when you compare that to the… what the same plot would look like for the warmer parts of Mexico, the warmer municipalities, and there's just a lot more adoption. All the dots in the lower plot are above the dots in the upper plot. I mean, really, the effect of being in a warm area is way greater than the effect of being of higher income in a cold area.

This is ex-ante adaptation. People adopt air conditioning in order to deal with heat because they know they're likely to get a lot of heat where they live. You don't go out and buy it in response to one hot day. You buy it and install it because you expect hot days to occur with some frequency.

But adaptation also has cost. So this is a graph electricity consumption in the Us. Now against temperature, and you see the similar sort of effect. Hot days cause a lot more electricity consumption. Again, not as much of an effect on cold days, because we use other forms of energy in the colder parts of the country for heating by and large.

If you turn… if you do the same calculations I was showing before, or if you take the kind of data I was showing before from Mexico and project air conditioning use under different climate change scenarios, and look at how much more people would spend on electricity per year.

for… you'd see that from using electricity more intensely, this is the 733 number on the lower right, this is saying per year people would spend 2 or 33 million more dollars by the end of the century under a high warming trajectory just from using air conditioning more intensively.

But they would also install more air conditioning, and when you account for the fact that they would have a lot more air conditioning and use it more intensively, it looks like almost $4 billion a year of extra expenditure just to stay cooler in a hotter climate. And again, there's offsetting benefits from less cold. This is just the cost side of it.

Um, but, you know, it's an important aspect of adaptation that it does have cost. This isn't something you chose to do anyway for free. You're willing to spend a cost to deal with heat or to deal with to deal with climate change.

If you want to think about how climate matters, one experiment you might want to run is just compare hotter and colder locations, because basically we're making colder locations look like hotter locations, and hotter locations look like even hotter locations, as we have climate change. So maybe we can compare outcomes in Iowa and Arizona and get a sense of what climate change will do.

This was a method that was used with some frequency, maybe 30 years back, and it just doesn't really work. There's too many things that are different between hot and cold places, and you can't quote-unquote control for all of them. Like, you can't deal with all of them. There's always unobserved ways that places are different. So we now know with pretty good certainty you can't just compare hot and cold locations. There are a whole bunch of ways that they're different.

Arizona is not Iowa for a whole bunch of reasons that have nothing to do with just the temperature in each location.

We can learn about the importance of weather, and we can do that very readily and with not even terribly sophisticated techniques, within a given location on the same day each year, let's say, sometimes that day happens to be hot. Some days it happens to be cooler, some days it happens to be rainier, some days it happens to be not.

That's as good as a random experiment as we're going to find. I mean, nature's throwing dice around the climate to get the weather every day, and you can then see what the weather does for all sorts of outcomes.

This is easy for economists to do, because we have great data on weather, and we can link it to all kinds of outcomes and see if weather matters. And the upshot is weather almost always matters. Whether the outcome is any sort of health effect, or many types of health effects, whether it's mortality, whether it's income, whether it's crime.

whether it's migration for all sorts of things, ultimately, some form of weather matters, whether it's heat or cold or both, some form of weather is going to end up mattering. So we now know that we're very exposed to weather through all sorts of different channels, and we've established that.

But a lot of what I'm going to talk about is why that's not enough, and why we need to know a little bit more.

Okay, so those are obvious things that we know. So hopefully that's not too boring. Now we'll talk about less obvious things that we know, and then why we really… I don't think we can really know what we want to know.

So less obvious things we know. So we, I guess that we're pretty good at getting responses to weather. We have at this point a pretty good handle of of what an extra hot day in a year is going to do along a whole bunch of margins. The problem is that I would argue that tells us almost nothing about the effect of climate change.

A single hot day in a given location is just not the same experiment as changing the climate. It's just very different. Changing the climate changes the weather consistently, year after year, and you also see it coming in advance for years ahead. I mean, we have climate science. We can project the climate way out.

It was weather is just a random hot day. You don't do the same things. So one example, capital stocks adjust only slowly. So this means you might do more response to climate than weather. So imagine you're a farmer getting a single hot day that might be hurting your crops.

You don't install irrigation because of that. You install irrigation maybe incrementally over time or in response to knowing that this hot period is going to recur. You adopt air conditioning, not because just today is hot, but because you expect the heat to recur, to come again. Capital stocks take time to adjust.

So, something, whether that looks very costly today might not be as costly with climate change once our capital stocks have adapted to that climate change.

On the other hand, resources get depleted over time. So imagine you're a farmer, and the way you deal with hot weather, as many farmers do, is to water your crops more intensively. Imagine, as many farmers do, that you get your water from groundwater. Well, you can do that for a single hot day, or a hot week, or hot month, or even hot year.

But if every year is now going to be like that, eventually you're pumping costs are going to go way up, and you might even just flat run out of water. So there's a limit to the degree to which you can do that. And so it's a horse race. Like, which is more important? Is it the fact that you're going to install the right kinds of capital over time?

Or is it that the resources that you're using to adapt are going to get depleted over time? Even think about adjusting your schedule around extreme weather. You can do that for a day or two, but it's hard to do that day after day and maintain that adjustment. At some point, you need to go out and run the errand you're gonna run, or exercise, or do whatever it was you were gonna do.

Another difference between climate and weather is that we have knowledge of climate much further in events.

The best weather forecasts are really 10 to 14 days out. We have some idea of the weather at a coarse scale a few months out, but climate, I mean, it's really the average of weather is one way to think about it, or the statistics of weather, and we have a sense of climate, you know, decades out. Climate change is not a dramatic surprise at this point. We've been studying it for.

Depending on how you want to count in 100 years. So we know a good bit about climate from longer away. So we can make pretty big investments that we're not going to make for just two weeks from now is going to be hot. So building seawalls is a big investment. You don't just make randomly, and you could come up with countless examples like that.

So there's just these big long run investments that are not going to be if you're looking at data about how does weather matter, it's just not going to be there in the data. Our ability to make these investments.

And the last, what I think is a big one that was not fully appreciated maybe until more recently. Climate affects everywhere at once, and I'll come back to this a bit later in the talk. It'll be critical. Climate isn't just changing your weather. It's changing the weather everywhere around the globe. And so I again, like, I'll hammer this to some length, but.

If climate affects only you, the prices of everything you buy aren't really affected. I mean, so what? Where you live, it's a little bit hotter. That doesn't affect the broader economy, one little location. But climate change is the weather everywhere, all at one time. It's within the whole economy adjusts, and the prices of things adjust, and demand and supply adjust.

And those channels are much more important for climate than they are for weather. They're almost absent for a local weather event.

Okay, the next thing that we kind of know is that adaptation is actually really important. I mentioned earlier that it exists, but it's actually really important. So here's an example from a working paper of mine. So we're looking at the value of weather forecast for avoiding mortality from from extreme temperatures.

So what we're looking at is on the left is a cold day, a day that happens to be cold. In the middle is a day that's pretty temperate. This is all in Celsius, and on the right is a day that's pretty hot. And then the blue parts of each panel are a forecast days that had a forecast that was too cold.

And the red parts are days that had a forecast that was too hot. And vertically, we're looking at how does mortality in these days change? And so we're interested in whether if you have a forecast that was bad, is mortality any higher than if the forecast was perfect, which is a zero. There was no forecast error.

So if you look at the middle plot, there's really no difference. It doesn't matter whether forecasts are good or bad, the weather is pretty mild. There's not that much mortality anyway, and it just… forecast accuracy is not that important.

If you look at the cold days, the red side goes up. And what that's saying is if you have a forecast that's too hot, so too mild, you under forecast the cold, then mortality is higher than if you didn't. And the only reason a forecast could matter is if people are using it, because the forecast doesn't kill you. It's the weather outside that kills you.

So, people must be acting on the forecast and not doing as much as something, or doing too much of something else that then left them more exposed to the cold. If you look at a hot day on the far right side, similar story. You forecast it to be too hot. It doesn't really matter for mortality. Mortality is about the same as what it would have been otherwise.

If you, but if your forecast undershot the extreme heat, then again, like your mortality is a lot higher. And we've seen this in study after study now by the same team has been… we have an NSF grant to do this value of weather forecasting stuff.

And application after application and forecasts are really valuable in extreme heat. People seem to take very strongly on forecast, which also leaves them very exposed to forecast errors because they're really relying on forecasting extreme heat. The good news is forecasts are actually pretty good in heat and much better in heat than in cold.

But the bad news is, when there are errors, they're actually quite costly.

Mortality is actually more sensitive to cold. Cold is much more deadly than heat. It doesn't kill on the same day, it kills over subsequent weeks and months, mainly because people catch diseases, whereas heat tends to kill on the same day. A lot of the deaths from heat are deaths that kind of would have happened. You could see in the data, like, just statistically, that most of them would have happened in the next couple weeks anyway.

He's just not as lethal, then heat would be a lot less lethal if we had even better forecast of it, because people do seem to use the forecast. And that's all adaptation. That's people taking actions in advance of the heat based on their knowledge that the heat might be coming.

Okay, so those are some things that we kind of know. But why I don't think we'll ever really know climate damages. And I've already kind of hinted at that. But this is a picture of all the things you would need to know, and ignore the light gray, because those are just parts that really aren't about the damages. So at the top, and this is from the review paper we've been working on. So at the top you have climate change. Climate change is going to change present weather, so.

focus on the arrow to the left. I don't know if you can see the cursor on the screen.

But so climate change is going to change present weather, and present weather is going to directly affect output. So climate change affects the weather today. The weather today is different than it would have been without climate change, and that it has an effect on economic output and all sorts of stuff we care about.

Um, that's what's called direct damages. And we have a pretty good handle on that. That's pretty easy to get with data. We've got a lot of experience now getting at that with data, and we've got a pretty good handle on that.

The problem is that that's a pretty small part of the story. There's a lot of other arrows on here, and we're way less good at getting at that. So climate change makes today's weather different. That means that we take different actions today. That's the present transient action. Like, for instance, our labor choices. That's called adaptation, and that's going to affect output as well. We have some handle on that, not as good as the direct damages. We have some handle on that.

The big problem is climate change changes a lot of other kinds of weather. So start in the far right, climate change changes the weather in all past years as well. Like, imagine we're sitting in 2050, meaning climate change has changed the weather every day of everybody's lifetime at that point in 2050.

So if climate change is persistent, it's not just one day or one year. If weather every day, day after day for years is different, that's going to affect the pattern of investment all throughout the economy. People are going to act on that. There's going to be different air conditioning, irrigation, highways, rail networks. I mean, everything.

And then the way that you build a train tracks depends on the temperature and the location. And then that that is prior adaptation, things that people did in the past is going to affect output today.

Climate also affects future weather. Again, like, we know… we can understand climate change at a pretty great level. I mean, climate science is not the main barrier at this point in time. I'd say the economics is a much greater barrier. We know less on that side.

So we anticipate climate change. We know it's coming. We know it's coming on roughly on what timescales as well. And so we can act on that information. And so we can make investments today, and we can make different choices today. And those are all forms of adaptation.

And as problematically, climate change is widespread. It's not just local. It's affecting weather everywhere else around the planet, both the present, future and past weather everywhere else around the planet. That's affecting the whole economy of the planet. It's affecting prices of everything. People react to prices, and again, that's a form of adaptation.

So we have all these links of future and past weather and weather everywhere else, which, as economists, we recognize are important, and we're starting to investigate, but we don't have a great handle on. And honestly, they're going to be way more important, I think, than just the present weather link that we've been studying.

Some of them may make climate change less of a problem. A lot of them might make it more of a problem. We don't know the total answer yet, but I think they're likely to be the bigger part of the story.

And the ultimate problem, and this is what you formulate in the review, is that there's something called a trilemma. So we're familiar with dilemmas. So can I have a vanilla or a chocolate cake? I gotta pick one or the other. A trilemma. There's 3 options, and I want all 3 of them, but I can only have 2 of them. So there's 3 things that we want when we're trying to estimate the effects of climate change, when we're trying to know what will climate change do to the economy.

We want to make sure that we're not making, like, particular assumptions that everything just hinges on these, like, really fragile assumptions. We want robustness to all these economic assumptions we might have to make. Okay, so we don't want to, like, make any… we don't want our conclusions to be too fragile.

We want to make sure that whatever we calculate is actually the effect we care about, which is widespread climate change, persistent climate change, anticipated climate change. It's not just the effect of today being hot and then whatever happens tomorrow happens tomorrow. It's really… it's got to be climate change and not just weather.

And it's also got to be what we call well identified. So we want to make sure we got a causal relationship and not just a correlation that, you know, hot locations do this and cold locations do that, but maybe it's not climate, maybe it's something else. So we want to make sure it's really well identified. And we just can't have all 3 of those things.

So, for instance, like if you want to get a causal effect, great. Some days are hot, some days are cold. That's random. That's a causal effect of weather. And that's going to be robust to any sort of assumption. It's pretty easy to do those stats.

But however, it's not climate change. You've gotten one in three really well. You've totally missed out on two. And every other method we've come up with is pushing on one margin or the other, but it can't hit all three of them.

I'm not gonna bore you by walking through this in detail, but this is a plot that we made of different methods that people have undertaken with some representative papers. And what you want is to fill up the whole triangle. You want to hit all the corners of the triangle and get all three of these things that you want really well.

But the trilemma is you can't do that. So every one of the triangles falls way short of touching all the corners, because each paper comes short in at least one dimension, if not even two or three dimensions. The dominant approach is the bottom left of just looking up this variation in weather, and there you do great on A and C, like I already mentioned, but you do horribly on B because the weather's not climate.

So, we always have to sacrifice something or other, and we're just never going to know the full answer. There's just no method that's ever going to get us there.

So then, so kind of shifting into the paper that I think I was more asked to talk about. So why am I thinking about damages to climate change to date? That is not something economists have traditionally studied. As most of you are familiar with, you know, economists are really interested in the right price on emissions. What should the carbon tax be? There's a lot of my work has addressed that.

And really, what's relevant for that is not with DAM just to date had been. That's irrelevant. Our current emissions don't affect climate change to date. They affect future climate change. So we want to know the damages from future climate change. If we emit a little bit more today. But damages from climate change data are irrelevant. Like, you might want to calculate loss and damage compensation. You might want to know adaptation.

You might want a reporting device just to get people informed and make it salient in the news. Imagine that the Fed reported climate change damages along with inflation and employment on a regular basis, or the BEA or somebody like that. That would get a lot of press. So there's a whole bunch of reasons why you might want to know that.

On top of that, I'd also argue that something we can actually get at. Like, I've said that I don't think we're gonna get at future climate change's cost in any sort of semi-accurate way, but I do think that we can get at recent losses in a more credible way. And there's two reasons for that.

One is that the… part of the problem with future climate change is we gotta project way out of sample. I mean, we're projecting into the future. Imagine in 1900 that I told you that we'd have climate change in 2020, and I asked you to figure out what that would cost the economy, you would have… you would have gotten everything wrong. Like, you wouldn't have even imagined what the economy looked like and how weather might matter for the economy today.

It's just things are too different. But climate change to date, I mean, partly because they're fairly early in the process, the economy is not radically different. Our technologies aren't radically different from what they would have been without climate change. So we can, you know, somewhat say what the world would have looked like without climate change.

And that's much harder with future damages. We have to go what we call way out of sample. Like, we just don't have that kind of economy in our data.

The other big one is that linear effects might suffice. So climate change is not going to be a small change in weather. It's going to be small changes year by year, where they add up to a bigger change. And nonlinear effects might be quite critical, and we just don't have the data to pin that down. I mean, it's going to be almost a pure assumption. We just can't say what the degree of nonlinearity is going to be.

To date, climate change hasn't been that large in the bigger picture, and so a linear relationship is probably a good first-order approximation.

Okay, so how would you estimate the cost? So what you need to know a couple of things. So one is you want to know what the effects of current and past weather on income are going to be and current and past as climate change has affected current weather and affected weather in previous years. And I've argued you want to account for that because capital and all that adjust.

And you want to know the effect on whether both locally and around the rest of the country. And obviously, he had more data around the whole rest of the world. So you want to count from national weather patterns might matter, and how weather where you live might matter, and income is going to be what I observe in the data.

And then, but then the question is, how did climate change affect weather? I leave that to the climate scientists. There's a whole set of experiments they run for the IPCC for detection and attribution, where they simulate the world with and without humans, basically. And then I can average across these models and get on average, what would the weather have been in these locations without climate change?

Compared to the weather we actually had, and you get a climate change signature. I'll show you that in a bit. And then, once you've got those two pieces, it's combining them. You know, the effect of climate on weather, you know the effect of weather in different years and in different locations on income, and then you can calculate the difference in income.

The results from one. So hot days hurting come per capita on average. So the top row is the average effect of each kind of day, um, points to the left are going to be losses from a day of that type, an extra day of that type in a given year points to the right are going to be benefits.

And the top rows for a fictional average county in the United States, and then I'll talk about the bottom rows below that. The red ones are hot days. So hot days are to the left, hot days are costly. Giving the fictional average county in extra hot day is going to reduce income of about 0.1%.

Warm day is also costly, not significant, but, you know, it seems like it's likely costly. Cool days are beneficial, cold days are pretty noisy. I can't really say a lot about it.

Then below that, we can get a heterogeneity around the country. So how do different kinds of counties, how are they affected by heat or by cold? So if you're richer, how does that alter the effect? So being richer actually makes having a hot day where you live more costly. It makes having a cooler cold day more beneficial.

It makes you more exposed to each kind of weather.

being further north also makes different kinds of heat more costly and makes cold and cool days, you know, maybe it's maybe more beneficial, although it's kind of noisy.

The big one I'm going to come back to in a bit, agriculture. So hot, if you live in a hotter county, and if you live in a more agricultural county, you're more exposed to heat.

And then cool and cold days, not much of an effect. And this is consistent with other… there are plenty of studies in agriculture, but it's easy to observe… get data on yields and link that to weather, and we do see beyond a certain point for most crops that extreme heat is damaging, yields fall, and that is arguably what's being picked up in this sort of relationship.

Okay, I won't spend much time on this, but this is now saying, let's not just make the weather today hotter, but let's make the weather hotter where we live today, but do so also for each one of the 5 prior years. So we're making weather hotter every day for 6 years in a row. This story is basically the same, just the effects are bigger.

The main experiment that I want to point out before going to the climate side of it is is what if now you, the experiment is the thought experiment is going to be, we're going to take the entire country and make every county of give it one more hot day, or one more warm day all around the country at the same time for 6 years in a row.

So we're going to change the weather pattern around the whole country at one time. How does that change things?

So for the fictional average county, an extra warm day around the whole country is damaging. An extra hot day around the whole country is probably damaging, but it's noisy. You can't really say for sure. An extra cool day is probably helpful, and an extra cold day is… seems to be clearly helpful.

How does that change with the characteristics of a county? Richer counties are actually now less exposed, which might might have been what you'd expect. Being further north makes you less exposed to heat.

And really nicely, because this is the one we actually have other papers to corroborate, um, be more agricultural actually makes you less exposed to heat.

So why would that be? Well, if you're more agricultural, why are you more exposed to local heat, but less exposed to widespread heat? So what I can't do with the data I have is really get at the channels, but this this is suggestive of what's going on.

So imagine that you're a farmer, and you're a farmer in Iowa growing corn, and your county got hotter this year, but nobody else's county had different weather than normal. So you've got an extra hot day, nobody else did. Your yields are lower, and that's bad for you, so your income goes down. And that's what we saw on the on the top left plot.

On this plot, the experiment is all the other farmers also got an extra hot day, so all their yields went down. You like having more yields, that's bad for you. But if everybody else's yields go down, the price of corn goes up, and that can actually be good for your income. So the effects can really switch when you start thinking about everybody being impacted at once.

Your ideal would be, I don't get the extra hot day and everybody else does. Then my yields are still good, and my price, the price of my corn is now higher. The consumers of corn aren't going to be thrilled about that, but as a producer, that's going to work in your favor in terms of income.

Okay, so results from two. So so that that's the effect of weather on income. The other piece we needed was the effect of climate on weather. So if you take these climate models, and you and you average there the results from this experiment, you can get a trend in hot and cool days and cold days and warm days over time.

And we see hot days have increased… hot days around the US. The average US weather have increased over time. So we get more hot days now than we used to, more warm days now than we used to, cold days actually don't have much of a trend, a lot less cool days than we used to have around the country.

If we look at that by location, the purple is less of something, and the green is more of something. So top left is the effect of climate change on cold days around the country. And there's just a lot less cold days in most of the country than there would have been without climate change, except for a big hole in the middle of the country. I'll come back to that.

Uh, cool things, less cool days everywhere, all over the country. There's a whole lot of purple. Bottom left is warm days, a lot less warm days in the south because they've transformed into hot days, and a lot more warm days in the upper parts of the country, because they were cooler days.

And hot days have gone up basically everywhere, except for the Upper Midwest. And that's the lower right, with a lot of green.

What's going on in the Upper Midwest? This is a phenomenon known as the Midwestern warming hole, and there's explanations for that, I think partly involving the polar vortex. But this is something that climate scientists have known about, so it's kind of reassuring to see that in these data as well.

Okay, so now we're going to combine all this to calculate impact. So with the effects of weather and income per capita, we have the effects of climate change on weather. Let's see what the effect of climate change on income per capita is.

So, the left one is what I call a short one local calculation. So here we're ignoring the fact that climate change is persistent and widespread. We're acting like climate change just affects weather. Today, where you live, and forget about the past, forget about the future, forget about everywhere else.

The peripheral are places that are harmed, the green are places that are benefit. So if you just did this, like, local weather calculation, climate change is totally transient, totally local and contained, then many parts of the country, the hotter parts of the country, would benefit from climate change in the middle of the country would be harmed by climate change.

The middle panel is a long-run local calculation, so now recognize that climate change changed weather in past years as well. But forget about the rest of the country. It's only affecting you in isolation, your county in isolation.

It's the same pattern as before, just more exaggerated. The purples are more purple, the greens are more green, but the high-level story is the same. Hotter places benefit, colder places are hurt.

The right is the full calculation, and really the correct calculation. So now we're counting for the fact that climate change is long run, it's persistent, so where it's affecting everywhere in past years, and the data, and based on the models.

And it's also affecting the whole country, and it's affecting the whole country based on how the model states affecting it.

And now you get a lot of dark purple, and that dark purple is covering most of the country. So it's really that nationwide pattern of climate change that is really critical for finding a significant reduction in income per capita. And that pattern affects who wins and loses.

Once you account for how prices adjust to climate change, and it affects the scale of those losses, as I'll show you on the next slide. So the accounting for national patterns is actually really critical.

So climate change. So this is aggregating over the whole country now. The top row is the total effect of all kinds of weather, and the red is if you pretended climate change only affected us today where we live. Pretty small effect detectable, but small. 0.3% of income. Not a big deal.

The blue is the full calculation, where you recognize climate change is persistent and widespread. And there's a 12% reduction in income, and definitely not zero. Uh, somewhat noisy, could be 22%, could be 2%, but clearly detectable, clearly not, like, 8%. It's clearly something on an order that people might care about.

Where is that coming from? There's limited data that I can work with to really get at that. I think there might be better data available soon. So I do a whole bunch of experiments where I construct national weather in different ways. The one I want to focus on is that in the formal model in the paper.

There's two reasons why National Weather might matter. One is that it might affect your trade with other counties. The other is that it might affect your beliefs about future climate change, because you learn about future climate change from weather everywhere around the country, and that might affect your investment patterns. So I can get at trade using.

somewhat crappy data on the state-to-state trade, and I can get at the belief channel by using data on, basically, Facebook links across counties, social media links across counties. Uh, can tell you something about, like, your networks across counties.

And if you use the trade data, you exactly replicate almost the main result. I mean, that basically recovers what the main result was. So it's suggestive evidence that it's really trade driving the full effect.

Uh, so this is a new framework for figuring out how climate's affected the economy. The main takeaway annual losses might already be 12% of national income. And, you know, big ticket policies that we talk about, trade policies, tariffs, monetary policy, big immigration restrictions have effects around that same order of magnitude. So climate change is like that big kind of policy so far.

It might be less salient to a lot of people if it really is operating through price channels. People, I think, respond to how the weather outside their door is affecting them, but if I see prices doing something or other, it's really hard for me to attribute that to weather everywhere else around the country. It's sort of, you need stats to do that.

This is not a full accounting of climate consequences, it's just day-to-day weather around the country. There's a whole range of ways that climate can matter that might be important, both the kinds of weather it affects, the kinds of physical phenomena it affects, and the kinds of outcomes, like mortality.

And what you could really imagine is institutionalized in this calculation. The government has better data that they could work with, and there's gonna be new years of data coming out every year, and so you could just do this and report year by year what the effect of climate change was this quarter on.

Jobs on income, or whatever it might be, and just the benefit of that for just publicizing what climate change is doing to the economy, I think would be huge. It would just make it a much more salient outcome.

Okay, so the last part I want to talk about is what this might mean for policy. So a different way to do policy that might not require knowing the future damages from climate change.

So the idea here is going to be to use markets to project damages where all the government has to do is the kind of reporting I mentioned on this last slide. Just keep updating the calculation of recent climate change or the cost of recent climate change.

So the typical way environmental economists think about pollution policy.

We're sitting at our offices at our computers, we measure what we call the social cost of pollution, so the harm that pollution provides to the rest of society. So this would be the cost of climate change from emitting today.

Where the regulator then theoretically would implement the policy, and then firms are then going to see the tax or see the cap. They're going to consider their own cost of reducing emissions, and they'll make the right decisions based on that. And this is the idealized, perfect way that policy should work.

How it… so what we've done is we've centralized measurement of social costs, centralized measurement of the damages of climate change among people like me and among people like in the EPA.

But really, that might make sense for a lot of public health problems, but where, like, you don't know what the air pollution outside your window is, because you can't see it. But climate change, we're all exposed to weather. We all have a sense of what we can do in response to weather, and we all have some information about that.

So, in cases like that, where there's dispersed information on values, the economists usually want to use markets because they don't believe that any one person can really get… has all access to all the information. So can you use markets as a way of aggregating what all the people around the planet happen to know?

About the cost and benefits of climate change? So this is going to be basically like price discovery for the cost of an externality instead of price discovery for the value of a company. So what would a policy that could do that look like?

Okay, so the three goals of this new type of policy are going to be you want to efficiently price emissions like you do with an emission tax. You don't want to lose that. So conditional on the information you have, you get the right price on emissions.

I also want to incentivize removal of past emissions, which an emission tax can't do. So if you've emitted and paid the tax, you have no incentive to go pay more money to remove your past emissions if you want to go net negative. You've already paid the tax, you're done. And I want this policy to do better on that front.

I also want to actually get better information about social calls, because as I've argued, I don't think we're going to ever do a great job, and maybe this isn't a job we should be centralizing. And so what would it look like to let markets do that?

As emission tax achieves only the first goal, it can just get the right emission price. But the informational problem is all on the regulator and on whoever's advising them. And then there does nothing on removal of past emissions. You have to have a subsidy or something like that to go with it for things like air capture.

And here, one policy is going to try to do all three of these things at one time.

Okay, so the the model now is going to be agents are going to measure their own exposure to climate change. Our regulator is going to measure recent aggregate damages, just like I did in the work I just showed you. Agents are going to then trade an asset that I'll talk about in a bit that I call carbon shares. So it's a piece of paper tied to units of carbon in the atmosphere.

Firms are going to observe the price of these things in a market, and they're going to compare their cost of reducing emissions to that price, and then make the right emission decisions. So basically, instead of setting an emission tax, you report damage measurements year by year, and then this asset that's traded on a market will then.

transform that into what the projected damages in the future will be. And that's really where the trick is.

So the whole key is, what is this thing? What is this piece of paper that's transferable? So in your mind picture there's an emitter at a given time, and they're going to put down a deposit at the time of emission. It's not a tax, it's a deposit, and it's going to be partially refundable. And that goes to our regulator of some sort.

goes to the government. The government then gives them something in return. It gives them this piece of paper I'm going to talk about that's going to have a value that we'll call Q. So and then they can then sell this piece of paper on. They don't have to hold on to it. They can sell it on any market like any piece of paper that gives you a claim to own part of a company like a share in a company.

So the incentive to reduce emissions is that, as you'll see, this value, this piece of paper is less than the deposit. So you're losing money when you emit. You're giving up D, but you're getting something back that's less valuable, so you're losing money, so you don't want to emit. Um, but still, there's this other thing going on. So what is this other thing?

So, if… imagine that they emitted at time t, let's say in 2020, um, and at 2050, that carbon is still in the atmosphere. So what's going to happen year by year is that whoever owns the piece of paper tied to those emissions is going to receive part of that deposit.

The regulator's gonna say, in this year, it looks like climate change cost us this much, maybe using a method like what I showed you a bit ago. And they're then going to say, your deposit implied, you know, sort of a worst-case damage that was bigger than that, and we're going to refund you the difference. And then what's left in the piggy bank will persist into the next year.

So basically, year by year, we're gonna… you put down a deposit. It's something like, how bad could climate change possibly be? And then year by year, we'll see how bad it actually is over time as climate change unfolds. And then if climate change is really, really bad, you get almost no refunds back. And if it's really just not a thing at all, you get a full refund over time.

And so so keep that in mind, because we'll come back to that.

The other thing that could happen, though, is let's say in 2050, whoever owns that piece of paper might say, well, I mean, screw this, air capture is cheap, uh, damages are high, I'm not getting many refunds back. Let's go ahead and pay to remove carbon from the atmosphere. Let's suck it out of the atmosphere. And in that case, they get whatever's left in the piggy bank. They're basically pulling the carbon back out, and they get the whole piggy bank back.

Why would you do this? Well, if the cost of removing carbon is low, and if your piggy bank's getting drained by damage charges, rather than going back to you as refunds, you'd rather just pay the cost to get rid of the carbon from the atmosphere and get the whole piggy bank back rather than draining it over time.

So an example lifetime. So in 2020 you put the deposit down as an emitter, you receive a share worth that has some value in the market.

Oh, you might sell that to some shareholder who then buys it in the next year, and they then they then start receiving refunds year after year. And they're receiving refunds the next year, and they hold on to it, and eventually they pay some money to remove their carbon from the atmosphere and then they receive whatever's left in the piggy bank at that time.

So why does this work as a as a way to collect information about damages? So this is a dynamic deposit refund instrument. So deposit refunds are things that we're familiar with. So the most prominent one is like bottle laws in a lot of states. So you you pay a little bit more.

to buy a Coke can, then you get that money back if you bring it back to be recycled. So that's a deposit, you put a deposit down, and you get a refund back if you do the right thing with the Coke can.

This is a dynamic version of that, where the refunds are issued over time based on measurements of climate damages over time.

So the refund's gonna be, as I mentioned, gonna be big if climate change is not a thing, and gonna be nothing if climate change is really, really bad.

So why do you want to reduce emissions? Well, refunds are only partial. You expect climate change to be something. It's not going to be completely nothing. So you'd rather not have to put down the deposit and only get some of it back as a refund. You have an incentive just not to admit it all. And I can show that that incentive is what the emission tax should be. It's equivalent to the proper emission tax.

Once carbon's emitted, you can now do two things. You can bet on future refunds. So basically, you're betting on climate change not being that bad. Or you can take whatever's left of the deposit by just paying to remove the carbon from the atmosphere. So you have a choice that you can make in every period.

If you think about what the value of owning a share is, what's the price of a share? What should you be willing to pay for it? What you're buying is the right to get refunds in the future. So the value of a share is the value of those refunds in the future, just like the value of a stock is the value of a dividend of the dividends that a firm might give to you in the future, if you own a share in that firm.

So you're so a share is more valuable when you expect future refunds to be high, which means it's more valuable when you expect future damages to be low, and it's less valuable if you expect the future refunds to be low, so future damages to be high. So you don't want to just sit there and own this thing if you expect high damages.

But you're happy to own it if you expect low damages. So the market value of the share is the projection of future damage measurements. And so and what you're doing when you choose whether to admit or not is you're comparing your benefit from emitting to the cost, which depends on the market value of a share.

So if you get a valuable share back, then emitting looks pretty attractive. But if the share that you get back from the government is worthless, then it doesn't look very attractive. And it's going to be worthless when you expect damage measurements to be really high. So the market's now in the business of projecting future damages, not the regulator.

The regulators just got to measure what's been happening in recent years and report that.

So kind of a conclusion, summarize what I've kind of gone over. The comments have learned a lot about the direct effects of weather and a good bit about adaptation. We're never going to be able to affect the full effects of climate change in the way we'd like. It's just not something we're ever going to have the single right number for. It's just climate change is just not weather. We can get at the effects of weather really well, but climate is a much higher dimensional problem with all kinds of weather wrapped up into it.

It's anticipated persistent and widespread. We can make more progress than the cost of recent climate change, partly for reasons that just the economy looks like what the economy would have been without climate change, and so we can use more conventional methods. My estimate is that it's already costing the US around 12% of income, although it's somewhat noisy, likely through trade.

And I think we should think outside the box beyond emission taxes and cap and trade. Climate change is a different kind of problem, with a pollutant that accumulates and hangs around over time, and that we might want to pull back out of the atmosphere in the future, and we want to maintain an incentive to do that. So we might use markets to project future damages.

And really, what the government then needs to do is institutionalize the measurement of recent damages, and just report on that regularly, just like we do with jobs or inflation.

So, that's what I've got, so really thank you for taking the time, and I'm happy to hang around and take questions.

You don't know.

And Michael, you're you're muted. Yep.

Oh, I'm unmuted. Oh, great, great. Derek, thank you. Uh, that's, uh… that's a lot of information in your final, uh, scheme.

to use markets to actually measure damage projections. I, at least, am going to have to think about. It's fascinating. I have one sort of start question, and then I'm going to

turn to a few of the specific things that came up.

But as I mentioned to you earlier, what really got us involved in your research, as fascinating as all of this stuff is, is the…

is the ex-post, uh, measure of climate damages in the United States.

Right. Yep.

And that 12% really struck me. I'm not an expert in this, but it seems to be an awful lot larger

then some of the studies I've seen, for example, there was a study by Klossig in Brookings last year that came up with a far, far smaller number.

Um, how does your number compare to what we find in other studies? And I think it's a lot different. Why is it so different than what we see in many other studies? That's my question.

Okay, that's a great question. Yeah, I haven't seen many other studies trying to tally up the cost of climate change. The current cost of climate change in this kind of way. I've seen there's some work focused on agriculture. There's some work focused on like.

specific disasters and, like, sort of tallying up their cost and maybe linking those to climate change. But I haven't seen the exact comparable number for recent climate change, like shifts in daily weather. There are economists who spend a lot of time, as I mentioned, projecting future climate change, and there, the more recent work does suggest higher losses than the earlier work did.

And part of that is because recent work is using different methods, and part of it's because they're counting more for the widespread nature of climate change. And there, I mean, some of the work that I don't know if the numbers are totally credible, but I mean, those numbers would for future climate change, and some papers would make this look pretty small.

There are other papers where this would be pretty comparable to what they would expect. But again, I'm really… it's the trade channels that are key. I mean, if I just did the conventional thing and looked at just whether where I live today being affected and nothing else, I get, like, 0.3% of income. It's not that big a deal.

It's really accounting for this trade channel as the main contribution of the paper, beyond just the thought experiment of doing recent climate change. And that's something that really not a lot of work has gotten at very well yet, because economists have shied away from that a little bit, just because it doesn't fit in the kind of method that had been used.

But a lot more work is coming back around to that, and systematically finding trade to be pretty important.

Um, a related question, Bill Jones was just asking, does your model, which is correlating, essentially, temperatures, uh, with…

with income, is it accounting for extreme weather events like, like, uh…

hurricanes.

Yeah, it's a great question. So only insofar as they are directly correlated with an extra cold or extreme cold or extreme hot day. Things like hurricanes are probably not accounted. There might be some aspect of wildfire that might be accounted to the extent that it's correlated with extreme heat. But.

Most kinds of extreme events you might think of aren't necessarily directly in here.

Yeah, yeah.

and certainly nothing with like sea level rise.

Yep. So it's not taking into account extreme weather events, and those are important in some of the models I've seen.

No. Correct.

It's… the local… purely local stuff

you found, is it important at all, right? And that's what we think of these hot days and our health, and air conditioning expenditures. That doesn't matter.

Well, it does matter for your health. It does matter for your health and for mortality, and in terms of income, it's going to be the local effective climate on local weather is a pretty… pretty small potatoes. It's the fact that it's changing local weather everywhere at one time.

Yeah, it's this everywhere at one time. And that's what I want to get at.

Yeah.

changing… I mean, what kind of… do you have examples? Have you been able to identify what those linkages actually are, so that when we talk to people and we say,

hey, 12%,

Yeah.

The reason is your life is affected in the following ways. Are there concrete ways we can explain that to folks?

I would love to. I mean, there's a lot of work showing the effects of heat on all sorts of outcomes from productivity to agriculture to education. I mean, there's a whole literature in this now. In terms of within my study, I would really love to. It's just the data on… I'm using this data going way back on county-level income.

And, like, there's not even county-to-county trade data. If, like, they were a good county-to-county trade data, I could see so much more, especially if you were broken down by sector. I could figure out which sectors are driving it, and, you know, how it's, like, mapping through prices. But, I mean, even… I can get a… There's pretty strong suggestive evidence in the paper in various ways that I think, well, decently somewhat strong that it's trade is really the channel. But even that, like, it's just state-to-state trade data that even that's not that great. Like, it's… there are, I think, county-to-county trade data that are coming down the pipe, um, but they're not available yet, and without that, I really can't…

Drill in super deeply. I would love to. I would totally love to know exactly what's driving it, but I can't get too far under the hood with the data I have, unfortunately.

Yeah, yeah, I understand.

Maybe the international economist could help you out on that, but of course, between countries,

We do have country to country trade data, correct? That's right.

Yeah, we have that, but the thing is, I mean, what we know is it would be outrageous to say that shocks in Europe

have 8 times more important effect on U.S. incomes than U.S. shock.

you know.

But that's the thing. And so, and you see that in when I try constructing. So my my base case, I the the way I constructed National Weather for, let's say, for Pima County is I weighted weather in different counties based on how close they were to Pima County. So whether in closer counties got a higher weight, and whether in further counties got a lower weight.

And then when I did the trade variation, it was based on your trade linkages instead of on that. But when I do a variation where I take what we know from international economics about this, how the strength of linkages decays with distance from trade, I get much smaller effects, but that's because the US is just way more integrated than the international economy is. And so you would expect to get smaller effects if you did it that way.

Yeah, yeah. Uh, I… folks, I don't want to monopolize this. I… I did see that Ed earlier had a question on,

how successfully U.S. firms are adapting. He points out that the government

An insurance company seemed to be adapting. Derek, are there… are there other areas where companies just seem to be blind to adaptation? What's…

Yeah, I don't have an… I don't have an opinion on areas where companies are just completely ignoring it, but there is a lot of work now showing that company where companies are not ignoring it, like, even changing supply chains to be more resilient. Like, this is… The reorganization, both geographically and in terms of the kind of investments companies make is a very real thing. I mean, people… at this point, I don't think there are many, many people are very well aware that climate change is already here and coming down, and coming even more down the pipe, and…

I don't think that they're incentivized to completely ignore that by and large. I don't… I am not aware of a study that has… credibly gone sector by sector and said this sector is… I find no evidence of adaptation here, and evidence of adaptation there. There are… there is some work… there is one strand of work that does suggest there's not a lot of adaptation, but I can't say that I believe that conclusion, and that would be a more wonky discussion we could talk about later. But so I haven't seen a study that I believe that goes through that.

Okay. I see that Richard McAllister has raised his hand. Ed, how do we recognize Richard?

Hello!

I think he's unmuted, so I think you can just go ahead and talk, Richard.

Richard, fire away!

My… I… Yeah. Can you hear me? Okay?

Yep.

Yep, loud and clear.

Oh, yeah, you're welcome.

Okay? Thanks, Derek, for your presentation. And I I have to say up front, I got on about 10 min late, so you may have answered my question. But when you say through trade channels.

12% of income. Do you mean that the Us. Would have to import more than it currently does, and that would reduce its GDP. What? What specifically do you mean by trade channels?

Yeah, thank you. Yeah. And that may I don't know if it was would or would not have been clear. Yeah, that study was only for the Us. I had data on Us. Counties. So this is county to county trade around the Us. So the point I was making in the beginning.

is that climate change doesn't just affect the weather where you live, it affects the weather everywhere all at once. And so I find that most of the losses due to climate change to date in the US are due not to the effect of weather on individual locations, but due to changing the pattern of national weather.

And then there's suggestive evidence that why does the pattern of national weather matter? There's suggestive evidence that's because of trade linkages across counties within the US. The other plausible mechanism would be beliefs that I'm learning about future climate change by observing weather elsewhere around the country.

And the, you know, doing, like, a back of the envelope test of a horse race of beliefs versus trade. Trade seems to be the clear winner. I mean, it explains everything, and when they're together, beliefs explain nothing.

Yeah. Derek, maybe I can clarify a little bit more. Uh, Rich, for example,

Suppose, uh, uh…

Yeah, go for it.

in some county other than yours, there's bad weather and an agricultural shock.

And you happen to be buying that kind of food. Well, that bad shock would increase the price of food, and you would have to pay more for food,

Exactly.

Even though the weather didn't hit you, it would have affected you through this trade, this trade channel.

Okay. And as importantly, it's affecting the entire supply chain. So, like, just sticking with agriculture, that's an input to all sorts of other sectors. I mean, food and fiber, it goes into livestock, goes into textiles, goes into a lot of things you wouldn't even think of offhand. And so then that… that shock to agriculture, that negative shock, is going to then ripple through the whole supply chain.

And that's going to then affect income everywhere else around the country. And there has been work on that internationally, showing the agricultural shocks internationally ripple into other countries through supply chains in exactly that kind of way.

Yeah, yeah.

So it sounds like what you're saying is you're restricting this analysis that yielded 12%.

to intra-US income. Okay, understood.

Correct. Yeah, I'm not looking at. I'm basically ignoring the rest of the world and whatever effect climate change had, because the data I have is on US counties. Basically.

Yeah. Okay. Thank you. That was very helpful.

Yep. And by the way, ignoring that could go either way in principle. It's not… I mean, you could bet one way or the other, but in principle, it could go either way.

Yeah.

Other hands raised out there?

Well, um, I'll just ask her a quick question. Derek, have you had any, um…

I mean, without being specific, but have you had any conversations with policymakers about some of your research?

I mean, are you starting to see some interest in this? I mean, because one of the things that CCL's very interested in is

you know, how we can apply this to policy,

And, you know,

to be able to… to get some sense of how, uh, how policymakers might be reacting to this.

Yeah, um, I'd say that those conversations have happened with more frequency at other times of life than maybe in the last year or 2. There have been… I mean, people are, I think, not uninterested in climate change. As we discussed earlier, it may not be the front burner issue that it was at other points in time.

I've been pushing more subtly on the institutionalization of the measurement, just because I think that's something that's achievable. I think it would go miles toward an emission tax to have a framework for taking measurement, you know, making it just more official in the way that we've done successfully with a lot of economic statistics.

So that… yeah, it'd be great to be ambitious and actually get, like, a mission pricing and all that, but I do think institutionalizing just measurements of climate change would just be… that would go in more of an indirect way, miles toward eventually doing something with climate change. So that's where I have had some conversations around that.

Um, but, you know. Things take time.

Yeah, right.

Derek, I've got a… I've got a quickie on a… on a… some… a topic I know you've worked on.

Uh, that you mentioned here, uh, background, of course, as you know, the EPA has stopped using a social cost of carbon. Well, it says…

Mm-hmm.

It's so uncertain, uh, we ought to… we ought to treat, uh, uh, pollution as zero, zero cost.

You…

Right, they haven't stopped using it. They just zeroed it out, right?

They zeroed it out. You have… you have studied how risk and uncertainty

should affect the way we compute the social cost of carbon. Have you found that more risk means we should make it lower? Help us understand how risk and uncertainty affect that thought process.

Yeah.

Yeah, typically, the way we deal with risk and certainty in essentially every other walk of life is that we price it. We have very sophisticated tools for doing that, because we deal with risk all the time in insurance markets and in financial markets and in day-to-day decisions. We don't just act like we can't make a decision because we can't exactly quantify exactly what's going to happen. We… Price the risk. So there's two main things you gotta think about when you want to think about the effective risk uncertainty on the social cost of carbon. The first is just the mere fact that things are risky. Theoretically, it means that you should increase your savings, and one way of saving is to reduce more emissions. And so you're basically saving in terms of like the future climate.

So that channel works to increase the social cost of carbon, and it always does.

The other channel is more theoretically ambiguous. So what we often care about with risk.

is if I imagine that you own a stock, and this stock could do really well, it could do really poorly. Um, so it meant, just to pick a company, I mean, imagine it's, I don't know, IBM. So you own IBM stock. And what you really care about is not just the fact that it's risky, but is IBM likely to do really well in states of the world where I happen to not be doing very well?

In which case, it's really valuable to me, or is it likely to do really well in states of the world where I'm also doing gangbusters, and in which case I don't really need the money? In which case it's less valuable to me. So that correlation between kind of how we're doing and when this thing I'm holding pays off or not is really critical.

With climate change, that means for insurance reasons, what you… and this is how insurance works. Insurance pays out when you're not doing well, so you get a lot of money back exactly when you needed the money. That's why insurance is so valuable. So this insurance channel for climate change, the way to think about it, we're uncertain about what the future damages and climate change will be.

There's a world in which damages are really high and reducing emissions avoids a lot of lost income. And there's a world where damages aren't as high, because either we adapt a lot, or the climate isn't as sensitive as we thought it was. And then reducing emissions didn't avoid a lot of losses.

If we tend… if the states of the world were damaged, the first order effect, and I've shown this formally, the thing that we should care most about is the fact that when damages are really high in the future, and of reducing emissions avoids, it saves us a lot of money in the future.

Those are states of the world where we're also relatively poor, because there's a lot of damages from climate change, and so that's really valuable. States of the world where reducing damages didn't pay off very much are states in a world where we got lucky. There wasn't really bad climate change, or we adapted really cheaply, and it's not that important to us that we saved a lot of money.

Yep.

So, this correlation makes reducing emissions look like insurance, and we're willing to pay for insurance, because we don't like that kind of risk. And so, the fact that climate… reducing emissions acts like a form of saving, and the fact that it acts like a form of insurance because of this correlation of paying off exactly when you want it to.

Well, I was being facetious.

Typically, we think that risk and uncertainty raise the social cost of carbon rather than, and certainly you wouldn't just say it, you zero it out. I mean, it'd be pretty hard to tell the story about that.

And if we were to pin you down, what would you… if you had to put a number on the social cost of carbon, what would you do… what would you say right now?

Oh, I… I'm not going to offer that, because I think the damage side's just deeply uncertain, and I think that… and I think that's a… I think that price, that effective uncertainty is really high, um, but it's a pretty big adder, and my paper suggests that it's a big adder. Um, in terms of, like, what the baseline number should be, that's an add or two, that's a whole… I don't have a… Pick a number that I favor.

Okay, okay. Great. Um, we have exhausted you on St. Patrick's Day. Anyone else?

Oh, I see Bill had a quick question. If someone were to make the calculations that you would like to see done in the future with your model,

Who would you… who would you feel comfortable leaving that with?

The EPA…

The Fed… oops…

Does it matter?

I would like it to be something like the Federal Reserve Bureau of Economic Analysis, and I would personally locate it in the Bureau of Economic Analysis probably if I had to pick between them. But one of the two, it should be a group with access to basically the best data the government has that we're using to generate the economic statistics that we're using to guide decisions.

Bureau of Labor Statistics is the other big one. But I mean, they're really focused on labor statistics. It's not really a good fit. Bea is doing the GEP calculations. The Federal Reserve is doing lots of calculations that are comparable for because they need to know them for monetary policy purposes. And either one of those would do a really good job. I personally would put it in the BEA.

Because I don't see climate as being directly a core part of the Fed's remit and not something that they need to have thrown on them. But the BEA, this should be a calculation pretty comparable to the kinds of things that they could be good at doing.

Yeah, yeah. And they're the ones, I guess, who compile the, uh, the income data that would…

be relevant, you bet.

Exactly. And there were a few other groups that could also plausibly do it. I mean, that kind of institution, like, this is exactly what they're set up to do.

Yep.

And we used to not know how to do inflation. We used to not know how to do GDP. We had to figure all this out, and it took. We got better at it over the course of decades in each case, and in the beginning it was all speculative. It's something that we thought could be useful if we could figure it out, and we figured it out. So I think we could make a lot of progress with climate in the same way.

Fantastic. Again, thank you for spending this St. Patrick's

Afternoon, evening with us. I hope…

Yeah.

You are an incredible resource for citizens' Climate Lobby. I hope we can…

Stay in touch, and when technical issues arise, and we need…

Yeah, happy to.

Your advice, I hope we can be in touch with you, and maybe meet with you again, and bank some of this out.

really, really…

Yeah, happy to. Happy to really appreciate you all taking the time, too.

Right.

Thanks very much.

Really good. I see… I see, I see applauses flying up on the screen. Thank you so much, Derek. We appreciate it.

Thanks very much. Appreciate it.

Great. Great. Yeah, my pleasure. Thank you all for having me. Really appreciate the invitation.

Thank you.