A Historic El Niño Is Coming That Could Cost the World Trillions — Odd Lots

[00:00:00] Speaker 1: when i think about the cost of this forthcoming el nino you know i estimate somewhere around 10 trillion dollars of global economic losses conservatively over the next five years right up to 14 trillion dollars of economic losses if you take some of these tail but i could only 70 of that only you know about 10 trillion dollars of that is in sample meaning supported by existing data we have on el ninos and economic responses uh our global economy is larger and so these these losses are associated not just with the magnitude of the el nino but the size of our global economy [00:00:37] Speaker 2: there's just more stuff to be impacted hello and welcome to another episode of the odd thoughts podcast i'm tracy alaway and i'm joe weisenthal joe i don't know if i've ever told you about [00:00:54] Speaker 3: this but i'm a little bit of a weather nerd okay i did maybe a little but i knew i don't i don't [00:01:00] Speaker 2: even know if weather nerd is the right term i just get really excited by weather i love thunderstorms [00:01:05] Speaker 3: i love a good thunderstorm so great if i'm inside well this is the thing i love thunderstorms except [00:01:11] Speaker 2: when i'm on an airplane or something like that um i do not like economically damaging weather i feel [00:01:17] Speaker 3: like i need to throw that caveat in there i don't like uh economically damaging other you know what i am i'm like a weather truther no this is what i mean like i don't believe so for example i don't believe that there's any accurate way for example to uh measure wind chill it's like oh it's like 25 degrees out and it feels like 15. it's like how do you know what 15 feels like you know what i'm saying like i don't believe in some of these numbers also humidity what there's water in the air give me a [00:01:48] Speaker 2: break there's no water in there no that one i believe tell me you've never been to abu dhabi summer without telling me you've never been to abu dhabi no this last week in new york [00:01:56] Speaker 3: i felt like i know that one was a joke i do i am i am a uh a windshield uh skeptic even though i don't think that's the topic of our conversation today it is not the topic of our conversation is [00:02:07] Speaker 2: actually economically damaging weather yes and specifically the upcoming el nino which by the way it's it's not that that often that you see a lot of sell side notes start to mention weather phenomenon but just in the past you know few days i've had a thailand economics note from city talking about interest rates and quoting the potential impact of el nino potentially inflationary i've had a bank of england outlook from morgan stanley that mentioned el nino and also a bank of america fx outlook uh talking about the impact on em currencies so this is something that people are expecting to have an actual impact both on the global economy and the investment landscape you know i will just say [00:02:53] Speaker 3: even though i'm not a weather nerd the way you are i sort of have this thought that if you look at like the really grand sweep of history like thousands and thousands of years maybe even hundreds of thousands of years i sort of think there are two things that matter in human history one of them as you know i'm interested in is like basically technological evolution media technology and how it reshapes how we think and then the other one is sort of climate change weather and storms and i think that when we think about history many things are contingent either on like brief or long periods of weather change and it can reshape society in permanent long-term unexpected ways at the time and um have we ever even done any el nino episode in like 10 years or 11 years of doing this i don't think we have but it sounds like this is one that we should cover if it's showing up in all these uh [00:03:47] Speaker 2: if it's showing up on all these notes if we're going to cover one this should probably be the one because at the moment well you have different el nino forecasts yeah and you know there there's a range of uh probabilities involved here but a lot of the forecasting models are currently uh forecasting that this is going to be a potentially record-breaking el nino something we haven't seen in decades do you [00:04:07] Speaker 3: think there are any people who are like oh it's interesting you bring up the el nino you know i've always thought la ninos are the more uh underrated one do you think that's the type of guy no i guess i hope [00:04:18] Speaker 2: not but are you going to become like an el nino truther how do you know it's el nino unless you're standing in the pacific ocean measuring the water yourself i'm going to be coming out he's like you [00:04:27] Speaker 3: know you really should be paying attention to the upcoming lightning yeah oh good uh that'll be a fun [00:04:32] Speaker 2: transformation for you okay well we do in fact have the perfect guest we're going to be speaking to someone who not only follows the weather but also tries to measure its economic impact we have justin mankin he is professor of geography at dartmouth and director of the climate modeling and impacts group so justin thank you so much for coming on all thoughts yeah happy to be here thanks for having me i feel like maybe the first thing we should do is can you can you convince joe that el nino is is a real thing apparently he needs like a first person experience of it what is el nino yeah what [00:05:06] Speaker 1: is it yeah light light wind chill uh factor it is it is in fact real and measurable um you know where and they're both kind of actually aerodynamically related right so wind chill the feeling you get on your skin is owing to the fact that the wind is wicking away latent heat from your skin which is the heat associated with a phase change of water from a liquid to a vapor um and so that's why when you get out of a shower uh you feel cooler uh than before you get into it right so that evaporation off your skin induced by aerodynamic resistance i.e the wind that is the thing that's cooling you down which is why we can measure it and why we can tell you uh that cold exposure is is a bad thing um the el nino phenomenon is an ocean atmosphere it's kind of part of what's called the el nino southern oscillation which is a essentially a relationship between the tropical pacific and the overlying tropical atmosphere um and it has a tendency to be in what we call a neutral state which is kind of average conditions where you have these easterly trade winds so these are winds that are blowing from east to west across the tropics along the equator okay and what they're doing is they're dragging surface water surface water that's been warm because it's in the tropics by the sun right there the sun is just shining there all year round and so that water heats up and then the winds kind of drag that surface water and create a literally a pile uh such that the the sea altitude is higher in the west pacific than in the east pacific um and that pile of warm water is something we call the west pacific warm pool and so it just sits there parked by indonesia and australia um and then every once in a while those trade winds weaken and break down and that pile of warm water that actually penetrates quite deep into the western tropical pacific that pile of warm water just spreads itself out over the vast expanse that is the tropical pacific and so that's actually how we measure el nino is we're measuring it as a function of sea surface temperature anomalies so how much warmer than average our ocean temperatures at the surface you know at this time in this location um and you know there are different ways to kind of calculate what the regional average is and we can get into the that technical nitty-gritty but essentially all this water sloshes back eastward and in doing so reconfigures all of weather which is just energy redistribution from the sun on our our fluid planet right um it redistributes all of that energy differently than would have been the case if all that water were piled up in the western pacific i guess like a really simple analogy a listener could maybe grab onto is they've walked into a cold room where there's a little space heater plugged into the corner right and that little space heater yeah is maybe generating some heat that's warming up the room um but it's really only warming up that corner of the room that is akin to neutral or what are called la nina like conditions which is an amplification of the neutral pattern meaning meaning that that those winds are stronger than normal and uh the eastern tropical pacific is cooler than normal um when you slosh all that water back across the entire expanse of the tropical pacific it's akin to walking into a room that has wall-to-wall radiators instead the thermodynamics of that room that's a totally different experience for you than would have been the case if it was just that space heater in the corner and so when an el nino occurs what it is doing it is it is it is essentially you know that the tropics are the heat engine of global weather and climate and it's a complete reconfiguration such that so much more of our global atmosphere is exposed to that heat energy from the ocean that would have otherwise been buried in the western tropical pacific that was an [00:09:27] Speaker 3: amazing answer yeah i totally i understand something that i went 45 years without having any idea of and now i get it what and as someone who used to live in a in a drafty new england home i have a lot of experience with space heaters that only heat up that corner of the room what are the conditions or what are the drivers such that these el ninos are semi-regular why do they seem to come every few years why are they not a permanent pattern why is one year an el nino year and another year isn't yeah there's a lot [00:10:02] Speaker 1: of work on kind of the drivers of the breakdown of the winds in the eastern tropical pacific i think you know a a simple answer is that we live on a sphere that's rotating and has stratified and unstratified fluids in the form of you know the ocean and the atmosphere and and that essentially means that energy is propagated as wave phenomena right with peaks and troughs um and el nino is akin to a wave phenomena right it is a it is kind of a pattern uh what we call a mode of variability george philander who is this princeton geophysicist and and major scholar of el nino called el nino the trunk of the variability tree uh when it comes to climate variation in terms of kind of regulating these pulses of energy redistribution around our planet um and so you know a phenomena like el nino being modal meaning it has a canonical pattern uh temporally and spatially and then induces physical responses that we can characterize and use as a means of prediction right like we know what we should expect for peru we know what we should expect for australia and we're able to do that because we've diagnosed this mode of variability um and given it a name and characterized its length scales and characterized what we call its teleconnections meaning the kind of physical bridge between el nino as a phenomenon and what it implies for the indian monsoon or for the southwestern united states temperature and precipitation regime for that season and we've given these things and names and they become they've come instead of being purely a source of kind of stochastic uncertainty just kind of random variation in weather and climate they become a means of predictability and that's what's really powerful about investigating these type of phenomena and giving them names right we have el nino or el nino southern oscillation which is the kind of teleconnection with the overlying atmosphere we have the north atlantic oscillation we have the indian ocean dipole we have the madden juliet oscillation so we've we've like identified these patterns and their their kind of manifestation and that in terms of like their underlying drivers which is i think the question you're question you're asking i think it's just a function mostly of living on a complex planet with geophysical fluid dynamics and such that these wave phenomena just kind of emerge on these characteristic time scales the precise mechanism of action that kind of leads the breakdown of these easterly trade winds which like triggers what's called the bjerkness feedback which is this positive feedback loop so an amplifying feedback that tends to reinforce itself um is uh is you know can come from myriad different different places from from changes in ice cover to um you know a weakening of the temp of the the equator to pull temperature gradient all kinds of different things have some culpability and potentially causing a breakdown of these winds that allow the sea surface temperatures to slosh uh eastward again and [00:13:23] Speaker 2: reconfigure all severe weather on our planet so talk to us about what the forecast actually is for this particular el nino because i read these things like apparently the australian bureau of meteorology which would be interested in this because australia is particularly affected by el nino apparently but uh their index of relative the relative oceanic nino index or ronnie for short uh apparently has never reached the 3.3 thresholds um that they currently project before so that seems that seems extreme what are what exactly [00:13:57] Speaker 1: are we expecting here so far uh all of the earlier forecasts so there's a bunch of different ways to forecast el nino all of them kind of stem from some of the original innovative work in the late 70s and early 80s by mark kane and steve zediac who built the first kind of physical model of el nino to explain it as a phenomenon and then actually make predictions of its occurrence which is pretty remarkable um but all these are kind of based on um some of them are statistical in nature so kind of taking temperatures at time t and making a prediction given uh kind of autocorrelation in that time series what it should look like at time t plus one um there are others that are physical models that are actually simulating the relationship between oceanic temperatures winds the overlying atmosphere and what it does to the kind of pressure and circulation that actually gets air moving in the first place and so those ones are called process-based models and those are distinct from these statistical models but all of them regardless of how it's being modeled are kind of converging on a record el nino a record el nino at least since 1950 uh but arguably since much earlier right the 1877 1878 el nino um which was associated with uh the great famines and mike davis has this amazing book from 2002 about that that el nino event um that one i think is you know kind of indistinguishable in strength from el nino as we've already had such as the ones in 1982 and 83 uh the 97 and 98 el nino which listeners may remember as being associated with chris farley's snl skit of uh i i am the el nino right um and uh and you know the 2015 and 2016 el nino which i believe is the one on record depending on um your indices so we've had really strong uh el nino's but this one is looking like it's much bigger potentially um you know every uh forecast right now i believe puts the forthcoming the unfolding el nino because we are in a in a in an advisory and we have been since june according to the national oceanic and atmospheric administration's climate prediction center um so essentially you're looking at every forecast being uh having a warmer el nino than the 2016 el nino can i just ask because you piqued my [00:16:34] Speaker 2: interest just then but how did we actually discover el nino as a weather phenomenon because you know it it happens with some regularity but it's not like it happens every three to four years like right on time at the same intensity as the previous you know round so how did people identify [00:16:52] Speaker 1: this as an actual measurable phenomenon gosh such an awesome question um you know in terms of kind of like the modern instrumental and climate science era i don't actually know the answer to that but i do know something that's actually i'd argue even more interesting um so you know the the person who first uh modeled el nino you know um from uh from a physics perspective mark kane uh some amazing geophysicist at uh lamont doherty earth observatory at columbia university and his colleagues john chang and other climate scientists there at the time and then an um an anthropologist by the name of ben orlov who is uh an anthropologist at um at columbia they wrote this incredible paper i believe in nature um you know and back in the 90s or early aughts where they essentially showed that there was really good archaeological evidence that peruvian potato farmers as long as 500 years ago were forecasting el nino and they were doing it based on the visibility of the pleiades which is a a star system um you can see from the andes and essentially as el nino occurs what it's doing is it's reconfiguring the distribution of high versus low clouds in the eastern tropical pacific it's essentially taking convection that was happening near australasia and moving it eastward and in doing so you're creating all this these kind of high cirrus cloud shields over the andes and that reduces the visibility of uh this this this star pattern so they were for and and they would alter their their their planting dates based on that so there there's just kind of this remarkable cultural knowledge of being able to grab onto something and be able to make a forecast an informed forecast and so i'd argue there's a lot of cultural knowledge about these innate climate phenomena around the world that we have yet to kind of appreciate and diagnosed in the in the highly industrialized world where we're still relying on instrumentation [00:19:02] Speaker 3: yeah folk meteorology or folk weather is like extraordinary and like how much they're able to anticipate i was sort of i was what you know i imagined that there were probably various civilizations that interpreted in el nino as like the king has lost the mandate of heaven and now you know there's a drought now we have to have a new political order also this mike davis book that you mentioned looks fascinating it sort of speaks to the point of like late victorian holocausts el nino famines and the making of the third world talk about subsequent famines in india northern china and northeastern brazil definitely going to have to read that book um when we talk about a quote record el nino what is the measure is it the heat variation is it the expected change in uh rainfall or drought like what are the sort of the gauges that we use to say this is a record el nino or not or measuring that what what are the [00:20:00] Speaker 1: uh parameters of an el nino my understanding is it's all benchmarked to the magnitude of the sea surface temperature and okay um an anomaly you know is a departure from expectation or a departure from normal and so what constitutes normal becomes really important there uh for kind of setting the magnitude of how anomalous something is right and that's why you know when tracy was talking about the relative um um um you know nino index right that the roni right that's uh kind of shifting kind of a rolling baseline uh because we have secular warming in the tropical pacific owing to global warming that's occurring right as you heat up the planet the largest reservoir of heat storage just happens to be the 70 volume of water that constitutes our planet's surface so um and it takes a lot of energy to heat up you know a a volume of water it's much much more energy intensive to heat up the volume of water than it is an equivalent volume of air right um so that specific heat is just much higher so it's just a great energy reservoir and the ocean has been doing humanity an incredible solid by absorbing all this excess heat which is the problem of global warming right that we're not allowing our atmosphere to shed heat as efficiently as it otherwise would have been if greenhouse gases were um lower in our atmosphere and in doing so we're uh having to heat up to reattain equilibrium such that the energy coming in from the sun is balanced by earth light leaving out to space um that is not in balance at the moment and a lot of that excess heat is going into the ocean and that's how planetary temperatures increase such that in can equal out again um that's essentially a very simple global warming 101 but the um you know the way we're measuring um el nino as a phenomenon is by taking a regional average so choosing an area in the tropical pacific that is canonically associated with the manifestation of el nino and you can choose different regions and then benchmarking those average temperatures uh at in this month to you know expectation what have averages been sure um over over the top over time and so you can choose different regions so there's a nino three region which is kind of shifted towards the central pacific there's a nino four region which is shifted a bit more eastward there's a nino three four region which is kind of a typical region that's now used which characterizes el nino over something in between those two regions then there's the e-index which is taking a dimension reduction technique on a spatial field and collapsing it to say these are kind of the the dominant um directions of variability in that field and that e-index would kind of represent something differently in in the sense that el nino while we think of it as a singular phenomenon it has a geography associated with it and that geography means that it has flavors associated with it and so these flavors mean that different parts of the tropical pacific warm differently and look anomalously different and so not every el nino is the same and so it's consequences in terms of the hazard it brings for the united states or for hurricane occurrence in the trop and in the atlantic basin or whatever right it's not the same for every el nino event the the spatial pattern of the sea surface temperature anomalies also seems to matter and so there's been a lot of work on kind of diagnosing these different flavors of el nino and the e-index is intended to try to capture some of that so there are lots of different ways for measuring it but it's always a sea surface temperature anomaly in degrees celsius benchmark to some expectation or average of normal so one of the things that joe [00:24:04] Speaker 2: and i have learned in doing various agricultural episodes over the years is that when you have a supply shock the the timing matters right so you generally don't want to declare war in the spring planting season when all your soldiers might be off uh planting stuff instead does the timing of this particular el nino matter i keep reading that the peak is expected to come sometime in i guess autumn in the western hemisphere is that particularly uh unfortunate timing you know there's been a lot of [00:24:37] Speaker 1: work that looks you know the the planting and and kind of harvest of crops tends to be um something that is is is you know highly calendar dependent and there's a bit of flexibility that i think farmers have to be responsive to their expectations for for weather variability um and certainly you know ag forecasts have become important and and agriculture uh writ large has become really dependent on on el nino forecasts i think you know is there the the thing about el nino is that it does typically peak in december january and february so this is kind of the boreal uh meaning the northern hemisphere winter um and uh that has implications for the austral summer right for um the southern hemisphere summer which is in the opposite season um and so i think you know it also there's also kind of a time dependency associated with the impact so let me just concretize this with an example we've been in an el nino advisory since june that's also coincident with the typical onset of the south asian monsoon which is a determinant of feast or famine for billions of people right you can link monsoon onset to um you know farmer well-being and farmer suicides you can link the timing of monsoon onset to uh diesel prices within india and uh and and kind of the political fates of people around diesel subsidies because if the monsoon onset is delayed you have a bunch of farmers suddenly needing to irrigate crops where they otherwise wouldn't have had to and that that that irrigation is dependent on the generators that rely on diesel fuel um and so you can see price spikes around those types of things and and the political fortunes of of like municipal and kind of local politicians kind of surviving or not surviving in elections based on those kinds of things so there's like lots of cascading ways in which an el nino even at this early stage in its life cycle which tends to kind of start after what we call the spring prediction barrier which is kind of this period of time where where it's difficult to make a prediction about whether or not an el nino is going to occur um to you know next february so we're looking at kind of a wind down of this event in february march of 2027 what's interesting about it is that it's going to lock in all sorts of phenomena you know 2026 has a non-trivial chance of being the warmest year on record not simply because of global warming but because of this emerge in el nino so it is already kind of just increasing global temperatures over what they would have been in absence of this event and it's nearly locked in a virtually guaranteed record temperature for 2027 even though we haven't actually had that year unfold yet um it is like it's no coincidence that the hottest years on record have almost always been el nino years right when you take a warming trend and then you impose an anomaly on top of that you're just going to get the these outliers um and so uh these impacts are arguably already manifesting so the south asian monsoon was delayed uh that onset was delayed by this el nino and so while the kind of major teleconnections and the peak strength associated with this event is anticipated to be in the winter which is typically the case associated with with el ninos and just kind of the the life cycle associated with them um its impacts are manifesting now and will continue to manifest into 2027. let's talk a little bit about [00:28:26] Speaker 3: the uh economic costs that we associate with an el nino you know i like i said up until your first answer uh in this interview i didn't know what an el nino really was but that certainly doesn't mean i've never said it or read it before you know i used to be on tv you have a guest he was like oh the upcoming el nino joe is joking his beard pontificating about the el nino yes but it always seems like a sort of cyclical thing it's like okay it hits every few years and there's some costs and then it goes back to normal but you know if uh if there are whole books about the long-term consequences of any given el nino it sounds like at least in some serious ones there can be quite a tail quite a you know ripple effects when you go back and look at maybe some of the more recent big el ninos that we had i think you mentioned there was one in the early 80s how do you think about like tallying up i don't know what the the what the global economy the the price the global economy paid for it i want [00:29:33] Speaker 1: to call up a key collaborator um that i that i've had in this work uh chris callahan who's a professor at indiana university um and just a stellar stellar scholar i couldn't have asked for a better phd student to get to think about these hard things with um so you know chris and i chris had been thinking about el nino um and it's kind of projection uncertainty in other contexts meaning what does the future of el nino look like in a world with global warming um and how to how to understand that just be given that it's such an important part of year-to-year variation yeah and weather and climate globally um and it's essentially unknowable right it's very hard to know what el nino is going to look like in 2050 what it's going to look like in 2100 and how that picture changes as a function of what your background warming happens to be one of the things we had been investigating is just well we kind of had pivoted from thinking about future um to saying well we live in a world where global warming has already occurred we can simply document what has happened and so you know much of what my lab group does here at dartmouth is investigate retrospectively kind of build that documentary evidence of here house here here are the costs of climate change to date and in thinking about the cost of climate change to date one of the things you have to do is isolate those costs from the costs that would have happened anyway such as the cost from el nino which is an innate form of climate variability right el nino occurs in absence of people and their greenhouse gas emissions and so isolating that cost and kind of partitioning it from the cost accounting of global warming to date became a pretty important exercise so we're always in kind of detection and attribution science trying to say well what is consistent with the world as it you know has been versus the world as it would have been absent global warming and that means having a really clear picture of what we call internal climate variability such as the variability from el nino and so we were interested in this cost accounting um of el nino just to clarify the cost of global warming to date and when we went back and did this exercise using these causal inference techniques so this is all observationally built right we can take data of variation in el nino which is essentially a random treatment right it is stochastically generated while it has a life cycle you know against it you know our global economies aren't necessarily shaping sea surface temperature anomaly patterns in places like the tropical pacific right so it's effectively exogenous yeah to our from our economy right and so that's really crucial in this empirical setting of trying to identify well what's the effect of an el nino on an economy and how does that upward aggregate to a macroeconomic toll what you can do is just take a bunch of data of uh of something like gross domestic product which is summarizing all economic productivity goods and services produced by an economy we have pretty good data on that for at the country scale for a pretty long amount of time and you can look at differences in that economic growth so how did economic growth vary from year to year so you're not looking at absolute levels of gdp you're looking at changes in economic growth as embedded in changes in gdp and then you can look at something like el nino as a treatment and then you have another source of variation which is that not every country is equally linked to el nino physically um you know el nino is kind of remarkable in that it generates many different kinds of physical hazards from heat waves to droughts to floods to landslides to wildfires right and so in that sense like its impacts interestingly look a lot like the wider impacts from global warming which is a worthy discussion in and of itself but we'll put that to the side for a second and what you can do is you can exploit the kind of random variation we've observed in the real world of el nino occurrence and el nino magnitude how that relates to the teleconnection how tightly coupled locations weather and climate is to el nino as a phenomenon and then see what accounts for the variation in economic productivity as measured in changes in gdp and so that's what we did uh back in 2023 and published in a paper in in science where we showed that actually we can detect uh very sizable deviations in economic growth from el nino as a phenomenon and that scales in teleconnection strength meaning countries that are most influenced climatically and from a meteorological standpoint by the phenomenon as of el nino are most impacted um and negatively so um you know so peru with road washouts uh disrupting commodities from you know mine sites in the mountains to port uh induces a shock um and that that shock doesn't seem to be what an economist might call a level effect meaning a simple deviation that you then recover from uh in a in a subsequent year what we find in this empirical research is that observationally uh el nino seems to systematically depress growth meaning your country is growing at a rate and then an el nino occurs and then you are growing on a different trajectory and so those losses just accrue ad infinitum in time and that is a distinct finding than what was previously [00:35:35] Speaker 2: understood about el nino right so this is really important i think most people are used to thinking about weather related shocks as the sort of short-term thing that as justin just said impacts the price level but what the paper actually found was that there's this longer-term impact on essentially output like growth in the longer term why is that like what exactly is happening here because again you would expect like okay a road gets washed out it's probably inconvenient it's probably expensive to replace but it does get you would think it does get replaced relatively quickly and a lot of these [00:36:12] Speaker 1: negative effects would subside relatively fast i think a one-off shock might look that way right you know a yield loss owing to a heat wave or drought in one part of the upper midwest says nothing about major grain producing regions in russia or cotton prices given growth in the fergana valley in uzbekistan right like these these things are are disparate enough in space that maybe they're not uh experiencing simultaneous shock what i think is kind of interesting about el nino and why el nino is an interesting analog for the impacts of wider global warming which is a you kind of a universally experienced you know there's a high degree of correlation um in the impacts of global warming right it's not simply a heat wave here it's a heat wave here here here and here all at the same time all experiencing loss at the same time which is one of the reasons why i think it's this simultaneity of disparate regions experiencing hazards and experiencing losses that then can lead to to these um you know up aggregating market forces that systematically depress growth and alter where capital investment can go you know the efficiency with which you're able to take labor and capital and and mobilize it into productive parts of the economy it is hard uh to do that when there are lots of hazards being experienced around the world right so it's not one thing disrupting your supply chain yeah it's maybe a dozen things the 97 98 el nino was about 36 billion in economic losses asked as estimated at the time as a level effect a shock and then a recovery what we actually found is that there was empirical evidence for a systematic depression of economic growth and over a five-year period that amounts to about 5.7 trillion dollars in global economic losses instead by the year 2003 right five years after the 97 98 event what's really fascinating is we played this game where we use what's called a perfect model framework where we simply ask the question if we know that the impact of el nino is permanent um and of this magnitude given the observational data we have could we prove that it's permanent um you know so these kinds of statistics rely on the ability to pull a signal from noise which requires a certain amount of data and that amount of data affects the claims you can make robustly and so what we were trying to do there is say even if we had a known effect can we recover that known effect perfectly given our data record and what we found is that we couldn't which means that we cannot rule out the possibility that the economic impacts of el nino at the global scale are permanent meaning that every time an el nino occurs we're just knocked off our growth trajectory that would have been yeah absent that event which is kind of remarkable which implies that these costs kind of accrue ad infinitum in time in our paper we only went out to five and eight and ten years of growth accounting but it just says that the economic losses associated with these events are much higher than we understood before which means you know it's just another kind of piece of evidence that shows that weather and climate matter for our economy and our climate system is endogenous to that economic growth right it is greenhouse gas emissions induced by economic growth that is altering our trajectory of climate yeah and our climate also has a finger on the scale of what constitutes our economy and so that endogeneity i think is something really important to understand and represent um in particularly around policies related to climate variability and climate change that was a [00:40:26] Speaker 3: fantastic answer when you think about the impacts of this el nino or the el nino in the year 2050 or the el nino in the year 2100 how much you know so obviously like one uh you know presumably take a country like peru it's probably going to be richer in the year 2050 so it'll probably have more economic activity it may also have better infrastructure in the year 2050 such that road washouts are less likely um countries after you know you experience i presume even since uh 82 or 97 there's been you know hardening hardening against weather related uh shocks and so forth as countries learn about these patterns how do you model that aspect both that aspect both the fact that okay there is more economic activity in the sort of uh temporally connected areas and also presumably governments and companies get better about anticipating the el ninos and you know strengthening their infrastructure such that they're not as vulnerable [00:41:39] Speaker 1: it turns out it's really hard right the the you know projections uh of future are not falsifiable scientifically um and and and so i hope perused wealthier in 2050 and 2100 but yeah i mean yeah we didn't even talk about the fisheries implications yeah i think um you know one thing that's really i think what you're you're kind of raising is this this question of adaptation yeah um and and resiliency and and the way i i've been thinking about that is evolving right there there's a story we can tell that says okay you know if if there are you know market inefficiencies or we're shown to be vulnerable societally in these particular ways right we're going to we're going we're going to develop strategies to ensure resilience right those peruvian potato farmers noticed they were losing crops you know and they altered their sowing dates based on the visibility of the plei 80s overhead like that is an adaptation and you can imagine you know autonomous adaptations meaning individual level adaptations you can also imagine society-wide or planned adaptations such as shoring up kind of resilience and these things are linked to el nino forecasts right like one of the things you're seeing uh you know you said at the outset tracy of of of all these financial uh planners thinking about the implications of el nino for markets around the world and supply chains that is a form of kind of resilience strategy um you know building that that's occurring that is in indelibly linked to the science of enso forecasting and why el nino forecasts are so important right the 19 uh you know 82 83 el nino we didn't have the benefit of a tremendous amount of lead time and planning for its impacts right a tropical cyclone off in the ocean right being able to track it and have sufficient time to get resources injected into the places you need and people out of harm's way right like that requires lead time which is why these types of forecasts are so essential um because they allow us to to pursue resilient strategies that we otherwise wouldn't have been able to but this wider question of like are we well adapted to climate variability we have and shouldn't we expect that we'll be even better adapted to climate variability in the future i would say the evidence is pretty mixed there isn't a whole lot of evidence that we're well adapted to the climate we have we are finding documenting observationally that the costs of el nino are astounding despite the fact that we've been exposed to it for hundreds and hundreds of years right this is not something that is that is novel right while the coming el nino is is like seems to be out of sample and the costs associated with it are really hard to anticipate because it is so out of sample right like when i think about the cost of this forthcoming el nino you know i estimate somewhere around 10 trillion dollars of global economic losses conservatively over the next five years right up to 14 trillion dollars of economic losses if you take some of these tail but i could only 70 of that only you know about 10 trillion dollars of that is in sample meaning supported by existing data we have on el nino's and economic responses uh our global economy is larger and so these these losses are associated not just with the magnitude of the el nino but the size of our global economy there's just more stuff to be impacted right but what that tells me is that we are not optimally adapted to the climate we have and so there isn't much evidence that we have that we are um taking on board the feedback we're getting from the climate system um about how things are going and and pursuing strategies that really offset uh or mitigate losses um in a coherent way i think that's uh you know pretty pretty telling just given the documentary exercise that's not to say that we couldn't be better adapted i just think that we are not well adapted and there isn't widespread evidence of adaptation to either el nino as a form of climate variability or wider global warming since the impacts of global warming look a lot like the impacts of el nino and so i think there's like a bunch to kind of disentangle there but when we're thinking about these these future growth uh impacts on ends from enso we need a counterfactual world of what economic growth trajectory would have been absent el nino but given global warming um and that is a picture of future that i think is really hard to envision um but there are economic tools that people use to create those those those counterfactual worlds and then you can measure deviations as could be the case given what we know to be observationally true about the relationship between el nino and an economy [00:46:53] Speaker 2: yeah i know you just said there aren't a lot of examples of successful adaptations for el nino other than perhaps peruvian potato farmers sowing their crops early but give us give us some hope here or some idea of what a successful adaptation could look like in an ideal scenario where this is an endogenous phenomenon as you just described where you have it basically happening all around the world simultaneously what could we do to try to offset some of the economic impact there's improving resiliency [00:47:24] Speaker 1: in global markets such that these weather and climate shocks don't alter prices and interest rates and all these things that are huge determinants of human well-being then there's also like the direct hazards themselves, right? A stronger than average atmospheric river event that triggers a landslide along Route 1 in California, creating a road washout that is a massive disruption to communities there. A wildfire, you know, that rages through, you know, the bush in Australia, killing trillions of animals and raising insurance premiums on kind of wildfire protection for homes there, right? There are all these kind of ways you can process trace how El Nino hurts an individual or a community. And so I think that protection of human well-being from approximate sense of like, are you okay, given these hazards, that to me is a different question than how do you shore up international markets such that they are resilient to the impacts of El Nino. And I think just separating them is really, really valuable. I think the fact that people are aware of El Nino as a phenomenon and that it can induce these cascading shocks through agriculture or through mining or through the transport sector, and that all those things all around the world can up-aggregate to a macroeconomic toll, I think is important to know. And I don't know enough about how people think about these portfolios, but I know that risk diversification is a strategy to ensure returns. And I imagine that's something that people are embracing given modern portfolio theory or whatever. But I think this question of like, how do you protect people from hazard, that is a different question and requires different sets of warnings about the local manifestation of the hazards associated with El Nino. And to me, that means early warning systems, that means emergency management and response teams that are well placed and understand these consequences. That means that we have insurance markets that recognize the hazards, the unique geophysical hazards associated with something like El Nino and its regional dependencies and what it should imply for the types of products those people need to protect the most capital intensive asset they have, which is their home. I think it's many different kinds of things. And so none of this is out of the realm of possibility. There's a lot of these things we do anyway, right? We build reservoirs to mitigate drought risk already. Nevermind the fact that Lake Powell, the central bank of the American West water economy is at a near record low, right? He knows his audience. I guess the point is, is that, you know, I think there are lots of, there's lots of evidence that we, these adaptations are within the realm of possibility, but the constellation and configuration they take is kind of regionally dependent. And so I think kind of a coherent global strategy about managing the impacts of El Nino El Nino, you know, is an interesting test bed for adaptation strategies that would lend itself to wider global warming. And that's something that I'm, I'm really intrigued by. [00:50:56] Speaker 3: So you talk about this idea that what you've discovered through your research is that El Ninos are not these sort of cycles that you bounce back from, and then you resume to trend growth, right? That they change their trajectory. But if El Ninos are something that are very consistent and predictable throughout history, such that we've been, you know, there's evidence of farmers having tried to anticipate them going back over 500 years. Why should we not assume that, you know, it's seasonal, like I, you could say, for example, that cold winter months depress economic growth, which they probably do. People don't go out as much and shop and travel as much in the middle of December, but we don't really say, and so we could say, oh, if there were no December's, if it never got cold, then in the counterfactual, you know, we would have had much higher growth, which is true, but also absurd because the premise of not having a winter is absurd. If El Nino is something that will always and has always been with us, why do we even imagine there being a counterfactual trend rate of growth [00:52:14] Speaker 1: in a world without El Ninos? You know, that comes back to the kind of empirical identification strategy. Okay. Right? Like, we don't have the benefit of re-simulating the world as it has been and kind of the emergence of global society and the global economy absent El Nino. And so we rely on models to create those counterfactuals. We know that El Nino, while being cyclic in nature, it is not perfectly so. It is as good as a random treatment. And so we can separate the experience of the world into those associated with El Nino and those not. And we can examine the differences in economic productivity under those two regimes. So that's essentially what we're doing empirically. It's akin to a medical control trial, right? Where you randomize a population into a treatment group that gets a drug and a control group that doesn't, right? Or receives a placebo. And you compare medical outcomes, right? There are a year, there are people that may look the same in the treatment and control group. And there are people that may look different in the treatment and control group. But if you randomized effectively, those differences among individuals cancel out. And the kind of efficacy of the drug itself is just the difference between those two groups. El Nino is as good as randomly assigned, despite the fact that it is cyclic in nature. And so you can treat it as this natural experiment, this treatment, and compare it to the control of non-El Nino years. And that's effectively what we're doing here when we're kind of talking about this counterfactual, right? We know what economic productivity looks like. And so you're right to say that, you know, economic productivity looks different in winter than in summer or in spring, right? And it looks different in the face of a tropical cyclone than it does in the face of, you know, a civil conflict or something else, right? Like these things have impacts. But what we're doing here is kind of exploiting natural stochastic variation in it, despite its cyclic nature. And what I think is remarkable, and what I'm kind of suggesting here is that despite the fact that El Nino is not a novel phenomenon, right? It is not something that we have never experienced before, which global warming is, right? Global warming is something humanity has never experienced, and it is out of sample. And it is arguably making this El Nino that's unfolding, given the background warming, truly out of sample, a truly out of sample event, right? Something humanity has never experienced before. And I'm saying, despite the fact that we should be inured to experiencing El Nino, we have limited evidence of [00:55:09] Speaker 2: widespread adaptation and resilience to it. All right. Justin Mankin, thank you so much for coming on Odd Thoughts. Yeah, thanks for having me. Joe, I found that conversation fascinating. Love talking about the weather in all ways. But I think the point about this idea that like, okay, El Nino is something that we have experienced, it does come with some regularity, and yet there seems to be some proof that it has this lasting impact, not just on like the immediate economy, but on economic growth itself is a very important one when you consider that like, again, we're about to experience a potentially record-breaking El Nino and with the wider backdrop of climate change, right? [00:55:48] Speaker 3: It's super interesting. I learned so much about that, so much in that conversation that I didn't know this term that I just heard throughout my life that I really, oh, it gets warmer or something like that. I didn't really know what it was or why it was or what direction. I thought it was really interesting. Okay, what makes an El Nino different from another random weather event? The global simultaneity of it as being key, like, okay, other events you just bounce back from. I mean, everything must have some effect. You never like bounce back from anything that perfectly back to trend. But the simultaneity, of course, we saw that during COVID and the ripple effects from that. And so when you have this wet weather pattern that is truly global in nature and sort of very, I guess, predictable in the sense that you know what regions they're all going to get hit, it sort of makes sense to think about it in these terms of like, oh, there are real long-term costs, especially if they're intensifying during a trend of global warming. [00:56:50] Speaker 2: Yeah. Nowhere to run, nowhere to hide kind of idea. The other thing I was thinking, of course, back to some of our early agricultural episodes from this year, all of this is happening against a backdrop of higher fertilizer prices and lower supply. Like, you still have, I think, 30%, 40%. I can't remember the exact numbers of global fertilizer supply stuck in the Strait of Hormuz. And that can't be helpful either when you're talking about the biggest El Nino on record. [00:57:18] Speaker 3: Yeah. No, it's going to be a, well, now we are well-armed. Maybe we're not well-adapted for the El Nino, but at least we are at least well-armed to understand news stories that will be emerging over the weeks and months. And the sell-side analyst notes. And the sell-side analyst notes in the months ahead. [00:57:36] Speaker 2: All right. Shall we leave it there? [00:57:37] Speaker 3: Let's leave it there. [00:57:38] Speaker 2: This has been another episode of the Odd Thoughts Podcast. I'm Traci Allaway. You can [00:57:42] Speaker 3: follow me at Traci Allaway. And I'm Joe Weisenthal. You can follow me at The Stalwart. Follow our producers, Carmen Rodriguez at Carmen Armand, Dashiell Bennett at Dashbot, Kale Brooks at Kale Brooks, and Kevin Lozano at Kevin Lloyd Lozano. And for more Odd Thoughts content, [00:57:57] Speaker 2: you should check out our daily newsletter. You can find that at bloomberg.com/oddlots. [00:58:02] Speaker 3: And you can chat about all of these topics 24/7 in our discord, discord.gg/oddlots. [00:58:09] Speaker 2: And if you enjoyed this conversation, then please leave a comment or like the video, or better yet, subscribe. [00:58:14] Speaker 3: Thanks for watchin'.