 Thank you very much. It's a pleasure to be back in Helsinki So I'm gonna go and talk about southern Africa uniquely and this builds on some of the work that Johannes already Presented so we're gonna focus on countries that are part of the Zambezi River Valley So we've modeled the whole the Zambezi River Valley and in terms of runoff and so forth and that Johannes already showed and then we're gonna look economically at Zambia, Malawi and and Mozambique and And this builds on as Ken was saying a series of efforts that we did At wider that were targeted at a bunch of different questions The questions that we were looking at originally are things like what are the implications of climate change for growth and development prospects to about 2050 in terms of these these three countries. What are the impact channels? What options are available to try to adapt or reduce impacts and that was all Published in a special issue and there've been other ongoing work in the region this presentation We want to focus on just just for one question and one issue Which is what are the implications of effective global mitigation for the case of the economies of these economies by 2050? And and Johannes gave you a preview of Some of those results, so we're gonna look at that And so I'm gonna leap forward to to the conclusions. Tell you what I what I decide and then we're gonna Sort of Provide some more detail on those conclusions because now there's more papers coming out This was really one of the first papers to really try to address this issue of what are the relatively short-run? impacts of reducing emissions by 2050 So the we find that the effective global mitigation generates two sources of benefit even by 2050 First we get less distorted climate outcomes And they're generally more favorable and there's less variability, which is consistent with what Johannes showed the second is that successful global mitigation policies reduce fossil fuel prices Especially oil the producer prices and as a consequence structural fuel importers such as countries in Zambezi River Valley get a term to trade benefit and it's quite significant So we're gonna use this multi-sector modeling framework that that Ken was missing mentioning This is the the sacred framework at the top. We have global change This is driven by a sort of a combined climate and economic model It's gonna produce for us precipitation Temperature it also produces global product prices. So it's gonna produce for us the oil price going to produce for us agricultural Prices on on a global basis and we then we spread that into particular regions. And so we get Rainfall turning into runoff and stream flow this might have implications for for flooding in in Mozambique you and other places you have sea level rise which is Interacting as in the case of Florence with with cyclones and and and storm surges Over here. This is what Johannes focused on. What does it do for hydropower? What does this do for for agriculture in terms of crop yields and production? What are implications for infrastructure? So it's quite a detailed look at at what climate change might mean or do and then we take it and we sum it all up And and get the interactions within a full economy model of each of the countries in focus So that we're able to you know really get a notion of what's of what's happening We're using the uncertainty approach developed at at MIT and these are we're going to 2050 These are temperature outcomes globally In 2100 and and this is the big this is the what we want to avoid Which is my no policy or what we're calling unconstrained Emissions and this is the distribution of potential sort of the range of outcomes that are expected This is more or less we've published a little while ago 2009 It is more or less what we still have roughly today, which is you know unacceptably large temperature Gains by the end of the of this century So you know an median or a mode of around five degrees centigrade, which is which is really quite quite a lot And and you note that this is what we call level one stabilization This is also what Johannes focused on. This is about 450 parts per million co2 Add in some other greenhouse gases like methane and you're getting to about 560 parts per million co2 equivalent in the atmosphere and this corresponds with a you know more favorable outcomes Some of them below kind of two degrees centigrade that's compared with 81 to 2,000 But but the point of the matter is that this distribution here is extremely distinct from That distribution there so you know obviously mitigation over this century matters really a lot for the what kind of climate we get at the end of the century and But it's it's also the case that even by about 2050 it matters materially So those distributions that that separation doesn't just occur right at the end of the century it occurs you know, it's they're starting to separate throughout the century and and so this is December this is temperature in the summer. So this would be January February March Temperatures and we can see that under unconstrained emissions you have Higher temperatures and a greater variation in the in the temperature outcomes at around 2050 If you go to precipitation you get the same result generally unconstrained emissions and level one stabilization And this is in the spring Yeah September October November Generally we're getting some drying in the in the spring But the drying under unconstrained emissions is is more pronounced and there's higher variants in the range of outcomes So this is sort of more uncertainty in what might actually happen. If we if we continue to put the same amount of Greenhouse gases into the into the atmosphere This is pretty consistent with what Johannes was talking about these are the runoff Anomalies or changes in levels of runoffs at various points in the in the In the Zambezi River Valley and and what you see is is you know distributions that that sort of vary more or less Around zero sometimes there's less runoff here. Sometimes there's a little bit more such as such as here up here in the Shirae generally we get Greater variants in the in the blue curves. So these are are again distribution functions So there's sort of greater likelihood of or a broader distribution of outcomes And and in all of these cases we have these tails Where a runoff goes up by by quite a bit and this is on an annual basis But you also have these kinds of tails of big increases in runoff at at shorter at shorter time steps So one of the things that we find and particularly in in Mozambique, which is sort of already being observed is big changes in in flood probabilities and floods being You know really potentially damaging as you know, North Carolina is finding out sort of right now So under level one stabilization we get an increase in in the sort of number of greater than 50 year floods That we would expect, you know in in any one year, but under Unconstrained emissions that that really increases quite a bit in Mozambique is a downstream Country, it's you know the farthest to the east and so it's picking up, you know a lot of this variation It's also quite flat and and that's that's pretty damaging So, you know just to skip to you know one end result in in respecting the time These are this is the kind of implication that we get for GDP in 2050 and So this again, this is a density So this is some kind of measure of likelihood of the outcome This is a change in the deviation from some kind of mythical no climate change baseline Which is which is set here and what we find is that you know There is some probability of positive outcomes, right? It's it's possible that you know It could warm up, but but you still get more GDP than you would have otherwise But the great likelihood is that you get a reduction in GDP somewhere in the range of Zero to five percent and we focus merely on GDP as as a matter of time And being able to compare and then we have this kind of bad tail of outcomes and that's Corresponding to this big potential increase in in flood probabilities So if you're just getting floods, you know on a repeated basis that are washing out infrastructure your your growth prospects are Reduced pretty substantially. So this bad tail here of outcomes is the consequence of that increase in in flood probability if you Have successful global mitigation you end up Sort of reducing the very so you increase you shift the distribution to the right you get fewer impacts And this is this is purely All we're doing here is changing the the climate scenario We're not changing the economic scenario at all So this is purely a temperature and rainfall kind of kind of impact the climate impact And and so we shift the distribution off to the right and we tighten it dramatically as a consequence of Sort of a more certain Climate future so that that's one set of benefits the other set of benefits that we Get are these world price effects so The most significant one that emerges out of out of the modeling exercises is in is it really in the price of oil Which is the major traded? Energy commodity so you know as has been observed by many at this point And this is starting to show up in the literature as well So right now if we're going to limit global temperature rises to about two degrees centigrade above pre-industrial We can only burn about up to one-third say of proven reserves what we already have In the ground so that would mean that if we're going to do that then we have to leave two-thirds in the ground And this changes the economics of oil extraction, you know quite substantially You know as soon as Saudi Arabia can you know Considerors or believes that you know only one-third of proven reserves can actually be burned Then they have an incentive to get rid of their reserves quite quickly and sell them while while they while they can So Otherwise those reserves would be would be stranded so for this reason So this this can then comes in at the at the top this is the world product prices And this is coming out of the the epa model Developed in MIT, and this is what epa shows and there's you know a lot of reasons to be Worried about to talk about this, but this is the ratio of the price of oil under unconstrained under level one stabilization to unconstrained emissions and under unconstrained emissions basically the price of oil is rising More or less throughout the century So price is what's holding it back in under level one stabilization. You've either got policy or other factors Constraining the consumption of oil and you're getting a lot less consumption. And so you have a lower price other other products are moving around less in in in oil you tend to take a lot of impact in Price and not quite as much in quantity because there's a fair amount of profit involved in extracting oil So so I already be extracts at $5 a barrel and sells at $65 a barrel They're perfectly willing to keep producing at $35 a barrel In with with coal extraction your your price is much closer to your marginal cost So there isn't that much price effect. There's a lot of quantity effect. However If you're if you're limiting and with natural gas we get this funny Bend because natural gas generally emits less so you end up with Shifting to natural gas early on in the process and then you shift away from natural gas And then this is really a carbon capture and storage effect from from 2050, which is which is sort of assumed So we also assume in this so we impose these fossil fuel prices out to 2050 We also assume in this case and this is partly because we think this is true and partly because this is only so much We can cram into one already quite complex model That we have specially different differential treatment of these mostly low-income countries So the world successfully mitigates our case economies Muslim beak Malawi and Zambia Are exempted so we're not they're not mitigating directly within our models And we have just to keep things also a little bit simpler if you Follow Muslim beak you know that there's a major coal find a major natural gas find and we those are still in process of being Developed very hard to know what's actually going to happen. So we just abstract from those So here's what we get So once we add that back in and we get this big reduction in in global fuel prices that at import this imports of fuels in Muslim beak are about 12% of total imports fuels and derived products up to 20% of Total imports it's a it's a significant Import item and getting it cheaper Gives you a significant terms of trade benefit which Results and basically more investment more growth and it shows up pretty strongly by By 2050 We get the same effect The same sets of effects in Malawi. So in Malawi, you don't get quite as much There's not as much climate change impact overall You do get the tightening of the distribution you shift it to the right. This is the pure precipitation and temperature effect and then we get another jump in Malawi as a structural fuel importer being able to to import fuels Less expensive Lee It's the same story in Zambia with slightly different look Distributions aren't The variance isn't reduced quite as much there is variance reduction But it's not quite as apparent But again the same qualitative story where we got a slightly tighter distribution We shift it to the right and then we shift it to the right again with respect to the The fossil fuel price effect so Our conclusions are as I said effective global many mitigation generates two sources of benefit one more certain And and more favorable economic outcomes by 2050 Second we get terms of trades effects that that especially African low-income economies often Benefit from quite strongly. So I have in now. Yeah, that was the end of this presentation But there was a new paper that came out last month so just recently looking at the same issue and and they come up with They don't focus on GDP. They focus on food security But they they worry about an increased risk of food insecurity under stringent global time global climate change mitigation policy And this is driven by coming. This is coming straight out of their article Relatively mild climate change impacts by 2050 There's quite a bit of in these older mitigation scenarios quite a bit of bioenergy that comes in under a Stringent global climate change mitigation policy that bioenergy competes for land it competes for water and it drives up food prices And then in this particular case Not only is art is the carbon tax imposed on Emissions from energy sources like electricity generation it's imposed on agriculture as well, which is which is quite hard so you you have a Production type effect even in low-income economies So why is it that we get? Such there I think there are quite different. We would get increased We would get increased food security and and they're getting less and there's there's a number of reasons right off the top They have there they're imposing all of the mitigation even on low-income economies whereas we're not and that's that's one significant Difference, but there are some other differences, which I think are important and we need to Some more work. The first is you know, there's just not as much There's not a much detail in the impact modeling and so they tend to get a little bit less impact because they're looking at fewer channels They're not they're not picking up everything that's happening So as a consequence when you get the benefits of mitigation you have you have smaller impacts to begin with and And you don't get as much benefit from from the mitigation That that you do observe so that that's one effect the other Effect which is Something that that especially in the models used the global models used for the IPCC There are two Potential impact channels from food price increases The first is if your food price goes up and all you do is buy food then you are worse off Right, you know, this is basically me right food prices go up. I buy food That's bad for me It's not very bad for me because I don't you know my my share of consume But for urban poor consumers who are working in service sectors in the urban area a higher food price is Typically going to reduce their their food, you know, but ability to buy and increase food food food Insecurity in rural areas It's a it's a completely different story in that most economic activity is involved is selling food So, you know, if you're selling wheat and the wheat price goes up You're the price of bread might go up a little bit, but your income is likely to offset that And what we find in most literature the 2008 2009 food price crisis and the continuing high food prices is that because most People who are food insecure are in rural areas this income effect Tends to outweigh the the consumption effect and that as a consequence actually higher food prices improve food security because You get you're getting greater wages in agricultural production in rural in rural zones This shows up both theoretically and in the empirics that we do have and one of the problems with the models that Looked at this has a go at all and and others is a multi-model study. Is is that impact channel just doesn't even exist? There is no way that higher food prices could increase income and come back around and and Improve food food and food security just it just doesn't happen. Whereas in the in the framework that we're using it does The final thing that that happens is often Either the model just doesn't have fuel at all. It's an agricultural sector model So many of these in this paper would be looking at the agricultural sector uniquely There is no fuel price. It doesn't exist if it does exist it's in a general equilibrium model typically and And that model often you know because it's so broad in terms of products. It's it becomes More aggregate in terms of areas. So often they will use Sub-Saharan Africa as one aggregate region And if you think of Sub-Saharan Africa as one aggregate region inside a model Then it's basically the South African economy with oil, you know, that's it. It's South Africa Nigeria combined And and and so actually that aggregate region is a net oil exporter So it's completely different from the situation which most low-income countries find themselves in And where most poor people would be living Ethiopia, for example is you know, one of the most populous countries in Sub-Saharan Africa the most populous low-income country and it is a structural fuel importer But that effect would be would be completely disguised. So there's work to be done to to to Work through this this full set of issues, but but that's that's where I wanted to in the talk today. So thank you I'd like to open the floor for specific questions for Channing on his presentation Thank you. So my question picks up a little bit on on your own comments just there about the spatial and temporal desegregation so I'm interested specifically in your research about how the models How you translate Changes in in climate risk and weather risk into economic impact and whether you're imposing your different climate scenarios on essentially the same economic scenarios or is there any Essentially, is there any adaptation on the economic side as the scenarios change and then maybe as a broader point if the Effects are so heterogeneous heterogeneous across space We see kind of a longer-term modeling suggests so Esteban Russi handsburg has some models showing You know if if we allow movement across space we can reduce or minimize climate impacts That's in the very long term my own research in the short term shows it following flooding people don't tend to Relocate they go back To the same places what I'm just interested about that how the modeling and the empirics kind of come together and how the Short-run and the long-run come together and we we kind of bridge those divides Thanks very interesting actually what you mentioned about the modeling You know structure and you mentioned that the you have the ability to capture the temporal dimension for less than a year I just want to know how is this implemented if as far as I know that the the economic model is Kind of an annual Model so if it is a CG model then it is dependent on a on a Sam and that is based on an annual data I would like to see how do you go for less than a year Impact and my my other question to aunt is about the food price impact and you mentioned that yes higher food prices might Be positive for rural household who basically produce food and they sell it But how do you then go with the majority of rural small-scale households who basically produce food for all consumption? And they produce cash cash crops for for the sale and therefore the the the higher food prices might still be negative for them Thank you. We have one more question in the back Thank you. You're saloon from Brookings institution I was struck by the difference on the impact of climate change on food security specifically on the price effect the last commentator also mentioned the impact of food price on food security and Native porters I was wondering if if you Took the modern took into consideration that most rural Households in Africa are not net They are net importers of food So I would assume The impact on food security comes from more from the production shock than the price given that they are net importers of food Thank you. So Channing if you could address these three. Yeah, good questions So this is the framework that that we used, you know, it's not perfect, of course but it's not up there I Anyway, so we have the Sort of a global model. We're reasonably well doing well in terms of linking so the the there's a global CG model that's producing the emissions the emissions are going into a a Model of intermediate complexity for the for the climate which is then producing our climate outcomes So we're linked up there It's very detailed on on on energy, of course because partly used for for mitigation purposes It has agriculture in there, but it's fairly aggregate So we don't get that much in the way of differentials across, you know products. It's it's an it's an agriculture Aggregate right those outcomes are coming and I forget the exact time step, but but it's a it's a pretty close Time step and that that's being fit fit into models that that will have You know relatively short-term time steps The you know Johannes could talk a lot about the kind of river and stream flow Models that we use like, you know, the crop models actually will take inputs at shorter time steps Then then we're able to to produce right and they'll do it across You know across space So we get to handle with a fair amount of detail what's going on both spatially and and temporally right and we could handle for example You know Precipitations coming in in similar amounts, but later in the year right and that different different temperature levels So so that kind of stuff we can handle the flooding is inherently fairly short time step and that that you know There's work to be done to make that to make that better So You know in terms of that that aspect that's there the the model what we show here is there's a fair amount of internal or autonomous adaptation that's that's happening right so The we'll take crop model and we'll largely convert to some kind of a productivity measure But then the model decides okay, you know if if you're able to move production. We're regionally differentiated in production across activities So if the north is is well suited for you know better going to be better for growing maize then then the center As as the climate is evolving then then we're going to get not autonomous move into the north as opposed to to the center This is happening That gets to your second question, which is the movement across space In these models at this moment. It's it's not there's not a lot of friction, right? So We we do get a fair amount of movement And this showed up less here than in then in the South Africa work that we did where it shows up Really quite strongly now is it unrealistic You know, I'm not entirely sure I think the the the argument the counter argument is we are looking at fairly long periods of time here and Pretty young populations, right? So, you know Yeah, if you're 30 years old and you've been farming here for 18 years and then you know You're probably likely to be staying Where you are you have your family and your house and so on and so forth But but you're at the the the long end of the population distribution most of the population coming in is it's 15 I think 50% of the population below 15 in these in these countries And and you know You're gonna be getting married and figuring out where it is You're gonna live and if there's areas that are that are doing better than others What one of the things that we do find in in now is a fair amount of migration between rural areas as as as you know conditions change and this is I think partly due to these young Population dynamics, so so it's not over the time frames that we're talking about I think I think that's that's plausible because we're not we're not moving everybody. We're just moving at the At the margin Yeah, so I've done those. Okay. Yeah Yes, it's true most many many rural households if not most would be You know often net buyers or they're they're producing for their own account and they're not You know selling Or you know we were I spoke very much about food prices and they may be they're not really selling much food They're selling other cash crops off often to the often to the economy So if we go back first to the to the drivers that's being You know put pointed to for most of these food insecurity things It's greater competition for land and greater competition for water the fixed factors that are inside of Agriculture so those those those those competition factors are going to affect cash crops as well as food crops And you're going to see you know you get the same get the same effect whether you know how strong it is and how varied it is That that's different. You would have to model that but but you would get this kind of cash crop effect The if you're going to be positively affecting most rural households Then some things have to be happening other than just the first-order instance of the price rise, right? because you're you're you're a net buyer of Food and you might not be selling much cash crop and what what tends to be happening is You do get supply response as a as a consequence of higher agricultural prices and this kicks in we we do observe it You're generally Improving and then that supply response is converting into a labor market effect, right? So most of these households, I mean if they're Sitting out there and they need to buy food and they're not selling it They got to be selling something in order to to have the income in order to buy the food often They're selling their labor So if they're getting better wages then then they can if that wage is going up by more than food price They're going to they're going to be able to buy More food and that is you know again This isn't an area of research, you know where we know for sure That's what happened But the analyses that have been done when you look at in detail at say the big rise and food prices that occurred from 2003 to about 2013-14 generally you're getting positive Effects on on food security because you're stimulating the rural economy where most Poor people are actually living right and that that's that's what's going on