← The Sustainability Story
The Sustainability Story
Livio Stracca, PhD: Pioneering Short-Term Climate Scenarios for Financial Decision-Making
00:31:55
Loading summary
Transcript29 lines
- [00:00]
Dr. Livio Straka
Foreign.
- [00:05]
Deborah Kidd
Welcome to the Sustainability Story, a podcast where we explore the intersection of all things sustainable with the investment industry and the global financial system. My name is Deborah Kidd and I'm your host for today's episode. My guest today is Dr. Livio Straka. Dr. Straka joins us to talk about the launch of a unique drug tool in climate risk management, the first publicly available short term climate risk scenarios. Released in May by the Network for Greening the Financial System, this new tool helps regulators and financial firms understand how climate policies, extreme weather events and economic shocks could disrupt financial systems not decades from now, but within business and policy planning horizons. Dr. Straka is the Deputy Director General for Financial Stability at the European Central Bank. His career at the ECB spans more than two decades. He has published extensively on macroeconomics, international economics and international finance. Dr. Straka holds a PhD in economics from the University of Leicester, a law degree from King's College London, and he holds a degree in statistics. Dr. Straka also chairs the Network for Greening the Financial System Work stream on scenario design and analysis and in this role he leads global efforts to develop climate scenarios for regulators, financial institutions and supervisors. Today we're going to talk about Dr. Straka's work with NGFS and and for our listeners who may not be familiar with the organization. NGFS is a global coalition of central banks and financial supervisors who develop best practices and tools to strengthen the financial system's response to climate and environmental risks. So welcome Livio. I'm delighted that you are able to join us today.
- [02:16]
Dr. Livio Straka
Thanks for inviting me. I'm glad to be here.
- [02:19]
Deborah Kidd
Before we talk about the short term climate scenarios, I want to ask I'd love to hear about your sustainability story and what led you to your work at ngfs.
- [02:31]
Dr. Livio Straka
That's a good question. So I would say I stumbled on sustainability in the sense that I was working on most other issues and when I move department a couple of years ago within the ecb, I was asked at the time DCB had been asked to chair the scenario workstream of the ngfs. Essentially the choice fell on me. I guess partly was also due to the fact that I have a background in macro modeling as well and that is important for climate as well. But I would say most of it was by chance. And to be honest also I not really worked on climate and sustainability before. I mean I was certainly curious about the topic but I not seriously worked on it. So I had to catch up a lot at the beginning. I had to learn a lot of new Vocabulary, new concepts. So it was a steep learning curve at the beginning, but I think now, after I would say three years working on Clement, I feel I'm really into it now. That is a bit my journey. So I arrived a bit by chance, but then I really invested on it and now I enjoy and attach a lot of importance to this straddle work.
- [03:50]
Deborah Kidd
Yeah, you mentioned your work in macro modeling. And for those who haven't seen the short term scenarios, the modeling and the development of those have been, it's quite intricate. Those are very complex models. And I want to talk a little bit more about that later. But right now I wanted to dive into kind of what led to the development of the short term scenarios. I know NGFS has developed long term climate risk scenarios and those have been around for several years and they're periodically updated that the short term scenarios are new. And you've said that nobody else in the world has a tool like this. So can you talk a little bit about that and why NGFS decided to launch these short term scenarios?
- [04:46]
Dr. Livio Straka
Yeah, so I mean, I agree with what you said essentially. So the traditionally, not only the ngfs but also other entities, for example the IPCC had developed long term scenarios. But I think that over time it became clear that in our world, which is the economic and financial world, decisions are taking place on a much shorter horizon. And that's also, by the way, the world of politics. So all the decisions that are important for our authorities, because the NGFS essentially is a club of supervisory regulatory authorities and central banks. So these are entities that take decisions within a horizon of five years or a few years, I would say. So it became quite natural to want to shorten the horizon of these scenarios. But partly also this was also an outside request. So at some point we did a survey of users in. So asking the question is what do you like to see in the NGFS scenarios? And I think the second most mentioned kind of need was the short term scenarios. So I decayed both from within but also from outside pressure.
- [06:15]
Deborah Kidd
Well, let's talk about the scenarios themselves. There are four that tend to has been released under this framework. And could you tell us a little bit about each of those four?
- [06:26]
Dr. Livio Straka
Sure. So first of all, so the process was organized in two separate steps. So we first had a bit of thinking of what are kind of plausible but scenarios that we could have. And so the NGFS released a conceptual note. And in this conceptual note we had actually five scenarios. There was also a process of feedback, consultation. But then after these scenarios were broadly seen as meaningful. We then launched a procurement for AFL modeling of those. A consortium of modelers want this procurement. And then we work together to produce these short term scenarios. And we have released four of them, so the fifth one might be released later. So we are a bit behind on the fifth one, but the four have been published. So two of them are related to transition risk and two of them are on the physical risk side. And here the idea is, of course you could envisions all kinds of different scenarios. So there is an infinite amount of scenarios you can think about. So our point was, let's do something that it looks like meaningful and has a clear narrative and so it can be taken as useful starting point. So something that is rather obvious to do so in the error transition. The two transition scenarios essentially assume that countries, rather than following the currently planned climate policies, they rather converge to the net zero path. So let's say we said, okay, assume that countries rather falling short of the party's targets, they actually implement policies that go there. And that if you think about this is an obvious, is an obvious scenario to think about. So the two transition scenarios differ by the way this is done. So in the first one, which we call highway to Paris, this is done in an orderly way and also efficient, while in the second one we assume that so the process is still convergence towards the net zero path, but is more disorderly, happens abruptly in the middle of the horizon, and also it is done less efficiently. So the carbon task that implements this transition is more distortionary in the second scenario. So it's both more abrupt and lower quality transition. And so this is what we call the sudden wake up call, because there is this element of surprise in the middle of the horizon. And then on the physical risk side, we mostly took the historical distribution of physical risk damages for each region and we just took a tail of that distribution. So in particular we said, okay, assume that we have a one in 50 years manifestation of the distribution. So something that has more or less a 2% probability of materializing and let's see based on the historical data, what kind of effect this could have on the economy. So these are, the two fiscal risk scenarios are related to that. And the difference between these two is that one is a purely physical risk. The second one is also an interaction between transition physical risk in the sense that we assume that climate events, very negative, very bad climate events also put kind of or create a friction or if you want or say, make it more difficult for emerging economies in particular to export raw materials and the raw materials that are critical for the green transition are used by advanced economies. So in this way, basically climate events in emerging economies made the transition more costly in advanced economies. That is the fourth scenario, which is called the Emerging Realities.
- [10:44]
Deborah Kidd
So the scenario results are quite specific to geography and macroeconomic indicators and industry. And I want to help our listeners understand what types of outcomes or results they will see from using these scenarios. For example, know that the disaster scenario shows up to a 12.5% in GDP losses in Africa, with the effects not just contained in Africa, but rippling through the global economy. What were some of the other notable results that you saw in the scenarios?
- [11:27]
Dr. Livio Straka
No thanks. So I would say the physical risk scenarios all have rather large effects. You measured the 12% for Africa, but you also have a 5% for Europe. And all other main regions have similarly large effects. I mean, Asia is even larger. I think it's 7% if I remember correctly. So in general, the physical risk side at least I was surprised on the upside. So the effects are large and probably larger than what most people expect is not only, I guess, the intensity, that depends on geography, also the nature. So some events, we call it wet events, think of floods, so these are hit more some geographies, for example East Asia, then you have dry events like droughts or heat waves, heat more Africa and so on. So it's about the quantity and the quality of the effects that is different and geography dependent. If you think about the impact on inflation, which is also variable, we look at because many of us are central banks, so we naturally look at inflation interest rates. So we find that the impact on inflation is rather muted for the physical risk scenarios, but it's actually larger for the transition risk. So the transition risk scenario is, I mean, you can think of those as abrupt shifts in carbon prices affecting the cost of energy in the short term and thereby inflation. So you have larger inflation spikes following the transition scenarios and this also prompt some reaction from central banks and so you have a rise in interest rates and so on. So what users can find in these short term scenarios, which by the way are very easily accessible, okay, even a person with low IT skills like myself could easily download the data. So what user will find is essentially a set of macro variables for their jurisdiction, again easily downloadable, plus also a number of financial variables. So the scenario also contain variables such as probizo default equity prices, interest rates, risk premia. So there is also a financial module that creates these financial variables. So a user can use can look at these financially real variables. And important to note, for many cities we also have sectoral variables. So you can look at new agriculture, or you can look at, say the energy production, divide it into coal or renewables, and so on. So there is quite a degree of sectoral degradation.
- [14:24]
Deborah Kidd
So there is a base case scenario for each country and region. And then a user would run the scenarios relative to that base case scenario and see how things change over the next five years depending on which scenario that they've run. And so is that how it works?
- [14:48]
Dr. Livio Straka
Yes. So we are often asked also about the probability of various scenarios happening and so on. And I guess the best way to think about this is to say that the baseline is probably the best possible estimate of current and future policies. I mean, unfortunately it's not updated because it takes time to produce this scenario. So this part has been updated to the end of 2023, so it's a little bit outdated now. But this is what we have. So the users can think of the baseline as the name says, the base case. So the most likely path, if you want to climb a policy, or at least the path that countries are committed to, and that is kind of the benchmark, and then you can tickle the short term scenarios, there are kind of perturbations around this most likely path that has some kind of stress or tail nature. So these are not things that are most likely to happen, but it's more like a tail event that, that is useful to think about for in case you want to run a stress test, for example. So you want to see, okay, on the physical side is approximately 2% probability event. How well am I prepared for this? So this is what the short term scenarios also tell you. But some users, we want to actually to use a baseline because the baseline is again the base scenario. So it will depend on the use of these scenarios by different users.
- [16:36]
Deborah Kidd
And so how might a central bank use these scenarios to adjust its monetary or fiscal policy?
- [16:48]
Dr. Livio Straka
Well, monetary, fiscal policy are all part of the scenarios. So they are all endogenous.
- [16:55]
Deborah Kidd
Okay.
- [16:55]
Dr. Livio Straka
So as I said before, for example, in the transition risk, you have inflation going up and you also have central banks reacting to that and increase interest rates. So it's all part of the model of the transmission channel. Yeah, so this is probably my first kind of pass on this.
- [17:19]
Deborah Kidd
And for financial institutions like investment firms or maybe insurance companies, would a use of this tool be something like an investor is perhaps looking to buy real estate, looking globally? And could they use these short term scenarios, physical risk scenarios, possibly to evaluate the kinds of geographic risk risk that might affect properties that they are considering.
- [17:54]
Dr. Livio Straka
So let me say first that with the release of these scenarios, we aim to have some rigorous approach that puts some discipline in these kinds of exercises. But it's not going to be for most of these exercises, it's not going to be usable one by one. So some adaptations will be needed. And so you mentioned, for example, residential or residential or real estate, real estate prices, for example. This part, for example, is not very well developed in this scenario. So one would need an additional step complement some variables. Yeah, so. So not all variables that are conceivable are released in the short term scenarios. And also users may need to adapt to different baseline, for example. So as I said, the baseline is for many countries rather outdated, but this also can be adjusted. So users can also shift the baseline and can consider these scenarios as deviations from the baseline. So deviation from a different baseline. So if GDP say is 5% below the baseline, but the baseline is different, then the users can compute the GDP, which is the new baseline, minus 5%. So just to make a really concrete example, so it's a tool that is quite adaptable and we would expect the users made some adjustments, but hopefully the short term scenarios go kind of 80, 90% of the way towards having a coherent scenario for each particular exercise. So it's something that is adaptable, but the core hopefully makes sense as it should be preserved.
- [19:53]
Deborah Kidd
Thank you, that's helpful. So on your website you have the data that users can download and then they can make the adjustments to that data and run a revised scenario.
- [20:07]
Dr. Livio Straka
So running a revised scenario entirely is not going to be easy because it's not going to be possible without having the whole modeling suite that we have. So running the short term scenario takes months of work and it also includes, by the way, a very strong review process. So we ask members to check the results for their jurisdictions. So everything has been totally checked by each member. DJ Fest is a member led organization, so we have. So the climate scenario Warstream has 200 members and members are in the first line to check the results. So this is just to say that producing new scenarios globally is a very heavy exercise. So the users are not going to be able to replicate the whole modeling suite and hold the results. But what they can do is to make some basic adjustments. So just to make a concrete example, so suppose that you are a financial institution, you are interested in a particular sector that is not covered by the short term scenarios. You might want, for example, to take the largest sector that is covered in this scenarios or even kind of gp, and try to map so to make an auxiliary regression between your target sector and GDP or your target sector and the sector that is including scenarios. And see given what this is in the shortest scenarios, what the sector you are particularly interested in would do under this scenario. So I hope it's clear. So suppose that for example, you are in the steel industry and suppose the steel is not covered by the short term scenarios, but you know, the steel is say more volatile in the GDP, that this means that if GDP say goes up by 5% or down, say was it the 5% then the steel industry will go down by more than 5% and so on. So we would expect users to run some kind of auxiliary analysis, but keeping the overall structure of the short term scenario unchanged because the user will not be able to run the scenarios in totality. Is that clear?
- [22:30]
Deborah Kidd
Yes, and thank you for clarifying that. The work that's gone into the scenarios is really quite amazing. It's complex and the, the inputs are extensive. But I do recall your point that anyone, even with limited technical skills can run and use these scenarios. I know your long term scenarios are updated periodically and I wondered how you envision seeing this tool evolving over time.
- [23:05]
Dr. Livio Straka
Thanks for the question. So the long term scenarios, indeed we plan to update them every three years. We are considering updating them now for next year also because this year there's going to be an important update of the country's climate policies. The so called national determined contributions in the party's process. I mean you may be more familiar than I am on this. So we had this planned update and we strive continuously to improve them and we receive a lot of feedback from users also on the long term scenarios. We know that many users are worried about tipping points or non linearities that we are kind of maybe underestimating some risks. We have an excellent team of climate scientists working with us on these long term scenarios, so I'm quite confident that they are reflecting the current science. But of course science evolves and we need to be up to date. But just to also be clear with the users who sometimes criticize certain choices, sometimes it's not that we don't want to change models or scenarios, but sometimes the science is still developing, is the science itself that is moving. And science, good science is always slow, so cannot rush science. So for example, in the area of tipping points, this is very interesting, we are monitoring this. I mean there is a big conference going on I think right now on tipping points. I sent a staff member to attend. So we are clearly interested in this. But it Seems to me that the science is not yet firm. We should not crush it.
- [24:56]
Deborah Kidd
Livia, is there a scenario that you think applies or characterizes the current global response to climate change?
- [25:04]
Dr. Livio Straka
Or so is your question about the most likely scenario, for example. So what I think is most likely to happen, yes, probably is a bit what they also said before that. So both in the short term scenarios and in the long term scenarios we have this assumption of NDCs. So this is a national determined contributions, which is what countries pledge to do in the context of the Paris process. And this is probably the baseline and of course in the current values. So these planned climate policies are very far from guaranteeing an S0 transition. So we are quite far. So we have to assume if things don't change substantially, that we are deviating an important way from the net zero path. So that's unfortunately where we are now probably. The other bit that I find also kind of believable in particular in the long term scenarios is what we call the fragmented world. So we have a. So one of the seven long term scenarios is one where certain areas do the transition and others not. And so you have this kind of deviation, the discrepancy between different areas and that is also a plausible outcome. Last thing I wanted to mention is we have worked a lot also with our consortium to refine the estimates of the economic damage. And that is a very fast evolving science. So like 10 years ago, so we had relatively small numbers in mind, but now it seems that science is converging towards larger numbers. So we have updated those numbers. This is what is called the damage function. And so we have updated those numbers. I mean we have sometimes been criticizing for doing it too quickly, but again, other users or other observers think that we are doing too little there. So we have received obviously feedback on both sides. But again, so we try to keep up with the science. And this is also. I also wanted to stress this a bit the DNA of our institution. So we are central banks mostly. So we are really used to be very analytical, cautious. So you will not see us jumping on a new theory easily. So we want to test things, be sure that the sciences sound is a bit how we operate. So I know that sometimes it's frustrating because of course climate activists or other activists have set the stories in mind. But also users should understand that we should not jump on any kind of fancy theory, but we need to be very grounded as much as possible in science.
- [28:18]
Deborah Kidd
It's a hard balancing act to strike and there's a lot of information on your Website both about your work and your approaches to climate risk assessment and the long term and short term scenarios. There's a lot of information that we didn't get to discuss today. Is there anything in particular that you want to point our listeners to on the website which is ngfs.net I guess.
- [28:48]
Dr. Livio Straka
It depends on the purpose of listeners. So I mean for those who want to know more in general about the short term scenarios, there is, there is a lot of documents online and even the slides that we publish, the NGFS published with the release are very informative. I think by now I'm sure there are also online videos presenting those. So for those who want a general, so becoming acquainted with this work at relatively little cost, you can kind of derive probably useful information. What you probably for us is particularly interesting is those who actually use the scenarios. So these may be financial institutions, maybe authorities. So these, you know, so we are particularly interested in that experience and we try to collect feedback and there is also on the Yaza portal, so there is a link somewhere in our website. We also point to Q and A possibilities. So users struggle with something. They can email questions and we have somebody in the team of modders who are responding to these questions. So we are always kind of interested in particular in that. And this I should have mentioned from the beginning this was the first version. It really should be seen as a prototype. So we hope to learn from users where it can be improved. So we know that there are already, we know already there are some aspects that could be improved. But for the next release we hope to make significant and positive adjustments.
- [30:38]
Deborah Kidd
Great. So that's important to know. You're really looking for feedback from the users of the scenarios to help improve the process and the scenario outcomes over the long term or as you continue to evolve the short term scenarios. Well, Livio, thank you so much for joining us today. It's been really a pleasure to have you as my guest and I'm excited for the information and the work that you've done with the short term scenarios for that to kind of become part of the decision making for financial institutions and central banks.
- [31:16]
Dr. Livio Straka
No thanks. Thanks so much for, for inviting us. I mean giving us this forum is important for us and you know, the more we can spread the word, the more feedback we, we, we elicit, the better. So it's also very useful for us and you know, also test for your questions every day. We always, I mean some of the questions are, are, are the same and, and whatever you say, but some are new. So today I received a couple of questions that I not received before. So already this is useful for us to absorb. So thanks so much for having us. It was a real pleasure.