S&P Global Inc. (SPGI) Earnings Call Transcript & Summary

Hello, everyone, and welcome to today's webinar. My name is Prerna Divecha, and I'm the Global business lead for climate and ESG Credit Risk Solutions at S&P Global Market Intelligence. I'm pleased to moderate and speak at today's webinar, which is titled, Empowering Financial Resilience: A Deep Dive into Climate Risk Analytics. Let me quickly run through a few housekeeping items first. We want this to be an interactive session. So feel free to submit your questions via the Q&A button at the bottom of your screen anytime during the presentation. Also do note the widgets at the bottom of your screen, there is a related content widget in particular, which has valuable resources and thought leadership. The webinar will be close captioning in English. You may click on the CC icon on the media player to activate it. And for any follow-ups or to schedule a private meeting with our product specialists, you may also chat with us by clicking on the icon at the bottom of the screen. Also please note that our activities of S&P Global Market Intelligence are independent and separate to S&P Global Ratings, and S&P Global Ratings maintains a separate analytical and commercial set of activities. Now before we begin, I want to introduce my fellow speakers. We have Alban Pyanet, partner at Oliver Wyman, with whom we've collaborated on a joint climate risk offering called Climate Credit Analytics; and we also have James McMahon, our CEO of The Climate Service, an S&P Global company. Now let's begin. As we anticipate Climate Week and COP28, I wanted to kick this session off by underscoring the importance of an imperative transition that we must make to a low-carbon economy. Let me first start by stating the obvious, which is when you look at the macroeconomic costs of in action. Now that would substantially outweigh those -- if there was a disorderly transition versus an orderly transition. Based on a survey done with the G20 economies, you would note in the chart to your left, that if current policies were to continue as is, i.e., a scenario without further climate action, that is expected to lead to GDP losses exclusively just due to physical risk of 8% and 20% by 2050 and 2100, respectively. Now in contrast, if we were to consider a net 0 transition scenario, which means that we limit the temperature increase to 1.5 degrees by 2050. GDP losses are reduced to an estimated 4% by 2050 and 3% by 2100. Now transition risk, therefore, one can say, has a moderately negative impact on our world GDP in this scenario through 2050. But if action is delayed, the scale and sharpness that is required and the adjustments that were required will be growing disproportionately. On the right, you would also see that several G20 nations have noted that increased temperatures would cause destruction through various channels such as negative effect on GDP through decline in productivity, detrimental effect on investments and financial stability as a result of uncertainty and in growth. There will be also additional channels of transmission where there will be risk to agriculture, tourism and energy sectors. There are also concerns on rising energy prices due to transition policies and climate change and the challenge related to low sales efficiency in energy in several countries. That said, the transition policies are really creating new avenues of economic growth and employment, while also accelerating investment towards adaptation and mitigating climate scenarios. There are increased employment opportunities, especially in industries like energy, construction and transportation sectors. There are countries with rich mineral reserves that have highlighted opportunities for job creation. There is also a change expected in the international trade system as well. So all in all, transition is likely to have a positive impact on investments, but would lead to increase in fiscal gaps, I would -- as you would note from the top left chart. But then you'll also note from the bottom left chart that there are financing consults, like technology and data gaps, labor market frictions and all of these are deemed as obstacles in the transition pathway. There are emerging economies, especially that are going to be seeking low-cost financing to implement transition and adaptation and so that one has adjust transition, so to speak. Even then, short-term costs of transitioning are small compared to longer-term prosperity that we expect to bring, especially when you compare it to the baseline of our hot house world scenario. Now both the public as well as private financing markets, acknowledging this, as you can see from the chart on your right. The global investment in low-carbon energy transition has totaled to $1.1 trillion in 2022, which is a new record and a huge acceleration from the year before as the energy crisis that we saw and policy action drove faster deployment of clean energy technologies. Renewables, top of the chart. We had the largest investment there on record as well at $495 billion in 2022. And then the electrification of the transport system came a close second with $466 billion in 2022. And then if you think about the countries where these investments are really taking place, China was by far the leading country. It accounted for nearly half of that global investment. And U.S. was a distant second at $141 billion, followed by Germany, France and the U.K. Now, however, to adequately displace fossil fuels and to limit global temperature rise to no more than 1.5 degrees, we still need the energy supply investments to accelerate, and that too significantly. If you look at the energy supply banking ratio that we see here for every dollar of bank financing activity supporting fossil fuel, we currently have 0.8 supporting low-carbon energy as per BloombergNEF. In 2021, where we have these last starts available, G-SIBs or globally significant important banking institutions, underwrote a $1.1 trillion of energy supply transactions with $499 billion being low carbon and $581 billion being fossil fuels. The NZBA, which is the NetZero Banking Association collectively underwrote $1.2 trillion of energy supply financing, of which $56 billion was low carbon. Which then brings me to our next section, really. We are, of course, seeing that financial institutions are awakening to the significance of transition financing. However, and quite rightly so, the related credit considerations are complex. And given the near-term impact on risk can be related to credit deterioration. There is a need for a balanced and integrated climate impact assessment. It's paramount to encourage the flow of capital in the right direction. And so how do we really do this fairly complex analysis in a relatively intuitive and confident manner. The answer really lies, I would say, in ensuring that various relevant modeling dimensions are considered. And on this slide, we've outlined most of these, if not all. On the data side, companies need granular information on carbon emissions and on assets geolocation for the firms they are exposed to carbon emissions are often not reported in a standardized way or not reported at all, especially for smaller companies. So availability of a large standardized data set or consistent benchmarks becomes quite critical to being able to conduct that analysis for starters. With regards to geolocation of assets, one can refer to a large database again, like what S&P provides or perform a mapping exercise to identify major climate-related financial risks, sometimes multiple with different impacts. The assessment is also driven by the type of scenario you consider in your analysis, the transition speed, it's influenced, obviously, by policy, by carbon tax levels, which differ not only by country but by industry and also by the speed of carbon tax evolution over the foreseeable future. Much is also going to depend on how companies will respond to transition. So as one can imagine, assessing alternate part to future counterparty behavior, different impacts on the production of goods and elasticity of demand in response to price changes, asset standing. All of this can really have varying implications on the exposure of the borrower's creditworthiness. And traditional risk analysis or credit risk models may not be able to capture all of these dimensions, especially given the extended time horizons. So really, if we think about this as the starting point, what else really needs to guide the framework is 2 other considerations, one for large and medium corporates. These would tend to be sector-based, high carbon emitting sectors where one can really tailor approaches that can capture nuances on transition trajectories and incorporate that company's specific behavior. Whereas then if you think about small and medium enterprises, whether it's lack of data, that's going to be a significant issue, credible benchmarks and a consistent and reliable approach would become necessary. And then, of course, the fact of the matter is that there is not a direct way to really back test results. Because after all, this transition is something we're all doing for the first time. But then the good news is that in this case, a scenario-based framework can become very useful to estimating our potential outcomes given the heightened uncertainty. So therefore, the idea is going to be for one to really start from current credit worthiness of the company and then assess how this is impacted by various shocks coming from transition and from physical risk events and ultimately, also from the overall macroeconomic considerations. So in this slide, what we are seeing is that we're focusing on climate-related risk analysis as there are already well-established techniques that link credit risks or defaults to business and financial risk. So in terms of approach, one focuses on relevant scenario variables, for example, the temperature pathways and then establishes its impact on intermediate company-specific drivers and finally translate them into financial impact that can be used in the usual sort of credit risk models. The framework that you see should be able to capture flexibility, not only in terms of time horizon, but also on the ability to flex variables in the shorter term, if macroeconomic conditions around us were to change for some reason. Another key consideration is the reality that defers across industries, right? So when we're seeing in the oil and gas segment, we're not seeing the same level of dramatic effect in other segments such as mineral mining or consumer staples. And then lastly, of course, the impact on credit risk change cannot be factored in without considering the corporate's ability to borrow and the changing interest rate environment around [indiscernible]. And so the framework should be able to incorporate all of these facets, and that's when it is more likely to be considered more holistic and dynamic in its ability to capture both climate risk as well as climate opportunities. I will now pass the floor to my colleague, James, who will touch upon aspects involving modeling some of the physical risks in particular. We will then pass on the floor to Alban, who will bring us home with an approach developed by Oliver Wyman and S&P Global to quantifying the credit risk impact of both physical and transition risk. Over to you, James.

Wonderful. Thank you so much, Prerna. Well, let's see, how would you quantify physical risks and the financial implications. Seems simple to me, it's just that you have to model the whole bio, geochemical and dynamical systems of the Earth. Couple that with the economic systems of the earth from macro to micro, and then just do that in the context of the sociopolitical trends. So it sounds simple. Let's take a look at how we might approach this. I want to just anchor us back in the TCFD framework, which I imagine is familiar to most of you, task force and climate-related financial disclosures. And the things that I love about this are the way that it breaks it down into the different impacts on different types of money, if you will, the income statement versus balance sheet, separates out risks and very importantly, opportunities. And then over on the left, you can see that the risks are separated into transition physical risks. I'll be concentrating on the physical risks for the moment. This is the general methodology that we found very useful to start from the assets from the individual locations and do a bottom-up analysis. So we started -- actually before this picture, this picture is hazards, vulnerabilities and risks, which is a framework taken from the catastrophe risk modeling community. But before that, you have to understand where the assets are, and that is sometimes quite a challenge because companies don't always know where all their assets are or disclose them in a way that's easy to aggregate and access. So getting the assets in the first place is a really big challenge. But then once you have the assets, you need to go asset by asset, location by location and understand, first, on the left, what are the various hazards or perils that this asset is exposed to? I'll go through some of the ones that we think about on the next slide. But you can see that this one shows temperature increasing sort of gradually and linearly as it tends to do over the course of the next 2 decades. And then once you understand the hazards that, that individual location is exposed to, that is not the end of the story because, as you can imagine, imagine a region-wide drought, for example, and you have 2 different assets, one is an office building and one is an agricultural asset like a field of wheat. Those would respond to that drought in very different ways financially. So that's where you need to have an impact function or as we call it, that expresses the vulnerability of that type of asset to that type of hazard. So already, we're talking about some complexity here. And as you can see, this can get quite nonlinear as the extent of the severity of the hazard grows, you can have some quick roll-offs in the impairment of the financial ability of that asset to produce its desired results. And that's finally how you get to the expression of risk in financial terms, and that's where you can see the dollars there. And then also in the percentage of the asset's value that is at risk can see on the top number, the 43.6%. So I'm just going to dive one click deeper. And on the right, you can see a flow chart that illustrates an example. So on the left, you can see the physical risk here that hazard itself is a coastal flood in this example. And how would that impact an asset type of an urban high-rise just as an example. Well, you can see that we would have multiple impact functions for different pathways of damage that could be created. On the top 2 are examples that would hit the expense -- increase expenses and hit the profit and loss -- so you can see cleanup costs of the issue as well as repair costs for anything that was permanently damaged. And then the next one down is rental income loss. That's an example of decrease in revenue, still in the P&L, but decreased in revenue. The bottom one impairs the asset, and that would be a foundation damage, for example, and that could impair the asset and reduce the book value. So you can see that for one type of physical risk, you can get all these different pathways and you need to really consider the impact functions for each different type. Over on the left, you've probably read this already, but the physical hazards that we found useful to consider first, I mean, there are many different ways that climate change can cause physical hazards, but some of the big ones are temperature extremes, not the means but the extremes is where it usually causes problems. Drought also generally the extremes, wildfire, water stress, which includes withdrawals, coastal flooding, which is a convolution of sea level rise, but also the storm surge that happens at the coast generally during flooding during a high-wind episode of some sort fluvial flooding, that's river-based flooding. Fluvial flooding, which is generally rain-driven or flash flood on impervia surfaces like cities and tropical cyclones. These are some of the main ones. There are plenty of others that are not on this list, cold waves, hail, that type of thing. And those are affected by climate change, and we personally are modeling those as we speak. You can see that we've got a number of asset types that we've already considered. And I think that this illustrates that how complex this gets from alfalfa plants to hydropower plants and the impact functions are in these light grade boxes. I think it's important to anchor to the literature, some of the questions that have already come in from you all are, how do you justify this? This is a very complex field. How do you justify this to stakeholders? And the answer is really just transparency. You have to be able to say, "Here's where the data came from. Here's what we did to it, here's why we think that's rational. And as Prerna mentioned, a little hard to back test this stuff still. But if you build on rational and transparent and defensible models from the bottom up, it's about as good as you can do. I just want to end with pointing out some of the areas that are still an active research mode. And so on the left, you've got some physical ones. On the right, you've got some transition ones. The very top one on the left is physical thresholds and tipping points. Those are really deblished to model. And what I'm talking about there is, for example, perhaps you heard the news story about the Atlantic Meridian overturning circulation, potentially -- well, it's slowing down, we measure it slowing down, could potentially slow down a lot or possibly even stop it does that from time to time in the geological record. And that would have a very, very significant and potentially rapid effect. There are other thresholds that we possibly are approaching may already have crossed you, for example, think rapid Antarctic ice sheet melting. It's very unclear what's going on there right now. So critical, critical to model is correctly and very difficult. So working on that very quickly very intensely. I want to talk about cascading hazards for just a moment, which is that generally, we're not affected by the means as much as by the extremes that we're not well adapted to, and the extremes tend to happen when multiple hazards happen at the same time. So think wind plus hurricane plus flooding plus rain damage and river damage from the runoff in the rivers or the land slides. Multiple things happen at once, it's very hard to model. We're working on it a probabilistic approach is probably the right way. The last thing that I'm going to talk about is over on the right, at the bottom, you can see coupled physical and transition risk. So the physical risk and transition risk happened together and they trigger each other, and it's very important and hard to model. But as Alban will talk about in a moment, the very first step is looking at them together. The next step is actually coupling their interactions together. So with that, I'd love to pass it to my colleague, Alban and take it away.

Thank you very much, James. Thank you very much to the S&P team and for everybody to join this webinar today. So in the next couple of slides, we will be talking about how we can do climate risk analytics in a holistic manner, which means covering both transition and physical risk. And we'll take the example of credit risk in that case, and see how this -- the impact of climate scenario can materialize on the creditworthiness of companies. And we'll talk also a little bit about how we can -- are we leveraging all the good work that James has been talking about in the previous section as well as the well data that S&P Global is bringing to the table on this topic. So on this page, you have a very high level overview of the methodology that we have developed and deployed, especially at banks, but also other financial institutions and in some cases, corporates. I'll talk at a very high level. And then on the next page, we'll go through a few examples and case studies of what that means in practice. The -- on the left-hand side, you see the inputs. And this is a big part of the methodology because as you will see, there is a number of data attributes and at items that are required. It's important to mention that we believe to do our climate risk analytics in a credible and robust manner. You have to go bottom up, and you've seen that already with James in the previous section that doing it bottom up is relevant to capture the effects of climate risk. That's true on an asset-by-asset basis. That's also true for transition risk, given the difference of product mix, emission profile, et cetera. So on the input here, you will see that the goal is really to be able to surface the key drivers of risk and identify the right data attributes. So for this, we leverage the wealth of data sets that S&P has a disposal. So you think we can think about the company financials, for instance to -- as a starting point, or to do a climate scenario analysis and the forecast. So 2 million companies that include public and private companies, a number of private companies are covered as well, as you can see here. The emissions data quite important. So you're probably all familiar with Scope 1, Scope 2, Scope 3 emissions data. That's what we're sourcing here and adjust and leveraging in the analysis. A company that is more in carbon-intensive will be more impacted in the transition risk scenario than another one. The physical risk data is really what James talked about in the previous section, i.e. looking at asset by asset for many different companies and many different asset types across 7 physical risk hazard and look at how they can be impacted. The transition plans of the companies, of course, this analysis is highly dynamic and companies will continue to evolve and adapt to the scenario in question, therefore, you would want to capture what the trajectory can be. And then importantly, we have the industry-specific data that needs to be captured. And that's going to be quite important and very sector specific. So if you're thinking about the power generation sector, what the energy mix of the company will look like is going to be quite relevant. The core company will be different from nuclear-powered -- nuclear energy mix for our company, for instance. Similarly, if you're thinking about car manufacturing, for instance, the trajectory in terms of ramping up electric vehicle production is going to be quite relevant. Same for the fleet for an airline, for instance, and the age and the carbon efficiency of this fleet will drive a lot of the effects there. So we combine all this data together with the portfolio data of the institution. So generally, an identifier, PD rating, the traditional kind of credit risk metrics, if you want. And then we feed that into the methodology that you see in the middle of this page. And this is really about -- this is really a bottom-up approach, as we discussed before. And the box that you see in the middle is really the core of the engine. So we'll assess transition risk and physical risk through the impact on the key drivers of performance. So price, volume, unit cost, capital expenditure and asset value. So for instance, the unit cost will be impacted by the carbon price, which can be triangulated through the carbon price from the scenario as well as the emission intensity of the companies. But it will also be impacted by the additional insurance costs, for instance, from the physical risk elements and attribute for this specific company given their specific assets. And the idea here is really to capture that in a bottom-up manner, differentiated in terms of the specific exposure to climate risk and differentiated across customers and borrowers in a given sector. Then once you have that, then we will compute the full financial statements, income statement, cash flow statement, balance sheet, on a scenario adjusted basis for every point in time and every company that are covered here. And then once you have that, there's a number -- there are a number of things you can do. Generally, you will want to have the financial metrics coming from that. So the evolution of the credit score to get to a view of the impact of the climate scenarios on the creditworthiness of the counterparts, but also the valuation of the equity, for instance, of the bond, which is often also of interest to our clients. And then once you have that, then there are a number of outputs that you can drill into. So the -- what are the drivers driving the various results, what's contributing to the overall effect. What are the financial statements of the company given you have the full financial statements, you can go and check what's happening in a given year for a given company in a given business line, for instance, and of course, the credit metrics, so the evolution of the potential credit scores in that case. So this is just to give you a sense of the overall methodology. On the next page, we have an example of what that can mean for a specific company, in that case, in the power generation. It's important that -- to mention that the framework we discussed on the previous page is really tailored to each sector of the economy so that you can reflect the appropriate dynamics of the sector under a climate scenario. So here, we take a power generation example. You see the input data on the left-hand side, company financials, fuel mix, emission profiles, volume of production, transition plan, capital expenditure capacity, regulated and unregulated capacity, the asset exposure. All of that is data that we source from the S&P environment. And then we feed into the drivers of performance for the specific company. So if you think about the drivers of performance and the transmission mechanism for specific power generation company, it's going to be highly dependent on the price of electricity, which often tends to go up, especially in transition scenario due to the increased demand and increased investments. A big driver will be regulated a specific utility will be. So the unregulated utility will be much more exposed to a change in cost than a regulated utility. So this is important to have this view as well. The volume of production will depend on the company transition plans, whether they have the ability to spend some of the investments required to achieve the transition plan, and of course, the scenario demand for a given fuel in a given region. And you remember what James told you in the previous section that you also have physical risk effect that can, for instance, create additional downtime for specific power plants. So this is what we capture as well in terms of volume. And then the unit costs will really be about how the increased transition risk and physical risk will impact the cost profile of those companies. So the cost of production will depend on the emission profile of the energy mix. So the higher the -- the more intensive the energy mix -- the more carbon intensive the energy mix, the more impact you will have on the cost. And similarly, the more exposed you are to physical risk, the more you will have increased insurance costs and damage costs, et cetera. I'm not going to go through all the remaining drivers in the interest of time, but just to give you a sense of what that looks like to capture transition and physical risk in a holistic manner. And the idea here is really to build this linkage between the climate scenarios and the drivers of performance fundamentals perspective, meaning like the really reflecting the dynamics of the sector and the transmission mechanism. If I go on the next page, you can have a view of the -- we put a couple of examples here and try to illustrate the contribution of physical risk and transition risk to the overall impact. So you will see, for instance, the first company here is a data center owner, and we run the model with transition risk only and with transition and physical risk here. And this is a pretty carbon-intensive company. So they are impacted compared to their peers. They are impacted by transition risk quite a bit to the extent of a couple of notches. And -- but they're also exposed to physical risk even in a transition scenario. That's due to the fact that the data centers tend to have high exposure from extreme heat since the since they tend to have a higher back ratio compared to other asset type and the cooling requirement associated with those type of assets. So this is the type of characteristics that we try to bring to life here to assess what the impact will be. And you will see that the incremental effect from physical risk will basically exacerbate the transition risk effect by an additional 2 notches in 2050. On the other hand, you have a second company here, a large oil and gas producer in that case. This company, as you would expect, is quite impacted by transition risk, especially in the net 0 scenario. And this company has a lot of exposure in offshore oil and gas platform. And actually, those this asset, in particular, have limited exposure to some of the hazards, other hazards like wildfire and flood and drought and extreme heat. Therefore, the marginal credit risk, if you want for that specific company is going to be negligible compared to the transition risk part. I'm going to keep this page in the interest of time, and I will speak quickly to this page, which has a couple of examples, again, in the power generation sector. So here, you have 3 examples of utilities. The first one has a lot of -- it's a diversified power generation companies, but with a lot of renewables and basically solid financials, ambitious transition plans, ability to capture the demand due to their cash flow and debt headroom. So they tend to be able to materialize these transition plans, capture the extra demand from the climate scenarios in the transition scenario and end up doing quite well in the transition scenario and are being upgraded by 1 and 2 notches. Then another example at the very bottom of this page is a carbon heavy producer. So here a lot of coal and a lot of gas. And the because of the climate scenarios and the immediate implementation of the carbon price, the coal power generation tends to decrease very quickly, which decreases their cash flow, which decreases their ability to invest and tend to be -- tend to impact them and really jeopardize the transition. And in that case, they end up in a situation that are close to default. And the last one in the middle is a nuclear producer. Nuclear tends to be to have limited impact. Therefore, the rating tends to be quite stable across the scenario. So I believe we are out of time for this section. So I think I will hand it over back to Prerna to proceed with the Q&A. Thank you very much for your attention.

Thanks, Alban. A few questions that are coming along I see that we may probably not be able to go through all of the questions live today, but we'll get back to you using our sort of response channels. The first question and probably the most commonly asked question. Alban, I'll direct this one to you. I see that you're capturing transition plans in the models. How do you ensure that these are credible?

Yes, that's a great question. So I think a couple of elements here. The -- we want to keep this approach as fact-based and data-driven as possible. So the way we look at transition plan is we basically assess what they mean in terms of required to achieve those transition plans. And then we compare that to the ability of the company to invest. That means if BP said they would have 30 gigawatts of solar in 2030, we will compare that to the cash flow under the scenario as well as their ability to issue debt to finance this investment. That way we have a view of what that -- like whether those -- whether they will achieve the transition plan in a fact-based analytical manner. And of course, the user would have the ability to adjust some of those parameters and enter even their own transition plans if they have a view of what the transition plans could look like.

Thanks. That makes sense. Another, I guess, very, very common pain point that we see with a lot of our clients is about the missing data elements. And that would be the case for us, a lot of small, midsized companies. So the question that has been posed is what you do if there are missing data elements and how do you ensure again that the modeling is credible. I'll let Alban yourself answer this one as well.

Yes. No, of course, happy to speak to this one. Generally, the approach we've been taking is leverage as much of the data you have as possible. Therefore, we really proceed with the waterfall approach. So for the companies where you have a lot of data attributes, then you would basically leverage those. For those where you have a significant amount of data, but not everything you need, then you can take some level of work around. So for instance, if you don't have the emission of the company, you can take an emission intensity for peer companies, for instance. And then we do that at multiple levels of the waterfall, if you want, up to a point where you don't they had to do anything bottom up in which case we fall back on the more capital effort based on the extrapolation of a similar set of the results of a similar set of companies.

All right. I think we're almost out of time now. So for those of you who want to review anything that we've covered. This session has been recorded, and you will receive a link tomorrow to access it on demand at your convenience. When we close out this webinar, you will also be routed to a survey, which will take you less than a minute to complete and we'd appreciate your feedback to sell -- to help us with some of our future content. So I guess all that's left to say is thank you to my fellow speakers for presenting, and thank you all for taking the time to attend today's session. We look forward to you joining us again soon. Thanks very much. Bye-bye.

Thank you.