Assessing the Role of Investors in the Realization of Climate Mitigation Pathways - MATLAB
Video Player is loading.
Current Time 0:00
Duration 29:28
Loaded: 0%
Stream Type LIVE
Remaining Time 29:28
 
1x
  • Chapters
  • descriptions off, selected
  • en (Main), selected
    Video length is 29:28

    Assessing the Role of Investors in the Realization of Climate Mitigation Pathways

    Stefano Battiston, University of Zurich

    Financial institutions face growing demand from investors and regulators to assess climate risk on their portfolios. Financial authorities recommend basing these assessments on NGFS scenarios (future trajectories of carbon prices and production output across the high- and low-carbon sectors of the economy, depending on future introductions of climate policies).

    Even with these assessments, several challenges remain to integrate climate risk into risk management. The circularity between materialization of risk and risk perception can cause uncertainty about which scenario may occur. While risk is endogenous to finance in other contexts, in climate risk it impairs the use of standard risk assessment tools.

    This talk is a collaboration with scientists of the IAM community involved in the development of NGFS scenarios and explains how a new framework can provide scenarios that assess the role of investors in the realizations of climate mitigation pathways.

    Published: 7 Oct 2021

    Today we'll be talking about climate transition risk, and in particular how the climate scenarios that are available and recommended by can be used to assess climate risk. Let me first recap very briefly where we are. So scientific evidence about climate change has been there for about two decades.

    And we knew about the physical risk of not doing anything about climate change, that's called unmitigated climate change, and also the scale and the pace that of the transformation that is required in order to mitigate climate change. There are some specific characteristics of climate risk, which are discussed in these papers cited here.

    The most important one is I would say, the endogeneity. And this means that the perception of risk impacts on the risk itself, which is typically not the case whether your perception of the outcome of a dice doesn't change the outcome. But in climate risk, this is the case.

    Climate risk has been only recently recognized by financial authorities as a source of financial risk. This is a major step ahead. However, we will see there is much work to do. On the one hand, we see a remarkable growth of sustainable finance under various labels. Lots of assets under management are labeled as ESG or sustainable in different ways.

    However, at the same time, most countries are actually lagging behind in terms of the achievement of the Paris Agreement. So the scale and the pace of the transformation to mitigate climate change implies that a portion of assets that are affected is quite large. It also requires a proactive role of the financial system, which is, as we discussed, we cannot give for granted.

    And as a result, there is a transition risk that is stemming in particular for potential changes in the market expectations about future scenarios. So this is where financial risk becomes really a key aspect. But in order to assess climate transition risk, and for those who are not so familiar, so climate physical risks has to do with essentially the physical component of climate change and how it affects physical assets. And transition risk has to do with the risk associated with a low-carbon transition and the expectations about that.

    So we need a conceptual framework to assess transition risk. And this is the operational procedure that we propose based on our stream of scientific work. So the first step is to classify the economic activities, particularly according to the so-called climate policy relevant sectors. Second step is to look into the transition scenarios.

    Third step is to use them to derive stress tests or, more in general, risk measures. And the fourth step is to actually look at this a little bit from taking a step back and look at the problem with endogeneity of the scenarios.

    So I'm going to guide you quickly through these steps. The first is the classification of economic activities. Normally, you would think that all about transition risk, you could think, is included in the accounting of greenhouse gas emissions. However, and in these slides for your review as a background, there are the definitions of scope one, two, and three.

    However, for instance, for an oil company, most of the emissions are in the scope three. That means they are in the downstream of to the production. They are actually embodied in the use of the product. Well, scope one only captures the emissions that are released during the production process.

    However, scope three is essentially subject to internal modeling and reporting that is essentially not strictly comparable across companies. So it's problematic for a portfolio analysis. This motivates the analysis of technologies rather than simply emissions. So, for instance, a company could reduce its own scope one emissions by, for instance, a electricity company, by expanding a line of business in electricity trading, which doesn't have any direct emissions. But that doesn't mean that it's actually getting ready for the transition to the low-carbon economy.

    So this is why tracking technologies and technological profile of firms is important. So if we want to track technologies, though, we need to use classification of economic activities. The standards are in Europe called NACE. In the United States called NAICS. They're all part of the ISIC family, it's an international system, which, however, was designed many years ago without having climate in mind.

    So can we grow these sectors in a few categories with distinct features in terms of transition risk? I'm going to give you an example of the problem. Where the activities that are associated with revenues, that have revenue associated with fossil fuels? Well, if we look at the sections, and these are common for both the NACE and the NAICS.

    So some of the activities are here in section B, mining and quarrying. So you see mining of coal, crude petroleum. Obviously activities related to fossil fuels. But in section H, transport, there is also transport via pipeline. If you dig a bit deeper into that, this is essentially transportation, long-distance transportation, of natural gas.

    Then within the manufacturing sectors, there is the manufacture of coke and petroleum. So that's in a completely different section. And, yet, again in another section, utility, electricity, and gas, you find manufacture of gas. So all these activities have in common the fact that they either support or carry out extraction production, transportation sales of primary energy derived from fossil fuels.

    So they are all in the same bucket, in terms of transition risk. So we have addressed this problem by, first of all, identifying some key dimensions of transition risk and then going one-by-one through all the sectors, and there are about 1,000, and classify them according to these dimensions. The dimensions are, what is the role of the activity in the value chain? Is this primary, secondary energy or is it non-energy goods and services?

    What kind of emissions? Direct or indirect? CO2 or methane? Or could it contribute to negative emissions like afforestation? Is there a specific policy process? That's very important for risk because if there is an authority that is regulating those activities as a group, you want to separate from activities that are regulated by a different authority.

    Think of oil subsidies and taxes as opposed to the regulation of electricity by the electricity authority, or of the policies related to housing. So they go in really different buckets in the policy process. And then eventually there is also the level of substitutability of business fuels as an input.

    When you do this through all of the sectors, you come up with about nine main categories and then some further more granular categories. I'm showing the first six, which differ along the dimensions I mentioned. And then the last column to the right contains the list of NACE codes of the activities that fall into that bucket.

    And so these sectors are essentially fossil fuels and utility, electricity, transportation, buildings. And then energy intensive are all those that are substantially affected by a carbon tax and that do not fall already in the previous category. Now the mapping table to the NACE and CPRS is available as an Excel file, open source, at this link.

    And this classification has been used by a number of financial supervisors. Here's the list of studies that you can consult. And to summarize, this methodology addresses the question, what is the portion of assets in a portfolio that is exposed to each category of transition risk? There are two related questions.

    One related question is, can we additionally put together all those assets that are primarily affected in an adverse manner by transition risk so they're affected by high-transition risk, you could say? Yes. We can do that. And another related question is, can we identify the activities that are potentially at least aligned to the European taxonomy of sustainable investments? Which is, at least in Europe, and willingly, intentionally becoming an international standard for green investments.

    And, yes. The answer is yes. We have to provide coefficients at the sector level to do so. And so the first version of the taxonomy alignment coefficients are available at this link. The transition risk coefficients will be available very soon next month in the reference here.

    The second step I mentioned is transition scenarios. What is important to understand about this is that these are not predictions. They describe instead what the economy, and also the land use, might look like in the next decades. But, however, we also need to understand these are not arbitrary trajectories in the future. They have been developed by making sure that they are compatible with the following constraints.

    One is coming from the laws of physics, essentially that a cumulative emissions determine statistically the degrees of global warming that the planet is going towards in the next decades. And then, also, the technological constraints that are coming from limits to technological coefficiency of engines, for instance, or how much energy you can extract from different types of fuels, how fast you can build capacity in renewable energy plants and so on.

    A final remark is that the type of models that are used to produce these scenarios, which I'm not-- I don't work myself directly within these models. I'm working on the next stage of the workflow. These are so-called large scale, very important, large-scale integrated assessment models, not to be confused with aggregated, integrated assessment models, which are, instead, being the object of other type of works. If you have heard of it, no-doubts model, that belongs to this other category. Here it's a different class. It's a class in which there is a large detail on the energy sector, in particular.

    And the NGFS, which is the Network for Greening the Financial System, is a platform of financial supervisors. Over 80 national and international authorities and central banks and organizations have joined this platform. In collaboration with the community that works on these integrated assessment models, and, in particular, within the framework, if you want, of the international-- sorry, intergovernmental panel for climate change, the IPCC, have identified a group of scenarios and, first of all, a group of archetypical scenarios.

    So these are called high-level scenarios. In this chart they're represented along two dimensions. On the horizontal is physical risk and vertical is transition risk. And then within those-- the high level, there are then some sub scenarios in each scenario. So, for instance, the most optimistic scenario is when everything goes well or in the best way. There is little physical risk and little transition risk because the world manages to agree early on, on low carbon transition pathway, which is predictable and, therefore, anticipated by market players.

    However, there is also a scenario in which-- although, the transition is achieved, this is done at the very last moment because of the political economy of it and this leads to a situation of transition risk. Higher levels of transition risk. In concrete terms, what do they imply, these scenarios? Let me give an example. On the right-hand side I'm plotting trajectory of the output of electricity based on coal, in particular in the region China under one specific model called REMIND is the one developed by the Potsdam Institute for Climate.

    And as you can see under the scenario current policies, which assumes it's only the policies that are already being introduced will stay, but nothing else, it's a very conservative one. The output of this sector peaks in 2035, whereas in an immediate 1.5 scenario, we'll have to peak immediately. And conversely, there is a kind of symmetric situation for low-carbon technologies. For wind, for instance, the output would go up in all scenarios, but much earlier and further up in the immediate transition.

    So now this is the link to finance because when you carry out a financial evaluation of a firm based on projected future profits, clearly profits are a function of revenues and costs. They're both a function of output. And it's clear that the profits of a company under-- that is operating in this sector under one scenario and the other scenario are very, very different. So this is where-- is the link between the scenarios and finance.

    So to summarize up to this stage, the transition risk is very important, is resulting the way we think about this is that it results from a changing financial actors' expectations. Yesterday, the financial actors are thinking to leave in current policy scenario. This morning they wake up and start believing that we are in the 1.5 scenario because of some new information that is coming. Maybe they get convinced by the new policies in green technologies of the United States, administration, or the ones by the European Commission.

    And then they redo the calculation of the value of both equity and bonds of firms based on the new scenarios. And then results become very different. And the difference is the shock that is essentially the loss or the gain, depending on which out of-- which activity the firm is engaged in, associated with transition risk. So this is a summary of what I said before. And just to maybe mention that I want to give-- the defining characteristics of scenarios are the targets, in terms of temperature, and the timing of the introduction of the policy, as well as the level of reliance on carbon dioxide removal.

    So now we can move to the next step, which is, how do we carry out a financial risk analysis? First of all, we need to take into account that typically as a financial institution, if I'm a financial analyst of the financial institution, I want to analyze the counterparty risk, credit risk, typically, or even market risk, but always at the counterparty level. So the firm that I'm talking about matters. And it's structured in terms of what are the technologies in which this firm is engaged.

    So what is needed here is a breakdown of the revenues, possibly the capex, capital expenditures, but typically this is not disclosed. So the fault of that and lack of that, the revenues, can be broken down across the granular categories of the climate policy-driven sectors. And then we can essentially project the future output and therefore the-- eventually estimate the future profits of the firm as a result of the combination of the output across the different activities.

    Once we've done that, we can compute the value of a financial instrument, for instance, equity or bonds, under the baseline scenario. And then redo that under a transition scenario. One, there are various ways of doing this. The simplest is, and probably less controversial, is to carry out the evaluation of equity based on discount to future profits. And for bonds and loans, one can use the compound future profits trajectory and combine that with a very simple structural model of default, whereby if assets are not exceeding the liabilities then you have a default by the maturity of the bond and you have a default.

    And you can compute the PD, probability of default, and also the loss given default. It is in the future after the transition in the transition scenario. It is also possible to calibrate, to use implied levels, implied probability, and implied loss given default using market data for calibration. Good. And so this adjustment is essentially the final output of all this exercise where you recompute the-- so you compute the adjustment in the valuation under the assumption that market expectations are changing from baseline to transition.

    And you are going to see what is the outcome. Visually I can represent this in the following way. These are two scenarios. And this is-- take it as a baseline. The other one is the transition scenario. So the output of this firm is pretty different. You compute the valuation, taking the discounted flow of dividends in the future in one scenario. In the other scenario, and essentially the difference, take into account, of course, the discounting over time, will drive the magnitude of the shock.

    In full probability of default and loss given default, the model is a bit more articulated, so I don't have the time to present it here. But there will also be a kind of a handbook providing this forthcoming. Let me spend the last four minutes presenting-- well, first of all, this is an example output of what you can do with this methodology for giving-- for selection of sectors. You have the value of the PD in the baseline and in the transition scenario, so you see how the PD of high-carbon sectors tend to increase and the LGD also tend to increase, and, of course, for the valuation of a bond both entering into the formula and is the resulting shock, which is typically negative for the high-carbon.

    And so, interestingly, these shocks under scenarios that are relatively not extreme can easily go to between 20% and 60%. So they're pretty substantial shocks. So this is an important issue for investors to consider. But the last thing I want to mention is that once we have done all this exercise, then we have to reflect a little bit on the meaning of these transition scenarios. So we have recently published a paper on science together with precisely those researchers that produce those scenarios.

    And what we have highlighted is that those scenarios are generated-- first of all, they're a very important step ahead. It's immense, the progress that they allowed the whole financial sector to make. However, there is a missing link, which I will illustrate here visually. So we start from the climate scenarios, temperature, emissions, and so on, then it goes to the output of the different sectors, in terms of energy and so on. Then this is translated, as we have just discussed, into an adjustment of risk measures, for instance, value at risk and other quantities.

    This, in principle, should lead to a reallocation of capital into low-carbon-- from high-carbon to low-carbon. However, this reallocation of capital is not guaranteed to be sufficient. It's not guaranteed, nothing guarantees that it is sufficient, to-- to provide enough investments into the low-carbon that will make-- will make it consistent with the scenarios. So in other terms, if-- and I will explain in a moment.

    So, imagine investors are looking at those scenarios among which there is the orderly transition and they get the perception that it's possible that the most likely scenario is that there is a transition that will happen smoothly without higher risk for the high-carbon firms. Then they have no incentive to reallocate capital or to demand from those firms to change their technological profile. But that means that essentially the economy will stay as it is and we know that this is clearly incompatible with the scenario.

    So the scenario becomes self-defeating. So this is very important. So we have proposed a framework to interface the modeling framework of the integrated assessment models with a climate financial risk modeling. And the result of these coupling is, essentially, the following. The trajectories are taken up by the CFR and produce, depending on the expectations of the investors, levels of interest rate, which are specific to the high-carbon and low-carbon sectors, which are then fed back into the IAM to compute adjusted trajectories.

    Here is an example. So in the top chart, you see how a solid line is the standard trajectory for low-carbon, blue, and high-carbon, the red. And in the presence of the other module that takes into account the reaction of the investors, well, the transition could actually occur earlier and be even smoother than without the financial system. So that's very good news. Financial system can actually enabling the transition.

    However, this is not the only possibility. In this bottom-left chart, we are showing the mirror case where the policy's introduced now, but investors don't find it credible. And the resulting trajectories, taking into account the reaction of investors, is actually a delayed transition with possibly jumps in the value of the assets. So this implies that we really have to think about the meaning of these transition scenarios. There are a number of policy implications, which I put there for those who are interested. But I'll stop here and I'm ready to take your questions. Thank you.