Editing IPCC:AR6/WGIII/Chapter-15 (section)

=== 15.6.1 Addressing Knowledge Gaps with Regard to Climate Risk Analysis and Transparency ===

<div id="h2-13-siblings" class="h2-siblings"></div>

Achieving climate mitigation and adaptation objectives requires ambitious climate finance flows in the near-term, that is, 5–10 years ahead. However, knowledge gaps in the assessment of climate-related financial risk are a key barrier to such climate finance flows. Therefore, this section discusses the main knowledge gaps that are currently being addressed in the literature and those that remain outstanding.

Climate-related financial risk is meant here as the potential adverse impact of climate change on the value of financial assets. A recent but remarkable development since AR5 is that climate change has been explicitly recognised by financial supervisors as a source of financial risk that matters both for financial institutions and citizens’ savings ( [[#Bolton--2020|Bolton et al. 2020]] ). Previously, climate change was mostly regarded in the finance community only as an ethical issue. The reasons why climate change implies financial risk are not new and are discussed more in detail below. What is new is that climate enters now as a factor in the assessment of financial institutions’ risk (e.g., the European Central Bank or the European Banking Authority) and credit rating ( [[#15.6.3|Section 15.6.3]] ), and, going forward, into stress-test exercises. This implies changes in incentives of the supervised financial actors, both public and private, and thus changes in the landscape of mitigation action by generating a new potential for climate finance flows. However, critical knowledge gaps remain. In particular, the underestimation of climate-related financial risk by public and private financial actors can explain that the current allocation of capital among financial institutions is often inconsistent with the mitigation objectives (Rempel et al. 2020). Moreover, even a correct assessment of risk, which could provide incentives for divesting from carbon-intensive activities, does not necessarily lead to investing in the technical options needed for deep decarbonisation. Therefore, understanding the dynamics of the low-carbon transition require to fill in at the same time gaps about risk and gaps about investments in enabling activities in a broader sense.

'''Physical risk.''' On the one hand, unmitigated climate change implies an increased potential for adverse socio-economic impacts especially in more exposed economic activities and areas ( ''high confidence'' ). Accordingly, ''physical risk'' refers to the component of financial risk associated with the adverse physical impact of hazards related to climate change (e.g., extreme weather events or sea level rise) on the financial value of assets such as industrial plants or real estate. In turn, these losses can translate into losses on the values of financial assets issued by exposed companies (e.g., equity/bonds) and or sovereign entities as well as losses for insurance companies. The assessment of climate financial physical risks poses challenges in terms of data, methods and scenarios. It requires cross-match scenarios of climate-related hazards at granular geographical scale, with the geolocation and financial value of physical assets. The relationship between the value of physical assets (such as plants or real estate) and the financial value of securities issued by the owners of those assets is not straightforward. Further, the repercussion of climate-related hazards on sovereign risk should also be accounted for.

'''Transition risks and opportunities.''' On the other hand, the mitigation of climate change, by means of a transition to a low-carbon economy, requires a transformation of the energy and production system at a pace and scale that implies adverse impacts on a range of economic activities, but also opportunities for some other activities ( ''high confidence'' ). If these impacts are factored in by financial markets, they are reflected in the value of financial assets. Thus, ''transition risks and'' ''opportunities'' refers to the component of financial risk (opportunities) associated with negative (positive) adjustments in assets’ values resulting directly or indirectly from the low-carbon transition.

The concepts of ''carbon stranded assets'' (see e.g., [[#Leaton--2011|Leaton and Sussams 2011]] ), and ''orderly'' vs ''disorderly transition'' ( [[#Sussams--2015|Sussams et al. 2015]] ) which emerged in the NGO community, have provided powerful metaphors to conceptualise transition risks and have evolved into concepts used also by financial supervisors ( [[#NGFS--2019|NGFS 2019]] )and academics. The term ''carbon stranded assets'' refers to fossil fuel-related assets (fuel or equipment) that become unproductive. An ''orderly transition'' is defined here as a situation in which market players are able to fully anticipate the price adjustments that could arise from the transition. In this case, there would still be losses associated with stranded assets, but it would be possible for market players to spread losses over time and plan ahead. In contrast, a ''disorderly transition'' is defined here as a situation in which a transition to a low-carbon economy on a 2°C path is achieved (i.e., by about 2040), but the impact of climate policies in terms of reallocation of capital into low-carbon activities and the corresponding adjustment in prices of financial assets (e.g., bonds and equity shares) is large, sudden and not fully anticipated by market players and investors. Note the impact could be unanticipated even if the date of the introduction is known in advance by the market players. There are several reasons why such adjustments could occur. One simple argument is that the political economy of the transition is characterised by forces pulling in different directions, including opposing interests within the industry, and mounting pressure from social awareness of unmitigated climate risks. Politics will have to find a synthesis and the outcome could remain uncertain until it suddenly unravels. Note also that, in order to be relevant for financial risk, the disorderly transition does not need to be a catastrophic scenario in terms of the fabric of markets. It also does not automatically entail systemic risk, as discussed below. Knowledge gaps in this area are related to emerging questions, including: What are, in detail, the transmission channels of physical and transition risk? How to assess the magnitude of the exposure to these risks for financial institutions and ultimately for people’s savings? How do transition risk and opportunities depend on the future scenarios of climate change and climate policies? How to deal with the intrinsic uncertainty around the scenarios? To what extent could an underestimation of climate-related financial risk feed back on the alignment of climate finance flows and hamper the low-carbon transition? Should climate risk be explicitly accounted for in regulatory frameworks for financial institutions, such as Basel III for banks and national frameworks for insurance? What lessons from the 2008 financial crisis are relevant here, regarding moral hazard and the trustworthiness of credit risk ratings? The attention of both practitioners and the scientific community to these questions has grown since the Paris Agreement. In the following we review some of the findings from the literature, but the field is relatively young and many of the questions are still open. [[#footnote-005|12]] Damages from climate change are expected to escalate dramatically in Europe ( [[#Forzieri--2018|Forzieri et al. 2018]] ) and in some EU countries there is already some evidence that banks, anticipating possible losses on the their loan books, lend proportionally less as a consequence.

'''Assessment of physical risk.''' There is a literature on estimates of economic losses on physical assets (see Cross-Working Group Box ECONOMIC in chapter 16 of AR6 WGII). Here we discuss some figures and mechanisms that are relevant for the financial system. Significant cost increases have been observed related to increases in frequency and magnitude of extreme events ( ''high confidence'' ) ( [[#15.4.2|Section 15.4.2]] ). At the global level, the expected ‘climate value at risk’ (climate VaR) of financial assets has been estimated to be 1.8% along a business-as-usual emissions path ( [[#Dietz--2016|Dietz et al. 2016]] ), with however, a concentration of risk in the tail (e.g., 99th VaR equals to 16.9%, or USD24.2 [[#footnote-004|13]] trillion, in 2016). Climate-related impacts are estimated to increase the frequency of banking crises (up over 200% across scenarios) while rescuing insolvent banks could increase the ratio of public debt to gross domestic product by a factor of two ( [[#Lamperti--2019|Lamperti et al. 2019]] ). Further assessments of physical risk for financial assets ( [[#Mandel--2020|Mandel 2020]] ), accounting in particular for the propagation of losses through financial networks, estimate global yearly GDP losses at 7.1% (1.13%) in 2080, without adaptation (with adaptation), the former corresponding to a 10-fold increase with respect to the current yearly losses (0.76% of global GDP). Finally, climate physical risk can impact on the value of sovereign '''bonds''' (one of the top asset classes by size), in particular for vulnerable countries ( [[#Volz--2020|Volz et al. 2020]] ).

Insurance pay-outs for catastrophes have increased significantly over the last 10 years, with dramatic cost spikes in years with multiple major catastrophes (such as in 2018 with hurricanes Harvey, Irma, and Maria). This trend is expected to continue. The indirect costs of a climate-related flooding event can be up to 50% of the total costs, the majority of which is not covered by insurance ( [[#Alnes--2018|Alnes et al. 2018]] ) (Section15.6.4). The gap between total damage losses and insurance pay-outs has increased over the past 10 years ( [[#Swiss%20Re%20Institute--2019|Swiss Re Institute 2019]] ). Indeed, the probability of ‘extreme but plausible’ scenarios will be progressively revised upwards in the ‘value at risk’. As a result it becomes more difficult to find financial actors willing to provide insurance, as was observed for real estate in relation to flood and wildfires in California ( [[#Ouazad--2019|Ouazad and Kahn 2019]] ). This progressive adjustment would keep the financial system safe ( [[#Climate-Related%20Market%20Risk%20Subcommittee--2020|Climate-Related Market Risk Subcommittee 2020]] ; [[#Keenan--2020|Keenan and Bradt 2020]] ), but transfer to taxpayers the onus of damage compensation and the financing of adaptation investments ( [[#OECD--2021c|OECD 2021c]] ) as well as build up latent liabilities.

'''Assessment of transition risk. Carbon stranded assets.''' Fossil fuel reserve and resource estimates exceed in equivalent quantity of CO 2 with virtual certainty the carbon budget available to reach the 1.5°C and 2°C targets ( ''high confidence'' ) ( [[#Meinshausen--2009|Meinshausen et al. 2009]] ; [[#McGlade--2015|McGlade and Ekins 2015]] ; [[#Millar--2017|Millar et al. 2017]] ). In relative terms, stranded assets of fossil fuel companies amount to 82% of global coal reserves, 49% of global gas reserves and 33% of global oil reserves ( [[#McGlade--2015|McGlade and Ekins 2015]] ). This suggests that only less than the whole quantity of fossil fuels currently valued (either currently extracted, waiting for extraction as reserves or assets on company balance sheets) can yield economic return if the carbon budget is respected. The devaluation of fossil fuel assets implies financial losses for both the public sector ( [[#15.6.8|Section 15.6.8]] ) and the private sector ( [[#Coffin--2019|Coffin and Grant 2019]] ). Global estimates of potential stranded fossil fuel assets amount to at least 1 trillion, based on ongoing low-carbon technology trends and in the absence of climate policies (cumulated to 2035 with 10% discount rate applied; USD8 trillion without discounting ( [[#Mercure--2018a|Mercure et al. 2018a]] )). With worldwide climate policies to achieve the 2°C target with 75% likelihood, this could increase to over USD4 trillion (until 2035, 10% discount rate; USD12 trillion without discounting). Other estimates indicate USD8–15 trillion (until 2050, 5% discount rate, ( [[#Bauer--2015|Bauer et al. 2015]] )) and USD185 trillion (cumulated to year 2115 using combined social and private discount rate ( [[#Linquiti--2016|Linquiti and Cogswell 2016]] )). However the geographical distribution of potential stranded fossil fuel assets (also called ‘unburnable carbon’) is not even across the world due to differences in production costs ( [[#McGlade--2015|McGlade and Ekins 2015]] ). In this context, a delayed deployment of climate finance and consequently limited alignment of investment activity with the Paris Agreement tend to strengthen carbon and thus to increase the magnitude of stranded assets.

'''Assets directly and indirectly exposed to transition risk.''' In terms of types of assets and economic activities, the focus of estimates of carbon stranded assets tends to be on physical reserves of fossil fuel (e.g., oil fields) and sometimes financial assets of fossil fuel companies ( [[#van%20der%20Ploeg--2020|van der Ploeg and Rezai 2020]] ). However, a precondition for a broader analysis of transition risks and opportunities is to go beyond the narrative of stranded assets and to consider a classification of sectors of all the economic activities that could be affected ( [[#Monasterolo--2020|Monasterolo 2020]] ). This, in turn depends on their direct or indirect role in the GHG value chain, their level of substitutability with respect to fossil fuel and their role in the policy landscape. Moreover, such a classification needs to be replicable and comparable across portfolios and jurisdictions. One classification that meets these criteria is the Climate Policy Relevant Sectors (CPRS) ( [[#Battiston--2017|Battiston et al. 2017]] ) which has been used in several studies by financial supervisors ( [[#EIOPA--2018|EIOPA 2018]] ; [[#ECB--2019|ECB 2019]] ; [[#EBA--2020|EBA 2020]] ; [[#ESMA--2020|ESMA 2020]] ). The CPRS classification builds on the international classification of economic activities (ISIC) to map the most granular level (4 digits) into a small set of categories characterised by differing types of risk: fossil fuel (i.e., all activities whose revenues depend mostly and directly on fossil fuel, including concession of reserves and operating industrial plants for extraction and refinement); electricity (affected in terms of input but that can in principle diversify their energy sources); energy intensive (e.g., steel or cement production plants, automotive manufacturing plants), which are affected in terms of energy cost but not in terms of the main input); and transport and buildings (affected in terms of both energy sources and specific policies). All financial assets (e.g., bonds, equity shares, loans) having as issuers or counterparties firms whose revenues depend significantly on the above activities are thus potentially exposed to transition risks and opportunities. Further, investors’ portfolios have to be part of the analysis since changes in financial assets values affect the stability of financial institutions and can thus feed back into the transition dynamics itself (e.g., through cost of debt for firms and through costs for assisting the financial sector). One outstanding challenge for the analysis of investors’ exposure to climate risks is the difficulty of gathering granular and standardised information on the breakdown of non-financial firms’ revenues and CAPEX in terms of low-/high-carbon activities ( ''hig'' ''h confidence'' ).

Several financial supervisors have conducted assessments of transition risk for the financial system at the regional level. For instance, the European Central Bank (ECB) reported preliminary estimates of aggregate exposures of financial institutions to CPRS relative to their total debt securities holdings as ranging between 1% for banks to about 9% for investment funds ( [[#ECB--2019|ECB 2019]] ). The European Insurance and Occupational Pensions Authority (EIOPA) reported aggregate exposures to CPRS of EU insurance companies at about 13% of their total securities holdings ( [[#EIOPA--2018|EIOPA 2018]] ). Further analyses on the EU securities holdings indicate that among financial investments in bonds issued by non-financial corporations, EU institutions hold exposures to CPRS ranging between 36.8% for investment funds to 47.7% for insurance corporations; analogous figures for equity holdings range from 36.4% for banks to 43.1% for pension funds ( [[#Alessi--2019|Alessi et al. 2019]] ). Another study indicates that losses on EU insurance portfolios of sovereign bonds could reach up to 1%, in conservative scenarios ( [[#Battiston--2019|Battiston et al. 2019]] ).

Given the magnitude of the assets that are potentially exposed, reported in the previously cited studies, a delayed or uncoordinated transition risk can have implications for financial stability not only at the level of individual financial institutions, but also at the macro level. The possible systemic nature of climate financial risk has been highlighted on the basis of general equilibrium economic analysis ( [[#Stern--2021|Stern and Stiglitz 2021]] ).

Some financial authorities recognise that climate change represents a major source of systemic risk, particularly for banks with portfolios concentrated in certain economic sectors or geographical areas ( [[#de%20Guindos--2021|de Guindos 2021]] ). Specifically, the concern that central banks would have to act as ‘climate rescuers of last resort’ in a systemic financial crisis stemming from some combination of physical and transition risk has been raised in the financial supervisor community ( [[#Bolton--2020|Bolton et al. 2020]] ). The systemic nature of climate risk is reinforced by the possible presence of moral hazard. Indeed, if a sufficient number of financial actors have an incentive to downplay climate-related financial risk, then systemic risk builds up in the financial system, eventually materialising for taxpayers ( [[#Climate-Related%20Market%20Risk%20Subcommittee--2020|Climate-Related Market Risk Subcommittee 2020]] ). While such type of risk may go undetected to standard market indicators for a while, it can materialise with a time delay, similarly to the developments observed in the run up to the 2008 financial crisis.

These considerations are part of an ongoing discussion on whether the current financial frameworks, including Basel III, should incorporate explicitly climate risk as a systemic risk. In particular, the challenges in quantifying the extent of climate risk, reviewed in this section, especially if risk is systemic, raise the question whether a combination of quantitative and qualitative restrictions on banks’ portfolios could be put in place to limit the build-up of climate risks ( [[#Baranović--2021|Baranović et al. 2021]] ).

'''Endogeneity of risk and''' '''multiplicity of scenarios.''' One fundamental challenge is that climate-related financial risk is endogenous ( ''high confidence'' ). This means that the perception of the risk changes the risk itself, unlike most contexts of financial risk. Indeed, transition risk depends on whether governments and firms continue on a business-as-usual pathway (i.e., misaligned with the Paris Agreement targets) or engage on a climate mitigation pathway. But the realisation of the transition pathway depends itself on how, collectively, society, including financial investors and supervisors, perceive the risk of taking or not taking the transition scenario. The circularity between perception of risk and realisation of the scenario implies that multiple scenarios are possible, and that which scenario is ultimately realised can depend on policy action. The coordination problem associated also with low-carbon investments opportunities increases the uncertainty. Further, not all low-carbon activities are directly functional to the transition (e.g., investments in pharmaceutical, IT companies, or financial intermediaries), thus not all reallocations of capital lead to the same path.

In this context, probabilities of occurrence of scenarios are difficult to assess and this is important because risks vary widely across the different scenarios. In this context a major challenge is the fat-tail nature of physical risk. One the one hand, forecasts of climate change and its impact on humans and ecosystems imply tail events ( [[#Weitzman--2014|Weitzman 2014]] ) and tipping points which cannot be overcome by model consensus ( [[#Knutti--2010|Knutti 2010]] ). On the other hand, everything else the same, costs and benefits vary substantially with assumptions on agents’ utility, productivity, and intertemporal discount rate, which ultimately depend on philosophical and ethical considerations ( [[#Nordhaus--2007|Nordhaus 2007]] ; [[#Stern--2008|Stern 2008]] ; [[#Pindyck--2013|Pindyck 2013]] ). Thus, more knowledge is needed on the interaction of climate physical and transition risks, the possible reinforcing feedbacks and transmission channels to the economy and to finance. Moreover, models need to account for compound risk, that is, the interaction of climate physical and/or transition risk with other sources of risk such as pandemics, such as COVID-19.

'''Challenges for climate transition scenarios.''' The endogeneity of risk and its associated deep uncertainty implies that the standard approach to financial risk, consisting of computing expected values and risk based on historical values of market prices, is not adequate for climate risk ( ''high confidence'' ) ( [[#Bolton--2020|Bolton et al. 2020]] ). To address this challenge, a recent stream of work has developed an approach to make use of climate policy scenarios to derive risk measures (e.g., expected shortfall) for financial assets and portfolios, conditioned to scenarios of disorderly transition ( [[#Battiston--2017|Battiston et al. 2017]] ; [[#Monasterolo--2020|Monasterolo and Battiston 2020]] ; [[#Roncoroni--2020|Roncoroni et al. 2020]] ). In particular, climate policy shocks on the output of low-/high-carbon economic activities are calculated based on trajectories of energy technologies as provided by large-scale Integrated Assessment Models ( [[#Kriegler--2015|Kriegler et al. 2015]] ; [[#McCollum--2018|McCollum et al. 2018]] ) conditioned to the introduction of specific climate policies over time. This approach allows to conduct climate stress-tests both at the level of financial institutions and at the level of the financial system of a given jurisdiction.

In a similar spirit, recently, the community of financial supervisors in collaboration with the community of climate economics has identified a set of climate policy scenarios, based on large-scale IAM, as candidate scenarios for assessing transition risk ( [[#Monasterolo--2020|Monasterolo and Battiston 2020]] ). These scenarios have been used, for instance, in an assessment of transition risk conducted at a national central bank ( [[#Allen--2020|Allen et al. 2020]] ). This development is key to mainstreaming the assessment of transition risk among financial institutions, but the following challenges emerge ( ''high confidence'' ). First, a consensus among financial supervisors and actors on scenarios of transition risk that are too mild could lead to a systematic underestimation of risk. The reason is that the default probability of leveraged financial institutions is sensitive to errors in the estimation of the loss distribution and hence sensitive on the choice of transition scenarios ( [[#Battiston--2020|Battiston and]] [[#Monasterolo--2020|Monasterolo 2020]] ). This in turn could lead to an allocation of capital across low-/high-carbon activities that is insufficient to cater for the investment needs of the low-carbon transition.

Second, IAM do not contain a description of the financial system in terms of actors and instruments and make assumptions on agents’ expectations that could be inconsistent with the nature of a disorderly transition ( [[#Espagne--2018|Espagne 2018]] ; [[#Pollitt--2018a|Pollitt and Mercure 2018a]] ; [[#Battiston--2020b|Battiston et al. 2020b]] ). In particular, IAMs solve for least cost pathways to an emissions target in 2100 (AR4 WGIII SPM Box 3), while the financial sector’s time horizon is much shorter and risk is an important factor in investment decisions.

Third, the current modelling frameworks used to develop climate mitigation scenarios, which are based on large-scale IAM, assume that the financial system acts always as an enabler and do not account for the fact that, under some condition (i.e., if there is underestimation of climate transition risk) can also act as a barrier to the transition ( [[#Battiston--2020a|Battiston et al. 2020a]] ) because it invests disproportionately more in high-carbon activities.

'''Macroeconomic implications of the technological transition''' . Global macroeconomic changes that may affect asset prices are expected to take place as a result of a possible reduction in growth or contraction of fossil fuel demand, in scenarios in which climate targets are met according to carbon budgets, but also following ongoing energy efficiency changes ( ''high confidence'' ) ( [[#Clarke--2014|Clarke et al. 2014]] ; [[#Mercure--2018a|Mercure et al. 2018a]] ). A review of the economic mechanisms involved in the accumulation of systemic risk associated with declining industries, with focus on fossil fuels, is given by [[#Semieniuk--2021|Semieniuk et al. (2021)]] . An example is the transport sector, which uses around 50% of oil extracted ( [[#IEA--2018|IEA 2018]] ; [[#Thomä--2018|Thomä 2018]] ). A rapid diffusion of EV (and other alternative vehicle types) poses an important risk as it could lead to oil demand peaking far before mid-century ( [[#Mercure--2018b|Mercure et al. 2018b]] ; 2021). New technologies and fuel switching in aviation, heavy industry and shipping could further displace liquid fossil fuel demand ( [[#IEA--2017|IEA 2017]] ). A rapid diffusion of solar photovoltaic could displace electricity generation based predominantly on coal and gas ( [[#Sussams--2017|Sussams and Leaton 2017]] ). A rapid diffusion of household and commercial indoor heating and cooling based on electricity could further reduce the demand for oil, coal and gas ( [[#Knobloch--2019|Knobloch et al. 2019]] ). Parallels can be made with earlier literature on great waves of innovation, eras of clustered technological innovation and diffusion between which periods of economic, financial and social instability have emerged (Freeman and Louca 2001; [[#Perez--2009|Perez 2009]] ).

Due to the predominantly international nature of fossil fuel markets, assets may be at risk from regulatory and technological changes both domestically and in foreign countries ( ''medium confidence'' ). Fossil fuel exporting nations with lower competitiveness could lose substantial amounts of industrial activity and employment in scenarios of peaking or declining demand for fossil fuels. In scenarios of peaking oil demand, production is likely to concentrate towards the Middle East and OPEC countries ( [[#IEA--2017|IEA 2017]] ). Since state-owned fossil fuel companies tend to enjoy lower production costs, privately-owned fossil fuel companies are more at risk ( [[#Thomä--2018|Thomä 2018]] ). Losses of employment may be directly linked to losses of fossil fuel-related industrial activity or indirectly linked through losses of large institutions, notably of government income from extraction royalties and export duties. A multiplier effect may take place making losses of employment spill out of fossil fuel extraction, transformation and transportation sectors into other supplying sectors ( [[#Mercure--2018a|Mercure et al. 2018a]] ).

'''Main regulatory developments and''' '''voluntary responses to climate risk''' . Framing climate risk as a financial risk (not just as an ethical issue) is key for it to become an actionable criterion for investment decision among mainstream investors ( ''high confidence'' ) ( [[#TCFD--2019|TCFD 2019]] ). Since 2015 financial supervisors and central banks (e.g., the Financial Stability Board, the G20 Green Finance Study Group, and the Network for Greening the Financial System (NGFS)) have played a central role in raising awareness and increasing transparency of the potential material financial impacts of climate change within the financial sector ( [[#Bank%20of%20England--2015|Bank of England 2015]] , 2018; [[#TCFD--2019|TCFD 2019]] ). The NGFS initiative has engaged, in particular, in the elaboration of climate financial risk scenarios.

Although disclosure has increased since the TCFD recommendations were published, the information is still insufficient for investors and more clarity is needed on potential financial impacts and how resilient corporate strategies are under different scenarios ( [[#TCFD--2019|TCFD 2019]] ). Several efforts to provide guidance and tools for the application of the TCFD recommendations have been made (using Sustainability Accounting Standards Board (SASB) Standards and the Climate Disclosure Standards Board (CDSB) Framework to Enhance Climate-Related Financial Disclosures in Mainstream Reporting TCFD Implementation Guide ( [[#UNEP%20FI--2018|UNEP FI 2018]] ; CDSB and SASB 2019). Results of voluntary reporting have been mixed, with one study pointing to unreliable and incomparable results reported by the US utilities sector to the CDP ( [[#Stanny--2018|Stanny 2018]] ).

There have been also similar initiatives at the national level ( [[#DNB--2017|DNB 2017]] ; UK Government 2017; [[#US%20GCRP--2018b|US GCRP 2018b]] ). In particular, France was the first country to mandate climate risk disclosure from financial institutions (via Article 173 of the law on energy transition). However, disclosure responses have been so far mixed in scope and detail, with the majority of insurance companies not reporting on physical risk ( [[#Evain--2018|Evain et al. 2018]] ). In the UK, mandatory GHG emissions reporting for UK-listed companies has not led to substantial emissions reductions to date but could be laying the foundation for future mitigation ( [[#Tang--2018|Tang and Demeritt 2018]] ).

A key recent development is the EU Taxonomy for Sustainable Finance ( [[#TEG--2019|TEG 2019]] ), which provides a classification of economic activities that (among other dimensions) contribute to climate mitigation or can be enabling for the low-carbon transition. Indirectly, such classification provides useful information on investors’ exposure to transition risk ( [[#Alessi--2019|Alessi et al. 2019]] ; [[#ESMA--2020|ESMA 2020]] ). Finally, many consultancies have stepped forward offering services related to climate risk. However, the methods are typically proprietary, non-transparent, or based primarily on carbon footprinting, which is a necessary but insufficient measure of climate risk. Further, ESG (environmental, social and governance) metrics can be useful but are, alone, inadequate to assess climate risk.

Decision-makers in financial risk management make increasing use of climate policy scenarios, in line with the TCFD guidelines and the recommendations of the NGFS. In order to reduce the number of scenarios to consider, Illustrative Mitigation Pathways (IMPs, Chapter 3), have been elaborated to illustrate key features that characterise the possible climate (policy) futures. The following considerations can be useful for scenario end-users who carry out risk analyses on the basis of the scenarios described in Chapter 3. It is possible to associate climate policy scenarios with levels of physical and/or transition risk, but these are not provided with the scenario data themselves.

On the one hand, each scenario is associated with a warming path, which in turn, on the basis of the results from WGII, implies certain levels of physical risk (AR6 WGII Chapter 16). However, climate impacts are not accounted for in the scenarios. Moreover, levels of risk may vary with the reason for concern and with the speed of the implementation of adaptation. On the other hand, while mitigation can come with transition risk, in the case of lack of coordination among the actors, as discussed earlier in this section, this is not modelled explicitly in the trajectories, since the financial sector is not represented in underlying models. The scientific state of the art in climate-related financial risk offers an analysis that is not yet comprehensive of both the physical and transition risk dimensions in the same quantitative framework. However, decision-makers can follow a mixed approach where they can combine quantitative risk assessment for transition risk with more qualitative risk analysis related to physical risk.

Figure 15.6 represents sequences of events following along a scenario both in terms of physical risk (left) and transition risk (right). Four groups of IMPs (more are considered based on the warming level they lead to in 2100. Current Policies (CurPol) considers climate policies implemented in 2020 with only a gradual strengthening afterwards, leading to above 4°C warming (with respect to pre-industrial levels). Moderate Action (ModAct) explores the impact of implementing the NDCs (pledged mitigation targets) as formulated in 2020 and some further strengthening afterwards, thereby limiting warming to less than 4°C (&gt;50%), but above 3°C (&gt;50%). In these two scenarios, there is no stabilisation of temperature, meaning that further warming occurs after 2100 (and higher risk) even if stabilisation could be eventually achieved. They are referred to as pathways with higher emissions. The warming levels reached along these two scenarios imply physical risk levels that are ‘Moderate’ until 2050 and ‘Very High’ in 2050–2100 (with low levels of adaptation). Noting, that ‘Moderate’ physical risk can mean for some countries (i.e., SIDS) significant and even hardly absorbable consequences (i.e., reaching hard adaptation limits). Transition risk is not relevant for these scenarios, since a transition is not pursued.

<div id="_idContainer023" class="_idGenObjectStyleOverride-1"></div>

[[File:32f78a53772c689be90825e4d2b62c37 IPCC_AR6_WGIII_Figure_15_6.png]]

'''Figure 15.6 | Schematic representation of climate scenarios in terms of both physical and transition risk.''' While the figure does not cover all possible events, it maps out how the combination of stated targets can lead to different paths in terms of risk, depending on implementation progress and policy credibility. IMP 1.5°C and IMP ''&lt;'' 2°C are representative for IMP-GS (Sens. Neg; Ren), IMP-Neg, IMP-LD; IMP-Ren; IMP-SP. Note that the figure defines ‘High’ progress as higher, but it is important that the physical risk varies by region and country. This means, that ‘Moderate’ physical risk can be significant and even hardly absorbable for some countries.

Illustrative Mitigation Pathways include two groups of scenarios consistent with modelled global pathways that limit warming to 2°C (&gt;67%) or lower, respectively. The two groups are representative for the IMPs defined in Chapter 3. In these scenarios, warming is stabilised before 2100. The warming levels along these paths imply ‘Moderate’ physical risk until 2050 and ‘High’ risk in 2050–2100 (with low levels of adaptation). Transition risk can arise along these trajectories from changes in expectations of economic actors about which of the scenarios is about to materialise. These changes imply, in turn, possible large variations in the financial valuation of securities and contracts, with losses on the portfolio of institutional investors and households. High policy credibility is key to avoiding transition risk, by making expectations consistent early on with the scenario. Low credibility can delay the adjustment of expectations by several years, leading either to a late and sudden adjustment. However, if the policy never becomes credible, this changes the scenario since the initial target is not met.

<div id="15.6.2" class="h2-container"></div>

<span id="enabling-environments"></span>