Climate assumptions in forward-looking statements

Here is a strong example from Nestlé of how to assess and identify physical climate risks that are relevant for an industry with global operations. In its Non Financial Statement 2025, Nestlé presents a modeled analysis of projected crop yield changes by 2040 compared to 2024 under an intermediate emissions scenario. The visualization maps expected increases or decreases in yields across key raw materials and sourcing countries. It covers commodities such as coffee, cocoa, maize, wheat, dairy and palm oil, and links them to specific geographies. The analysis shows how changes in temperature, precipitation patterns and extreme weather events can alter growing conditions. As a result, certain regions may experience yield reductions, while others may see moderate increases. For a company that relies on agricultural inputs across multiple continents, these shifts translate directly into supply availability risks, cost volatility and sourcing complexity. This is not a generic climate narrative. It is a forward looking assessment of commodity exposure. By modeling yield impacts at country and crop level, the company can identify where supply chains are most vulnerable and where strategic adjustments may be required. Consequently, physical climate risk becomes a procurement and capital allocation issue. It affects long term contracting, supplier engagement, regenerative agriculture investments, water management and geographic diversification. In other words, climate resilience must be embedded in core business planning, not treated as a separate sustainability topic. This example illustrates how climate risk assessment can move from disclosure to decision making. Source: Nestlé, Non Financial Statement 2025 

2025 climate data: what are key considerations for your risk assessment? According to the Copernicus Global Climate Highlights 2025: - Global temperature: +1.47°C vs pre-industrial - 2023 to 2025: first three-year average above 1.5°C - 50% of land areas: more strong heat-stress days than average - Lowest global sea ice extent on record What I see as the key risk signals (a) Warming is becoming persistent. The last 11 years are the 11 warmest on record. That changes probability assumptions. (b) Physical risk is becoming broad, not regional. Half the globe experienced more strong heat-stress days. This affects labour productivity, cooling demand and operational thresholds. (c) Polar amplification and sea ice decline continue. That influences long-term sea level, supply chains, Arctic routes, and insurance modelling. (d) Scenario analysis needs recalibration. If the current baseline is 1.4°C, your “1.5°C scenario” is no longer a distant case, it’s close to becoming reality. What you can practically do with this report? If you sit in finance, audit, or risk: - Use it to challenge your physical risk assessment assumptions. Are you modelling based on historic averages or current warming levels? - Revisit site-level exposure. Heat stress data (UTCI) can inform workforce risk mapping. - Review insurance premiums and disclosures and renewal discussions. Insurers are pricing trend, not political targets. - Check consistency between climate narrative and financial assumptions. If your report acknowledges 1.4°C warming, do your asset lives and impairment tests reflect that? - Lastly, use it as external evidence. This is independent, data-driven analysis from ECMWF. It strengthens governance documentation. #ClimateRisk #RiskManagement #FinancialReporting

How should Climate risks be reflected in Credit Loss Accounting such as IFR9 Expected Credit Loss (ECL) impairment standard? This is what this latest publication from European regulators represented by European Systemic Risk Board (ESRB) tries to answer. "𝘾𝙡𝙞𝙢𝙖𝙩𝙚 𝙧𝙞𝙨𝙠𝙨 𝙢𝙪𝙨𝙩 𝙗𝙚 𝙞𝙣𝙘𝙤𝙧𝙥𝙤𝙧𝙖𝙩𝙚𝙙 𝘪𝘯𝘵𝘰 𝘵𝘩𝘦 𝘮𝘢𝘤𝘳𝘰𝘦𝘤𝘰𝘯𝘰𝘮𝘪𝘤 𝘢𝘯𝘥 𝘰𝘵𝘩𝘦𝘳 𝘮𝘰𝘥𝘦𝘭𝘴 𝘶𝘴𝘦𝘥 𝘣𝘺 𝘱𝘳𝘦𝘱𝘢𝘳𝘦𝘳𝘴 𝘰𝘧 𝘧𝘪𝘯𝘢𝘯𝘤𝘪𝘢𝘭 𝘴𝘵𝘢𝘵𝘦𝘮𝘦𝘯𝘵𝘴, 𝘪𝘯𝘤𝘭𝘶𝘥𝘪𝘯𝘨 𝘣𝘢𝘯𝘬𝘴 𝘧𝘰𝘳 𝘵𝘩𝘦 𝘦𝘴𝘵𝘪𝘮𝘢𝘵𝘪𝘰𝘯 𝘰𝘧 𝘦𝘹𝘱𝘦𝘤𝘵𝘦𝘥 𝘤𝘳𝘦𝘥𝘪𝘵 𝘭𝘰𝘴𝘴𝘦𝘴" "𝙁𝙖𝙞𝙡𝙞𝙣𝙜 𝙩𝙤 𝙞𝙣𝙘𝙤𝙧𝙥𝙤𝙧𝙖𝙩𝙚 𝙩𝙝𝙚𝙨𝙚 𝙛𝙖𝙘𝙩𝙤𝙧𝙨 𝙞𝙣𝙩𝙤 𝙗𝙖𝙣𝙠 𝙢𝙤𝙙𝙚𝙡𝙨 𝘮𝘢𝘺 𝘱𝘳𝘰𝘥𝘶𝘤𝘦 𝘦𝘴𝘵𝘪𝘮𝘢𝘵𝘦𝘴 𝘰𝘧 𝘤𝘳𝘦𝘥𝘪𝘵 𝘭𝘰𝘴𝘴𝘦𝘴 𝘵𝘩𝘢𝘵 𝘢𝘳𝘦 𝘵𝘰𝘰 𝘣𝘦𝘯𝘪𝘨𝘯, 𝘢𝘴 𝘵𝘩𝘦𝘺 𝘸𝘰𝘶𝘭𝘥 𝘭𝘪𝘬𝘦𝘭𝘺 𝘳𝘦𝘴𝘶𝘭𝘵 𝘧𝘳𝘰𝘮 𝘥𝘦𝘭𝘢𝘺𝘦𝘥 𝘳𝘦𝘤𝘰𝘨𝘯𝘪𝘵𝘪𝘰𝘯 𝘰𝘧 𝘵𝘩𝘦 𝘴𝘪𝘨𝘯𝘪𝘧𝘪𝘤𝘢𝘯𝘵 𝘪𝘯𝘤𝘳𝘦𝘢𝘴𝘦 𝘪𝘯 𝘤𝘳𝘦𝘥𝘪𝘵 𝘳𝘪𝘴𝘬 𝘧𝘰𝘳 𝘦𝘹𝘱𝘰𝘴𝘶𝘳𝘦𝘴 𝘮𝘰𝘳𝘦 𝘷𝘶𝘭𝘯𝘦𝘳𝘢𝘣𝘭𝘦 𝘵𝘰 𝘤𝘭𝘪𝘮𝘢𝘵𝘦 𝘳𝘪𝘴𝘬𝘴. , 𝘯𝘰𝘵 𝘤𝘰𝘯𝘴𝘪𝘥𝘦𝘳𝘪𝘯𝘨 𝘤𝘭𝘪𝘮𝘢𝘵𝘦 𝘳𝘪𝘴𝘬 𝘪𝘯 𝘵𝘩𝘦 𝙚𝙭𝙥𝙚𝙘𝙩𝙚𝙙 𝙘𝙧𝙚𝙙𝙞𝙩 𝙡𝙤𝙨𝙨 𝙢𝙤𝙙𝙚𝙡𝙨 𝘤𝘰𝘶𝘭𝘥 𝘢𝘧𝘧𝘦𝘤𝘵 𝘵𝘩𝘦 𝘦𝘴𝘵𝘪𝘮𝘢𝘵𝘪𝘰𝘯 𝘰𝘧 𝙥𝙧𝙤𝙗𝙖𝙗𝙞𝙡𝙞𝙩𝙮 𝙤𝙛 𝙙𝙚𝙛𝙖𝙪𝙡𝙩 (𝘗𝘋), 𝘢𝘴 𝘸𝘦𝘭𝘭 𝘢𝘴 𝙡𝙤𝙨𝙨 𝙜𝙞𝙫𝙚𝙣 𝙙𝙚𝙛𝙖𝙪𝙡𝙩 (𝘓𝘎𝘋), 𝘵𝘩𝘳𝘰𝘶𝘨𝘩 𝘰𝘷𝘦𝘳𝘦𝘴𝘵𝘪𝘮𝘢𝘵𝘪𝘰𝘯 𝘰𝘧 𝘵𝘩𝘦 𝘤𝘰𝘭𝘭𝘢𝘵𝘦𝘳𝘢𝘭 𝘷𝘢𝘭𝘶𝘦." "𝘚𝘪𝘮𝘪𝘭𝘢𝘳𝘭𝘺, 𝙞𝙣𝙨𝙪𝙧𝙖𝙣𝙘𝙚 𝙘𝙤𝙧𝙥𝙤𝙧𝙖𝙩𝙞𝙤𝙣𝙨 𝙣𝙚𝙚𝙙 𝙩𝙤 𝙘𝙤𝙣𝙨𝙞𝙙𝙚𝙧 𝙩𝙝𝙚 𝙚𝙛𝙛𝙚𝙘𝙩 𝙤𝙛 𝙘𝙡𝙞𝙢𝙖𝙩𝙚 𝙧𝙞𝙨𝙠𝙨 𝘸𝘩𝘦𝘯 𝘦𝘴𝘵𝘪𝘮𝘢𝘵𝘪𝘯𝘨 𝘵𝘩𝘦 𝘧𝘶𝘵𝘶𝘳𝘦 𝘤𝘢𝘴𝘩 𝘧𝘭𝘰𝘸𝘴 𝘰𝘧 𝘵𝘩𝘦𝘪𝘳 𝘪𝘯𝘴𝘶𝘳𝘢𝘯𝘤𝘦 𝘤𝘰𝘯𝘵𝘳𝘢𝘤𝘵𝘴 𝘶𝘯𝘥𝘦𝘳 𝘐𝘍𝘙𝘚 17." See section 3.3. of the report (link below) for more details These confirmations were made as part of the broader assessment performed by representatives from European Central Bank, European Banking Authority (EBA), European Insurance and Occupational Pensions Authority (EIOPA) and other EU regulators of how climate risks are addressed in existing IFRS accounting standards and reflected in financial statements, identifying 4 relevant issues for financial stability: 1) The incomplete incorporation of climate risks in market 𝗽𝗿𝗶𝗰𝗲𝘀 can cause assets to be 𝗼𝘃𝗲𝗿𝗲𝘀𝘁𝗶𝗺𝗮𝘁𝗲𝗱 or liabilities to be 𝘂𝗻𝗱𝗲𝗿𝗲𝘀𝘁𝗶𝗺𝗮𝘁𝗲𝗱 2) Failing to fully include relevant climate risks in impairment tests for non-financial assets may distort their valuation 3) It may be 𝗼𝗽𝗲𝗿𝗮𝘁𝗶𝗼𝗻𝗮𝗹𝗹𝘆 𝗱𝗶𝗳𝗳𝗶𝗰𝘂𝗹t for banks and insurance corporations to incorporate 𝗰𝗹𝗶𝗺𝗮𝘁𝗲 𝗳𝗮𝗰𝘁𝗼𝗿𝘀 𝗶𝗻𝘁𝗼 𝘁𝗵𝗲 𝗺𝗼𝗱𝗲𝗹𝘀 used to estimate expected credit losses under IFRS 9 Financial Instruments or expected cash flows from insurance contracts according to IFRS 17 Insurance Contracts 4) Efforts should be made to 𝗲𝗻𝗵𝗮𝗻𝗰𝗲 𝗱𝗶𝘀𝗰𝗹𝗼𝘀𝘂𝗿𝗲 requirements about how climate risks are reflected in the financial statements.

The International Accounting Standards Board (IASB) has issued a near-final draft of "Disclosures about Uncertainties in the Financial Statements Illustrated using Climate-related Examples". 1- These examples aim to improve the reporting of climate-related and other uncertainties in financial statements. 2- They do not introduce new IFRS requirements but show how existing standards apply. 3- While there's no set effective date, the IASB expects entities to have sufficient time for implementation, typically months. 4- Regulators will likely expect these examples to be considered in the next financial statements. .. Example 1: Materiality of Climate Transition Plans (IAS 1/IFRS 18)   Entities facing significant climate-related risks may need to explain the lack of impact from their transition plans on financial statements if deemed material. In contrast, those with limited risks may find such explanations immaterial. This emphasizes the importance of assessing materiality based on specific facts and user expectations. .. Example 2: Impairment Test Assumptions (IAS 36) This example highlights the need to disclose key assumptions and perform sensitivity analysis during impairment tests for goodwill or indefinite-lived intangible assets, even when the recoverable amount exceeds the carrying amount. It stresses that all relevant key assumptions, not just discount or growth rates, should be disclosed. .. Example 3: Sources of Estimation Uncertainty (IAS 1)   This example clarifies that disclosures regarding assumptions that carry a significant risk of leading to material adjustments in asset and liability carrying amounts within the forthcoming year may be essential, even if these assumptions are not expected to be resolved within that period. .. Example 4: Climate-Related Credit Risk (IFRS 7) This example demonstrates the required disclosures related to credit risk when climate-related factors materially influence specific loan portfolios. It outlines the considerations for determining the materiality of such information and offers illustrative examples of relevant disclosures. .. Example 5: Materiality of Provisions (IAS 37)  This example shows that obligations like decommissioning responsibilities can be material even if the recognized provision seems quantitatively immaterial due to discounting. Factors such as the risk of early settlement are relevant in assessing materiality. .. Example 6: Disaggregation for Clarity (IFRS 18) This example shows how to disaggregate asset information within the same class, like Property, Plant, and Equipment (PP&E), based on distinct risk characteristics, such as varying GHG emissions, to offer relevant insights. ... Boards should recognize that these examples offer valuable insights into current IFRS disclosure mandates while maintaining existing requirements. Considering them is essential for effective financial reporting on climate and other uncertainties. .. More: 👇 Source: IASB, EY

1 in 10 years → Every 3 years. Climate change is impacting construction projects: Yet nobody’s talking about it… The issue: UK construction contracts reference "1 in 10 year weather events." But these events are now happening every 3 years. Your project will be in either a peak or a dip in the weather cycle. NEC contracts use Met Office data and locations to define these thresholds. Great in theory. Useless when climate patterns have shifted. This spring in Europe? Severe drought across parts of Europe. Last year? Extreme flooding. What this means for contractors working under a NEC contract: → Can't claim for weather delays if it's "less than 1 in 10 years" → But extreme weather is now the norm, not the exception → NEC's fixed methodology offers no discretion for new patterns → Reality: unpredictable extremes every few years. Z Clauses for weather will be needed. Real example happening now: UK groundworks contractors are weeks ahead of schedule. Beautiful dry conditions alertly in the year. Perfect for excavation. Great if you're building a reservoir, however will there be enough water to fill it in the future? The truth: Our contracts assume weather patterns from 30 years ago. Climate change will increasingly make those assumptions worthless. The NEC's objective approach was great in the 1990s but needs to adapt to new realities. Time to update: → Redefine what constitutes "extreme" weather → Adjust frequency assumptions in contracts → Plan for unpredictability, not monthly patterns → Build climate resilience into risk allocation The 1 in 10 year clause is dead. Climate change will kill it. Time our contracts caught up with reality. P.S. Want to get smarter at contracts, claims, and commercial risk? I drop sharp, no-BS insights straight from the top of the industry. Join my FREE newsletter here — don't miss what’s coming: https://lnkd.in/ga9WGi6C

“𝘚𝘩𝘰𝘸 𝘮𝘦 𝘸𝘩𝘦𝘳𝘦 𝘤𝘭𝘪𝘮𝘢𝘵𝘦 𝘳𝘪𝘴𝘬 𝘤𝘩𝘢𝘯𝘨𝘦𝘥 𝘢 𝘥𝘦𝘤𝘪𝘴𝘪𝘰𝘯.” That is now the defining question for UK banks and insurers. Not because of theory. Because the evidence has shifted.  • The 𝗡𝗮𝘁𝗶𝗼𝗻𝗮𝗹 𝗖𝗹𝗶𝗺𝗮𝘁𝗲 𝗘𝗺𝗲𝗿𝗴𝗲𝗻𝗰𝘆 𝗕𝗿𝗶𝗲𝗳𝗶𝗻𝗴 makes clear the UK is already experiencing systemic physical risk.  • 𝗟𝗦𝗘𝗚 data shows UK-listed assets among the most exposed globally to 𝗳𝗹𝗼𝗼𝗱 𝗿𝗶𝘀𝗸.  • S&P Global now frames the world as tracking toward ~𝟮.𝟯°𝗖 𝗯𝘆 𝟮𝟬𝟰𝟬, not some distant horizon.   Markets don’t reprice climate risk smoothly; they reprice it when assumptions break. That context matters for the PRA’s Policy 𝗦𝘁𝗮𝘁𝗲𝗺𝗲𝗻𝘁 𝗣𝗦𝟮𝟱/𝟮𝟱, which updates and replaces SS3/19 with SS4/25. Under SS4/25, supervisors are not interested in your framework. Not your disclosures. Not the fact that you’ve run scenarios. They are interested in a decision.  • A loan repriced.  • A location exited.  • A risk limit tightened. An assumption has been adjusted because future risk no longer resembles historical loss data. PS25/25 does not invent new requirements. What it does is raise the bar on how firms evidence judgement. The PRA is clearer and more explicit about how climate risks should be integrated into governance, risk management, and scenario analysis and, critically, how that integration influences real outcomes. This is what principles-based regulation actually means in practice. There are no prescribed actions. But supervisors will expect firms to justify why their approaches are reasonable, proportionate and credible given their exposures, especially where data is imperfect, and uncertainty is unavoidable.  • 𝗙𝗼𝗿 𝗯𝗮𝗻𝗸𝘀, this scrutiny increasingly shows up through second-order effects: changes in insurance pricing or availability feeding directly into collateral values, credit assumptions and liquidity resilience.  • 𝗙𝗼𝗿 𝗶𝗻𝘀𝘂𝗿𝗲𝗿𝘀, it shows up when historic loss experience is no longer a sufficient guide for underwriting, reserving or capital planning. For both, it shows up when climate scenario analysis exists, but doesn’t change decisions. Supervisors are not asking for perfect models. They are asking whether leadership understands the limits of those models and has acted accordingly. By 2026, the question is whether your decisions and ultimately your balance sheet show you understood climate risks. 𝗣𝗥𝗔 𝗣𝗼𝗹𝗶𝗰𝘆 𝗦𝘁𝗮𝘁𝗲𝗺𝗲𝗻𝘁 𝗣𝗦𝟮𝟱/𝟮𝟱 (𝗗𝗲𝗰 𝟮𝟬𝟮𝟱): 👉 https://lnkd.in/eyS45KXP #ClimateRisk #RiskManagement #PRA #SS4_25 #Banking #Insurance #CRO

𝗧𝗮𝗿𝗴𝗲𝘁𝘀, 𝘀𝗰𝗲𝗻𝗮𝗿𝗶𝗼𝘀, 𝗮𝗻𝗱 𝘁𝗿𝗮𝗻𝘀𝗶𝘁𝗶𝗼𝗻 𝗽𝗹𝗮𝗻𝘀 𝗺𝘂𝘀𝘁 𝗮𝗹𝗹 𝗯𝗲 𝗯𝘂𝗶𝗹𝘁 𝗼𝗻 𝘀𝗵𝗮𝗿𝗲𝗱 𝗹𝗼𝗴𝗶𝗰. 𝗧𝗵𝗮𝘁'𝘀 𝘁𝗵𝗲 𝗰𝗼𝗻𝗰𝗹𝘂𝘀𝗶𝗼𝗻 𝗼𝗳 𝘁𝘄𝗼 𝗻𝗲𝘄 𝗿𝗲𝗽𝗼𝗿𝘁𝘀 𝗽𝘂𝗯𝗹𝗶𝘀𝗵𝗲𝗱 𝗯𝘆 𝘁𝗵𝗲 𝗡𝗚𝗙𝗦. 🔸 Scenario analysis, target setting and transition planning share the same data foundations. All depend on forward-looking assumptions, sectoral emissions pathways, macroeconomic models, and regulatory policy expectations. Treating them separately introduces inconsistencies that compromise decision usefulness. 🔸 Scenario analysis and transition plans are forward-looking and aim to test financial resilience. When done well, they support setting credible decarbonisation targets, identifying high-risk assets, and testing how portfolios perform under multiple policy and physical risk futures. 🔸 Transition plans are more credible when based on real scenarios. Linking them to plausible pathways improves their robustness, especially if the institution’s own exposures, strategy, and risk appetite are tested against those pathways. 🔸 The NGFS recommends a unified data architecture. This means shared climate metrics, consistent modelling assumptions, and integrated systems that serve both functions. It also means governance structures that treat scenario analysis, target setting and transition planning as a single operational cycle. 🔸 NGFS sets out four essential elements for target setting. These are: a baseline, a quantitative target, a timeframe, and an implementation strategy. Many institutions still skip the last step, yet it’s the one that connects ambition to action. 𝗠𝘆 𝗧𝗮𝗸𝗲 Targets are not meaningful unless they're backed by transition plans. Transition plans are not credible unless they are tested against different scenarios. Risk analysis is incomplete if it doesn't include an analysis of the risks and opportunities associated with the transition plan against different scenarios and a baseline. Undertaking these as separate processes introduces inconsistencies and muddled thinking. That's why the NGFS is urging integration. One dataset, one governance structure, one strategy. Source: [1]  https://lnkd.in/ecxZHvnE [2] https://lnkd.in/ehRFZ4ka #ClimateRisk #TransitionPlans #ScenarioAnalysis #NetZeroFinance #NGFS #SustainableFinance ___________ 𝘗𝘭𝘦𝘢𝘴𝘦 𝘭𝘪𝘬𝘦 𝘰𝘳 𝘤𝘰𝘮𝘮𝘦𝘯𝘵 𝘰𝘯 𝘵𝘩𝘪𝘴 𝘱𝘰𝘴𝘵.  𝘍𝘰𝘭𝘭𝘰𝘸 𝘮𝘦 𝘰𝘯 𝘓𝘪𝘯𝘬𝘦𝘥𝘐𝘯: Scott Kelly

Most institutions now accept that climate risk is financially material. What’s less clear is how that materiality should translate into balance-sheet risk. For many banks and insurers the uncomfortable answer is: in capital and solvency assessments. Not as a new report. Not as a parallel exercise. But as part of how resilience is assessed under uncertainty. This is why ICAAP, ORSA, and climate risk are becoming inseparable. Climate risk doesn’t introduce a new objective. It challenges the assumptions embedded in existing frameworks. Assumptions about: the stability of cash flows the strength of business models the correlation between risks and the time horizon over which capital adequacy is assessed Which raises a difficult question for boards and risk committees: How far do we really need to go in integrating climate risk into capital and solvency assessments? In practice, integration is starting to mean things like: 🔹Using climate scenarios to test the resilience of capital and solvency positions 🔹Assessing whether existing buffers remain adequate under longer-dated, non-linear risks 🔹Understanding how climate risk amplifies traditional risks already captured in ICAAP or ORSA 🔹Being explicit about judgement, limitations, and uncertainty, rather than aiming for false precision What it does not mean: ▫️Producing a separate “climate ICAAP” or “climate ORSA” ▫️Treating climate scenarios as forecasts ▫️Relying solely on static metrics disconnected from decision-making ▫️Assuming current capital frameworks are automatically fit for purpose Boards are not expected to own the mechanics of capital modelling. But they are increasingly expected to: 🔹Understand how climate risk is reflected in capital and solvency thinking 🔹Challenge whether key assumptions remain valid 🔹Ask how management uses scenario results — not just whether they exist 🔹Ensure capital discussions reflect emerging sources of risk, not only historical experience In the next few years, the institutions that struggle most won’t be the ones with imperfect climate modelling. They’ll be the ones that can’t clearly explain: ▫️ how climate risk has been considered in capital and solvency assessments, ▫️ what judgements were made, ▫️ and how those judgements inform board-level decisions. For board members, CROs, and risk leaders: ▶️ What has been hardest when bringing climate risk into ICAAP or ORSA discussions? #ClimateRisk #RiskManagement #CapitalAdequacy

🌍 Definitely wonkish, but a must-read for anyone invested in our climate future! Before you dive in, I'd like to explain why this matters. This is about macroeconomic models—tools used to inform policymakers on crucial economic and environmental decisions. 📉💼 Now, all macroeconomic models have their flaws. They're built on assumptions and simplifications that help with forecasting or scenario analysis. 🔍 However, it becomes a major issue when certain assumptions or omissions misrepresent the problem they're analyzing. And that's exactly what's happening with the mainstream models often used for climate change and mitigation scenarios. 🚨🌡️ These models (called Environmental Dynamic Stochastic General Equilibrium (E-DSGE) models) aren’t just abstract concepts; they shape our climate action strategies. 🌎✨ This paper by Yannis Dafermos, Andrew McConnell. Maria Nikolaidi. Servaas Storm and Boyan Yanovski(👉 https://lnkd.in/es7a6ZjM) summarises the critique: 1️⃣ Banks as "Pure Intermediaries" is a Misrepresentation: E-DSGE models view banks merely as pass-through institutions of savings, not as money creators. This ignores banks' ability to create money and underestimates potential macro-financial instability (e.g., green booms or fossil-sector busts) from climate policies. 🌱💸 2️⃣ Demand’s Limited Role in Growth: In E-DSGE models, demand impacts the economy only in the short term. Long-term green investments are seen as costly to GDP since increased green spending must offset other demands. This makes net-zero goals look economically heavier than they might actually be. 📉🌍 3️⃣ Struggle with Disequilibrium & Climate Impact: Rational expectations in E-DSGE models mean agents are assumed to have near-perfect foresight. This framework limits understanding of short-term green investments or rising climate damages, ignoring critical disequilibrium effects in real markets. 📈⚠️ 4️⃣ Unrealistic Substitutability Assumptions: E-DSGE assumes fixed substitutability between fossil and green energy. This limits the capacity of green policies to phase out fossil energy fully and reduces the effectiveness of green monetary policies, even making them appear negligible. 💡🔋 5️⃣Downplaying Fiscal Policy Benefits: E-DSGE suggests green public investments "crowd out" private investments by raising interest rates or requiring higher taxes. This view downplays fiscal policy’s ability to complement private-sector green investments and reduce long-term climate risks. 🏗️🌱 6️⃣ Questionable "Optimal" Policies: E-DSGE models operate in a “second-best” world with inherent market distortions. This makes concepts like “optimal carbon tax” unclear—policies might improve welfare, but not necessarily. 🚫💰 See also a summary of these points below 👇 Better models (and common sense) can help us better than these models.

Regulators and banks are placing growing emphasis on the impact of climate risk on Expected Credit Loss (ECL). The PRA sets the tone with clear expectations: “Consider a broader range of downside climate scenarios, and climate-related variables, in the economic scenarios used in the ECL calculation, to allow for timely identification of borrowers and sectors more exposed to climate risk than the wider economy.” In practice, this guidance points to two main approaches: 1. Introduce dedicated climate-related scenarios and layer them into the ECL framework. 2. Embed climate risk into existing economic scenarios, adjusting key macro variables like GDP, inflation, and credit risk metrics to reflect physical and transition risks. The first route is conceptually elegant. Many banks, like HSBC, already assign probabilities to multiple macroeconomic scenarios and use them in weighted ECL calculations (see below). In that structure, adding a new climate scenario might seem straightforward (HSBC already have 2 Downside scenarios): assign it a probability and plug it in. But here’s the catch: Climate scenarios don’t come with probabilities. At least, not ones you can confidently defend. Climate narratives like “Orderly Transition,” “Disorderly Transition,” or “Hot House World” are plausible, not probable. Institutions have deliberately avoided attaching likelihoods, not due to lack of imagination, but to avoid misleading certainty about long-term, systemic outcomes. So, what about the second approach? If we modify existing macroeconomic scenarios to reflect climate risk, another problem emerges: Which climate assumptions go into which economic scenario? A recession can occur in a green transition, or in a world of extreme weather. Mapping climate futures onto economic narratives becomes an exercise in judgment—and the risk is, climate drivers get blurred into general macro assumptions and lose visibility. We might treat economic and climate scenarios as separate dimensions, and model them using joint probabilities. However we hit the same wall: You still need probabilities for each climate state—the very numbers we’re hesitant to define. So where does this leave us? Well, work in progress..