Climate risk is financial risk

This article is part of a series. You're reading part one.

Markets rarely collapse because risks suddenly appear. They collapse because risks remain invisible — embedded in asset prices, lending models, and regulatory assumptions until the gap between perceived value and reality can no longer be sustained. Climate change has entered that gap.

I.  The carbon mispricing problem

At the centre of this issue lies a fundamental distortion: carbon remains one of the most inconsistently priced externalities in the global economy, and the dispersion has not narrowed since the Stern Review first quantified it nearly two decades ago. It has widened and fractured across jurisdictions, asset classes, and time horizons in ways that directly distort capital allocation.

Consider the present landscape. In the European Union's Emissions Trading System, allowances have traded in a range of roughly €60 to €100 per tonne in recent years — itself a figure significantly below the levels at which comprehensive decarbonisation becomes economically rational across the full range of industrial sectors. In voluntary carbon markets, prices routinely sit between $3 and $20 per tonne, despite persistent and well-documented concerns about the environmental integrity of the underlying credits. In jurisdictions without carbon pricing — still the global majority — the effective price remains zero. Where fossil fuel subsidies persist, the implicit price of carbon is structurally negative. The International Monetary Fund estimates that total fossil fuel subsidies, including implicit subsidies from underpriced environmental damage, exceeded $7 trillion globally in 2022.¹

These are not marginal discrepancies. They represent a systematic failure of the price signal upon which capital allocation depends.

The consequences are already embedded in investment portfolios. Long-lived energy infrastructure — pipelines, LNG terminals, power generation assets, industrial facilities — was financed on assumptions about future energy costs and regulatory frameworks that are increasingly inconsistent with plausible policy trajectories. The question is no longer whether long-duration fossil assets will be repriced. It is whether the financial system can absorb the correction without transmitting it as a crisis.

The question is no longer whether long-duration fossil assets will be repriced. It is whether the financial system can absorb the correction without transmitting it as a crisis.

II.  Physical risk, transition risk — and the policy dilemma between them

Climate-related financial risk operates through several distinct but interconnected channels. Conflating them produces analytical confusion; separating them clarifies where systemic pressure is building and why no policy pathway eliminates the exposure.

Physical risk

Physical risk arises from the direct economic consequences of climate change itself: more frequent and severe weather events, coastal flooding, agricultural disruption, and the progressive degradation of infrastructure and collateral value. The Bank of England's 2021 Climate Biennial Exploratory Scenario estimated potential additional losses in the hundreds of billions for UK-exposed financial institutions under a delayed-transition pathway, before accounting for second-order sovereign and macroeconomic effects. These estimates are conservative; they do not capture non-linear threshold effects or cross-border contagion.²

Transition risk

Transition risk is more structurally complex, because it is endogenous to policy, technology, and market expectations rather than to atmospheric physics. It arises when the shift to a low-carbon economy — through carbon pricing, regulation, technological disruption, or changes in investor preference — causes a repricing of existing assets and revenue streams across energy-intensive sectors.

Liability risk

Liability risk is the least developed of the three in financial modelling terms, but its trajectory is becoming clearer. Litigation establishing causal links between corporate emissions and quantifiable climate damage — of which Lliuya v RWE in the German courts is the most closely watched precedent — creates the possibility that corporate balance sheets will face climate-related legal claims not currently reflected in valuation frameworks. As jurisprudence develops, liability risk will increasingly interact with both transition and physical risk channels.

The policy dilemma

The interaction between physical and transition risk creates a dilemma with no clean resolution. Delayed climate policy reduces transition risk in the short term by allowing gradual adjustment, but allows physical risk to accumulate at an accelerating rate. Abrupt policy action constrains long-run physical damage but compresses the timeframe for asset repricing, amplifying transition risk across exposed sectors simultaneously. No pathway eliminates both. The question is which risk profile the financial system is better equipped to absorb — and current evidence suggests the answer is neither.

III.  How climate risk enters the financial system

Climate risk does not appear on balance sheets as a discrete line item. It enters the financial system through a chain of economic mechanisms, each involving delay, opacity, and potential amplification.

The sequence typically runs as follows. Policy change or technological disruption alters the relative economics of energy production. This shifts corporate revenue and cost structures in energy-intensive sectors. Asset valuations adjust — sometimes gradually, sometimes abruptly as forward expectations reprice. Balance sheet stress transmits to banks and capital markets through deteriorating credit quality and falling collateral values. In a concentrated stress scenario, that pressure propagates through interbank and sovereign credit markets.

What makes this transmission mechanism particularly difficult for conventional risk frameworks to capture is a structural feature: the lags involved are long, non-linear, and jurisdiction-dependent, but the corrections, when they arrive, can be rapid and correlated across sectors simultaneously. That is precisely the combination that stress-testing models calibrated to historical data are least equipped to handle. Historical data, almost by definition, contains no analogue for a global energy system transition of this scale and speed.

Economists refer to the asset-level manifestation of this dynamic as stranded asset risk. But the systemic implications extend well beyond individual firms. If repricing occurs simultaneously across energy, utilities, industrials, transportation, and real estate — as a coordinated or cascading policy shift could produce — the effects would propagate through entire financial systems, not merely through the balance sheets of directly exposed companies.

There is also an information asymmetry embedded in the transmission chain. Companies with assets most exposed to transition risk are not uniformly incentivised to disclose that exposure in granular terms. Voluntary disclosure frameworks improve the quality of available information but produce strategically curated outputs. Investors and lenders are therefore pricing risk on the basis of data that is structurally incomplete — and the gaps are not random noise but, in many cases, systematic understatement.

IV.  Why the financial system remains structurally unprepared

It is tempting to attribute the mispricing of climate risk primarily to an information problem. Firms may lack adequate data, disclosure standards remain inconsistent, and long-horizon climate modelling involves genuine uncertainty. Information is a real constraint. But it is not the primary explanation.

The deeper problem is structural, and it operates at three levels.

The first is prudential capital regulation. The Basel III framework — the core architecture governing bank risk management globally — contains no explicit climate risk module. Capital requirements for loans to carbon-intensive industries are not systematically differentiated from capital requirements for loans to low-carbon ones, except where underlying credit models already reflect the valuation difference. Where asset prices have not yet adjusted — which is precisely the condition that creates systemic risk — regulatory capital does not reflect the embedded exposure. The consequence is direct: without a mechanism linking carbon exposure to prudential capital charges, financial institutions have no structural incentive to price transition risk into their lending and underwriting decisions, regardless of what TCFD disclosures may indicate.

The second is energy market design. Wholesale electricity markets in most jurisdictions still operate on a marginal-cost clearing mechanism developed for fossil-fuel generation, in which thermal plant — with high variable costs and relatively predictable output — dominated the merit order. Renewable energy has the opposite cost structure: high upfront capital costs and near-zero marginal operating costs. As renewable capacity expands, traditional market designs increasingly produce problematic signals — price suppression during periods of high renewable output, followed by sharp price spikes during periods of low renewable availability. The resulting revenue uncertainty increases perceived risk for long-duration infrastructure financing and raises the cost of capital for the very projects the energy transition requires. Market design is not a technical detail; it is part of the financial infrastructure that determines whether the transition can be funded efficiently.

The third is insurance market integrity. In some of the highest-risk zones — parts of coastal Florida, wildfire-exposed counties in California, flood-prone regions of Central Europe — private insurers are withdrawing from the market. Where private insurance retreats, public balance sheets assume the residual risk, often without the fiscal capacity to bear it transparently. The socialisation of unpriced climate risk is already underway. It is not yet visible in sovereign credit spreads, but the fiscal logic of eventual repricing is entirely consistent with financial history.

Without a mechanism linking carbon exposure to prudential capital charges, financial institutions have no structural incentive to price transition risk accurately — regardless of what TCFD disclosures indicate.

V.  Regulators are catching up — but not fast enough

Central banks and prudential regulators have recognised the problem with increasing clarity over the past five years. The Network for Greening the Financial System, whose membership now includes the majority of systemically significant central banks and supervisors, has developed climate scenario analysis frameworks and published guidance on supervisory expectations for transition risk. The European Central Bank has conducted dedicated climate stress tests of major euro area banks. The Bank of England has embedded climate risk within its broader financial stability mandate.

These are significant institutional developments. They represent an explicit acknowledgement, by institutions with genuine systemic authority, that climate risk is a macro-financial stability concern rather than merely a matter of voluntary sustainability disclosure.

They are also, by the regulators' own assessment, insufficient. The ECB's 2022 climate stress test found that approximately two-thirds of bank income was concentrated in sectors with elevated climate risk — and that most institutions still lacked the granular data or modelling capacity to assess their own exposures with adequate precision.³ Scenario analysis remains a supervisory exercise. It has not translated into binding capital requirements.

The gap between regulatory aspiration and regulatory capacity is widest precisely where it matters most: in the calibration of capital charges to climate-related risk. Until scenario outputs drive capital allocation decisions rather than qualitative risk narratives, the incentive structure facing financial institutions remains fundamentally unchanged.

VI.  The 2008 analogy — instructive but insufficient

Comparisons between climate-related financial risk and the global financial crisis of 2007–2008 are frequently invoked and may be genuinely instructive. Both situations involve the systematic mispricing of risk embedded in long-duration assets, obscured by complexity, and dependent on a correction that the financial system is not structured to manage gradually. In both cases, the gap between perceived value and underlying reality was allowed to widen precisely because the institutions most exposed were also those with the least incentive to close it.

But the analogy requires qualification, and the qualification matters for policy. The 2008 crisis was triggered by a relatively concentrated repricing event in the US residential mortgage market, with a clear and rapid transmission mechanism through structured credit products. The correction was acute, visible, and politically legible — which is precisely why it generated a substantial, if ultimately incomplete, regulatory response.

Climate-related risk has a different temporal and spatial structure. It involves multiple potential repricing events across different sectors, geographies, and asset classes, unfolding over a longer but uncertain time horizon, with non-linear tipping points that are difficult to identify in advance. It resembles less a discrete asset bubble and more a slow-moving structural adjustment that accelerates unpredictably.

That distinction has a crucial political economy implication. Acute financial crises create reform windows — moments of sufficient institutional disruption that fundamental changes to regulatory architecture become politically possible. Slow-moving structural risks tend to suppress that urgency. They permit the rationalisation that adjustment is occurring, that markets are adapting, that the correction, if it comes, will be manageable. Financial history suggests this rationalisation is frequently wrong. The gradual nature of the transition may delay policy responses until the underlying adjustment becomes abrupt — and by that point, the window for orderly adaptation may have closed.

VII.  The financial architecture of the transition

The energy transition is routinely described as an engineering challenge — building renewable generation capacity, electrifying transportation, and decarbonising industrial processes. It is equally a financial architecture challenge, and arguably a more difficult one, because the financial systems through which the transition must be funded were designed for the world the transition is replacing.

Capital must move at unprecedented scale into new infrastructure while legacy assets decline in value across a compressed timeframe. Achieving that shift without systemic disruption requires financial markets capable of pricing risk accurately, allocating capital efficiently, and transmitting stable revenue signals to long-horizon investors. None of those conditions is currently met in full.

Adequate adaptation implies at minimum three structural changes. First, carbon pricing must become more consistent, predictable, and comprehensive across investment horizons. Infrastructure decisions that commit capital for twenty to thirty years cannot be made rationally on the basis of price signals that vary by an order of magnitude across jurisdictions and fluctuate with political cycles. Second, financial regulation must incorporate climate risk into prudential capital frameworks — not merely into disclosure regimes. Disclosure without capital consequences improves information but does not change incentives. Third, energy market design must evolve to provide stable revenue frameworks for low-carbon infrastructure investment, replacing mechanisms that produce the volatility and uncertainty most corrosive to long-duration project financing.

These are not incremental reforms. They require deliberate institutional redesign across regulatory architecture, market structure, and international policy coordination. The scale of that challenge should not be minimised.

Conclusion

Financial history does not offer many examples of large-scale, systematically mispriced risks that resolved themselves without a significant correction event. Railway booms, dot-com equities, and mortgage-backed securities each followed a recognisable pattern: price signals diverged from underlying fundamentals, capital continued to flow against emerging realities, and adjustment eventually occurred at a pace and scale that the financial system was not prepared to absorb.

The energy transition is not a speculative bubble. It is a structural transformation driven simultaneously by atmospheric physics, technological change, and policy intervention — three forces that do not wait for financial markets to be ready. The financial architecture through which that transformation must be financed retains, in critical respects, the assumptions, incentive structures, and regulatory frameworks of the world it is replacing.

The capital required to finance the transition exists. The central question is whether the financial system can evolve its pricing mechanisms, prudential frameworks, and market structures quickly enough to direct that capital efficiently — and to absorb the repricing of legacy assets without amplifying it into systemic instability.

Understanding that challenge in its structural detail is the purpose of the series that follows.

illuminem Voices is a democratic space presenting the thoughts and opinions of leading Sustainability & Energy writers, their opinions do not necessarily represent those of illuminem.

See how the companies in your sector perform on sustainability. On illuminem’s Data Hub™, access emissions data, ESG performance, and climate commitments for thousands of industrial players across the globe.

Notes

1. IMF fossil fuel subsidy estimate: Gaspar et al., 'Fiscal Monitor: Climate Crossroads', IMF, April 2023. The $7 trillion figure includes both explicit subsidies and implicit subsidies from underpriced externalities including local air pollution and climate damage.

2. Bank of England, 'Results of the 2021 Climate Biennial Exploratory Scenario', May 2022. Estimates are scenario-dependent and represent illustrative stress outcomes rather than central forecasts.

3. European Central Bank, 'ECB economy-wide climate stress test', Occasional Paper Series No. 281, September 2021; ECB Banking Supervision, 'Walking the talk: Banks gearing up to manage climate and environmental risks', November 2022.

4. Lliuya v RWE AG, case pending before the Oberlandesgericht Hamm. The case asserts that RWE bears proportional liability for flood protection costs in Huaraz, Peru, based on its historical share of global greenhouse gas emissions.

5. NGFS scenario framework: Network for Greening the Financial System, 'NGFS Climate Scenarios for central banks and supervisors', updated June 2023. The NGFS currently comprises over 130 member institutions.