The intensifying impact of climate change on the financial system

Introduction

Climate-related financial risks (so-called climate risk), defined as the financial implications of climate change and the sustainable transition to mitigate global warming, are now widely recognised within the financial community. The development of international networks such as the Network of Central Banks and Supervisors for Greening the Financial System (NGFS), which now consists of 152 members across over 95 countries, highlights the importance of collective work to improve understanding of the financial impacts of climate change. International collaboration is crucial as the climate science community continues to warn of the devastating impact of global warming exceeding the 1.5°C limit.

Global warming has accelerated in recent years.^[1] Many countries are likely to miss their next climate objectives,^[2] increasing both transition risk from an abrupt transition and physical risk from a warmer global temperature trajectory. Climate risks are difficult to quantify because of the nature of climate change – its unprecedented nature, a lack of data, and complex mechanisms. Yet, climate change is already happening. Its intensification means that supervisors must act while acknowledging that current data may underestimate the real risk. The European Central Bank (ECB) has highlighted a paradox: despite improved data and a clearer understanding of how climate change drives economic and financial losses – thanks to research on climate transmission channels – policies to support the sustainable transition remain inadequate to address future risks.^[3] In this context, the role of financial supervisors is to ensure that the financial sector remains resilient to climate risks that are underway.

This paper relies on numerous studies published in recent years demonstrating that climate risk already affects the value of financial and non-financial assets. The increasing trend in both physical and transition risk drivers implies that the associated financial risks are growing, calling for risk management measures at both the institutional and systemic levels.

I  Transmission channels between climate-related events and financial risks

First, financial institutions must understand how the current climate crisis translates into financial risks. These risks materialise through different transmission channels that seek to explain the financial consequences of events and socio-economic changes associated with climate change and transition measures such as disruptive technology, policy change, natural disasters or lasting climate effects. These transmission channels have been described by several international organisations, such as the NGFS (Figure 1) and the Bank for International Settlements (BIS).^[4]

Figure 1: Transmission channels between climate risks and financial risks^[5]

The NGFS has developed several sets of climate scenarios to translate different assumptions about climate change-related events into economic and financial impacts, helping supervisors and policymakers anticipate the repercussions of climate change on the economy and financial system. While some limitations remain due to the complexity and nascent nature of this work,^[6] climate scenarios improve understanding of how the economy may react to an orderly transition, a delayed transition or a hot-house world (i.e. a scenario in which climate change remains unmitigated).

The Financial Stability Board (FSB) has developed a framework for assessing climate-related vulnerabilities, representing the different interactions between climate-related factors and the non-financial and financial sectors.^[7] It highlights the transmission channels and possible amplification effects through which climate change could affect the financial system. Within this framework, the FSB has defined three categories of climate risk metrics that should be monitored:

• Proxies: “provide an early signal on potential drivers of transition and physical risks. They include information on the likelihood and severity of hazards or potential alignment gaps of [greenhouse gas] GHG emissions between projected and reference transition paths of jurisdictions.”
• Exposure metrics: “provide insights on how climate risk drivers identified by proxies could transmit through the financial system and affect different sectors (financial and non-financial). These metrics build on a combination of climate-specific, non-financial, and financial-sector information.”
• Risk metrics: “build on the information contained in proxies and exposure metrics to quantify the scale of financial impacts as climate shocks transmit through the financial system by interacting with vulnerabilities.”^[8]

II  Early signals of climate risk

The early signals identified by the FSB include trends in absolute emissions and carbon prices (transition risk), and economic losses resulting from physical climate hazards (physical risk). This paper proposes to extend monitoring to forward-looking metrics such as the projected deviation from climate objectives for transition risk and the projected trajectory of warming for physical risk, in order to capture trends not contained in historical data.

A.  Physical risk

Economic losses from climate change-related physical hazards are a useful indicator of climate risk. Higher losses could lead to an increase in the probability of default (PD) of companies operating in the affected region or a deterioration of collateral value. Both factors increase credit risk for financial institutions. Such losses are increasing worldwide. In the EU, average damage from natural disasters has increased by more than 50% over the last 15 years,^[9] while the average damage in the United States has doubled over the same period.^[10] These indicators align with recent research literature^[11] showing that climate change is already having substantial economic effects, which are expected to intensify in the coming years: 1°C in global warming was found to have a negative cumulative effect of more than 20% on world GDP, while a 2026 ECB working paper also highlights that short-term economic risks due to extreme weather events will affect countries and sectors differently. For example, among the top European economies, Germany appears to be the most economically sensitive to heatwaves, with one day of extreme heat reducing GDP by up to 0.4%.^[12]

Additionally, climate scientists anticipate a potentially strong El Niño event in the second half of 2026.^[13] This may increase rainfalls in South America, the United States, and central Asia, while causing droughts in Australia, East Africa, Indonesia, and parts of southern Asia. The phenomenon is expected to amplify the effects of global warming in the coming year, potentially leading to greater economic impacts in 2027.

Finally, the International Association of Insurance Supervisors (IAIS) warns that the majority of economic losses from natural catastrophes remain uninsured and that this protection gap could widen in the coming years due to the increased frequency and severity of physical events. In 2024, 57% of those losses worldwide were uninsured.^[14] A widening of insurance protection gaps could increase financial systemic risk by shifting more risks to the banking sector through the loss of collateral value and business interruptions.

B.  Transition risk

The objective of the Paris Agreement to limit global warming to 1.5°C is likely to be missed in the coming years. In 2015, when the Paris Agreement was signed, a linear extrapolation suggested that the world would reach 1.5°C of warming (as a 30-year average) by 2042. However, applying the same methodology just a decade later, in December 2025, projected that this threshold could be crossed as early as 2029, 13 years earlier than expected.^[15] This represents an important early signal of a possible abrupt and disorderly transition and the associated transition risks, as governments may face more social pressure to implement ambitious policies to keep the effects of global warming on their constituencies manageable.

In addition, transition risks are likely to be amplified as more drastic efforts will be required to realign with a safer warming path. The further a jurisdiction deviates from its climate objective, the more abrupt the economic adjustments will need to be, triggering sharp readjustments of financial and non-financial asset values. In its Emissions Gap Report from November 2025^[16], the United Nations states that only “seven G20 members are on track to achieving their [nationally determined contributions] NDC targets, but few are on a clear trajectory towards their net-zero emission pledges”. This global deviation from a lower-risk trajectory could lead to the materialisation of transition risk in the short term.

III.  Materialisation of climate risk

Numerous studies have shown that climate risk has significant financial impacts,^[17] which are likely to increase in the coming years. For example, in 2025, the NGFS released a set of short-term scenarios simulating the economic and financial impacts of different combinations and magnitudes of transition and physical risks over the next 5 years.^[18]

The evidence referred to in the ensuing sections shows that climate-related events, whether from natural disasters or transition-related economic transformations, could translate into financial risks. While some studies currently report only modest impacts on traditional risk metrics, this should not be interpreted as evidence of low materiality. As global warming accelerates, historical data cannot capture the full scale and severity of future climate shocks. Financial supervisors must account for this uncertainty and, given the systemic dimension of climate change, should deploy additional macroprudential tools to ensure the stability of the financial system.

A.  Credit risk

The materialisation of climate risk via credit risk is a key transmission channel for the financial sector. Sudden shifts in economic conditions – driven by political decisions or the growing frequency and intensity of natural disasters – could increase default rates in banks’ portfolios. Additionally, collateral used as security for credit positions may lose value, amplifying losses for the financial institutions in case of a counterparty default.

A 2023 study demonstrated the link between one or multiple natural disasters and an increase in non-performing loans, highlighting the impact of physical risk on PDs.^[20] This effect is particularly significant for Latin America and the Caribbean, South Asia, and Sub-Saharan Africa. Moreover, European banks are starting to charge a transition risk premium for loans to counterparties that have not committed to reducing their emissions, indicating that transition risk impacts counterparty credit risk profiles.^[21] However, this premium remains small and has been identified primarily by banks that have set emission reduction targets for their portfolios. A survey conducted by the United Nations in 2025 highlighted that 61% of banks include climate risk in their credit risk management through PD modelling for example, but these efforts are “largely in the early, exploratory stages”.^[22]

In its 2025 short-term climate scenarios, the NGFS includes a climate credit risk model, CLIMACRED,^[23] which estimates the probability of default and firm costs of capital adjustments under a specific climate scenario. This advancement enables the assessment of credit risk impacts at a highly granular geographical and sectoral level. It shows that, within the next five years, PDs could increase significantly for carbon-intensive sectors (mainly coal, oil and gas production), with absolute increases compared to the NGFS baseline between 8% and 35% under both orderly and delayed transition scenarios. Applying this methodology to the Chilean banks reveals material impacts for the banking as a whole.^[24] This echoes the work done in the EU, which shows that PDs of half of firms in carbon-intensive sectors could more than double in the coming years if transition risk materialises.^[25] Regarding physical risk, the NGFS short-term scenarios also show higher credit risk for sectors such as agriculture, construction and energy production, with absolute increases in PD ranging from 5% to 25% relative to the NGFS baseline. For example, material losses from physical risk have been highlighted in the Uruguayan banking sector, reflecting its high exposure to agricultural activities.^[26]

B.  Market risk

Climate risk will not only materialise through higher credit risk, but also through market risk. If climate change disruptions intensify, sudden repricing could occur following climate events that expose assets’ vulnerabilities, impacting future cash flows. As market prices depend on participants’ expectation of future values, even medium to long-term climate events, such as the phase-out of fossil fuel activities and climate tipping points, may have an impact on today’s valuations.^[27]

A broad literature review shows that both physical and transition risks affect equity market prices, with transition risks having a stronger impact.^[28] Policy and regulatory events constitute new, lasting information, triggering the repricing of expected cash flows, causing market prices to adjust. Such adjustments depend on the sector of activity and nature of the information. For instance, the Paris Agreement led to a decline of fossil fuel equity returns by 10% to 15% over the three-month negotiation period, whereas Trump’s election resulted in equity returns in the sustainable energy sectors decreasing by 8% over the two days following the election.^[29] Physical risk impacts on equity prices are more heterogeneous and tend to be less pronounced. Existing studies primarily focus on direct impacts in specific geographical areas, potentially underestimating the broader effects of the simultaneous materialisation of natural disasters worldwide, including second-round and spillover effects.

Transition risk events and natural catastrophes may increasingly impact equity prices within a short-term horizon. A study by the Banque de France published in April 2026,^[30] based on the NGFS short-term scenario, shows that policy changes associated with an abrupt transition would primarily affect fossil fuel sector equity values, whereas other sectors may be less affected. On the physical risk side, dry events such as droughts, heatwaves and wildfires would mainly impact sectors such as agriculture and construction, while wet events such as floods and storms would negatively impact all sectors. These potential losses could reach EUR 33 billion under an abrupt transition scenario, and EUR 196 billion under a climate disaster scenario, significant for the French financial sector.

Climate risk also impacts bond prices by driving interest rates up. Due to its impact on agricultural and industrial productivity,^[31] global warming has been shown to affect inflation rates in many countries. This effect may be stronger in the short term and in low-income countries.^[32] Moreover, financial actors are beginning to price transition risk through higher sovereign yields for countries deviating from a net-zero trajectory.^[33] If countries continue, or increase, their deviation from climate objectives, sovereign bond markets could experience greater volatility than expected.

C.  Liquidity risk

In the short term, liquidity risk is a critical issue for banks. The failure of the Silicon Valley Bank in 2023 is a recent example. This part of the paper highlights the impact of climate factors on bank liquidity risk.

On the liability side, banks may face higher deposit withdrawals as businesses and households use their savings to fund transition needs, such as investments in emissions reductions or home renovations, or to recover from asset destruction following natural disasters. On the asset side, banks rely on their high-quality liquid assets (HQLAs) to access short-term liquidity. These assets can be easily sold or used in repurchase agreements (repo) to cover banks’ short-term liquidity needs. A recent study in India found a significant relationship between banks’ HQLA level and climate events,^[34] suggesting that increases in non-performing loans due to climate risk, associated with an increase in depositor withdrawals, lead to a reduction in HQLAs. Banks may sell these assets to access liquidity and sustain their activity if persistent adverse climate conditions (e.g. rising temperatures, changing precipitation patterns, increased disaster frequency) continue to put pressure on their counterparties and asset valuations, increasing their liquidity risk.

Moreover, as the understanding and the disclosure of banks’ climate risk improve, investors have started to account for it when assessing banks’ risk profile. This translates into higher repo rates for banks exposed to higher transition risk, increasing the funding cost for a bank to access short-term liquidity.^[35] The consideration of climate risk in the repo market could amplify the impact of other financial risk factors on bank liquidity. This compounding effect is likely to be currently underestimated in the financial sector.

D.  Compound risk

Compound risk refers to the amplified impact of multiple separate shocks occurring simultaneously or in succession, resulting in financial losses greater than the sum of their individual effects. The importance of considering compound risk in the context of climate risks has been highlighted by the NGFS.^[36] It can arise from the interaction of multiple successive climate events impacting the same area, or from the combination of climate-related and non-climate-related shocks. Understanding of compound risk remains a nascent field of risk management, and its quantification is still in its infancy, potentially leading to a massive underestimation of the financial impacts from climate risk.

Climate risk is materialising in a world already stressed by several crises. Recent examples include interactions between climate events and the COVID-19 pandemic,^[37] as well as increasing geopolitical tensions (e.g. war in Ukraine, Middle East conflicts). Geopolitical risks are currently disrupting energy supply chains, weighing on the global economy and leading to higher inflation rates.^[38] Climate risk could amplify these pressures through natural disasters that further increase supply chain pressure. For example, the conflict around the Strait of Hormuz threatens global food prices given its importance in the fertilizer supply chain.^[39] At the same time, global warming, boosted by El Niño conditions, increases the risk of global food insecurity, potentially creating systemic risks for the global economy and financial sector.^[40]

Conclusion

Increasing economic losses resulting from climate change, together with many countries’ current deviation from a Paris-compatible transition trajectory, should act as an early warning for financial supervisors to move from risk exploration to effective risk management measures.

The development and deployment of climate scenario analysis, based on transition plans and stress tests, can only partially address climate risk in the financial sector. As climate change will affect the financial sector as a whole, relying solely on a microprudential approach will not match the magnitude and severity of looming risks.

Microprudential tools must be complemented by macroprudential measures addressing the systemic dimension of climate risk and accounting for financial institutions’ interconnectedness, compound effects and the current underassessment of climate risk resulting from the limited maturity of climate scenario analysis. The remaining uncertainty surrounding climate risk should not justify a wait-and-see approach by financial supervisors. The warning signs are explicit, and proactive measures are imperative.

Thomas Larible, Research & Advocacy Officer at Finance Watch

[email protected]

+32 2 880 04 38