Why mutualisation of risk may not be possible in a world of systemic perils

For centuries, the UK insurance industry has thrived by mutualising risk, pooling premiums from diverse policyholders to cover losses from infrequent, localised natural disasters like floods or storms. This model assumes that risks are episodic and geographically uncorrelated, allowing insurers to spread liabilities across a broad base while maintaining profitability. However, climate change seems to be upending this framework. Extreme weather events, whether its floods, droughts, coastal erosion, or wildfires, are unquestionay becoming more frequent, severe, and systemic, creating interconnected perils that challenge the very foundation of risk mutualisation. Based on this thesis, I want to explore why the mutualisation of risk is increasingly incompatible with today’s climate-driven world and how the UK insurance industry might adapt to survive.

The traditional insurance model: mutualisation of episodic risks

The UK’s property and casualty (P&C) insurance sector relies on mutualisation to manage risks from natural catastrophes. Insurers collect premiums from a large pool of policyholders, using sophisticated catastrophe models to estimate the likelihood and impact of events like floods, based on historical data and exposure details (e.g. property locations and values). These premiums cover expected claims while building reserves for rare, high-impact events. Reinsurance, insurance for insurers, further disperses risk globally, with firms like Lloyd’s of London playing a pivotal role. Public-private partnerships, such as the Flood Re scheme launched in 2016, exemplify mutualisation by pooling premiums to subsidise flood insurance for high-risk properties, covering approximately 350,000 homes, where the industry has struggled to remain involved in a particular sector of the economy.

Historically, this model has succeeded because natural disasters were episodic and geographically diverse. Losses from a flood in one region could be offset by premiums from unaffected areas, and reinsurers could diversify risks across uncorrelated perils, such as UK floods and Japanese earthquakes. After major events, insurers often raise premiums to recoup losses, boosting short-term profitability. For example, UK insurers paid out £573 million for weather-related home insurance claims in 2023, a 36% increase from £421 million in 2022, driven by storms like Babet and Ciaran, with average claim costs rising 11% to £6,235. In 2024, this figure rose to £585 million. This ability to adjust premiums and rely on mutualised risk has kept the industry viable, but the systemic nature of climate-driven perils must now be undermining this approach.

Why mutualisation is incompatible with systemic climate risks

The mutualisation of risk depends on three key assumptions: risks are relatively rare, geographically independent, and predictable based on historical patterns. Climate change is eroding all three, making the traditional insurance model increasingly untenable.

1. Increased frequency and severity of events

Climate change is driving a surge in the frequency and severity of extreme weather events, overwhelming the actuarial assumptions of mutualisation. We've seen a 30% rise in global flood losses over the past decade, with models projecting increased flood volumes across riverine, rainfall-driven, and coastal surge events in a warming world - and it's only going to get worse. London seems particularly vulnerable, with 13% of properties face high or medium flood risk. Wildfires, once rare in the UK, are also escalating, with Europe seeing 370,000 hectares burned in 2024.

Drought, often overshadowed by floods, is a growing systemic threat too. A recent article in the Conversation notes that England faces a water crisis, with half its water companies projecting supply deficits by 2035 under a 4°C warming scenario. Droughts in 2022 and 2018 strained reservoirs, particularly in southeast England, where population growth and outdated infrastructure exacerbate shortages. The northwest of England has already declared a drought situation this year. Unlike floods, which cause immediate property damage, droughts create prolonged economic losses, disrupting agriculture, energy, and water-intensive industries. These losses are harder to mutualise due to their diffuse, long-term nature, yet they compound systemic risks when followed by floods or wildfires, as dry soils increase runoff and flood severity.

The sheer frequency of these events reduces the time insurers have to rebuild reserves between claims, straining the mutualisation model’s reliance on infrequent payouts. A warming world is a wetter world for floods, but droughts reflect the flip side: intensified drying over land. This dual challenge creates a cycle of losses that mutualisation struggles to absorb.

2. Correlated and systemic risks

Mutualisation assumes risks are uncorrelated, allowing insurers to offset losses in one region with premiums from others. Systemic climate risks, however, are increasingly correlated, both geographically and temporally. For example, a drought can dry out soils, increasing runoff during subsequent heavy rains, leading to severe flooding. Similarly, wildfires can destabilise landscapes, exacerbating mudslides or floods. The 2011 Thai floods, the largest insured flood loss on record, demonstrated this when interconnected river systems inundated manufacturing hubs, causing global supply chain disruptions and plummeting share prices for chip manufacturers.

In the UK, systemic risks are evident in regions like southeast England, where drought-induced water shortages could coincide with floods or coastal erosion. The Conversation article also highlighted how ageing water infrastructure and over-reliance on reservoirs fail to address these compounding risks, leaving sectors like agriculture vulnerable. There are warnings that under high-emission scenarios like SSP 8.5, parts of the planet could become uninhabitable by century’s end, implying widespread losses that defy geographic diversification. When risks are correlated, mutualisation fails because claims occur simultaneously, depleting reserves and overwhelming reinsurance capacity.

3. Unpredictability and historical data limitations

Mutualisation relies on historical data to predict future risks, but climate change is rendering past patterns obsolete. All of a sudden, places with no recent flood history may be highly vulnerable due to a lack of preparedness. For instance, areas further north on the US eastern seaboard, unaccustomed to hurricanes, face growing risks due to a poleward shift in storm tracks, a trend relevant to the UK’s coastal regions as well. Droughts, too, are becoming harder to predict, as the Conversation article notes that climate change disrupts traditional water supply patterns, with England’s reservoirs struggling to cope with prolonged dry spells.

Of course there's always a danger of false precision in high-resolution climate models, as smaller-scale projections carry high uncertainty. But at the same time, this unpredictability undermines insurers’ ability to price risks accurately, a critical component of mutualisation. Without reliable data, premiums may either be insufficient to cover losses or so high that they deter policyholders, shrinking the risk pool and weakening the mutualisation framework.

4. Socioeconomic and market dynamics

Mutualisation assumes a broad, diverse pool of policyholders, but climate change exacerbates socioeconomic vulnerabilities that fragment this pool. Areas most vulnerable to floods or droughts, sadly often emerging markets or regions with high poverty, face a vicious cycle of limited capital access and heightened climate risk. Drought-related water restrictions disproportionately affect low-income households too, reducing their ability to afford insurance. As premiums rise to cover systemic losses, coverage becomes unaffordable, driving policyholders out of the market and shrinking the risk pool.

Moreover, insurers are increasingly retreating from high-risk areas, creating insurance deserts. This leads to a protection gap where insurers avoid unprofitable regions/areas. In the UK, Flood Re mitigates this for floods, but droughts, wildfires, and coastal erosion lack similar schemes. This means that drought-vulnerable sectors like agriculture face coverage gaps, as insurers struggle to mutualise long-term, diffuse losses. Globally, insurers have exited markets like California (wildfires) and Florida (hurricanes), and the UK could see similar withdrawals in water-stressed or coastal areas, further eroding the mutualisation model.

Cascading impacts on financial markets

The breakdown of risk mutualisation has broader implications for financial markets. There must be unpriced climate risks in real estate, central to debt markets like mortgage-backed and municipal bonds, that could trigger significant revaluations. If drought or flood insurance becomes unaffordable, property values could plummet, affecting mortgage viability and bond ratings. In a worst case scenario, drought-induced losses in agriculture or industry could destabilise local economies, reducing property values and tax revenues, which underpin municipal bonds. In socioeconomically vulnerable areas, these impacts are magnified, exacerbating inequalities and market instability.

I have to believe that financial actors are trying to grapple with these risks, using company facility data to assess how floods and water scarcity affect corporate performance, but I worry that as a general topic, it's not well understood. Droughts, for instance, disrupt water-intensive sectors like agriculture, energy, and manufacturing, affecting supply chains and profitability. However, the uncertainty in climate projections complicates these assessments, delaying market adjustments until losses materialise.

Adapting to a systemic risk landscape

Despite these challenges, the UK insurance industry can adapt to navigate systemic risks, though mutualisation alone is insufficient. Several strategies to my way of thinking seem to offer pathways forward:

1. Innovative insurance products, like parametric insurance: paying out based on predefined triggers (e.g., rainfall deficits for droughts or excess for floods) reduces costs and speeds up claims. While more common in developing markets, it’s gaining traction in the UK and could address diffuse risks like drought; al microinsurance: affordable, limited-coverage policies could protect vulnerable sectors like agriculture, maintaining market presence in high-risk areas.
2. Advanced risk modelling: there have been noticeable advances in flood modelling using LIDAR, satellite data, and AI (captured in the EA's most recent release). Similar tools could model drought risks, integrating soil moisture, water supply, and demand data to improve underwriting accuracy. We should advocate for precision irrigation and water efficiency, which insurers could incentivise through discounts.
3. Public-private partnerships: expanding Flood Re to cover droughts, wildfires, or coastal erosion could stabilise markets. The Conversation article emphasises demand management, reducing leakage and promoting water efficiency, as critical for drought resilience, which insurers could support through risk-mitigation incentives. The Association of British Insurers (ABI) has called for sustained government investment in flood and water infrastructure.
4. Adaptation opportunities: opportunities abound in adaptation, such as investing in drought forecasting, water recycling, or resilient infrastructure like sea walls (grey, green and hybrid). Insurers could develop products for drought-resistant agriculture or water-efficient industries, aligning profitability with climate adaptation.
5. Regulatory reforms: unlike some European countries, UK insurers cannot build tax-deductible catastrophe reserves. Allowing such reserves could enhance resilience against systemic risks, which require long-term planning beyond annual premium adjustments.

The road ahead

The UK insurance industry faces an existential challenge as climate change drives systemic risks that render traditional risk mutualisation incompatible with today’s reality. The increased frequency, correlation, and unpredictability of floods, droughts, coastal erosion, and wildfires, coupled with socioeconomic vulnerabilities, threaten to shrink risk pools, deplete reserves, and drive insurers from high-risk markets. Without adaptation, coverage gaps could leave homeowners, farmers, and businesses exposed, with cascading impacts on real estate, debt markets, and economic stability.

We should acknowledge data improvements, which while valuable, are not the primary barrier; structural incentives and expertise in translating climate risk into financial decisions are critical. By embracing parametric insurance, advanced modelling, public-private partnerships, and adaptation-focused products, the industry can navigate this new landscape. However, failure to act risks widespread uninsurability, particularly in vulnerable regions like southeast England, where floods and droughts converge. The interplay of climate risk and socioeconomic inequality could exacerbate vulnerabilities, a concern echoed in the Conversation article’s call for proactive drought solutions. The UK insurance industry must evolve beyond mutualisation to remain a cornerstone of economic resilience in a warming world.