The Slow Repricing: Climate Change, Insurance Costs, and the Bill Nobody Sees Yet

Estimated reading time: 14 minutes

On 1 July 2026, the world’s reinsurers cut their prices again. Property catastrophe rates fell 20% to 25% for the best-performing North American accounts, according to Gallagher Re. Guy Carpenter’s global property catastrophe rate-on-line index has now fallen 16% across this year’s renewals. That marks the steepest annual decline since the late 1990s. This is the story of climate change and insurance costs.

Now look at the other end of the chain. The average American homeowner will pay roughly $3,057 for cover in 2026. That premium sits about 46% above 2021, near three times the pace of inflation. Two prices move in opposite directions during the same month. Both respond to the same physics.

This piece explains the divergence. It also explains why the divergence closes painfully, and who pays when it does.

The year the industry called lucky

Munich Re counted roughly $224 billion of global natural disaster losses in 2025. Insurers absorbed about $108 billion of that total. Swiss Re, working from a slightly different basis, put insured losses at $107 billion. Both totals fell sharply against 2024, when insured losses reached $147 billion.

Neither reinsurer read the decline as good news. No hurricane made landfall on the American mainland during the entire season. Thomas Blunck, a member of Munich Re’s board of management, stated the point plainly.

Two faces, one year

Balz Grollimund, Swiss Re’s head of catastrophe perils, called the shortfall “favourable variability rather than any easing of underlying risk”. Munich Re’s chief climatologist Tobias Grimm described a year with “two faces”. The first six months delivered the costliest loss period the industry has ever recorded. The final six months delivered the quietest in a decade.

Reinsurers therefore closed 2025 with strong balance sheets and a warning. The market heard the balance sheets.

A hundred billion, four years running

Global insured catastrophe losses have exceeded $100 billion in each of the past four years. Howden Re makes that observation in its June 2026 renewal analysis. The threshold once signalled a shock. It now describes a floor.

Swiss Re anchors the underlying trend at 5% to 7% real growth per year, measured since 1996. Apply that rate and 2026 lands near $148 billion. Extend it to 2030 and the trend implies $186 billion. Swiss Re’s modelled peak-loss scenario for 2026 reaches about $320 billion.

That last number deserves a pause. A single bad year could more than double the recent annual average. Reinsurers exist to absorb precisely that tail. They also price it twelve months at a time.

The trend line matters more than any single year

Swiss Re’s own arithmetic puts 2025 roughly $33 billion below the trend line. The industry booked $107 billion where the model implied $140 billion. Weather variability produced that gap. Risk reduction did not.

Capital markets read the gap differently. Investors saw two consecutive years of strong reinsurer returns and light catastrophe budgets. They deployed accordingly. The result now governs pricing across the whole chain, from Bermuda to a bungalow in Butte County.

Secondary perils became the main event

Munich Re attributes $166 billion of 2025 economic losses to non-peak perils. Floods, wildfires, hail and severe thunderstorms drove that figure. Insurers paid roughly $98 billion of it, a record for the category.

Swiss Re reaches the same conclusion from a different direction. Secondary perils generated 92% of global insured losses last year, an all-time high. American severe convective storms alone cost insurers about $42 billion, and have exceeded that mark for three consecutive years.

The Los Angeles wildfires topped every table. The Palisades and Eaton fires produced roughly $53 billion of economic loss. Insurers covered about $40 billion, the largest insured wildfire loss on Swiss Re’s records.

Peak perils hold the tail. Secondary perils supply the trend.

This distinction drives everything that follows. Hurricanes and earthquakes create the headline years. Hail, flood and fire create the arithmetic.

Secondary perils also carry a higher insured share. A tornado outbreak in Oklahoma converts into claims far more efficiently than a Myanmar earthquake. That efficiency looks like resilience from a distance. Up close, it looks like a bill.

Exposure explains most of it, but not all of it

Swiss Re decomposed 55 years of loss data by peril and by region. Exposure growth explains more than 80% of the long-term global increase in weather-related insured losses. More valuable property now stands in harm’s way, and rebuilding it costs more.

Jérôme Haegeli, the group’s chief economist, points to one simple reality. Developers keep placing more valuable property directly in harm’s way.

The residual still matters enormously. North American wildfire insured losses grow about 14% each year. Roughly 60% of that growth escapes any exposure explanation. European severe convective storm losses grow about 10% annually, and outpace exposure by a factor of two.

Hazard intensification and rising vulnerability fill the remainder. That residual is the climate signal, isolated and measured by the people who pay the claims.

Why reinsurance prices fall anyway

Reinsurers price one year at a time. They reset terms every January, April, June and July. That cadence lets a reinsurer exit a deteriorating risk within twelve months, then reprice the survivors.

Capital, not climate, therefore sets the near-term price. Aon counts global reinsurer capital at a record $790 billion as of 31 March 2026. Third-party capital reached $141 billion. Gallagher Re puts non-life insurance-linked securities capital at $135 billion, up 18% in a year.

Catastrophe bond issuance reached $15.6 billion by mid-June. Mid-year reinsurance demand rose more than 10%. Supply rose faster. Prices fell.

How a renewal actually works

A cedent, meaning the primary insurer, buys layers of protection above a retention. Each layer carries a rate-on-line, the premium expressed as a percentage of the limit. Brokers negotiate that rate, the attachment point, and the terms.

Reinsurers hardened all three variables in January 2023. They raised prices, lifted attachment points, and tightened wordings. Attachment points matter most. Higher attachment means the primary carrier absorbs more frequency before reinsurance responds.

Terms held. Prices did not.

KBW surveyed reinsurance executives in Bermuda before the mid-year renewals. Its analysts found pricing down closer to 20% than 15%. They also found the 2023 structural terms almost entirely intact.

Reinsurers, in other words, sell price concessions rather than structural ones. They will accept a lower rate-on-line. They will not lower the attachment point back to 2021 levels.

That asymmetry decides who carries secondary-peril frequency. Reinsurers now sit above the hail, the flash flood and the derecho. Primary carriers sit inside them. Homeowners sit inside the primary carriers.

The reinsurers still print money

Gallagher Re estimates sector return on equity near 19% for 2025. It expects 14% to 15% for 2026. Aon recorded an average first-quarter return on equity of 14.1%. Both figures clear the cost of equity comfortably.

Tom Wakefield, Gallagher Re’s global chief executive, describes a market of “strong capital, healthy returns and increasing competition”. Each element reinforces the others. Healthy returns attract capital. Capital intensifies competition. Competition compresses returns.

Where the cushion thins

Howden Re tracks economic value added across the sector. That measure subtracts the cost of capital from the return on invested capital. The cushion has narrowed through every renewal of 2026.

David Flandro, Howden Re’s head of industry analysis, names the contradiction directly.

Jim Williamson, chief executive of Everest Group, asks the sharper question. He wonders whether discipline holds, or whether “greed overcomes the fear factor”. The 1 January 2027 renewal will answer him.

The homeowner lives in a different market

Insurance reaches households through primary carriers, never through reinsurers. Primary carriers file rates with state or national regulators. Regulators approve those filings slowly, and sometimes reject them.

Reinsurance savings therefore reach policyholders with a lag measured in years. Reinsurance increases reach them faster, because carriers file for relief immediately.

Insurify projects a 4% average premium rise for American homeowners in 2026. That follows a 12% jump in 2025 and a cumulative 46% climb since 2021. Deceleration is not reversal.

The averages hide the map

State-level numbers scatter violently. Minnesota premiums rose 34% during 2025. Colorado rose 33%. Nebraska rose 25%. Oklahoma rose 24%. None of those states touches an ocean.

Insurify expects California premiums to rise about 16% during 2026. The United States Treasury compared premiums across ZIP codes ranked by expected climate loss. Households in the riskiest fifth paid 82% more than households in the safest fifth.

Climate risk already prices itself into household budgets. It simply arrives under a different name.

California shows the endgame

The California FAIR Plan insures homeowners whom private carriers refuse. Its policy count grew 44% between autumn 2024 and December 2025, reaching 668,600. That growth reflected withdrawal, not marketing.

The January 2025 wildfires cost the plan roughly $4 billion. The plan then levied a $1 billion assessment on its member insurers. That was its first assessment since the 1994 Northridge earthquake.

California regulators allow member carriers to recoup half of their share from policyholders. The surcharge appears on ordinary renewal bills across the state. Households in Sacramento therefore help pay for houses that burned in Altadena.

A 29% increase, and what it conceals

The FAIR Plan requested a 35.8% statewide rate rise. The California Department of Insurance approved 29.1%, effective 15 October 2026. Averages conceal distributions.

The wildfire component drives almost the entire increase. High-risk properties will see far steeper rises, and some wildfire premiums will roughly double. A minority of low-risk policyholders will pay slightly less.

Carmen Balber of Consumer Watchdog called the decision “a real blow for consumers”. The plan, meanwhile, now holds bonding authority under California’s 2025 stabilization package. A state insurer of last resort can now issue debt to pay claims.

Europe’s gap is wider, and much quieter

Europe recorded only about $11 billion of natural disaster losses during 2025. Insurers covered roughly half. The ten-year average sits at $35 billion, so 2025 flattered the continent badly.

That calm disguises a structural weakness. EIOPA reports that insurance covered only about a quarter of European extreme-event losses between 1980 and 2024. Its 2025 Eurobarometer found that just 17% of respondents hold natural catastrophe cover for property damage.

Petra Hielkema, EIOPA’s chairperson, rejects the historical method outright.

The German insurance association has warned that property premiums could double within a decade. EIOPA cites that warning in its own supervisory commentary.

Brussels reaches for a pool

EIOPA and the European Stability Mechanism proposed a Europe-wide risk-sharing framework in April 2026. The design pairs a catastrophe pool with a loan-based public backstop for extreme tail events. It would sit beside national schemes rather than replace them.

The backstop uses repayable, market-consistent financing. Brussels frames it as liquidity, not fiscal transfer. Ratings analysts read it as modestly positive for private insurers.

Italy moved first and moved harder. Rome now requires almost every registered business to buy catastrophe cover. Micro and small enterprises faced a 1 January 2026 deadline. The state export credit agency SACE absorbs up to half of insurers’ indemnities.

A private market failure becomes a public balance sheet

Each of these mechanisms performs the same conversion. A risk that private capital declines to price migrates onto a sovereign ledger. California issues bonds. Italy deploys SACE. Brussels proposes a pool.

None of these instruments reduces the underlying hazard. They redistribute its cost across time and across taxpayers. That redistribution buys adaptation time. It also mutes the price signal that would drive adaptation.

The transmission runs through the mortgage

Lenders require insurance. No policy means no loan. That single rule converts an insurance problem into a credit problem.

In Louisiana, high insurance costs sink 30% to 40% of mortgage applications. A 2026 survey of American buyers and sellers found that 21% of transactions collapsed over insurance issues. Nearly half of respondents hit insurance problems of some kind.

Ishita Sen, who studies insurance markets at Harvard Business School, traces the escalation. She warns that “some households may end up defaulting on their mortgages when insurance costs rise”.

From premium to price to collateral

First Street estimates that insurance underprices climate risk on 39 million American properties. That covers 27% of the housing stock. The firm projects roughly $1.5 trillion of housing asset losses across the next three decades.

Read that projection carefully. It does not describe fire damage. It describes repricing, as buyers finally subtract the cost of cover from what they will bid.

Insurance premiums once consumed about 8% of a typical American mortgage payment. They now consume close to a fifth. Collateral values follow cash flows.

The protection gap has a number

Swiss Re estimates the global natural catastrophe protection gap at $424 billion in 2025. That figure rose from $395 billion the year before. Its resilience index reached about 27%, up from 25% a decade earlier.

Nearly three-quarters of global catastrophe exposure therefore carries no insurance at all. In emerging economies, 80% to 90% of catastrophe losses go uncovered.

The 2025 regional data confirms the asymmetry. Asia-Pacific absorbed $73 billion of losses and insured $9 billion of them. North America absorbed $133 billion and insured $93 billion. Wealth, not weather, determines who recovers.

The math that breaks

Günther Thallinger sits on the board of management at Allianz. In March 2025 he published an argument about the limits of insurability itself.

Thallinger argued that entire regions are already losing access to cover. He then followed the chain outward. Nobody mortgages an uninsurable house. Nobody finances an unmortgageable one. He called the result a climate-induced credit crunch.

He later told CNBC that current policy paths point toward 2.7 or 3 degrees of warming. At that level, he argued, adaptation stops working. No engineer defends Amsterdam against three meters of sea-level rise.

Adaptation carries the better economics

Swiss Re calculates a median benefit-cost ratio of 1.86 across adaptation projects. Every dollar spent returns almost two. Allianz estimates that catastrophe losses typically cost about ten times the price of adaptation.

European flood defences already appear in the loss data. Swiss Re credits dikes, levees and land-use rules with constraining insured flood loss growth in Britain, France, Switzerland and Austria.

Adaptation lowers vulnerability. Lower vulnerability preserves insurability. Insurability preserves credit. The chain runs in both directions.

What to watch next

• Through October 2026. The Atlantic hurricane season tests reinsurers’ unusually healthy catastrophe budgets. Gallagher Re logged only $38 billion of insured losses to mid-June.
• 15 October 2026. California FAIR Plan rates rise 29.1%, with wildfire-exposed premiums rising far more.
• 1 January 2027. Renewals reveal whether reinsurance pricing discipline survives a fourth soft round.
• Through 2027. Brussels decides whether an EU catastrophe pool moves from discussion paper to statute.

The slow burn

Climate change does not bankrupt insurers. It reprices them, and it reprices everyone standing behind them.

Reinsurers reprice annually and shed risk within a year. Primary carriers reprice slowly and answer to regulators. Households never reprice at all. They pay, they cut cover, or they leave.

More than half of American homeowners surveyed by Insurify report financial sacrifices to afford cover. Nearly three in ten would drop coverage entirely if their lender allowed it. That number describes the real protection gap, and no pool in Brussels closes it.

The industry has already run the arithmetic. It has concluded that the risk remains priceable, for now, at a price. The unresolved question concerns who can still afford that price in 2035.

Primary sources

1. Munich Re, Natural disaster figures 2025, 13 January 2026.
2. Swiss Re Institute, sigma 1/2026: Natural catastrophes in 2025, 19 March 2026.
3. Swiss Re Institute, Natural catastrophe protection gap, June 2026.
4. Gallagher Re, 1st View renewal report, July 2026.
5. Aon, Reinsurance Market Dynamics, Midyear 2026.
6. Howden Re, 1 June 2026 property catastrophe renewals.
7. Guy Carpenter, via Global Property Rate-on-Line Index, July 2026.
8. EIOPA, Insurance protection gaps in a changing climate, speech by Petra Hielkema, 16 April 2026.
9. EIOPA and the European Stability Mechanism, joint proposal on a European catastrophe risk-sharing framework, 9 April 2026.
10. California Department of Insurance and the California FAIR Plan, dwelling rate filing approval, May 2026.
11. Levy Economics Institute, A Premium Crisis, Policy Note 2026/2, 15 April 2026.
12. First Street, climate risk and property valuation analyses, 2025–2026.