The insurance-linked securities (ILS) market reached a record $144.5 billion in outstanding capital at the half-year mark of 2026, according to Moody's Ratings, which described the sector as reaffirming its position as "a deep and resilient source of insurance risk capital" even as traditional reinsurance pricing falls at its steepest pace in years.

Citing data from Aon, Moody's said catastrophe bonds outstanding grew to around $63.4 billion, up 17% from a year earlier and nearly double their 2021 level, while reinsurance sidecars grew to roughly $23 billion, up about 50% since the end of 2024.

Aon's own most recent ILS report put catastrophe bond issuance over the 12 months to June 2026 at $24.9 billion, the highest 12-month total on record and 15% above the previous high.

Softer reinsurance pricing hasn't slowed growth

The record ILS growth is unfolding against a backdrop of rapidly falling traditional reinsurance pricing.

Guy Carpenter's Global Property Catastrophe Rate-On-Line Index fell 16% over the course of 2026's renewals, the steepest annual decline since the late 1990s, leaving pricing roughly 22% to 23% below its 2024 hard-market peak, though still meaningfully above the last soft-market low reached in 2017.

Meanwhile, abundant capital and several consecutive quarters without a major individual catastrophe loss have driven the softening, with catastrophe bond spreads following traditional reinsurance pricing lower even as the average expected loss of new cat bond issuance has edged higher.

Moody's noted that pricing has compressed most sharply on remote tail-risk layers, where expected losses run below 2%, while compressing less on higher-frequency, lower layers where the risk of loss is greater, a pattern that mirrors the traditional reinsurance market's own demand dynamics.

Despite the pricing pressure, the asset class continues to deliver strong returns. The Swiss Re Global Cat Bond Index returned 11.4% in 2025, a third consecutive double-digit year, and 4.1% in the first half of 2026, with low correlation to equity and bond markets.

Investors are moving up the risk curve

As spreads compress on lower-risk instruments, Moody's said investors are allocating more capital toward instruments with higher expected losses, including aggregate covers, frequency protection and secondary perils such as wildfire, flood and severe convective storm.

That shift is broadening the range of risks insurers can transfer to capital markets, covering perils that have historically been harder to place, though it also increases investor exposure to modelling uncertainty and loss volatility, since multi-peril aggregate bonds have historically been one of the main sources of investor losses.

The market also broadened geographically during the first half of the year. Gothaer issued the first catastrophe bond covering solely German flood risk, and Zurich returned to the cat bond market after a 13-year absence. Twelve (12) sponsors issued their first catastrophe bonds during the period, according to Swiss Re, while sponsors increasingly widened the range of perils covered within single placements, with SageSure's Gateway Re 2026-1 spanning named storm, earthquake, severe convective storm, wildfire and winter storm on an all-perils basis.

Data centres are the next frontier

Moody's flagged data centre and digital infrastructure risk as an emerging area where capital markets are likely to play a growing role, projecting at least $3 trillion in global data centre investment over the next five years.

That estimate sits alongside separate figures from Swiss Re Institute, which projects global insurance premiums tied to data centres will nearly double by 2030, from $10.6 billion to $24.2 billion, while capital spending by the five largest hyperscalers alone is forecast to exceed $600 billion in 2026, a 36% annual increase.

Insurers and brokers have already begun building dedicated capacity for this risk. Aon expanded its Data Center Lifecycle Insurance Program to $5 billion in July 2026, covering construction, property damage, business interruption, liability, cyber and project cargo risk across a data centre's full asset lifecycle.

Marsh has separately launched Stratus, a $10 billion property-focused exchange for operational digital infrastructure backed by 30 capital providers. No catastrophe bond dedicated specifically to data centre risk has yet been issued, though Moody's expects this to become a market focus over the coming year, given how efficiently cat bonds already transfer modellable natural catastrophe risk to capital markets.

Why casualty and MGA-linked structures carry different risks

Beyond catastrophe risk, Moody's pointed to continued growth in casualty-oriented sidecars and structures supporting business originated by managing general agents, where underwriting quality can be harder to assess given MGAs' often limited operating histories and less public disclosure than traditional carriers.

Asset-intensive life reinsurance sidecars have also expanded quickly as life insurers partner with asset managers to support long-dated annuity liabilities, carrying meaningful asset risk since higher-yielding, less liquid assets often back those liabilities, and typically collateralising reserves rather than the full reinsured limit, a structurally different risk profile from property catastrophe ILS.

The combination Moody's describes, record capital growth alongside the steepest reinsurance pricing decline in over a decade, is not necessarily contradictory. Cheaper reinsurance and cheaper cat bonds are both symptoms of the same abundant-capital environment, and sponsors locking in multi-year protection now are explicitly hedging against the pricing eventually turning, which Moody's notes remains genuinely uncertain heading into the January 2027 renewals: a single major hurricane landfall could reprice risk quickly even with a favourable El Niño pattern suppressing Atlantic activity.

For an industry that has spent years discussing the "convergence" of insurance and capital markets as a slow structural trend, this report reads as evidence that convergence is now simply how a growing share of catastrophe, and increasingly non-catastrophe, risk gets priced and placed.