Hurricane season forecasts: Sound and fury, signifying nothing?

Atlantic hurricane season is upon us; but what makes this year different; and what scale of storm would qualify as a market-turning event?

Weather forecasts are amazingly helpful tools when looking at the next 24-72 hours, but the longer the outlook, the less reliable forecasts tend to be, given the toughness of the task and the severe data limits of even the most sophisticated modern models.

For this reason, Atlantic hurricane season forecasts have limited utility for re/insurers at this still-early stage in the storm season. A forecast of an active season, with many hurricane strength storms breaking out at sea, can still translate into a mild season for claims.

Firstly, this is because many storms are formed and dissipated without ever making landfall. Secondly, even the perfect storm can make landfall in a remote or under-insured area, even within Florida’s renowned coastal exposure concentrations.

We saw this with last year, with the late arrivals of Hurricanes Helene and Milton, both tremendously powerful storms that made US landfalls and caused tragedy and devastation, but not nearly as much as they might have for re/insurance companies, had they hit bigger concentrations of insured property.

What the forecasters say

2025’s season is expected to bring above-average storm activity, though this activity is expected to be less severe than the record-breaking 2024 season, reinsurance brokers and rating agency commentators concurred upon in the early days of June.

Forecasters from nine weather institutions predicted an average of 16 named storms, eight hurricanes, and three major hurricanes this season—exceeding both the long-term average (1950–2024) and the 30-year climate norm (1991–2020), a report from Moody’s observed.

The science for this is based on a transition between currently weak La Niña conditions into something Moody’s described as “a more ENSO-neutral condition” – I’ll leave the science there, with apologies to meteorologists and cat modellers. 

What the brokers have said

Renewals at 1/1 saw US property catastrophe reinsurance rates decline by some 5% to 15% on a risk-adjusted basis, broker reports suggested. For a more up-to-date market snapshot, the pre-season June Florida renewals also saw rates decline even further from last year.

Guy Carpenter’s 1/1 rate-on-line indices, for instance, showed a decline across regions globally for property catastrophe treaty business. For US property cat business, it decreased 6.2%, the broker said.

Howden noted at 1/1 that US property cat pricing is beginning to decline from historically high levels, but that structural changes introduced during the hard market “are likely to endure”, the broker suggested.

Insurers are expected to face continued earnings volatility in 2025, Howden observed, as they absorb the majority of catastrophe losses, potentially soaking up claims and sparing their reinsurers from market-turning losses, due to persistently high attachment points.

What counts as big, nowadays?

Helene and Milton were the largest hurricanes to make US landfalls in 2024. These two huge storms accounted for a substantial portion of insured losses, with estimates ranging from $20bn to $36bn each.

So-called secondary perils – such as floods, hailstorms and wildfires – added approximately $50bn in the aggregate to the industry’s claims total. Added together, globally, total insured losses from natural catastrophes in 2024 have been pegged at around $125bn.

Fifteen years ago, this would have been considered a market-turning figure. Now, in hindsight, we know that $125bn has not been enough to turn a softening market.

We can also assume that core dynamics such as retention points remain more-or-less untouched, and that the contribution of secondary peril events may remain relatively consistent or perhaps continue to grow, given that it is made up of many smaller events.

Individual hurricanes would logically need to be a level above Milton or Helene, which failed to tip the market, to qualify as market turning. That could mean something north of $50bn of insured losses for an individual storm to be in the frame as a market-turning event.

More than one such huge storm in the same season, making landfall in the ‘right/wrong’ place, would dramatically increase the likelihood of a market turn. But at this point, not even the bravest or foolhardiest of modellers would predict such a thing.

AdvantageGo’s exposure and risk management solution, Exact, empowers exposure and risk management professionals to better understand their exposure and plan for future extreme weather events accordingly.