Gallagher Re’s Bowen: SCS, data centres and ENSO focuses after ‘manageable’ Q1 leaves cat budgets intact

A “manageable” first quarter for natural catastrophe losses has left the re/insurance industry in a stronger position heading into the costlier parts of the year, according to Steve Bowen, chief science officer at Gallagher Re.

Global insured losses reached $20bn in Q1 2026, around 26% below the decadal average, while economic losses were estimated at $58bn, according to Gallagher Re’s “Q1 2026 Natural Catastrophe and Climate Report”, published today.

“It’s another example of a quarter where the industry is in a stronger position than it has been historically going into the more expensive Q2 and Q3 of the year,” says Bowen (pictured).

“It continues this trend of manageable quarters, and we’ve seen that play out in terms of much more capital flowing into the market.”

Despite the relatively modest insured losses, he emphasised the gap between insured and economic losses, suggesting the industry could be doing better even for US risks.

Gallagher Re’s report confirmed that the protection gap for the first quarter stands at $39bn, reflecting the large share of uninsured damage globally.

Pricing resilience and capital strength

The subdued start to the year has reinforced expectations that reinsurance pricing will remain soft unless the industry faces an extreme loss scenario.

Bowen says it would take a materially larger shock to shift the market – something like the hurricanes Katrina, Rita Wilma combined shock of 2005.

He points to historical benchmarks such as 2005 and 2017 as comparable stress scenarios for the market.

“We now estimate it’s going to take at least $115bn to $125bn, either a single event or a series of events above what we would annually expect, for that to really start to show some meaningful change in pricing,” he says.

“That essentially suggests you’d have to be pretty well north of $200bn for the year,” he added, referencing the significant share of so-called non-peak perils that have made up the bulk of cat losses in the absence of large hurricanes.

Rethinking ‘peak’ perils

The reinsurance broker’s report also raises questions around how the industry defines peak risks, particularly in Europe.

Despite windstorm losses reaching their highest economic level for a first quarter since 1999, the region has not seen a major insured loss event above $10bn since 2007.

Bowen says this has shifted market focus, and questions whether it can still be called a peak peril.

“We’re not trying to be provocative for the sake of being provocative, but when you look at the data, it’s been coming up on 20 years since we’ve seen a $10bn event,” he says.

“Most of the conversations now are around flood and severe convective storm. It’s not necessarily around European windstorm anymore,” Bowen says.

He adds that this reflects how underwriting priorities are evolving based on actual loss experience rather than legacy classifications.

SCS losses driven by structural factors

One of the report’s central themes is the continued escalation of severe convective storm (SCS) losses in the US.

Since 2008, SCS insured losses have risen sharply, but Bowen stresses that this is not primarily a story about climate risk or weather patterns.

“Eighty to 90% of the annual growth we’ve seen this century is due to non-weather factors,” he says.

“It’s not downplaying the effects of climate change, but that alone simply does not explain what’s driving this increase.”

The report highlights a combination of macroeconomic and societal drivers, including higher construction costs, social inflation, litigation, and the expansion of housing in high-risk regions.

Bowen points to the role of rising material costs to replace or repair storm damage, even amid wider inflation.

“You go back to 2008, when oil prices spiked, and that feeds directly into asphalt roofing costs. Those prices never really came back down, so you’ve embedded a higher loss environment,” he says.

Data centres as concentration risk

The rapid expansion of data centres is another emerging risk theme, particularly as facilities are often located within storm-prone regions.

Bowen warns that exposure is growing faster than resilience in some cases.

“If you’re talking about data centre facilities that are $10bn to $20bn per location, even a partial loss is going to be significant,” he says.

First and second order effects of damage to a data centre are a growing concern for the re/insurance sector, while data centre construction has boomed due to rising technology demand, in large part fuelled by the rise of artificial intelligence.

“Given where they’re being built, you have to assume there’s going to be some level of catastrophe risk that they face,” Bowen says.

He highlights concerns around building standards and infrastructure resilience, particularly in areas without strict codes, singling out several US states that are home to many data centres.

“Virginia has the most, then Texas and then Illinois,” he warns, referring to inland flash floods that led to significant losses in 2016, before the data centre boom.

Outlook: ENSO and seasonal uncertainty

Getting into storm season, the return of El Niño conditions and implications for global catastrophe activity is a big focus.

“All eyes are going to be on ENSO [El Niño–Southern Oscillation],” Bown says.

“Historically, El Niño means fewer Atlantic storms, but it doesn’t guarantee you won’t get a landfall.”

He adds that risks may shift geographically rather than disappear.

“The Pacific tends to become more active, and the question becomes whether that puts markets like Japan more in play – you’re not removing risk, you’re redistributing it,” ,” he adds.