Commercial Property Insurance Rates Fell 31% Through a $35 Billion Storm Year. Texas Fell Fastest.

How far has commercial property insurance rate on line actually fallen?

On the national view in the Advocate app, US property rate on line starts near $0.50 in the autumn of 2025 and trends down steadily to roughly $0.33, a decline of about 31 percent across twelve months. Extending the window to two years shows the same shape at greater amplitude: rate held in the high forties to low fifties through 2024 and 2025 before breaking down sharply from February 2026.

The detail that matters is what sits on top of the curve. The national chart overlays natural disaster markers, wind, hail, tornado, and flood, across the same period. Those markers do not thin out as the line descends. They are as dense at the bottom of the curve as they were at the top. Loss frequency did not fall away and leave room for price to follow. Price fell through continuing loss activity.

That is the anomaly the episode set out to explain, and Dowson framed it as a question about durability rather than direction.

Was 2026 actually a bad severe convective storm year?

Not a record year, but a real one, and the distinction turns on which measure you use.

Gallagher Re put year to date US severe convective storm insured losses above $35 billion as of mid August 2026, a figure that includes public and private losses and preliminary estimates from the outbreak that swept the central and eastern United States from August 9 to 12. In dollar terms that sits below the five year average for the peril, which Gallagher Re places at roughly $50 billion for the 2021 to 2025 period. On that measure alone, 2026 looks unremarkable.

Frequency tells a different story. Six separate severe convective storm outbreaks have each produced more than $1 billion of insured loss so far in 2026, against seven for the whole of 2025. The year is not finished, and the $35 billion is a developing number rather than a settled one.

Which markets saw commercial property rates fall the fastest?

Eighteen months ago, the three state comparison looked exactly as exposure would predict. Texas priced at $0.47, California at $0.35, and Washington at $0.31. The most exposed market carried a visible premium over both controls.

Across the trailing six months of placements, Texas sits at $0.34, roughly 28 percent below where it was. California sits at $0.31 and Washington at $0.27, both down but by materially less. Every market got cheaper. Texas got cheaper close to twice as fast as either control.

The consequence is convergence. The Texas premium over California, about twelve cents eighteen months ago, has compressed to under 10 percent of rate. For a commercial property insurance buyer, that is the opposite of the intuitive result. The reasonable expectation going in is that carriers ease pricing in benign markets while holding the line where the tornadoes are. The data shows the reverse: the tornado exposed market is the one being discounted hardest.

Did newer buildings just dilute the average?

This is the objection that has to clear before any of the above means anything. If newer, cheaper to insure buildings entered the Texas book over the period, the average rate on line would fall without a single individual price moving.

Two tests were run against it. Both failed to support the mix explanation.

The first isolated two building age cohorts and tracked each on its own, removing composition as a variable. Pre-1980 buildings moved from $0.69 in February 2025 to $0.53 in February 2026, a drop of roughly 23 percent for the same vintage of risk. Buildings constructed after 2001, the lower risk cohort, moved from $0.31 to about $0.29 over the identical window, roughly 7 percent. Both cohorts got cheaper independently, which rules out pure mix.

The second test was stronger, and the result was directionally counterintuitive. It asked what today's average would look like if the split between old and new buildings had been frozen where it stood eighteen months ago. If composition were inflating the decline, holding the split constant should shrink it. Instead the decline got larger, running from $0.48 down to $0.33.

The conclusion is narrow and it is the load bearing one for everything that follows: this is not carriers writing a safer book and reporting it as a rate cut. Same buildings, same risk, lower price.

Why did $35 billion in losses not move the price?

Because of where the loss landed, not how large it was.

A primary carrier, the company that issued the policy, purchases reinsurance that attaches above a retention. The retention is the amount of loss the carrier absorbs entirely from its own balance sheet before anything transfers to the reinsurer. It functions like a deductible, but at the level of the whole company rather than an individual policyholder. If a carrier's retention on a given treaty is $50 million, the first $50 million of any one loss in that layer is the carrier's own money, and the reinsurer writes nothing until a single event breaches that threshold.

The larger the retention, the more has to happen in one event before reinsurance capital is touched at all.

Severe convective storm does not produce that kind of event. It does not arrive as a single enormous claim the way a major hurricane landfall does. It arrives as tens of thousands of individually modest claims, scattered across dozens of carriers and dozens of states, spread across an entire season. Even in a year that totals well past $35 billion in aggregate, no individual carrier's share of any one outbreak is large enough to punch through its own retention.

So the loss sits with the primary company for its full duration. That is not a surprise to anyone. It is background loss, budgeted annually and already priced into the model as a normal cost of doing business.

And rate is set off reinsurance capital, not primary capital. Through this entire loss year, reinsurance capital was never touched by convective storm activity at all. The loss total and the rate curve were running on two separate clocks, which is why $35 billion of real, paid loss exerted no upward pressure on price.

Why is capital competing hardest for the most exposed risk?

Here the analysis moves from mechanism to motive, and the honest answer is that two explanations fit the same data.

Reinsurers have had a strong year. Profits are healthy, capital is abundant, and the question facing the market is where to allocate it. The observable behaviour is that capital is competing hardest for precisely the risk that generated this year's loss numbers. That is counterintuitive on its face: money would be expected to move away from a peril that just produced $35 billion of loss, not toward it.

The first explanation is a genuine re-rating. Carriers went back, examined the risk with better information, and concluded it had been overpriced all along, leaving room to come down and still make money. That would be a legitimate assessment.

The second is simpler. Capital needs somewhere to go, Texas had room to absorb it, and when everyone arrives at once, supply expands and price falls. That is not a risk judgment at all. It is a capacity outcome.

From a building owner's seat the two are indistinguishable, because they produce the identical renewal quote. The difference only surfaces the next time reinsurance capital tightens, when you find out whether the pricing holds or snaps back.

The data does offer a hint, though not proof. A genuine re-rating of how a state's worth of convective storm risk should be priced is a slow process. It requires updated models, new data, and sign off across multiple layers inside a carrier. A 28 percent move in eighteen months is faster than that kind of deliberation typically clears.

What does this mean for your next property renewal?

The practical read is that a decreasing renewal is not evidence that your risk profile improved or that your loss history is being rewarded. The softening is running on reinsurance capital supply, and it is moving fastest in the markets with the most exposure, not the least.

That reframes three questions worth asking before signing:

1. Is this renewal priced off your own loss experience or off reinsurance capital conditions, and if it is the second, what in your submission are you actually negotiating?

2. National commercial property rate on line fell roughly 31 percent over twelve months, and further in high exposure markets. What did your specific program get, and against what benchmark do you know whether that is competitive?

3. If this is capital finding a home rather than a durable re-rating of the risk, what is the plan for the renewal after capital tightens?

Whether the trend continues is genuinely open. What the data settles is that the softening is not stopping at the risky business. That is where it is moving fastest.