Seventeen points between staying and moving

Good morning. Two numbers landed yesterday and the distance between them is the letter.

Seventeen points between staying and moving, on the same product, in the same year. That is from the September ICE Mortgage Monitor, published yesterday morning, which tracks property insurance costs inside mortgage escrow accounts across a very large sample.

The same report has property insurance up 8.7 per cent over the year and at another record in the second quarter, with the rate of increase slowing. Both things are true at once: it is still going up, and it is going up less steeply than it was.

A gap like that invites a simple reading, which is that shopping saves you seventeen per cent. It does not say that, and the difference matters if you are about to spend a Saturday on the telephone.

People who switch are not a random sample of homeowners. They are disproportionately the ones who received a large increase, which is what prompted them to look. They are also the ones whose houses and claim histories let another carrier take them on. A household that nobody else will write is a household that stays where it is, and its premium goes in the other column.

So the honest version is that among people who shopped and could move, the ones who moved paid less. Which is still worth knowing, and is a smaller claim than the headline.

Here is the part I have watched go wrong for thirty years, and it is the reason I am writing about a price story at all.

Two quotes are only comparable if the numbers underneath them match. Same dwelling limit. Same loss settlement basis, replacement cost against actual cash value. Same deductible, and the same treatment of wind and hail, which in a good many states carries its own percentage deductible rather than a flat one. Same special limits. Same endorsements, or a note of which ones are missing.

A quote that comes in four hundred dollars cheaper has usually changed at least one of those, and the one it changes most often is the deductible, because that is the lever that moves the premium fastest without looking like less coverage.

A week ago I told you to take your dwelling limit, multiply by 1.43 and look at the result. The multiplier is sound, and the instruction was not.

Applying a reconstruction cost increase to an old limit assumes the old limit was right when it was written. Mine may not have been. If a limit was already short in 2019, multiplying it forward carries the shortfall along and understates the gap rather than measuring it.

The multiplication tells you roughly how much costs moved. It does not tell you whether your starting point was correct, and I presented one as though it were the other.

ICE Mortgage Monitor, September 2026. Freddie Mac PMMS, 3 Sep 2026. Federal Reserve, 29 Jul. BLS, 12 Aug.

I have no view on whether you should move carriers. I have never had one, and I would be suspicious of anybody who formed one without seeing your declarations page.

What I will say is that the six figures above take about four minutes to write down and they change the conversation from a price hunt into a comparison. Without them you are choosing between two numbers that describe different things.

The bulletins are new this week. If they are useful I will keep them, and if they read as filler I would rather know now than in a month.

Monday: the rerun of my own estimate should be back, with the two hundred and eighty square feet in it, and I will print whatever it says.