Hey there —

Here's this week's number, and it got bigger while I was checking it: the effective property tax rate across the 100 largest metros runs from 0.28% to 2.58%. About nine times, end to end.

I had it written down as seven. Re-deriving it this week turned up two metros below the one I'd recorded as the floor. A small embarrassment, and a useful one: the number you carry around from last time is a number you haven't checked.

Nine times, on a line item that comes out of your rent every year, forever. It has to be the effective rate, though — tax over market value. A published millage applies to assessed value, and that means something different in nearly every state.

So I did the obvious thing: took the 95 of them with both rent and price data, ranked them on rent-to-price, then re-ranked them on rent-to-price minus the property tax rate. Six of them drop twenty-five places or more. Rochester falls 48. Syracuse 38. Then Buffalo, New Haven, Hartford, Albany.

Every one of them is in Upstate New York or Connecticut.

That grouping is the actual finding, more than any single metro's number. When six markets fall off a cliff together and share two state governments between them, you aren't choosing between two cities — you're choosing between two state legislatures.

Two things came out of it that I didn't expect.

The first is that the lowest rate in the country doesn't win. Honolulu has it — 0.28%, the floor of the whole distribution — and counting it moves Honolulu eleven places. Birmingham, at 0.32%, moves from 20th to 6th. The difference is the denominator: Honolulu's median value is $854,584 and Birmingham's is $263,665. A rate isn't a bill. The bill is a rate times a value, and a tax advantage can only improve a yield that already exists.

The second is worse, and it's the one to actually act on. Every rate I've quoted you is the owner-occupant's rate. Those figures are computed over people who live in the house. South Carolina assesses a legal residence at 4% of appraised value and a rental at 6%; homestead exemptions are worth nothing to you; and in a state that resets assessments on sale, the tax line on the listing is a record of how long the seller has owned, not a preview of what you'll pay.

So the practical move this week is three questions to a county assessor, and the phrasing matters: ask for the rate on non-owner-occupied residential property, in those words. Then pull one comparable rental's actual parcel record. Then ask what happens to the assessment when the property changes hands. About ten minutes.

Do it before you make an offer and it's a price you set correctly. Do it during diligence and it's a re-trade you might lose.

One more, if you already own several in one county: a tax regime is a correlated exposure. One assessor, one millage vote — every unit under it moves on the same day, with nothing on the other side.

One caution, because this is the week's most likely misreading: none of this is an argument for chasing a low tax rate across state lines. The tax you can see is already priced into what you paid. This is about the tax you can't see yet.

My question this week: have you ever had a property tax bill come back materially higher than the one you underwrote? What was the gap, and did you catch it before closing or after? Hit reply — I read every one, and the "after" stories are the more useful half.

Martin