A mortgage is the most heavily shopped product in household finance. Buyers compare rates across lenders, run the arithmetic on points, and argue over an eighth of a percent that will cost them a few dollars a month.
Then they arrive at closing and pay a stack of fees they have never compared to anything, because the disclosure showing them arrives when the house is already bought and the alternative to signing is not buying it.
The one line item with no competitive pressure
Title search, title insurance, escrow administration, settlement, recording, courier, notary. Individually small, collectively several thousand dollars, and structurally different from the interest rate in one decisive way: the buyer usually does not choose the vendor. The lender or the agent does, and neither of them pays the bill.
That is a market with a disclosure regime and no price discipline. Fees are itemised in detail — the paperwork is genuinely transparent — but transparency only constrains a price if the person reading it can act on it, and by the time it is read the transaction has momentum nobody wants to interrupt over four hundred dollars.
Title insurance is the clearest case. It is a one-time premium against a risk that is largely eliminated by the search performed immediately beforehand, in a market where loss ratios run far below what any other insurance line would sustain. The industry's answer — that the search is the product and the premium funds it — is coherent, and it also happens to describe a fee that has risen with house prices despite the underlying work being increasingly automated.
Recording and municipal fees have moved for their own reasons. Counties that deferred technology investment for years are funding it through transaction charges, which is a reasonable public finance decision that lands entirely on whoever happens to be buying a house this year.
The effect on the household is a lump at exactly the wrong moment. Buyers arrive at closing having already stretched for the deposit, and the settlement stack takes another few thousand out of reserves at the point where reserves matter most — the first year of ownership, before anything has been repaired. That is the same squeeze visible in the rising cost of insuring the asset once the keys change hands.
It also lands unevenly. As a percentage of the transaction the stack is close to fixed, which means it is regressive: the buyer of a modest house pays nearly what the buyer of an expensive one pays, against a much smaller base. For a first-time buyer the settlement costs can exceed a year of the savings that produced the deposit, which is not visible in the aggregate picture of household balance sheets because it happens once and then disappears into the cost basis.
Lenders are not the obvious villain here, and several have started unbundling the stack in their own marketing, which is the first competitive pressure the segment has faced. The obstacle is that a fee borne once, at a moment of maximum transaction fatigue, generates almost no consumer anger relative to its size — unlike a monthly payment that reprices every renewal, which households notice and eventually act on.
Shopping does work, where anyone does it. Buyers who obtain independent quotes for title and settlement routinely find spreads of a thousand dollars or more on identical work in the same county. What the market lacks is not the right to compare but any moment at which comparing is convenient, and the industry has shown no urgency about creating one.
