Think in terms of banks not borrowers.

Houses are a much better long term bet for buyers, but car depreciation is more predictable. Lend someone 500k to buy a house and during a recession it might tank in value by 20-40% while a bunch of people walk away. Worse the’ll likely pull out their equity before defaulting in mass.

Cars value on the other hand is more counter cyclical. When the economy is bad people still get into accidents etc but they now want to buy used cars instead of new ones. Even better it’s just less money being lent out at higher interest rates.

That's why banks require huge deposits though, or charge you for insurance for this exact risk. If the house tanks 20% but you had to put in a 20% deposit in + whatever payments you made up until that point, you eat all of the losses first before the bank.

In 2024 only 52% of US buyers put 20+% down on a house and which is up from 2022. Worse as a share of outstanding money new loans with low down payments obviously count for more. IE a 30 year loan on a house at year zero likely has 10+X as much outstanding vs a 25 year old loan due to principal payments and inflation on the average value of homes.

PMI is there to reduce risks, but PMI isn’t an infinite pool of money independent of the rest of the economy. It’s on someone’s balance sheet.

Maybe in silly-con valley.

Around most places the land goes up very slowly, but the house’s depreciation is reduced by steady repairs.

I can envision a future (maybe even the present in some areas) where increasing prices of skilled labor and/or materials results in a structure’s price increasing even though it is aging.

We have something somewhat like that where structure prices ARE increasing for a number of reasons - including the cost to build a new one.

I mean, the simple fact of inflation and urban flight during the covid spending bubble did indeed cause prices to go up quite a lot for many properties.