That is a really good point.

I also have a pet theory that CPI understates the cost of housing due to people who are pushed out of entering the market altogether (e.g. homelessness rising, people living with parents longer)

Sort of the same way some people say unemployment figures understate the true number by ignoring discouraged workers / people who have stopped looking altogether. I have a hunch there are many discouraged folks who have stopped looking for housing altogether.

You should take a look at Pia Maloney and Eric Weinstein work.

CPI is broken because it doesn't factor in a "change in taste or prefence"

CPI treats individuals as a giant unmovimg population mean.

To counter they change to use vector fields.

Of course preference is included in CPI. Weightings are adjusted periodically as consumer spending patterns shift, and price elasticity is explicitly factored on a quarterly basis in chained cpi

Housing is not in CPI. They use "owners' equivalent rent" which is a subjective (cooked) metric.

Well, it’s based on a survey. (And surveys are cursed instruments [1].)

> Generally, owners' equivalent rent is obtained through surveys asking homeowners the following question: "If someone were to rent your home today, how much do you think it would rent for monthly, unfurnished and without utilities?" [2]

I would guess that some owners’ ideas of what their house would rent for might lag the market?

[1] https://carcinisation.com/2020/12/11/survey-chicken/

[2] https://www.investopedia.com/terms/o/owners-equivalent-rent....

So what housing cost metric should be looked at? The cost for a retiree living in their paid off home? The mortgage payment for brand new homeowner with minimal down? The cost to split an apartment for 2 people? What objective measure should be used?

CPI will never been perfect. They could even break it down into 3 categories: 1. Poor man’s CPI: ramen, rent & public transit 2. Middle class’s CPI: house, beef & cars 3. Rich man’s CPI: penthouse, luxury goods & sports teams

All of them? The government has this data already it just needs to be aggregated by someone motivated.

How do you combine them into one number, in a way that isn't "cooked"? Proposing we mash up all data ever into one figure, without detailing how, is useless

Why would they necessarily be combined?

If you look at the Bureau of Labor and Statistics they will happily provide you with more detail, for instance the median salary in a field for the top 10% of employees versus the top 90%.

I don't know how you'd keep that updated though, you can hardly say "the amount the richest 10% spend is the rich CPI" because it's a silly statistic. I don't know how you can reasonably say "the minimum amount required to be rich", it just seems silly on its face.

No, I don't think there's much of a way to make it work. It's probably more directly useful to understand whether people are saving money or how much debt they have, though that's going to lag.

I mean, or just ask people. The government is fairly good at doing that.

I don't think they have to ask anyone. IRS and other agencies could pull all this information. IRS knows how much rent my landlord makes. County knows how many units are on the parcel my landlord owns and the landlords property tax burden. Lender reports mortgage data to the government, now you know my landlords mortgage. By aggregating just this info, most of which like the county parcel data already being public, you can get the true costs towards housing paid by everyone in the country living in housing.

For income once again you can turn to IRS tax return data. I believe IRS can also sum your accounts as well as review transaction history. Creditors also report debts to IRS.

All told I believe the federal government can access all the information needed to know exactly what almost every American is paying for housing, how much they are being paid, how much they are spending, and how much debt they have.

Medians aren't crooked.