[vivafrance]

You’re wrong. House prices aren’t included anymore in CPI. Imputed rents are. These are very different things.

Anyways, I’m sure all this falls on deaf ears. If you used the same calculation the Fed used in the 80s to measure inflation we’d currently be at 10%. You can pick and choose and weight whats in the cpi basket to get any number you want and the government is incentivized to make it appear lower.

[JumpCrisscross]

> House prices aren’t included anymore in CPI. Imputed rents are

CPI calculates something called "owners’ equivalent rent of residences" [1]. This is reasonable as nobody purchases a new house every year; instead, one "uses" a portion of the home value over time. When house prices go up, this measure goes up by a similar measure.

Housing was put into CPI in 1954, when it was included as a user-cost item. It was replaced in the early 1980s because user-cost methods include "ex ante expected gain" while usage pricing "includes actual ex post realized capital gains on the house" [1].

> If you used the same calculation the Fed used in the 80s to measure inflation we’d currently be at 10%

Well, yes. You'd be excluding everything invented since 1980, e.g. all modern technology. We don't spend the same fraction of our budget on hams and eggs, as the 1980 definition measured, and most people have health insurance costs now.

If someone insists on living like it's 1980, I suppose observing the old metric would be perfectly valid.

[1] https://www.bls.gov/opub/hom/pdf/homch17.pdf page 104

[2] https://www.brookings.edu/wp-content/uploads/1980/06/1980b_b... page 558

[vivafrance]

> When house prices go up, this measure goes up by a similar measure.

That’s only true if you assume interest rates aren’t falling. Artificially low interest rates have inflated the actual price of housing while keeping monthly payments fairly steady.

The Fed is then able to say “look, no inflation!” despite actual prices rising very rapidly.

[JumpCrisscross]

> Artificially low interest rates have inflated the actual price of housing while keeping monthly payments fairly steady

There's a case to be made for each approach. Given we have excellent ways of measuring house prices, but no great ones for cost of living, having CPI measure the latter and thus usage--instead of user costs--seems reasonable. (We don't include stock appreciations and declines in inflation, for example.)

[djrogers]

Those interest rates didn’t fall over the past decade, they stayed very consistently low, so the rise in house costs caused a rise in mortgage payments and imputed rents.