" From what I recall, renters are surveyed for their actual rents, and homeowners are asked to estimate what their house might rent for, and those numbers are t

" From what I recall, renters are surveyed for their actual rents, and homeowners are asked to estimate what their house might rent for, and those numbers are taken in to account in CPI." Hence my original point that housing costs are included through a heuristic. The point being that a survey asking "how much do you think you could rent your house for" is really un-scientific. If I'm not participating in the market, there is no market, and the price is a guess. Is that guess good within an order of magnitude? Sure. Is it good within 5%? No.