That’s not what it says. It’s that only 7% of inflation is attributed to housing. Ask the question: if you buy a house, either with a mortgage or without, does your cost of “rent” increase every year? No?
(Yes, other costs may increase, your property taxes, perhaps maintenance, but otherwise the base cost is the same year after year. Why should the overall calculation of inflation for everyone include a fictitious cost which you don’t pay?)
(Let’s take it a step further: the UAW bargains for COLA in their contract. Well, COLA is calculated on the fictitious “rent increase” that is baked into this number, so it actually increases wages more than the actual rise in inflation. That, of course, pushes actual inflation higher, leading to round 2. Lather, rinse, repeat.)
I don’t know what the correct methodology is, but including a cost which isn’t an actual cost doesn’t seem to be the right way to approach it.
But that is exactly how it works for other hard goods, including cars, washing machines, TVs, etc. Inflation stats look at the increase in cost for new purchase , not some theoretical cost you (don’t) pay for using your 5 year old TV. Housing costs, it seems, could follow that model: use actual housing costs (rents and housing price increases for similar square footage & amenities, nationalized to account for geographic differences). Why would this be bad?