Reality Index · category
The single biggest category in the Reality Index, and the place CPI's methodology choices have the largest impact on the headline number.
Housing is where weighting, not the price measure, drives the gap. Home prices grew 6.84× from 1980 to 2025 per FHFA's House Price Index, and rent grew 5.37× per BLS's own CPI subindex — but most owners carry older mortgages or own outright, so what households actually pay for shelter rises far slower than home prices. Tenure-weighted, the Reality Index housing composite grew 5.10× — close to CPI's own shelter measure, versus CPI All Items at 3.91×. The Reality Index's gap with CPI comes from how much weight housing carries in the basket, not from a dispute over housing prices.
The methodological argument: CPI uses Owner's Equivalent Rent (OER) — what homeowners would theoretically pay if they were renting their own home — instead of measuring actual home prices and mortgage costs. OER systematically understates housing inflation when home prices rise faster than rents, which is essentially always since 1995. The Reality Index headline rate uses FHFA home prices plus Freddie Mac mortgage rates for housing carrying cost, not OER.
Each card below links to the full chart page for that item — including retail dollar series, BLS CPI subindex for that category, and the long-history backstop where available. Multiples shown are 1980-anchored unless the underlying data series doesn't extend that far back.
Home prices
FHFA House Price Index
Rent of primary residence
BLS CPI subindex
Composite construction (v0.2). A tenure-weighted blend of what owners and renters actually pay each year, indexed to 1980 = 100. The owner side combines a new-buyer carrying cost (10%) and an existing-owner carrying cost (90% — a 12-year vintage moving average of past mortgage rates, so most owners carry older, cheaper financing). The renter side uses BLS Rent of Primary Residence (which rose 5.37×) spliced with Zillow ZORI from 2015 to track market rents more currently — the spliced renter rose 5.71×, since BLS rent lags the market. Owner and renter are weighted by the Census homeownership rate each year. The 2025 composite is 5.10×, close to CPI’s own shelter measure.
What the headline rate uses. The Reality Index headline uses this tenure-weighted composite (5.10×) — the population measure of what households actually pay for shelter. The new-buyer carrying cost alone — principal + interest on an 80% LTV mortgage at each year’s PMMS rate, plus property tax (1.1%), insurance (0.35%), and maintenance (1.0%) on the FHFA-scaled median home — grew just 4.30×, because mortgage rates fell from 1981’s 13.7% peak to today’s ~6.6%. But new buyers are only ~10% of owners in any given year, so the headline weights them lightly. (The American Dream Index, which measures entry cost for a new buyer, uses that 4.30× figure instead.)
Why OER is contested. Owner's Equivalent Rent is the largest single component of CPI (about 26% of the headline rate). Its construction — survey homeowners about what they would pay to rent their own home, smooth the answers, deflate by amenity-adjustment — is methodologically defensible but produces a number that has lagged actual home price growth by ~3% per year since 2000. Critics have argued for decades that this single methodology choice is the largest source of CPI's understatement of real cost-of-living growth.