It looks like they applied CPI directly to the cost of housing. That is not correct. CPI has a component of "owner's equivalent rent", which should be removed

It looks like they applied CPI directly to the cost of housing. That is not correct. CPI has a component of "owner's equivalent rent", which should be removed before cost-correcting the price of housing. Otherwise, you are removing the increase of the item you are searching for an increase of (Housing accounts for 1/3 of the CPI correction.) Their graph is instead showing that the buying power of a dollar is roughly constant when 1/3 of the CPI basket-of-goods is corrected with CPI. Calculated Risk has a more sane analysis, you will note that the graphs are corrected with "CPI less Shelter": http://www.calculatedriskblog.com/2016/11/real-prices-and-price-to-rent-ratio-in.html http://www.calculatedriskblog.com/2016/11/real-prices-and-pr... He doesn't include price/sqft, but note that there is a 44% increase in real prices between the 90s and now. Square footage has not gone up by that much in 20 years. As a side note: AEI is a fairly political org, I wouldn't rely on their accuracy for anything involving stats, or math in general.