While commodities and education are being undercounted, housing actually lowers inflation readings. Tony Crescenzi addressed this relationship earlier this year in an article called " How Housing's Surge Is Suppressing CPI ." The way BLS accounts for housing expenses is an oddity that creates all sorts of problems. Before 1983, CPI measured housing inflation by looking at what it actually cost to own a home: house prices, mortgage rates, property taxes, even maintenance. After 1983, BLS changed the housing component, using the concept of "owner's equivalent rent." It's a measure of what homeowners could get for their homes if they rented them. It accounts for 23% of the overall CPI and about 30% of core prices, according to BLS. Since the housing market began soaring, rental properties have languished. Vacancy rates rose, and rents came down in price. This had the surreal effect of pushing CPI measures down. At exactly the time housing became extremely expensive, the BLS measure of this component made inflation appear to be going lower. What happens to the CPI inflation if we back out this absurdity? Tim Iacono, host of the blog The Mess That Greenspan Made, did the number crunching. As the chart below shows, if we revert back to the pre-1983 measure of housing costs, the core CPI spikes to a more realistic 5.3%.
The CPI's miscalculation is even worse than you might imagine. As utility costs go up, inflation is further under-reported in the CPI (core or headline). Why? BLS perversely removes the value of any landlord-provide utilities in its calculation of owner's equivalent rent. And you wondered why I mock the Labor Department's statistical methodology. Talk about adding insult to injury.