Wednesday, September 21, 2011
I've seen a lot of talk lately about how the CPI is way over-stating inflation, since its largest component—owner's equivalent rent—has increased over 7% since the end of 2006. Housing has contributed to inflation? How can that be if we all know that home prices have collapsed. Ha, ha.
The chart above helps to understand what is really going on. Due to the extreme volatility of housing prices in the 1970s, in the BLS decided in the early 1980s to switch from using home prices as an input to the CPI to instead using "rents," since they don't tend to change as much. This decision remains controversial, but as the chart shows, rents (which are estimated) have been much more stable than prices, and over time the two have tracked pretty well. If the CPI is overstating inflation today because it uses rents instead of prices, then it was hugely understating inflation in the late 1990s and early 2000s. Regardless of whether you had used housing prices as an input to the CPI, or rents, by now the cumulative amount of inflation you get is roughly the same. The difference between prices and rents evens out over time, which stands to reason.
This chart uses all the available data from Case Shiller, and that data only includes the prices of the 10 largest metropolitan markets, so the fact that prices have increased about 16% more than rents since 1987 is not to be given too much importance. The important thing is to see how the two series track each other over time. It's my understanding that rents today (which are widely reported to be rising) are becoming expensive relative to the cost of purchasing a house. So an ideal version of this chart might show that owner's equivalent rents have risen more than housing prices over the past several decades. But in any case, it disproves the allegation that the BLS is mis-reporting inflation. Rents and housing prices have come back into line, after a period in which housing prices were grossly inflated.
Posted by Scott Grannis at 11:18 AM