November 16, 2010
Measuring Inflation: Shelter vs. Housing
By Cindy Ivanac-Lillig
One of the most common measures of inflation is the Bureau of Labor Statistics' (BLS) Consumer Price Index (CPI). This is a broad based measure of price levels in our economy. Many have argued that the Fed missed the recent housing bubble because it relied too heavily on the CPI, which does not capture home values. Although home values are not included, the CPI does take into account the value of "shelter."
The two main components of the CPI's "shelter" value are actual rents and owners' equivalent of rent. The latter figure is an estimate of the value of the home to the homeowner as measured by what they would hypothetically pay in rent. So in short, it is true that the asset value of a house is not included in the CPI. If housing prices began to rise, CPI would not reflect this increase unless it translated into rising rents in the area.
Rent equivalents are estimates and require a fair amount of judgment and careful sampling to obtain. If you are more interested in the methodology, check out the BLS site.
In general, price level indices, such as the CPI, are not designed to capture asset prices, but rather prices of things everyone uses ("basket of goods and services"). Assets are traditionally viewed as investment vehicles. But after this crisis, many are asking if asset prices need to be looked at either in a combined/hybrid index or separately. How much of an impact do asset prices have on economic activity in the short run? It seems to me that most economists still think that price indices should be kept to just the current basket of goods and services, but there are those out there that argue the opposite -- check out this article from the Centre for Economic Policy Research.
What do you think?