As BusinessWeek asked way back in 2005 before the bubble burst:
Wondering why inflation figures are so tame when real estate prices are soaring? There is a simple explanation: the Consumer Price Index factors in rising rents, not rising home prices.
Are we really getting a true reading on inflation when home price appreciation isn’t added into the mix? I think not.
I find the idea that house price appreciation and depreciation is not factored into inflation figures stunning. For most people it’s their single biggest lifetime expenditure, and for many today mortgage payments are their single biggest monthly expenditure. And rental prices (which are substituted for house prices) are a bad proxy. While house prices have fallen far from their mid-00s peak, rents have continued to increase:
Statisticians in Britain are looking to plug the hole. From the BBC:
A new measure of inflation is being proposed by the Office for National Statistics (ONS).
It wants to create a version of the Consumer Prices Index that includes housing costs, to be called CPIH.
The ONS wants to counteract criticisms that the main weakness of the CPI is that it does not reflect many costs of being a house owner, which make up 10% of people’s average spending.
While a welcome development (and probably even more welcome on the other side of the Atlantic) it doesn’t make up for the fact that the explosive price increases during the boom years were never included. And it isn’t just real estate — equities was another market that massively inflated without being counted in official inflation statistics. It would have been simple at the time to calculate the effective inflation rate with these components included. A wiser economist than Greenspan might have at least paid attention to such information and tightened monetary policy to prevent the incipient bubbles from overheating.
Of course, with inflation statistics calculated in the way they are (price changes to an overall basket of retail goods) there will always be a fight over what to include and what not to include.
A better approach is to include everything. Murray Rothbard defined inflation simply as any increase to the money supply; if the money is printed, it is inflation. This is a very interesting idea, because it can reflect things like bubble reinflation that are often obscured in official data. The Fed has tripled the monetary base since 2008, but this increase in the monetary base has been offset against the various effects of the 2008 crash, which triggered huge price falls in housing and equities which were only stanched when the money printing started.
Critics of the Austrian approach might say that it does not take into account how money is used, but simply how much money there is. An alternative approach which takes into account all economic activity is nominal GDP targeting, whereby monetary policy either tightens or loosens to achieve a nominal GDP target. If the nominal target is 1%, and GDP is growing at 7%, monetary policy will tighten toward 1% nominal growth. If GDP is growing at a negative rate (say -2%), then the Fed will print and buy assets ’til nominal GDP is growing at 1%. While most of the proponents of this approach today tend to be disgruntled Keynesians like Charles Evans who advocate a consistent growth rate of around 5% (which right now would of course necessitate the Fed to print big and buy a lot of assets, probably starting with equities and REITs), a lower nominal GDP target — of say, 1% or 2% — would certainly be a better approach to the Fed’s supposed price stability mandate than the frankly absurd and disturbing status quo of using CPI, which will always be bent and distorted by what is included or not included. And for the last 40 years monetary policy would have been much, much tighter even if the Fed had been pursuing the widely-cited 5% nominal GDP target.
I don’t think CPI can be fixed. It is just too easy to mismeasure inflation that way. Do statisticians really have the expertise to determine which inflations to count and which to ignore? No; I don’t think they do. Statisticians will try, and by including things like house prices it is certainly an improvement. But if we want to be realistic, we must use a measure that reflects the entire economy.