 |
| Robert Shiller - World Economic Forum Annual Meeting 2012 (Photo credit: World Economic Forum) |
Los Angeles, Dec.6, stock tips .- This is a very long piece from economist Tom Lawler on long term house prices. I've written about this before , The upward slope of Real House Prices, and Tom digs much deeper into the data!
From Tom Lawler:
One of the most widely abused home price series used by folks is the long-term “real” home price chart constructed by Robert Shiller in his “Irrational Exuberance” book. While he and others have sometimes characterized his “time series” of “real” home prices back to 1890 as being a good representation of “constant quality” home prices, in fact that is not even remotely the case. There are especially serious issues with the “older” home price series used by Dr. Shiller, with the “most troubling” being that used for the 1890-1934 period. Indeed, the authors of the book from which the home price index used by Dr. Shiller came from actually argued that this index did NOT reflect the behavior of “constant-quality” home price (with evidence to support that argument), and suggested using a materially different home price index. More on this point later.
When Dr. Shiller decided to explore home prices, and especially what appeared to be a housing “bubble,” he was “most surprised” that there were no “good” data going back prior to 1987 – with “good” defined as the Case-Shiller HPI.. But he wanted to “show” last decade’s house price runup in a long-term historical context. So he decided to take various home price sources, and attempt to construct a “long run” home price index by “concatenating” time series based on these sources.
Not surprisingly, from 1987 to the present Shiller used the “national” S&P/Case-Shiller home price index, which is a market-value weighted (using values from the various decennial Censuses, which unfortunately are not available for 2010. This “national” index is estimated to cover just over 70% of the US housing market.
From 1975 to 1987 Shiller used the “national” FHFA (formerly OFHEO) home price index, which is a unit-weighted index (using annually-updated estimates of states’ share of the housing stock) based on repeat sales transactions AND appraisal information on refinances of homes backing mortgages owned or guaranteed by Fannie Mae or Freddie Mac. Aside from the obvious “dataset” limitations (GSE only) and use of appraisals, the “unit-weighting” can result in materially different behavior than “value-weighting” over time.
From 1953 to 1975 Shiller used the home price index from the BLS’ Consumer Price Index, which was based on a sample of new home purchases financed with FHA-insured loans. It probably is not the case, however, that new home prices financed with FHA-insured loans, which often represent a very small sample of total home purchases, reflected trends in “overall” home prices. Indeed, the BLS characterized the data base used as representing “a small and specialized segment of the housing market and presents BLS with increasingly serious estimation problems . Here is an excerpt from a BLS paper discussing the 1983 change in the treatment of shelter costs for homeowners in the CPI.
“In the current (now ‘old’) CPI, the data are for house sales on which financing is insured by the Federal Housing Administration. These data fall far short of the ideal. Processing delays often mean that several months elapse between the time a house sale occurs and the time it is used in the CPI. For some geographic areas, especially those in the Northeast, the number of FHA transactions is very small. In addition, the FHA mortgage ceiling virtually eliminates higher priced homes from consideration. The impact of the ceiling- and especially changes in the ceiling - may be quite substantial, possibly resulting in a downward bias in the house price indexes used in the CPl.”
From 1934 to 1953 Shiller couldn’t find a good published source, but his grad students found median asking prices advertised in newspapers in five cities -- Chicago, Los Angeles, New Orleans, New York, and Washington, D.C. – and his home price index from 1934 to 1953 is a simple average of these median home asking prices.
From 1890 to 1934, Shiller used a home price index for 22 cities by Grebler, Blank, and Winnick (in their seminal book, “Capital Formation in Residential Real Estate: Trends and Prospects,” 1956) derived using data from the “Financial Survey of Housing” by the Department of Commerce in 1937. Here is a description of the question asked of home owners. ... Continue to read.
No comments:
Post a Comment