Tuesday, February 19, 2008

Tracking Housing Prices, Why The Numbers Conflict

Courtesy of WSJ.com

(See Corrections & Amplifications item below.)

By David Wessel
From The Wall Street Journal Online

Predicting how much worse the U.S. housing market will get is tough. The future is never certain. But when it comes to home prices, getting a clear picture of the recent past turns out to be surprisingly hard as well.

That's confusing to homeowners, who fret about the value of what for many is their single largest asset. There is a huge psychological difference between a slower climb in the value of one's house and an outright decline -- and, as a result, a difference in the political reaction.

Tracking home prices is harder than tracking the price of stocks, which are traded constantly in public view on exchanges. And it's harder than tracking the price of toothpaste. That just involves sampling posted prices on grocery-store shelves and Web sites.

The two best -- though far from perfect -- measures of housing prices are the Office of Federal Housing Enterprise Oversight's index and the gloomier Standard & Poor's Case/Shiller index. Both are based on a concept, developed in the 1980s by Karl Case of Wellesley College and Robert Shiller of Yale University, that looks at repeat sales of the same houses.

Ofheo's index says home prices rose nationally by 1.8% between the third quarters of 2006 and 2007. But the S&P/Case-Shiller national index of home prices was down 4.5% in the same period. The Ofheo index showed a 2.16% increase in house prices in Chicago; the Case-Shiller index showed a decline of 2.48%.

Those discrepancies persist even though both barometers avoid distortions that occur in other widely cited measures -- such as the National Association of Realtors' median home price -- that reflect the mix of homes actually sold in a given month as well as the change in prices. Such measures rise in months when a lot of high-end houses are sold and fall at times when a lot of low-end houses are sold.

The Realtors' measure fell 6% in 2007. The group says the index was pulled down by a drop in the number of high-end home sales, which have been hurt by disruptions in the market for mortgages exceeding $417,000, the maximum mortgage giants Fannie Mae and Freddie Mac are allowed to guarantee.

The big picture here is clear: House prices rose rapidly in the early years of this decade. They have stopped rising in many places. And, in many markets, they are now falling. (Even Ofheo's index showed a quarterly decline at the end of 2007.) And prices don't appear to have touched bottom yet. But Charles Calomiris, a Columbia University economist, says, "Too much weight is being attached to the Case-Shiller index. ... Housing prices may not be falling as much as some economists say they are."

With house prices so central to the economy right now, there is intense public (as well as scholarly) interest in why these two carefully constructed measures differ.

Ofheo gets a steady stream of inquiries from ordinary homeowners trying to figure out what's happening to the price of their houses, and offers an online calculator to make estimates. Ofheo's quarterly numbers -- to be released monthly beginning in March -- go into the Federal Reserve's estimates of household wealth. Case/Shiller is increasingly prominent and is the basis for future contracts that allow investors to bet on the price of houses.

There are a couple of very big differences. The Ofheo index relies on data collected by Fannie Mae and Freddie Mac, which Ofheo regulates, so it excludes loans too big for Fannie and Freddie to guarantee (those exceeding $417,000) or too shaky (the riskiest of the subprime). Case/Shiller includes those, but its data are limited to 20 major markets because it relies on the costly process of going to local property records for data. One of Mr. Calomiris's complaints is that house prices in these markets may be doing worse than those in other places.

A recent dissection of the two indexes in 10 metropolitan areas by Ofheo economist Andrew Leventis, posted on the agency's Web site, sheds some light on other differences. Part of the discrepancy is technical, such as different approaches to adjusting data when there's a long interval between repeat sales of a house.

But puzzles remain. It turns out, for instance, that prices of low- and moderate-priced homes with mortgages that aren't guaranteed by Fannie and Freddie are falling particularly sharply, buoying the Ofheo index -- even though that index includes plenty of other of low- and moderately priced homes in the same neighborhoods.

Of course, by the time the experts get the measures perfected, we'll be onto a bubble in some other asset market.

Email your comments to rjeditor@dowjones.com.

-- February 15, 2008
Corrections & Amplifications:

In addition to its widely followed 20-city survey of home prices, S&P/Case-Shiller publishes a national home price index based on data from more than 100 metropolitan areas.

No comments: