Existing home sales overcounting "minor": NAR

WASHINGTON Tue Feb 22, 2011 3:56pm EST

A vacant home for sale is pictured in Yonkers, New York, October 26, 2010. REUTERS/Mike Segar

A vacant home for sale is pictured in Yonkers, New York, October 26, 2010.

Credit: Reuters/Mike Segar

WASHINGTON (Reuters) - The National Association of Realtors said on Tuesday any overcounting of U.S. existing home sales in its data was "relatively minor", dismissing claims it could have overstated sales by as much as 20 percent.

The Realtors group said it was consulting with a range of experts to determine whether there was a drift in its monthly existing home sales data.

"At this point, we believe any drift in our data to be relatively minor," the NAR said.

The NAR issued the comment following a Wall Street Journal report that the NAR was examining the possibility that it overcounted home sales dating back as far as 2007.

According to the paper, NAR's home sales count was at odds with calculations by CoreLogic, a California real estate analysis firm. CoreLogic says NAR could have overstated home sales by as much as 20 percent.

"Our early discussions to assess any data drift have included CoreLogic and others. Since then, NAR has discovered factors which imply that CoreLogic's assumptions about existing home sales data drift are too high," the NAR said.

CoreLogic, which the NAR regards as a competitor, said it had confidence in its data.

"CoreLogic remains confident in the accuracy of our data for home sales and the validity of our methodology which relies on actual home sales data from public records," it said.

An overcount of home sales may mean that there is a bigger backlog of unsold homes and that it will take longer for the U.S. housing sector to climb out of its deep hole.

The U.S. housing market crash, sparked in part by shoddy lending practices, was at the heart of the economic meltdown that started in the United States and spread around the world.

While noting the difficulties associated with data measurements, housing analysts generally viewed the NAR's existing home sales as reliable.

"But you never know what is going on inside that black box. We don't know what is really going on inside CoreLogic's black box either," said Karl Case, economist at Wellesley College, who helped to develop the Standard & Poor's/Case-Shiller home price index.

Case noted the NAR was proposing downward revisions to existing homes sales data.

"And that has been a pretty good indicator I've thought over the years of what is going on in the market and I don't see any reason to doubt that it is telling us pretty much the truth," said Case.

U.S. home prices could fall another 25 percent from current levels, influential economist Robert Shiller said on Tuesday.

"My intuition rates the probability of another 15, 20, even 25 percent real home price decline as substantial. That's not a forecast but it's a substantial risk I think," Shiller, co-founder with Case of the Case-Shiller index, told reporters on a conference call.

U.S. single-family home prices fell for the sixth month in a row in December, bringing them closer to the low seen in 2009, according to the S&P/Case Shiller composite index of 20 metropolitan areas.

The NAR will release January's existing home sales report on Wednesday, which will also include annual revisions to sales rates and month's supply for the last three years.

In addition to reviewing its data, the NAR said it was working on a new methodology to re-benchmark data on a more frequent basis. The review will not, however, result in any notable changes to previous characterizations of monthly sales changes, or to price data.

(Reporting by Lucia Mutikani and Corbett Daly; Editing by James Dalgleish)