Flawed Housing Forecasts

Yesterday morning at 10 a.m., CNBC real-estate correspondent Diana Olick hit the gray streets of Washington, D.C., to report the latest housing data from the National Association of Realtors. (Strange, I was able to get the data while remaining inside.) The news? The Pending Home Sales Index, an indicator of future activity, edged up in October, although it was still off 18.4 percent from October 2006. Olick also dutifully reported NAR's annual year-end forecast, which was picked up by the wire services. While the housing market may be in the dumps, said NAR economist Laurence Yun, "Now that mortgage conditions have improved, some postponed activity should turn up in existing-home sales over the next couple of months, and I expect sales at fairly stable to slightly higher levels."

Economic forecasting is exceedingly difficult. The consensus estimates compiled by the Wall Street Journal and other outfits on measures like GDP growth and unemployment are frequently incorrect. As a profession, economists project growth when a recession is about to start and project recessions when the economy is poised for continued expansion. And as stand-up economist Yoram Bauman puts it, "Macroeconomists have successfully predicted nine of the last five recessions." There are some institutional reasons for this: Many economists are associated with corporations, Wall Street firms, and trade groups, where it doesn't pay to be bearish. Others fall into the trap of extrapolating existing trends into the future. But given the complexity of the contemporary world, the huge range of variables, the unrelenting flow of data, and the fallibility of humans, it's likely impossible to forecast consistently with any accuracy. And it's especially difficult to project economic activity when the economy reaches inflection points—times when the economy is about to go from expansion to contraction, or vice versa.

But within the fraternity of financial and fiscal forecasters, the seers at the National Association of Realtors—longtime chief economist David Lereah and his successor Lawrence Yun—may be uniquely ill-equipped to deliver sobering forecasts. They work for a trade group whose mission is to buck up the spirits of real-estate brokers. And real-estate brokers—who live to sell, promote, and market—are constitutionally disinclined to hear anything but good news. Indeed, as I noted last summer, Lereah's penchant for putting out positive spin on dismal housing numbers inspired a blog and led critics to dub him the Baghdad Bob of real estate. Lereah has moved on. But Yun has picked up where he left off.

In addition to claiming that the sun is shining brilliantly even as rain pours down from the heavens in a mighty stream, Lereah and Yun have also hazarded optimistic, educated guesses about the future. In February 2005, Lereah published a book that is my candidate for Longest Title Ever: Are You Missing the Real Estate Boom?: The Boom Will Not Bust and Why Property Values Will Continue To Climb Through the End of the Decade—And How To Profit From Them. Naturally, the boom busted soon after publication, and property values began to descend.

Clearly, the housing market encountered an inflection point in 2006, with housing prices and activity peaking. But you wouldn't have known it from looking at NAR's December 2005 forecast for 2006, in which the group said existing-home sales would fall 3.7 percent to 6.84 million and new-home sales would fall 4.8 percent to 1.23 million. Modest declines after an impressive multiyear run. Instead, as the data from December 2006 show, existing-home sales for 2006 fell to 6.47 million, down 8.6 percent (more than twice the rate forecast), while new-home sales fell 17.8 percent to 1.06 million. Last December, the 2007 forecast was once again guardedly optimistic. "Most of the correction in home prices is behind us, but general gains in value next year will be modest by historical standards," said Lereah. For 2007, existing-home sales were "expected to rise steadily from the current cyclical low and reach an annual total of 6.4 million, which would be 1.0 percent lower than this year's total." NAR wasn't so sanguine about new-home sales, however, projecting they would fall 9.4 percent in 2007 to 957,000. But once again, the forecasts were far too optimistic. As NAR reported today, existing-home sales are projected to come in at 5.6 million for 2007—a fall of 12.3 percent, not the 1 percent fall projected last December. New-home sales for 2007 should amount to 788,000—down about 25 percent from 2006 and far below the projection made a year ago.

For next year, NAR projects existing-home sales will rise ever so slightly to 5.7 million, and that the median home price should rise 0.3 percent. New-homes sales, however, are expected to continue to fall sharply, to 693,000, which would represent a 12 percent drop. The forecast presumes that for the largest chunk of the market—the existing-home-sales market—the bottom has come, and it sits poised at an inflection point. It's possible. But the tendency of housing professionals to call the housing bottom for the past two years—again and again—has been pretty dismal. The good folks at NAR aren't the worst forecasters ever. That title goes to the two guys who, in 1999, projected the Dow Jones Industrial Average would hit 36,000 in three to five years. But given the recent history and the slope of the trend lines, it's probably prudent to prepare for the worst rather than hope for the best.