Continuing the business press's slavish devotion to seasonally adjusted figures in government reports to the exclusion of looking at what actually happened, Martin Crutsinger at the Associated Press, aka the Administration's Press, began his Tuesday dispatch on May's new-home sales report from the Census Bureau as follows: "Sales of new homes rose in May to the fastest pace in five years, a solid gain that added to signs of a steadily improving housing market."
Except for two "little" things: Fewer homes were actually sold in May than were sold in April, and May's reported increase in seasonally adjusted annualized sales only came about because of a tax break which ended in April 2010:
(Links: Actuals; seasonally adjusted annual rates)
Pending possible revisions, 46,000 homes were sold In April; 45,000 were sold in May, a 2.2 percent decline. Yet annualized sales increased by 2.1 percent.
The seasonally adjusted annual rate increased artificially primarily because of just one of the five years used in the seasonal calculations. In May 2010, new home sales dove to 26,000 after a relatively strong 41,000 in April because of the expiration of new homebuyer tax breaks. That makes the May 2013 drop not only not look so bad, but seem better than "normal" -- hence the seasonally adjusted increase.
The track record of the past 50 years for April to May changes in new-home sales has about as many increases as decreases, and four instances of no change. This year's actual May decline contradicts the core assertion in Crutsinger's opening paragraph, namely that May result was "a solid gain that added to signs of a steadily improving housing market."
No it wasn't. As I noted yesterday:
This housing “recovery” is tepid at best. The current new-home market is better than it has been in any of the past five years, but worse than every year on record preceding it, with only 1982 being a borderline call.
This news won’t be presented that way in the press.
And, as seen, it wasn't.
Cross-posted at BizzyBlog.com.
