AP Reporter's Tone and Stats Obscure Housing Market in Possible Recovery

Yesterday's Existing Home Sales report for February issued by the National Association of Realtors had better than expected news: On an annualized basis, sales were up. They were expected to go down. Someone interested in getting to the bottom of things would have found that the improvement reported by the NAR may be an early indicator a broader recovery in existing-home unit sales and sales prices.

That appears to be the last thing the Associated Press's Martin Crutsinger was interested in yesterday. In his report, he instead seemed determined to do all he could to portray the increase as a one-month respite in a long-term gloomy scenario. Additionally, he, in my opinion, presented changes in annualized sales volume as if they were one-month changes in actual sales, causing readers to possible believe that the housing market remains more in the doldrums than it really is.

Here is how his report began:

Sales of existing homes increased unexpectedly in February after six months of decline, but private economists said it was too soon to say the prolonged slide in housing is coming to an end.

The National Association of Realtors said sales of existing homes rose by 2.9 percent in February to a seasonally adjusted annual rate of 5.03 million units. It marked the first sales increase since last July, but even with the gain sales were still 23.8 percent below where they were a year ago.

Prices continued to slide. The median sales price for single-family homes and condominiums dropped to $195,900 (not seasonally adjusted -- Ed.), a fall of 8.2 percent from a year ago, the biggest slide in the current housing slump. The median price for just single-family homes was down 8.7 percent from a year ago, the biggest decline in four decades.

"The hemorrhaging has stopped but the recovery will be long, slow and painful," said Bernard Baumohl, managing director of the Economic Outlook Group. "It's unlikely that we will see any sustained jump in home purchases, must less higher prices, until mid 2009 at the earliest."

Brian Bethune, an economist at Global Insight, said, "A quick bounce-back in the housing markets is simply not in the cards."

The "not seasonally adjusted" note I inserted is there for two reasons: First, anyone who downloads the related NAR spreadsheet (the download link for it is at this NAR page) will see that the sales-price data is clearly labeled "not seasonally adjusted." Since Crutsinger correctly labeled the sales rate of 5.03 million as seasonally adjusted, he should have changed the label when switching topics to the national median sales price.

The second and more important reason for the note above is that Crutsinger erroneously gave his readers the impression that the sales figures he recited later in his report were actual one-month sales changes:

By region of the country, sales surged by 11.3 percent in the Northeast and were up 2.5 percent in the Midwest and 2.1 percent in the South. The only region of the country to see a sales decline was the West, where sales dropped by 1.1 percent.

But the percentages he cited are not the one-month changes in actual regional sales -- they are changes in much-less volatile annualized, seasonally adjusted sales.

With all due respect, readers deserve to know that changes in actual sales in February compared to January, and compared to February 2007, were as follows (figures Crutsinger reported [or would have reported had he disclosed every region's annual sales change] are in parens for comparison purposes):

South:
Feb. v. Jan., +8.7% (+2.1%)
Feb08 v. Feb07, -18.3% (-22.0%)

Midwest:
Feb. v. Jan., +22.6% (+2.5%)
Feb08 v. Feb07, -14.6% (-19.5%)

Northeast:
Feb. v. Jan., +24.4% (+11.3%)
Feb08 v. Feb07, -23.3% (-26.4%)

West:
Feb. v. Jan., -1.8% (-1.1%)
Feb08 v. Feb07, -23.6% (-29.2%)

All USA:
Feb. v. Jan., +12.2% (+2.9%)
Feb08 v. Feb07, -19.4% (-23.8%)

Now stay with me on this: Actual sales of existing homes in February, the figures Crutsinger is purporting to be passing on to his readers, were really 12.2% higher than than they were in January, compared to the 2.9% the AP reporter claimed. It's hard to gauge the real-world significance of the actual February vs. January increase, because the NAR spreadsheet did not have January 2007 data. But it may be that the increases noted, especially those of over 20% in the South and Midwest, are significant enough to lead to firmer home prices in future months -- sooner instead of later. These are the kinds of things AP reporters are supposedly being paid to find out.

Comparing February of 2008 to February of 2007, you'll see that every region of the country had a smaller decrease in year-over-year unit sales than the one Crutsinger reported for the entire country.

Either what you see above is what Crutsinger should have reported, or he should have clearly said that the sales figures he was presenting were annualized.

It "just so happens" that the actual sales figures are, with only one exception (the West, Feb. v. Jan.), either more favorable or less unfavorable than those the AP's Crutsinger reported.

How convenient.

Cross-posted at BizzyBlog.com.