“Bubble Watch” digs into trends that may indicate economic and/or housing market troubles ahead.
Buzz: Are we seeing a significant early-spring slip in the pace of new-home purchases nationwide and in 13 Western states?
Source: April data from U.S. Commerce Dept.
Trend reported: U.S. new-home sales in April were off 1.5 percent from March. In Western states, sales were down 8 percent.
Dissection: Pundits often forget this monthly data is essentially a very rough guestimate. And the report clearly states the shortcomings — though such warnings don’t stop analysts and media types from jumping to broad conclusions without accepting the noteworthy limitations of this survey-based study.
You know that “plus or minus” margin-of-error measurement we often hear about when discussing, say, political polling? A typical accuracy spread in these voter-preference polls is plus-or-minus 3 percentage points.
Now, this survey of builders’ output has a huge variance — a caveat outlined in the report. It’s not that homebuying is “volatile” as some observers like to note. It’s that this initial estimate of builders’ sale success is, at best, a sneak peek at what’s happening and it’s also subject to large revisions down the road.
So how big is that margin of error? Huge!
Yes, U.S. new-home sales in April were off 1.5 percent from March … but that’s PLUS OR MINUS 11.8-percentage points — yes, 11.8 percentage points! That means buying was up as much as 10.3 percent in the one-month period or down 13.3 percent. That means the “result” was not “statistically significant,” in data geek speak.
As for year-to-date selling patterns, was the national trend really up 11.6 percent — the survey result — or up 35.3 percent (best case) or down 12.1 percent (worst case) when you toss the plus-minus into the math?
And the margin of error for Western states? Even larger! Take year-over-year data … Western developers were somewhere between popping champagne corks — sales up 56.6 percent, best-case scenario — or popping anti-depressants — sales down 18.8 percent, worst-case scenario.
Note: The Commerce Department report is blunt about its limits: “It takes six months to establish a trend for new houses sold. Preliminary new home sales figures are subject to revision due to the survey methodology and definitions used.”
PS: CoreLogic data for Southern California’s new-home sales, based on actual deal closings, showed purchases were down 1 percent in April vs. a year ago.
Another view: Analysts at IHS Market wrote “national and four regional estimates each failed the statistical significance tests” but concluded: “New home sales were somewhat softer than expected through April.”
How bubbly? On a scale of zero bubbles (no bubble here) to five bubbles (five-alarm warning) … ONE BUBBLE … because the trend hints of possible house hunter reluctance to buy.
