The other piece of the housing puzzle relates to something that the media seems to ignore. We never seem to get anyone to look at new home sales and new home construction together. Each is usually analyzed separately. It's easy to see why - if we look at the data. If homebuilders started construction on new homes at an annualized rate of 582,000 in June and new homes are being sold at an annualized rate of 384,000 in June - what does this tell you? It tells you that homebuilders are building approximately 50% more homes than they are selling - and this is now happening month after month after month.
This leads to some obvious observations - home inventories/vacancies must be increasing - and this is exactly what we're seeing. I'm also starting to see articles about banks that are now bulldozing new home construction - since they can't sell the properties (http://www.cnbc.com/id/30580830).
As Chris mentions below - because the new home sales number is a 'sampling' - the 11% 'spike' is + or - 13.2%. This means that the number could be -2.2% or +24.2% - or any number in between. Makes you wonder if the data is even worth reporting. Media outlets could care less - and trumpet this 'good news'. The bottom line is that if we compare sales to a year ago - we're looking at a 20%+ decrease.
The media spin machine continues........
jg – July 28, 2009
In Session - New home sales: 'Really good news'
By: Chris Martenson
New home sales: 'Really good news'
By Les Christie, CNNMoney.com, July 27, 2009
NEW YORK (CNNMoney.com) -- Sales of newly constructed single-family homes spiked 11% in June to an annualized rate of 384,000 homes, according to a report released Monday.
The gain over May was much greater than expected.
A consensus of housing industry analysts had forecast seasonally adjusted sales of 352,000, according to Breifing.com.
However, sales are still 21% below the levels of a year ago, when new homes sold in June at an annualized rate of 488,000, according to the report released by the U.S. Department of Housing and Urban Development. Four years ago, during the height of the housing boom, the sales rate for June was 1,374,000, nearly three-and-a-half times higher than last month.
"Spiked" and "much greater than expected" sure make things sound great. While I think it is plausible to begin to look for a bottom somewhere in housing (personally I think it is still a year or more away) I do object to the use of hyperbole to paint an inaccurate picture of the situation.
The data is horribly noisy and, worse, it is sampled data that is seasonally adjusted and has serious methodology issues (e.g. cancelled orders are not removed from the reported figure) making the Census Bureau New Home Sales figure among the least reliable of them all.
At least they have the good graces to tell us as much right on the release itself:
Sales of new one-family houses in June 2009 were at a seasonally adjusted annual rate of 384,000, according to estimates released jointly today by the U.S. Census Bureau and the Department of Housing and Urban Development. This is 11.0 percent (±13.2%)* above the revised May rate of 346,000, but is 21.3 percent (±11.4%) below the June 2008 estimate of 488,000.
The reported figure of 11% above last month is, uh, plus or minus 13.2%. In other words the release tells us flat out that their methodology is inaccurate enough that the correct answer could just as easily be +24.2% as is could be -2.2%. Or maybe it's +11%. Hard to say for sure. All we know is that each of those figures is equally likely under the loose methodology employed by the Census Department.
At any rate, below is the chart that, for some reason, never accompanies a mainstream news financial article but which the Wall Street pros get to check out before they make their decisions. I personally do not (yet) see anything here to suggest a bottom or turnaround.
Does that look like a "spike" to you? How would you characterize that wiggle at the end there? Is it any more dramatic than any other wiggle on the chart?
Anyone care to go long a double fistful of homebuilder stocks based on what you see here?