The Government's New Home Sales Report For January: Either Fraud Or Incompetence

The Census Bureau released its new home sales report for January today.  It showed nearly a 10% increase in sales from December to January and an increase over January 2013.  The only problem with this report is that it has holes in the numbers that are wider than the Mariana Trench is deep.

To begin with, please keep in mind that the headline numbers reflect a seasonally adjusted annualized rate (SAAR).  This means that the numbers collected by the Census Bureau are fed into a statistical model that spits out a result and we have no idea whatsoever how the result was calculated.  This is common across all Government economic reports and results in a high degree of reporting errors and bias to the upside, especially when a rising trend is followed by declining trend, such as is the case with the current housing market.

Instead of looking at the SAAR, it's more useful for analyzing the data by looking at the unadjusted monthly data, which is included in the Govt report  - LINK - but never reported by the media or discussed by Wall Street analysts.  As I'll show, it is this aspect of the data that is an inconvenient truth and I suspect it will eventually be removed from the report, just like the Fed removed M3 from its reports.

If you look at the link, you'll see that in January a total of 34,000 homes were preliminarily estimated to have been "sold."  I say "sold" because the Census Bureau records a sale when a contract is signed - not when a home is delivered, escrow clears and title is transferred.  Currently most big homebuilders are reporting a 25% cancellation rate on homes "sold."  If we apply this rate to the 34k number, we get 26k (rounding up) as the actual number of homes that might eventually be delivered and constitute a real sale, or cash generating economic event.   If we annualize this number, we get an annualized sales rate based on January's contract signings + likely cancellations of 312,000.  Note that this varies significantly from the 468k SAAR reported by the Govt.

Even if I give the numbers the benefit of seasonality, there's no way a number which is based on January's contract signings and includes cancellations would come anywhere near 400k.  One more important point of note.  When a contract "sale" as reported by the Govt is cancelled, the Govt does not subtract this from previous "sales" reports.  From the Census Bureau site:  "The Census Bureau does not make adjustments to the new home sales figures to account for cancellations of sales contracts" (LINK).   You'll also note that, as I stated above, the Govt admits that when the market is declining this report and the methodology used overstates the results.  This is what is happening now.

A second source of fraud/incompetence is that the reported increase of sales for January is completely inconsistent with the mortgage purchase application data released weekly by the Mortgage Bankers Association.  Since the early fall of 2013, this report has been showing a decline almost every week.  Since 2014 began, it's been showing double digit year over year declines almost every week.  Now, we know from this data - (LINK) - that mortgages are used in close to 95% of all new home purchases.  January 2013 to January 2014 showed a double digit decline in mortgage purchase applications.  Same for December.  How is it possible that new home sales increased 10% from December to January and 2% from January this year from January 2013?

It has been suggested that perhaps investors started buying new homes to rent out.  While it is possible, that theory is entirely inconsistent with the rate of return model being used by these investors, who require the low cost basis of distressed homes to make their ROR models work.  New homes are significantly more expensive than a distressed home, or even non-distressed existing homes, and therefore it is highly improbable that investors are flocking to buy new homes.

Instead, it would appear that the Government report is seeded in fraud or incompetence.  One last point, we've have had the "bad weather" narrative shoved in our face ad nauseum with every economic report that is showing weakness during January.  However, you'll note that the Government is reporting that the northeast and the south - the two regions which were hit with several bad weather days in January - are both registering increase home "sales" for January over December.

How is it possible that consumers in both the northeast and south decided to stay home in January and not spend money on anything except a new home?  Are they buying these homes from Amazon.com and Ebay?  By the way, online sales tanked hard in January too.  The answer is:  fraudulent or incompetent reporting.