The Seasonally Adjusted Annualized Rate (SAAR) economic numbers are now manipulated beyond the definitional meaning of the word “absurd.” This is especially true with the housing market and auto sales reports. – Investment Research Dynamics
Today the NAR released its “pending home sales” index. On a “seasonally adjusted annualized rate” basis, it showed 1.3% gain over June. June’s original report was revised lower from +.8% to -.2%. Mathematically, this downward revision enabled the National Association of Realtors to report a gain from June to July. Keep in mind this is on a “seasonally adjusted” and “annualized rate” basis.
Now for the real story – at least as real as the reliability of the NAR’s data sampling techniques. In the same report the NAR shows the “not seasonal adjusted” numbers. (click on image to enlarge) On a year over year basis for July, pending home sales were down 2.2%. They were down 13% from June. This is significant for two reasons. Using a year to year comparison for July removes seasonality and it removes the “seasonal adjustments.” Just as important, if you look at historical data for existing home sales by month, “seasonality” between June and July is non-existent – i.e. in some years June sales exceed July and in other years July exceeds June.
The not seasonally adjusted data series is much more reflective of the real trend in the housing market that has developed this summer than is the manipulated SAAR number vomited by the NAR’s data manipulators. The 13% from June to July should shock the hell out of housing market perma-bulls.
FURTHERMORE, the not seasonally adjusted numbers are consistent with the highly correlated mortgage purchase applications data. “Pending” sales are based contracts signed. Concomitantly with signing a contract – the NAR reported that 80% of all existing home buyers in July used a mortgage – the buyer needs to file a purchase application. But the Mortgage Bankers Association reported that mortgage purchase applications hit a 6-month low in July. The mortgage applications data contradicts the NAR’s pending home sales report on a SAAR basis but is entirely consistent with the pattern in the not seasonally adjusted data.
The not seasonally adjusted data are pointing to a rapidly developing housing market implosion – 13% drop in contracts signed from June to July in a two-month period that has little if any seasonality and with 30-yr fixed mortgage rates hitting all-time lows.
Just like the big bubble which finally exploded in 2007-2008, I was early in my call on Housing Bubble 2.0 (HB 2.0). Because it takes a lot of capital and “inertia” to move the housing market, directional movements take time to develop and they become fast-moving trains with no brakes – until they either hit a wall or hit the ground. But change in direction happens suddenly.
When prices are moving up, the market becomes very illiquid on the “offered’ side and buyers become ravenous. This occurred because the Fed dedicated $2 trillion of it’s QE to the mortgage market and the Government made Government-guaranteed mortgages much easier for buyers by taking the down payment requirement down to 3% and in some cases 0%. But when the market rolls over, supply quickly builds and demand disappears and the market becomes very illiquid on the “bid” side. The market is about to become very illiquid on the buyer side of the equation.
I made this call in my latest Short Seller’s Journal this past week:
The housing market is heading south now as well. It’s been my view, and I’ve supported this view with detailed analysis of new and existing home sales on my blog, that both the Government (new home sales report) and the National Association of Realtors (existing home sales report) are using their mysteriously calculated “seasonal adjustments” to inflate the true level of homes being sold on a monthly basis. MOREOVER, and this point is crucial to understand, to the extent that there are flaws in the “seasonal adjustments,” the “annualized rate” calculation compounds these flaws by a factor of 12.
As an example, last week’s new home sales report, which showed an unexpected and absurd 72,000 (SAAR) new homes sold in July vs expectations and 154,000 more homes sold vs. July 2015. However, the report also shows the “not seasonally adjusted, not annualized number for July, which never makes its way into the media reports. In that section it shows only 16,000 more homes vs the 154k SAAR headline sold year over for July AND a decline in sales from June to July of 6,000 homes. In other words, the sensationalized headline reports were manufactured out of thin air from the “seasonal adjustments” applied to the monthly numbers and then converted into an annualized rate
As you can see, the Government’s new home sales report is utterly unbelievable. In fact, the Mortgage Bankers Association has reported that mortgage applications to purchase homes hit a 6-month low in July. New home sales are based on contracts signed. With 93% of all new home buyers using a mortgage, if mortgage applications are not being filed, contracts are not being signed. It’s really that simple.
The NAR’s existing home sales report was well below consensus expectations and showed a 3.2% drop in existing home sales from June and a 1.6% drop from July 2015. These numbers are based on closings. Again, if mortgage purchase applications dropped in June and July, we can expect (or at least should expect )that existing home sales reports for at least the next two months will show further declines. Furthermore, the NAR uses the same statistical “adjustment” model as the Government. To the extent that the NAR’s SAAR numbers showed a decline, the true decline is likely much greater.
After the employment, GDP and inflation reports, the home sales reports from both the Government and the National Association of Realtors are among the most highly manipulated economic data reports. The data is heavily modeled and massaged via the “seasonal adjustments.”
The truth from the ground, based on the extensive footwork due diligence I conduct plus emails from readers around the country reporting similar observations, is that the inventory of home listings of soaring (the published inventory reports by design have 2-3 month lag), prices are dropping quickly, the time it takes to sell a home is increasing significantly and, most important, the potential pool of middle class home buyers no longer have an income level that will support the size of mortgage it takes to “buy” a home.
Short-sell ideas are starting to work again. The short-sell selections in my Short Seller’s Journal have now worked four weeks in a row. My pick from 3 weeks ago is down 6%. At one point it was down 10%. The pick from two weeks ago gave subscribers a quick 13% drop after it reported earnings and it’s still down 10%. My pick from last week is down nearly $2 (2.3%) after 2 1/2 days of trading but the puts are up 42%. I am also making several homebuilder short recommendations now each week.
You can subscribe to the Short Seller’s Journal by using this link: SSJ subscription.