The National Association of Realtor's Pending Home Sale Index for November was released Monday morning. While it showed a slight increase using the seasonally adjusted annualized data, it missed expectations by a wide margin. In addition, if you dig beyond the headline numbers, the data in the report shows an underlying downtrend for housing sales that appears to be accelerating.
The Pending Home Sales Index measures housing market activity based on signed sales contracts for existing homes in a given month. For November, the NAR reported an index level reading of 101.7 vs. 101.5 for October. October's number was revised lower from 102.1 to 101.5. Because of the downward revision to October's report, the index gained .2% in November vs. October. The Wall Street consensus was for a 1.5% month-to-month increase. In other words, not only were contract signings lower in October than originally thought, but were weaker than market analysts' expectations.
In using the seasonally adjusted annualized data, a negative trend in contract signings on existing home sales is emerging. The number reported and revised for October shows a 1.2% drop from September's index (102.7 as reported) and 2.2% drop from the October 2012 number (103.8 as reported). While the November report shows a small .2% increase from the lower-revised October number, it dropped 1.6% vs. November 2012. In that I expect a downward revision next month to November's number, it's clear using the adjusted numbers that there's downtrend in contract signings.
It gets even more interesting if you look at the not adjusted numbers, which represent the actual number of reported contract signings for each month. You can find these numbers in the link I provide above. Before I review the numbers, I would like to point out that the seasonally adjusted number, per the NAR website, represents what the annualized number would be if the number calculated with seasonal adjustments for the specific month were to be maintained for 12 months.
The methodology for making the seasonal adjustments is not disclosed; however, for the past several months the housing market data from both the NAR and the Census Bureau (the CB developed the seasonal adjustment model used by the NAR) has been more often than not revised lower in subsequent months. To the extent that the number for the month being reported has been overestimated, the overestimation error is further compounded when the number is annualized.
One big source of overestimating the data for a given month with contract signing data is cancellations. As I've shown in previous Seeking Alpha articles, the cancellation rate for new homebuilders has been increasing over the last several months. It is likely, especially given the increase in mortgage rates during November, that the November pending home sales report will be revised quite a bit lower, as it was for October. This dynamic results in seasonally adjusted annualized reports that can be misleadingly overstated, affecting stock values and analyst expectations.
For this reason, whenever provided, I like to analyze the trend in the unadjusted monthly data, which the NAR provides. From the data in the link above, you can see that contract signings for November were 4.5% below the same number for November 2012. This decrease is significant because it does not contain seasonal "noise." Adding to the significance of the year-over-year drop for November using the unadjusted data is the 18% drop from October to November. While there is no doubt of some seasonal bias in this result, it is significantly higher than the drop in the unadjusted number for October/November 2012 of 13% (106.1 for October, 88.2 for November). As you can see, using the unadjusted actual data, the pending home sale data for November is showing a marked decline not reflected in the seasonally adjusted annualized data.
My point with this analysis of today's pending home sales report from the NAR is that a detailed look at the data reveals a negative trend that is not apparent from the headline news reports. More significantly, as I've presented in several recent housing market articles on Seeking Alpha, my analysis of today's pending home sales numbers is consistent with the downtrend in the housing market that is reflected in the data for existing home sales, new home sales and mortgage purchase applications.
Based on everything I look at, it's my view that the homebuilder stock are overvalued because their prices reflect highly inflated expectations for both the housing market and for homebuilder earnings. Please see some of my previous articles if you are interested in ideas to express a bearish view on the housing market.