by Calculated Risk on 1/26/2010 11:11:00 AM

Case-Shiller released the November house price index this morning and most news reports focused on the small decrease, not seasonally adjusted (NSA), from October to November. As I noted earlier, the seasonally adjusted (SA) data showed a small price increase from October to November.

House price data has a clear seasonal pattern, and I think the headlines for this data should be the Seasonally Adjusted number.

The following graph shows the month-to-month change of the Case-Shiller Composite 10 index for both the NSA and SA data (annualized). Note that Case-Shiller uses a three-month moving average to smooth the data.

Click on graph for larger image in new window.

The Blue line is the NSA data. There is a clear seasonal pattern for house prices.

The red line is the SA data as provided by Case-Shiller.

The seasonal adjustment appears pretty good in the '90s, however it appears insufficient now. Still the SA data is probably a better indicator than the NSA data - and to be consistent I've kept reporting the SA data.

The second graph compares the Case-Shiller Composite 10 SA index with the Stress Test scenarios from the Treasury (stress test data is estimated from quarterly forecasts).

NOTE: This is using the Seasonally Adjusted (SA) composite 10 series.

The Stress Test scenarios use the Composite 10 index and starts in December. Here are the numbers:

Case-Shiller Composite 10 Index, November: 157.66

Stress Test Baseline Scenario, November: 140.86

Stress Test More Adverse Scenario, November: 128.54

This puts house prices 12% above the baseline scenario and 22% above the more adverse scenario.