by Calculated Risk on 8/17/2011 03:35:00 PM

DataQuick has developed a new resource for tracking home sales that is updated weekly. It can be accessed by going to the DataQuick news site and clicking on the National Home Sales at the top (or directly here: National Home Sales Snapshot).

DataQuick provides sales and median prices using the most current 30/31 days based on 98 of the Top 100 US MSAs (excluding Louisville and Wichita). DataQuick estimates this is almost two-thirds of all US home sales. These are real sales counts (as opposed to the NAR approach that will probably be changed later this year). DataQuick also provides median prices.

Right now DataQuick is combining existing and new home sales, and they were kind enough to break out the data for me.

This resource will allow us to track sales weekly, although we have to remember this data is Not Seasonally Adjusted (NSA), and both the NAR and Census Bureau headline numbers are reported on a Seasonally Adjusted Annual Rate (SAAR) basis.

I decided to compare the DataQuick numbers to the NAR and Census Bureau reports. Both the NAR and Census Bureau report monthly, and the DataQuick data is weekly.

Click on graph for larger image in graph gallery.

This graph shows the NSA data for DataQuick and the NAR. The dashed line is scaling up the DataQuick numbers by .6625 (their estimate of sales coverage).

The NAR recognizes that their estimate are too high - and they are planning revising down their estimates this fall. However this suggests that the NAR might be overestimating by even more than the 10% to 15% that many analysts think. This also suggests sales in 2011 are very weak.

The second graph compares the DataQuick new home sales numbers with the Census Bureau. The Census Bureau reports when contracts are signed and DataQuick reports when the purchase is closed. So there are some timing issues.

The dashed line is using the same scaling factor as for existing homes. The new home data from the Census Bureau appears to be fairly close to the DataQuick numbers - although it is hard to tell because of the large spikes due to the homebuyer tax credit and because of the lag between contract signings and closings.

This appears to be very useful data for tracking home sales.