by Calculated Risk on 10/16/2015 04:42:00 PM

From housing economist Tom Lawler:

Based on publicly-released realtor/MLS reports from across the country released through today, I project that US existing home sales as estimated by the National Association of Realtors ran at a seasonally adjusted annual rate of about 5.56 million in September, up about 4.7% from August’s preliminary pace and up 9.0% from last September’s seasonally-adjusted pace. I expect the NAR’s estimate of the inventory of existing homes for sale at the end of September to be down about 1.7% from August’s preliminary estimate, and down about 1.3% from last September. Finally, I expect the NAR’s median existing home sales price estimate for September to be up about 5.5% from last September.

Post-Mortem on August’s Existing Home Sales Report: NAR’s Estimate of the YOY % Change in Unadjusted Sales Seems Reasonable

On September 21st the NAR estimated that US existing home sales ran at a seasonally adjusted annual rate of 5.31 million in August – far below both the “consensus” forecast and my projection based on realtor/MLS reports available as of September 15th. As I acknowledged following the NAR EHS report for August, part of my “miss” reflected a “misread” of the likely seasonal factor used to adjust the “raw” sales data. However, my projection was also off because the “sample” of local realtor/MLS reports I had available as of September 15th proved to be a poor representation of the larger sample of local realtor/MLS reports for August that include reports subsequently released. Based on this larger sample, the NAR’s estimate of the YOY % change in existing home sales in August seems broadly consistent with local realtor/MLS reports.

If, in fact, my September projection for existing home sales is correct, then one might ask: why have there been such large month swings in seasonally-adjusted home sales over the past several months? My gut is that some of these swings have been less related to volatile markets, and more related to difficulties in accurately estimating the true “seasonal” component of existing home sales. Statistical estimates of this seasonal pattern of home sales have changed considerably over the past 10-15 years, and it is quite possible that some of the observed change in the “seasonal” pattern may actually be related to other forces (witness, e.g., the huge increase in the amplitude of “seasonal” swings in home prices since the housing collapse, which most analyst attribute to the combination of the surge in distressed sales and the seasonal pattern of the distressed-sales share of total home sales).

CR Note: The NAR is scheduled to release September Existing Home Sales on Thursday, Oct 22nd.  The consensus forecast is for 5.36 million (this will move up after this is posted).  Take the over!