| Home Sales Overcounted |
|
Posted Under: Data Watch • Home Sales • Housing |
You've heard about it. But, it's not as scary as they say.
When the National Association of Realtors (NAR) issues its latest report on existing home sales next week (9 am central time, Wednesday, December 21), it's likely to show substantial downward revisions for both the pace of sales and the number of existing homes in inventory, dating back to 2007. Some see this as horrible news about the economy, but the changes are really not important at all.
First, as the chart shows, over the past several years existing home sales have not fallen nearly as much as either new home sales or housing starts. (The chart is indexed so that each indicator's average level in 2005 equals 100.) For example, while existing home sales are now roughly 30% below the 2005 average, new home sales are down about 75% and housing starts are down around 70%. In other words, it's something we have noticed and wondered about. All a downward revision for existing home sales will do is move them closer to what other indicators were already saying about the housing sector. We have been saying for some time that a "normal" pace of existing home sales would be about 5.5 million per year. So far this year, sales have averaged just shy of 5.0 million, which seemed oddly good given that it may take several more years for the housing sector as a whole to return to normal.
Second, according to the NAR, the downward revisions to sales (because of double counting or population estimates) will be matched by a downward revision to inventories. This will leave the months' supply of existing homes – how many months it would take to clear out the current inventory, at the current pace of sales – roughly unchanged. In addition, prices will not be revised.
Third, and perhaps most important, is that existing home sales make almost no difference for economic growth. If you remember Econ 101, GDP only counts new production, not the re-sales of used items. Selling an existing home is just a transfer of an asset from one owner to another, and does not reflect the current production of goods and services. The only part of this transaction that impacts economic data is the commission earned by a realtor.
There are several reasons for next week's benchmark revision, including changes in coverage areas for the multiple-listing services (MLSs), shifts in population across coverage areas, a drop in the share of sales by owner, the seeping of some new home sales into MLSs, and the recording of some sales on more than one MLS. In 2005, another benchmark revision by NAR reduced 1999 sales by 11%. That was not considered important news for the economy and neither should the revision next week.
|
|