It wasn't at all what I expected. With accusations over the weekend by various journalists that the Realtors have been overestimating existing home sales numbers, I was ready for a full-court PR press at this morning's monthly sales report lockup. Not so much.
NAR chief economist Lawrence Yun opened with, "This is going to be a fun press conference!" After detailing the usual monthly report, and carefully acknowledging a nearly 33 percentage point seasonal adjustment in sales numbers (!), Yun addressed the brewing controversy. He explained how they last "benchmarked" sales using Census figures back in 2000, but that they couldn't do that in 2010 because the Census changed some of its survey questions. He admitted that the last revision to data in 2000 showed a 13 percent "drift," as in the NAR's numbers were 13 percent too high.
I won't get into all the reasons for the "benchmark" troubles...okay, it's really 3 reasons: 1) Consolidation of many MLS's (multiple listing service—where Realtors list properties), so a rise in the numbers could be due to the MLS's area getting bigger, not more sales, 2) Properties being listed on multiple MLS's (so sales are double counted), and 3) A big drop in For Sale By Owner properties, so with more sellers using agents, there are more properties on the MLS to sell.
Anyway, Yun said that since the Census had dropped its survey questions, they would have to come up with an entirely new way to count. He said they would call in economists, housing experts, even the media to figure out how to come up with the right numbers. Hopefully, he said, that would be by this Summer. It was slightly disconcerting in that he didn't offer much of a defense of the NAR's recent sales data.
In an interview after the presser, I pressed Yun on whether or not we should believe any of the NAR's numbers. Prices, he said, were solid, no question. As for sales: "Right now I don't know what the level of revision will be, whether it will be in single digits or whether it will be in double digits, but the measurements have been fairly accurate, but it remains to be seen about what is important—to use outside data that is non-Realtor, non MLS for a benchmarking process."
He also added that the revision in the data wouldn't change the directions in the month to month activity, so I guess even if the actual number is wrong, if it's wrong every month, then the change is right? Okay.
Yun admits, "One could make an educated guess that it's most likely going to be a downward revision, based on all the sources," (he didn't say what the "sources" were). By how much, he declined to guess on that one.
Here's my problem: I have talked a lot recently about "noise" in the numbers, in prices, sales, inventory, whatever; the unique nature of this recent housing boom and bust has created literally immeasurable circumstances. Sales of foreclosures and short sales (when the bank allows a homeowner to sell for less than the value of the mortgage) hit 37 percent of all January sales, the highest in a year. Investors made up 23 percent of all buyers, again very high. That's why prices dropped so much, down 3.7 percent annually to a median price of $158,800. That's the lowest median price since 2002 and a record low for January.
If distressed sales are driving the market, then organic buyers are far fewer than they should be. Don't get me wrong, we need to sell these distressed homes, but think about it: What do investors do with the homes? Sure, some will put them up for rent, but many will wait a few months and put them right back out on the Spring market.
Inventories always go up in the Spring anyway, which means we'll be looking at a huge glut, and even more pressure on prices. And don't forget that as banks are now ramping up their foreclosure processes, more repossessed homes will hit the market as well. Numbers? No clue. That's why they call it the "shadow inventory."
All of this is far more important to assess when gauging housing's recovery than the number of homes that supposedly sell each month.