Thursday, September 13, 2007

Sales volumes are down! But by how much?

UPDATE: Reader Cal informs me (see comments) that when DataQuick quotes year-over-year in a press release they re-run the prior year's numbers under the current methodology. However, they do not go back and update press releases that were issued prior to the re-calculation. Thus, this information below is relevant when you are running your own numbers based on DataQuick historical data but not a factor when considering DataQuick's current releases with year-over-year stats.

Everyone is talking about how DataQuick's sale volume numbers are so scary low this August compared to last August. But have we forgotten that DataQuick changed their methodology last Januay? Yup! The numbers under the current DataQuick calculations actually come out higher than they used to. This is what DataQuick says about the new method of counting the number of sales:

To count as an "arm's-length" sale for our sales counts, the logic we've used insisted that there be a seller, a buyer, and that money changed hands. We've now expanded this to include transactions where there was a purchase loan if no price was apparent.

We're also now including multiple sales transactions. If three homes were bought in the same transaction, we now count them as three home sales, not one sale.

These changes increase monthly sales counts by an average of 10 percent. Intra-family transfers are not included, nor are foreclosures until a home is re-sold to a new buyer.

So the number of sales DataQuick now reports each month will be significantly higher than it would have been prior to January, 2007 and thus the decline is even worse than it appears to be at first glance. Check it out here.

8 comments:

Anonymous said...

Nice catch!

Cal said...

The 2006 numbers that they use to compare Year over Year data are calculated using the new method.

The change they made affects all the data quoted in a press release moving forward from February 2007 (the January 2007 release). So they didnt go back and change the 2006 press releases or anything like that, but when they quote the 2006 numbers in a 2007 press release it is using the new methodology. So the sales drop reported is representative of the sales closed in August.

Anonymous said...

Fabulous - now Dataquick is moving the goal posts too.

The NAR did the same thing a few years back with the "affordability" calculation. When it became clear that the affordability was approaching damn near 0%, they changed the methodology to using adjustable mortgage rates with a much smaller down payment.

WTF? This sort of statistics publishing should be shamed and outed as pure manipulation. Oh wait, that's just what you *did* here... Good job!

Kate said...

Cal:

Thanks so much for the additional information! Can you provide a link?

Cal said...

Kate,

I cannot provide a direct link to a phrase of DQ saying exactly what I said. I wondered this exact same thing back in February and Jon Lanser admonished me for not reading the press release correctly. I went back and read both (the February stats release and the methodolgy announcement) and it still wasn't clear.

Jon Lansner is "the man" in my book so give a lot of weight to everything he says. After comparing current releases to past releases just to be sure the numbers are in fact changed, I was satisfied that the numbers reported in 2007 were using the new methodolgy.

Kate said...

Thanks Cal!

DR said...

I'm still confused on the median issue. For example, in DQ's "LA Times Chart" DQ reports that SFH median prices increased 9.1% 8 06 to 8 07. However, downloading this data into excel and sorting on the price change % and finding the median price change at 2218 unit sales ( ~ 1/2 of the 4444 homes sales) reuslts in a median of minus 3%. A swing from +9.1% to -3% is significant. Can anyone shed some light on this?

Kate said...

DR:

Are you using historical numbers to get your year over year? Because Data Quick re-runs their historical numbers before publishing year over year. They don't republish the historical numbers using their new algorithm, but they do run the new algorithm on the old numbers before calculating year over year.