San Diego Housing Bubble News and Analysis

The mix has changed. Instead of heavy buying in East and South County, the buying is much slower and a larger percentage of it is in North County Coastal and Inland.

Thus the median goes up when measured area wide. The NAR also provided a solution in their statement the other day where they admitted that the condo and entry level market is virtually no existant.

Hence, if before, the market was one entry, one middle-class and one upper middle class and now it's just one middle-class and one upper middle class, the medium moves up, even if each home get's less.

I don't know the answer but my thoughts are that the sales volume for lower/middle housing levels is down, while the volume of sales for mid/upper housing levels is up. This will end up raising the median for the county overall.

You know my feelings on macro level stats and announcements. They are useless. Focus on the area you want to buy or sell in. Even within a zip code you can have that happen. I recall a post thread awhile ago where it was pointed out that Poway was holding up well or something like that. When I looked at all of the listings that sold in Poway for that particular month of the year, and compared them to the same period a year before, I found the same phenomenah, that the median was affected by a few high end sales.

Just my thoughts...

I really really really think sales volume is much more important then median pricing.

Realtor

OK. The 1.7% is only for resale houses (detached).

The title of the article is misleading since it says "Resale homes up 1.7%". In the text this is changed to "houses" as stated in the data from DataQuick. The U-T has done this type of pro-real state reporting for years even if it means that the information is not entirely accurate or clear.

no_such_reality & SD realtor,
Yes, this is a good theory. But let's assume for a moment that it is true that a larger percentage of sales have shifted to North County Coastal and NC Inland. The micro-medians in those areas for SFR Detached was down -1.7% and
-4.0% respectively. Shouldn't that contribute to a negative pull on the overall San Diego median also? Is DataQuick pulling the wool over our eyes?
New_Renter

FWIW, I have often been baffled by Dataquick's analyses and commentary. They must be using different data than the rest of us can get because they always seem to run counter to everything else I see.

For instance, Dataquick commented not long ago that during the last recession the average drop was 15%, but I could only identify just a couple of areas that "only" dropped by 15% on a house-to-house basis of comparison. The averages I saw on a regular basis were more like 20 - 25%, and I saw a couple market segments drop by nearly 40% off peak.

When Dataquick's John Karevoll finally starts talking about big losses (if he ever does) THEN you'll know that we've hit bottom and it's time to buy.

Here's how:

Imagine there are two neighborhoods, POOR and RICH. The Median in Poor in 2006 is 100k. The Median in Rich is 200k. In 2006, there are five sales, 4 in Poor, and 1 in Rich:

100k Poor
100k Poor
100k Poor
100k Poor

200k Rich

So the median is 100k.

In 2007, prices decline in both markets by 20%. Poor people stop buying, rich retirees jump in:

80k Poor
160k Rich
160k Rich
160k Rich
160k Rich

So the new median is 160k, up 60% from the previous year, although prices overall declined 20%.

When the volume of sales declines in cheap areas and remains steady in expensive areas, the mix of homes sold shifts toward the high end. Consequently, the overall median can go up, even while price in all areas declines.

trex,
Thanks for this great explanation, makes perfect sense. I better take a statistics refresher course! It also vividly points out the inherent flaws in using Median.

It seems the Case-Shiller index for SD is about the best we have at this point in time to get a more accurate read on the true underlying price action. With Median per Sq. Ft. being 2nd best (as Rich has so well pointed out). Unfortunately Case-Shiller seems to be about 2 months behind the others. BTW, does anyone know where you can look at charts of the various Case-Shiller indexes online?
New_Renter

thanks for showing the math trex.

Also guys remember... data can be shaped in SO many ways. Was the dataquick information only detached resale or did it include new housing?

Again, I know it is hard to turn a blind eye to the media but try not to let it get to ya. The overall problem is that by far the less informed majority of consumers do not realize this. So what can you do? Keep a cool head. The bigger question is, what is reality then? How do we gauge things. Everyone has different answers. My personal answer is the sales volume, active/pending ratios, cancelled/expireds verses solds... that sort of thing. Keep it simple, keep in confined to the zip code you are in, etc...

SD Realtor

How it possible that the overall median price for resale homes is up 1.7% ($565K vs. $555.5K) when every single constituent area is down significantly:

It's not possible. The problem is the UT's table is screwed up. Note the zone-specific 2006 vs. 2007 medians and then the listed percentage. They're completely messed up.

Central San Diego resale SFR's are listed as -5.8% yoy. It should be +5.8%.

East county is listed as a 9.6% drop when it's actually a 0.96% drop.

And so on...

Whether online or in print, you'd think somebody might review this before it gets published.

A closer look shows a sloppy cut and paste job. The percentages given for resale condos are actually the numbers for resale SFR's.

Amateur hour.

"How it possible that the overall median price for resale homes is up 1.7% ($565K vs. $555.5K) when every single constituent area is down significantly":

I don't know if this is why, but I read recently that the median went up because less people are able to buy starter homes. I thought it was strange too, especially since house prices(asking prices) in my area are falling. Before January '07, there was nothing on the market below 400k, now there are 15 and climbing. Two are actually asking 299k for sfr..........ttfn

This is yet ANOTHER REASON why we need to have a single, objective, and open-to-the-public real estate database. Our current system is like trying to construct the Tower of Babel.

The internet revolution can't get to the RE fast enough. I hope, one day, all of these info will be public and you can easily do a search using your favorite search engine to get all the data you want and construct your own analysis using a spreadsheet app.

There's really no statistic that will be able to tell you where the bottom is. You won't even know where the bottom is until you past it. People who like to call bottom will call bottom all the way down. Just like people who like to call top called top all the way up. I'd say, don't worry about the bottom and just buy a house when it make sense relative to income and rent. The rest will fall into place.

Oh Jeez, great catch Jim! What a joke. This data is pretty important right now, so you'd think at least DataQuick would have caught the mistake, it's their numbers after all...

Alex's example still shows us how it could happen that the overall median could rise, while the constituents could fall, but when you think about it, for that to happen there would have to be a MASSIVE shift in the mix. This tells me that the shift in the mix, while no doubt is happening, is more subtle (i.e. slower moving). The fact that North County Coastal median was up 4.5% and volume constant, while South County median was down -4.7% and volume WAY down (174 vs. 262) is really where the change in mix is showing up in the data.