July 7, 2007
Please stop reporting median sales
(c)
copyright, View From Silicon Valley, 2007. All rights reserved.
You may recall last weeks' missive started with:
"Santa Clara County's median at this time last year was $765K compared to 2007's $795.25K.
While the headlines trumpet "new all-time high," we've only had a gain of +3.95% in 52 weeks. With short-term money
paying more, this rate of change should scare people."
We soon received an interesting reply from a long-time reader:
* * * * * * * * *
"Please stop reporting the median SALES price. It's irrelevant and misleading. When sales
volumes plummets at the low end, ASKING prices can drop across the board and be hidden by a few high-end sales.
Think
of it this way: if you fire everyone making less than the median income, the median will go up even though no one is making
any more money. In fact, the median can go up even if the (remaining) employed people make less.
They key is that no
one is reporting that asking prices have fallen. Now that's something I'd link to.
Thanks for listening.
Patrick"
*
* * * * * * * *
--OK, but how do you really feel?
After some thought, we replied:
* * * * * * * * *
Did you read the whole article?
I agree median has limited usefulness,
particularly with declining volume. The average, which we also track, is similarly limited.
BTW, your example
on salaries has been happening for the last three or four years. Total jobs are still ~150Ku below the peak.
Folks who were below the median saw their jobs filled with discount H-1B's, or off-shored completely, while top execs made
ever more money. This results in a rising median wage, on top of a even stronger rising "average" which, compounded
with rising real estate, obscure the truth of this condition.
I desperately wish "asking prices have fallen" here in
Silicon Valley was true. Our last missive cites a price example of a case for which this is not true. (We
wrote about another one at $1.99M for ~1,500 square feet +cottage sold in ~5x days, a couple weeks ago: http://www.viewfromsiliconvalley.com/id336.html)
It is different in San Mateo (420 -730 sales per month YTD'07) and Santa Cruz County (165 -200 sales
per month), but here in Santa Clara County, we're running 1,500 -2,200 sales per month. It takes more than "a few high
end sales" to significantly skew the median.
As much is we earnestly wish it wasn't true, and as likely as it is that high
prices are to soon drive me out of the valley, asking prices are definitely not falling in our neighborhood. Not only
is unit price up, but the quality of what you can find today at a given price point is worse than it was last year, which
was worse than the year before. (Which was the main point of the last article. We happened to use the current
median price as a gauge of what example to pick for the y-o-y comparison...)
Looking outside our neighborhood,
Santa Clara County shows 29 of 50 zips codes have a y-o-y median price gain. If the median is really off as a measure,
it's off in a lot of zip codes. (We think this is more a measure of the continuing widespread nature of hysteria
rather than a real sign of strength...)
At some point, I think (/hope) median prices will fall off a cliff. If
you can reliably produce a statistic showing unit asking prices are already declining, we would gladly trumpet it to the masses.
The
last chart you produced suggested asking prices were volatile, but flat. If asking prices were really going down while
the median was up, we should have a glut of listings. However, DQ show listings are +28% y-o-y for the last 90 days
and +28% y-o-y. MLS shows +7% listings since Sep'06 --the farthest back we have their data. "Up" is "bad," of course,
but these numbers do not scream "glut."
Unless the data has changed since a few weeks ago, your data falls short of
"proof." I think you can appreciate our unwillingness to trade one stat which doesn't prove the desired point for another.
FWIW,
we think housing prices will (finally) start to fall when we start seeing layoffs here in the valley. Rumors suggest
this time may soon be at hand...
Also, this makes for a good missive. Would you want to be cited as the other
side of the argument? Or anonymous?"
* * * * * * * * *
To which the response was:
"Thanks,
I have to
admit I don't usually read past (the headline of) a report of rising median sales prices, because it usually means uncritical
acceptance of realtor-spew.
Good point about the salaries. The lower levels have been offshored, and I know people
who helped.
I'm pretty sure asking prices have indeed fallen in Silicon Valley. Here is real asking price data from
the MLS:
http://patrick.net/housing/contrib/Bay_Area_Prices.txt
Sure, you can cite me."
Conclusion:
When
you actually read the missive, the message remains:
Median prices may be up but:
+subjective "quality" is down
+cash
flow (which is negative)
+cap rates are maybe 2.5%.
The ONLY basis on which current housing prices make sense is
if you assume prices will keep going up. Even then, we submit 7.5%, per year, is the mininum rate to constitute
a justification on this basis. (i.e., enough to earn back the ~2.5% "lost" on cash flow plus the 5% T-bills
would pay during each successive year. Factoring in pportunity cost adds another percentage point or so to this
figure.)
