View from Silicon Valley
Something seems fishy...
Home
Santa Clara Co. median (updated Sep 6th)
San Mateo Co. median
Santa Cruz Co. median
Santa Clara Co. stats (updated Aug23)
SEMI B:B to Jul'08 (updated Aug30)
SIA Data '04 -Jun'08 (updated Aug15)
Wafer area vs.SIA$ 2Q08(updated Aug30)
VC Funding -4Q07 (updated Apr27)
SV Stats (Updated!)
Links
About Us
 
November 1, 2006
 
Something seems fishy...
 
(c) copyright View from Silicon Valley, 2006.  All rights reserved.
 
 
 
Tracking weekly housing statistics published in our local paper, you get used to being surprised on Saturday morning.  After several weeks of declines in the summer, Santa Clara County sales prices stabilized over the last eight or ten weeks.
 
With weekly sales prices flat, an objective observer would expect to y-o-y price increases to steadily decline.  However, over the last couple weeks, Santa Clara County y-o-y gain figures improved slightly even as prices stayed flat.
 
This small divergence catches your eye when you know (from reading this web site, at least ;=) that Santa Clara County has steadily reported positive y-o-y prices, while San Mateo and Santa Cruz Counties y-o-y's have slipped more and more into the red over the last three or four months.
 
We decided to physically check the y-o-y numbers.  The results were surprising.
 
First, we look up the equivalent week from 2005.  Fortunately, we keep track of these on the web site and easily pulled it up as follows:
 
 
October 16, 2005:
Median Price        Sep 29'05 from'04 $/SqFt Volume from'04
All homes           $663,000   20.5%   $453   2,452  -9.8%
Total resale houses $705,000   19.5%   $477   1,614  -9.5%
Total condominiums  $485,000   24.4%   $421     585 -18.9%
Total new homes     $679,000    7.8%            253  +18.2%
 
Then the current week is:
October 28, 2006:
Median Price        Oct04'06  from'05 $/SqFt Volume from'05
All homes           $670,000    2.3%   $463   1,875 
-26.2%
Total resale houses $738,500    5.5%   $498   1,175 
-29.1%
Total condominiums  $483,750    0.3%   $446     447 
-20.7%
Total new homes     $584,750    4.6%            253 -21.2%
 
 
Ignoring the 5-day difference, the 2006 table should read:
October 28, 2006:
Median Price        Oct04'06  from'05 $/SqFt Volume from'05
All homes           $670,000    1.5%   $463   1,875 
-23.5%

Total resale houses $738,500    4.8%   $498   1,175 
-27.2%
Total condominiums  $483,750   -0.3%  $446     447  -23.6%
Total new homes     $584,750  -13.9%           253   0.0%
 
In other words:
 
+"All homes" y-o-y price is only +1.5% vs. +2.3% with volume slightly less negative.

+"Total resales homes" y
-o-y price is only +4.75% vs. +5.5% with volume slightly less negative.
 
+"Total condominiums" y-o-y price actually flips to negative -0.3% and y-o-y volume is worse.

+"Total new homes" y-o-y price flips from +4.6% to -13.9%!  Did somebody "miss" a -$94,250 price decrease? (Assuming DataQuick is not actively manipulating the results...)
 
Also note, "New homes" y-o-y volume is reported at the same 253 in 2006 as in 2005.  While we can observe this week's 253 is the highest figure since July 20, 2006 we cannot conclusively say the figure is, or is not, accurate.
 
As an aside, each weekly 2005 new home report after September 29, was higher than 253.  In other words, "new homes" sales volume steadily grew in the last three months of 2005.  If we get something similar in 2006, it may serve to help hold up total sales volume but will probably also depress actual and y-o-y prices.
 
Running the same comparison for San Mateo County:
October 29, 2005:
Median Price        Oct 06'05 from'04 $/SqFt Volume from'04
All homes           $758,500   14.9%   $538     749 -12.2%
Total resale houses $805,000   15.8%   $570     556 -14.1%
Total condominiums  $538,000   22.3%   $496     178   +4.1%
Total new homes     $954,500   11.8%             15 -57.1%
 
October 28, 2006:
Median Price        Oct06'06  from'05 $/SqFt Volume from'05
All homes           $760,000   0.0%  $568     675  -14.7
%
Total resale houses $790,000  -1.9%  $591     527 
-11.3%
Total condominiums  $539,000   1.9%   $503     139  -23.6
%
Total new homes     $612,000 -35.4%            9   -40.0%
 
 
The corrected 2006 figures should be:
Median Price        Oct06'06  from'05 $/SqFt Volume from'05
All homes           $760,000  -0.2%  $568     675   -9.9
%
Total resale houses $790,000  -1.9%  $591     527  
-5.2%
Total condominiums  $539,000   0.2%   $503     139  -21.9
%
Total new homes     $612,000 -35.9%            9   -40.0%
 
Again, y-o-y prices are worse than reported, with "all homes" flipping to a negative number, while y-o-y volume is slightly better.
 
Once more for Santa Cruz County:
October 22, 2005:
Median Price        Oct 06'05 from'04 $/SqFt Volume from'04
All homes           $695,000   13.9%   $540     363   +0.3%
Total resale houses $742,000   17.4%   $537     251 -10.0%
Total condominiums  $544,500   21.0%   $588      63  -3.1%
Total new homes     $636,500   44.7%             49 +172.2%
 
October 28, 2006:
Median Price        Oct06'06  from'05 $/SqFt Volume from'05
All homes           $681,000   -2.3%   $466*   226 -36.5%
Total resale houses $727,500   -2.6%   $466    163 -33.2%
Total condominiums  $518,000   -5.2%   $466     39 -41.8%
Total new homes     $637,000    2.4%             24 -46.7%
 
 
The corrected 2006 figures should be:
October 28, 2006:
Median Price        Oct06'06  from'05 $/SqFt Volume from'05
All homes           $681,000   -2.0%   $466*   226 -37.7%
Total resale houses $727,500   -2.0%   $466    163 -35.1%
Total condominiums  $518,000   -4.9%   $466     39 -38.1%
Total new homes     $637,000    0.1%             24 -51.0%
 
Santa Cruz County shows the reverse of the other two.  Three of the four y-o-y prices are slightly better while three of four y-o-y volume comparisons are worse.  However, the better prices in Santa Cruz were better by a much smaller margin than in the other two counties...
 
Conclusion:
Is this a smoking gun?  Does it prove, once and for all, that real estate industry-generated stats are manipulated?  Not quite.
 
It is, however, another indication the statistics err on the side of higher prices.  
 
Even assuming the implied bias is just a coincidence, it bears continued scrutiny.
 
 
 * * * * *
The above is not intended as advice to buy, sell or hold any stock, bond, real estate nor any other financial product or service.  Invest at your own risk.