(c) View from Silicon Valley,
2007. All rights reserved.
The last couple months' statistics seemed a little
off. Some of it we could "prove" via inconsistencies compared to earlier versions or reports from other sources.
Some of it just didn't "feel" right.
Accepting that our staff is naturally suspicious and/or
skeptical (and sometimes even paranoid), we tried to approach this month's statistics with an open mind. Even so, we
found a "bunch" of issues.
First, it was the monthly DataQuick report for median price
of a resale house. Regardless of your opinion about the usefulness of the median, the least everyone would expect
is the same number reported from one day to the next. If the median was $810Kwith
1,171 houses sold "through July 31," when the DQ stats were initially published, then you expect to see those same
numbers in the monthly statistics table. On the other hand, the paper reported $805.5K and 1,191 last Sunday.
Can
we have a turn making up the numbers next month?
Then
they changed the line "MN150" line item to "MN150 market value" and omitted the NASDAQ line item completely. This makes
it impossible to track the relative value of the MN150 over time and you have to find the NASDAQ separately.
The
monthly table already omits "Local CPI" and "US CPI" every other month. Even though the rest of the world
gets these figures every month, we've come to accept the folks running the paper at the so-called "Heart of Silicon Valley"
regularly can't find them or can't get them into their paper.
For
the second consecutive month, the paper did not to publish a figure for "Bankruptcies
Filed." Why? Who knows? Is the City of San Jose not keeping track? Or not making the figures available?
Can we get some kind of explanation?
Most
"interesting" this month, however, is a number of line items published with exactly the same figure as the
previous month. In the same order as in the table:
New
Business Applications (810)
Air
Passengers (983Ku)
OK, actuals were June =982,885 followed by July =982,863 for a difference of 22 (=0.000022, or 0.0022%)
Average
Hotel Occupancy (64%)
This is almost certainly wrong since Total Hotel Rooms sold went from 84.7K to 88.1K (= 3.9%)
Approved
Construction ($241.1M)
Non-residential
Construction ($156.9M)
Residential
Construction ($156.9M)
---Of
course, this leaves the residential /non-residential unched at 65.1%
For
those of you find of numbers, the table's total 21 total lines, fully 11 are suspicious this month (= 52%). If
you count the NASDAQ and two CPI items, then 14 of 24 items are corrupted (=58%).
Is
this part of some grand conspiracy by the local paper and/or the various reporting agencies? At this point, we should
cobble together an intricate conspiracy among the "powers" to deprive the little guy of fair and accurate data. It would
probably generate more page views and ad revenue.
However,
it's almost certainly not a conspiracy. It's probably some semi-random combination of sloppy work, bad luck, poor timing
and incompetence.
If
the rest of us could do such shoddy work and keep our jobs, then that would be a real story.
* * * * The above and any linked article,
website or advertisement are not intended as advice to buy, sell or hold any stock, bond, real estate nor any other financial
product or service. Buy and sell at your own risk (just like we do.)
