[115553] in cryptography@c2.net mail archive
Interesting New Developments in SocGen
daemon@ATHENA.MIT.EDU (Jon Callas)
Thu Feb 21 12:16:49 2008
From: Jon Callas <jon@callas.org>
To: Cryptography <cryptography@metzdowd.com>
Date: Thu, 21 Feb 2008 07:13:10 -0800
<http://news.bbc.co.uk/2/hi/business/7255685.stm>
Excerpt:
An internal investigation into billions of euros of losses at
Societe Generale has found that controls at the French bank
"lacked depth".
The results of the investigation also show that rogue trades
were first made back in 2005.
<http://news.bbc.co.uk/2/hi/business/7256102.stm>
Excerpt:
Societe Generale made a profit in 2007 despite a trading scandal =20
that
cost the bank 4.9bn euros ($7bn; =A33.7bn).
The French bank said it made a net profit of 947m euros for the =20
year,
although this was down 82% from 2006.
I think these two things are very interesting from a viewpoint of =20
security and economics. This fellow had been making unauthorized =20
trades for two to three years, and when it all came tumbling down, it =20=
knocked off at most 82% of one year's profits. (I say at most because =20=
it's reasonable to think that in a year of subprime issues, they'd =20
have been down 30-50%.)
Compare and contrast with Nick Leeson's sinking of Baring's, which was =20=
a mere $1.4bn. We can even double-ish that to say $3bn to account for =20=
the intervening time.
Both cases were unauthorized trades spinning out of control in =20
attempts to cover small losses, but Baring's was sunk for the =20
(adjusted) $3bn and SocGen merely loses 80% of one year's profits at =20
$5bn.
Does this suggest that what is really needed is a way to detect losses =20=
that could spin out of control before they do, as opposed to direct =20
security mechanisms?
Jon
---------------------------------------------------------------------
The Cryptography Mailing List
Unsubscribe by sending "unsubscribe cryptography" to majordomo@metzdowd.com
