Message-ID: <30691873.1075843952565.JavaMail.evans@thyme>
Date: Thu, 31 May 2001 15:59:00 -0700 (PDT)
From: per.sekse@enron.com
To: jeffrey.shankman@enron.com, mike.mcconnell@enron.com
Subject: Unit Contingent Outage Options - Hedge for Notional P100 Risk
Cc: vince.kaminski@enron.com, vasant.shanbhogue@enron.com, john.best@enron.com, 
	david.hoog@enron.com, greg.whalley@enron.com, per.sekse@enron.com, 
	chip.schneider@enron.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Bcc: vince.kaminski@enron.com, vasant.shanbhogue@enron.com, john.best@enron.com, 
	david.hoog@enron.com, greg.whalley@enron.com, per.sekse@enron.com, 
	chip.schneider@enron.com
X-From: Per Sekse
X-To: Jeffrey A Shankman, Mike McConnell
X-cc: Vince J Kaminski, Vasant Shanbhogue, Amitava Dhar/Enron@EnronXGate, John Best, David Hoog, Greg Whalley, Per Sekse, Chip Schneider
X-bcc: 
X-Folder: \Mark_McConnell_June2001\Notes Folders\Notes inbox
X-Origin: MCCONNELL-M
X-FileName: mmcconn.nsf

Working with Amitava Dhar,  Vasant Shanbhogue and John Best,  we completed 
the following analysis to determine max loss for our portfolio of 
transactions for the balance of 2001.

A Model simulation yields max loss of less than $60 million for our total 
portfolio in 2001.

 Comments:  We ran several simulations in the model with 10,000 trials 
showing a max loss without hedges of $49,500,000 - $52,000,000. In 
otherwords, the  highest possible loss experienced in all runs was less than 
$52 million.  We also ran one simulation at 30,000 trials and the max loss 
did not vary   significantly from the earlier trials. Our conclusion from the 
results of the model runs is that the 1 in 100,000 trial would produce a max 
loss (P99.999) of less  than $60 million vs the notional P100 payout of $280 
million.

 This does not address the "Venus falling into the Sun" scenario. Since 
running the simulation does not produce a max loss approaching the P100, we 
had  to go outside the model and create a spread sheet that would force a 
scenario that generated the full loss.

B.  Worst case scenario of all units down on day 1 hits maximum payout of 
$280 million over 5 days.

 Comments: In the spread sheet we assumed an unrealistic scenario that all 
units go down on day one for 5 days. Taking the total MWh covered for each  
unit, we then calculated the power price necessary to hit the maximum dollar 
payout for each unit given a 5 day outage. The 5 day price spike for each  
power market ranged from $1,500 in Entergy to $700 in PJM. Under this 
unrealistic scenario we would reach our maximum payout of $280 million in 5 
days.

C. Power option hedges generate offsetting cash value of $ 215 million, 
reducing our net P100 exposure to less than $65 million

 Comments: We then layered our proposed hedging strategy on top of the 
unrealistic spread sheet scenario in 2 above. Assuming the 5 day price spike  
replicated the forward curve in the MidWest during 1999, a spot market at 
$1,500 pulled forward prices up to a range of $250-$300/MWh. The option   
hedges we would put in place would cost $14-15 million and would yield $215mm 
in positive MTM to offset the payout  of $280 million, giving us the net  
Maximun risk of $65 million.

We need to present our calculations to Vince tomorrow morning to get his sign 
off and can then go over the analysis in detail with you and Greg immediately 
afterwards. We are still not recommending we put the full hedge in place to 
cover this unrealistic scenareo, but the analysis shows we can significantly 
cover the tail risk in our book without having insurance in place. Our 
recommendation remains to put a smaller hedge in place while we continue to 
pursue a insurance partner during the month of June.

Speak to you tomorrow. 
Per