Frank:
One of the most likely scenarios for a VAR blowout would be a severe cold 
front hitting the country in the middle to latter part of the winter.  In 
such a circumstance, cash may separate from prompt futures similar to how 
Midwest power traded $5000+ on specific days last year while prompt futures 
were $200.  The correlation between prompt and cash is normally very strong, 
and is indicated by the small VAR associated with a spread position 
currently.  But in the winter that may change.
Another thing to keep in mind while developing this scenario is the 
assymetric risk presented by having a spread position on.  Assuming we enter 
the winter with normal to below normal storage levels, a position of long 
cash, short prompt futures has a long tail only on the positive p&l side.   
While such a trade in an efficient market has expected payout of 0, the 
payout probabilities may look like the following:

20%  $ -.05
40%     $ -.02
20% $ 0
19% $ .03
1% $ 1
   


   
	Enron North America Corp.
	
	From:  Frank Hayden @ ENRON                           07/20/2000 02:12 PM
	

To: Fletcher J Sturm/HOU/ECT@ECT, John Arnold/HOU/ECT@ECT
cc:  
Subject: Stress Testing

RAC is working on developing some "canned" stress tests regarding VaR.  For 
example, one test could be called "hurricane", were the prompt month is 
"stressed" on both price and vols, holding all other inputs constant. 

Anyway, I would like to know of any likely/realistic stress scenarios you can 
think of....

Let me know,
Frank