fyi.  also, I just spoke with Sarah Wesner (ext. 35261) who will coordinate 
with you.  You can review the note with her and let's talk about their 
guaranty (will our guaranty format work?).  Also, I'm almost inclined to go 
with their commitment letter (if revised) in lieu of a lengthier agreement.  
SS
----- Forwarded by Sara Shackleton/HOU/ECT on 06/22/2000 08:46 AM -----

	Sarah Wesner@ENRON
	06/20/2000 07:24 PM
		 
		 To: Joseph Deffner/HOU/ECT@ECT
		 cc: Sara Shackleton/HOU/ECT@ECT
		 Subject: Differences

Joe - attached is a first shot at getting a handle on all the differences 
between the proposed and existing margin lines.  I plan to do more analysis 
with respect to covenants and warranties but this hits the main points.  The 
analysis between rates will be very complex and may be something best left to 
the research Group.  We can go through why this is such a complex issue to 
model.  What I do plan to do is update historical info. so that based on the 
past we can make general statements such as over time Fed Funds + X bp <,>,= 
to LIBOR + X bp.

A quick synapses of difficult items:

Commissions vs. Margin lines - will trading desk use facility for many trades 
or less frequent trades for smaller amounts
Short on Financial gas vs. Long on physical gas - cash flow differentials and 
how we can model the cost of cash to fund the working capital (as you 
explained to me today)
What positions the trader take and how our equity position is at that exact 
time.  Will we trade in a similar fashion in the future?


From what I understand the trading desks do not have ready access to the 
above  information in a format which lends itself to modelling of reasonable 
complexity or do not wish to share it with us (understandable considering the 
sensitive nature of the business.)  The ideal situation would be for the 
desks to decide themselves whether a line makes sense for them based on the 
pricing provided.  As mentioned above, I believe that we should be able to 
explain why a certain base rate would be attractive but am not sure how to 
estimate the value of a line to the traders without detailed information.  
Perhaps the best division of labour would be for the desks to decide what 
loan would be most valuable based on the desk's trading patterns while we 
could focus on getting the line in place as quickly as possible.  If you 
disagree, please let me know and I will try to rethink how to model this 
appropriately.  

Sarah