Alan,

>>> Day-Ahead & Hour-Ahead Markets

 Market separation rules aside, we should resist any mandatory Day-Ahead 
Market.  The results for day-ahead clearing markets have been bid gaming to 
influence price in the day-ahead market.  In PJM and NY-ISO parties submit 
false load and generation schedules in set price (most often this is a low 
ball load estimate),then later the true load is scheduled.  I think there is 
less of a problem with the hour-ahead market since the opportunity to submit 
a false bid to set price and then replace it later is very limited if not 
nonexistent.  

>>> Unbalance Schedules

 The general direction we are going for Eastern ISO's is to allow unbalance 
schedules in the real-time market.  Tim indicated a few weeks ago that he 
still wants this feature.  Normally there would be no imbalance penalties, 
however, when reserve margins are so slim that the system's security is in 
jeopardy, penalties would be triggered.  The result is that parties can 
choose to use the real-time market or they can choose bilateral schedules or 
a combination.  They are not compelled to do either.  As long as reserves are 
adequate, then why should we care.

>>> Unit Commitment

 Although not on your list, I think the primary problems that the Day-Ahead 
Market is trying to solve is unit commitment.  While this may be the intended 
purpose, it hasn't worked very well.  Centralized unit commitment would 
address the problem, but then the operator decisions have a large impact on 
prices without having any consequences.  We need a proposal for making sure 
that resources will be on line and that the operator knows they will be in 
place.  A possible approach may be as follows (as suggested by Mike Roan):

The Operator publishes its expected real-time price for the next 48 hours 
based on information supplied by generators regarding their availability and 
bid price (one part) and the ISO's estimate of hourly loads.  The estimates 
roll forward every hour to that a continuous two day outlook is provided to 
all market participants.  The Operator provides the demand/supply curves to 
the market so everyone has the same information.
As units are committed and load forecast is adjusted, the estimate of 
real-time prices changes, providing signals for units to use in deciding to 
commit units.  As expected price rises, more units are committed and price 
goes down.
At a given point in time, such as 4 hours ahead of real-time, the Operator 
locks units in as committed.  If there are inadequate units committed, it 
pays to add capacity (call contracts).  If not enough is voluntarily bid, the 
Operator orders units on under emergency rules contained in 
interconnection/integration agreements with the generators. 
The cost of the call contracts or of units ordered on is included in the cost 
of imbalance energy in real-time for the hours for which those costs were 
incurred.

This procedure is a continuous bid type of approach, which provides signals 
and information to the market.  It or some other procedure like it is needed 
to address the Operators concerns that it will be able to operate the system 
and to deal with the overly conservative tendencies of a not-for-profit ISO.

Steve

 



	Alan Comnes
	03/09/2001 01:10 PM
		 
		 To: james d steffes@enron, smara@enron.com, jeff.dasovich@enron.com, 
rick.shapiro@enron.com, tim.belden@enron.com, mhain@enron.com, 
steve.walton@enron.com
		 cc: 
		 Subject: CAISO Market Stabilization Plant

Here are some initial reactions and questions to the CAISO market 
stabilization plan, which is attached.

Give me or Sue Mara your comments and then we can turn it into message points 
for  media / advocacy folks.

In addition to reviewing the plan, I listened in to part of a stakeholder 
call that occured Friday 10 a.m. PST.

DRAFT MESSAGE POINTS:

With limited exceptions, the plan does nothing to increase supply or decrease 
demand.  It primarily addresses costs and market "stability"

INTERESTING QUOTES:
 
"Over the edge into cost based regulation" Duke Power

Although the plan provides for cost control and for improving market 
"stability" (whatever that means), the ISO staffer (Byron Wortz) who 
presented the study admitted it "Will do nothing to increase supply this 
summer"

A staffer, Lorenzo,  stated (paraphrased)" all out of state suppliers are 
cost of service based and do not need to recieve market based rates

ON SPECIFICS

The Ugly:

Resource-specific cost based bid caps (RSBC or bid caps):  cost of service 
regulation that disincent supply
The ISO will hold in-state resources hostage--cuts any exports when reserves 
fall below a stated critera

Imports can only participate as price takers.  Thus if the highest-cost 
in-state resource is below the opportunity cost of power, needed imports will 
not come in:
 >>this will decrease supply!!
 >> Staffer stated ISO hopes to sign "longer term deals" if necessary to 
secure out-of-state power but acknowledged the proposal needs more work.

The Bad:

The ISO is considering pay as bid, which is complex and does not increase 
supply.  However, on the call they said they are leaing to a single price 
auction.  However, with bid caps the market will still be distorted.

The Good?:

The ISO wll implement an hourly day-ahead and hour-ahead market, essentially 
filling  a void created by the PX's demise.
 >>( Enron has strongly advocated market seperation in the past but do we 
want to consider supporting a day ahead market?)

Market separation rules (MSR).  ISO is leaning towards option 2 in which SCs 
submit bundled schedules but that market seperation would be suspended after 
congestion process is conducted; i.e., final schedules may be imbalanced.
 >>What do we think of this?
.