EXHIBIT 99.290

Introduction:

This document expands upon the PX's initial response to the ISO's Congestion
Management Reform Recommendation. Whereas previous comments were limited to
concerns with the ISO's proposal, the following proposal is intended to provide
solutions to those concerns.

This document describes revisions to specific portions of the ISO's
recommendations. It is not intended to replace their entire proposal. Instead,
it describes how certain processes need to be modified to realize the ISO's
goals.

The complications involved in introducing changes to a complex, interdependent
model are understood. However, the reforms detailed below are specifically
designed to interact seamlessly with the aspects of the ISO's proposal that do
not require modifications.

As can be inferred from the previous statements, the PX believes that much of
the Comprehensive Market Reform design should be retained. The decisions to
build upon the existing California market structure and to uphold the original
fundamental design principles have significant value, as does the proposed use
of a consistent network model throughout the Real-Time and forward markets. The
introduction of a single commercial network model should eliminate most of the
inconsistencies that currently exist between the forward and spot markets. The
introduction of more zones/LPAs along with enforcement of nomograms in the
forward markets should facilitate the reduction of intrazonal congestion in the
forward markets thereby reducing Real-Time intra-zonal congestion. However, the
ISO should release the details of this model as soon as possible and give
stakeholders the opportunity to provide feedback.

Design consistency should extend beyond network model. When finalizing the
details of their recommendation, the ISO should make certain that policies are
uniform across markets and do not discriminate against any single class of
stakeholder.

Forward congestion will be better managed with the introduction of nomograms in
CONG (the ISO's congestion management software). The creation of new zones/LPAs
and enforcing nomograms will allow the market to eliminate much of the current
intra-zonal congestion in the forward markets. Validating preferred schedules
against the nomogram constraints should reduce the amount of Real-Time
congestion.

Utilizing recallable transmission will potentially further enhance congestion
management efficiency. We support the concept of a product that allows the
market to access the unused transmission capacity reserved by ETCs. However,
more details of the proposed market are required to fully judge its impact.

Concept details are essential if the market is to accurately evaluate the ISO's
proposal. While an idea may seem desirable in concept, if it is not properly
implemented much of the potential value may be lost. When defining market
mechanics it is important that the




ISO pursue the simplest workable solution. Unnecessary levels of complexity are
significant barriers to market participation and thus, market liquidity. The ISO
should also seek to avoid market proliferation. Increasing the number of markets
into which the same capacity can be sold fragments the markets and may adversely
affect liquidity.Those aspects of the ISO's proposal that have been identified
as overly complex are simplified in the enclosed recommendations.


Comments on the Long Forward Market:

The Long Forward market design doesn't consistently conform to the ISO's
self-stated goals of enhancing market efficiency through decentralized decision
making. Instead, the market structure includes processes that abandon market
solutions - even in situations when the marketplace is best suited to
efficiently resolve reliability requirements.

One such process is Local Reliability Service (LRS) procurement. The ISO's
proposal relies on a two-day-ahead "auction" to procure generation within each
Local Reliability Area (LRA). Unlike other markets within California, loads
within LRAs do not submit bids for this generation. Instead, the ISO procures
capacity on their behalf based in its forecasts of loads and requires that
portions of the acquired generating capacity be scheduled in Day-Ahead based on
nomogram constraints. Capacity purchased by the ISO in the 2-Day-Ahead market,
but not identified as "minimum reliability energy" (and thus required to be
scheduled in Day-Ahead) may be dispatched by the ISO as needed in Real-Time.

Two distinct and separate products are being acquired by the ISO in this
process. Capacity required to be scheduled as minimum reliability energy in the
forward market is not capacity. It is instead, a firm energy purchase and should
be treated as such. Committed capacity that is reserved by the ISO and is able
to be dispatched through Real-Time would be better obtained through existing
processes such as the Ancillary Services markets

Because the decision to schedule Minimum Reliability Energy (MRE) is a forward
energy allocation decision, the ISO needs to let SCs complete this task if it is
to remain consistent with the goals of maintaining the decentralized decision
making process. Thus, SCs will be responsible for scheduling load and generation
within LRAs according to current Day-Ahead and Day-Of timelines, removing the
need for a 2-Day-Ahead auction.

As all nomogram constraints will be placed in CONG, it will validate whether
enough generation within each LRA has been scheduled to meet the daily load
requirements. If insufficient generation has been scheduled CONG will either
increment the necessary generators based on their adjustment bids, or it will
decrement those loads that exceed the nomogram limitations.




CONG will ensure that scheduled generation and load satisfy each nomogarm on a
market-wide basis rather than by SC. CONG would determine "usage charges" for
flows across nomogram constraints. An SC would pay the usage charge for bringing
power across the nomogram. Conversely, an SC whose schedule allows other SCs to
import more across the nomogram, would be paid the usage charge for its
counterflow. This is necessary to conform to nomogram requirements efficiently.

Qualified market operators could facilitate the LRS procurement process by
running forward markets for LRS capacity and energy to help participants develop
schedules that meet nomogram requirements and avoid the associated usage
charges.. Buyers and sellers would be able to trade LRS capacity and energy
using many different contract types and would self-provide the resulting
positions in the Day-Ahead market.

Market efficiency would be further enhanced if nomogram constraints were traded
like transmission rights. If this were the case, loads could purchase the rights
to schedule portions of their loads entirely against energy imported from other
Locational Price Areas (LPAs) and avoid the cost of expensive in-area
generation. Nomogram rights, like FTRs would be both physical and financial.

If this structure were introduced, loads within LRAs would have sufficient
incentives to acquire nomogram rights and to schedule in the forward markets in
order to hedge against ISO charges. The ISO would also remain consistent with
its stated goals.

If after running Day-Ahead and/or Day-Of congestion management, the ISO felt
that an insufficient amount of in-area generation had been scheduled (or if the
ISO expected additional load within a LRA to appear in Real-Time), it would
acquire what it refers to as committed capacity. This process would be
facilitated by the introduction of committed capacity (CC) as an additional
Ancillary Service.

The CC ancillary service will be treated in the same fashion as existing AS
types with the exception that CC will be purchased on a zonal/LPA basis rather
than a market-wide basis. SCs will be allowed to self-provide the service
according to the same rules that govern self-provision today. Additionally, SCs
who have available capacity will be able to place bids to supply CC to the ISO,
who would procure it along with its other Ancillary Services. As with minimum
reliability energy, SCs will have the incentive to self-provide this service in
order to hedge against ISO charges.

Because some level of market power will exist within LRAs until new generation
or transmission is built, market power issues must be addressed. Market power
issues are the same whether the ISO's or the PX's proposal is adopted: a price
must be determined that protects loads from undue market power, provides
sufficient revenues to generators and provides locational price signals in
accordance with FERC requirements. The PX




feels that the ISO and its stakeholders are capable of developing a workable
solution to this issue and thus does offer a specific proposal at this time.

While this approach differs significantly from the ISO's, it contains enough
benefits to warrant inclusion in their proposal. This model's primary advantage
over the ISO's is its greater consistency with the original California design
principles. In it, the ISO's primary roles are information supplier (to
facilitate the efficient scheduling of load and generation) and supplier of last
resort (in situations where insufficient generation was scheduled or additional
load was expected to appear in Real-Time). Physical scheduling is accomplished
through decentralized decision making.

In the ISO's approach, however, they are responsible for both forecasting and
obtaining MRE and CC. Although the right for SCs to self-provide these services
has been added to their proposal, the ISO's timelines make it more likely that
the majority of scheduling decisions will be made by the ISO. This is because
their proposal requires SCs to finalize specific arrangements in advance of the
Day-Ahead market. In addition, it is not clear that the ISO's proposed cost
allocation mechanism would provide sufficient incentives for self-provision.

The PX's approach relies on market forces to efficiently allocate LRA resources
within existing timelines. This ensures that procurement of MRE and CC does not
interfere with distort other Day-Ahead scheduling decisions. When the amount of
energy and capacity that is likely to be allocated through this process is
considered, the importance of adopting an approach that compliments, rather than
supplants existing market structures is made clear. ISO statements that the
procurement of MRE and CC could even extend to competitive LPAs in certain
situations only underscores the importance of adopting the PX's approach.


Comments on the FTRs:

The ISO is to be commended for responding to stakeholder input and limiting the
term of FTRs to one year. This will provide required price certainty as the
market becomes familiar with the implications of new zonal boundaries.
Additionally, annual FTRs are more simple to link to zone creation /
redefinition policies. As boundaries solidify, the sale of longer-term rights
should again be examined, however.

Some modifications to the ISO's proposed transmission rights policies are
required though. The ISO should release majority of FTRs in their annual primary
auction. The specific percentage should depend on the amount of rights that can
be released with a predetermined annual level of certainty. If this is done,
secondary markets will develop to reallocate transmission rights. The ISO can
compliment this process by releasing shorter-term seasonal FTRs, as they become
available.

It is understood that shorter-term FTRs may help the market adjust for seasonal
variances in transmission capacity. However, questions must be answered before
the nature and




amount of shorter-term FTR to be released is determined. How significant a
barrier are shorter-term FTRs to the development of liquid secondary markets?
How often is transmission capacity significantly affected by weather or other
considerations (are monthly auctions required or are longer terms such as
quarterly better)? How much of a change in transmission capacity is needed and
how long must the new levels be sustained before the ISO determines that the
capacity in question can be auctioned as an FTR?

Of additional concern is the ISO's proposed method of allocating costs to FTR
holders. The ISO has stated that in the event of topology or operating transfer
limit changes they intend to "keep market participant's financially whole by
creating a balancing account of net gains and losses." Balancing accounts are to
be cleared by each applicable PTO on a monthly basis. The details of this
proposed arrangement need to be provided, but an initial examination of the
information reveals deep concerns.

If the ISO publishes a "library" of shift factors as is stated in their
proposal, and the factors are published in advance of the primary auction, SCs
should be capable of reacting to system changes. For instance, SCs could
purchase enough FTRs in the primary auctions to cover the range of potential
shift factor changes. Those who were unwilling to purchase rights in the primary
auction to ensure against a potential event could purchase required FTRs in the
secondary markets that will be provided by such qualified exchanges as the PX.

If SCs are given information about shift factor changes in time to adjust their
schedules and their transmission rights, they are capable of making educated
business decisions and dealing with the resulting financial implications. Having
a central balancing account where profits are used to offset others losses is
not only unnecessary, it is contrary to the most fundamental free-market
concepts.

Instead of pursing this course, the ISO should treat financial rights in the
same manner as they propose to resolve scheduling priorities: SCs should be
responsible for procuring additional FTRs when the ISO implements new sets of
shift factors. Those who choose not to hedge by purchasing additional FTRs on
the secondary market would expose portions of their schedules to both financial
and physical delivery risks.


Comments on the Forward Market:

We support the efforts of the ISO to preserve the original design concepts of
the deregulation process, which we feel are well represented in the proposed
enhancements to the forward market. However, certain changes could further
improve the recommended modifications.

One such change is the introduction of Inter-SC adjustment bids. While the
implementation of this functionality is outside the scope of the congestion
reform project, its effects are not. Although the ISO's market separation study
clearly shows that the existing market structure is an efficient one, Inter-SC
trade adjustment bids can only




enhance efficiency for those SCs who choose to use them. When the efficiency of
the existing market structure is combined with the fact that decentralized
decision making remains a cornerstone of California's proposed deregulation
model, arguments against any relaxation of market separation become
overwhelming.

There is a proposed enhancement that would threaten market efficiency, however.
Implementation of the ISO's proposed congestion activity rule would create as
many, if not more, issues than it would resolve. While it is understood that
this rule is proposed to provide the ISO the opportunity to always select the
lowest cost scheduling iteration, this is not the appropriate way to insure
against potential market behavior. Rather than reducing the potential for
damaging market behavior, the congestion activity rule would simply increase the
likelihood that both iterations would become distorted. The level of uncertainty
that the implementation of the activity would introduce increases this tendency.
Stakeholders will be less willing to enter into new, binding transactions if
there is the danger that the ISO will force them to be cancelled with any degree
of frequency. If the ISO is concerned about limiting the potential gaming
affects of having two congestion management iterations, it should either adopt
different activity rules or consider eliminating the second iteration entirely.


Comments on Real-Time:

Based on our current understanding, the PX supports the ISO's proposed Real-Time
changes. However, more information is required to fully evaluate them. For
instance, the anticipated dispatch model hasn't been adequately defined and
should be released to the marketplace for examination. Additionally, since the
ISO is proposing to move to a transmission constrained economic dispatch
optimization model to dispatch Real-Time energy, references to Real-Time bid
stacks should be removed from the draft proposal to eliminate confusion. The PX
support of a Real-Time optimization model is predicated on the understanding
that only Real-Time bids will be optimized.

If the optimization model is used to re-dispatch previously scheduled energy
that doesn't have explicit bids (other than in system or local emergencies) many
of the fundamental principles of California's deregulation process are
undermined.


Conclusion:

The PX commends the ISO for their attempts to develop a comprehensive congestion
management solution. However, we urge the ISO to incorporate these reforms into
their proposal. The integration of these concepts with the ISO's existing
proposal will ensure that the maximum benefits of any reform are enjoyed by all
Californians. Failure to do so, however, will significantly blunt the benefits
of any changes.