The difference between these positions has had a critical impact on the
process of industry restructuring. In the first instance, it is assumed
that markets will find economically efficient solutions to generation,
transmission, and consumption processes. In the second, it is assumed
that the laws of physics and mathematical modeling will define a better
(in fact the best or optimal) solution to the same problem. The first
approach or model has been identified with concepts of zonal transmission
pricing or, more recently, real-flow or flow-based transmission pricing.
The latter model has been associated with locationally-based marginal
pricing (LMP).

The objective of this white paper is to argue that these two positions,
while having been argued as polar extremes, can, in fact, be combined
to provide both a robust and liquid forward market as well as a secure
power delivery system. Figure 1 illustrates the breakdown of these various
market and operational elements by their relative time domains. Market
functions occupy the greatest amount of time, while operational functions
occur over much shorter periods of time. For example, actual delivery
occurs over the very short “real time.”

The Forward Market

Turning electricity into a commodity that can be traded is not trivial.
In the forward market, however, it need be no different from any other
commodity. The objective of such a market is to provide a means of hedging
risk in both consumption and production. The same conditions apply to
natural gas (a highly similar product) and corn. The forward market trades
the commodity as a financial product, anchored in the knowledge that the
first player in the chain is the producer, the last is the consumer, and
in between the product may trade multiple times. The spot market provides
the final price at the time of delivery and represents the realization
of the expected prices of all of the prior trades. Some of the traders
may guess high on the realization and some low, but the key is that the
market players, looking forward in time, chose to transact, each for its
own reason and each for its own perceived benefit while taking its risk
profile into careful consideration.

The delivery of electricity diverts from traditional commodities in that
it is traditionally considered to be a product that is produced and consumed
at the same instant (i.e., it is not storable). While this is technically
correct from the physical perspective of the electrical engineer/operator,
it is not the case from a market perspective. Electricity is actually
stored as a variety of intermediate products (e.g., as municipal water
stored in water towers). Nonetheless, it is clear that in real time there
is a need for a mechanism to assure that the energy contracted for in
the forward market is deliverable in a manner that will assure the integrity
of the delivery system (stability).

Figure 1
Critical element is time domain – we see that the functions then fill
into this picture.
See
larger image

The Interface

The question then becomes how to create a forward market that functions
and is “deliverable” while assuring the physical operation of the system.
The answer lies in defining, carefully and consistently, the interface
between the forward market and real-time operating mechanics. This definition
must focus on the question of the management of actual or potential congestion
on the transmission system. All the discussion surrounding this congestion
issue, in fact, really reflects the essential underlying debate between
the two models as to how to operate and control the market for electricity.

The polarization of the debate has led to the need to develop a “compromise”
or a “hybrid” model. It is now recognized in the U.S. that such an alternative
needs to allow for a robust forward market as well as, an accurate, cost
allocative mechanism for real-time operations. Indications of this recognition
can be found in the hybrid efforts of ERCOT (Texas), SPP (Oklahoma and
Nebraska), the MidWest ISO (Minnesota to Ohio), and activities in Florida,
which are aimed at responding to the FERC Regional Transmission Organization
(FERC Order No. 2000) requirements .

The Hybrid

Why is the hybrid a necessary outcome? The answer is clear that neither
model alone allows for both the creation of a forward market with a stable
delivery system. The imperfections in the first model can be seen in its
application in California. This approach did not pay sufficient attention
to the potential for congestion within the pricing zones in the state.
The result has been increased costs that must be averaged across all users.
Imperfections are also seen in the second model as applied in PJM. While
assuring that no costs will be averaged, it has not allowed for the creation
of a liquid forward market for delivered energy. Its proponents claim
that the existence of a liquid market at the western hub is sufficient,
but the selling players in the market will not take the risk of delivery
to the eastern markets (the largest markets) because they are unable to
lock in a price for transmission (delivery) of the product. There are
few, if any, useful delivery rights available in either the primary or
the secondary markets.

While calling the proposed model a hybrid may be a misnomer, it is a
useful paradigm for moving forward in the development of liquid regional
energy markets. The key to any hybrid is the recognition that there are
two distinct problems that must be solved and that the handling of congestion
– the congestion management system (CMS) – is at the core of the interface
between these two problems.

The CMS

The purpose of the hybrid proposal is to develop a market for transmission
capacity that will foster liquid, competitive wholesale and retail markets
and provide locational pricing signals for efficient expansion of the
grid. It represents a flow-based congestion management system (CMS) –
Real Flow – that establishes a market for physical transmission property
rights (PTRs) and that internalizes the physics of the electric grid into
market design. The hybrid CMS is additionally designed to permit use of
the existing control area infrastructure in loose pools and to institute
a Regional Transmission Organization (RTO) that will coordinate the reliable
operation of the grid, but play a minimal role in market operations. The
hybrid CMS can also be scaled to operate seamlessly in multiple RTOs across
the Eastern Interconnection.

The premise of the hybrid CMS is that congestion can be managed, preemptively,
in forward markets without RTO intervention. This is accomplished by selecting
a set of commercially significant flowgates (CSFs), whose capacities market
participants purchase and trade in the form of physical transmission rights
(PTRs). Participants trading energy purchase PTRs based on the flow impacts
of their transactions on the CSFs to hedge congestion risk. This aligns
market activity with the physics of the grid. Market participants are
required to submit balanced schedules a day ahead of their use. Changes
are permitted until the close of the hourly market, as long as the changes
have the required physical transmission rights and therefore create no
additional congestion. This market design ensures that, in the absence
of unforeseen system outages, congestion will be managed in advance by
market participants themselves, hence the concept of preemptive congestion
management.

Real Flow permits commercial entities to operate exchanges for energy,
transmission and ancillary services where market participants will trade.
The commercial transactions of the forward market operate independent
of the RTO except in so much as the RTO can make additional PTRs available
to the market. The forward market for electricity takes on the characteristics
of all other forward commodity markets. The product is traded in a financial
transaction that at all times is potentially deliverable (i.e., the participants
in the transaction have the ability to close the transaction with all
three components firmly committed: the source, the transmission path,
and the sink). At the same time, prior to the day-ahead scheduling requirement,
these same elements – generation, transmission and demand – may be traded
independently.

Trading and Market Rules

The key elements of the hybrid market structure are, as stated above,
the rules and realities of passing from the forward market, with its acknowledged
simplifications, to a real-time operating mechanism that can assure delivery
to end users. To satisfy these conditions requires, initially, a set of
trading and market rules proposed with the following conditions.

First, the forward market closes a day ahead. The time period has been
chosen to match the general scheduling time frame of today’s operators.
At this point in time, the requirement is that all transactions that go
to delivery are scheduled and that the schedule be balanced to include
the point of injection, the point of withdrawal and the PTRs required
for delivery. The key point is that when PTR purchases are required there
is no anticipated congestion in the schedules that are handed to the operator
at this time.

Second, the adjustment market occurs between the time of closure of the
forward market and the real-time operating period. During the adjustment
market period the schedules submitted may be modified (and new schedules
added) as long as there is no change in anticipated (based on PTRs) congestion.

Third, the real-time operating window begins at some point prior to the
hour in which the energy is to be delivered. This may be as much as an
hour and as little as five minutes prior to the hour. During this time
period the operator is in full control of the system and uses voluntary
incremental and decremental bids (incs. and decs.) for both the supply
(generation) and demand (consumers) as a means of balancing the system
in real time. The system operator uses the rules of locational-based marginal
pricing to operate the system, minimizing the total cost of real-time
operations, using the inc and dec bids, while assuring that the schedules
submitted are delivered. The operator calculates and reports an LMP for
every monitored node in the system.

These data are then used in calculating imbalance charges. It is critical
to note that any congestion that arises due to unanticipated outages in
the system is adjusted for in real time by the operator through the cost
minimization structure described above. These costs are not, however,
charged in real time to any individual participant in the market (as is
the case with imbalances) but rather are averaged across all system users.

In addition, after the close of the real-time operating window, imbalances
(the difference between scheduled and actual deliveries and demands) that
occurred during operations are then posted and open to trading between
potentially offsetting parties. Any trading requires that the parties
account for locational differences in both generation and demand and thus
account for the price of congestion in their transaction. Any residual
imbalances are charged (or paid) by the RTO at the LMP in effect at the
time of the transaction.

The benefits of the hybrid as described above are obvious: the assurance
of a liquid forward market and locationally specific real-time operations.
The disadvantage is that there is some minimal uplift associated with
the requirement that transmission, once scheduled with PTRs, is financially
firm regardless of any emergency changes in the transmission system.

Conclusion

In summary, the electric power industry in the United States has been
moving forward along a poorly lit path. Different models have been proposed
and implemented. Each has had strengths and weaknesses but neither has
been perfect under all circumstances. The U.S. model of “allowing a thousand
flowers to bloom” has clearly allowed options to emerge, but the time
has come to try to find a model that is, at minimum, a logical transition
and at best an end game for the market structure of what is arguably the
key infrastructure industry in the United States today. The hybrid model
discussed here begins to lay the foundation for the development of both
a liquid forward market and a stable delivery system.

Footnotes

1 The structure of the California market would be an imperfect example
(and one presently being corrected) of this first position, while the
Pennsylvania New Jersey Maryland (PJM) tight power pool that has evolved
into an Independent System Operator structure would be an example of the
second position.

2 The first category is associated with the interests of independent
power producers and marketers (the non-incumbents in the industry). Richard
Tabors, of Tabors Caramanis & Associates, and MIT are associated with
these alternatives. The second category is associated with the interests
of the operators of the power system and the remaining vertically integrated
utilities (the incumbents in the industry). Professor William Hogan of
LECG/Navigant and Harvard University is associated with the latter.

3 This CMS creates a commercial model based on an approximation of the
physical transmission system. This proposed hybrid is based on the work
of a breadth of individuals and group. The initial concepts of a flow-based
congestion control system was called “flow-bat” and introduced by Paul
Barber of Citizens Energy Corporation. The current discussion was initiated
by the work of Ed Cazalet and his colleagues at the Automated Power Exchange
(APX), a California-based commercial power exchange software corporation.

5 A flowgate represents a point (or coordinated multiple points) of physical
connection within the power system that represent both a monitoring point
and a point of potential congestion in the system. A Commercially Significant
Flowgate is being defined as a flowgate that, when constrained, has price
differentials on the two sides that are of sufficient magnitude to alter
the economics of commercial transactions. The precise definition will
be required for each region that adopts this methodology.

6 This practice of averaging and charging to all customers is referred
to as an “uplift” based on the vocabulary developed in the U.K., or as
“socialization” of costs in some literature in the U.S.