But while the hub development process in the power market is progressing, it is far from complete. And as market participants in other energy commodities have learned, the decision to trade at a particular hub using a particular contractual instrument, can be as important - if not more important - than the price of the trade itself.

In fact, the power business already has some battle scars from the shifting of trading hubs and trading instruments. For example, earlier this year the imposition of Location Marginal Pricing (LMP) for 1,600 buses at PJM¹ effectively split the established "Seller’s Choice 500KV line" delivery point into a west hub and an east hub. The western hub has a significant level of activity, but limited reliable access to high demand areas, while the east delivery point has the high-demand access, but is occasionally constrained during on-peak hours. Liquidity dried up in both halves of PJM and took several weeks to recover. A number of traders that were caught with out-month positions at the old 500KV hub delivery point incurred substantial financial losses to unwind their positions.

In this paper, I will provide background on the current state of hub trading in the power market, focusing on the hubs which are responsible for the majority of the industry’s trading activity.

I will also address these questions:

• How have Hubs Evolved in Other Energy Commodities? What lessons can be elicited from the commoditization of crude oil, petroleum products, natural gas?

• How will Hubs Evolve in the Wholesale Power Market? Why some hubs could succeed while others fail? And finally,

• How will Traders Manage Hub Trading Delivery Risk? And how will they leverage the evolution of hub-based trading in power to improve trading performance?

What are Hubs?

A good definition and description of a power hub can be found on the PJM webpage, as described below:

A hub is an aggregation of representative buses grouped by region. Hubs create a common point for commercial energy trading.

1. What is the business reason for Hubs?

• Hubs create a common point for commercial trading contracts to settle with or without going to physical delivery.
  
• Hubs are intended to create price signals for geographical regions of the control area by aggregating a group of representative buses.
  
• The creation of hubs reduces the risk of delivering to one particular bus whose price is more volatile during a constrain than a collection of bus prices at a weighted average.

2. Can a hub be a source or a sink?

In the case of PJM: yes.

Where are the Hubs?

Potentially the power business could have a very large number of hubs. FERC has jurisdiction of 166 utilities identified on NERC’s map of regions and control areas each which theoretically could become a hub or even more than one hub depending upon the utility’s transmission system. But, according to trade publications that track spot prices for next-day power, only about 20 points are actively traded.

Five of the points: Mid Columbia, COB, Palo Verde, Four Corners, and Mead are located in the west. There are seven points in the central U.S. area: ERCOT, Ameren, ComEd, MAIN north, MAPP, Into Entergy and SPP. The remainder are situated in the east, and include: Into Cinergy, north ECAR, PJM-west, NEPOOL, NYPP, Into TVA, southern Florida and the Florida-Georgia border. Although some trading occurs at all of these points, power trading is primarily concentrated among three major hubs in the east: Cinergy, Entergy and TVA, and three points in the west: COB, Palo Verde and Mid-C.

Clearly power trading volumes are concentrated at a small number of locations. But is this a negative aspect? Why is the market behaving in this manner? And what is occurring at the remaining points? For answers to these questions, we can look to the experience of other energy commodities.

How Have Hubs Evolved in Other Energy Commodities?

For crude, petroleum products and natural gas liquids, there are only a scant number of critical market hubs, each with its own trading standards and execution tools. In the crude oil business, there are four major market centers: Cushing, OK; West Texas; Gulf Coast; and the New York Harbor. Petroleum products utilize four: the New York Harbor, the Houston Ship Channel, Los Angeles and Group 3 (which basically covers the midwest). In natural gas liquids, market centers are located at: Mt. Belvieu (Texas), Conway (Kansas), Sarnia, Ontario, and the Los Angeles Basin. For crude, products and natural gas liquids, other points such as production facilities, refineries and storage terminals generally trade in an active location-arbitrage marketplace at market transportation differentials, relative to these major market center points. This relative pricing structure is what makes hub-based trading work across an entire market.

A futures market has only been successful at two of these trading locations - Cushing for crude oil, and the New York Harbor for petroleum products. On the other hand, electronic trading has been very successful in the natural gas liquids market, with over 40% of spot trades consummated online.

Natural gas prices are reported daily at about 100 points, and the commodity is traded very actively at over 30 major locations. Yet natural gas futures have only been truly successful at the Henry Hub. Hub-based electronic trading has captured a large portion of the next-day trading market, with 150 to 250 trades each day in the U.S., and an additional 200 trades in Canada.

Key Characteristics of Successful Hubs

When we look at successful energy trading hubs across the various commodities, six key characteristics are consistently present:

1. First, the location is a natural supply/demand balancing point for a particular market. It is not necessary that a large number of transmission systems come together at a single point, although it does seem to be an advantage.

2. Second, a successful point requires reliable contractual standards for delivery and receipt of the energy commodity. Players must be able to rely on delivery mechanisms where the costs and the ‘risk of curtailment’ are known quantities.

3. Third, there must be transparent pricing at the point. Participants must be confident that trading is on a relatively level playing field, and that no single player nor group of players can grossly manipulate the market price.

4. Fourth, there must be homogeneous pricing across the hub. If prices vary across the hub and there are no trading conventions to handle the situation, sellers will always seek to deliver the commodity at the cheapest location, while buyers will always seek out the most expensive point. If left inadequately addressed, the point will probably not be actively traded.

5. Fifth, there must be convenient tools to execute trades and aggregate transactions. This includes execution tools like electronic trading systems and futures exchanges, plus scheduling mechanisms to provide the capability to adjust a portfolio of supply to a portfolio of demand.

6. And finally, yet most important, there must be a critical mass of buyers and sellers that respond to the five characteristics listed above, and actively trade the market on a consistent basis. This is the definition of liquidity, which is clearly the most critical requirement of a successful trading hub. And it is the importance of liquidity that drives markets to concentrate liquidity to as few locations as possible. Thus, we can conclude that the concentration of liquidity at a limited number of points is not a problem for the market. This geographic concentration of hubs functions efficiently for crude, petroleum products and natural gas liquids.

Likewise, we can also conclude that each commodity is unique, requiring trading instruments and execution mechanisms designed specifically for each individual energy commodity. For example, natural gas has evolved a more diverse hub trading structure than the other energy commodities.

We believe that this diverse structure is due to a number of factors, the most important of which are: (a) the physical characteristics of the natural gas delivery system, and (b) the relative value of gas transportation versus the gas commodity value.

In other words, most of the physical gas supply system is targeted toward specific regional markets with unique pricing dynamics, and even more important - the cost of gas transportation and storage is high relative to the cost of gas at the point of production. Thus, the market for gas is more localized than for the other energy commodities, resulting in a greater number of actively traded hubs and a very active location arbitrage (basis and EFP) market. This also drives the gas market to electronic trading tools, which make it easier to discover prices and execute trades at a large number of trading points.

Wholesale Power Market Hub Structure

From what we know about other energy commodities, we would expect the wholesale power market to be characterized by:

• A large number of geographically diverse hubs used by the market to effect delivery of the physical commodity;

• An active location arbitrage trading marketplace;

• A robust electronic trading market, consummating hundreds of trades each day; and

• Pricing at locations distant from hub locations, based primarily on transmission cost differentials from major trading locations.

However, essentially, the wholesale power market possesses none of these characteristics.

The market is confined to very few locations and arbitrage is a limited financial game frequently tied to very few hubs or the natural gas market. The top 10 to 20 players are responsible for the vast majority of the market activity. Electronic trading beyond the mandatory California PX system is negligible, and up to two thirds of even daily trades are booked out, resulting in minimal physical deliveries. Additionally, there is a faint relationship between pricing at the major hubs and pricing at nearby non-hub delivery locations.

This schematic is more representative of the true structure of the power market. Over 85% of power trading is conducted at these 10 trading points. Cinergy, Entergy and TVA are the core of the market east of the Rockies, with ERCOT, PJM, ComED and NEPOOL filling out most of the remaining daily marketplace. In the west, most bilateral trading is at COB, Palo Verde, and Mid Columbia, while the California PX dominates the next-day market.

A location arbitrage market exists, but it is almost exclusively between the seven eastern points as a group, and the three western points. There is virtually no arbitrage trading between east and west.

When combined with existing and planned futures contracts also shown on this schematic, the points on this slide nearly encompass virtually the entire power trading market as it exists today. The 10 cash market hubs, combined with the six futures contracts (plus one announced location at PJM), and the NYMEX gas futures contract which is used by a number of traders as a surrogate for fuel costs, make up the structure of the wholesale market.

We believe that the variance between this market structure and its expectations (listed previously) occur mainly due to:

• Unreliable and expensive transmission discourages transmission between hubs to capture pricing differentials;

• Unwieldy tagging rules make it difficult for hubs to be used as true physical aggregation and balancing points;

• Several hubs have been defined so broadly that prices vary across the hub, defeating the requirement of homogeneous pricing across the hub; and

• Primarily due to the June price spike, the number of potential counterparties at any one hub has been reduced by market withdrawals, cutbacks, and increasingly stringent credit requirements. In other words, there is not a critical mass of players at some of the hub locations.

The power market still has a long way to go before the regulatory framework and business processes have evolved to the point where a liquid, hub-based marketplace for physical power exists. This immature state of the power market is not due to any unique physical properties of electricity, but instead has primarily resulted from the lack of progress in the development of an truly open transmission network.

Implications For Power Marketers And Traders

The structure of today’s power market is unstable. It is an evolving market where the risk of being surprised by unforeseen events remains relatively high:

• Active hubs are at risk of being refined, drying up, or fragmenting into multiple pricing points. Long-term trades must be structured to provide an exit strategy for either side of the trade if delivery rules or trading conventions at the contractual delivery point change materially.

• Due to the difficulties in moving power over the transmission grid, location basis risk may be as high, or higher, than position trades in the outright commodity. It is difficult to transport power to correct regional imbalances.

• The limited number of trading points today are susceptible to the domination of a very few players.

• Because physical power underlies only a relatively small portion of spot market trading, pricing on futures contracts may not converge with the physical spot market; putting hedging strategies in question.

• The relationship between hub prices and the value of wholesale power at adjacent points will be volatile.

This view of the power market is not presented here to discourage trading, hub development, or long-term transactions. On the contrary, the intent is to encourage market participants to manage these risks prudently by utilizing trading tools, contractual provisions and information systems that are designed to minimize these risks. Furthermore, the intent is to encourage progress toward greater access to the transmission network by the trading community, and the development of greater standardization in business processes and contractual provisions.

In Order 888, the Federal Energy Regulatory Commission (FERC) stated its goal "to remove impediments to competition in the wholesale bulk power marketplace and to bring more efficient, lower cost power to the nation's electricity consumers." Clearly the immaturity of today’s trading hub structure is exactly this - a considerable impediment to open competition.

In conclusion, the FERC, state commissions and the power trading industry must work together to insure that the market’s regulatory framework encourages a viable, bilateral, hub-based power marketplace.