David A. Foti and Martin F. Nellius III

Retail energy markets entail a unique set of risk management challenges.

The emergence of unregulated retail energy has spurred the formation of a number of new entrants who hope to reap the benefits of a promising new market. Serving the deregulated retail market entails unique risks that energy marketers have not had to concentrate on in the past, specifically full requirements consumption risk, tariff risk, and certain operational risks such as mass-market billing. Successfully dealing with these risks will require a focused risk management initiative but should result in a competitive advantage for those retail energy marketers that are able to execute effectively.

Retail energy marketing was truly born when California opened up its $20 billion1 power market in 1998. The initial rush to capture market share resulted in such schemes as free power give-aways and unholy alliances with Amway. The initial enthusiasm waned when retail energy providers (REPs) found that they were unable to offer significant discounts to the utilities' standard offer due to the mandated Competitive Transition Charge (CTC) pass-through. As the economics became clear, marketers either rethought their business plans or left the market altogether.2 For example, Enron Energy Services abandoned the residential market in late 1998 after racking up big losses on advertising programs to entice customers to sign up.

By the end of 2001, a repeat of the California shake out was playing out on a national scale. The three most promising national players-Enron Energy Services, Shell Energy Services, and New Power Company- respectively halted, dramatically scaled back operations, and stood on the edge of financial failure. While a myriad of factors punished these major players, some of their problems could have been ameliorated with a structured risk management program.3

Retail Energy Marketing vs. Wholesale

Wholesale energy marketing and trading currently has three main incarnations, depending on corporate structure and objectives: on-system wholesale trading in support of production (characterized by asset control and physical trading capabilities), wholesale marketing (characterized by a strong origination & structuring skills), and on- and off-system wholesale and financial trading (characterized by complex modeling skills. ()

Wholesale gas marketing and power marketing, which were unleashed by FERC orders 436 and 888 respectively, are the immediate antecedents to the retail energy market. What specifically differentiates a retail from a wholesale marketing transaction? "

Breakdown of Tariff Risk

Tariff Risk Tariff risk is that risk which the marketer incurs downstream of the uplift. This risk can be broken into Timing Risk (I - III) and Magnitude Risk (IV-VI) as illustrated below. Timing Risks Each of these items have to do with transition period timing. Period I represents the pre-transition duration. During pre-transition customers only have access to the local utility's bundled rate. Period II is the length of transition. Typically during the transition period, customers have the option to either to keep on taking the utility's bundled rate, or standard offer, or procure generation from a third-party marketer though an unbundled tariff. Period III represents the duration of the competitive transition charge (CTC). The CTC surcharge is typically charged to all customers starting at the beginning of the transition period regardless of whether they chose the standard offer or unbundled tariff. The CTC is meant to compensate the local utility for its stranded costs under as a result of deregulation. An inaccurate guess on the timing of these events will result in a long or short generation for the retail marketer. Also, there is the potential for losses related to unrecoverable CTC charges if period III lasts longer than expected. Magnitude Risks Periods IV, V and VI have to do with the magnitude of the tariff charges. Area IV represents the CTC surcharge. CTC amount varies from utility to utility depending on their specific stranded-costs and each state can have its own methodology for determining whether CTC is a fixed or variable charge (e.g. Illinois is residual while Texas is fixed1). A strategy among some retail marketers is to give a discount off of the current rate which includes CTC in the hopes of making up for some early losses when CTC ends. Thus an inaccurate forecast of CTC amount can cause deals to be mis-priced and create losses into the future. Zone V is representative of the generation prices the local utility passes though. These prices have become increasingly volatility as utilities have passed through higher generation fuel costs through their periodic fuel trackers. A retail marketer will be exposed to this risk at least through the start of transition and then possibly thereafter if the standard-offer rate is lower than the unbundled rate. Area VI covers the local utility's transmission and distribution (T&D) charges. These charges are generally determined by cost-of-service based regulation. If the retail marketer chooses to give its customers a total fixed-priced deal it must incur the T&D price risk. On a system-wide level T&D prices are comprised of many factors. Each of these factors must be taken into account in order to calculate a system-wide forecast of an individual utility's rates into the future. System-wide detail is probably not adequate in most circumstances, therefore the analysis should drive down to the tariff level. -D.A.F. and M.F.N. Unit 2004, with true-up thereafter.

The primary differences are load shaping, scheduling, and swing. For instance, a full requirements deal is by definition retail since the marketer would have to follow their demand by scheduling constantly. Wholesale is blocks of power. The standard traded quantity in ERCOT is 50 MW 5x16, meaning 16 hours a day, 5 days a week, 50 MW flat, no fluctuations in demand. Add in odd size blocks (52 MW for instance) and ancillary services such as spinning reserve, line losses, load following (or swing optionality for us financial types) and you get retail generation. Adding transmission and distribution, CTCs, system benefit charges, and other tariff type things gets you to retail bundled," says Mark Courtney, the former director of origination for Enron Energy Services.

Unique Retail Energy Risks

The characteristics that differentiate a retail energy deal also imbue it with certain unique risks beyond those incurred by a wholesale marketer. These risks help define the retail energy market and can be placed into two broad categories: volumetric risk and tariff risk.

The main components of volumetric risk are consumption, forecasting and portfolio build risk.

Consumption risk falls into two categories: load profile and usage. () Since retail deals are by definition full-requirement deals, it is incumbent on the marketer to estimate both the customer's load shape and overall usage when pricing a new deal. Load shape will affect the demand charges the marketer incurs and the ratio of on-peak to off-peak power that he has to buy.

Usage will affect the marketer's energy charge. A large shift in either factor has the potential to change a deal's cost structure. Usage risk can be further divided into hourly and daily swings. Fluctuations in day-to-day and hour-to-hour loads, based on weather conditions and various plant operations, can result in mismatched hedges or higher cost supply products from generators.

A close cousin to consumption risk is forecasting error. While consumption risk is an ex ante risk, forecasting error is ex post deal consummation. Forecasting error is the difference between a customer's actual day and month-ahead usage and his forecasted usage. Differences here can leave a marketer either long or short subject to the extreme volatility of the spot market.

Bid-Offer Spreads: A Hedging Device

The Bid-Offer spread was suggested above as one of several possible mitigants for many of the listed risks. How exactly does a retail energy marketer use the spread as a hedging device? Generically, the Bid-Offer spread represents the profit a market-maker or intermediary demands for creating liquidity. This spread is composed of the intermediary's variable cost per deal plus any liquidity risk they may bear. In liquid markets the bulk of the spread will be variable cost-based and in illiquid markets it will be dominated by the liquidity risk factor. Newly deregulated energy markets are often characterized by a lack of liquidity. A player who wishes to create markets in such a nascent environment has at least three options for spread determination (in ascending order of complexity): Calculate the value-at-risk for the time period estimated it would take to liquidate the position and add the associated capital reserve charges and expected discounted sales value to the spread. Calculate the historical volatility of the underlying instrument and do a Monte Carlo simulation to set the mid-offer spread so that some percentage (e.g. 95 percent to 99 percent) of all deals done will have a profit greater than zero. This method may not always be viable since there may be no available data on the historical performance of the risk examined (e.g. unbundled transmission or distribution rates). Decompose the illiquid risk into its component parts and individually hedge those components that are liquid. In this case the hedging cost would be included in the spread. For the illiquid components proceed as #2 above. This method can be problematic since the cross-correlations between the component pieces must be known in order to properly take into account the portfolio effect. -D.A.F. and M.F.N.

Finally, portfolio build risk is associated with the difference in price from the time the customer offer is made (or accepted depending on contract terms), and hedgeable levels are reached within the portfolio. These levels are subsequently backed with an analogous purchase. The risk associated with build volume can also pertain to the time it takes the portfolio size to reach optimal levels. This risk can have a potentially significant impact, since the granularity of retail loads are extremely difficult to hedge in the wholesale market, leaving a portion of the retail portfolio exposed. This unhedged volume is subject to the volatility and directional movement of the power market. It will eventually reach a level that can be backed, though not exactly, in the wholesale market. Using the components of volatility, time, price and volume, an exposure can be determined. The cost of this exposure should be passed to the customer.

Of course, all the other risks a retail energy marketer faces-business, commodity, credit, etc.-while significant and problematic, cannot be considered unique to this market segment. For these, the retail energy marketer can look to the best practices employed by their wholesale counterparts for mitigation strategies.

To protect themselves against the unique risks they face, retail energy marketers will need to be innovative and leverage off of their existing core competencies in risk management to build new capabilities. Specialized retail-hedging groups and programs should be created to properly capture and manage this unique exposure. Outside experts can be brought in as well. Regardless of whether this is done internally or externally, the retail energy market leaders of tomorrow will be the ones who combine their expert knowledge with proper risk assessment and control.

1. http://www.eia.doe.gov/cneaf/electricity/esr/t07.txt
2. In March 1998 there were more than 300 REPs registered with the CPUC, as of June 2001 there are 28.
3. E.g. EES reportedly suffered market risk with their short position in California during the 2000-2001 power price spike, SESCO suffered credit and operational risk with their billing operations in Georgia, New Power has suffered credit and operational risk (billing and charges of "slamming" for customer acquisition in ERCOT).
4. These risks are not new risks - they existed under prior to deregulation, but their market effects were softened as a result of the regulatory paradigm.

Articles found on this page are available to Internet subscribers only. For more information about obtaining a username and password, please call our Customer Service Department at 1-800-368-5001.