Setting Fixed Electric Rates and Bills by Chris Trayhorn, Publisher of mThink Blue Book, May 15, 2006 Today many utilities strive to increase customer satisfaction by offering new products and services to meet customer demands. One of the most basic benefits customers want is certainty, as evidenced by their participation in services that reduce risk and increase certainty such as gas line and appliance protection plans, surge suppression programs, fixed unit rate programs and fixed-bill programs. Consumer Demand for Fixed-Rate Products Fixed Unit Rate Product Energy suppliers and utilities in many markets have offered fixed unit rates to their customers. These offerings of a fixed price per therm of natural gas or per kWh of electricity are popular with customers, offering them the opportunity to lock in the rate they pay for all the energy they use during a contract period. A fixed unit rate eliminates the consumers price risk; however, their bills will still change as weather causes their usage to fluctuate. Fixed-Bill Product Fixed-bill programs guarantee the entire payment amount of the customers energy bill during the contract period, thus eliminating both price and weather risk for the consumer. In a recent survey, E Source found that 25 percent of residential customers were very interested in a fixed bill; a higher ranking than surge suppression, appliance warranties or any other product in the survey.[1] Fixed-bill programs are currently offered in at least 13 states by regulated gas and electric utilities as well as by unregulated energy suppliers. These programs are well-accepted in every market, as evidenced by strong renewal rates of about 90 percent. Risks Associated With Fixing an Electric Rate In order to offer a fixed rate to electric customers, utilities or energy suppliers must manage the associated risks. Examples of these risks and how they are manifested follow. Case 1: Unit-Cost Risk In this example, a hypothetical utility plans a fixed unit rate program for an estimated 50,000 consumers. Each consumer is expected to use 10,000 kWh annually. The rate to be offered to the consumer is $0.08 per kWh, while the estimated cost is $0.04 per kWh. The utility expects to generate $20 million in margin. To demonstrate unit-cost risk, we assume that all variables behaved exactly as planned except the cost to supply the consumer, which jumped from $0.04 to $0.06 due to the doubling of oil as feed fuel (see Figure 1). This unexpected price spike reduces the unit margin from $0.04 to $0.02. The risks driven by changes in unit costs are known as unit cost risks. These risks can be due to market pricing, changing fuel costs or unexpected changes in generation mix such as losing a base-load plant. Case 2: Marketing Risk To demonstrate marketing risk, consider this example of a fixed unit rate program under the same expectations as those in the previous case. In this case, the utility actually secures the cost of the energy by purchasing 10,000 kWh of electricity for each consumer at $0.04 prior to marketing the program. In this example, only 25,000 consumers signed up for the program. The costs of the excess secured-energy supply cannot be recovered from nonprogram customers. The excess supply is sold at a loss of $0.02 per kWh (see Figure 2). Marketing risk is tied to consumer acceptance of a program and the ability to execute an appropriate marketing plan. Components of marketing risk include impacts of press coverage, failure of models predicting consumer purchase behavior and failures in the marketing process such as bad messaging. Case 3: Consumption Risk To demonstrate consumption risk, we start with the same planned fixed unit rate program proposed in Case 2. As in the previous case, the utility actually secures the cost of the energy by purchasing 10,000 kWh of electricity for each consumer at $0.04. However, the average consumer used 16,000 kWh. In this example we assume the additional energy was purchased at market prices averaging $0.10 per kWh (see Figure 3). In this case, the revenue increased by $24 million and the cost increased by $30 million, thereby reducing the margin by $6 million. The risks driven by changes in consumption are called consumption risk. It is this risk that the price paid for supply will be different than expected that causes concern with a fixed unit rate program. There is no need to differentiate the different types of consumption risks for a fixed unit rate program, since the customer will pay for all the energy used at the fixed unit rate. Case 4: Cross-Product Risk Cross-product risk is a mathematical term. It does not refer to a specific risk related to marketing consumer products, but rather considers the interplay of risks. Consider a complicated example involving consumption increasing and marketing errors. In this example 60,000 consumers sign up and use an average of 16,000 kWh. As before, the utility arranges supply for the expected 50,000 consumers at $0.04. Additional power is purchased at an average of $0.10 per kWh (see Figure 4). The compounding effect between the marketing risk and the consumption risk is referred to as cross-product risk. In this example all the unexpected additional customers drive additional supply to be purchased at a loss. The loss on purchasing for unexpected consumption of planned consumers is consumption risk. The loss on unexpected consumption of additional unexpected consumers is the cross-product risk. Additional Risks of a Fixed-Bill Program Each of the risks associated with a fixed unit rate program must also be managed for a fixed-bill program. In addition, the utility or energy supplier must also consider the volume risks associated with changes in consumption. Consumption risks are divided into those risks caused by changes in temperature and those caused by changes in consumer behavior. However, these types of consumption risk are handled differently in a fixed-bill program. In a fixed-bill program, the customer pays a set amount per month for the contract period. This fixed-bill amount is calculated using a fixed unit rate and expected use at normal weather. The changes in consumption caused by changes in temperature are considered weather risk and must be carefully managed in a fixed-bill program. In many cases, this weather risk is held by the energy provider and provides earnings stabilization without the purchase of weather derivatives. Because each consumers fixed-bill amount depends upon their expected use at normal weather, there are additional risks associated with the quality of the model used to predict that normal weather usage. A poor-quality model may introduce selection bias and temperature bias. Using a model that accurately predicts how the consumer uses energy at different temperatures can minimize these biases and the risks associated with them.[2] Impact of Risks on Electric Providers The impact of these risks on any specific electric provider will depend on its unique circumstances. We will discuss two distinct scenarios describing electric providers that bracket the range of possibilities. In practice, each providers situation will fall somewhere between the two scenarios. Secure Supply Scenario In this scenario, the provider is generally a base-load heavy generator with significant net exports of base-load power. The provider usually has a large capital-cost recovery charge and a low fuel cost built into their normal unit costs. For example, a utility with a large portion of their generation provided by nuclear and coal plants will have a $0.10-perkWh price, of which $0.02 to $0.03 is fuel cost. Assuming a $0.02-perunit distribution cost, the capital-recovery cost is $0.05 to $0.06 per kWh. This type of provider can fix their costs to provide the exact amount of energy needed with low or no risk. They have little risk of purchasing either too much or too little energy. When the provider has the great majority of its supply in inexpensive, price-certain generation, all risks can be categorized as opportunity costs. If more power needs to be delivered, it defers external sales that may be more profitable, but it does not trigger loss conditions. Since the unit-cost risk of supply impacts most of the other risks in fixed-bill and fixed-rate programs, it is no surprise that the most prominent early electric fixed-bill programs were found in the southeastern United States utility companies. These companies have large nuclear and coal plants with low unit-cost risk. In other words they are operating in the Secure Supply Scenario. Risky Supply Scenario In the risky supply scenario, the provider is either a net purchaser of power, or a generator dependent on a price-volatile fuel mix such as oil or gas. The provider generally has a small capital-cost recovery charge and a high fuel cost built into its normal unit costs. For example, a utility with a large portion of its generation provided by natural gas and oil plants will have a $0.10-per-kWh price, of which $0.04 to $0.05 is fuel cost. Assuming a $0.02-per-unit distribution cost, the capital-recovery cost is $0.03 to $0.04 per kWh. More importantly, if the fuel costs double, the total cost of power increases by 50 percent. This type of provider may be exposed to grid price risk for some or all of its supply and to the risk associated with changing fuel prices, because most grid supply is provided by otherwise-unused resources that have higher unit costs than those used to supply their own customers. Due to this complexity, energy providers in the Risky Supply Scenario have been cautious about offering fixed-unit rate or fixed-bill programs. Techniques exist to minimize risks in the scenario of risky supply. Feed-fuel purchasing can be hedged for generation mixes. Many grid simulation and hedging techniques have been proven to build hedging strategy for purchased energy. Plant outage risk and storm damage recovery risks can be handled through a combination of insurance instruments and other hedging techniques. Managing the Risks Determining an Appropriate Fixed Unit Rate Figure 5 shows a sample simulation of unit costs for a utility with net purchasing, plant outages, fuel-price volatility and weather for a typical northern utility with significant purchased power. Using advanced simulation methodology, it is possible to determine an appropriate fixed price per kWh that covers the expected costs and risks associated with either a fixed unit rate or a fixed bill. Figure 6 shows the sample cumulative probability distribution of a hedged fixed-bill programs impact on normal earnings versus not having the program. This simulation shows there is approximately 19 percent probability that in any given year the utility will achieve a higher margin without a fixed-bill program and approximately 81 percent probability that this utility will achieve a higher margin in any given year if they run a fixed-bill program. Since the fixed bill carries additional risk beyond price risk, volume modeling is extremely important. The financial performance and hedging creates numerous cross products with other risks. Hedging of price risks requires extreme precision in the modeling and risk simulations. Modeling risk can lead to hedging imperfections and, in the case of a fixed bill, selection bias risks. In these simulations, dependencies need to be considered and conditional branching of purchasing logic needs to be included. For these reasons, and the sheer size of the problem, the use of canned simulation packages or spreadsheets is not sufficient. In addition, proper calculations are very long-running, requiring parallel processing and other supercomputing techniques. Conclusion With the advanced simulation techniques available today, it is possible to identify and mitigate risks in order to set fixed electric rates either to be offered directly to the consumer or to be used in providing a fixed-bill program. Endnotes Residential Marketing Survey 2004, E Source, December 2004. Understanding Selection Bias, The Energy and Utilities Project, Volume 5, May 2005. Filed under: White Papers Tagged under: Utilities About the Author Chris Trayhorn, Publisher of mThink Blue Book Chris Trayhorn is the Chairman of the Performance Marketing Industry Blue Ribbon Panel and the CEO of mThink.com, a leading online and content marketing agency. He has founded four successful marketing companies in London and San Francisco in the last 15 years, and is currently the founder and publisher of Revenue+Performance magazine, the magazine of the performance marketing industry since 2002.