EMF SR 3: Survey of Utility Real-Time Pricing Projects in the U.S.
Recent advances in metering and control technologies have reduced the costs of spot pricing to a level where it is being considered a viable option for many utilities. As those utilities find themselves scrambling to keep their largest customers, a handful of utilities in the U.S. such as Pacific Gas and Electric Company (PG&E), Niagara Mohawk Power Corporation NMPC), and Southern California Edison Company (SCE) have begun offering their users what are known as “Real-Time Pricing (RTP)” programs, where rates are calculated each hour to reflect the cost of providing power for a given incremental load. Prices are preannounced, typically twenty-four hours in advance for most recent programs, and vary on an hourly basis. Methods for implementing RTP programs vary from utility to utility. Some calculate rates based on actual supply conditions, while other plug in predetermined pricing scenarios according to a defined set of environmental conditions. Most of the utilities in the U.S. that offer RTP are doing so on an experimental basis, usually involving fewer than 20 users. Depending on the utilities generation costs and the customer’s usage patterns, saving under RTP plans can often be in the range of 10 to 30 percent. <br /><br />Experience with real-time utility rates is growing. In this report three major RTP experiments in place at PG&E, NMPC, and SCE are reviewed. The experience to date with RTP has been positive. Those RTP rates that have entered utility operations most effectively have been for larger industrial and commercial users where economic incentives play a larger role. It is notable that commercial sector enterprises such as a food distribution center, a commercial laundry, and a large office building with an energy management system have responded more to hourly price signal than large process industries. <br /><br />RTP is still in its early stages of use, as compared to other service options such as interruptible service and time-of-use (TOU) rates. Previous economic analyses have demonstrated that spot pricing holds the greatest potential for optimally improving service quality to customers with diverse values, but offering pure spot pricing to utility customers is not practically feasible for a number of reasons. RTP, which approximates spot pricing more closely than other service menu options, holds great potential for efficiency improvements, if employed on a broader scale. Whether enough users will be able to respond to hourly price signals is a question that remains to be determined. <br /><br />In this report an attempt is first made to estimate the potential efficiency gains and price sensitivity of a customer to RTP. First, I examined the advantages of RTP when compared with TOU rate due to the superior adaptability of spot pricing to unpredictable fluctuations in demand over time. As marginal costs increase steeply and price elasticities increase, the load leveling effects of RTP are larger than those of TOU rates. The results of a simple Monte Carlo simulation demonstrate the following: <br /><br />1. RTP is very effective in its response to demand fluctuation. For example, PG&E’s load factor under hourly RTP is 68.6 percent, approximately 9.6 percentage points higher than that under TOU rates (with three pricing periods). Unit generating costs were about 1.5 percent lower under RTP. <br /><br />2. The effects of load leveling and cost gains of RTP are very sensitive to price elasticities and load fluctuations. As price sensitivity and the randomness of hourly load increase, improvements in load factors and cost savings resulting from RTP increase. <br /><br />3. When unit cost savings under RTP are estimated in comparison with a three-period TOU rate and hourly TOU rate, cost savings are: 77-82% due to load adaptiveness and 23-18% due to the difference in the length of the pricing period.<br /><br />The second objective of this report is to estimate the effects of RTP on a customer demand and the price sensitivity of a RTP customer using real hourly load and RTP tariff data for a refrigerated warehouse. The resulting price elasticities seem unrealistic. This detailed study is limited to only one customer. More engineering and economic data samples are needed. These responses are short-term. A necessary condition for customers to make capital investments is the establishment of an RTP program. Unless such a commitment is made, customers will be very reluctant to make capital investments. Nonetheless, and RTP rate design brings substantial savings to customers even if they do not respond to RTP by making appropriate investments. Actually, as a marketing strategy the utility can provide RTP to customers who have access to alternative power sources to keep them on its system. <br />