3. What is Demand Response?

There are two general types of demand response programs that have been used to reduce demand when energy prices are high or when supplies are tight:

· "price-responsive" programs (in which customers choose how much load reduction they can provide based on either the electricity price or a per-kilowatt (kW) or Kilowatt-hour (kWh) load reduction incentive), and




· "reliability-triggered" programs (in which customers agree to reduce their load to some contractually-determined level in exchange for an incentive, often a commodity price discount).

Both types of programs motivate customers to reduce their loads in exchange for some type of benefit such as reduced energy rates, bill credits, or exemptions from rotating outages. Increasingly the line between these two types of programs has blurred. This blurring occurs because high market price forecasts often coincide with high temperatures and high system or local peak demands, which are two drivers of reliability concerns. When system demand is very high, reserve margins can be low, which puts the ability of the system to serve all the load online at risk in the event of an unexpected generation or transmission outage. When reserve margins fall below acceptable levels, reliability-triggered programs are called upon.

There is not currently a "day ahead market price" established by the California Independent System Operator (ISO), which has limited our ability to offer rates to customers tied to actual market prices or to test a customer's true "price responsiveness" to market prices. Several of the programs the utilities have characterized as price-responsive for 2005 use forecasted temperature or system demand levels to decide when to trigger program operation, rather than a forecasted market price. The "price response" from the customer comes as a result of the utility offering bill credits or other discounts to the customer as a result of reducing its load, rather than the customer responding to a market price or their tariffed electricity rate. Thus the price signal customers are responding to is indirect.

For purposes of this decision, any demand response program that is designed to be triggered the day ahead, whether for price, temperature, or system demand conditions, will be a day-ahead notification program and will count towards meeting the utilities goals for price responsive demand. In contrast, reliability-triggered programs are called on a shorter time frame, the day of, hour of, or as late as 15 minutes before, being needed. It is these programs, designed to truly respond to any emergency conditions, that will be considered "reliability-triggered" programs for today's decision. This delineation is somewhat different than how we, and the utilities, have characterized programs in the past, but helps to clarify the types of programs we are focusing on and why. In comments, some parties criticized that the draft decision allowed day-ahead triggered programs to be counted towards meeting "price responsive" goals. We recognize that the day-ahead programs are not tied to market prices and are thus not truly price responsive. However, because participation is voluntary and does not generally carry penalties for not reducing demand when called, but instead provides a payment for reductions, we believe that these programs operate as price driven programs, with the price set administratively rather than by the market. Reliability programs, on the other hand, carry with them strict penalties for non-performance, making performance obligatory once the customer enrolls. We also clarify, as recommended by Natural Resources Defense Council, that demand response programs should result in a net reduction in demand. To the extent that future demand response programs plan to utilize onsite generation, only programs that rely on clean distributed generation technologies that meet or exceed the California Air Resources Board's 2007 standards and result in a net reduction in demand are eligible for demand response funding.

In theory, price-responsive programs are called on before reliability programs and serve to reduce system load and the need to call on reliability-triggered programs (historically, the interruptible tariffs). The availability of price-responsive load to reduce demand with a slightly longer lead time (generally the day ahead) is an important tool in meeting day-to-day demand requirements; because they have some lead time notice requirements, day-ahead notification programs are valuable for reducing predictable high peak loads. Reliability-triggered programs, like interruptible rates, have much shorter notice times, and serve as an important tool in mitigating unexpected shortages, local distribution problems, or transmission constraints that could result in system failures.

Every rate schedule provides a price signal that causes a customer to place load on the system consistent with that signal. Although all large customers are currently enrolled on TOU tariffs, the current volumetric TOU rates for the largest customers do not send a strong signal to reduce load during the critical peak period because the energy price differentials between peak, mid-peak and off-periods are generally less than 3 to 1. In addition, the summer peak period is currently applied to a fixed afternoon period, generally from May through September, whereas the most critical peak loads are of much shorter duration. Without modifying our rate design, customers will not have strong ongoing price incentives to systematically move their load during critical peak demand periods off of the system. If we truly want to reduce our critical peak demand, we must modify our rate design to provide a stronger price signal to customers to shift load out of the critical peak. We have begun this process through the joint ALJ/Assigned Commissioner ruling issued on December 8, 2004.¹

Several commenters criticized the draft decision's statements about the incentives large customers have to move load off peak, stating that the price differentials are significant to shift off peak. We clarify that our point in making these statements was to highlight the fact that current rate design does not send customers price signals to differentiate the expectation of a critical peak load versus a day with regular load conditions.

As a result, the day-ahead notification programs that we will approve for 2005 will focus on providing incremental peak demand reduction driven by day ahead high temperature, price, or demand level forecasts. The reliability-triggered programs we will approve for 2005 will focus on providing quick response and targeted locational load reduction capability. We will also carefully review and approve technology and technical assistance programs to automate customer response to demand reduction signals, and education of customers about their power to reduce their bills by driving their load off peak.