Lisa Wood, Ph.D., (email@example.com) has recently joined the Brattle Group in the Washington, D.C., office. She specializes in economic and regulatory issues in the electric industry. She wishes to acknowledge Joe Wharton, Frank Graves, and William Moss of the Brattle Group, for their insightful comments for this article.
There is a timely and renewed national interest in price-responsive demand among utilities, independent system operators, policy makers, and regulators. The Federal Energy Regulatory Commission and Department of Energy co-sponsored a conference on demand response in February. Independent system operators in California, New York, and New England initiated widely publicized price-responsive retail load programs last summer. And individual utilities have recently introduced a variety of price-responsive demand programs. Puget Sound Energy developed what has been perhaps the best-known program, which placed 300,0000 residential customers on a time-of-use price as the default rate. It was fully voluntary, but customers had to opt off, not on.
On the energy policy side, officials like FERC Chairman Pat Wood are touting price-responsive demand as a way to potentially increase the effectiveness of wholesale markets. In the FERC’s “standard market design” white paper issued in March, demand response is included as an element of market design. The Senate energy bill approved and sent to the House in April would require utilities to offer real-time pricing to customers that request it. Both this bill and the energy bill passed by the House in August include tax credits and accelerated depreciation for advanced metering systems.
Price-responsive demand programs are not new to the electric industry. What’s new is the high-level public discussion of their desirability—and even necessity. Public policymakers have long been willing to offer price-responsive demand programs to customers on a purely voluntary basis. But with a few notable exceptions—including some residential load-management programs in Florida and Gulf Power’s residential dynamic pricing program—small customers saw no reason to overtly choose them, and rate-regulated utilities gained little from marketing them. As a result, retail price-responsiveness languished in the wings.
Clearly what is new is the impact that price spikes and disruptions in emerging competitive bulk power markets have had on customers’ bills, system reliability, and the financial viability of delivery companies. Energy companies, policymakers, and customers—even residential customers—are now recognizing that price responsiveness may be a relatively effective way to improve market performance, reduce bills, and improve reliability in regions with constrained capacity.
The need to foster price-responsive demand by innovative public policy should not be surprising to those who have followed the research on customer preferences for electric options. During the past five years, two large EPRI-sponsored studies measured the willingness of different types of customers to adopt price-responsive demand programs. These studies showed little willingness of customers to adopt time-of-use rates relative to fixed rates. As expected, the preference for time-of-use rates did increase with customer size: Large business customers had the greatest preference for such programs, while residential customers had the least.
The 2000 EPRI study concluded that, on average, customers prefer fixed rates to time-of-use rates, and time-of-use rates to hourly pricing. However, these studies did not explicitly separate the impact of “inertia,” or the preference for the status quo, from a real preference for fixed rates. In other words, the studies leave open the possibility that small customers consider electricity such a low priority that they lack the time or desire to consider a different rate offering, rather than that they really prefer fixed rates. If the environment changed, is it possible that customers would see price-responsive demand in a new light?
Are Customers Opposed—or Complacent?
Puget Sound Energy’s experiment, having completed its first year in June, strongly indicates that, under the right circumstances, residential customers are quite willing to participate on a price-responsive demand rate and change their behavior in response to price signals. Residential customers bought in because the Bellevue, Wash.-based utility offered price-responsive demand in a new way: The program was unusual in having advanced communications and metering in place to measure customer usage by time periods. By the spring of 2001, customers knew from media coverage and the utility’s educational efforts that the state of Washington was facing an energy crisis caused by the combination of a major drought and the energy cost crunch linked to California’s power shortages. With approval from the Washington Utilities and Transportation Commission, Puget Sound switched 300,000 residential customers to a time-of-use rate. The rate was voluntary but it was the default; if customers wanted to return to the flat rate, they had to opt out.
The results of this experiment are striking. First, less than 1 percent of residential customers chose to opt out of the time-of-use rate. Second, residential customers adjusted their electricity consumption in response to time-of-use rates in the very first month of the program and have continued to respond at the same level. On average during the first six months of the program, they reduced electric usage during the peak periods by about 4.5 percent and increased it during the off-peak period about 5.5 percent. In addition, the price elasticity estimates are the proper sign (negative) and in the range of –0.34 to –0.61, indicating that customers will decrease their energy consumption by 3 percent to 6 percent in response to a 10 percent increase in price. (In this case, the elasticity of demand estimates are approximate because they are based on how customers respond in each time-of-use period to the changes in all other prices taken together, rather than on how they respond to one price change, as is typical.) These econometric results are in line with the results from the large body of literature on time-of-use experiments conducted in the 1970s.
For residential customers, inertia for the default option rather than a preference for fixed rates may be the primary driver of choice; customers simply may tend to “choose not to choose.” Based on estimates from a 1998 EPRI study that simulated residential customer choices for alternative rate designs, under a best case scenario about one of five customers would “opt in” to a time-of-use program. However, that assumes that customers are aware of the program, there are no transaction costs for switching, and inertia is not a factor. To the extent that these assumptions are false, the actual opt-in rate will likely fall to perhaps one in 20 customers.
At a time when policymakers at both the state and federal levels are promoting demand-response, one important question looms: What should be the “natural” or default rate design in our new world of competitive wholesale power markets where forward prices exhibit more sharply defined seasonal, weekly, and daily price patterns, and spot markets exhibit random price spikes? There is little evidence that we’ve made a dent in residential customers’ strong inertia. However, we do have evidence that these customers are both willing and able to adjust their energy consumption, particularly via home automation.
Getting Their Buy-In
If inertia is the barrier, one way to overcome it is by introducing a default rate design that takes advantage of the energy-shifting potential among residential customers. Policymakers and utilities need to provide all reasonable protections for customers. But particularly in states that are pursuing sustainable competitive retail markets, we do not want to be “killing the market with kindness.”
A recent nationwide survey of residential customers conducted by Quantum Consulting found that residential customers are willing to participate in load-control programs. For example, about 50 percent of non-participants said they were willing to sign up for air conditioner load control, 33 percent of non-participants said they were willing to purchase water heater load control, and those already participating in such programs wanted to sign up more appliances.
Despite these findings, utilities historically have not experienced such high participation rates for load control programs. Again, like the time-of-use programs, non-participation may be the result of inertia for the default option rather than an explicit decision not to participate. In the case of direct load control, non-participation may also be related to specific program design elements such as inability to override an interruption.
There is exciting new evidence that residential customers may be responsive to policymakers’ calls for modifying their energy usage for the public good. However, federal and state policymakers may need to send a unified, clear signal. If price-responsive rate programs are proven to be economically efficient and residential customers can respond to price signals, policymakers may be willing to support them as the “new plain vanilla” default rate option for mass-market customers. That would dispel the myth that residential customers simply don’t want time-of-use rates, and put us over the enormous “inertia” hurdle.
We can then turn to the other major issues that need to be addressed before launching price-responsive demand programs: the communication, controls and metering costs, the overall program cost-effectiveness, and the net benefits for all stakeholders.