Steven Brown, Senior Associate Editor
There may not have been a better place or a better time in recent history to launch a large-scale time-of-use (TOU) billing program than the Pacific Northwest at the turn of the 21st century.
This was the setting:
The West Coast, particularly Southern California, was in the midst of a well-publicized, and not soon-to-be-forgotten, energy crisis. Wholesale power prices were hitting never-before-seen peaks. Residential customers of San Diego Gas & Electric opened monthly electric bills that were double—and in some cases a good deal more than double—what they likely would have been in the days prior to deregulation. Southern California Edison and Pacific Gas and Electric customers were protected from soaring bills by a rate freeze, but those companies had their own problems to deal with; both appeared to be on the verge of bankruptcy.
Residential power customers in California were willing to sacrifice comfort for lower bills. And both inside and outside California's borders, electric consumers were showing a receptivity to utility-sponsored programs that would give them at least the impression of being in control of what seemed like an out-of-control energy market.
Utility customers in the Pacific Northwest might not have been as hammered by crisis as their neighbors to the south in California, but there were other reasons an innovative pricing program might flourish there. These customers were at the top of the tech-savvy, environmentally conscious curve—perfectly suited for an energy conservation program with an Internet-based component.
It was a logical setting for an imaginative, time-sensitive billing program, and Puget Sound Energy (PSE) was well-equipped to provide one for its rate payers. That the program might also lower electric bills was not just icing on the cake, but a prime expectation of PSE customers in this time of crisis.
Rise and Fall of a Shining Star
PSE launched its pilot TOU billing program, dubbed Personal Energy Management (PEM), amid this setting. The utility had in place a solid infrastructure to support such a program. They had the interval metering component necessary for TOU pricing, the wireless automatic meter reading (AMR) network for timely data collection, and the customer information system working behind the scenes.
Participating customers would be billed on a tiered schedule, paying slightly more than fixed-rate customers during peak hours, about the same during the midday hours, and less during off-peak, or "economy," hours (9 p.m. to 6 a.m. and certain holidays). Customers would also be able to log onto an Internet website and track their electricity usage from hour to hour. The program boasted upward of a quarter of a million participants, although it must be noted this was an "opt-out" program.
The ballyhooed pilot program garnered industry awards and headliner status at power industry trade shows. It was a bright and shining star in an otherwise gloomy Western power scene, and it had industry participants excited about the prospects of demand response in general.
Then, in November 2002, PSE announced that it would end the PEM program ahead of schedule. Participants had begun receiving TOU summary reports in October, and, according to published reports, 94 percent of those participants learned they had been paying more than flat-rate customers had—although only about 80 cents more per month. Upon seeing these results, participants began opting out of the program at a reported rate of 1,300 customers per day.
"We believe in the fundamental principle that customers who use energy efficiently should be rewarded," Steve Reynolds, PSE's president and CEO, said in a release announcing the pilot program's early end. "But it is clear that the program as currently designed is not giving many of our customers the rewards they expected and that we believe it should."
The bright, shining star had apparently fallen.
What Went Wrong?
So the question arises: What went wrong with PSE's program? Those knowledgeable on the subject of TOU pricing point to a number of factors that may have doomed PEM from the start. Unrealistic customer expectations, a weak rate structure and lack of enabling technology are chief among those factors.
Bill Uhr, managing partner of UHR Technologies, said he believes high expectations coupled with low price differentials hurt an otherwise innovative program. UHR Technologies provides energy pricing and marketing consulting services to the utility industry.
"When you have a watered-down time-of-use rate, you can't expect big savings or big losses," Uhr said. "Basically, you don't see anything."
Participants in the PEM program were on a rate schedule that billed them 6.4 cents/kWh at peak times and 5 cents/kWh during economy hours. (See Figure, previous page, for a comparison of utility time-sensitive rate structures.)
"The thing that was astounding was the fact that they (PEM participants) became so upset about such a trivial amount of money—less than the price of a Coke," Uhr continued. "The only way anyone would get upset about that is on principle. Customers were making some self-sacrifices, and then not only does their bill not go down, it goes up by 80 cents. But even if it went down by 80 cents, big deal."
The "sacrifices" Uhr referred to came in the form of PEM participants shifting such high-energy tasks as washing and drying clothes and dishes to late evening hours to take advantage of the economy rate. Uhr believes that for a system like this to work, physical effort shouldn't be part of the equation. There should be an automated component on the demand side.
Brian White, mass market specialist at Gulf Power, a Southern Company, is another proponent of adding automatic controls to TOU and other demand response programs. The GoodCents Select price-responsive load management program in place at Gulf Power combines TOU rates with a programmable "smart" thermostat and automated load control to help customers achieve savings without expending effort (other than the effort of programming the thermostat twice a year). Gulf Power's voluntary program has nearly 4,000 participants, 95 percent of whom say they are either "satisfied" or "very satisfied" with it. White believes the automated nature of Gulf Power's program has had a big effect on the program's success—and likely had a sizable effect on PSE's lack thereof.
"You have to think about the total cost to the customer in a program like this," White said. "The effort customers have to put forth is a cost to them. If the economic benefits don't outweigh the cost of doing laundry at night or altering behavior in other ways, it's not going to be successful.
"It has to be effortless. It should be a 'set it and forget it' type of program," White continued. "Their problem (at PSE) was the customers had no way of responding, other than physically doing something or avoiding doing something."
White also agreed that the differences between PSE's peak and off-peak rates were probably too small to drive the bill reduction customers probably expected. Gulf Power's GoodCents rate structure ranges from a low of 4.2 cents/kWh (in effect 28 percent of hours annually) to a high of 10 cents/kWh (in effect 12 percent of hours annually). The program also includes a "critical peak" rate of 30.9 cents/kWh, which is in effect only 1 percent of hours annually.
Can TOU Work?
A program like PSE's Personal Energy Management doesn't receive such a large amount of initial praise without having done some things right. PSE showed that an expensive AMR system could be used for much more than simple data collection. The utility also showed that a complex system and billing program could be implemented fairly rapidly.
And, PEM's premature end notwithstanding, proponents of TOU and demand response believe similar programs can survive, and thrive, in the residential market. It's simply a matter of proper application of available technologies in the appropriate locales at the right times.
Gerald Mimno, strategic alliances director for Internet-metering.net, a division of AES-Intellinet, said that PSE's program may have been a victim of circumstance. The conditions that existed in 2000 were conducive to such a program, but they were transitory conditions.
"Their program was created as a result of market volatility, things they hadn't seen in 10 years in terms of droughts and other conditions on the West Coast," Mimno said. "Once those conditions went away, the program had less justification.
"You could almost say this type of program should be held in reserve and only turned on during those periods when it really makes sense in terms of the physics and economics," he said.
Mimno said he believes recent moves by the Federal Energy Regulatory Commission, namely the proposal for Standard Market Design, bode well for TOU and demand response. Demand response aggregation, he said, can be economically viable under FERC's proposed rules.
Uhr said TOU is definitely viable in the residential market, but a shotgun approach won't work. "It takes a very targeted program that goes after a certain segment of the market with high critical peak prices," he said. "It doesn't have to be castles, but it can't be really small homes either; there's nothing there to work with."
An analogy between PSE's pilot program and the failure of deregulation in California—which ironically had much to do with the launch of PEM—can be easily drawn. The amount of attention paid to California's experiment likely soured a nation on the idea of power deregulation. Will the price of Personal Energy Management's fame be the same? Demand response proponents hope not.
"I'm disappointed the Puget Sound project failed," White said. "I wish it hadn't because it might give the impression that customers aren't interested in variable rates.
"But what we've shown is that the variable rate is only half the issue," he said. "The other half is ease of responding to price. That's just as important. Time of use, combined with enabling technologies, is the way the mass market should buy electricity."