Innovative ratemaking is the talk of the town. Trials of time-of-use rates, demand charges and time varying pricing are playing a growing role in the transformation of the electric power sector.
California will deploy default time-of-use (TOU) rates in 2019 at an unprecedented scale. Landmark regulatory debates across the country have been resolved in recent months by stakeholder agreements to explore new ways to use rates to control spiking peaks. And research is beginning to point toward what works and what doesn’t.
“For TOU rates, the magnitude of the impact on peak demand is consistent with the size of the ratio between off-peak and on-peak rates,” Brattle Group Principal Ahmad Faruqui told Utility Dive.
His conclusion is based on a Brattle study of 300 TOU pilots and trials. He also concluded that “TOU rates solve yesterday’s problem." California’s move to TOU rates "will not be a failure but will produce peak demand changes that will be too little, too late," Faruqui said.
"California needs dynamic pricing," he said. Unlike TOU rates, which include a modest price differential on each day, dynamic pricing involves alerting customers to steeper increases in per-kWh rates in advance of specific peak demand events. Instead of a small differential every day, the larger gap between peak and off-peak pricing is meant to drive more significant reductions during the highest demand days.
Jayant Kairam, Environmental Defense Fund (EDF) California Clean Energy Director, sees TOU rates as a way the state can reliably and cost-effectively deploy DER. Research done for state regulators described TOU rates as “a type of load modifying demand response” that could help save “up to $700 billion annually by 2025” by optimizing now-curtailed renewables, he said.
But Faruqui wrote recently that deployment of advanced metering infrastructure (AMI) has made possible more complex rate designs that incorporate time-varying pricing.
Dynamic pricing has the potential to be a “win-win-win opportunity” by better aligning pricing and costs with price signals that guide customer usage, he noted. That can “help utilities achieve lower generation and distribution costs as well as renewable energy integration.”
The question for policymakers is whether consumers are ready for dynamic pricing.
TOU rates in Ontario and California
Faruqui said the rollout of default TOU rates to 4 million customers in Ontario is “the biggest full-scale deployment anywhere in the world except Italy.”
Results validate Brattle’s research on TOU rates which, unlike dynamic pricing, have fixed prices for peak and off-peak time periods.
Recent research from the University of Waterloo concluded the Ontario TOU rates were not successful because they only reduced peak demand by 2.5% to 3%, Faruqui said. But the research failed to explain that “the small reduction was because the peak price to off-peak price ratio of 1.5-to-1 is a very mild differential.”
Brattle research found nothing inherent in TOU rates that limits peak period load reduction.
“Our data from 300 trials and pilots around the world shows the bigger the peak price to off-peak price ratio, the bigger the peak load reduction because the bigger the incentive, the stronger the customer response," Faruqui said.
Ontario’s load reduction is not a failure of its TOU rates but a product of the rate structure, he added. Ratios of 4-to-1 or 5-to-1 on very high demand days, when prices spike, would drive “significantly more peak load reduction,” Faruqui said. “But stakeholders rejected a change from more traditional ratemaking and the [Ontario Energy Board] accepted that.”
What can California learn from Ontario?
California can learn three lessons from Ontario's TOU rate experience, Faruqui said. First, 4 million customers can move to TOU rates without disruption. Second, the impact on peak demand is predictably proportional to the off-peak price to peak price ratio.
Third, TOU rates designed to be revenue neutral produce no overall reduction in energy consumption, he said. “If the peak rate is higher and the off-peak rate is lower, the average price is the same so there is no incentive to reduce consumption.”
The Ontario experience also shows TOU rates are likely to be inadequate to serve the power system of the future, Faruqui believes. “With both central station and distributed resources, static peak and off-peak TOU rates will be outdated because peak demand will be about net demand, not gross demand.”
Dynamic pricing will be needed to address new load uncertainty, he said. “As soon as demand starts spiking and the price starts rising, customers with enabling technology can respond to that price signal to help manage the demand spike.”
Dynamic pricing would more effectively address California’s growing concerns with over-generation and sharp late afternoon load ramps, he said. “TOU rates solve yesterday’s problem. The future is dynamic pricing enabled by smart technology that brings variable renewables and DER onto the grid at the right time and the right price.”
Faruqui’s prescription is a three-part rate. Dynamic pricing or another form of time-varying rate would be used for energy costs. Fixed and demand charges would address capacity costs.
Three-part rates have long been effectively used for commercial and industrial customers and can be equally effective for the more than half of U.S. residential customers that now have AMI, he said.
“Energy efficiency and solar advocates won’t like them because the volumetric portion of the rate will be lower. But why should the customer pay more for energy than it costs to produce?” Faruqui asked.
EDF’s Kairam sees TOU rates as today's answer to California’s concerns about renewables curtailment. The price signal will “optimize DER on the grid and avoid bad large asset bets,” he said.
Without TOU rates, California could curtail up to 12% of its renewable generation. With them, it can reduce curtailment 600%, according to Lawrence Berkeley National Laboratory research cited by Kairam.
California’s investor-owned utilities are preparing 2018 pilots that will lead to the full-scale 2019 rollout. Kairam is concerned with their structure.
Pacific Gas and Electric (PG&E) will test only one on-peak/off-peak rate and San Diego Gas and Electric (SDG&E) and Southern California Edison (SCE) will test only two, he said. And he is not sure the price differentials and proposed peak period lengths are adequate to motivate changes in usage, drive DER growth, and encourage investment in new technologies.
All three utilities told Utility Dive they are following commission-imposed guidelines.
Vote Solar Distributed Generation Regulatory Policy Program Director Rick Gilliam said TOU rates make sense as the next move away from current residential volumetric rates because they reflect utility costs. But the design can be challenging.
A bigger peak to off-peak price ratio will have a greater impact on peak load but, even with AMI in place, could disadvantage low and moderate income (LMI) customers with less electricity consumption flexibility, he said.
Dynamic pricing, he said, only makes sense “from an economist’s viewpoint.” Residential customers “in the real world lack the advance knowledge about their own use” to respond successfully, Gilliam said.
It is also important to distinguish between types of dynamic pricing, he added. Peak time rebate savings is a “carrot” approach that rewards customers for load reduction. Critical peak pricing (CPP) is a “stick” approach because customers pay more if they do not reduce usage.
Tomorrow’s dynamic pricing
Pepco Holdings Inc. (PHI) Delmarva Power subsidiaries have successfully deployed dynamic pricing rates, Faruqui said. And the Oklahoma Gas and Electric (OGE) Smart Hours program “is by far the most successful U.S. dynamic pricing rate.”
Following deployment of AMI in its Delaware and Maryland territories in 2012-13, PHI implemented the Peak Energy Savings Credit Program, according to Strategic Manager of Customer Relations Stephen Sunderhauf.
TOU rates are “old school” and PHI wanted a price structure that reflects current market conditions because “it sends a better price signal,” he said.
All eligible residential and small business customers are automatically enrolled but can choose not to participate, Sunderhauf said. Those who respond to day-ahead text, email, or phone notifications receive a $1.25/kWh rebate for load reductions during the 16 or so hours per year when demand response (DR) is needed.
The load reduction may be to meet utility needs or the utility may ask for reductions in response to a DR event called by PJM Interconnection, the regional system operator, Sunderhauf said. In that case, the utility may be paid for participation in PJM’s DR market.
Between 60% and 70% of PHI’s over 750,000 customers typically respond to the opportunity to earn the incentive for shifting their energy consumption, Sunderhauf added. “There are many reasons to participate but the incentive matters.”
TOU rate peak period prices are the same regardless of system conditions. The resulting general peak demand load reduction can be helpful, he said, “but it is nothing like the real peak load reductions that come from a dynamically altered price signal.”
Power market regulators have been reluctant to initiate change because dynamic pricing concepts are new and electricity is regarded as an essential commodity, Sunderhauf said. Some doubt the load reduction reliability and the price signal’s impact.
But, like food and gasoline, power prices can be responsive to wholesale market conditions, he argued. “And If it is an essential commodity, shouldn’t people have the price signals and incentives they need to make good decisions?”
Vote Solar’s Gilliam said PHI’s non-mandatory, incentive-backed program is a limited form of dynamic pricing.
“Dynamic pricing more typically penalizes the customer for not reducing load and does not offer a reward for reducing it,” he said.
Sunderhauf responded that “any pricing that tracks grid conditions dynamically is dynamic pricing.” Dynamic pricing may be “the future of rate design” because it can “evolve as customers become familiar with it”, he added.
“Market conditions, technology and policy will dictate much of what happens, but some form of dynamic pricing is likely to be around for a long time," Sunderhauf said.
National Consumer Law Center (NCLC) Senior Policy Analyst John Howat, an LMI customer advocate, has doubts about critical peak pricing, but does not reject PHI’s peak time rebate.
“TOU rates may be easier and more familiar to consumers but opt-in dynamic pricing with an incentive is much better than default dynamic pricing with a penalty," he said.
TOU rates may be easier and more familiar to consumers. But “decisions about the best, least cost means to achieve a policy objective can only be made on a case by case basis.”
OGE's dyanmic pricing
OGE’s opt-in Smart Hours program is closer to Gilliam’s definition of dynamic pricing.
Approximately 100,000 of OGE’s 625,000 residential and small business customers are enrolled, said Bryan Scott, director of price and load.
Participating customers move from standard rates of about $0.05/kWh, off-peak, and $0.20/kWH, on-peak, to peak period rates that can be $0.05/kWh, $0.10/kWh, $0.20/kWh, or $0.40/kWh, depending on the forecasted peak load.
Customers are notified of the next day's rate by text, email, or phone. “The customer can consume electricity and pay the higher price or not use power and avoid paying it,” said Mike Newcombe, products and services portfolio manager.
OGE went to the $0.40/kWh price “six or seven times last summer, but the vast majority of days had the $0.20/kWh price,” he said. “The average customer saved approximately $150 for the summer.”
Dynamic pricing has cut OGE’s average peak load of 5900 MW by 160 MW and reduced participating customers’ contribution to peak by 40%, according to Scott. The average contribution to peak load for program participants dropped from 4 kW to 2.5 kW.
Combined with energy efficiency and commercial-industrial peak reductions, OGE’s peak load is down approximately 300 MW, Scott said. That has allowed the utility to avoid new investment in thermal generation.
“The objective was to manage our peak load growth and we achieved that,” he added. “The program is not a silver bullet for all utility pricing problems but it has been effective for OGE, though recruitment is slowing.”
Newcombe said only about 2% of enrolled customers choose to leave the Smart Hours program and the satisfaction of people in the program “is higher than that of standard customers.”
But, NCLC's Howat said, “consumers are not clamoring for dynamic pricing. The push is coming from environmental advocates, utilities, or vendors. If policymakers move beyond TOU rates to dynamic pricing, they must be sure customers are protected.”