California’s new distributed energy resource law aims to turn the state’s vast fleet of customer-sited resources into formal utility capacity, moving virtual power plants out of the emergency pilot stage and into the core resource adequacy market.
Signed into law by outgoing Gov. Gavin Newsom in late September, SB913 directs regulators to enhance market integration pathways for aggregated DERs to qualify for resource adequacy. It addresses hourly capacity crediting, multiple devices behind one meter, and recognition of battery exports.
The bill’s “secret sauce,” explained Raghav Murali, who leads Tesla’s state energy policy work, is that it breaks DERs out of the emergency response “box,” and into utility planning. VPP programs have long been viewed as “industry handouts or nice-to-haves, and in turn are being labeled as pilot programs or demonstrations,” he added, speaking on a panel hosted by Advanced Energy United.
Jomo Thorne, PG&E’s director of load management strategy and policy, is cautiously optimistic that the new law will allow VPPs to scale beyond historical barriers. In the past, unrecognized resource adequacy value has been one barrier to the utility’s ability to scale VPPs, he told Latitude Media. To get past that, the framework for implementing SB913 will have to be measurable, operationally workable, aligned with existing programs, and avoid duplicate compensation.
“SB 913 itself does not automatically expand our existing virtual power plant programs,” he cautioned.
Measurement challenges
Historically, VPP participation has been based on an individual home or building’s utility meter, rather than individual devices, said Eric Lyon, policy lead for California and the West at Everyday Electric. That creates enrollment friction when a utility can’t pinpoint which device has capacity to spare, and complicates attribution when multiple providers operate behind a single meter.
SB913 authorizes multiple enrollment, including in separate programs, as long as the same load reduction or export isn’t counted twice.
That will really come down to the question of using device-level data to measure performance, Lyon said. For aggregators like Everyday Electric that infer HVAC consumption from compressor runtimes rather than dedicated sub-meters, for example, the question is whether regulators will accept algorithmic estimates as accurate enough to verify performance of individual devices and settle payments.
That said, the move to device-level data is not a foregone conclusion; regulators have a lot of discretion in how the bill’s measurement provisions are implemented. For example, the CPUC is only required to allow device-level data for participation “if the commission has determined that device-level telemetry provides sufficient and accurate measurements,” as compared with meter-level readings.
This distinction is essentially an “existential” issue for Everyday Electric, Lyon added: “I don’t think we can meaningfully reduce barriers to participation as required by this law, or allow multi-device enrollments without shifting to device-level data as the primary, if not the only source, for measuring and compensating DERs.”
For more on the new market potential for VPPs, listen to this live episode of the Open Circuit podcast, recording at Latitude Media’s Flex Summit:
Measurements will also underpin accreditation, meaning how much dependable capacity an aggregator can actually sell. Enrolled megawatts aren’t the same thing as a commitment to respond. AEU’s Brandon Garcia said regulators often assume that if total enrolled capacity doesn’t respond to a dispatch request, VPPs can’t hit their marks.
“We had to really educate and explain…if we say we’re going to get a gigawatt of power, and we only come up with 800, that’s actually not a bad outcome if the state is only asking for 500,” he explained. “Part of the program structure, part of the technology, the way that it’s built, is that it assumes those things.”
SB913 directs capacity determinations to be tech-neutral, performance-based, and weather normalized. But it also explicitly puts the onus on the aggregator to deliver the promised supply, in the specific hours it’s needed. That matters commercially, because an aggregation’s value will depend not just on its maximum response, but when that response is available.
Exports and the net metering question
A well-designed framework could unlock underused storage and bidirectional EVs by rewarding export capacity during peak demand hours, Thorne explained.
Regulators will have to establish how payments for that capacity works, including how they will interact with existing net metering and billing credits, to avoid paying twice for the same kilowatt-hour. For utility planners, Thorne added, that means distinguishing between baseline customer battery behavior and what a device can be expected to deliver in response to a dispatch event.
On the industry side, however, there’s concern that overly restrictive rules could leave useful battery capacity on the sidelines. Murali, at Tesla, said California has historically “over-indexed on this issue of double-counting and incrementality.” But he acknowledged that in order for programs to actually bring down electricity bills for ratepayers — which was one of the law’s central aims — it will be important to make sure devices aren’t paid twice. Now, he added, the industry should step up and take more of a role in proposing solutions.
Over the course of the next 18 months, these questions will play out in two key CPUC proceedings, explained Brian Turner, senior director at Advanced Energy United. Implementation will span CPUC demand response proceedings addressing telemetry and enrollment, resource adequacy proceedings addressing qualifying capacity methodologies, and CAISO’s work to update wholesale market participation rules.
The deadlines set in the bill don’t guarantee that things are ironed out in parallel, however. By June 30, 2028, the CPUC must make recommendations for CAISO about changes to its market rules — but wholesale market changes don’t need to be operational by that date.
That said, Thorne said PG&E expects full implementation in time for the 2029 resource adequacy operating year.
The industry’s trade-off
These laws were a long time coming.
Proponents of VPPs in California worked to pass similar legislation several times in the past, but they were vetoed. What’s different this time around, explained Brandon Garcia, Advanced Energy United’s California director, is that the affordability crisis had “reached a crescendo” and legislators were frustrated by the lack of progress on bringing down rates.
Garcia also pointed to the fact that SB913 comes several years after California’s contentious net metering fights, which centered on how much rooftop solar customers should receive for exporting electricity to the grid, and whether those payments shifted costs on to other customers. “A lot of that left a bad taste in certain legislators’ mouths,” he said, ”and I think the industry needed that time to grow a little bit more.”
For developers, the longer-term prize is an industry that’s less dependent on discrete program budgets and expiration dates, and can plan with more certainty. The industry was stymied by the question of future funding for the Demand Side Grid Support program, for example, which provides incentives for load reductions and backup generation during extreme grid events.
Market integration could provide a more sustainable framework for compensation, but Murali also said that it also changes expectations for the industry. “If we are providing firm-ish capacity that the utilities can call upon and rely upon, it is also incumbent on the industry to then demonstrate a higher level of accountability than we’ve historically advocated for in VPP programs,” he said.
