When electricity demand spikes on a sweltering summer afternoon, grid operators have long reached for the same solution: fire up a gas peaker plant. It works, but it’s expensive and carbon-intensive — and it’s a habit that utilities and regulators are increasingly trying to break.
A growing number of U.S. states are now pursuing a different kind of backup power — one with no smokestacks, no turbines, and no fixed address. The scale of what some states are now committing to is worth paying attention to.
What exactly is a virtual power plant?
A virtual power plant is a network of distributed energy resources coordinated by a central controller — think home batteries, commercial battery storage, and industrial facilities that can dial down their electricity use on demand. None of these assets sit in one place, yet together they can behave like a single power plant, responding to grid signals within seconds. When the controller sends a signal, hundreds or thousands of nodes respond in unison, and resource owners receive compensation for their participation.
That makes VPPs significantly cheaper and cleaner than natural gas peaker plants for handling short-term demand spikes. The catch is definitional: “virtual power plant” still lacks a universally agreed meaning, which creates real ambiguity when policymakers try to write targets or track progress across programs.
Massachusetts sets an ambitious target
On March 13, Massachusetts Governor Maura Healey signed an executive order directing the state to develop 3.5 gigawatts of demand-management resources — including VPPs — by 2035. For context: the entire six-state New England grid recorded a peak demand of 26.1 GW in 2025. California’s largest VPP network, the biggest in the country, peaked at roughly 0.5 GW last July.
The order takes a broad view of what counts toward that target, encompassing virtual power plants, electric vehicle charging management, energy efficiency programs, and demand response initiatives. Some energy policy observers note that most items on that list are, in effect, forms of a VPP — which is part of why the explicit categorization matters.
A September report will inventory existing demand-response programs, establishing a baseline before the state pursues new ones. Larry Chretien of the Green Energy Consumers Alliance called the approach a promising start, though he added his organization is “always impatient.” Reaching 3.5 GW could reduce the need for expensive grid infrastructure and bring the retirement of peaker plants meaningfully closer.
Minnesota takes a different — and contested — approach
On May 13, Minnesota utility regulators approved Xcel Energy’s Capacity*Connect plan: 200 megawatts of neighborhood-scale batteries, each ranging from 1 to 3 megawatts, deployed across the grid. Unlike most VPP models, Xcel will own and operate every battery in the network.
The company argues that ownership is central to reliability. By controlling the assets directly, Xcel says it can place batteries precisely where the grid needs them, store energy when prices are low, and dispatch it when demand rises — maximizing benefits for all customers rather than just individual battery hosts.
Critics are not convinced. John Farrell of the Institute for Local Self-Reliance argues that utility ownership undermines cost efficiency and removes consumer control from the equation, and that decentralized, consumer-owned models create stronger incentives to keep costs down.
The program represents a notable departure from typical VPP structures, which are built around participating customers. Xcel’s own news release does not use the phrase “virtual power plant” — a reflection of both the program’s unusual design and the broader definitional ambiguity surrounding the term.
How big is the VPP landscape nationally?
A January 2025 report from Lawrence Berkeley National Laboratory inventoried approximately 180 VPP projects across the United States. California leads the field with 62 projects, followed by Colorado with 16 and Massachusetts with 15. The geographic spread signals that VPP development is no longer concentrated in a handful of early-adopter states.
Combined, these projects represent a potential capacity of 19 gigawatts — nearly three-quarters of New England’s peak demand. That figure suggests VPPs have moved well past the niche-experiment stage. Autumn Proudlove, managing director for policy and markets at NC State’s Clean Energy Technology Center, describes 2025 as a year of “steady uptick” in legislative and regulatory activity, with Massachusetts and Minnesota among the most significant examples of a broader pattern.
The road ahead: promise, tension, and open questions
The potential upside of scaling VPPs is substantial. Enough distributed capacity could reduce costly investments in transmission infrastructure and bring the retirement of fossil fuel peaker plants within reach — a goal that advocates like Chretien are explicit about pursuing.
Significant tensions remain. The debate over utility versus consumer ownership of distributed battery assets will shape whether these systems deliver broadly shared benefits or concentrate control — and cost savings — in fewer hands. Minnesota’s Capacity*Connect program will serve as an important test case on that question.
Definitional inconsistency is another persistent obstacle. Without a standard meaning for “virtual power plant,” comparing programs across states and setting coherent national policy both become harder than they need to be. Targets like Massachusetts’s 3.5 GW goal require program development and investment starting now, not in 2034. The states that move earliest — and most deliberately — on VPP frameworks will likely shape how the rest of the country approaches this transition.
Carlos is an engineer with strong expertise in technical and industrial topics. He previously worked at international companies such as Siemens and speaks Spanish, German, English, and Italian.
