States should begin changing their time-varied electricity pricing to better match real-world conditions.

Electric utilities offer or require time-varied rates, especially for commercial and large institutional customers.

Time-of-use rates are an economic signal to shift a customer’s power demand to happen when the power grid can most easily meet the need.

Traditionally, on-peak hours are afternoon and evening hours, like noon to 8:00 P.M. This is when homes run appliances and adjust their building thermals, while commercial and institutional buildings are also operating.

But thanks to distributed solar photovoltaic (PV) production, the peaks are meaningfully changing.

Distributed solar now reduces the power grid’s system-level demand curve between late morning and 4:00 P.M. On some days in New England, it’s actually pushed the real-time electricity market price below zero.

This means utility on-peak hours from noon to 8:00 P.M. don’t even crudely approximate the actual system peak.

The attached chart shows several recent days in New England when it’s been cold (so there’s a solid demand floor), but also sunny.

There are two peaks to most days, when power gets costly – early mornings when most people wake, and in the evenings when they return from work. Solar in the middle keeps power costs low.

Utilities should look into changing a customer’s economic incentive to use less power during these new peaks, which make the on-/off-peak distinction tricky to manage.

This raises questions:

  • Should each day have not one but two on-peak and off-peak windows each?
  • Should there be a single on-peak starting at 4:00 P.M. and ending between 7-9:00 P.M.?

The smaller the behavior adjustment window and required mindfulness period, the more easily customers can meet it.

Changing the windows also allows for a stronger price difference between on-/off-peak rates, for a greater economic effect.

This would accelerate the expansion of battery storage deployment, whose typical <4 hour duration is enough to store solar overproduction and discharge at sunset.

In the process, this would reduce system peak demand costs that are socialized across ratepayers, and also reduce the marginal emissions created in serving extra load.


In related news, a new Stanford study recognizes that most electric vehicle (EV) owners charge at home overnight, and in US western states it’s begun costing the power grid.

This, because of the mid-day overgeneration of solar PV, pressing the system demand curve below its overnight level.

As the transportation sector electrifies, it’s becoming cheaper for the system if electric fleets charge mid-day.

Encouraging employers and building owners to install EV charging will be a key strategy to address this.

Incentives in the rate structure can make a difference in how costly it is for a host to provide charging, which can influence their choice to offer it.

The study’s lead author notes: “All states may need to rethink electricity pricing structures as their EV charging needs increase and their grid changes.”

The attached graphic depicts a few days on the New England grid, where this issue isn’t nearly as severe as it’s becoming further west, where there is more solar irradiance and greater numbers of EVs.

In the western US, once 50% of vehicles are electric, the study suggests the difference in when drivers charge EVs represents roughly an extra nuke plant or an extra gigawatt of battery storage being needed.

Time-of-use ratemaking should get fresh scrutiny, for reasons of both supply and demand in the new energy economy.