Skip to content

Demand Response Pricing

Pricing policy — instantiates Rebound-Aware Efficiency Governance

Varies price continuously by time, load, or scarcity so responsive demand moves off the peaks efficiency would let it pile onto — reshaping when the resource is used rather than what it costs on average.

Demand Response Pricing makes the price move. Instead of one standing charge, it varies the price of a resource by time of day, live load, scarcity, or congestion, and updates it on a fast, rolling cadence — so flexible users face a high price exactly when the system is stressed and a low one when capacity sits idle. Its distinctive job in this archetype is temporal: efficiency tends to make a resource cheap and convenient enough that demand piles onto the peak, and a flat price cannot discourage that pile-up. A time-varying price reshapes when the service is used — shaving peaks, filling troughs — rather than trying to shrink the total by charging more on average. What sets it apart from a standing price incentive is precisely the cadence: the signal is continuous and responsive, not a lever pulled once and held.

Example

A public EV-charging network has a problem born of its own success. Efficient fast-chargers made topping up cheap, so nearly everyone plugs in at 6 p.m. on the way home; the evening peak overloads local substations and the operator is being pushed to fund a costly upgrade. Instead it prices the peak: charging is expensive from 4–9 p.m., cheap overnight, with a critical-peak surcharge on the handful of strained days each year. Drivers who can wait shift to the overnight trough; the same total energy flows, but the peak flattens and the substation upgrade is deferred. The scheme changed when the demand landed, not how much of it there was — and it did so by moving the price on a schedule that tracks the load, not the calendar.

How it works

The mechanism couples a price to a live system signal — a clock, a load reading, a scarcity index — and lets it swing. Responsive demand self-selects into the cheap windows; inflexible demand stays put and reveals itself. The essential design choice, and the source of all its leverage and all its risk, is the cadence and volatility of the signal: a price that updates every five minutes wrings the most peak-shifting out of flexible users but is hard to plan around, while coarse time-of-use blocks are predictable but leave value on the table. It manages demand's shape, never its ceiling.

Tuning parameters

  • Schedule granularity — flat time-of-use blocks up to real-time five-minute pricing. Finer granularity captures more peak-shift but demands more metering and more from the user.
  • Peak-to-off-peak ratio — how wide the price spread runs. A bigger spread moves more load but bites harder on those who can't move.
  • Update cadence and notice — how often the price changes and how much warning users get. Predictability raises participation; surprise raises backlash.
  • Surge ceiling — an optional cap on how high the scarcity price may spike, trading some peak-shifting for protection against extreme bills.
  • Enrolment default — opt-in versus opt-out onto the dynamic tariff, which largely decides how many respond at all.

When it helps, and when it misleads

Its strength is extracting more service from fixed capacity: by moving flexible demand in time it shaves the peak that efficiency inflates, defers capacity expansion, and prices scarcity honestly moment to moment. It is the right tool when the problem is when demand happens, not how much.

It misleads when the demand isn't actually flexible. Only responsive users shift; inelastic ones — the person with no home charger, the shift worker — simply pay the peak, which makes a naïve scheme regressive.[1] Complexity and unpredictable bills erode the very response the scheme depends on, and a real-time price can be gamed. The classic misuse is to deploy dynamic pricing as a disguised revenue raiser — lifting the average price under the banner of efficiency while little load actually shifts. The discipline is to measure load-shift, not revenue, as the success metric, and to shield inelastic essential users with a protected block or a lifeline rate.

How it implements the components

  • demand_guardrail — the time-varying price acts as a soft guardrail on the peak, steering responsive demand away from the stressed window rather than capping the total.
  • adaptation_cadence — its defining feature: the price re-sets on a fast, rolling schedule tied to live load, which is what lets it track scarcity as it moves.

It does not set the standing price level or any corrective fee — that is Price Incentive Adjustment — nor meter and bill per unit of use, which is Usage-Based Pricing. The heterogeneous response it relies on is estimated by Elasticity Experiment, and protection of those who cannot shift belongs to Essential-Access Rebound Review.

References

[1] Peak-load pricing — charging more when shared capacity is scarce and less when it is idle — is a long-standing result in utility regulation. Its efficiency case is strong; its equity case depends entirely on protecting users whose demand is not actually flexible. ↩