Time-of-Use and Dynamic Pricing: Promise vs. Reality

Time-of-use (TOU) pricing has been a fixture of commercial and industrial rate schedules for decades. The logic is straightforward: because the cost of generating and delivering electricity varies significantly by time of day and season, rates that signal those cost differences should theoretically induce customers to shift consumption away from expensive peak periods, reducing the need for peaking capacity investment and lowering overall system costs. What has taken longer is the translation of that logic into residential rate design—and the early evidence from mandatory and default residential TOU programs reveals a gap between theoretical promise and practical performance that rate professionals should understand clearly.

The Economic Case for Time-Differentiated Pricing

The economic case rests on a simple observation: a utility’s marginal cost of serving load varies enormously by time. On a summer afternoon when every available generating unit is running and the system is constrained, the marginal cost of an additional kilowatt-hour may be five to ten times higher than the marginal cost of serving load at 2 a.m. when base load plants are running at partial capacity. Flat volumetric rates that charge the same price at all hours send no signal about this cost variation and therefore provide no economic incentive to shift consumption when shifting would benefit the system most.

The capacity investment implications compound this problem. Utilities size their generation and transmission infrastructure to serve peak demand—a relatively small number of hours during the year when load reaches its maximum. If customers respond to peak price signals by reducing consumption during those hours, the utility can defer or avoid capacity investments that benefit all customers through lower revenue requirements.

What the Evidence Shows on Customer Response

The empirical record on residential TOU response is extensive but requires careful interpretation. Controlled pilots—where customers opt in to time-differentiated rates—consistently show meaningful demand response: peak period reductions of 10 to 15 percent are typical, with some programs achieving higher reductions among customers with enabling technology like smart thermostats or home energy management systems.

The picture becomes more complicated when utilities move from opt-in pilots to default or mandatory TOU enrollment. Several large-scale default TOU deployments have found substantially smaller demand response than pilots predicted—on the order of 3 to 5 percent peak reduction—with significant variance across customer segments. Lower-income customers, renters, and customers without enabling technology show materially lower response rates, raising both efficiency and equity concerns about broad default TOU enrollment.

The California experience is instructive. The state’s move to default TOU rates for residential customers beginning in 2019 produced smaller-than-projected peak demand reductions, prompting ongoing debate about optimal rate design, the role of enabling technology, and the appropriate balance between TOU benefits and customer bill impact variability. Customers on default TOU face higher bill volatility than those on flat rates—a feature that is economically rational but politically contentious and disproportionately burdensome for households with less flexibility to shift consumption.

Dynamic Pricing: Beyond TOU

True dynamic pricing—real-time pricing (RTP) and critical peak pricing (CPP)—takes time-differentiation further by varying prices based on actual system conditions rather than predetermined schedules. Under real-time pricing, retail rates reflect hourly wholesale market prices; under critical peak pricing, a sharply elevated rate applies during a limited number of declared critical peak events per year.

Critical peak pricing has a stronger empirical track record than broad TOU for achieving large demand reductions during the specific hours when system stress is highest. Because the elevated price applies only on declared event days—typically 10 to 15 times per year—customers can bear significant bill exposure on those days while maintaining predictable costs on the other 350 days of the year. The behavioral response to clearly signaled, high-stakes events is generally larger than the chronic low-level response to standard TOU differentials.

Real-time pricing has been more successful among large commercial and industrial customers, who have the metering infrastructure and operational flexibility to respond to hourly price signals, than among residential customers. The information processing burden of hourly price awareness without automated response technology is simply too high for most households.

The Enabling Technology Question

The performance gap between pilots and broad deployment points to a central issue: time-differentiated pricing works best when customers have automated tools that do the response work without requiring conscious daily decisions. Smart thermostats, smart water heaters, EV charging management systems, and home energy management platforms can shift load automatically in response to price signals—capturing the system benefits of TOU without burdening customers with constant attention to pricing periods.

The utility industry has not yet resolved how to handle the enabling technology gap equitably. Programs that are most effective for technology-enabled customers tend to be least accessible to lower-income households, renters, and older customers—creating a distributional concern that mirrors the net metering equity debate discussed in our related article on Net Metering and the Cost-Shift Debate.

Rate Design Implications

For rate departments, the evidence argues for differentiated TOU strategies rather than one-size-fits-all default enrollment. Rate designs that pair TOU periods with bill protection guarantees during the transition period, customer education campaigns, and optional enabling technology programs have shown higher customer acceptance and better demand response outcomes than mandatory enrollment without support structures.

The interaction between TOU pricing and distributed generation compensation also requires careful design. Net-metered solar customers benefit from TOU rates that pay higher export credits during peak periods when their generation is most valuable—but the same rate design can create adverse incentives if solar production does not align with the utility’s actual peak periods. Rate professionals designing TOU structures in high-DER environments need to model these interactions carefully rather than treating the rate elements independently.

The long-term trajectory of time-differentiated pricing is clearly toward greater granularity and real-time signal transmission as advanced metering infrastructure becomes universal and customer-side automation technology matures. The question for rate design is not whether to move in that direction but how to manage the transition in ways that preserve equity and customer acceptance while capturing the system efficiency benefits that make TOU pricing economically compelling in the first place.