ABSTRACT

This paper reviews the international experience with real-time pricing of electricity for households. In the economics literature, real time pricing is regarded as the “best” form of pricing from the vantage point of maximizing economic efficiency. However, from a customer perspective, it has not found much traction in the U.S. or Canada. The widest deployment has been in Europe, where it has been offered by competitive retailers, but even there it is a relatively recent phenomenon. It is being tested by a network in Australia. In the paper, I compare the impact of real time pricing on peak demand with simpler forms of time-varying rates, such as critical-peak pricing and time-of-use rates.

WHAT IS REAL TIME PRICING?

In wholesale markets, electricity prices change from minute to minute in “real time,” giving rise to the term: real time pricing (“RTP”). Even where wholesale markets don’t exist, RTP can be defined by equating it with variations in the marginal cost of energy, which is sometimes measured by “system lambda” in production costing models. Often, real-time prices (“RTP”) or marginal costs also vary locationally, thus giving rise to the term, location-specific, marginal cost pricing.

When these wholesale prices are passed through to retail customers, they are also called RTP. Sometimes, they are simply called highly dynamic prices.

WHERE IS RTP BEING OFFERED?

In Canada, RTP was offered in two provinces, Alberta and Ontario. But it was difficult to implement RTP without smart meters, even if the load shape for the entire system or for the residential class was used. Once smart meters were introduced, it was technically possible for consumers to sign onto RTP. But it is hard to find any evidence that they did. In Ontario, the overwhelming majority preferred to buy electricity on the Regulated Price Plan (“RPP”), which was a three-period time-of-use (“TOU”) rate. Recently, a fourth period has been introduced which offers a much lower rate than the off-peak period. This is designed to meet the needs of electric vehicle (“EV”) drivers and those customers who work during the night shift. About 90 per cent of customers are on the standard TOU rate, 10 per cent are on the inclining block rate (which pre-dates the TOU rate), and less than 1 per cent are on four-period TOU rate.

In Alberta, the Regulated Rate Option (“RRO”) used to be the standard rate for residential customers who did not want to switch to a retail energy provider. In January 2025, it was replaced with the Rate of Last Resort (“RoLR”). The RRO was a flat rate that fluctuated from month to month based on conditions in the wholesale market. RoLR is a fixed rate from January 1, 2025, until December 31, 2026, and it can only go up or down by a maximum of 10 per cent at the end of each 2-year term. It is set every two years to avoid sudden spikes in pricing for customers.^[1]

In the U.S., RTP has only been offered to residential customers in one state: Illinois. Two investor-owned utilities are offering it: Commonwealth Edison (“ComEd”) and Ameren Illinois. In 2007, Commonwealth Edison began to offer RTP to residential customers and later, it was followed by Ameren Illinois.

The RTP signal only applies to energy sales. Transmission and distribution costs continue to be recovered through a traditional rate design.

In Europe, RTP is being offered in at least five countries: Denmark, Netherlands, Norway, Spain and the United Kingdom. In Denmark, Netherlands, Norway and the United Kingdom, it is offered by retail providers of electricity.^[2]

In Spain, it was the default tariff from October 2015 to December 2023. It was discontinued as the default tariff because customers complained.

In France, RTP used to be offered prior to the 2021–22 energy crisis. Currently, it is not being offered. In its place, retailers offer Critical Peak Pricing (“CPP”).

In Australia, a form of RTP called Dynamic Operating Envelops^[3] is being tested in a pilot called Project Edith.^[4] It is operated by Ausgrid^[5], a network that serves New South Wales, and Reposit Power. It is only offered to customers that have installed solar panels and applies to imports from and exports to the grid. Prices are set for every five-minute interval and can be positive or negative (rewards).

HOW MANY CUSTOMERS HAVE ADOPTED RTP?

In Illinois, less than 2 per cent of customers have signed up for RTP. That’s despite the findings of one study, which found that an overwhelming proportion of customers would have lower bills if they got on the RTP rate as compared to non-time-based rates.^[6] Specifically, the study concluded:

“Using 12 months of energy-use data from smart meters, anonymized by zip code, Environmental Defense Fund (“EDF”) and Citizens Utility Board (“CUB”) calculated what the 2016 electricity bills of 300,000 ComEd residential customers would have been under the Hourly Pricing program. The study found:

• 97 per cent of the households studied would have saved money, comprising total savings of $29.8 million.
• The average ComEd customer would have saved $86.63 for the year, or 13.2 per cent less than they paid under traditional billing.
• The top 5 per cent of savers would have cut their bills by an average of $104 a year, or 31 per cent.
• Of the customers who would have lost money (roughly 3 per cent of the sample), the median increase in bills was an estimated total of $6.23 for the year.
• There are no significant differences between the effects of real-time pricing on the bills of customers who have low-incomes and other customers.

“In sum, the vast majority of ComEd customers would have financially benefitted in 2016 from participating in the Hourly Pricing program.”

Despite this positive finding, why have only 1–2 per cent of customers signed up to receive service on RTP? No one has an answer to this question. Surveys of those customers who are on RTP show that most of them have figured out that there are some hours in the day that are more expensive and other hours are less expensive. They respond to RTP rates as if they were on a time-of-day (“TOD”) rate, which defeats the purpose of sending hourly pricing signals. In other words, the same results would have been obtained if the customer had been a much simpler rate design. Interestingly, relatively few customers in Illinois are even on a simple TOD rate.

In Denmark there are approximately 40 retailers of electricity. They provide more than 143 pricing products to customers including flat prices, spot prices, and spot prices with a cap. Some 70 per cent of the customers have chosen some form of RTP. But RTP only applies to the energy portion of the bill. Most distribution utilities offer TOD rates. Distribution costs account for a third of the customer bill while taxes account for a similar percentage.

The typical Danish bill has five elements:

• Cost of electric energy. Most customers get the hourly price defined by the day-ahead spot market, but consumers can also choose to get a fixed price or to get the hourly price with a price cap.
• Cost of distribution. These tariffs typically feature time-of-day variation. Typical tariffs have three levels depending on the season. Winter is the peak season. Here’s an example from Radius, which operates the distribution grid in the Copenhagen area.^[7]
  • Summer: Off-peak (9pm-6am) 12.15øre/kWh =1.9c/kWh, Normal (6am-7pm) 18.22øre/kWh=2.8c/kWh, Peak (5-9pm) 47.38øre/kWh=7.2c/kWh. The peak to off-peak price ratio is 3.9:1.
  • Winter: Off-peak (9pm-6am) 12.15øre/kWh =1.9c/kWh, Normal (6am-7pm) 36.45øre/kWh=6.5c/kWh, Peak (5-9pm) 109.34øre/kWh=16.8c/kWh. The peak to off-peak price ratio is 9:1.
  • On top of the volumetric charge, Radius has a fixed charge of 537kr=83$ per year.
• Cost of transmission. In 2023, it was a flat rate of 11.2øre/kWh.
• Taxes. There is a fiscal energy tax (69.7øre/kWh=10.7c/kWh in 2023, which is planned to decrease gradually to 56.10øre/kWh=8.6c/kWh by 2030) and a value-added tax (VAT)which amounts to (25% of the total bill including the tax).
• Subscription charge. Distribution companies add a charge for the meter (around $100/year), while retailers may also, in some cases, add a fixed charge depending on the customer’s tariff plan (typically around $4–6 /month).

Customers pay the bill to the retail company who then subsequently pays the Distribution System Operator and the Transmission System Operator.

In Norway, about 75 per cent of all customers are on RTP but the prices are not actively communicated to customers. There is no default price for electricity. Distribution utilities set prices for grid services and customers pay a bill for electricity and a bill for the grid, but some retailers combine the two prices into a single price.

In Spain, RTP began to be offered as the default tariff to customers who did not switch to a retailer from October 2015 onwards. Approximately half of the customers were on the default tariff. However, prompted by concerns about price volatility, RTP is no longer the default tariff. In January 1, 2024, RTP was replaced by a three-period TOD rate whose prices change daily, since they are indexed to the wholesale market prices.^[8]

In the United Kingdom, Octopus Energy^[9] offers RTP. Their product offering is called the Agile Octopus. It is called an “innovative beta smart tariff.” According to information on the company’s website, it helps “bring cheaper and greener power to all our customers but is directly impacted by wholesale market volatility.” Agile features half-hourly prices that can spike up to 100 p/kWh at any time, although on average a typical household in the winter of 2022/23 would have paid around 35 p/kWh average.^[10] In US currency, that would represent a peak price of $1.30/kWh, compared to an average tariff of 45.5 cents/kWh. The prices are set between 4 and 8 pm on the previous day and reflect wholesale market prices.

The beta smart tariff is being marketed to customers who are in a position to shift large amounts of their energy away from the peak periods by using smart technologies like solar and batteries, EVs and thermal energy storage. It features a price cap that ensures that prices won’t rise above 100 p/kWh, or US $1.38/kWh.

The figure below shows the type of price variation associated with the Agile tariff.^[11]

Figure 1: Real-time price variation in the UK^[12]

WHAT HAVE BEEN THE RESULTS?

In Illinois, a price elasticity of -0.05 has been estimated for RTP.

In Denmark, no studies have been published that quantify customer response to RTP. It’s also unclear how much money customers save through RTP. Many of them defaulted onto RTP when they switched to retail suppliers.

In Holland, according to ANWB Energie, customers on RTP have saved an average of more than 200 euros per year. For households with an EV, the savings have exceeded 1,000 euros per year.^[13]

The figure below shows the changes that occur in load shapes with RTP. It shows that consumers shift their power consumption away from the peak hours to the off-peak hours.

Figure 2: Load shifting in response to RTP in Holland^[14]

Customer bill savings are relative to what they would pay based on flat rates, not TOD rates. In Norway, which has an abundance of hydro power, there is some variation in hourly prices but not as much as one might see elsewhere, as seen in the figure below.^[15] The price variation is in line with variations between peak and off-peak TOD rates that exist in many countries. Thus, it is not surprising that hourly load does not vary much in response to hourly prices.

Figure 3: Price variation ($/MWh) and load (MW) in Norway^[16]

The chart above shows how load varies with prices, the left vertical axis represents price in $/MWh, the right vertical axis load in MW and the horizontal axis shows time across months.

Econometric studies have found really low-price elasticities with RTP in the -0.01 to -0.07 range. However, on the coldest days, there has been no price elasticity. Research has shown that the average household does not respond to RTP. But certain segments in the general population do respond to RTP, including those who check the hourly prices frequently on apps or those with electric vehicles.

A field experiment with peak-time rebates^[17] was carried out in Norway to measure customer response to the rebates, which were provided to customers on RTP.^[18] There were two sets of results, one for the 2-hour peak period and one for the 13-hour peak period, as shown below. The figure below shows the reduction in peak demand that took place as the ratio of peak to off-peak prices was dialed up.

Figure 4: The arc of price response in Norway (2 and 13 hours)^[19]

It’s useful to benchmark these results against those from other experiments and full-scale deployment. As of the time of writing, some 400 time-varying rates have been implemented across the globe and their impacts on peak demand have been reported. A meta-analysis of this data is contained in Arcturus.^[20] The meta-analysis yields six “arcs of price response” that (a) plot the percent reduction in peak demand against the ratio of peak to off-peak rates and (b) that differ based on whether (i) the relationship being measured is based only on the price signal or (ii) is paired with an enabling technology and (iii) whether it pertains to TOU, critical-peak pricing (“CPP”)or Peak Time Rebate (“PTR”).^[21]

In general, the higher the ratio between peak and off-peak prices, the higher the price response. However, the relationship between price response and the price ratio is not linear, it’s curvilinear. Price response rises with the price ratio but at a diminishing rate.

Enabling technology such as a smart thermostat boosts price response. Finally, price response also depends on the type of price signal being conveyed to the customer — i.e., it varies by TOU, CPP and PTR.

When compared with the meta-analysis in Arcturus shown above, the impacts from Norway are a lot lower, as shown in the figure below. The green dots come from analyzing the data from 400 deployments of time-varying rates (“TVRs”) across the globe in Figure 5.

Figure 5: The arcs of price response for various time-varying rates^[22]

Figure 6: The arc of price response Norway (2 and 13 hours) and Arcturus^[23]

In Spain, between 2015 and 2023, RTP was the default rate, as noted earlier. Roughly half of the households are on it but 77 per cent of them were not even aware of being on RTP. Most RTP customers didn’t know what price they were paying. The ratio of the highest to the lowest prices during the day is shown in the figure below. It does not exceed 2:1. Price variation across the sample period is shown in Figure 8.

Figure 7: Ratio between the highest and lowest price each day in Spain^[24]

Figure 8: Average daily prices over the same period in Spain (Euro/MWh)^[25]

Econometric analysis of load shape and price data for 4 million households has failed to measure any statistically significant value for the price elasticity of demand.^[26]

HOW MUCH ADDITIONAL PEAK LOAD REDUCTION STEMS FROM RTP VERSUS TOD PRICING?

The answer depends on how much RTP varies across the hours of the year and whether, during the peak hours, RTP values are higher than TOU and CPP prices. It’s likely RTP prices will be higher than TOU peak period prices since the latter are averaged over 600–1,000 hours during the peaking season. However, RTP prices and CPP prices might be quite similar in magnitude since CPP prices focus on the top 50–100 hours of the peaking season. Thus, one should not expect to get much incremental load response from RTP over and above a cost-based CPP rate.

To get the highest response from RTP, three things need to occur.

1. Customers need to be fully informed about the hourly prices, preferably via an app on their phone and on a web portal.
2. They must be educated about the benefits of RTP and be internally motivated to spend time checking the app.
3. They must learn to program their end use loads to automatically respond to prices, a process referred to as “getting prices-to-devices”. This can be done with smart thermostats and EV chargers but just because something can be done does not mean that it will be done. The same concepts apply to the CPP rate.

It is worth mentioning that Oklahoma Gas & Electric (“OGE”) in Oklahoma has implemented a more advanced concept of CPP known as variable-peak pricing or VPP. VPP should not be confused with virtual power plants.^[27] VPP features four levels of CPP, based on system conditions. OGE sends the prices directly to the customer’s thermostat, thereby implementing the variable prices-to-devices concept but much simpler than RTP. The customer can, if they wish, program the thermostat such that its temperature settings vary with the prices but it is not required to do so. In addition, it is worth noting that OGE does not control the customer’s thermostat.

The program is opt-in. It has signed up 10 per cent of the household population. Bill savings are substantial and so are the reductions in peak demand.

CPP sign-up rates are much lower. In California, estimated take-up rates are approximately 2% of household customers.^[28]

SHOULD UTILITIES OFFER CUSTOMERS A CHOICE OF RATE DESIGNS?

In general, it is a good idea to offer a choice of rates to customers.^[29] No two customers are alike. There are demographic and psychographic reasons for why some are happy with flat rates, some with time-of-day rates and some with dynamic pricing rates. Each rate being offered should be cost-based. When these rates are plotted in the risk-reward space, they create an efficient pricing frontier, as shown in Figure 9.

Figure 9: Risk-reward trade-offs along the efficient pricing frontier^[30]

The terms are defined in Table 1 below.

Table 1: Rate design options^[31]

CONCLUSIONS

RTP deployment has not made much progress in the US even though simpler forms of time-varying rates are gaining traction in much of the globe, primarily static time-of-use rates but also dynamic rate designs such as critical-peak pricing rates and peak-time rebates.

RTP deployment is proceeding at a much more rapid pace in Europe, primarily due to the presence of retail competition. However, at the time of writing, there is no empirical evidence to suggest that RTP is generating greater reductions in peak demand or in the amount of load shifted from peak to off-peak periods than much simpler forms of time-varying rates or that it is bringing about greater bill savings for customers. 

• * The author is an Economist-at-Large who has published widely on time-varying rates and related topics such as demand forecasting, demand-side management, load flexibility, EVs and DERs. He has advised more than a hundred clients located on six continents, published widely on the topic and has been featured in many newspaper commentaries. He has worked at a number of institutions, including the California Energy Commission, EPRI, Barakat & Chamberlin, Charles River Associates and The Brattle Group.

  ¹ Alberta Utilities Commission, “Electricity rates: The AUC ensures regulated customers receive safe and reliable service at just and reasonable rates” (last visited 14 July 2025), online: <auc.ab.ca/current-electricity-rates-and-terms-and-conditions>.

  ² Nordic Energy Research, “Evaluation of Nordic Electricity Retail Markets” (12 April 2024), online: <nordicenergy.org/publications/evaluation-of-nordic-electricity-retail-markets>.

• ³ Enea, “Project Edith: Project Overview Report” (July 2022), online (pdf): <ausgrid.com.au/-/media/Documents/Reports-and-Research/Project-Edith/Project-Edith-2022.pdf?rev=42030a3921274632910a9fbf6ff1e2ac>.

• ⁴ Renew Economy, “Project that calmed network fears about rooftop solar wins innovation award” (26 September 2023), online: <reneweconomy.com.au/project-that-calmed-network-fears-about-rooftop-solar-wins-innovation-award>.

• ⁵ Ausgrid, “Making electricity accessible for all” (last visited 14 July 2025), online: <ausgrid.com.au>.

• ⁶ Environmental Defense Fund, “Data reveals real-time electricity pricing would help nearly all ComEd customers save money” (14 November 2017) online (blog): <blogs.edf.org/energyexchange/2017/11/14/data-reveals-real-time-electricity-pricing-would-help-nearly-all-comed-customers-save-money>.

• ⁷ Radius, “Tariffs and network subscriptions” (14 July 2025), online: <radiuselnet.dk/elnetkunder/tariffer-og-netabonnement>. One Danish Krone equals 0.16 US dollars.

• ⁸ Red Eléctrica, “Voluntary price for the small consumer (PVPC)” (last visited 14 July 2025), online: <redeia.com/es/actividades/operacion-del-sistema-electrico/precio-voluntario-pequeno-consumidor-pvpc#:~:text=Es%20el%20nuevo%20sistema%20de%20fijación%20del%20precio,a%20la%20anterior%20Tarifa%20de%20Último%20Recurso%20%28TUR%29>.

• ⁹ Octopusenergy, “Join the UK’s most popular energy supplier: Enter your postcode to get a quote” (last visited 17 July 2025), online: <octopus.energy>.

• ¹⁰ Octopusenergy, “The 100% green electricity tariff with Plunge Pricing” (last visited 14 July 2025), online: <octopus.energy/smart/agile>.

• ¹¹ Octopusenergy, “Dashboard” (last visited 17 July 2025), online: <agile.octopushome.net/dashboard>.

• ¹² Ibid.

• ¹³ Solar Storage Mazazine, “ANWB: more conscious use of energy through dynamic contracts with hourly prices” (26 October 2023), online: <solarmagazine.nl/nieuws-zonne-energie/i35666/anwb-bewuster-gebruik-energie-door-dynamische-contracten-met-uurprijzen?trk=article-ssr-frontend-pulse_publishing-image-block>.

• ¹⁴ Ibid.

• ¹⁵ Euenergy, “Electricity prices no Norway” (last visited 17 July 2025), online: <euenergy.live/country.php?a2=NO1> .

• ¹⁶ Ibid.

• ¹⁷ Peak-time rebates (PTR) have been successfully deployed in Maryland by BGE, Delmarva Power and Pepco. All customers are defaulted onto PTR. More than 80 per cent of the customers actively engage with PTR and reduce their demand during critical system hours and are compensated at $1.25/kWh.

• ¹⁸ Matthias Hofmann & Karen Byskov Lindberg, Evidence of households’ demand flexibility in response to variable electricity prices – Results from a comprehensive field experiment in Norway, Energy Policy, 2024.

• ¹⁹ Ahmad Faruqui, “Flexible Demand in Norway: A conversation with Matthias Hofmann” (last visited 26 July 2025), online: <energycentral.com/energy-management/post/flexible-demand-norway-conversation-matthias-hofmann-ucF8AEYtaQpYfVS>.

• ²⁰ Sanem Sergici et al, “DO Customers Respond to Time-Varying Rates: A Preview of Arcturus 3.0” (2023) Brattle, online (pdf): <brattle.com/wp-content/uploads/2023/02/Do-Customers-Respond-to-Time-Varying-Rates-A-Preview-of-Arcturus-3.0.pdf>.

• ²¹ Customers who reduce their load during critical time periods are offered a rebate under a PTR program, as opposed to being exposed to a higher price under a CPP rate.

• ²² Supra note 20 at 9.

• ²³ Supra note 21.

• ²⁴ Natalia Fabra et al, “Estimating the Elasticity to Real Time Pricing: Evidence from the Spanish Electricity Market” (2021) 111:1 AEA Papers and Proceedings 425; Natalia Fabra, “Real Time Pricing for Everyone: Evidence from the Spanish Electricity Market” (2019), online (pdf): <mreguant.github.io/em-course/materials/day4/slides_rtp.pdf>.

• ²⁵ Ibid.

• ²⁶ Natalia Fabra et al, “Estimating the Elasticity to Real Time Pricing: Evidence from the Spanish Electricity Market” (2021) 111:1 AEA Papers and Proceedings 425; Natalia Fabra, “Real_Time Pricing for Everyone: Evidence from the Spanish Electricity Market” (2019), online (pdf): <mreguant.github.io/em-course/materials/day4/slides_rtp.pdf>.

• ²⁷ OGE Energy Corp., “Pricing Options” (2025) online: <oge.com/wps/portal/ord/residential/pricing-options/smart-hours>.

• ²⁸ Based on conversations with staff at one of the three investor-owned utilities.

• ²⁹ Ahmad Faruqui & Cecile Bourbonnais, “The Tariffs of Tomorrow: Innovations in Rate Designs” Institute of Electrical and Electronics Engineers, online: <ieeexplore.ieee.org/document/9069846>.

• ³⁰ Author’s conceptualization.

• ³¹ Author’s conceptualization.