Modernizing Cost Allocation to Keep the Grid Affordable

Imagine if your annual income tax liability was based on your single largest paycheck instead of your total income. Those who earn commissions or employees who earn annual bonuses would owe far more than someone with the same annual income but who earned it from a steady weekly paycheck. And we would logically expect companies and employees to seek out compensation structures as a reaction to this illogical tax structure.

In most states, that is not far from how the costs of the power grid are allocated. Many traditional methods divide generation costs among residential, commercial, and industrial customers based largely on how much electricity each group uses during a tiny number of peak hours. Residential customers have more variable demand than commercial and industrial customers. This means that residential customer rates are set to recover a larger share of generation costs than annual residential energy use would suggest.

On average, residential electricity rates across the United States are, in fact, higher than those of commercial customers, as shown in the exhibit below. According to US Energy Information Administration, rates for residential customers were already nearly twice as high as those for industrial customers in 2019. And while rates for all customer classes have increased in the subsequent years, families continue to pay nearly twice as much as industrial customers.  

Some of that gap reflects real differences in the cost of serving each customer class: residential customers are generally smaller, more dispersed, and often more reliant on local distribution infrastructure. But current allocation methods widen the gap because they do not fully account for how each class uses the system over time.

Cost allocation is the step in a ratemaking process where regulators decide how a utility’s approved costs are divided among customer classes. And getting it right is even more urgent now.

Electricity affordability is becoming a front-line issue for regulators, consumer advocates, and utilities. In the first half of 2026 alone, states took 362 separate actions on energy affordability, most commonly to prevent large-load cost shifts or to help lower-income customers. At the same time, the grid is changing quickly. Data centers and other large customers are driving new demand in some states, while wind, solar, storage, and demand flexibility are changing when electricity is produced and when the system is most stressed.

While there is broad support for concepts like the Ratepayer Protection Pledge, it is unlikely that these commitments to prevent large-load cost shifts can be met using many existing cost allocation practices. When the same few peak hours are used to allocate costs across customer classes, they can obscure important differences between customers with variable demand and large load customers who use electricity consistently throughout the year, potentially assigning costs in ways that do not fully reflect how each class uses the system over time.

How can modernizing cost allocation methods help with affordability?

Cost allocation happens after regulators decide how much revenue the utility may collect and before they design the rates customers ultimately pay. It usually involves many steps, and its complexity makes it an almost invisible, yet highly consequential, driver of electricity costs affecting regular ratepayers.

Most regulator-approved cost allocation methods assign the cost of building power plants and transmission lines based on a tiny number of peak hours — between 1 and 12. That may have been a reasonable shortcut when the grid was dominated by large fossil fuel plants and planning was focused primarily on meeting a predictable peak. It is a weaker fit for a grid where different resources provide value at different times, and where large customers may have the capability to use batteries or backup diesel- or gas-powered generation to avoid power use during those hours.

Time-based methods can help by using more of the data utilities already have. Instead of asking only who contributed to a few peak hours, regulators can examine when the system is stressed, when specific types of generation assets are used, and which customer classes rely on those power supplies across the year. This approach can help make cost responsibility more transparent and better tied to system use.

Duke Energy’s 2026 North Carolina rate cases show what that shift can be worth. Under the utility’s conventional cost allocation method, residential customers are assigned 48% of generation costs. Analyses for Duke Energy Carolinas using hourly generation and load data for residential customers reduced the allocation to 38%, enough to reduce the entire residential bill by 6%. This may seem small, but it is enough to offset the average US residential rate increases from 2024 to 2025.

What are regulators doing about it?

Regulators are already beginning to examine whether older, peak-focused production cost allocation methods still reflect how today’s grid is used and who drives costs. In several states, commissions and stakeholders are moving from general affordability concerns toward more concrete questions on cost allocation.

• Iowa shows that hourly methods can be implemented and sustained over time. In the MidAmerican case, the Iowa Utilities Board accepted a move away from a traditional approach after finding that the hourly method better balanced energy and demand components of generation cost allocation. According to MidAmerican, the hourly method adopted in 2014 now results in residential customer savings of 20%–22% relative to the alternative peak-focused allocator. The Commission has retained this method through several subsequent proceedings to this day.
• Colorado has moved in the same direction. In 2024, the Colorado Public Utilities Commission approved a modified method for Xcel Colorado based on class usage during the top 1,000 hours. The Commission described it as a measured step toward a more dynamic understanding of cost causation, drawing on a broader set of peak hours that can adapt as electrification shifts when peaks occur.
• Wisconsin offers a lighter touch that any commission can take. In recent large load proceedings, the Wisconsin regulator required utilities to produce cost-of-service analyses based on hourly unit dispatch and class load curves, building the record for comparing methods before committing to a new one.

Together, these examples point to a practical near-term path: Regulators do not need perfect information to begin modernizing cost allocation. Instead, they can evaluate time-based production cost allocation methods now, adopt them where they improve fairness, and require utilities to collect the data needed to compare methods where the record is not yet sufficient. In relevant rate cases and large-load tariff proceedings, commissions can require side-by-side analysis of current methods and time-based alternatives. Where utilities lack the necessary hourly load, generation, or dispatch data, commissions can require a clear data collection plan over the next two to three years.

As the grid changes, affordability debates cannot focus only on how much utilities spend. They also need to ask whether the cost allocation method still fits the system utilities are building today and tomorrow. Modernizing cost allocation is one practical way regulators can bring rate design closer to today’s grid realities and better protect customers as new load and new resources reshape the power system.