For the past two years, a tidy story has taken hold in electricity policy circles.
Power prices are rising. Data centers are booming. Manufacturing is reshoring. Therefore, the thinking goes, large new electricity loads must be driving up costs for everyone else.
It’s an elegant narrative. It offers a clear villain. It flattens a complex system into something that feels intuitive. And it has justified a growing list of bad ideas: moratoria on new connections, punitive tariffs aimed at large customers, and a creeping instinct to slow demand rather than expand supply.
There is only one problem.
The data don’t support it.
A new, comprehensive analysis from Lawrence Berkeley National Laboratory and The Brattle Group provides the clearest empirical look yet at what has actually driven retail electricity price changes in the United States over the past five years. Its conclusion should force a rethink of how we talk about affordability, load growth, and the future of the grid.
In short: large loads are not the primary driver of recent electricity price increases. In many cases, they have done the opposite. They have lowered prices.
What the Study Actually Looks At
The study focuses on national and state-level retail electricity prices from 2019 through 2024. Not forecasts. Not modeling scenarios. Actual outcomes.
Using public utility data, state statistics, a broad literature review, and regression analysis, the authors evaluate a wide range of possible drivers: fuel prices, transmission and distribution spending, renewable mandates, net-metered solar, extreme weather, and changes in electricity demand.
This matters because electricity prices are not set by a single lever. They are the product of layered systems: capital recovery rules, fuel exposure, regulatory decisions, cost allocation fights, and political trade-offs. If you want to understand why prices rise, you have to examine the whole machine, not just one moving part.
That is what this study does.
The Finding No One Wants to Lead With
Among the report’s twelve core findings, one stands out because it cuts directly against today’s dominant narrative:
State-level load growth has tended to depress retail electricity prices in recent years, by spreading fixed costs over greater demand.
That result holds even after controlling for fuel price exposure, renewable policies, and behind-the-meter solar. In the statistical analysis, a 10 percent increase in statewide electricity load was associated with roughly a 0.6 cent-per-kilowatt-hour reduction in average retail prices.
This is not a theoretical claim. It shows up clearly in the data.
States that experienced strong load growth from 2019 to 2024 generally saw inflation-adjusted electricity prices fall. States with stagnant or declining demand often saw prices rise.
North Dakota offers the cleanest illustration. Rapid commercial and industrial growth, including energy production, data processing, and agricultural loads, coincided with declining real electricity prices. Fixed system costs were spread across more megawatt-hours. Customers benefited.
This is basic utility economics. Yet it has been conspicuously absent from recent policy debates.
Why the Intuition Breaks Down
The intuition behind the “large loads raise prices” claim feels reasonable at first glance. More demand requires more infrastructure. Infrastructure costs money. Someone has to pay.
But that intuition collapses once you account for how electric systems actually recover costs.
Most utility costs are fixed. Poles, wires, substations, control systems, and much of generation capacity must be paid for whether electricity use is high or low. When load grows faster than capital spending, average costs fall. When load stagnates or shrinks, those same costs are concentrated over fewer kilowatt-hours, driving prices up.
This is why the study finds that recent price increases were not primarily caused by load growth. Over the past five years, national demand growth was modest. Meanwhile, costs surged elsewhere.
What Really Drove Prices Higher
If large loads weren’t the culprit, what was?
The study is strikingly consistent on this point. The biggest contributors to recent retail price increases were:
- Transmission and distribution investment, driven by aging infrastructure, supply-chain inflation, and resilience upgrades.
- Natural gas price volatility, particularly during and after the Russia-Ukraine war.
- Extreme weather and wildfire mitigation, especially in California and along the Gulf Coast.
- Policy-driven costs, including renewable portfolio standards and certain net-metering structures.
In California, wildfire mitigation and liability costs now make up a significant share of utility revenue requirements. In New England, pipeline constraints magnified gas price spikes. In Florida, storm hardening and recovery dominated residential price increases.
None of these trends were caused by data centers. None were caused by reshoring factories. None were caused by large customers signing long-term power contracts.
If anything, the study finds that commercial customers often benefit the most from load growth, which helps explain why residential prices have risen faster than commercial and industrial rates. That distributional issue is real and worth addressing. But it is a rate-design problem, not an argument for suppressing demand.
How the Narrative Goes Off the Rails
The danger of the prevailing narrative is not merely that it is wrong. It is that it pushes policy in exactly the wrong direction.
If you believe load growth inherently raises prices, the logical response is to slow it down. Delay interconnections. Penalize large customers. Require them to self-supply or pay special surcharges. Treat electricity demand as something to be rationed.
But if load growth can lower prices when supply is adequate and planning is competent, then suppressing demand becomes self-defeating. You lock in higher average costs. You strand existing infrastructure. You make affordability worse, not better.
The study is explicit on this point. Load growth does not always reduce prices. In constrained systems, where new generation and delivery infrastructure cannot be built in time, growth can increase costs in the short run. PJM’s recent capacity auction is a clear example.
But that is not an argument against load growth. It is an argument against under-building.
The Question We Should Be Asking
The real policy question is not whether large loads should exist. They already do. And they will continue to grow, driven by data, electrification, and industrial demand.
The question is whether the system will be allowed to respond.
Will generation, transmission, and distribution be permitted and built fast enough to keep pace? Will planning reflect long-run demand rather than assuming stasis? Will cost allocation track actual causation instead of political convenience?
The Berkeley-Brattle study does not prescribe answers. But it decisively dismantles the idea that affordability requires restraining demand. In many cases, the opposite is true.
Capacity Is the Constraint
The lesson here aligns with a simple reality we keep avoiding: capacity matters.
Electricity does not become expensive because people use too much of it. It becomes expensive when systems are designed around scarcity, delay, and the avoidance of hard decisions.
Large loads are not a threat. They are signals. They tell us where value is being created and where infrastructure should follow. Treating them as villains rather than opportunities guarantees worse outcomes.
This study does not romanticize growth. It restores a basic truth to the conversation. When systems are allowed to expand, scale works in favor of consumers. When they aren’t, everyone pays more.
The numbers are now on the table. It’s time the policy caught up.
