Why Reindustrialization Requires More Electricity

Why Reindustrialization Requires More Electricity

LEARN / REINDUSTRIALIZATION

Why Reindustrialization Requires More Electricity

Factories need power that is available continuously, at a location, on a date. Each of those three conditions is now constrained, and the administrative queue that governs grid connection has become a longer lead time than building the plant itself.

Key points

Industrial load is not commercial load

An office building draws power on a schedule that follows daylight and the working week. A production line draws power whenever it is running, and the economics of a capital-intensive plant push toward running it continuously. The distinction matters because the grid is planned around peaks and around the resources available to meet them, and a load that does not vary is a load that must be served in every hour including the ones when generation is scarcest.

Firmness therefore matters more to a factory than annual energy does. A plant that loses power for four hours does not lose four hours of output. It loses the batch in process, the thermal state of its furnaces, and whatever restart sequence the equipment requires, which for continuous processes can run considerably longer than the outage that caused it.

A further constraint sits underneath the electricity question entirely. Cement, glass, primary metals, and much of chemicals require process heat at temperatures that electrification does not reach economically with current equipment. For those industries the energy question is not how the electricity is generated but whether the process can use electricity at all.

The demand forecast, and its track record

The North American Electric Reliability Corporation publishes an annual ten-year assessment built from planning data submitted by system operators. Its demand forecasts are the closest thing the sector has to an official consensus, and the useful way to read them is in sequence rather than one at a time.

Data55 → 80 → 132 → 224 GW
Ten-year summer peak demand growth, as forecast in NERC’s Long-Term Reliability Assessments for 2022, 2023, 2024 and 2025. The 2025 assessment projects 224 GW, a 69 percent increase over the 132 GW projected a year earlier and roughly a 24 percent rise over 2025 peak demand. Source: NERC, 2025 Long-Term Reliability Assessment, published January 29, 2026.

Series and provenance
Agency
North American Electric Reliability Corporation
Program
Long-Term Reliability Assessment
Series
2025 LTRA, published January 2026
Measure
Ten-year summer peak demand growth forecast
Units
Gigawatts of peak demand growth over ten years
Adjustment
Committed projects only
Period
Assessments of 2022, 2023, 2024 and 2025
Latest
224 GW in the 2025 assessment, against 132 GW in 2024, 80 GW in 2023 and 55 GW in 2022
Source tier
Primary
Retrieved
August 9, 2026

NERC counts a large load only once a project advances from speculative and exploratory stages into development commitments, so the series lags real but uncommitted load.

Open the source

Four successive assessments, each built by the same institution using the same method, quadrupled the projection. Winter growth followed the same path, rising to 245 GW against 149 GW the year before. NERC attributes most of the increase to data centers, and its own report notes that thirteen of twenty-three assessment areas now face elevated or high resource adequacy risk, with planned resource additions failing to keep pace.

Measurement noteNERC counts a large load in its forecast only once the project has advanced from speculative and exploratory stages into development commitments. The figures therefore describe committed load rather than announced intentions, which makes the successive upward revisions harder to dismiss as enthusiasm. It also means the forecasts lag: load that is real but not yet committed does not appear until a later assessment.

Supply moved the other way over the same period. Existing capacity from fossil-fueled generators fell by 21 GW between 2024 and 2025, while battery, wind and solar contributed 23 GW of additional capacity at peak hours. Confirmed and announced retirements over the coming decade total more than 105 GW of peak seasonal capacity.

The interconnection queue

A generator cannot connect to the transmission system until the operator has studied the effect on the network and assigned the cost of any upgrades required. The list of projects awaiting those studies is the interconnection queue, and Lawrence Berkeley National Laboratory compiles it annually across the seven ISOs and RTOs and roughly fifty non-ISO balancing areas, covering about 98 percent of installed American generating capacity.

Data2,060 GW · 5 years · 13%
Capacity actively seeking grid connection at the end of 2025, roughly twice the installed U.S. generating fleet. Median duration from interconnection request to commercial operation exceeded five years for projects built in 2025. Only about 13 percent of capacity that has submitted a request historically reaches operation. Source: Lawrence Berkeley National Laboratory, Queued Up, 2026 edition.

Series and provenance
Agency
Lawrence Berkeley National Laboratory
Program
Queued Up, Electricity Markets and Policy
Series
2026 edition, data through end of 2025
Measure
Capacity seeking transmission interconnection
Units
Gigawatts; years; percent completion
Adjustment
Active requests, duplicates removed
Period
Requests through December 2025
Latest
2,060 GW active; median over 5 years from request to operation for projects built in 2025; about 13 percent historical completion
Source tier
Synthesis, compiled from ISO, RTO and utility data covering about 98 percent of U.S. capacity
Retrieved
August 9, 2026

Covers generation requests seeking transmission connection only. Excludes load interconnection requests, distribution-connected projects and behind-the-meter projects. A factory connecting as new load appears nowhere in these figures.

Open the source

The duration figure is the one that binds industrial siting. Median time from request to operation has risen from under two years for projects built between 2000 and 2007 to over five years for those built in 2025. A firm deciding where to put a plant is choosing among locations whose power availability depends on a process that now takes longer than the plant’s own design and construction.

The completion rate compounds the problem. If roughly seven of every eight gigawatts in the queue never reaches operation, then queue volume is a poor guide to future supply, and a developer reading 2,060 GW as a pipeline is reading a list of applications rather than a schedule of deliveries.

Measurement noteBerkeley Lab states that its dataset includes only generation requests seeking transmission connection. It excludes load interconnection requests, distribution-connected projects, and behind-the-meter projects. A factory or data center applying to connect as new load does not appear anywhere in the 2,060 GW figure. Commentary that cites the queue as a measure of how long it takes to power a new industrial facility is citing a series that does not contain industrial facilities.

Institute analysisLoad interconnection is the less-measured half of the constraint and the half that determines industrial siting. Generation queues are compiled nationally, published annually and studied closely. Load queues are handled utility by utility, disclosed inconsistently, and largely absent from public analysis. The result is that the constraint most directly binding on new manufacturing is the one for which the least evidence exists.

Long-lead equipment

Approval is not energization. Connecting a large load requires transformers, switchgear, breakers and conductor, and the largest of those items are built to order by a small number of manufacturers. Where the equipment lead time exceeds the study timeline, clearing the administrative constraint moves the binding constraint to the supply chain rather than removing it, and the project date does not move.

NERC’s own assessment names supply chain alongside interconnection and approval delays as a threat to infrastructure growth, which is the same observation from the reliability side of the problem.

Common misconceptions

That generating capacity is the constraint. Capacity that cannot be connected does not serve load. With more than 2,000 GW awaiting study and a five-year median to operation, the network and the process governing it bind before the generation fleet does.

That the queue measures a pipeline. With a historical completion rate near 13 percent, the queue is a list of applications. Treating its volume as forthcoming supply overstates by roughly a factor of eight.

That electrification resolves industrial energy demand. High-temperature process heat in cement, glass, primary metals and much of chemicals cannot presently be supplied economically by electricity, so for those industries the constraint is thermal rather than electrical.

What the evidence says, and where it is contested

ContestedWhether the current demand forecasts will be realized is genuinely disputed. One position notes that utility load forecasts have overshot before, that speculative data center announcements can be submitted to several utilities at once and counted more than once, and that a portion of the projected load will never materialize. A second notes that NERC counts only committed projects, that the revisions have been consistently upward across four assessments, and that a forecast repeatedly revised up is more likely to be lagging reality than leading it. Both positions are describing real features of the data, and the disagreement will be settled by which loads actually energize rather than by further analysis of the forecasts.

Where each of these is developed

This entry states the energy constraint from the manufacturing side. The Institute’s Energy & Power subject takes each claim further.

What Is the Electric Grid?
Why a siting question is never about “the grid” but about a specific balancing authority.

Firm Power and Why It Matters
How accredited capacity is calculated, and why no standard method exists.

Why Electricity Demand Is Growing Again
The forecast revisions in full, and the one utility that measured the gap between requested and committed load.

The Interconnection Queue
Why about one part in eight of queued capacity is ever built.

Transmission and Why It Is Not Built
Construction down to a fifth of its rate, and why cost allocation is the obstacle.

Long-Lead Equipment and Transformers
The equipment constraint quantified, and the single domestic steel producer beneath it.

Related Institute research

Powering the Buildout: Energy Demand Projections for Data Centers, Reshored Manufacturing, and Defense Production
The energy constraint, quantified across three sources of new load.

The Conversion Problem: A 50-Year Decision Being Made by Default
How thermal generation converts heat to power, and why the cycle choice is being made without examination.

What Is Reindustrialization?
Why capacity is a system in which the scarcest input sets the ceiling.

Sources

Reference entry maintained by the Institute for American Manufacturing & Technology. Figures are drawn from primary sources and cited above. Where the Institute states a position rather than a finding, it is marked as such.

Private Professional Network

Institute for American Manufacturing & Technology

Join the IAMT Network. Connect with researchers, policymakers, and industry leaders working on American manufacturing, energy, and technology policy.

Apply for Membership →