The Reindustrialization of America Is a Decade-Long Tailwind

Every cycle has noise. Underneath this one is a signal worth separating out: the forces reindustrializing America are structural, they compound each other, and they run for a decade or more.

Sourcing for an active project? Call 601.892.5017 or email collin.t@fabtekindustries.com — send a drawing or a scope and we’ll respond fast.

U.S. manufacturing construction reset structurally higher

U.S. manufacturing construction spending averaged roughly $82 billion a year in 2021. By 2024 it averaged $235.6 billion — nearly a threefold expansion, the fastest of any nonresidential sector in the modern data era. Spending has cooled from that peak as the first wave of semiconductor megafabs completed, running near $175–190 billion into 2026 — still more than double the pre-2022 baseline. The takeaway isn't "the boom ended." It's that the country's baseline rate of factory-building reset structurally higher.

The grid investment layer underneath

For the first time in two decades, U.S. electricity demand is genuinely growing — record consumption in 2025 and again in 2026, driven by data centers, industrial electrification, and onshored manufacturing. Serving it is a capital program with few precedents: the Edison Electric Institute projects investor-owned utilities will invest $238.8 billion in 2026, a 17% jump after a record 2025, and roughly $1.4 trillion through 2030. Globally, $5.8 trillion in grid investment is forecast for 2026–2035. Data center construction alone hit $84 billion in starts through July 2026 — nearly triple the prior year — and pulled power infrastructure construction from $9 billion in 2021 to $37 billion in 2025.

Why the reindustrialization tailwind doesn't mean-revert

• The grid is old regardless. Large power transformers average roughly 38 years in service; tens of millions of distribution units are past design life. Replacement demand exists independent of growth.
• Supply chains are being rebuilt deliberately. Transformer capacity, electrical steel, switchgear — after two decades offshore, domestic capacity is being rebuilt with support from both parties, because the alternative is multi-year import queues for grid-critical equipment (a risk we unpacked in Domestic Fabrication and Supply Chain Risk).
• The load is contracted. Hyperscale campuses, fabs, and battery plants announced today carry 2027–2032 energization dates. The demand is signed; the infrastructure to serve it isn't built yet.
• The constraint is physical. Factories, steel, transformers, and skilled people all take years to add — which is exactly why the backlog can't clear quickly even at record investment rates.

Positioning for a supply-constrained decade

FabTek made its bet on this thesis early — committing the business to the power grid and expanding from 50,000 square feet in 2024 toward 500,000 by 2030, reinvesting in capacity ahead of demand rather than behind it. Fabricated steel for substations, transformer components, and datacenter power distribution sits directly in the path of every dollar described above. America is rebuilding its industrial base; it will take a decade, and it will be supply-constrained the whole way. The companies that positioned early — on both sides of the buyer-supplier relationship — will compound the longest. If you're building for that decade, so are we.

Frequently asked questions