Market Data

Two-Speed Inflation: Goods and Labor Have Quietly Decoupled

The textbook fear is a wage-price spiral, goods and labor costs chasing each other upward. The data shows the opposite: goods inflation and labor inflation have decoupled, running at very different speeds. That changes both the macro story and the plant-level decision between capital and labor.

The inflation narrative most people carry is the wage-price spiral: workers demand higher wages to cover rising prices, firms raise prices to cover rising wages, and the two chase each other up. It is a real phenomenon in some cycles. It is not what this data shows. Manufacturing materials are up about 30.9% over the past year, while wages are up about 4.2% and energy is roughly -1.9%. Goods inflation and labor inflation are running at completely different speeds. They have decoupled, and the gap changes both the macro read and the decision a plant makes on the floor.

Why the decoupling matters for the macro story

A wage-price spiral is self-reinforcing and hard to stop, which is why central banks fear it. Decoupled inflation, where goods run hot on their own drivers while labor stays contained, is a different animal: it is driven by specific supply-and-demand stories in individual commodities, electrification for copper, feedstock for resins, rather than by a general labor-cost dynamic. That distinction matters because the two call for different responses. A spiral demands broad demand restraint; a goods shock concentrated in a few materials calls for supply-side patience and targeted hedging, not a sledgehammer to the whole economy.

The plant-level consequence: capital versus labor

Decoupling reaches all the way to the floor through a decision every plant makes: whether to solve a problem with a machine or with people. When goods inflation runs far ahead of wage inflation, capital equipment, which is a good, is getting expensive faster than labor is. That quietly tilts the make-versus-automate calculus back toward labor at the margin, the reverse of the usual "automate because labor is expensive" instinct, because in a two-speed world it is the machine, not the worker, whose price is climbing fastest. A plant that automates on autopilot without checking the relative cost trend may be buying the input that is inflating hardest.

The fear is a spiral. The reality is a divergence. Goods and labor are not chasing each other up; they are running at different speeds, and the gap is the story.

The caveat, and what would change it

Decoupling is a snapshot, not a permanent state. If a tight labor market eventually pushes wages to catch up with goods, the gap closes and the spiral fear becomes more relevant, which is why the quits rate and wage trend are worth watching as the leading edge of any reconvergence. And a single hot commodity, copper, is doing much of the goods-side work, so the goods number is more concentrated than it looks. But as of now the two speeds are real and wide, and a plant that plans as if a wage-price spiral were underway is preparing for the wrong cycle.

The goods side, over six years

Over the 36-year record copper producer prices have moved decisively rather than oscillated: from 127.60 at the close of 1990 to 557.23 today, up 337%, and now at the top of its 36-year range. A change of that size across a span this long is a level shift, not a cycle, and planning that assumes a return to the 1990 figure is planning against the whole record.

Use the equipment payback calculator against current labor cost to test the automate-or-not decision in a two-speed cost world. Weigh capital vs labor

Published 2026-08-06.