Opinion

Reshoring Pencils for Some Products and Is Fantasy for Others

Reshoring is not a yes-or-no question, and treating it as one is how companies make expensive mistakes in both directions. It is selective, and the landed-cost math draws the exact line between the products that pencil and the ones that are pure press release.

Opinion | By Lena Fournier, Trade & Policy. The argument here is the columnist's own; every figure links to the live series behind it, and opinion is not measurement.

The reshoring debate is broken because it is framed as a referendum: is reshoring happening or is it hype? The honest answer is that it is neither and both, because reshoring is not one decision. It is thousands of separate product-level decisions, each with its own landed-cost math, and that math sorts them cleanly into the ones that pencil and the ones that are fantasy. The inputs that do the sorting are all live: currency, with the peso at 17.1357 (Aug 7, 2026); loaded US labor, benchmarked off $30.35/hour; freight, tracking diesel at $5.26/gal; and the capital to build a line, visible in machinery imports at $78.39B.

What puts a product on the right side of the line

Reshoring pencils when four conditions line up. Labor content is low or automatable, so the wage gap that favored offshoring barely applies. The product is bulky or heavy, so freight and the long pipeline penalize distance hard. The foreign currency has strengthened, eroding the offshore cost advantage without anyone renegotiating. And the volume is high enough to amortize the capital cost of a new line. Hit those four and the landed-cost math often favors home, not out of patriotism but out of arithmetic. Miss them, and reshoring is a slide in a strategy deck that the spreadsheet never supported.

The honesty check nobody runs

There is a simple test that separates real reshoring from theater, and it is the machinery import data. You cannot reshore production on equipment you have not bought, so a genuine trend has to show up as sustained strength in capital-goods purchases before any output shifts. When the announcements run hot but machinery imports stay flat, the intent may be sincere but the capital commitment that would make it real has not landed. Here the long lens is unusually encouraging: machinery imports are climbing and, on their full five-year archive, the highest in the five-year archive, more than double their February 2022 level. Whatever you think of the rhetoric, the capital goods that turn reshoring from slogan into installed line are actually being bought, and that trajectory is a better lie detector for the story than any number of ribbon-cuttings.

Reshoring is not a belief. It is a spreadsheet that comes out differently for a heavy, automated part than for a light, hand-built one, and the announcements blur a line the math draws sharply.

The year the argument ended

I have been skeptical of reshoring rhetoric for years, and the machinery import series is the reason I have changed my mind about the direction, if not about the selectivity. Look at the shape. For three years, 2021 through 2023, capital-goods imports went nowhere: down 2% across the entire stretch, while the announcements piled up and the ribbon-cuttings continued. That gap between rhetoric and capital was the honest case for cynicism, and I made it. Then it broke. From 2023 to today, machinery imports are up 119%.

You cannot fake this series. A press release is free; a machine tool clears customs with a value on it. Whatever anyone believes about whether reshoring is real, somebody is buying an enormous and rapidly growing quantity of production equipment and bringing it into the United States, and they are not doing it for the photo opportunity. This is the single most persuasive number in the reshoring debate, and it is one almost nobody cites, because it sits in trade data rather than in a policy announcement.

Note the irony, because it is the kind of thing that gets lost in the referendum framing: the clearest evidence that America is reshoring production is a surge in IMPORTS. The equipment that makes domestic manufacturing possible mostly is not made domestically, so the first phase of any genuine reshoring wave has to show up as a widening capital-goods import bill. Anyone reading the trade deficit as a scoreboard will read this exactly backwards.

Stop arguing the referendum

The useful posture is to abandon the yes-or-no framing entirely and run the ledger product by product at today's currency, freight, and wage inputs, then rerun it whenever those move, because any of them can flip a marginal case. Reshoring is real for the products where the math says so and fantasy for the ones where it does not, and no amount of policy enthusiasm or offshore-cost nostalgia changes which bucket a given part falls into. The companies that win this decade will be the ones that computed the line instead of arguing about the referendum. For the live count of how many product families currently clear the break-even, the reshoring signal page keeps the tally.

Use the nearshoring landed cost calculator to sort your own products into the ones that pencil and the ones that don't. Run the product-level math

Published 2026-08-06.