The Factory Recovery Has a Parts Bill

American factories entered October with a useful problem: customers wanted more, while supplying them became more expensive. September’s manufacturing index edged down to 54.5 from 54.6, still indicating expansion, while the input-prices index jumped 6.8 points to 77.9. As Trade Radar reported on October 2, these are diffusion readings describing the breadth of reported changes. Factory prices did not rise 6.8%, and output did not grow 54.5%. The decimal places are precise; their meaning still requires adult supervision.

The tariff friction appears further down the purchasing list. A machinery respondent said Canadian tariffs had increased cross-border costs and disrupted established supply chains. An electrical-equipment respondent said new tariffs against Canada had raised the cost of capital equipment and assemblies. Those are distinct directions of the trade dispute, and individual business accounts rather than a national estimate. But they locate a concrete mechanism: trade barriers can affect the equipment and components manufacturers need to produce, alongside the goods they sell. Trade Radar identifies procurement trouble inside an expanding industry.

That expansion deserves equal billing. New orders rose to 55.3 from 53.7, backlogs increased, and the reporting identified AI infrastructure spending and inventory rebuilding as supports for activity. It also described energy and supply risks associated with the Iran conflict. The prices-index jump therefore cannot be handed entirely to tariffs, however convenient that would make the argument. The narrower conclusion is consequential enough: stronger demand coexisted with reported tariff-related sourcing costs. For businesses, a fuller order book does not settle the cost of filling it. For consumers, this is evidence of upstream pressure, without a measured increase in shelf prices. Trade Radar

An earlier manufacturing survey shows why expansion and operational improvement require separate inspection. In May, the manufacturing PMI rose to 54.0 from 52.7, its highest reading in four years. Yet the supplier-deliveries index stood at 60.6, indicating slower deliveries, with aluminum, electrical and electronic components, resins, semiconductors and steel products among the scarce inputs. Reuters, published by MarketScreener, reported those conditions in June. The headline was cheerful; the delivery schedule had reservations.

The contradiction had a specific statistical mechanism. Longer delivery times likely helped lift May’s PMI because slower deliveries are normally associated with stronger demand. In that episode, however, snarled supply chains were lengthening the wait. The measure could register a feature usually associated with economic strength even when disruption helped produce it. That does not establish the same mechanism in September. It explains why September’s expansion reading, set beside its sharper input-cost pressure, is insufficient on its own to pronounce procurement healthy. MarketScreener; Trade Radar.

May also exposed consequences beyond the headline index. Food, beverage and tobacco manufacturers reported diesel costs hurting profitability, while the manufacturing employment index recorded its 32nd consecutive month of contraction despite increasing orders. Those historical details matter because they identify separate tests of a recovery: margins and hiring can diverge from demand, and energy costs can add pressure alongside tariffs. They do not establish September’s employment or profitability. They give readers a reason to ask what accompanies growth before treating it as an all-purpose certificate of industrial health. A factory can receive more business and still have less reason to celebrate each delivery invoice. MarketScreener

Investment announcements face a related test before they become production. In August, Treasury official Chris Pilkerton and Small Business Administrator Kelly Loeffler told Reuters that tariffs and deregulation were encouraging foreign manufacturing investment, while acknowledging that components had to be available. Their explanation of what drove investment was an official claim. The dependency was illustrated more concretely at Hanwha Philly Shipyard: CEO David Kim said each ship relied on more than 1,000 suppliers, roughly two-thirds based in the United States. Industrial expansion arrives with a purchasing network attached. Reuters, published by Investing.com

Kim described a $5 billion investment pledge for the coming years that could increase employment to 10,000 from around 2,000. More than $200 million had already been invested in workforce, capabilities and capacity. Keeping those numbers in their respective tenses is essential: spending completed, spending promised and jobs possible measure different stages of expansion. The supplier network helps explain what sits between them. Capital can support additional capacity, but a pledge alone does not establish that the components needed for that capacity are available. Investing.com

The sourcing gap was explicit elsewhere. Pilkerton recounted visiting a quantum refrigeration company acquired by Bluefors whose executives said they could not find a domestic supplier for dry compressors. The administration’s supplier initiative remained a pilot without detailed guidance for identifying vendors. These August accounts describe component availability; September’s machinery and electrical-equipment respondents describe disruption and cost. Together, they show two distinct procurement constraints on industrial ambitions, without establishing that the Canada dispute affected either named company. The practical test of progress is therefore demanding but ordinary: watch equipment costs, supplier capability and realized capacity alongside orders and pledges. An announcement can contain five billion dollars. A purchasing department still needs someone who can deliver the part. Investing.com; Trade Radar.