Manufacturing Costs

Up 52.84% for the year, down 2.73% in three months: July metals prices tell two stories.

July 2026 primary nonferrous PPI rose 52.84% over the year and fell 2.73% over three months. Nonferrous mill shapes rose 30.21% and 4.07%, respectively, so the second category does not share the short retreat.

Editorial evidence cutoff: September 9, 2026. Published September 29, 2026. Observation periods are stated throughout; older figures are retrospective evidence.

A supplier can point to a 52.84% annual increase while a buyer points to a 2.73% decline over the latest three months. In July 2026, both numbers accurately described the same primary nonferrous metals producer-price index.

That is not an error in the arithmetic. It is a dispute waiting to happen when a contract, a budget and a market discussion use different comparison windows without saying so. The annual increase describes the distance from July 2025. The quarterly decline describes the distance from April 2026. Neither percentage cancels the other.

The July comparison is relevant to September 2026 negotiations because different reference months can support very different adjustments on the same order.

One benchmark, two defensible arguments

The primary nonferrous metals index was 52.84% above its July 2025 level and 2.73% below its April 2026 level. A buyer whose latest quote was set in April could reasonably care about the retreat since then. A supplier whose agreement resets from the previous July could correctly calculate a large positive adjustment under that reference period.

Which comparison should govern is a commercial and contractual question. The data alone cannot answer it. What the history can establish is that using an annual inflation headline to describe every current transaction loses information about the path between its endpoints.

The distinction also prevents a misleading editorial shortcut. A quarterly decline does not prove that prices have become low. An annual increase does not prove they are still rising over the most recent months. Both statements depend on a baseline that should be visible wherever the percentage appears.

This is a recurring pattern

The archive contains 427 months in which both a twelve-month and a three-month change can be calculated, ending January 1991 through July 2026. The primary nonferrous index had opposite annual and quarterly directions in 121 of those matched windows.

Some disagreements could reflect tiny changes. To avoid allowing rounding noise to carry the argument, we required annual growth above 0.5% and a quarterly decline below minus 0.5%. Fifty-six windows still met that stricter test. Before 2020, there were 49 such windows among 348 eligible observations.

Those are overlapping observations, not 56 independent economic episodes. A persistent retreat from a high level can produce several successive months with the same sign conflict. That is useful for a contract review: the mismatch can remain visible across multiple reset dates rather than appear as a single anomalous print.

A second metals index refuses the same story

Nonferrous mill shapes were 30.21% above July 2025 and 4.07% above April 2026. Unlike primary nonferrous metals, this neighboring category rose on both horizons. That is a useful counterexample to a broad claim that the entire nonferrous sector was already retreating.

The two measures should retain their proper names. The primary nonferrous series is not a copper-and-brass-only benchmark, and the mill-shapes series is not an aluminum-only price. Broad categories can contain a changing mix of products whose own contracts behave differently. A familiar metal name attached to a broad index makes an already difficult comparison less reliable.

The primary nonferrous history and nonferrous mill-shapes history provide the underlying observations. Their legacy route wording should not dictate the description of what the source measures. For a September review, the contrast argues for checking the purchased product and the actual reference dates before extending one market's short retreat to another.

The six-month check blocks a bigger claim

If the latest quarter is falling, has a longer downward trend begun? The saved data does not establish that. Primary nonferrous metals were still 5.14% higher over the six months ending July 2026. Nonferrous mill shapes were 13.05% higher over the same interval.

Across the full primary nonferrous history, replacing the quarter with six months leaves 30 windows with annual growth above 0.5% and a six-month decline exceeding 0.5%. The broader phenomenon survives, but the current six-month comparison does not point downward.

That failed extension belongs in the article. The strong finding is a recent quarterly retreat inside a large annual increase. Turning it into a claim that the entire cycle has reversed would use the data for more than it can support. Seasonality is another reason to be precise: these producer-price measures are not seasonally adjusted.

The reference month can matter more than the headline

Consider a buyer preparing a new annual agreement and another reviewing an order quoted three months earlier. They can read the same index release and need different answers. Applying a single percentage to both transactions confuses a description of the market with the measurement required by the decision.

A transparent adjustment identifies the base observation, the comparison observation, the affected portion of the price and the timing of the reset. It also explains what happens when the measure falls and what occurs if an observation is revised. Those details make the calculation reproducible rather than argumentative.

None of this requires assuming that a supplier is opportunistic or a buyer is unreasonable. Their cost positions may genuinely differ. Purchased inventory, firm supply commitments and long production lead times can separate a supplier's realized costs from the latest published index. That difference calls for evidence about the order, not an accusation based on one percentage.

Report the path before recommending the posture

The useful chart marks three dates: the year-earlier observation, the recent peak or reference month, and the latest reading. It lets readers see why a large annual increase can coexist with a short retreat. A single annual bar cannot show that path.

For purchasing teams, the implication is to ask what period an adjustment is intended to cover before debating its size. For writers, it is to resist presenting the latest percentage as a complete market description. A strong headline can retain a contradiction without pretending one side is false.

Here the contradiction is the story: plus 52.84% over a year and minus 2.73% over three months. Both are true. The work is deciding which question each number answers, and which additional costs or commitments the index leaves outside the frame.

Sources and calculation

Calculations use the saved BLS primary nonferrous metals series and nonferrous mill-shapes series, January 1990 to July 2026. Annual and quarterly changes share the same ending month. No annualization, inflation adjustment or inference about supplier margins is applied. Threshold counts use unrounded changes.

Sources and evidence

Evidence period: January 1990 to July 2026; latest change ends July 2026. The frozen evidence record lists the source files and verified hashes available September 9, 2026. Source revision: 5e4fb7726c3d40060c0151c086c903baae856cab. Later live-data updates do not alter the historical evidence in this article.

data.bls.gov/timeseries/WPU1022

data.bls.gov/timeseries/WPU1025

Published 2026-09-29.