Manufacturing Costs

Paper rose 2.41% while resin fell 8.08%. Summer purchasing inflation had no single direction.

May to July 2026 paper rose 2.41%, resin fell 8.08% and manufacturing hourly earnings rose 0.43%. January to July resin remained up 15.78%, qualifying the short decline.

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

From May to July 2026, the paper producer-price index rose 2.41% while plastic resins and materials fell 8.08%. Manufacturing production-worker hourly earnings rose 0.43% over the same interval. For a September purchasing review, those three paths make a single automatic inflation adjustment difficult to defend.

The comparison does not establish whether a finished paper package became more expensive than a plastic alternative. It establishes something narrower and useful: the relevant input benchmarks moved differently, even before conversion, delivery and the buyer's specifications entered the calculation.

Two months produced a 10.49-point materials split

The paper index moved from 276.352 in May to 283.012 in July, using the unrounded source values. Plastic resins and materials moved from 318.635 to 292.875. Percentage changes within each series put the difference in their movements at 10.49 percentage points.

Their raw index levels cannot be compared as prices. An index near 300 is not a statement that one material costs more per kilogram than another near 280. The useful comparison rebases each May observation to 100 and follows its own percentage change.

On that basis, July paper is 102.41 and resin 91.92. The divergence becomes visible without implying a common unit price. A September review should retain the May-to-July dates beside the numbers, because changing the starting month changes the conclusion about the larger price cycle.

The monthly path supports the endpoints

Paper rose 1.75% between May and June and another 0.65% between June and July. Resin fell 2.86% and then 5.38%. The opposing net changes therefore do not arise solely from two selected endpoints concealing an opposite middle-month pattern.

That is a modest but useful check. The window contains only three monthly observations and two monthly intervals. It cannot establish a persistent relationship, a forecast of September quotations or a general rule that paper and resin move inversely.

The time path does show that an estimator using a common materials escalator through this part of the summer would miss distinct observed movements. Whether those movements reached a particular supplier depends on the purchased material, its pricing terms and the timing of the supplier's own commitments.

The year-to-date comparison tells a different story

Resin's decline from May does not mean it was cheaper than at the beginning of 2026. From January to July, the resin index was up 15.78%. Paper was up 10.13% over the same January-to-July span.

That comparison is a useful challenge to the most dramatic interpretation of the summer split. A recent resin retreat was occurring within a substantial increase from January. Calling it broad resin deflation without naming the window would hide that fact.

For a contract review, the relevant baseline may be its agreed reset date rather than January or May. The data can reproduce either comparison, but it cannot choose the contractual reference. The honest article shows the path and explains why a buyer with a May quote and one with a January quote can legitimately focus on different changes.

Paper is not a finished box

The paper measure is BLS series WPU0913. It covers paper, not paperboard, corrugated containers or every finished packaging product. The resin measure, WPU066, covers plastic resins and materials, not the final price of a molded container.

That source distinction limits the conclusion about substitution. Two finished packages may require different material quantities, conversion processes, tooling, protective performance and delivery arrangements. The input indexes alone do not establish which option provides a lower cost for an equivalent job.

The paper history and resin history should therefore be labeled by their official scope when used in a purchasing document. The older wording retained in a URL is not evidence that the paper series measures containers. A benchmark can be useful without being stretched to cover products outside its definition.

A transparent basket makes the weighting visible

Consider a purely illustrative $100,000 May cost basket with 40% assigned to paper, 40% to resin and 20% to manufacturing hourly earnings. Applying the observed changes fully to those fixed base amounts produces $97,816.29 in July, a decline of about 2.18%.

That is not a measured packaging-cost index. It assumes fixed quantities and full movement of each selected benchmark into its assigned cost bucket. It excludes the possibility of fixed-price contracts, different product specifications and costs not represented by those three measures.

The illustration earns its place because its answer depends visibly on the weights. A paper-heavy basket and a resin-heavy basket would not have the same result. Publishing one blended number without its weights would conceal the assumption that determines which material's change dominates.

Use September to reconcile the actual purchase

The practical next step is to identify the material specification, the portion of the quoted cost exposed to it and the reference period in the agreement. A supplier citing rising materials should be able to explain which materials and which dates support that statement.

The same standard applies to a buyer demanding a reduction because resin fell. A supplier may have bought material earlier, held a fixed supply commitment or incurred other changes. The benchmark is evidence to investigate, not proof of the supplier's realized cost or margin.

The summer record supports a precise conclusion: paper rose, resin fell and hourly earnings edged up between May and July. The January comparison prevents that finding from becoming an exaggerated claim about the whole year. Together, they provide a better basis for a September price discussion than a single undefined materials-inflation allowance.

Sources and calculation

The analysis uses BLS paper WPU0913, plastic resins and materials WPU066 and manufacturing earnings CES3000000008, in the version available September 9, 2026. The BLS schedule shows July PPI released August 13; August PPI was scheduled for September 10 and is excluded. Percentage changes use unrounded source observations.

Sources and evidence

Evidence period: May to July 2026; January to July sensitivity check. 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.

fred.stlouisfed.org/series/WPU0913

fred.stlouisfed.org/series/WPU066

fred.stlouisfed.org/series/CES3000000008

Published 2026-09-29.