Manufacturing inventory and PPI — 2024-2026 mid-market read.

01 What Census M3 actually shows

The Census Bureau's Manufacturers' Shipments, Inventories and Orders (M3) release is the most operationally useful monthly read for any private manufacturer above $25M in revenue. Three series matter for the Q1 2026 reading. New orders ran $630B in March 2026, up roughly 7% versus March 2025. Total inventories ran $956B, an all-time high but up only 1.3% year-over-year. Unfilled orders ran $1.54T, an all-time high and up roughly 11% year-over-year. The composite reading is what matters: orders accelerating faster than inventory builds, and a backlog that is widening month-over-month into 2026.

The single most consequential derived series sits underneath those three: the manufacturing inventory-to-sales ratio. It ran 1.60 in January 2024, drifted down to 1.58 through most of 2024, broke to 1.56 in mid-2025, and printed 1.51 in March 2026 — the lowest reading since the immediate pre-COVID period. The destocking cycle that defined 2023-2024 is over. The current cycle is lean inventory + rising demand + accelerating input prices, which is operationally a very different posture than the "we're drowning in 2021 vintage cost-base" cycle most operators are still pattern-matching against.

The unfilled-orders trajectory is worth one more sentence because it is the cleanest leading indicator for the back half of 2026. Backlog grew $155B in the trailing twelve months — concentrated in transportation equipment, machinery, and electrical/computers. That is a positive demand signal for manufacturers in those NAICS codes, and a clear capacity-allocation constraint for everyone supplying into them. The next two quarters' margin story will be substantially determined by whether mid-market suppliers in that ecosystem can convert backlog into shipped revenue at the pricing level the input-cost environment now requires.

02 PPI by stage of processing — the cost wedge that defines 2026

The single most important data point in this entire read is the divergence between PPI stages of processing across the trailing 28 months. Crude materials (the BLS WPSID62 series — unprocessed raw inputs, before any value-add) climbed roughly 22% from January 2024 to April 2026. Intermediate materials (WPSID61 — partially processed inputs, the typical "components and supplies" line on a manufacturer's P&L) climbed roughly 11% over the same period. Finished consumer goods (WPSFD49207 — the price manufacturers actually get) climbed roughly 7%.

That is the cost wedge in one sentence: input prices ran roughly three-to-one ahead of output prices over 28 months, and the gap accelerated sharply in Q1 2026. Between December 2025 and April 2026 alone, crude materials PPI jumped from 252 to 300 — a 19% move in four months, the steepest stretch of the cycle. Intermediate materials moved from 254 to 283 over the same window, up 11%. Finished consumer goods moved from 261 to 267, up only 2%. The pass-through engine is not keeping pace, and the squeeze is concentrating itself in the converters who buy intermediate inputs and sell finished goods — exactly the mid-market processing layer.

The Q1 2026 acceleration in crude PPI is tariff-driven, commodity-driven, or both depending on the input — copper, steel, aluminium, refined petrochemicals, and certain agricultural intermediates have all moved sharply in the same window. The PPI manufacturing output series (PCUOMFGOMFG) confirms the asymmetry from the seller side: it climbed from 246 in December 2024 to 273 in April 2026, a +11% move over sixteen months, but that aggregate hides wide dispersion by NAICS — durable-goods manufacturers with backlog leverage are pricing through, undifferentiated commodity converters are not.

03 Industrial production — the recovery that finally took

Total industrial production (FRED INDPRO) ran 102.5 in April 2026 (2017=100 base) — an all-time high. Manufacturing industrial production (IPMAN) ran 98.7 in April 2026, still below its 2018 cycle peak but the highest reading since mid-2023. The IPMAN year-over-year growth rate is the cleaner read on momentum: persistently negative through all of 2024 (-0.5% to -1.7% monthly prints), flat-to-slightly-positive in Q1 2025, then breaking above +1% by mid-2025 and printing +1.4% in April 2026.

In operating terms, manufacturing IP turning positive year-over-year in mid-2025 after eighteen months of contraction is the single most important sentiment marker in the dataset. The 2023-2024 contraction was real and prolonged. The 2025-2026 recovery is real and incremental. It is not a "demand is exploding" cycle. It is a "the bottom is in, capacity utilization is normalising back toward 78-80%, and customers are reordering at trend pace" cycle. The implication for capex underwriting is direct: greenfield capacity additions remain hard to justify on demand alone; brownfield productivity and inventory-positioning investments clear much more easily.

Underneath the aggregate, two NAICS clusters explain most of the variance. Transportation equipment and aerospace have been a steady-positive contributor since mid-2024 — Boeing backlog re-acceleration and EV/auto-electrification capex are visible in the M3 unfilled-orders backlog. Electrical equipment and computers/electronics turned positive sharply in late 2025, driven by data-center power infrastructure (the Eaton 10-Q is the cleanest public-cohort read on this — Q1 FY26 revenue +16.8% YoY, capex stepping up to 2.6% of revenue). Primary metals, chemicals, and consumer-durable household have been the laggards — still close to flat year-over-year despite the broader recovery.

04 What the listed manufacturer cohort says

The public industrial cohort's Q1 2026 filings provide the cleanest readthrough into how the mid-market should be positioning. Four names worth examining: Parker-Hannifin (PH), Rockwell Automation (ROK), Emerson Electric (EMR), and Eaton (ETN). Each tells a different part of the story.

Parker-Hannifin — diversified industrial pass-through

PH posted $5.49B in Q3 FY26 revenue (calendar Q1 2026), up 10.6% year-over-year. Operating margin held at 22.4% versus 22.3% in the prior-year quarter — flat margin despite tariff and PPI pressure. Inventory grew 12.6% to $3.18B; inventory days extended marginally from 81.2 to 82.5. Capex ran a disciplined 1.6% of revenue. The PH read: strong demand, modest pre-tariff inventory build, gross-margin pricing power intact at the platform level. This is what large-cap manufacturing pass-through looks like in 2026 — and it is the operating posture mid-market peers should be benchmarking against.

Rockwell Automation — best-in-class margin expansion

ROK is the best-in-class read of the cohort. Q2 FY26 revenue $2.24B, up 11.9% year-over-year. Gross margin expanded 160 basis points to 50.2%, operating margin expanded 330 basis points to 23.7%. Inventory grew only 4.3% year-over-year, and inventory days actually compressed from 103 to 99. The pattern is unmistakable: pricing actions and productivity initiatives running ahead of input-cost inflation, with inventory tightening rather than building. ROK had the strongest cost-of-revenue discipline in the cohort and is the cleanest case study for what mid-market manufacturers should be trying to replicate.

Emerson Electric — the pre-positioning posture

EMR posted Q2 FY26 revenue of $4.56B, up a more measured 2.9% year-over-year. Inventory grew 10.6% to $2.45B and inventory days extended from 97 to 103 — a +6 day build that is explicitly pre-positioning for tariff timing on imported automation components. Net margin expanded to 13.5% from 10.9% on operational mix and post-Aspen Tech integration synergies. The EMR pattern is the one most mid-market manufacturers should think about copying: deliberate, controlled inventory build into a known tariff cycle, with pricing actions covering the carrying cost.

Eaton — the data-center cycle

ETN tells the demand-leading-pricing story. Q1 FY26 revenue $7.45B, up 16.8% year-over-year — easily the strongest growth in the cohort. But gross margin compressed roughly 280 basis points (from 38.4% to 35.6%) as imported component costs ran ahead of pricing. Inventory grew 17.2% (slightly ahead of revenue growth) but inventory days actually compressed from 100.6 to 96.5 — strong inventory turn even at the elevated demand level. Capex stepped up to 2.6% of revenue, signaling capacity additions for data-center power infrastructure. The ETN read: extraordinary topline growth but margin pressure from input mix; mid-market manufacturers in the data-center / electrification supply chain should expect a similar pattern.

05 What this means for $25M–$200M manufacturers

The macro data, the PPI stage-of-processing data, and the listed cohort all point in the same direction for mid-market private manufacturers. Three operating implications come out of the read, sequenced by the question they answer.

Reprice now if you have not in the trailing six months

PPI for intermediate materials is up 9.9% year-over-year on the April 2026 reading, accelerating from a 4.7% reading just six months earlier. For any manufacturer who set list prices in late 2025 and has not raised since, the gross-margin compression mathematically equals roughly 4-6 percentage points on the COGS line — and that compression is going to land in calendar Q2-Q3 2026 numbers if pricing does not move. The ROK case study is the right reference: pricing actions and productivity-led cost-out running concurrently, with the pricing action sequenced ahead of the input-cost arrival rather than after. We see most mid-market manufacturers running the reverse sequence — wait until the input-cost hit lands, then attempt to recover via a delayed price increase, which leaves a one-to-two-quarter margin hole.

Pre-position inventory for the tariff cycle, but do it deliberately

The total business inventory-to-sales ratio at 1.32 (March 2026, FRED ISRATIO) is the lowest since 2014. Manufacturers are running lean inventory positions into rising demand, which is operationally tight. The Emerson pattern — deliberate, identified, partially-financed pre-positioning for known tariff cycles — is the model. The Eaton pattern — inventory growing in line with revenue but tightening on a days basis — is also valid for fast-growing names. The wrong pattern is the panicked across-the-board inventory build, because the working-capital cost of carrying broad inventory at 2026 rates (SOFR + 350-450 bps for typical mid-market credits) eats most of the protection. The right pre-position is SKU-by-SKU, vendor-by-vendor, with the carrying-cost math run on each line.

Defer the speculative capex, accelerate the productivity capex

The listed cohort is running capex at 1.6%–2.6% of revenue — restrained by historical standards, but with the highest spend concentrated in clear-demand categories (Eaton on data-center power). Mid-market private manufacturers should mirror this: speculative greenfield capacity (a new line, a new plant) is hard to underwrite when IPMAN is only running +1.4% YoY and capacity utilization is still in the 78-80% range. Brownfield productivity capex — line automation, scrap reduction, energy efficiency, MES integration — clears IRR thresholds at much lower demand assumptions and de-risks the margin profile against the next input-cost shock. Run the capex sequencing through the same SKU-level landed-cost lens we use on inventory; the projects that survive are typically the ones that touch the highest-volume / lowest-margin SKUs first.

06 Sequencing — what to do in the next 90 days

Three operating moves are sequenceable inside one quarter for any mid-market manufacturer reading this in spring 2026.

1. 01
   Run a 30-day pricing audit across the top-50% SKU revenue base. Pull the trailing-12 unit cost trajectory by SKU using the BOM and standard-cost roll. Compare to the trailing-12 list-price trajectory. Any SKU where the COGS line has moved >5% but list price has moved <2% is a flagged candidate for repricing. Push the repricing window inside calendar Q3 2026 — once the Q1 2026 PPI acceleration starts arriving in landed costs, lead times tighten and the customer-conversation window narrows.
2. 02
   Build the 60-day pre-positioning inventory matrix. Identify the SKUs where (a) you have visibility on a known tariff or input-cost step-change in the next 90-120 days, (b) your supplier has the capacity to ship ahead of that step-change, and (c) the carrying cost at your working-capital rate is less than the cost-step you are protecting against. For most operators, the matrix surfaces 20-40 SKUs — not the full catalog. The Emerson pattern is the reference here, run at mid-market scale.
3. 03
   Re-rack the capex schedule on a 90-day cycle. Pull the current 12-month capex schedule. Tag each project as (i) speculative-greenfield, (ii) maintenance, or (iii) productivity-with-defined-payback. Defer category (i) by one cycle unless the demand is contracted. Accelerate category (iii) into the next quarter if the IRR clears your hurdle rate at current PPI assumptions. Run the lender covenants alongside — most mid-market revolver covenants tighten on a debt-to-EBITDA basis, and accelerating productivity capex pre-EBITDA-recovery requires a deliberate conversation with the lender.

07 What we are watching into Q3 2026

Three signals matter into Q3 2026. First, whether the PPI crude-materials series continues its Q1 2026 acceleration or rolls over — the April 2026 print of 300 (1982=100) is a 25-year high and the trajectory is not sustainable indefinitely. A roll-over reading is what gives mid-market manufacturers margin room without further pricing actions; a continued acceleration is what forces a second round of repricing inside calendar 2026. Second, whether unfilled orders continue to widen — the current 11% YoY growth concentrates in transportation, electrical, and machinery, and the durability of that backlog determines whether the demand recovery has legs into 2027. Third, whether the Fed's rate path continues to ease — the current 3.63% Fed Funds rate (May 2026) is supportive of working-capital builds, but any reversal directly compresses the pre-positioning math.

We publish this read quarterly. The Q3 2026 update is scheduled for August 2026 and will incorporate the July 2026 PPI release, the Q2 2026 M3 release, and the Q2 2026 listed-manufacturer 10-Qs.