What Manufacturing Inventory-to-Sales Ratios Can and Cannot Tell a Buyer

The Census Bureau's August 2026 inventories and sales release, published October 15, put the total business inventories/sales ratio for manufacturers, retailers and merchant wholesalers at 1.33 on seasonally adjusted data, against 1.40 a year earlier. For manufacturers alone, the full report's summary table shows a seasonally adjusted ratio of 1.48 for August, a preliminary estimate, against 1.56 a year earlier, even though the same table shows manufacturers' inventories higher than a year before: shipments grew faster. When a port strike threatens or a supplier asks for longer lead times, figures like these invite a quick conclusion about how much stock stands between a buyer and a shortage. This article sets out what the manufacturing measures count, how the annual and monthly series differ, and what a buyer needs from a supplier before any ratio changes a sourcing decision.

What does an inventory-to-sales ratio measure?

The ratio is arithmetic with a unit attached. A research manuscript by economists Valerie Ramey and Daniel Vine describes an inventory-sales ratio as a stock divided by a flow, measured in units of time such as the number of days of sales held in inventory. The Census Bureau reads its monthly ratios the same way: its program definitions relate end-of-month inventory values to a month's sales and describe the result as an indication of the months of inventory on hand, a reading the page gives for retail ratios. By that arithmetic, a manufacturers' ratio near one and a half means stock worth roughly six weeks of shipments at the current pace, for the sector as a whole. Read that as a rough measure: the same definitions value inventories at cost but shipments at selling value, and count purchased materials and work in process alongside finished goods.

The unit also depends on what sits in the denominator. Ramey and Vine note, as Ramey and West's 1999 handbook chapter had pointed out, that real and nominal aggregate inventory-to-final-sales ratios trend very differently; they add that by the real measure the aggregate ratio was no lower in 2003 than in 1950, and that the nominal trend is affected by relative price changes and can be very misleading. For more homogeneous goods such as manufactured goods, the same note expects nominal and real ratios to present similar pictures, but it flags a different aggregation problem, which it credits to earlier research: shifts in the share of sales across sectors with different ratios can make an aggregate ratio behave differently from the ratios beneath it. The Census figures discussed here are current-dollar values, and the August release says its inventories are adjusted for seasonal and trading-day differences but not for price changes. A move in the all-manufacturers ratio can therefore reflect a change in the mix of industries as well as a change in stock or shipments, and the ratio alone does not say which.

How do the annual and monthly Census measures differ?

The annual measure comes from the AIES. The Census Bureau's description of the survey says it replaced and integrated seven annual business surveys in 2023 and is sent to approximately 359,000 companies covering domestic nonfarm employer businesses. Its 2024 table, Manufacturing: Inventory Valuation Statistics for Employer Firms, values total inventories at cost or market as of December 31 and before any LIFO adjustment, includes stock wherever it is held, Foreign Trade Zones and bonded warehouses included, and excludes goods owned by others and held on consignment. The table also reports the value of inventories subject to LIFO valuation and the LIFO reserve, the gap between current cost and the LIFO carrying value, and shows manufacturing at the 2- through 6-digit NAICS levels. When a listed supplier reports its inventories on a LIFO basis, add back its disclosed LIFO reserve before setting its figure beside a Census total stated before any LIFO adjustment, and remember that a company's consolidated figures still differ in scope, and can differ in date, from an industry estimate.

The monthly measure works differently. In the monthly program's definitions, manufacturing inventories are end-of-month stocks at any stage of fabrication, from purchased materials to finished goods, valued at cost on any basis other than LIFO, while manufacturers' shipments are net selling values f.o.b. plant, with freight and excise taxes excluded. The August report adds that manufacturers contribute no estimate of sampling variability, because the mail panel behind the monthly manufacturing survey is not a probability sample from a known frame, so the manufacturing figures carry no published standard errors.

The two series are linked through benchmarking, though not in 2026 for manufacturing. The AIES program page says AIES estimates serve as a benchmark for Census indicator programs, the Manufacturers' Shipments, Inventories and Orders survey among them. The August report states that its wholesale and manufacturing data are unrevised from their most recent monthly reports, and that revised historical data and seasonal adjustment models for the manufacturing survey will remain unchanged for the rest of 2026. A buyer comparing the manufacturing ratio over time should record which release each figure came from and expect its history to change whenever that revision is published.

What can the ratios show a buyer, and where do they stop?

At sector level, a ratio is a reasonable prompt. A manufacturers' ratio lower than a year earlier says the sector carries less stock against its shipments, which is a reason to ask critical suppliers about cover before a disruption rather than after. What the ratio cannot do is locate the stock. A peer-reviewed study of pandemic-era U.S. imports reported that retail inventories stayed low partly because goods ordered weeks and months earlier were not yet in inventory, and it described the inventory-to-sales ratio as consolidating several factors at once. A low or high ratio can reflect stock in transit, orders not yet placed, a swing in sales or a change in prices, and the ratio alone does not separate them.

The annual table has the same ceiling. The AIES table notes make the company the reporting unit, one or more establishments under a single owner, and say every estimate is subject to sampling variability, with large coefficients of variation calling for caution; a cell marked D is withheld to avoid disclosing individual companies, and S marks an estimate that does not meet publication standards. A year-end figure for a six-digit industry is a weighted estimate across employer firms on one date, not a count of any supplier's shelves.

How should a buyer record a ratio beside supplier evidence?

A ratio becomes useful when it is recorded with everything that changes its meaning and set beside the supplier evidence it cannot replace. The register below is our proposal for doing both once per review.

What supplier records answer the stoppage question?

For a buyer weighing a port stoppage or a request for longer lead times, the question is how many days of demand a named supplier can cover for named parts. Start with your own exposure: map which of your inbound shipments and which suppliers' imported inputs route through the affected ports. Then ask each exposed supplier for on-hand quantities of your part numbers by plant and warehouse, on-hand stock of the imported inputs behind them, work in process, open purchase orders for those inputs with expected receipt dates, the quantity committed to your orders and its allocation policy if supply runs short, and the lead time it would quote today. Ask how much stock sits in transit or in a bonded warehouse. Where a supplier cannot share these records, score the gap in the risk review instead of filling it with an industry ratio, as set out in our guide to supplier risk assessment methods, and make the same records a standing item in supplier relationship reviews, so a disruption does not start the conversation from nothing.

Where does a public inventory ratio belong in supplier risk work?

Before the next review of a critical part, note the latest manufacturers' ratio with its release date and revision status, look up the year-end inventory for the supplier's six-digit industry and its flag in the panel above, and send the supplier the requests in the last column of the register. If the records cover fewer days of demand than the disruption you are planning for, plus the time a port backlog takes to clear, that gap is what to act on: bring orders forward, agree an allocation commitment in writing, route critical parts through another gateway or by air, and check the force majeure and allocation clauses in the contract. Size any extra order to the gap you measured, not to the headline ratio.