What the U.S. Inventory-to-Sales Ratio Measures

The Census Bureau publishes a monthly Manufacturing and Trade Inventories and Sales report, usually shortened to MTIS.

Its total-business ratio puts the reported inventory stock at the end of the month over the reported sales flow for that month.

In simple form, ratio = month-end inventories ÷ monthly sales. Both sides are dollar values, so the ratio has no dollar unit. A value of 1.30 says the measured stock is 1.30 times the sales recorded during the month.

The calculation is straightforward, but its interpretation depends on the exact table and population. A retail-only ratio and the combined manufacturing-and-trade ratio do not describe the same businesses or stages of distribution.

Census describes a retail inventories-to-sales ratio as a comparison between end-of-month inventory and monthly sales.

It can be viewed as an approximate number of months of sales on hand for that retail series.

For the combined total, treat the result as an aggregate relationship, not a literal promise that all firms have identical stock coverage. {source:censusMtiDefinitions}

The ratio is descriptive. It does not say whether management intended to build stock, whether goods are in the right locations, or whether they can be sold at the reported value. Those questions require other data and business context.

MTIS combines data from three Census surveys: the Manufacturers’ Shipments, Inventories, and Orders Survey; the Monthly Wholesale Trade Survey; and the Monthly Retail Trade Survey.

Together they cover manufacturers, wholesalers, and retailers in the United States. {source:censusMtiOverview} {source:censusMtiDataCollection}

This breadth is helpful when a reader wants a monthly view across business stages. It also means the aggregate is not a measure of only store shelves, consumer purchases, or one industry's supply chain.

The published sales total combines distributive-trade sales with manufacturers’ shipments.

A product can be recorded as it moves through different businesses, so the total is not the same as sales to final consumers or personal consumption expenditures.

The component surveys use different collection methods and units.

Census publishes aggregate statistical estimates, not a census of every shipment or every individual firm’s inventory. Survey coverage and estimation therefore matter when interpreting small changes. {source:censusMtiDataCollection}

The ratio describes the businesses covered by this report. It is not a measure of household pantry stocks, service-sector inventories, or the complete value of every good held anywhere in the economy.

Inventory is measured at a point in time: the value held at the end of the reporting month.

Sales accumulate over an interval: the value recorded across the month. Dividing a point-in-time stock by a monthly flow compares two different kinds of measurement.

If the stock and flow have similar valuation bases, their ratio can be read as a sales-month equivalent. For example, 1.30 means the end-month inventory value equals 1.30 times one month’s sales value.

That shorthand helps people picture relative coverage, but it is not a literal depletion clock.

Sales do not arrive evenly each day, every item does not sell at the same pace, and some inventory may be unfinished, seasonal, committed, or in transit.

Census definitions value manufacturing and retail inventories at cost, while their sales measures reflect sales values.

The dollar amounts therefore do not necessarily share an identical valuation basis, another reason not to read the ratio as a physical stockout clock. {source:censusMtiDefinitions}

An aggregate also combines businesses with very different turnover speeds. A durable-goods manufacturer, a wholesaler, and a grocer can have distinct production and delivery cycles even if they contribute to one total.

The denominator is monthly sales, not a forecast of future sales.

Treating a ratio of 1.30 as a guarantee of exactly 1.30 months before stock runs out silently assumes that sales continue at a fixed pace and that every dollar of inventory is equally available.

MTIS reports dollar estimates in current dollars, not a constant-dollar measure adjusted to remove price changes. {source:censusMtiDefinitions}

A change in the ratio can therefore reflect changes in quantities, prices, or the mix and valuation of goods, as well as changes in sales. {source:censusMtiDefinitions}

The report also publishes seasonally adjusted data. Census uses adjustment factors to account for recurring seasonal patterns. {source:censusMtiDefinitions}

For sales, the factors also address trading-day and holiday differences. The Bureau cautions that these adjustments are estimates and may become less precise when patterns change. {source:censusMtiDefinitions}

When reading the headline total-business ratio, use the same adjustment basis as the published series. Do not divide a seasonally adjusted inventory estimate by an unadjusted sales estimate and treat the result as the official ratio. {source:censusMtiDefinitions}

The source estimates can be revised as survey information and annual benchmark data are incorporated. A percentage or ratio from a later release may not exactly match the number first reported for that month. {source:censusMtiDefinitions} {source:censusMtiOverview} {source:censusMtiDataCollection}

Record the data month, release vintage, adjustment status, and level of aggregation when comparing readings. This makes it possible to tell a genuine change in the series from a change in the data version or definition. {source:censusMtiDefinitions}

Assume a fictional group of businesses has $520 million of inventory at the end of a month and $400 million in sales during that month. The illustrative ratio is $520 million ÷ $400 million = 1.30.

Now keep the stock at $520 million but increase monthly sales to $500 million. The ratio becomes 1.04. Nothing in this calculation says that the inventory stock was sold down; only the sales denominator changed.

Alternatively, keep sales at $400 million and let month-end inventory rise to $600 million. The ratio becomes 1.50. This time the numerator changed while the sales flow stayed fixed.

These figures are invented solely to demonstrate the arithmetic. They are not Census estimates, a current reading, a forecast, or a claim about firms’ intentions.

The example shows why it is helpful to inspect the numerator and denominator separately.

A falling ratio can arise from faster sales even when inventory is unchanged; a rising ratio can arise from slower sales even when inventory is unchanged.

<!-- learn:illustration -->

A rising ratio means measured inventories have increased relative to monthly sales, based on the selected series and adjustment basis. It can be a reason to ask whether stock is building, sales are slowing, or both are happening.

A falling ratio means inventories have decreased relative to sales. It may reflect stronger sales, lower stocks, or a combination. The ratio alone does not tell which of those changes occurred.

Business decisions can produce similar readings for different reasons. Firms may increase stock before an expected busy season, hold more goods to manage delivery risk, or find that sales fell faster than production adjusted.

The same ratio can also hide different conditions across industries. A total may be steady while one sector’s ratio rises and another’s falls.

Looking at the components can explain how an aggregate was formed, but it still cannot reveal a particular firm’s plans.

For that reason, describe the ratio as a relative stock measure. Avoid saying that it proves oversupply, predicts a recession, identifies the cause of weak demand, or guarantees a specific number of months before goods sell.

The ratio does not measure physical units. It compares dollar values, and the underlying goods can differ in price, quality, stage of production, and sales speed.

It does not isolate prices from quantities. Since the published dollar estimates are not converted to constant dollars, higher prices can change inventory and sales values even if the number of goods is unchanged.

It does not show whether stock is finished merchandise ready for a customer. Manufacturing inventories include raw materials, work in process, and finished goods; those stages serve different purposes and are not equally available for sale.

The aggregate does not show how inventory is distributed among locations, whether goods are already committed to a buyer, or whether they are suitable for the current demand mix. Such details are outside what this total ratio can establish.

It also is not the same as inventory investment in GDP. {source:beaChangePrivateInventories}

MTIS compares an end-month inventory value with monthly sales; BEA inventory investment tracks the change in private inventory stocks over a period in the national accounts. {source:beaChangePrivateInventories}

One is a relative stock-to-sales measure and the other is a production-accounting flow.

Start by naming the series: total business, retail, wholesale, or manufacturing. Keep the month, seasonally adjusted status, and release vintage consistent.

Then check whether the ratio moved because inventories changed, sales changed, or both.

Next compare a sector with its own history and seasonal pattern. A ratio that is typical for one industry can be unusual for another because production cycles, shelf life, lead times, and accounting practices differ.

If you want to understand production rather than inventories relative to sales, see industrial production and capacity utilization.

If you want to compare consumer-facing retail receipts with household spending in the national accounts, see retail sales and PCE consumer spending.

For the separate GDP question, inventory investment and final sales explains why a period’s net stock change enters GDP accounting.

That guide does not turn the MTIS ratio into a measure of real output or consumer demand.

Use the ratio as one descriptive clue and look at its components before offering an explanation. It can organize a question about stock relative to sales; it cannot settle that question by itself.