Factory orders is the common name for a monthly US economic report on the value of new orders received by manufacturers, together with their shipments, unfilled orders and inventories. It is published by the US Census Bureau, whose formal title for it is the full report on Manufacturers' Shipments, Inventories and Orders, often shortened to the M3 survey. Economists and investors use it to judge demand for manufactured goods and the direction of manufacturing output in the months ahead.
The report covers both durable goods, meaning items expected to last three years or more, and non-durable goods such as food, fuel, chemicals and paper. It follows about a week after the advance report on durable goods, which gives an early reading for roughly half of the total. Similar statistics exist elsewhere, and monthly German factory orders are among the most closely watched figures in the euro area.
What the report contains
The Census Bureau surveys manufacturers across all industries and publishes four main series, each in current dollars and with seasonal adjustment.
- New orders. The value of orders received during the month, net of cancellations. This is the headline number, reported as a percentage change from the previous month.
- Shipments. The value of goods delivered, which corresponds to sales and feeds into estimates of current production.
- Unfilled orders. The backlog of orders received but not yet shipped. A rising backlog suggests that factories will stay busy.
- Inventories. The value of materials, work in progress and finished goods held by manufacturers, as described under inventory.
For most non-durable industries the survey does not collect orders separately. Such goods are usually shipped from stock within days, so new orders are taken to equal shipments.
Timing and revisions
The report is released about five weeks after the end of the month it covers, normally at 10:00 a.m. Eastern Time, and appears on every economic calendar. Because the durable goods portion has already been published, the new information is limited to non-durable goods and to revisions of the durable figures. For that reason the release tends to move markets less than the advance durable goods report or survey-based measures such as the ISM Manufacturing Index, both of which arrive earlier.
Monthly figures are revised in the following month and again in annual benchmark revisions. The figures are not adjusted for inflation, so part of any rise in a period of rising prices reflects higher prices, not higher volumes.
Ratios and a worked example
The monthly change in new orders is calculated as:
Here is new orders in the latest month and is new orders in the month before. If orders rise from 584 billion dollars to 590 billion dollars, the change is percent.
The report also gives the inventories-to-shipments ratio:
With inventories of 860 billion dollars and monthly shipments of 590 billion dollars, the ratio is . Manufacturers are holding stock equal to about 1.46 months of sales. A rising ratio can mean that goods are piling up unsold, which often comes before cuts in production.
How markets read the data
Analysts look past the total to several sub-series. Orders for civilian aircraft and defence equipment are large and irregular, so a single airline order can swing the monthly total. Figures excluding transportation give a steadier view. Orders for non-defence capital goods excluding aircraft, often called core capital goods, are treated as a guide to business capital expenditure, and shipments of the same category enter the official estimate of equipment investment in gross domestic product.
A result well above or below the consensus estimate can affect share prices, the bond market and the dollar, mainly through its effect on expectations for growth and for policy at the Federal Reserve. Strong orders are normally read as favourable for shares of industrial companies and unfavourable for bond prices.
Limitations
Manufacturing is a small share of the US economy compared with services, so the report covers a limited part of total activity, though a part that moves strongly with the business cycle. The data are volatile from month to month, are subject to sizeable revision and are published late. Analysts therefore read them alongside industrial production, purchasing managers' surveys and other economic data, and prefer three-month averages to single readings.