Key finding: The U.S. Census Bureau reports 326 firms in Iron and Steel Mills and Ferroalloy Manufacturing in 2023. Of those, 150 had no paid employees.
Those 150 firms made up 46% of all firms in the group.
But they accounted for only 0.02% of sales, value of shipments, or revenue.
Census calls firms with no paid employment or payroll nonemployer firms.
The 150 nonemployer firms reported $27.4 million in the Census dollar measure.
The total for all 326 firms was about $123.1 billion.
Employer and Nonemployer Firms
| Firm type | Firms | Share of firms | Sales, value of shipments, or revenue | Share |
|---|---|---|---|---|
| Nonemployer | 150 | 46.0% | $27,401,000 | 0.02% |
| Employer | 176 | 54.0% | $123,027,197,000 | 99.98% |
| Total | 326 | 100% | $123,054,598,000 | 100% |
Source: U.S. Census Bureau, 2023 Nonemployer Statistics by Demographics. NAICS 3311 covers Iron and Steel Mills and Ferroalloy Manufacturing.
What the Data Show
Firm count and economic share tell very different stories.
Nonemployer firms made up almost half of the firm count.
Yet their share of the Census dollar measure was only 0.02%.
Employer firms made up 54% of the firms.
They accounted for 99.98% of the dollar measure.
This does not mean employer firms made 99.98% of U.S. steel.
The Census measure is sales, value of shipments, or revenue. It does not measure physical steel output.
Method
RecurSave used Census data for NAICS 3311.
This industry is called Iron and Steel Mills and Ferroalloy Manufacturing.
The figures cover 2023.
Census combines two sets of data in this table:
- employer firm data from the Annual Business Survey;
- nonemployer firm data from NES-D.
RecurSave calculated the share of firms like this:
Nonemployer firms ÷ total firms × 100
The share of the dollar measure was calculated like this:
Nonemployer dollar value ÷ total dollar value × 100
The percentages use the published Census values before rounding.
Limitations
Census defines a nonemployer business as a business with no paid employment or payroll.
That does not mean the owner does no work. It also does not mean the business is inactive.
Employer and nonemployer data come from different Census systems.
Some firms can move between the two groups during a year. Census says this can cause some firms to be counted twice in the combined data.
Census estimates this issue affects about 92,000 firms across the United States. It does not give a figure for NAICS 3311 alone.
Census also rounds some firm counts.
The percentages here use the published counts.
The $27.4 million nonemployer value also has a G flag. Census says this means disclosure protection changed the value by less than 2%.
The finding shows a difference in firm count and economic share.
It does not explain why that difference exists.
Why the Finding Is Useful
A firm count does not show how much economic activity those firms represent.
That matters here.
Firms with no paid employees made up 46% of the published firm count.
Their share of the published dollar measure was only 0.02%.
Putting both figures side by side makes the difference clear.