Founder Data ReportAmerican Workplaces Keep Getting Smaller. The Share With Fewer Than Five People Has Risen Every Year Since 2017.
Get clarity — download the free Entrepreneur Survival Guide Summary.
The number of places people work has grown far faster than the number of jobs since 2017, and the share of workplaces with fewer than five people has risen in every single year of the run — through a pandemic that cut employment and never once stopped the count of workplaces rising.
Every year the Bureau of Labor Statistics sorts the country’s private workplaces by the size of their payroll. Lined up from 2017 to 2025, the series moves in one direction and does not turn around. The share of workplaces with fewer than five people rose from 62.1% to 67.1% — up in every one of the eight year-on-year steps, with no reversal. The average workplace shrank from 12.72 people to 11.27. The cause is not job losses: over the same span the number of workplaces grew 23.8% while the number of jobs grew 9.6%.
- The share of private workplaces with fewer than five people rose from 62.1% in 2017 to 67.1% in 2025.
- It rose in all eight year-on-year steps. There is no year in the run where it fell.
- The average workplace went from 12.72 people to 11.27.
- Private workplaces grew 23.8% over the span while private-sector jobs grew 9.6% — that gap is the whole mechanism.
- The number of workplaces with fewer than five people grew 33.7%, faster than the total, so this is genuine growth at the bottom rather than larger places sliding down.
- The pandemic accelerated the trend but did not create it: the share rose 1.61 points over the three years before it and 1.78 points over the four years after.
One direction, eight years, no reversal
Size-class detail is published once a year, for the first quarter, which makes this an unusually clean series: the same quarter, the same definitions, nine times. The share of private workplaces with fewer than five people on the payroll reads 62.1% in 2017 and 67.1% in 2025, a rise of 4.98 points. More telling than the endpoints is the path between them: it rose in every one of the eight steps.
The average workplace fell from 12.72 people to 11.27 over the same period. That line is not as clean — it drops sharply in 2021 and then partly steadies — and the reason is worth stating rather than smoothing over. In the first quarter of 2021 employment was still well below its pre-pandemic level, so the same workplaces counted fewer people. That is a real effect on the average and it is visible in the chart.
Bars run on a shared axis to 14 people. Figures are average monthly employment across the first quarter divided by the number of establishments in that quarter. The sharp step down in 2021 is the pandemic: employment fell that quarter while the number of workplaces kept rising. Unlike the share of workplaces under five people, this line does not fall every year — it is flat in two years and ticks up slightly in 2025, when employment grew marginally faster than the count of workplaces.
Source: U.S. Bureau of Labor Statistics, Quarterly Census of Employment and Wages, private ownership, all industries, United States, first quarter of each year 2017-2025, by establishment size class, retrieved from the BLS open CEW data API on September 12, 2026. Size-class detail is published for the first quarter only. QCEW counts establishments — single physical locations — not firms.
The count of workplaces never stopped rising
The arithmetic behind a falling average is not complicated: either the jobs went away or the places multiplied. Here it is decisively the second. Between 2017 and 2025 the number of private workplaces rose 23.8%, from 9,420,257 to 11,657,676, while private-sector employment rose 9.6%. Workplaces grew roughly two and a half times as fast as jobs, and an average whose denominator outruns its numerator by that much has to fall across the span — whatever it does in any single year.
There is a second reading of a rising small-business share that has to be ruled out before the first one can stand, and the data rules it out. A share can rise simply because larger workplaces shed staff and drop into a smaller band — a compositional illusion rather than new activity. If that were the story, the under-five count would grow no faster than the total. It grew 33.7% against 23.8% for all workplaces. The bottom of the distribution is not just absorbing arrivals from above; it is where most of the growth is happening.
All three bars run on one axis to +40%, so they can be read against each other. This is the arithmetic of the shrinking average: the number of workplaces grew roughly two and a half times as fast as the number of jobs, and the smallest workplaces grew faster still.
Source: U.S. Bureau of Labor Statistics, Quarterly Census of Employment and Wages, private ownership, all industries, United States, first quarter of each year 2017-2025, by establishment size class, retrieved from the BLS open CEW data API on September 12, 2026. Size-class detail is published for the first quarter only. QCEW counts establishments — single physical locations — not firms.
The pandemic sped it up. It did not start it.
The single largest year-on-year move in the series is 2020 to 2021, when the under-five share jumped 1.59 points — about three ordinary years of movement in one. It would be easy to read the whole trend as a pandemic artifact, and easy to be wrong. Cut the series at that step and it runs the same way on both sides: 1.61 points over the three years before, 1.78 points over the four years after. The direction predates 2020 and survives it.
One more fact sits underneath all of this and is easy to miss. Across the whole run, including the pandemic year in which private employment fell below its 2017 level, the number of private workplaces rose every single year. Jobs took a year off. The count of places doing business never did.
Both years run on one axis to 70% of all workplaces. Slate is 2017, orange is 2025. Only the smallest band grew its share; the other two both gave ground, which is what a distribution shifting downward looks like.
Source: U.S. Bureau of Labor Statistics, Quarterly Census of Employment and Wages, private ownership, all industries, United States, first quarter of each year 2017-2025, by establishment size class, retrieved from the BLS open CEW data API on September 12, 2026. Size-class detail is published for the first quarter only. QCEW counts establishments — single physical locations — not firms.
The country keeps adding places to work faster than it adds jobs. That is not a recession signal and it is not a boom signal. It is a change in the unit the economy is built from.
What a founder should take from a trend line
Be careful about what this does and does not say. It is a count of workplaces, not of companies, so some of the growth is large firms opening more and smaller locations rather than founders starting businesses. It says nothing about whether any of these workplaces are profitable, and nothing about whether the people in them are better off. A trend in the size of the unit is not a verdict on the quality of the unit.
What it does say is that the small operation is not a residual category waiting to be consolidated away. In every year of this run — ordinary ones and a catastrophic one — the economy has arranged itself into more pieces, and a larger share of those pieces has been tiny. If you are running something with three people in it and treating that as a phase to be exited as quickly as possible, the structural trend is not pushing against you. It has been moving your way for the better part of a decade.
Source: U.S. Bureau of Labor Statistics, Quarterly Census of Employment and Wages, private ownership, all industries, United States, first quarter of each year 2017-2025, by establishment size class, retrieved from the BLS open CEW data API on September 12, 2026. Size-class detail is published for the first quarter only. QCEW counts establishments — single physical locations — not firms.
Frequently Asked Questions
Are small businesses becoming more common?
By workplace count, yes. The share of U.S. private establishments with fewer than five employees rose from 62.1% in 2017 to 67.1% in 2025, according to the BLS Quarterly Census of Employment and Wages, and it rose in every one of those eight year-on-year steps.
What is the average number of employees per business, and is it falling?
The average U.S. private establishment employed 11.27 people in the first quarter of 2025, down from 12.72 in 2017. It is falling because the number of establishments has grown 23.8% over that period while employment has grown 9.6%.
Is this just a result of the pandemic?
No. The 2020-to-2021 step is the largest in the series at 1.59 points, but the same trend runs on both sides of it: the under-five share rose 1.61 points in the three years before the pandemic and 1.78 points in the four years after it.
Are businesses getting smaller, or are big companies just shedding staff?
Mostly the former. If a rising small share simply reflected larger establishments dropping into smaller size bands, the under-five population would grow no faster than the total. It grew 33.7% between 2017 and 2025, against 23.8% for all private establishments.
Does this count businesses or locations?
Locations. QCEW counts establishments — economic units at a single physical location — so a firm that opens a second site adds a second establishment. Part of the growth in workplaces is therefore existing companies opening more and smaller sites, not only new businesses forming. The series also excludes businesses with no employees entirely.
Published by The Lonely Entrepreneur — the community and coaching platform for entrepreneurs who are building alone. lonelyentrepreneur.com
This article is for educational purposes and is not a substitute for professional financial or legal advice.