Founder Data ReportHalf the Businesses That Closed Last Year Had Been Open Six Years or More
Get clarity — download the free Entrepreneur Survival Guide Summary.
The survival curve is real and it is being read backwards. Census firm-age data on 523,958 closures says surviving buys you a great deal in the first five years and very little after that — and that the floor under mature firms has been rising for a decade.
523,958 American employer firms closed in the year to March 2023. 49.3% of them were five years old or younger. The other 50.7% had been in business six years or more, and 13.6% had been trading for at least twenty-six. That is not a 2023 story: in every one of the eleven years the Census Bureau has published since 2013, between 50.7% and 55.5% of the firms that closed were at least six years old. The survival curve everyone quotes is true and it is being read backwards. Your odds improve with age; they do not stop mattering.
- 523,958 employer firms closed in the year to March 2023. 50.7% of them were six years old or more.
- The annual closure rate falls from 21.0% in year one to 9.7% by year five — and then to only 6.3% by twenty-six years.
- 13.6% of every closure in the country was a firm that had been trading at least twenty-six years.
- The floor is rising. Firms aged 26+ closed at 4.5% in 2013 and 6.3% in 2023, up in nine of the ten steps.
- Closures cost 2.69 million jobs in 2023.
Half the closures are not startups
Split the 523,958 firms that closed by how old they were when they went. 258,117 were between one and five years old. 84,448 were six to ten. 110,391 were eleven to twenty-five. 71,002 — 13.6% of every closure in the country — had been trading for twenty-six years or longer, and 13,422 of those had been in business since before 1978. Firms in their first year are not in this count at all: a business cannot have employees in March and no employees in the March before it was born, so the Census Bureau records no closures at age zero.
Segments are shares of every firm closure in the country in the year to March 2023, and they sum to 100% by construction — the age bands are exhaustive and the identity was checked against the published national total on firms, establishments, employment and closures. The oldest band includes firms already trading in 1978, whose exact age the series cannot resolve.
Source: U.S. Census Bureau, Business Dynamics Statistics, firm-age tables (BDSFAGE), 1978-2023, downloaded from www2.census.gov/programs-surveys/bds/data/ on September 18, 2026. Employer firms only. A firm death is a firm that had employees in the previous March and none in this one.
The danger falls fast, then almost stops falling
The rate is the useful number, not the count. Of the one-year-old firms that were trading when the year began, 21.0% had closed by the end of it. At two years old, 14.9%. By five, 9.7%. That is a fall of 11.3 points, and it is the part of the curve everyone talks about. What nobody mentions is the rest of it. From year five to twenty-six-plus — twenty-one more years of customers, reputation and retained earnings — the rate goes from 9.7% only as far as 6.3%. About one firm in 16 at that age still closed. The only group that does better is the one that has been trading since before 1978, at 4.3%, and that is forty-five years of trading to buy 2.0 points.
Each bar is that age group’s closures divided by the firms of that age that were trading when the year began — the firm count published for the end of the year leaves out the firms that died, so dividing by that instead would overstate every bar. Bars run against a 22% axis and the exact rate is printed on each. Orange marks the first five years, slate everything after, and the colour is the age band and nothing else. Firms aged zero are omitted: they cannot yet have closed.
Source: U.S. Census Bureau, Business Dynamics Statistics, firm-age tables (BDSFAGE), 1978-2023, downloaded from www2.census.gov/programs-surveys/bds/data/ on September 18, 2026. Employer firms only. A firm death is a firm that had employees in the previous March and none in this one.
Four more years of trading takes the annual risk from 21.0% to 9.7%. The twenty-one years after that take it to 6.3%.
And the floor under you is rising
The most uncomfortable number here is not a level, it is a direction. In 2013, 4.5% of firms aged twenty-six or more closed during the year. In 2023, 6.3% did, and the rate was up in nine of the ten year-on-year steps. Mature firms are closing more often than they were a decade ago, in a decade when the number of employer firms grew from 5,064,208 to 5,593,727. One honest note on the chart below: it tracks only the firms aged twenty-six or more whose exact age the series can pin down, and leaves out the group that has been trading since before 1978. Put both groups together and the series runs 4.19% to 5.77% instead — but a shift-share says 23% of that move is the pre-1978 group shrinking as a share of the band rather than anyone’s risk changing, so the narrower series is the honest one to chart.
The vertical axis runs from 4% to 6.5%, not from zero, so that a 2-point move in a small rate is visible; the gridlines are labelled so the scale can be read. Value labels are thinned to every other year because eleven of them do not fit. Each point is that year’s closures among firms aged 26+ divided by that year’s firms aged 26+.
Source: U.S. Census Bureau, Business Dynamics Statistics, firm-age tables (BDSFAGE), 1978-2023, downloaded from www2.census.gov/programs-surveys/bds/data/ on September 18, 2026. Employer firms only. A firm death is a firm that had employees in the previous March and none in this one.
What the three numbers are worth to you
None of this argues that age does not help. It helps enormously, and it helps almost entirely in the first few years. The argument is about what you do with the years after that. If you run a fourteen-year-old business and you have quietly retired the question of whether it could fail, the data says 7.2% of the businesses that age closed during the year to March 2023. The concentration risk, the one customer who is a third of revenue, the supplier with no alternative, the systems only you understand — those do not get safer because the company got older. They get more expensive.
All three figures are that age group’s own closures divided by its own firm count. The first and second are the same measure at different ages, so the difference between them is the whole of what four more years of trading was worth in 2023.
Source: U.S. Census Bureau, Business Dynamics Statistics, firm-age tables (BDSFAGE), 1978-2023, downloaded from www2.census.gov/programs-surveys/bds/data/ on September 18, 2026. Employer firms only. A firm death is a firm that had employees in the previous March and none in this one.
Two limits worth stating. The Business Dynamics Statistics counts employer firms only — if you have no payroll you are not in these 5,593,727, which is why the numbers here are smaller than the thirty-odd million businesses the IRS sees. And a firm death in this data means a firm that had employees in one March and none in the next: an owner who sells, merges or lets the last employee go and carries on alone all count the same way. It is a measure of businesses that stopped employing, which is close to but not identical with businesses that stopped. The jobs are real either way — 2,686,413 of them went with the firms that closed in 2023.
Frequently Asked Questions
How many US businesses closed in 2023?
523,958 employer firms closed during the year to March 2023, according to the Census Bureau's Business Dynamics Statistics. Those closures accounted for 2,686,413 jobs. The figure covers firms with employees only.
Are most business closures new businesses?
No. 50.7% of the firms that closed in 2023 had been in business six years or more, and 13.6% had been trading for at least twenty-six years. In every year since 2013 the six-years-or-older share has been between 50.7% and 55.5%.
What percentage of businesses close each year?
Across all ages, 8.6% of the American employer firms trading in March 2022 had closed by March 2023. The rate depends heavily on age: 21.0% for one-year-old firms, 9.7% at five years, 6.3% at twenty-six or more.
Does a business get safer the longer it survives?
Yes, but almost all of the improvement is early. The annual closure rate falls from 21.0% to 9.7% between year one and year five, and only as far as 6.3% across the following twenty-one years. It never reaches zero.
Are older businesses closing more often than they used to?
Yes. Firms aged twenty-six or more closed at 4.5% in 2013 and 6.3% in 2023, and the rate rose in nine of the ten year-on-year steps.
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.