The 524,000-closure claim checks out

A post making the rounds this week claimed that in 2023, roughly 524,000 U.S. employer businesses closed outright rather than selling, and that more than 11,000 of them were mature firms, eleven-plus years old, with 10 to 99 employees, the profile a search fund or independent sponsor actually wants to buy. Numbers like that travel fast and get repeated faster, usually without anyone checking the source. We checked it.

The method was simple: download the Census Bureau's Business Dynamics Statistics tables directly, the same public microdata the claim was supposedly built from, and recompute every figure ourselves rather than trust a screenshot. Two tables did the work: the economy-wide time series for the headline number, and the two-way firm-age-by-firm-size breakdown for the mature-firm segment, summing firmdeath_firms across the age buckets 11 to 15, 16 to 20, 21 to 25, 26-plus, and left-censored (firms that predate the series' 1978 start, and are therefore 11-plus by construction), against the size buckets 10 to 19 and 20 to 99 employees.

The total held exactly: 523,958 employer-firm deaths in 2023, measured March to March. That is the highest reading in the 46-year history of the series, edging out the prior peak of 511,115 set in 2009, the worst year of the Great Recession. Out of roughly 5.6 million employer firms operating in the U.S. that year, about one in eleven closed.

The mature-firm number held too: 11,288 firms aged eleven-plus years with 10 to 99 employees closed in 2023, up 48.8% from the 7,584 that closed in 2010 and 25.1% above the 2015-to-2019 pre-pandemic average of 9,024. Those 11,288 firms employed roughly 493,000 people at the moment they shut down, work at roughly 31 businesses a day that fit a serious acquirer's profile.

A business that sells and keeps operating under a new owner is not counted as a death. These 523,958 firms really did close instead of sell.

What the original post left out, and what actually strengthens its case once you understand it, is the definition Census uses. A BDS firm death requires every establishment the firm owns to cease operations and employment to fall to zero. A business that sells and keeps operating under new ownership is explicitly not counted as a death. So this is not a number contaminated by ordinary M&A activity. These 523,958 firms, and the 11,288 mature ones inside that total, really did shut their doors instead of finding a buyer. That is exactly the framing the original claim used, and it turns out to be the correct one.

The one place the post oversimplified was the record claim. Look at the mature-firm series across all 46 years and 2002 stands out as an outlier at 14,100, well above every neighboring year. That is not a real spike in closures. It is a documented processing artifact in the underlying Longitudinal Business Database: Economic Census years, which end in 2, retime how firm births and deaths get recorded, producing a spurious bump that Census and NBER methodology papers on the LBD redesign have already flagged. Exclude that one artifact year and 2023 is still the record, the highest genuine reading the mature-firm segment has ever posted. Include it, and 2023 is merely the second-highest. Either way the underlying claim, more mature businesses are closing than at any point in the modern series, holds.

Zooming out to the decade helps explain why 2023 landed where it did rather than treating it as an isolated spike. Mature-firm closures ran 7,722 in 2013, dipped to 7,471 in 2014, then climbed almost every year after: 8,254 in 2015, 8,580 in 2016, 9,624 in 2017, 9,207 in 2018, 9,456 in 2019. The one year that breaks the intuitive pattern is 2020: with PPP loans and eviction moratoria propping up small businesses through the worst of the pandemic, you would expect closures to fall. Instead the mature-firm segment posted 9,967 closures, its highest reading up to that point, ahead of even the pre-pandemic 2019 figure. 2021 eased slightly to 9,417, then the climb resumed hard: 10,454 in 2022, 11,288 in 2023, an 8.0% jump in a single year. If that slope held, the 2024 vintage, expected in late September 2026, would print somewhere around 12,200.

Within the mature-firm total, the split by size matters for anyone actually shopping in this segment. Of the 11,288 closures in 2023, 7,164 were on the smaller end, 10 to 19 employees, and 4,124 were larger, 20 to 99 employees. The larger band is the one where financing, staffing complexity, and succession problems compound fastest, and it is also the band where a well-capitalized buyer has the most room to add value simply by professionalizing operations a departing owner-operator never got around to.

None of this proves every one of those 11,288 businesses was a good acquisition target. Census does not track why a firm closed, only that it did. Some were businesses nobody wanted at any price. But the scale is real, verified against the primary source rather than taken on faith, and it is the segment we built this database to track.