Baby Boomer Business Owners Retiring: What the Data Shows

The claim and where it comes from

The "silver tsunami" argument runs as follows. A large share of private business owners belong to the Baby Boom generation, they are reaching retirement age together, and therefore a wave of owner-operated companies will change hands, close or be sold over the coming decade. The argument is a familiar opening for lower-middle-market, search-fund and independent-sponsor theses.

The owner-age percentages repeated in that argument usually trace back to one federal source: the owner-characteristics tables of the U.S. Census Bureau's Annual Business Survey (ABS). The ABS is a joint project of the National Center for Science and Engineering Statistics (NCSES), within the National Science Foundation, and the Census Bureau, and it "measures business owner demographics", among other topics ^[1]. Other figures that travel with the claim, such as dollar estimates of business wealth "in transition", come from private models, not from a survey that measured them.

This article states what the Census tables measured, separates those measurements from the inferences built on them, and gives a worksheet and checklist for using owner-age data in an investment committee (IC) memo. The structural case for succession-driven deal flow is covered in the founder succession thesis and the succession and ownership guide; this page only checks the numbers.

What Census actually measured

The most widely shared version of the statistic is a Census graphic titled "Business Owners' Ages". It reports that 51 percent of business owners were 55 and older, 43 percent were 35 to 54, and 6 percent were 34 and under ^[2]. Its source line reads "2019 Annual Business Survey, data year 2018". The graphic was published in 2020, so the underlying observations are now eight years old.

The note on the same graphic is the part that is usually dropped. It states that the percentages "do not represent all U.S. business owners" and that the estimates "only represent responding owners for employer businesses (4.1 million)" ^[2].

Who is in the denominator

Four restrictions define the population behind the percentage:

1. Employer businesses only. The ABS methodology states that the survey "excludes the collection of data from nonemployer businesses", and its frame is firms with payroll during the reference year and receipts of $1,000 or more ^[3]. Sole proprietors without staff are outside it.
2. Some industries are excluded. Crop and animal production (North American Industry Classification System codes 111 and 112) and public administration (92) are excluded from sampling ^[3].
3. Owners, not firms. Firms report characteristics for "up to four persons owning the largest percentages in the business" ^[3]. A company with three owners contributes three people to the count. The statistic is therefore the share of owners, not the share of companies, that are 55 or older.
4. Respondent firms only. The owner tables are titled "Owner Characteristics of Respondent Employer Firms" ^[4]. They describe owners of firms that answered, not a count of every owner in the economy.

The newer numbers, checked against the Census tables

Post-2018 owner-age shares circulate on secondary sites, usually without a table reference. Although the owner-characteristics tables are titled by sector, sex, ethnicity, race and veteran status, they also contain an "Age of Owner" topic, coded QDESC=O09 (question OWNRAGE), as the Census table index pages show ^[5].

The Census Data API (api.census.gov) now requires a key for every data query, so the figures below were read on 2026-10-28 through the data.census.gov table service, which serves the published tables without one (pattern: data.census.gov/api/access/data/table?id=<table ID>&g=010XX00US, filtered to QDESC=O09). For each vintage we kept the national row for all sectors (NAICS = 00) and for all owners on every demographic dimension. Each year links to its table. Reference year 2023 is the newest vintage published as of that date.

One naming convention matters. Census names each survey by collection year, so "the 2024 ABS, for example, covers reference year 2023" ^[3]). The table pages confirm the same pattern for the 2022 ABS (reference year 2021) and the 2023 ABS (reference year 2022) (^[6][7].

The "55 and older" column is our arithmetic: the 55-to-64 and 65-or-over owner counts divided by owners reporting an age. Census publishes the two bands separately, with standard errors between 0.1 and 0.5 percentage points on each published share, and does not publish a combined 55-plus figure or its standard error.

Three observations follow from the table:

• The 2018 graphic reconciles. The 2018 table reports 4,114,139 owners reporting an age, matching the graphic's "4.1 million", and the computed 50.8 percent rounds to its 51 percent.
• The circulating 52.3 and 51.3 percent figures are real, but they belong to specific reference years. 52.3 percent is reference year 2021 (the 2022 ABS); 51.3 percent is reference year 2023 (the 2024 ABS). A memo should state the reference year, not the survey name alone.
• Owners who skipped the question are outside the percentage. Each table has an "Item not reported" row (28,288 owners in 2023; 140,835 in 2018) that carries 0.0 percent. The published shares are shares of owners who reported an age.

What the numbers do not say

The tables measure the age of owners. None of the following should be attached to a Census citation:

• Intent to sell. A 67-year-old owner may intend to sell, transfer to family, wind down, or keep working.
• Business value or size. An owner of a two-person firm counts the same as an owner of a firm with several hundred employees.
• Firm counts. Because up to four owners per firm are counted, the table cannot be converted into "X million companies owned by people over 55" without assumptions the table does not support.
• Employment share. We did not find a verified Census figure for the share of employment in firms whose owners are 55 or older, and this article does not state one.
• The whole economy. Nonemployer businesses, the excluded industries and non-respondent firms are outside the universe.

The other Census source commonly used for market sizing, Statistics of U.S. Businesses (SUSB), does not fill these gaps. It covers "U.S. business establishments with paid employees", excludes crop and animal production among other industries, and reports firms, establishments, employment and payroll by enterprise size ^[8]. It has no owner-age dimension, so SUSB counts cannot be multiplied by an ABS owner share to produce "firms with older owners by size band" without stating that this is an assumption. The Census SUSB source page and the Census ABS source page describe each dataset's fields in more detail.

Retirement flows versus ownership stock

A second strand of the claim uses retirement data. Pew Research Center, analysing the monthly Current Population Survey, reported that about 28.6 million Baby Boomers (born 1946 to 1964) were out of the labour force because they had retired in the third quarter of 2020, 3.2 million more than the 25.4 million retired a year earlier. Before 2020 the number had grown by about 2 million a year on average since 2011 ^[9].

These figures describe people leaving the labour force, not which retirees owned a business: context for the cohort's size, not evidence about how many businesses changed hands.

How common is owning a private business at all

The Federal Reserve's Survey of Consumer Finances (SCF) answers a different question. In the 2025 survey, "Ownership of (private) business equity was 15.4 percent in 2025, up almost 1 percentage point from 2022" ^[10]. By age of the family's reference person, 18.9 percent of families aged 55 to 64 held business equity, 15.4 percent of those aged 65 to 74 and 9.1 percent of those 75 or older ^[11].

The unit is the family, not the owner or the firm. The often-repeated "20 percent of families owned a privately held business" is the 2022 figure, which counted a family as an owner "if the family reported owning a business or if either the reference person or the spouse or partner was self-employed"; among those business-owning families, 52 percent owned nonemployer firms and 78 percent employed fewer than five people ^[12]. The 2025 report's narrower definition excludes "the 5 percent of SCF families in which the reference person is self-employed but does not report a business interest" ^[10], so 20 percent (2022) and 15 percent (2025) are not a decline. Either way, the SCF population overlaps only partly with the ABS universe of employer firms, and SCF and ABS figures should never be combined in one ratio.

Why the cliff has not been a cliff

The table supports one narrow observation. Across reference years 2018 to 2023, the share of reporting owners aged 65 or over rose from 20.4 percent to 23.1 percent, while the 55-to-64 share fell from 30.4 to 28.2 percent. The combined 55-plus share stayed between roughly 51 and 52 percent. That pattern is consistent with older owners continuing to own their businesses as they age, rather than a sudden exit of a whole cohort.

It does not establish why. Each vintage describes a different set of respondent firms, not the same owners followed over time, and the ABS does not record exits, sales or closures by owner age. We found no verified primary source measuring the rate at which owners aged 55 and older sell or close their businesses, so whether they are delaying sales, transferring to family or closing quietly remains open.

Modelled estimates of business wealth "in transition" fall in the same category. They are outputs of private models built on assumptions about ownership, valuation and timing, and should be labelled with the model's author and method, not presented as if they had the standing of Census measurements.

Claim vs what the data supports

The worksheet below applies the checks above to common phrasings. The claims are illustrative composites, not quotations from any particular publication.

From age to a screen

Age data can still inform sourcing if it is treated as a hypothesis flag rather than a fact about intent. A short method:

1. Start from a defined universe (sector, geography, size proxy) built from establishment or registry data, not from the ABS share.
2. Collect owner-age proxies per company, each with its source and date: the year the business was founded or acquired, officer names in state registry filings, professional profiles stating career start dates, and published succession announcements.
3. Assign a confidence label to each proxy: stated (the owner's age or retirement plan is written on a retained page), inferred (derived from a founding or graduation year), or unknown.
4. Never convert a proxy into a fact. A founding year of 1988 does not establish that an owner is 60, still owns the company, or wants to sell.
5. Leave unsupported fields open instead of filling them with a national average.

Fictional example. An independent sponsor screens 400 commercial HVAC (heating, ventilation and air conditioning) contractors in one region. For the fictional Quillmere Mechanical, a state registry filing lists the same president since 1991 and the company website states "family-owned since 1991". The analyst records "owner tenure 35 years (inferred, two sources)", age "unknown", intent "unknown". For the fictional Thessaly Peak Climate, a local newspaper profile quotes the owner saying she plans to retire within five years; the analyst records intent as "stated, one source, dated". Only the second record says anything about intent, and it still requires a conversation to confirm. Neither record should be described in the IC memo as "a Boomer-owned business looking to exit".

The ABS also asks when owners acquired the business (YRACQBUS), so a tenure distribution for respondent owners can be pulled in the same way for context ^[5].

Checklist for using these numbers in an IC memo

• Cite the Census table and query, not a secondary article, and give the reference year, not only the survey name.
• State the universe in full: owners of respondent employer firms who reported an age, up to four owners per firm.
• Say "share of owners", never "share of businesses".
• Report the 55-to-64 and 65-or-over bands separately if precision matters; label any combined figure as computed.
• Do not combine ABS shares with SUSB firm counts or SCF family shares without labelling the result as an assumption.
• Do not attach intent, valuation or employment share to an age statistic.
• Label any "wealth in transition" figure as a model, with its author and method.
• Use Pew retirement counts only as cohort context, with their 2020 date.
• For company-level screens, record each owner-age or intent field with source, date and confidence label, and leave unknowns open.

Docket's research records follow the same discipline: each answer quotes a retained page, and fields that cannot be supported stay open.

Sources

1. Business Owners' Ages (graphic) — U.S. Census Bureau, 2020; owner-age shares (51/43/6 percent), the 4.1 million respondent-owner caveat, source line "2019 Annual Business Survey, data year 2018". Accessed 2026-10-28.
2. Annual Business Survey (ABS) — U.S. Census Bureau; ABS measures business owner demographics; joint NCSES and Census project. Accessed 2026-10-28.
3. ABS Methodology — U.S. Census Bureau, page last revised December 12, 2024; employer-only scope, excluded industries, up to four owners per firm, collection-year naming. Accessed 2026-10-28.
4. Census Data API, 2023 ABS variable OWNCHAR — U.S. Census Bureau; table title "Owner Characteristics of Respondent Employer Firms". Accessed 2026-10-28.
5. ABS Characteristics of Business Owners: 2024 Tables — U.S. Census Bureau; 2024 ABS covers reference year 2023; Age of Owner (OWNRAGE) and Year Acquired (YRACQBUS) topics. Accessed 2026-10-28.
6. ABS Characteristics of Business Owners: 2023 Tables — U.S. Census Bureau; 2023 ABS covers reference year 2022. Accessed 2026-10-28.
7. ABS Characteristics of Business Owners: 2022 Tables — U.S. Census Bureau; 2022 ABS covers reference year 2021. Accessed 2026-10-28.
8. data.census.gov table ABSCBO2018.AB1800CSCBO, Age of Owner (QDESC=O09) — U.S. Census Bureau; owner-age counts and shares, reference year 2018. Accessed 2026-10-28.
9. data.census.gov table ABSCBO2019.AB1900CSCBO, Age of Owner (QDESC=O09) — U.S. Census Bureau; reference year 2019. Accessed 2026-10-28.
10. data.census.gov table ABSCBO2020.AB2000CSCBO, Age of Owner (QDESC=O09) — U.S. Census Bureau; reference year 2020. Accessed 2026-10-28.
11. data.census.gov table ABSCBO2021.AB2100CSCBO, Age of Owner (QDESC=O09) — U.S. Census Bureau; reference year 2021 (55-plus computed 52.3 percent). Accessed 2026-10-28.
12. data.census.gov table ABSCBO2022.AB2200CSCBO, Age of Owner (QDESC=O09) — U.S. Census Bureau; reference year 2022. Accessed 2026-10-28.
13. data.census.gov table ABSCBO2023.AB2300CSCBO, Age of Owner (QDESC=O09) — U.S. Census Bureau; reference year 2023 (55-plus computed 51.3 percent; 4,208,231 owners reporting). Accessed 2026-10-28.
14. Statistics of U.S. Businesses (SUSB): About — U.S. Census Bureau; coverage of establishments with paid employees, excluded industries, no owner characteristics. Accessed 2026-10-28.
15. The pace of Boomer retirements has accelerated in the past year — Pew Research Center (Richard Fry), November 9, 2020; retired Boomer counts from the Current Population Survey. Accessed 2026-10-28.
16. Changes in U.S. Family Finances from 2019 to 2022: Evidence from the Survey of Consumer Finances — Federal Reserve Board, October 2023; 2022 broad ownership definition (20 percent) and size mix. Accessed 2026-10-28.
17. Changes in U.S. Family Finances from 2022 to 2025: Evidence from the Survey of Consumer Finances — Federal Reserve Board, October 2026; 15.4 percent of families held private business equity in 2025; definition change. Accessed 2026-10-28.
18. SCF historical tables based on internal data, 2025 dollars (Excel) — Federal Reserve Board, October 9, 2026; Table 9, business equity holding by age of reference person. Accessed 2026-10-28.