US business counts diverge by 1.6 million, Fed paper finds
A new Federal Reserve working paper documents a 1.6 million gap between the Bureau of Labor Statistics and Census Bureau reports on U.S. private sector employer establishments by 2023. This significant discrepancy is primarily driven by differences in coverage and how multi-unit firms are counted across the two datasets.
Divergent data, different realities
Since the 1990s, a growing divergence has emerged between the Bureau of Labor Statistics (BLS) and the Census Bureau regarding the number of U.S. private sector employer establishments, reaching approximately 1.6 million by 2023.
This significant gap, documented in a new Federal Reserve paper, stems from two primary drivers.
First, the BLS frame includes a large and increasing number of employers providing services to the elderly and persons with disabilities (NAICS 624120) not captured by the Census Bureau.
Second, many firms are recorded with substantially more establishments in the BLS frame.
These methodological differences lead to a fundamentally altered view of the U.S. establishment landscape, affecting the measured size distribution of businesses and quantitative policy analyses.
Two frames, two views
The divergence between the BLS and Census Bureau establishment counts arises from fundamental differences in how each statistical agency constructs its business frame.
The Census Bureau primarily relies on Internal Revenue Service (IRS) tax filings (Form 941) and periodic surveys, which are used to build the Longitudinal Business Database (LBD).
In contrast, the BLS frame is constructed from state-level unemployment insurance (UI) records and the Multiple Worksite Report (MWR), forming the Quarterly Census of Employment and Wages (QCEW).
The unique linked microdata from the LEHD program, housed within the Census Bureau, allows researchers to compare these distinct methodologies directly.
More than just a statistical quibble
The discrepancies between BLS and Census data are not merely statistical anomalies; they fundamentally alter the understanding of the U.S. establishment size distribution and impact quantitative policy analysis.
This highlights the critical need for data harmonization to ensure accurate economic measurement and consistent policy evaluation.
Without such efforts, policy decisions risk being based on an incomplete or distorted view of business dynamics.