Skip to main content

Census BFS / July 2026

Business Formation Observatory

Follow monthly Census business applications and employer-business formations across the nation, states, and industry groups. Applications, observed formations, and projections are different measures.

Monthly panel
115
January 2017 - July 2026
Geographic field
51
50 states + District of Columbia
Industry field
20
published national buckets
Series custody
7
57,960 auditable cells
Business applications
578,926
+8.1% from June / seasonally adjusted EIN applications
Projected formations
29,959
+0.7% from June / modeled employer formations within four quarters
High propensity
151,857
+1.4% from June / application subset, not observed firms
Actual-formation edge
December 2022
later cohorts remain projected in the spliced path

Monthly formation desk

Business formation field

Selected
United States
Series
BA_BA
Focus
Jul 2026
Value
578,926
Observed
BA_BA / national / trajectory

Applications trajectory

Selected Employer Identification Number applications included in the Census BFS business-application universe.

Business Applications for United StatesMonthly source counts from Jan 2021 through Jul 2026. The chart shows the U.S. source series.0144.7K289.5K434.2K578.9KUnited States / source countJan 2021May 2022Oct 2023Feb 2025Jul 2026
United StatesCounts are not rates. Broken paths preserve unavailable source cells.

Source interpretation

One cohort, several distinct measures

Business Formation Statistics describe selected EIN applications and employer formations traced from those application cohorts. An application is not an operating business, a projected formation is not an observed startup in the publication month, and a state or industry count is not a rate. These series do not measure concentration, market power, legal antitrust markets, welfare, competitive effects, or policy causation.
01

Applications

Business Applications are a selected subset of IRS Form SS-4 EIN applications. They indicate business-initiation activity, not operating firms or employer startups.

02

Application subsets

High-propensity, planned-wage, and corporate applications are overlapping nested constructions. They must not be added to Business Applications or to one another.

03

Formation outcome

Actual four-quarter formations use first payroll tax liability to identify employer businesses arising from an application cohort after the outcome window matures.

04

Projection and splice

Projected formations are econometric estimates. The spliced series uses actual formations through December 2022 and projected values at the current edge.

Evidence boundary

Timely formation signal, not an antitrust conclusion

Census BFS and Business Dynamics Statistics are distinct systems. BFS identifies formations through first payroll tax liability linked to EIN applications, while BDS identifies annual employer-business entry using employment and a March reference period. Their counts and timing are not interchangeable.