The Census Bureau’s November 12 release puts two different business measures beside each other. October brought 501,609 business applications, a seasonally adjusted 6.0% decline from September. The agency also projected 27,867 employer businesses to emerge within four quarters from October’s application cohort, 1.8% fewer than the comparable September projection. The first number counts tax-ID applications; the second estimates future employer formation. Neither is a count of all businesses that opened their doors in October. U.S. Census Bureau U.S. Census Bureau
An application is an administrative signal, not a storefront count
Census defines a business application as an application for an Employer Identification Number (EIN) made predominantly for business purposes. The series is based on IRS Form SS-4 filings. It is a timely indicator of people and organizations taking a formal step, but an application alone does not establish that the applicant began operating, hired anyone, earned revenue, or even became a lasting firm. U.S. Census Bureau U.S. Census Bureau
The coverage has boundaries too. Businesses that operate without an EIN can be absent, while the statistics exclude applications in specified industries and public entities. So 501,609 is neither a census of every new venture nor a direct headcount of entrepreneurs. It is a defined administrative series. The distinction matters whenever a headline translates “applications” into “startups” without describing the measurement step in between. U.S. Census Bureau
The formation projection looks ahead from the application cohort
The release’s projected-formation figure asks a different question: how many new businesses with payroll tax liabilities are expected to originate from applications filed during a particular month within the next four quarters? For the October cohort, Census put that projection at 27,867, down 1.8% from its September-cohort projection. It is explicitly forward-looking; it does not estimate how many firms formed during October itself. U.S. Census Bureau U.S. Census Bureau
This is why the application decline of 6.0% and projected-formation decline of 1.8% should not be read as competing estimates of the same event. The application measure counts incoming filings. The projection estimates a later outcome for a cohort, using a series designed to capture new employer businesses. Payroll evidence also arrives on a different schedule: Census identifies employer formations through first payroll liability, and its projection model weights applications by their estimated likelihood of becoming employers. That lag and modeling step are why an application count cannot immediately answer how many applicants eventually become employers. U.S. Census Bureau
A one-month comparison carries seasonal and revision context
October’s application total was seasonally adjusted, and its stated comparison is with September. Seasonal adjustment tries to account for recurring calendar patterns so that month-to-month changes are easier to interpret. It does not turn the estimate into a complete or final account of business activity. The release also labels statistical significance as not applicable or not measurable for these percentage changes, so the reported 6.0% decline should not be promoted into a test of a lasting shift by itself. U.S. Census Bureau
There is another detail in this release: Census says its latest annual update revised underlying industry codes for the prior five years where better administrative information was available, updated formation-series information using its Business Dynamics Statistics program, and recomputed seasonal factors. Those updates can change seasonally adjusted historical values as well as the latest month; Census says seasonal factors are recomputed annually, and adjusted application series are revised. U.S. Census Bureau A comparison with September in this edition should therefore be understood as the current published comparison, not necessarily the number a reader saw in an earlier vintage. U.S. Census Bureau
How to read the headline without overreading it
A careful summary has three parts: the reference month (October), the publication date (November 12), and the unit being counted or projected. One concise reading is: “Census reported 501,609 seasonally adjusted EIN business applications for October, 6.0% fewer than in September; it separately projected 27,867 new payroll businesses within four quarters from October’s applications, 1.8% below the September-cohort projection.” The wording is less dramatic than calling either number an immediate startup count, but it preserves what the data actually say. U.S. Census Bureau
For readers following hiring or local business conditions, the next useful question is whether this application activity later appears in employer formation and payroll measures. These statistics can help show an early pipeline and its estimated transition toward employer status. They cannot tell a household whether to start a business, predict a particular firm’s prospects, or establish that the national economy is accelerating or slowing on their own. The release is a timely signal with a specific scope, a projection horizon, and a revision process—not a verdict on every new business.
Sources & further reading
- Business Formation Statistics, October 2026 ↗U.S. Census Bureau
- Business Formation Statistics FAQs ↗U.S. Census Bureau
- Business Formation Statistics Methodology ↗U.S. Census Bureau
- Business Formation Statistics ↗U.S. Census Bureau