U.S. Business Applications Fall 7.8% in August to 531,728

U.S. business applications fell 7.8% in August to a seasonally adjusted 531,728, reversing much of the sharp increase recorded a month earlier. The decline pulled the monthly total back from July’s elevated level after a strong midsummer rise in filings.

The monthly figure measures applications for Employer Identification Numbers that meet the Census Bureau’s criteria for its Business Application series. It is a measure of business initiation activity, not a count of companies that opened their doors, hired workers or survived long enough to become established employers.

The August figures are part of the U.S. Census Bureau’s Business Formation Statistics, a monthly data product built from qualifying EIN applications and subsequent payroll-tax records. Census uses those records to track which applications eventually become employer businesses and to produce estimates of likely employer formations.

August reverses much of July’s jump

The drop came after a particularly strong July. Census had reported 578,926 seasonally adjusted applications for that month, an increase of 8.1% from June. The August decline therefore marks a clear reversal in the latest monthly direction rather than an extension of July’s increase.

Business Formation Statistics can be revised as newer information is incorporated and seasonal factors are updated. That means figures from an earlier monthly release should not always be used to reconstruct the percentage change printed in a later vintage. The latest report’s month-over-month move is the appropriate measure for assessing August.

The broader series has also changed markedly from its pre-2020 pattern. Monthly applications moved to a higher range during the pandemic period and have remained elevated compared with much of the late 2010s. The series can still be volatile from month to month, which makes a single decline more useful as a near-term signal than as evidence of a lasting shift in business creation.

That distinction is especially important because an EIN application can be associated with a genuine future employer, a nonemployer business or an effort that never develops into an operating firm. Census therefore publishes narrower application categories with characteristics historically associated with a greater likelihood of becoming payroll businesses.

An application is not the same as a new employer

The core Business Application series is built from IRS Form SS-4 filings for EINs after specified exclusions. Census removes filings such as those connected with tax liens, estates and trusts, certain financial filings, applications outside the 50 states and the District of Columbia or without usable geographic information, and some industry categories with very low rates of transition into employer businesses.

Census also publishes High-Propensity Business Applications, which have characteristics associated with a higher probability of becoming an employer. Another subset identifies applications that report a planned date for first wages, while a separate series covers applications from corporations. Those measures allow readers to distinguish the broad volume of startup-related filings from applications that more closely resemble future payroll businesses.

The formation side of the report goes further by using the first appearance of payroll-tax liabilities to identify actual employer business formations. Because that process takes time, Census publishes projections for how many employer businesses are expected to emerge from each month’s application cohort within four or eight quarters.

For July, the Census Bureau projected 29,959 employer business formations within four quarters from that month’s applications, up 0.7% from June. The projected figure was far smaller than the total number of July applications because only a fraction of qualifying EIN filings are expected to become employer firms within the forecast window. A rise or fall in total applications therefore should not be read as an equal change in the number of businesses that ultimately begin operating with employees.

The Census methodology is designed to make that distinction explicit. Applications provide an early indication of intent, while employer formations rely on later tax information showing that a firm has actually incurred payroll-tax liability. The two measures describe different stages of the business-creation process and can move differently over short periods.

The next release will test whether the pullback persists

The August decline leaves the near-term direction less clear after July’s jump. A single weaker month can reflect ordinary volatility in filing activity, seasonal adjustment and changes across industries or regions. Evidence of a broader slowdown would require additional weak readings or deterioration in the narrower high-propensity and employer-formation measures.

For economic monitoring, the attraction of Business Formation Statistics is their speed. The data arrive shortly after the reference month and can provide an early view of changes in entrepreneurial activity before slower measures of firm births and employment are available. The trade-off is that applications sit near the beginning of the process, so they are better treated as an upstream indicator than as a direct count of newly operating businesses.

Census publishes the monthly series for the nation, Census regions and states, along with national industry detail. It also provides seasonally adjusted and not seasonally adjusted measures. Those breakdowns can help show whether a national movement is concentrated in particular parts of the economy or is more widely shared.

The next scheduled Business Formation Statistics release covers September 2026 and is due October 14 at 10:00 a.m. Eastern Time. That report will provide the next comparison point for determining whether August mainly reversed July’s unusually strong reading or marked the start of a softer stretch in U.S. business-application activity.