I have written from time to time over the years about concerns that the number of start-up firms in the US economy has been stagnant (for example, here, here and here). This is a matter of general concern, because start-up firms are often the ones providing a jolt of new goods and services, new jobs, new competition, and new energy in a dynamic economy.

However, there are some preliminary sources of data which suggest that since the end of the pandemic, the rate of new US business start-ups may be on the rise. Here are two of them.

A relatively new source for this information is the Business Formation Statistics published by the US Census Bureau starting in 2019. If a new organization is planning to hire employees, it needs to apply for an Employer Identification Number (EIN) from the Internal Revenue Service, so that the IRS can collect taxes from these employees. Here’s the monthly data on such applications, which has been extended back to 2011.

As you can see, the new applications for EIN numbers dropped during the pandemic recession, then rebounded much higher, but since have settled into a level well above the pre-pandemic levels.

There’s no guarantee that applying for an EIN will actually lead to hiring someone. Thus, the Census Bureau looks over the EIN applications and applies a few criteria to estimate the number of “high-propensity applications”–that is, those where it seems especially likely that hiring will actually happen. For example, if an organization has formed a legal corporation, has an announced date when it expects to begin paying wages, and is in an industry where employees are common like accommodation and food services, construction, manufacturing, retail, professional, scientific or technical services, educational services and healthcare, then it’s likely to be classified as “high propensity.” As you can see, the share of high propensity applications was about 100,000 per month pre-pandemic, but seems to have stepped up to a higher level of 150,000 per month.

There’s some other evidence to back up the belief that something is happening here. The Census Bureau also calculates what is called the Business Dynamics Statistics, constructed from its Longitudinal Business Database. Basically,, this data uses Census data and links the records on individual companies over time. To keep the records of individual companies confidential, the detailed data can only be accessed by qualified researchers through a network of Census Bureau Research Data Centers. Also, this data is available at quarterly intervals and with a longer lag than the EIN data. The most recent data release is up through the end of 2022. But some overall trends are apparent.

If you go back 10 years to 2012, in a given quarter, the number of “establishment births” is about 2.9-3.2% of the total number of existing firms. An “establishment,” in the jargon here, is a new geographical location where a firm is operating, and so it includes both new locations of existing firms as well as brand-new firms. But since the second quarter of 2021 through the end of 2022, the quarterly rate of establishment births has rise to about range of 3.8-4.1% of the total number of existing firms. Meanwhile, the rate of “establishment deaths” hasn’t changed much, other than an upward spike when the pandemic hit in the second quarter of 2020.

I should stress that these patterns are fairly recent, and the US economy is still climbing out of the pandemic. As these sources of data continue to accumulate, along with other sources like the National Report on Early Stage Entrepreneurship in the United States done by the Kauffman Foundation, we’ll know more.

But a reasonable if provisional hypothesis is that one reason the US economy has managed to keep unemployment rates low and to avoid an outright recession since the pandemic is an underlying surge in new start-up firms.