Podcast
Important Information:
Has the Wave of New Businesses Crested?
Important Information:
Chen Yeh provides an update on the longevity and broader impacts of the surge in start-ups during the COVID-19 pandemic. Yeh is a senior economist at the Federal Reserve Bank of Richmond.
Related Links
Transcript
Tim Sablik: My guest today is Chen Yeh, a senior economist at the Richmond Fed. Chen, welcome back to the show.
Chen Yeh: Thanks for having me, Tim. It's always great to be back.
Sablik: Today, we're going to be talking about the surge of new business activity following the COVID-19 pandemic. Longtime listeners might remember that you were on the show back in January of 2023 to talk about this topic, and today, we've got some more data and updates to share with our audience.
Before we get into all of that, could you provide a quick reminder of what happened to startup activity during and after the pandemic?
Yeh: Yeah, of course.
When we talk about startup activity, we're actually talking about business applications. These are applications that the IRS receives for so-called Employer Identification Numbers. The reason why I'm making this distinction is because, despite having the word "employer" in them, it's not obvious these startups will actually employ people.
In a nutshell, what happened was there was a huge increase in these business applications at the start of the pandemic. This increase some people also refer to as the pandemic startup surge. Just to give you some numbers, first quarter of 2019 these EIN applications increased by more than 65 percent. At its peak, the IRS was receiving up to one and a half million EIN applications in a given quarter.
What is more interesting is that these businesses, as I was saying before, it's not clear whether they actually will employ people. But there are entrepreneurs that are filing these applications that indicate whether they're planning to hire. When you look at those applications, those actually increased also by a lot, almost 50 percent.
Another interesting point is that these business applications have remained quite high. Even far into the end of 2024, the IRS was still receiving about 400,000 applications. These heightened levels actually stayed.
Sablik: Yeah, definitely a really interesting development that researchers here in the Fed and outside the Federal Reserve System have been studying and watching very closely.
In our last conversation, I know I had a lot of questions for you about what this surge in business activity might mean for the economy. For the most part, it was too early to tell. In fact, we have a clip from that conversation which we're going to play now.
[Sound effect]
Sablik: Do we know yet anything about what type of startups these are?
Yeh: It's difficult to say. We don't have the data yet to verify that. Typically, you verify this with confidential micro data from the Census Bureau, and that data tends to be released two or three years after the actual time episode. So, I would say we'd have to wait at least another year.
[Sound effect]
Sablik: Okay, we're back.
Thanks to the relentless passage of time, it's now two years later. As you promised, you now have some answers to the questions I asked you before, so let's go through them.
First, there was some uncertainty about whether or not the new businesses being created during the pandemic represented mostly self-employed entrepreneurs or people starting firms with plans to hire. What did the latest data suggest? Did the startup surge translate into a hiring surge?
Yeh: First of all, it's good to be held accountable.
The first thing that I want to emphasize, like what I talked about before, is that ex ante it's not clear whether these EIN applications translate to establishments hiring actual employees. Just to give you an example, sometimes when you want to get a business loan at a bank, they require you to have an EIN. So, you want to get that loan, but you're not necessarily planning to hire these employees, right?
In terms of this translation you were talking about, Tim, we did definitely observe one. At the end of 2021, so a little bit later after the EIN application surge, we actually saw a surge in establishment openings — these are establishments that are actually hiring people. So, this increased by about 45 percent compared to, let's say, the first quarter of 2019.
Importantly, these are not just businesses opening a location and then, let's say, hiring one person. They were actually hiring a lot, what we call entrant job creation. These are jobs created by just-opened businesses. Businesses like that increased by almost 40 percent. Just to give you an idea of the magnitude, at the end of 2021, opening establishments alone were adding about one million jobs in a given quarter, which is very high but not historically unprecedented.
All these statistics were a clear sign that something happened to American entrepreneurial spirits, right? There was a lot of optimism, a lot of businesses [that] were created. A lot of jobs were created.
The one thing I would push back a little bit on, I would say, is whether the pandemic surge can be labeled as the cause of the ongoing increase in entrepreneur activity. When you go back to data in the '80s or '90s, what you see is a decline in what we call a secular trend or a long-run trend in business dynamism. What I mean by that is fewer establishments entering the economy [and] jobs being created by entering firms, or young firms, for that matter.
What was interesting is when you start looking, let's say, from 2012 onward, that declining trend broke a little bit and actually started increasing a little bit. That trend has basically continued until literally this day. The pandemic surge definitely caused higher levels even compared to that trend. But when you look at the data now, we're reverting back to that trend. So, these higher levels of establishments entering and creating jobs, that's hard to attribute to the pandemic because that trend was already taking place way before, in particular around 2012. So, in that sense, I would say the pandemic surge was relatively short lived.
Sablik: That's really interesting. I think that might have been one thing we talked about last time as well, about these longer running trends.
Another thing we discussed previously was what effect the new firms would have on productivity. One reason that economists pay particular attention to startups is that, in the aggregate, they disproportionately contribute to productivity growth. Not all startups grow, as we discussed before, but some grow and innovate very rapidly.
In 2023, it was too early to tell which kind of startups made up the pandemic surge. Do we have a better idea now?
Yeh: I would say yes and no. The data from the Census Bureau has only been updated until March 2022, so it's still a bit early to dive into the distribution and look at the micro data.
Something we can do, though, is look at industry-level data that's provided by the Bureau of Labor Statistics: the Quarterly Census of Employment and Wages. When you look at this pandemic surge in EIN applications, you can look at quite narrow industries and look at which industries saw a particularly large surge that even deviated from the aggregate trends — non-store retailers or information services or apparel manufacturing.
Out of all the industries that saw a big EIN application surge, only one of these sub-industries — in particular, information — outperform the aggregate. The information sector contains industries such as publishing or broadcasting and data processing and hosting and related services. It's also the only sector that has been showing some productivity growth. Relative to, let's say, 2017, its productivity increased by 11 percent. When you look at broad sectors like manufacturing or wholesale and retail trade, their productivity actually stagnated or even declined.
So, I don't have the data, but my estimation is that the aggregate effects are quite small. The information sector, despite growing by 11 percent, is a relatively small sector, right? Just give you an idea, information on its own has about two or three percent of aggregate employment. So, they grow a lot, but it's fairly small. So, my sense would be that the effect on the aggregate is relatively small.
Sablik: Another question that we discussed, and you've already sort of alluded to, was whether the surge in new business activity represented a lasting break from the pre-pandemic trend or more of a temporary blip. Could you go into a little bit more about what you found there?
Yeh: I would say the first two to three years after the pandemic, we definitely saw a boost, that we were deviating in a positive sense from that increasing trend. But when you look at data after, let's say, 2023 onward, I would say we're reverting back to that trend. In that sense, this pandemic surge was short lived.
I wouldn't say that the pandemic surge was a blip. It was short, but we still saw huge levels of job creation, right? Adding a million jobs for a quarter is quite impressive.
Sablik: For a couple years.
Yeh: Exactly.
Sablik: We've been talking about these EIN applications. The data there comes from the Census Bureau's Business Formation Statistics, which is a relatively new dataset. It's only been around for around 20 years or so. As a researcher, what has the pandemic episode revealed to you when it comes to working with and interpreting those data?
Yeh: Unlike other datasets it's high frequency, so it gets updated every month. So, if you want to talk about cyclical movements, it's a nice dataset. Most importantly, the lag — how much we're behind in terms of data — is only a month.
In terms of what we can learn from it, how we can interpret those data, I think the latest episode has shown us that this surge in EIN applications actually did result in a translation of employer establishments. So, it does tell us something about whether entrants are willing or going to create a lot of jobs. In that sense, it can be used as a predictive tool, which wasn't possible with the data that we had before.
Sablik: Are there still other questions about this pandemic startup surge that you're waiting on more data to answer?
Yeh: As someone who is interested in firm dynamics, I'm always interested in doing research on what we call gazelles. These are young or startup firms that grow by a lot, let's say, in the first four to five years. What I would be very interested to look at is whether this distribution of gazelles had changed. Do we see changes in the magnitude of their growth, or do we see compositional shifts across space or industries? That would be something that I would like to look at and that we could probably start in one or two years.
Sablik: Well, hopefully then we'll have you on in at least another year or two to chat about that. But for today, Chen, thanks so much for joining me.
