Two economies of young firms

In a previous post, I looked at data on job creation by startups and showed that, in any given year, startups create more jobs on net than all other firms combined. The result that startups are key to job creation is now well documented, here and elsewhere. Figure 1 is the chart I used (click for larger image):

Figure 1

Last week, the Atlanta Fed's macroblog ran a post reviewing some recent research that highlights the need to be careful when aggregating:

Over [2006], expanding firms more than 10 years old added a whopping 11 million jobs--about three times as many jobs as created by new firms. Of course, some older firms were downsizing or closing--contracting mature firms destroyed an estimated 10 million jobs.

Those who read my post on gross job flows will not be surprised by these numbers. Net jobs figures hide a huge amount of churning. It would be a big mistake to assume that older firms are uniformly shedding jobs or doing nothing.

In light of these data on startups and older firms, I thought it would be interesting to look at what happens to young firms immediately after their first year. Figure 2 shows job creation (number of jobs added by expanding establishments) and job destruction (number of jobs eliminated by shrinking establishments) for 1-year-old firms--that is, firms that were startups in the previous year.

Figure 2

So some new firms immediately begin shedding workers while others keep expanding. Though startups are creating 2 to 4 million jobs per year, in their second year of existence some of them are responsible for destroying almost a quarter of those jobs. Luckily, other firms of the same age keep expanding and about make up for it--though it's interesting that 1-year-old firms have been net job destroyers since 2001.

Figure 3

Figure 3 shows the same data for 2-year-old firms. By this age, firms have become net job destroyers (in every year except 2000). Figure 3 shows 3-year-old firms.

Figure 4

It doesn't take long for each startup cohort's surge of job creation to disappear. This isn't a big surprise--the high failure rate of startups is not a secret. In some sense, the data almost paint a picture of two economies. In one, successful startups enter and steadily create jobs, becoming somewhat permanent. In the other, firms enter, struggle, and shrink or collapse within a few years, constantly contributing to labor market reallocation but providing little in terms of stable labor demand. A nice discussion of this is provided by this BDS briefing:

Young firms have higher employment growth rates, if they survive, than older firms. . . . However, younger firms experience much more employment loss due to establishment exit, nearly 20 percent at very young firms, than do larger firms. 

The pattern for young firms is thus one of "up or out" with very rapid net growth for survivors balanced by a very high exit rate. . . . Lumping together all firms of the same age is clearly misleading, given this "up or out" dynamic. Young firms obviously are doing both better and worse than more mature firms in terms of growth and survival.

So while it's still true that startups are crucial for net job creation, their involvement in the job market is highly volatile and impermanent.