With unemployment at stubbornly high levels, Washington's attention is focused on job creation. Recent research has shown that the formation of businesses is the source of much net job creation. But when discussing job creation by entrepreneurs, most observers focus on successful new businesses, assuming that most jobs are created from the growth of the best-performing startups. What's often overlooked is the fact that soon-to-be-unsuccessful entrepreneurs create even more jobs than their luckier counterparts.
Business Formation Is the Source of Net Job Creation
Recent statistics gathered by the U.S. Census Bureau show that almost all net job creation comes from the formation of new businesses, not from the growth of existing businesses. In fact, as the figure to the right shows, except for the job creation by businesses more than 26 years old, the act of formation accounts for all of the net job creation in the economy.
(Of course, as I have argued elsewhere, this job creation pattern might exist by construction. Companies at least one year old both create and destroy jobs, but new businesses only create jobs. As a result, net job creation and gross job creation are the same for new businesses, making net job creation much higher for new businesses than for existing ones.)
Moreover, the number of jobs accounted for by new business formation is substantial. In 2005, the latest year for which the Census Bureau has made such data public, the act of new business creation was responsible for 3.6 million net new jobs.
Most of the Net Job Creation Is from Failed Businesses
Just over half of all new businesses are gone by their fifth birthday. Those failures produce more net new jobs as part of the formation process than the ones that survive. Census data from the latest available cohort—those businesses founded in 2000—show that businesses that didn't make it until 2005 created 1,611,177 net new jobs when they were started, whereas the businesses that did survive until 2005 created 1,475,331 net new jobs in their start year. That is, 52 percent of the net new jobs created through formation were in the businesses that would fail within five years, and 48 percent were created in businesses that would survive.
Of course, over the subsequent five years, some of the survivors added employees. But others laid people off. On balance, over the 2000-to-2005 period, the businesses from the 2000 cohort that lived to age five added 679,447 net new jobs. Bear with me here, but that means that while the survivors did indeed add jobs, failures generated more by forming in the first place. This matters more than you might think.
Putting Two and Two Together
If most of the net new jobs in the economy are created by entrepreneurs forming new businesses and the majority of those jobs are produced by entrepreneurs whose businesses don't survive five years, then the act of business creation by unsuccessful entrepreneurs is a key part of how our economy creates new jobs. To have the kind of net job creation that would bring down our currently high unemployment rate, many soon-to-be-unsuccessful entrepreneurs need to start businesses.
However, policymakers thinking of using entrepreneurship as a source of job creation need to recognize the implications of these patterns. When entrepreneurs start businesses, they often have to incur costs, such as paying newly hired employees, before they generate revenues.
The negative cash flow that startups experience from incurring costs before generating revenues has to be financed. The data show that the primary source of this financing is the founders' savings. This means that a lot of the soon-to-be-unsuccessful entrepreneurs are creating a big chunk of net new jobs by using their savings to hire employees—that is, using their wealth to give their employees paychecks.
Of course, the entrepreneurs that provide this job creation through wealth transfer aren't unsuccessful at the time they create and pay for the new jobs, nor do most of them anticipate being failed entrepreneurs. But a bunch of them will fail after having paid others out of their savings to get their businesses going. If Washington decides to stimulate job creation by encouraging people to start businesses, it needs to pay attention to where the money to pay employees is actually coming from.