In recent years the BLS has developed a new database where they can track the jobs created by an establishment over time. It is called Business Dynamics and you can read about it here.
Research using the new database has altered the basic view of where jobs are created. The old view that jobs are created by small businesses has been pretty much discredited by this approach.
As the chart below demonstrates over recent years, large firms have accounted for a greater share of employment than small firms. Since the 2000 recession their share has been relatively constant, implying that small and large firms have created about the same number of jobs.
The new research finds that net job growth appears to stem from the new establishments that survive. Of course the problem is that many new establishments do not survive. After 5 years less than half of the new establishment are still around and after 16 years less than a quarter survive. So while it is true that small firms create many new jobs, they also destroy many jobs.
Just as an aside this chart shows one of the problems I have with many conservative economist. They believe that government is inefficient and wasteful. Actually, they are probably right. What I have problems with is their belief that the private market is efficient, especially if it is taken to mean that it is not wasteful. As far as I am concerned the private market probably is just as inefficient and wasteful as government. Less than half of new firms survive five years. Just think of how wasteful that is– for years businessmen try to provide a good or service that the market is not demanding. It may be the best we can do, but it is still extremely inefficient and wasteful.
Was the stock market efficient last week when it was bouncing up and down several percentage points daily only to end up barely changed for the week?
Years as a business economist taught has me that markets almost always overshoot and undershoot, either producing too much or too little and asking prices that or either too low or too high. My one investment rule that may be original is that any shortage everyone sees a few years down the road will never materialize because too many people will try to take advantage of the supposed shortage and actually end up creating a surplus.
What the new data also provides is a record of the jobs created by new establishments — what the birth part of the birth -death model in the monthly employment report estimates. What this data shows is that the republican claim that new jobs are not being created appears to be based in fact.
However, that trend of new establishments creating fewer jobs is a decade old. The number of jobs created by new firms peaked in 1999 and has declined every year since and the record in recent years is right on the trend established over the past decade. Thus, it appears that the shortage of new jobs from new establishments in recent years does not appear to stem from recent developments. Rather, it is a trend that has been firmly in pace for a decade.
When I look at this data I do not come up with a good explanation for the decade long decline in job creation by new establishments. The data only goes back to the early 1990s, so maybe what we are seeing is that job creation was unusually high in the 1990s because of the IT bubble and all the recent decline represents is just a return to normal. But I doubt it, and am open to other suggestions.