The Cult of Small Business

7/19/2012 12:24:00 PM

I hear an endless amount of rhetoric about promoting small business from policy makers these days. It is as if there was a law saying that you can't deliver a speech about the economy or taxes without using the phrase "small business" at least three times every fifteen minutes. Here is my political interpretation: democrats appeal to small business to avoid the anti-business label while still opposing giveaways to big business, while republicans appeal to small business so that their pro-business agenda doesn't look like giveaways to big business. Small businesses are therefore just a political least common denominator--too small to be a threat to anyone, they are therefore praised by everyone.

But then there are all these statistics which, at face value, seem to suggest that small business is pretty important. Here are the big points:

Small businesses employ a majority of the workforce

This is a line I've heard a lot. It is true, sort of, although the proportion dropped to 49.5% beginning in 2007. This may be misleading, however, since to get this statistic you have to define "small business" to include any company with less than 500 employees. Just for reference Bain Capital, with $66 billion in assets--certainly enough for most people to call it "big business"--has roughly 400 employees and therefore technically qualifies as small business. (See the small business administration here for a regurgitation of the conventional talking points about the wonders of small business)

Small businesses account for more than 80% of new job creation

As I recall, Hilary Clinton made a big deal of this statistic during her campaign. It is absolutely true--more than 80% of all new jobs are at firms with fewer than 50 employees (this is a much more reasonable definition of small business, too).

Unfortunately, this fact is rather imprecise. We'd like to interpret it to mean that if we enact policies that increase the number of small businesses, then we will also increase the pace of job creation. Unfortunately, that's not what this says. Just consider, for example, a law that imposes sharp tax hikes on firms that employ more than 50 employees, which is similar to some of the labor market policies in some Eurozone countries (the much heralded "structural reforms" mostly means changing these types of policies), and something that has been proposed in the US in the form of tax cuts to businesses with less than 500 employees--such a policy would succeed in boosting the number of small businesses, but only because it disincentivizes them from growing into large companies. Hence, the policy would reduce employment growth, not increase it, relative to a policy that treats all sizes of businesses equally.

John Robertson in this post says of the cult of small business that
" is tempting to be skeptical of claims that talk about any large group of individuals or firms as if they are a single, homogeneous unit."
Translation: Why the heck are we categorizing firms by the number of employees they already have? It turns out that a much better correlation is between firm age and job creation--young firms account for the vast majority of job creation, and the correlation with "small business" occurs only because younger firms tend to start out as small business before they grow into big ones. Abandoning the rhetoric about small business could help sharpen our focus: we need policies that promote new business creation, not small business creation.

But...hold on a second: Although young businesses account for most new job creation, they also account for most job destruction. See here:
The green shows the percent change in employment among the 10% of firms with the biggest job losses, and the purple shows the percent job growth among the 10% of firms with the biggest job gains. From left to right we go from young firms to old firms. This shows that both job destruction and job creation decrease with the firms age, which is consistent with the now considerable amount of empirical research showing that newly created jobs are much more likely to be permanent at old firms (or large firms) than at new (or small) firms.

The St. Louis Fed has a nice piece on the topic here. Some good commentary from Bruce Bartlet here. Finally, an alarming trend in the pace of new business formation at macroblog here.