Self-Employment Is Not A Good Proxy for Entrepreneurship

by Reihan Salam

Catherine Rampell asks:

Given a climate so favorable to entrepreneurs, what might account for the fact that self-employment and other measures of entrepreneurship are lower in the United States than in many other developed countries? Any suggestions, readers?

In an earlier post, Rampell wrote the following:

A more direct measure of entrepreneurship might be the share of workers who are self-employed. Compared to other developed countries, the United States figures are poor to middling, for both native-born and the huddled masses coming from abroad. Greece has the highest rate of self-employment for native-born citizens, and Poland has far and away the highest self-employment rate for foreigners.

The fact that Greece and Poland have high rates of self-employment should give you pause. Apart from Australia, the economies with a high concentration of tiny firms are peripheral economies or economies that creates barriers to the growth of young firms. 

A new report on entrepreneurship from the Organization for Economic Cooperation and Development finds that the smallest businesses — those with fewer than 10 employees — account for almost all of the businesses in most developed countries. The United States is on the low end of the distribution, though, with only about three-quarters of its businesses being so tiny.

Why would we want large numbers of tiny firms and a high self-employment rate? Think about it: firms start out tiny and they either grow quickly or they stagnate, at one employee or a handful of employees. In a competitive marketplace, we’d presumably see these minnows outcompeted by more successful firms, which would then grow beyond ten employees, perhaps by hiring some of the formerly self-employed who had been competing in the same space. 

The Kauffman Foundation has addressed the phenomenon of what it calls “jobless entrepreneurship“:

“Since it began, the recession has triggered annual declines in the rate of employer enterprise births,” said Carl Schramm, president and CEO of the Kauffman Foundation. “Far too many founders are choosing jobless entrepreneurship, preferring to remain self-employed or to avoid assuming the economic responsibility of hiring employees. This trend, if it continues, could have both short- and long-term impacts on economic growth and job creation.”

Tino Sanandaji has gone further, arguing that self-employment correlates negatively with high-growth entrepreneurship:

This paper uses two newly assembled datasets to demonstrate that the common practice of relying on self-employment to proxy for entrepreneurship often gives to rise to misleading inference. I determine the source of wealth of all billionaires listed on Forbes Magazine’s list, identifying 996 individuals in over fifty countries who became rich by founding new firms. 

Using these individuals to define the per capita rate of entrepreneurship, I show that entrepreneurship rates correlate negatively with self-employment rates. Countries with higher income, lower taxes and less regulation have higher entrepreneurship rates but less self-employment. I attempt to account for these results theoretically using a model where efficient financial markets and a favorable policy environment lead to a better allocation of capital to talent, raising wages, and thereby driving the least productive self-employed individuals to seek employment. 

This evidence is supplemented with data from a recently administered survey of 12,000 Swedish twins. The survey asks individuals to identify as self-employed or entrepreneurs based on their intentions to innovate and grow their businesses. Whilst the self-employed have lower incomes than employees with similar characteristics, entrepreneurs have higher incomes. These relationships hold both in the cross-section and within family. 

 Sanandaji’s method of determining the per capita rate of entrepreneurship is quirky, but interesting nevertheless. 

The Agenda

NRO’s domestic-policy blog, by Reihan Salam.