The question of whether immigrants are more likely than people born in the United States to start high technology companies has been receiving a great deal of attention since Senators John Kerry and Richard Lugar introduced the Start-up Visa Act bill in Congress.
To justify the need for this program, several authors have argued that immigrants are more likely to become high tech entrepreneurs than people born in the United States. For example, in his BusinessWeek column, Vivek Wadhwa wrote “I published a research report back in 2006 showing that over 50 percent of Silicon Valley engineering and technology startups were founded by immigrants (as were 25 percent of such startups nationwide), I concluded that immigrants were more likely to be entrepreneurs.”
Recently I took a look at what data from the Bureau of Labor Statistics (BLS) (which collects data on the entire U.S. labor force) show about the odds that immigrants and non-immigrants become the self-employed heads of incorporated businesses. (I focused on incorporated self-employment because it is a better representation of business formation than unincorporated self-employment.)
Following the standard practice of defining high tech as those industries with above average (or above the median) research and development (R&D) spending, I identified the following industries as “high tech,” using the 2010 Science and Engineering Indicators from the National Science Foundation:
• Internet publishing and broadcasting
• Publishing industries (except internet)
• Internet service providers and data processing services
• Professional and technical services
• Computer and electronic products
• Chemical manufacturing
• Health care services (except hospitals)
• Motion picture and sound recording industries
• Management of companies and enterprises
• Machinery manufacturing
Counter to the frequently made argument, the data don’t show that immigrants are more likely to become high tech entrepreneurs. True, in 2009 in six industries with above average R&D intensity (Internet publishing and broadcasting, publishing industries except the Internet, chemical manufacturing, machinery manufacturing, health care services except hospitals, motion picture and sound recording), immigrants were more prone to becoming incorporated self-employed than people born in the U.S.
But in four high tech industries (computer and electronic products, Internet service providers and database services, professional and technical services, and the management of companies and enterprises), the native born were more likely than the foreign born to be the self-employed head of a corporation.
Even if we just concentrate on a subset of high tech industries – those that relate to computers – there’s no clear pattern in favor of immigrants or the native born. Immigrants have greater odds of running an incorporated business in Internet publishing and broadcasting, but not in Internet service providers, data processing services or computer and electronic products.
Moreover, the gaps between the groups are large. The native born are 6.6 percentage points more likely than immigrants to be the self-employed head of an incorporated Internet service provider or data processing service, but 4.7 percentage points less likely to run their own incorporated Internet publishing and broadcasting business.
There’s no evidence of the supposed greater odds of high tech entrepreneurship among immigrants if we substitute unincorporated self-employment for incorporated self-employment. Using this alternative measure, three of the industries (chemical manufacturing, management of companies and enterprises, and machinery manufacturing) in which immigrants were more likely to run their own incorporated businesses now show no difference between immigrants and those born here. Three of the industries (the publishing industry except Internet, Internet publishing and broadcasting, and motion picture and sound recording studios) where immigrants were more likely to run their own incorporated businesses show lower rates of unincorporated self-employment than those born here.
Only in the computer and electronic product industry did the native born have a higher rate of incorporated self-employment but a lower rate of unincorporated self-employment.
A final point concerns the stability of these patterns over time. In 2008, the native born incorporated self-employment rate in machinery manufacturing was 2.3 percent, versus 1.6 percent for those born outside the U.S. But in 2009, the numbers were 2.8 percent for those born in the U.S. and 3.9 percent for those born elsewhere. In chemical manufacturing, the 2008 rates were 1.4 percent for the American-born and 1.0 percent for immigrants. But in 2009, the numbers were 1.2 percent for those born in the U.S. and 1.5 for those born outside the country. (We can’t say much about changes in the information-based industries because the definitions were changed between 2008 and 2009).
In short, the data don’t support the assertion that immigrants have greater odds of becoming high tech entrepreneurs – at least not in the United States.