Immigration, innovation, and the connection between them are hot topics once again. With the labor market in the doldrums, America's technical supremacy under challenge from abroad, and the political issue of immigration reform heating up, many argue that immigrants have a special propensity for innovation and entrepreneurship that can help spark badly needed economic growth.
President Barack Obama expressed this view in its purest form in his highly publicized speech on immigration  in El Paso, Texas, on 10 May. "Look at Intel, look at Google, look at Yahoo, look at eBay," he said. "All those great American companies, all the jobs they've created, everything that has helped us take leadership in the high-tech industry, every one of those was founded by, guess who, an immigrant. So we don't want the next Intel or the next Google to be created in China or India. We want those companies and jobs to take root here." The president also quoted the native-born Bill Gates, a man, he said, who "knows a little something about the high-tech industry," to the effect that excluding those "able and willing to help us compete" will damage the nation's "competitive edge."
How accurately does this picture reflect immigrant entrepreneurs' role in the formation of job-creating, technology-based companies? On the one hand, a widely quoted study  by Vivek Wadhwa of Duke and Harvard universities and his co-authors finds "at least one immigrant key founder in 25.3% of all engineering and technology companies established in the U.S. between 1995 and 2005 inclusive."
Recent research by David Hart , director of the Center for Science and Technology Policy at George Mason University in Fairfax, Virginia, and Zoltan Acs  director of George Mason's Center for Entrepreneurship and Public Policy, finds a considerably lower figure. Obama's statement and the Wadhwa study focus on companies in the computer and information technology fields. The research by Hart and Acs examines companies in a much broader range of science- and technology-based industries that are important to the American economy (and to the employment prospects of scientifically and technically trained people), including pharmaceuticals, medical devices, optical equipment, industrial and agricultural chemicals, guidance and navigation systems, aeronautics and space, testing instruments, and much more.
"There's always a tendency to generalize from the outlier," Hart said to an audience of policy and immigration experts at a 13 May presentation sponsored by Georgetown University's Institute for the Study of International Migration . Simply founding a company does not guarantee that it will generate employment; most start-ups fail, and only "a small number of companies produce almost all the new jobs in the economy," he said.
Unlike the great majority of companies, the strong job producers, which he called "high-impact" companies, are generally "young" and "fast-growing." Surveying a nationally representative sample of "high-impact" companies in high-tech industries, he and Acs found 16% to have at least one foreign-born founder. Though "substantial," this figure stands near the "low end" of scholarly findings to date on immigrant entrepreneurship, Hart said. Nor did Hart and Acs's data reveal immigrant-founded companies to be faster growing, more innovative, or more research-oriented than those founded only by native-born Americans.
Methodological differences produced the discrepancy between the estimates, Hart explained at the presentation. The 25% figure is an extrapolation from data collected in Silicon Valley. "Because [that research] concentrates on the region of the United States in which high-tech immigrant entrepreneurs are most likely to be found, one cannot generalize easily from it" to the rest of the country, Hart and Acs write in a paper titled "High-Technology Immigrant Entrepreneurship in the U.S.,"  which is forthcoming in Economic Development Quarterly.
Hart and Acs, on the other hand, surveyed "a national random sample of high-tech, high-impact companies" and achieved a response rate of 29%, Hart said. They define "high-impact" companies as those with sales that "have at least doubled" over the 4 years between 2002 and 2006" as well as an "employment growth quantifier" of two or more. This metric combines "absolute and percent change in employment" in order to avoid distortions from the use of either measure alone, they write. Using the U.S. government's Standard Industrial Classification code, they identified 49 "high-tech" industries. Their survey yielded data on approximately 1300 companies and 2000 founders.
Approximately 16% of the companies they surveyed reported at least one founder born outside the United States; some companies have multiple founders. Overall, "about 12.8% of the founders were foreign born," a figure close to the foreign-born proportion of the U.S. population at large, Hart and Acs report. Companies with foreign-born founders are "disproportionately located in states with large foreign-born populations."
The survey did not ascertain the age or circumstances of the immigrant entrepreneurs' arrival in the United States but found foreign-born founders generally to be "deeply rooted" in the United States, Hart said. On average they had been here 25.9 years at the time of the survey, with only 15% being here less than 15 years. More than 75% are U.S. citizens.
The immigrant founders hail from 55 different countries. India, the United Kingdom, Canada, and Japan topped the list, with 40, 25, 15, and 15 individuals respectively. Iran, Mexico, Burma, Croatia, and Tanzania are among the homelands that appear. China, Taiwan, and Hong Kong, counted separately, contributed 8, 5, and 2 founders, respectively. Foreign-born founders have more education than native-born founders and twice the chance of holding a doctorate -- hardly surprising, Hart noted, given that higher education is a major means of entering the United States; two-thirds of the foreign-born founders earned their highest degree in this country. The immigrant- and native-founded companies appear about equally likely to support R&D and to hold or have applied for patents.
In preparing their study, Hart and Acs identified some individuals of extraordinary talent and drive. One man, for example, "is just an idea factory" responsible for numerous commercially viable inventions, Hart said. Unable to exploit his abilities because of conditions in his homeland, "he's just spun off company after company" since coming to the United States. Others in the study are merely people who took advantage of opportunities.
Overall, the data clearly show the valuable contribution of immigrant founders of "high-impact," high-tech companies to the economic life of the United States -- but not any special propensity for innovation or spectacular entrepreneurial success. In fact, it may be America's "institutional environment that is truly unique and not the entrepreneurs," write Hart and Acs in an article in Issues in Technology Innovation. 
What can we learn by following President Obama's suggestion and looking at Intel and the other companies he mentioned? We can learn that none matches the popular paradigm of the highly skilled newcomer applying imported technical expertise to attain entrepreneurial triumph. Three of the great high-tech innovators to whom the president alluded -- Google's Sergey Brin, Yahoo's Jerry Yang, and eBay's Pierre Omidyar -- are indeed foreign-born. All, however, arrived in the United States not as trained technologists but as small children, Brin and Omidyar at age 6 and Yang at 10. All were educated in the United States; Brin and Yang even attended America's oft-maligned public schools.
The two men whom the Intel Corp. names as its founders,  Gordon Moore and Robert Noyce, are natives, respectively, of San Francisco, California, and Burlington, Iowa; Noyce's family immigrated to America on the Mayflower. Andrew Grove, an early employee who, according to the company, "participated in the founding" and became a highly successful CEO, arrived in the United States as a 20-year-old refugee from the 1956 Hungarian uprising and earned his degrees at City University of New York and the University of California, Berkeley, before starting the career that led to Intel.
So what does all this suggest about policies that will foster innovation? Immigration policy "should be seen as one component of a broader strategy to expand high-impact entrepreneurship as a whole. Effective policies for education, research, antitrust and a variety of other elements of the entrepreneurship environment ought to complement a skills-oriented immigration policy," Hart and Acs write.
"It's very hard to pick who's going to be a successful entrepreneur," Hart said at the presentation. Harder still would be designing an immigration system to discern the future entrepreneurial potential of 6-year-olds. A more plausible plan might be to maintain and enhance an environment in which everybody who has a bright idea that may be commercial, regardless of birthplace, has a decent chance to test it in the marketplace.
Beryl Lieff Benderly writes from Washington, D.C.