A conference on The Role of Immigrants and Foreign Students in Science, Innovation, and Entrepreneurship took place on April 27 in Cambridge. Ina Ganguli of University of Massachusetts Amherst; Shulamit Kahn of Boston University, and Research Associate Megan MacGarvie of Boston University organized the meeting, which was sponsored by the National Science Foundation Science of Science and Innovation Policy Program and the Ewing Marion Kauffman Foundation. These researchers' papers were presented and discussed:
Michael Roach, Cornell University; Henry Sauermann, European School of Management and Technology and NBER; and John Skrentny, University of California, San Diego
U.S. Immigration Policies and the STEM Entrepreneurial Workforce
Large technology firms and startups are often credited with driving innovation and job growth, and there are heated debates about whether policy reforms are needed to attract and retain high-skilled immigrant workers. Using a novel survey that follows a cohort of 2,203 science and engineering doctorates from U.S. research universities, Roach, Sauermann and Skrentny examine the sorting of immigrant STEM workers into employment in startups or established firms. They report three key findings: First, prior to graduation foreign-born students are significantly more likely to be interested in working in a startup than U.S. citizens. After graduation, however, immigrant doctorates are significantly less likely to work in a startup over an established firm, even controlling for ability, degree field, and other individual characteristics. Moreover, the researchers find evidence that natives sort into startups or established firms based on their ex. ante career preferences, while immigrants do not, suggesting that immigrants face constraints in preference-based sorting. Second, there is no significant difference in the wages of immigrant and native STEM doctorates in a given type of firm, although startup employees make significantly less than established firm employees. Finally, the researchers find differences in visa types between startup and established firm immigrant STEM workers, with H-1B being the most prevalent type among established firm employees. "Self-sponsored" visas such as OPT or spousal green cards are more common among startup employees, perhaps because these immigrants don't require an employer-sponsored visa to work in a startup. Overall, these findings suggest an untapped pool of immigrant doctorates who are interested in startup employment but instead work in large established firms. The findings have implications for current debates on immigration reform and for efforts to strengthen the human capital base of entrepreneurial ventures.
J. David Brown, U.S. Bureau of the Census, and John S. Earle, Mee Jung Kim, and Kyung-Min Lee, George Mason University
Are Immigrants More Innovative? Evidence from the Annual Survey of Entrepreneurs
Do immigrant entrepreneurs account for a disproportionate share of job creation and innovation among start-up firms? The received wisdom that immigrants create hugely successful and innovative companies has been influential in many popular discussions, and it is easily illustrated with particular anecdotes, but as yet there has been relatively little evidence from systematic analysis of large, representative datasets. A number of studies have examined immigrant self-employment, but few have studied job creation, and as far as we know none have studied innovation. Brown, Earle, Kim, and Lee aim to contribute to the emerging literature on the impact of immigrant entrepreneurship on the U.S. economy. They focus on the founders of new businesses during their initial "entrepreneurial phase." The outcome variables analyzed include initial employment, subsequent employment growth, and innovation activities. Because of the high skewness in firm performance outcomes, however, their main focus is on the top end of the distributions, the high growth "gazelles" and the biggest innovators. The researchers measure high growth among entrants as the top x percent of the employment distribution at various ages. For innovation, they measure not only the incidence of various product and process changes, R&D, and patenting, but also the degree to which the type of innovation involves exploration vs. exploitation. In estimating differences in these job creation and innovation outcomes between immigrant-owned and native-owned start-ups, the researchers are able to control for a wide variety of founder characteristics, including gender, age, education, etc. Data on race/ethnicity permit some disaggregation of immigrant country of origin. For firms with multiple founders, they also examine immigrant participation in founding teams, and the impact of diversity on this dimension for firm performance. The researchers are also able to examine immigrant-native differences in the roles played by a number of factors that may be jointly determined with job creation and innovation outcomes, including start-up capital, choice of industry, region, legal form (including franchising), and founder roles in the new company. Finally, the researchers use self-reported motivations for founding the business to distinguish growth-oriented from "lifestyle" entrepreneurship. For all of these variables, Brown, Earle, Kim, and Lee are interested both in characterizing immigrant relative to native entrepreneurs and in measuring how they influence or mediate the immigrant-native entrepreneur differences in job creation and innovation performance.
Stefano Breschi, Bocconi University, and Francesco Lissoni and Ernest Miguelez, Université de Bordeaux
Returnee Migrants' Self-selection: Evidence For Indian Inventors in the U.S.
Based on an original dataset linking patent data and biographical information for a large sample of U.S. immigrant inventors of Indian origin, specialized in ICT technologies, Breschi, Lissoni, and Miguelez investigate the rate and determinants of return migration. For each individual in the dataset, they estimate both the year of entry in the United States, the likely entry channel (work or education), and the permanence spell up to either the return to India or right truncation. By means of survival analysis, they then provide exploratory estimates of the probability of return migration as a function of the conditions at migration (age, education, patenting record, migration motives, and migration cohort) as well as of some activities undertaken while abroad (education and patenting). The researchers find both evidence of negative self-selection with respect to educational achievements while in the U.S. and of positive self-selection with respect to patenting propensity. Based on the analysis of time-dependence of the return hazard ratios, return work migrants appear to be negatively self-selected with respect to unobservable skills acquired in the U.S., while evidence for education migrants is less conclusive.
Kirk B. Doran and Chungeun Yoon, University of Notre Dame
Immigration and Invention: Evidence from the Quota Acts
Since (Hicks, 1932), economists have noted that inventions often economize on labor, so scarce labor should encourage more invention. But (Acemoglu, 2010) notes that in canonical macroeconmic models, plentiful labor encourages invention. The stakes of this debate are high in the policy context of mass immigration. Doran and Yoon provide the first causal evidence of the effect of mass immigration on invention, using variation induced by 1920s quotas, which ended history's largest international migration to the U.S.. Inventors in cities exposed to fewer low-skilled immigrants applied for fewer patents, an effect driven by fewer patent applications relevant for the industries that lost the most immigrant workers.
Gaurav Khanna and Munseob Lee, University of California, San Diego
High-Skill Immigration, Innovation, and Creative Destruction
Economists and policy-makers have long debated the impacts of immigration on the wages of domestic workers. Despite the important theoretical implications of how immigrants may affect consumers in destination countries, there is little empirical evidence about the effects on consumer welfare. Khanna and Lee study how high-skill migration to the United States under the H-1B visa program affected consumer welfare via changes in prices and the varieties of products produced by firms. This new evidence will help quantify the overall benefits from immigration, and inform policy on high-skill migration. The researchers combine the Nielsen data with the universe of H-1B visas granted between 2006 and 2016, which includes both new visas and renewals, and find that firms that hired more H-1B workers produced more products and lowered the prices of the products produced. This will potentially increase the welfare of consumers that now have access to more and cheaper products.
Anna Maria Mayda, Georgetown University; Francesc Ortega, Queens College and Graduate Center CUNY; Giovanni Peri, University of California at Davis and NBER; Chad Sparber, Colgate University; and Kevin Y. Shih, Rensselaer Polytechnic Institute
The Effect of H-1B Visas on Firms: Evidence from Publicly Traded Firms
The H1-B visa program is the main channel to hire skilled foreign workers in the U.S. There is an annual cap on the number of new H-1B visas awarded to for-profit firms. Mayda, Ortega, Peri, Sparber, and Shih estimate the effects of H-1B visas on the performance of the firms that receive them. The researchers obtained case-level data from the USCIS on the universe of H-1B visas awarded over the period 2000-2012 and merged them with data for all publicly traded companies in the United States. The resulting dataset is a panel of firms including information on sales, employment and profits, along with estimates of the stock of H1-B workers in the firm at each point in time. To address endogeneity bias the researchers adopt an instrumental-variables approach based on changes in the annual cap interacted by the degree of dependence of each firm on H-1B visas (as in Kerr and Lincoln, 2010). The researchers' preliminary findings suggest that within-firm increases in H1-B employment lead to growth in firm size (measured by employment and sales) and in profitability.
Sari Pekkala Kerr, Wellesley College, and William R. Kerr, Harvard University and NBER
High-Skilled Immigrant Networking and Innovation
Networking and the giving and receiving of advice outside of one's own firm are important features of entrepreneurship and innovation. Kerr and Kerr study how immigrants and natives utilize the potential networking opportunities provided by the Cambridge Innovation Center (CIC). CIC is widely considered the center of the Boston entrepreneurial ecosystem. The researchers surveyed 1,334 people working at CIC in three locations spread across the Boston area and CIC's first expansion facility in St. Louis, MO. Survey responses show that immigrants value networking capabilities in CIC more than natives, and the networks developed by immigrants at CIC tend to be larger. Immigrants report substantially greater rates of giving and receiving advice than natives for six surveyed factors: business operations, venture financing, technology, suppliers, people to recruit, and customers. The structure and composition of CIC floors has only a modest influence on these immigrant versus native differences.
Ina Ganguli, and Patrick Gaule, CERGE-EI
Will the U.S. Keep the Best and the Brightest (as Post-docs)? Career and Location Preferences of Foreign STEM PhDs
Ganguli and Gaule discuss findings from a recent novel survey of current doctoral students in a major STEM field — Chemistry — conducted in 2017 at 50 U.S. institutions about their career and location preferences. First, the researchers estimate the career preferences of foreign and U.S. STEM Ph.D. students for different types of post-graduation jobs — postdocs, industry, or teaching positions — using both hypothetical choice methods and more standard Likert measures of preferences for different careers. Second, they examine students' location preferences using based on a hypothetical choice method. The researchers proceed to use these counterfactual job questions to construct a revealed-preferences ranking of universities as locations to do a chemistry postdoc. Ganguli and Gaule's findings show the foreign and U.S. Chemistry Ph.D. students have significantly different preferences for careers, with foreign students being much more likely to prefer doing a postdoc. Foreign students also value a U.S. location more than U.S. students, and their revealed preference ranking of locations for a postdoc differ in many ways. The researchers' results suggest the U.S. will manage to retain talented foreign graduate students and no brain exodus is in sight.