«Discussion Paper 77-2013 MIGRATION AND INNOVATION – A SURVEY Sheida Rashidi Andreas Pyka Universität Hohenheim | Forschungszentrum Innovation und ...»
Partridge and Furtan (2008) investigate the link between innovation and immigration in Canada and find that highly-skilled immigrants with language proficiency in English or French have a significant impact on the innovation flow at the provincial level in Canada.
Innovation outcome is measured with patents. They found that a 10 percent increase of highly-skilled immigrants led to 7.2 percent increase in the overall number of patents in the province (Partridge and Furtan (2008), p. 128).
Regarding brain circulation and commuting entrepreneurial networks the work of Saxenian et al. (2002) study the role of US educated immigrants who span their activities across borders and create economic opportunities. They explore the scope and organization of the local and transnational networks that are built by immigrants, particularly by the first generation of Indian, Chinese and Taiwanese immigrants in Silicon Valley. In their survey three issues are addressed: (i) The involvement of Silicon Valley’s foreign-born professionals in the region’s entrepreneurial economy, (ii) the nature of professional connections that first-generation immigrants are building to their native countries, and (iii) the extent to which immigrants are becoming transnational entrepreneurs and establishing business operations in their native countries. Their conclusions imply an extensive evidence of brain circulation between California and fast growing regions in India and China.
Ozgen et al. (2010) empirically investigate the link between migration and innovation in Europe. According to Ozgen et al. immigration may enhance innovation through five channels: (i) a population size effect, (ii) a population density effect, (iii) a migrant share effect, (iv) a skill composition effect and (v) a migrant diversity effect. The first three mechanisms result from the fact that due to immigration local demand rises. Additionally, since migrants are mostly attracted to the larger urban areas where job opportunities are best, they contribute to urban population growth, and thus strengthen the forces of agglomeration which encourages more innovation. The fourth mechanism, the skill composition effect refers to the way through which immigrants change the human capital stock of the host regions, because immigrants bring in new knowledge. According to Borjas (1999), immigrants are not a randomly selected sample of the population. There is a self-selection process in which the skilled workers who migrate may also be more entrepreneurial and less risk averse and considerably young (Ozgen et al. (2010), p. 3). Their mobility generates spillover benefits to the host countries and enhances the innovation activities there. Finally, the fifth mechanism stems from the larger cultural diversity in the host economy.
Ozgen et al. (2010) empirically study the effects of immigration on the innovativeness of the regions in Europe based on data from 12 countries (Austria, Belgium, Denmark, France, Germany (west), Ireland, Italy, Netherlands, Portugal, Spain, Sweden and UK). They construct panel data of 170 regions across 12 countries in Europe (NUTS2). Innovation outcomes are approximated by the number of patent applications per million inhabitants.
Their results suggest that: (i) population size is insignificant; (ii) population density is significant but has a negative sign; (iii) the share of immigrants is statistically insignificant and not necessarily associate with innovation; (iv) the average skill level of migrants is positively correlated with patent application. An increase in the average skill level of migrants 5 has a positive and statistically significant effect on patent applications and (v) cultural diversity in the regional population is significant which means that there are positive externalities in culturally more heterogeneous regions. They find that an increase in the diversity index by 0.1 percent increases patent applications per million inhabitants by about
0.16 percent. Ozgen et al. conclude that in European regions with culturally diverse settings, higher competitiveness and availability of knowledge spillovers add to innovativeness. Their study also shows that there is a critical level of cultural diversity and that innovation is positively affected only if cultural diversity is above.
The Nomenclature of Units for Territorial Statistics (NUTS) is a geocode standard for referencing the subdivisions of European countries for statistical purposes. The NUTS 1 level refers roughly to states or large regions, level 2 to provinces, and level 3 to counties.
proxied by migration from source countries from which emigrants are on average higher skilled Niebuhr (2009) investigates empirically the relationship between cultural diversity and innovation in Germany. She used employment data instead of population data and differentiates between three levels of education: no formal vocational qualification, completed apprenticeship, university degree as well as 213 nationalities. By considering the cultural diversity of the labor force at different qualification levels, Niebuhr intends to verify whether education matters, i.e. taking into account that it might be only cultural diversity of highly qualified workers which affects the process of innovation. Her results confirm that German regions with a higher diversity in their workforce are characterized by higher levels of innovation activities.
A further study by Fabling et al. (2011) tests for New Zealand whether firms located in areas with a relative more immigrants are also more innovative. They find a positive relationship between innovation outcomes and workforce characteristics such as the proportion of migrants, the proportion of people new to the area, the proportion of migrants with high-skills and the employment density. However, this positive relationship is not evident for all innovation outcomes. Moreover, they did not find these relationships for neighborhood areas.
The missing direct link between innovation and local workforce characteristics implies that the spillovers from immigration to innovation are in their results not as strong as on previous studies (Fabling et al. (2011), p. 20). However, the results of Fabling et al. reflect the distinctive features of New Zealand’s immigration patterns and innovation system. According to Fabling et al. (2011) it could be related to New Zealand’s relatively small size and low population density that the scope of spillovers and dense networks is limited.
Hansen & Niedomysl (Hansen & Niedomysl, 2008) focus on the migration of creative persons in Sweden and address three issues: (i) creative class members move more often compared to other migrant groups; (ii) creative persons are more selective in choosing their destination and consider the salutatory culture critical for their decision; (iii) and have different reasons to immigrate. Using two different datasets, the authors identify the creative class which allows for comparison between them and other groups. Their empirical work illustrates that migration rates of the creative class are only marginally higher compared to other groups. Moreover, most migration activities for the creative class take place just after finishing university and that the creative class also moves for jobs rather than place.
Neil Lee and Max Nathan (2010), explore the impact of diversity on innovation in the population of London. London is known as one of the most diverse cities in the world, where 300 languages are spoken by schoolchildren (Gordon et al. 2009, 2007), and 31 ethnic minority groups and 38% of the working-age population were born abroad (Spence, 2008).
Like Niebuhr (2009) the authors check if culturally diverse firms in London are more innovative and what forms of diversity are associated with what form of innovation. In order to measure cultural diversity they focus on two specific aspects of diversity, country of birth and ethnic group. They construct three diversity measures: (i) LABS’ (London Annual Business Survey) coverage of workforce and ownership characteristics, (ii) country of birth and (iii) ethnicity. In order to measure innovation they develop four broader innovation measures related to product and process innovation: exploring new products, modifications of existing product ranges and new equipment and new working methods. Their results illustrate that London’s diversity is an economic asset. They find that diversity and innovative activity are much stronger associated for process innovation than product innovation. The role of “ethnic entrepreneurs” is of particular importance in knowledge-intensive firms in innovative product differentiation and in process innovation.
De Grip et al (2009) analyze the determinants of labor migration after graduation as well as five years after graduation in 12 European countries. They analyze the country choice of the graduate migrants. They find that not only wage gains are determining migration decisions, but also differences in labor market opportunities, past migration experience. Additionally they show that international student exchanges are strong predictors for future migration.
Surprisingly, their results show that job characteristics like skills utilization in the job and involvement in innovation do not affect migration decisions. Regarding the country choice, only countries like the U.S., Canada and Australia appear to attract migrants due their larger R&D intensity. Graduates with better grades are more likely to migrate to these countries.
Miguélez & Moreno (2010) analyze the contribution made by collaborative networks and the labor and geographical mobility of inventors to the process of knowledge creation and regional innovation performance. For this purpose a knowledge production function framework at the regional level is applied which considers inventors’ networks and their labor mobility as independent variables. They use patent data to identify individual inventors, and create a new dataset of individuals with information on personal address(es), their patenting histories, the owners of their patents (be it a firm, a university or other public institution, or the inventors themselves), and the co-authors in their patents. They find strong support for the positive relationship between regional labor market mobility and regional innovation intensity. The influence of networks is also fairly important, but the strength of these ties (measured with the network density) was found to have a negative influence on innovation.
However patenting activities do not explain the mobility pattern of individuals nor their cooperative relationships.
Table 2 summarizes the still rare, however, diverse results of empirical studies of the relationship between migration and innovation.
Table 2: Empirical evidence on migration and innovation
Knowledge-based economies are characterized by new patterns of competition on an international and global scale. Growing highly-skilled mobility has raised the competition among countries in winning the best talents. Highly-skilled migration shows a positive and increasing trend since the beginning of 1990s. Traditional settlement countries, in particular the United States, benefit from immigrant’s population. Their national policies are able to attract highly-skilled immigrants. Studies confirm that non-US citizens contribute extensively to economic development of the U.S. economy. For example empirical studies concerning the registered international patents by immigrants or the contribution of transnational networks between immigrants and their homelands confirm this observation.
Compared to North American countries, Europe shows higher inflows of international migrants. However, it most immigrants arriving in Europe do not hold high skills in terms of education. Also, so far not enough studies of the European situation exist which allow for a better understanding of the contribution of immigrants to European economic development.