Migration, particularly skilled migration, is a driver of knowledge diffusion and it can induce productivity shifts as measured by new exports

Over the past few years, economic migration has increasingly become a highly debated political issue, in both developed and developing countries. This issue was particularly highlighted during the 2016 US presidential elections and the 2016 referendum in the UK on membership of the EU. There is a growing sentiment that migrants add additional pressure to existing infrastructure and leave less jobs for locals. However, these fears are often not based on evidence and overlook the gains to host economies (watch Mushfiq Mobarak discuss these gains in a video here).

Migration’s relationship with productivity

An important outcome of migration is productivity. In fact, there are many ways through which migrants can affect productivity: through encouraging efficient specialization, innovation, and skill-complementarities arising from diversity (Peri 2010, Hunt and Gunther-Loiselle 2010, Alesina et al. 2016). Our research explores a novel aspect of this debate by investigating whether and how migrants can contribute to the export diversification of countries by fuelling the emergence of new export sectors (Bahar and Rapoport 2018).

All international flows of goods, capital, and people have the potential to convey critical determinants of productivity: technology and knowhow. Imported goods, for example, contain knowledge as to how they have been produced; such knowledge can, under certain conditions, be used to copy, improve, and eventually export these goods. Foreign direct investment can also be a form of transporting technology. When a multinational corporation establishes a branch in a new location, this tends to come with transfers of firm-specific knowhow through short-term migration and/or training for local workers. Admittedly, such knowledge is intended to stay within the firm, yet there is a vast literature studying how these investments tend to spill over to other firms in related sectors.

Mechanism: Knowledge diffusion

This column focuses on the role migrants play as vehicles of information between source and receiving countries that leads to increased productivity as measured by exports. There is less systematic evidence of how migration flows are one driver of knowledge diffusion. Yet, there are many reasons to believe that migration may actually be the most effective of such driver, as migrants are not bounded to a particular firm and can bring tacit knowledge – a type of knowledge that requires direct human interaction to be transferred appropriately and hence cannot be embedded in goods or written down on a website or a piece of paper (Polanyi 1956, Arrow 1961).

One noteworthy consensus in the literature looking at the patterns of knowledge diffusion is that diffusion is highly geographically dependent. For instance, patent writers tend to cite other patents that originated in the same location more frequently, and firms tend to benefit from spillovers mostly from other nearby firms. This suggests that a large chunk of the knowledge from which firms grow, or from which innovation benefits, is tacit, because otherwise distance would not play any role (Jaffe et al. 1993, Bahar et al. 2014). If people are vehicles of knowhow, then the pattern of international knowledge diffusion should logically follow the pattern of international migration.

Evaluating the impact of migration on comparative advantage

Building on this insight, we used international migration and trade data between over 100 countries during 1990 to 2010. Our main finding is that a 10% increase in the immigrant stock from country exporters of a certain good can explain a 2% improvement in the probability of the receiving country exporting this same good to the rest of the world, competitively and from scratch. Our results therefore imply that the inflow of immigrants coming from countries with comparative advantage in a given product can lead to a strong increase in the likelihood that the receiving country starts to export that same product competitively in the following ten years. We deal with endogeneity concerns by instrumenting for migrant flows using a bilateral migration gravity model.

We interpret these changes in the composition of the export basket of countries as sectorial productivity shifts. This is because only an increase in productivity can explain a country being able to export a good to the world in a significant amount from scratch, keeping global demand constant. Interestingly, qualitatively similar results are obtained for sending countries (i.e. migrants also explain the appearance of new goods in the export basket of the sending country).

Skilled versus unskilled migrants

We find that this process is strongly driven by immigrants that are considered skilled (i.e. have completed enough years of education to earn at least a college degree). In fact, when comparing the ability of migrants to shape the export basket of countries through sectorial specific shifts, we find that a skilled immigrant is about ten times more effective than an unskilled one. Our evidence also shows that this effect is driven mostly by instances where the receiving country is a non-OECD one, where we find that skilled immigrants are about 20 times more effective in driving this effect than unskilled ones. 

Importantly, these results are not driven by a simple increase in the supply of labour, since the data show positive effects for goods both above and below the average capital intensity (the capital-to-labour ratio in the economy) measure. Interestingly, our results suggest that the export effect is concentrated in products that are neither particularly knowledge-intensive nor particularly knowledge-unintensive. That is, migrants do not play a role in explaining productivity shifts of products that are ‘simple’, as knowledge would diffuse with ease in those cases regardless of migration. And at the other end, where products are complex, it is difficult for knowledge to diffuse, in spite of migration flows, if migration is not accompanied by complementary inputs.

Policy implications

Based on our work, there are many reasons to believe that migration may actually be a highly effective driver of knowledge diffusion across borders. This is, if anything, the most important conclusion of our study. Therefore, this study carries important policy implications. Since knowhow and technology are the most important determinant of productivity growth, policies incentivising temporary migration, for example, may be a highly efficient way for firms and countries to access new knowledge. Further investigation of the type of migration policies that would be most effective for this purpose is, in fact, a crucial part of our ongoing research agenda.


Alesina, A, J Harnoss, and H Rapoport (2016), "Birthplace diversity and economic prosperity", Journal of Economic Growth 21(2): 101-138.

Arrow, K (1969) “Classificatory Notes on the Production and Transmission of Technological Knowledge”, American Economic Review 59(2): 29–35.

Bahar, D, R Hausmann and C Hidalgo (2014), “Neighbours and the evolution of the comparative advantage of nations: Evidence of international knowledge diffusion?”, Journal of International Economics 92(1): 111–123.

Bahar, D and H Rapoport (2018), “Migration, knowledge diffusion and the comparative advantage of nations”, The Economic Journal Early View article.

Hunt, J and M Gauthier-Loiselle (2010), "How much does immigration boost innovation?", American Economic Journal: Macroeconomics 2(2): 31-56.

Jaffe, A., M Trajtenberg and R Henderson (1993), “Geographic localization of knowledge spillovers as evidenced by patent citations”, The Quarterly Journal of Economics 108(3): 577.

Peri, G (2012), "The effect of immigration on productivity: Evidence from US states", Review of Economics and Statistics 94(1): 348-358.

Polanyi, M (1962), Personal knowledge: Towards a post-critical philosophy, London: Routledge.

Knowledge diffusion Skilled migrants Export diversification