Decomposing the immigration-trade link using price and quantity margins: the role of education

ABSTRACT

This paper examines the immigration-trade links for OECD countries for the 2000–2015 period. By decomposing the overall effect of the presence of immigrants on the export of the hosting countries according to the methodology proposed by Hummels and Klenow (2005), we investigate if there is a migration-driven increase in the number of export relationships (extensive margin) and/or simply a growth of pre-existing export relationships (intensive margin) decomposed by an increase in quantity and price. We consider the country of origin of immigrants and their level(s) of education. Our results confirm the existence of a pro-export effect of immigrants attributable to the contribution of international networks regarding the reduction in information costs, which is especially strong for developed countries as sources of migration. However, when we look at the heterogeneous impact of different levels of immigrants’ education, we find that the pro-trade effect is due to the contribution of less- and more educated immigrants to the reduction of trade costs, while the medium educated do not spur bilateral trade significantly. We address the potential endogeneity problems by adopting an instrumental variable (IV) approach based on shift-share IV and pull-and-push factors that could induce international migration.

KEYWORDS:

• International trade
• extensive and intensive margins
• international migration
• skilled migration
• gravity equation

JEL CLASSIFICATION:

• F14 (Empirical studies of trade)
• F22 (International migration)

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ Immigrants can generate additional demand for goods from their source countries, both directly, through their own consumption (as when, in their utility function, they attach a higher weight to the consumption of products from their origin country), or indirectly, by affecting natives’ preferences (through a diffusion of preferences effect) (Rapoport 2018).

² In the original UN dataset, among the 232 countries or areas included, 213, representing 92% of the total, had at least one data source on the total migrant stock since the 2000 census round, 75% of countries or areas had at least one data source on the age of international migrants, and 81% of countries or areas had at least one data source on the origin of international migrants. Moreover, the dataset refers to ‘immigration’ in a country as the stock of residents in that country who hold a foreign citizenship; not considering people that have acquired local nationality. It also only refers to people residing formally, neglecting undocumented immigrants. Moreover, the data do not include undocumented immigrants and economically active and inactive migrants.

³ In the Appendix Table A1 we present the number of trading partners by OECD country in our sample

⁴ The dataset does not provide information about Norway.

⁵ As a robustness, we estimated our models also in real terms. In the Table A2 of the Appendix, indeed, we estimated our models ‘in real terms’ by including in the equation the real GDP of both exporter and importer countries, to see the effect of the migration on the quantity exported by countries, and the results are in line with those of our preferred specification.

⁶ The EM-DAT database is maintained by Centre for Research on the Epidemiology of Disasters (CRED) within the Université catholique de Louvain (UCLouvain) in Belgium, providing data from 1900-present on more than 200 countries. The dataset follows a country level data collection methodology and the criteria for an event to be recorded (at country level) are there must be (1) at least 10 or more reported deaths due to the event and/or (2) at least 100 or more people affected and/or (3) there must be a declaration of emergency by the State and/or (4) a call for International Assistance. Data are available at EM-DAT, CRED/UCLouvain, Brussels, Belgium – www.emdat.be.

⁷ $0.34\times 1/24,714\times 1.4billionUS\mathrm{\$}\approx 19,260dollar.$.

⁸ The classification of developed countries is made according to the classification of the United Nations Statistical Division (originally published as Series M, No. 49, and now commonly referred to as the M49 standard).