Net international migration: A panel analysis of economic determinants

Abstract

Various factors can motivate and encourage individuals to leave their country of origin and engage in the process of international migration. Large migration flows over the past few years, which are a consequence of the so-called refugee crisis in 2015, have resulted in a significant increase in academic interest in international migration. Although many factors can encourage international migration, people’s desire to increase their standard of living is undoubtedly one of the most important causes.

This empirical study analyses the economic determinants of net international migration. It was conducted using techniques of econometric analysis of panel data on an extensive balanced panel data sample covering 136 countries over a period of 30 years (1990‒2019). Although this study focuses on the analysis of the economic determinants of net international migration, it is necessary to point out that in this case it is not a one-way impact. Apart from the fact that economic indicators affect international migration, it is indisputable that there is an effect in the opposite direction, i.e. that migration flows significantly affect the economic performance of the origin and destination country. This influence can be realised through different mechanisms. The potential presence of reverse causality generates a problem of potentially endogenous regressors, which must be considered when selecting model estimation techniques.

The estimation of the models was performed using the following two techniques that allow cross-section dependence: (i) standard common correlated effects pooled estimator (CCEP), which is based on the application of the ordinary least squares method; and (ii) modified common correlated effects pooled estimator, which is based on the application of the two-stage least squares method, allowing the presence of endogenous regressors.

The obtained findings suggest that the impact of the unemployment rate on net international migration is negative. Estimates generated by the standard CCEP technique (as well as by the modified CCEP technique) show that an increase in the unemployment rate of 1% results in a decrease in net international migration by about 0.03 migrants (0.06 migrants) per 1,000 inhabitants. Education has a positive impact on net international migration. The standard CCEP technique (modified CCEP technique) points to the fact that a 1% increase in education results in an increase in net international migration of about 0.01‒0.02 migrants (0.003 migrants) per 1,000 inhabitants. The level of development of the migrant network has a positive effect on net international migration. The standard and modified CCEP technique show that increasing the stock of migrants by 1% increases net international migration by about 0.04‒0.05 migrants, i.e. by about 0.01 migrants per 1,000 inhabitants. Finally, estimates obtained using the standard CCEP technique suggest that a 1% increase in per capita gross domestic product (GDPpc) results in an increase in net international migration of about 0.01 migrants per 1,000 inhabitants, while the results of the modified and more credible CCEP technique show that growth in GDPpc by 1% implies a decrease in net international migration by about 0.02 migrants per 1,000 inhabitants. The negative effect of GDPpc could be rationalised by the existence of an inverted‒U relationship between GDPpc and emigration (mobility transition curve). Acceptance of such an explanation requires that the following conditions be met: (i) there is an inverted-U relationship between emigration and GDPpc; (ii) the countries in the sample fit within the growing part of this relationship; and (iii) when GDPpc grows, emigration increases absolutely more than immigration.