Can International Migration Accelerate Development? A Global Dynamic General Equilibrium Analysis

Policies to facilitate international migration and targets for reductions in remittance costs faced by migrant workers are set to be part of the emerging post-2015 development agenda. This is a recognition of significant linkages between international migration and the achievement of the post-2015 development goals, and is a response to the fact that total remittance flows to developing countries are already a multiple of international development assistance flows. Global demographic shifts over the coming decades are bound to magnify the economic incentives for South-North migration and reinforce the economic case for a reduction of existing barriers to international labor mobility. The domestic labor supply has already peaked in high-income countries as a whole. It is set to decline steadily over coming decades while hundreds of millions of new workers are projected to enter the labor force by 2030 in developing countries as a group. Moreover, given the considerable variety in demographic dynamics and labor productivity levels across developing regions, there is potentially also considerable scope for mutual gains from further South-South migration. Correspondingly, forward-looking assessments of the prospective economic impacts of future changes in policies toward cross-border migration flows deserve a high priority on the global development research agenda. Aims This study adopts a global dynamic computable general equilibrium simulation approach to provide a regionally differentiated quantitative assessment of the incremental economic benefits resulting from a marginal relaxation of existing restrictions on international migration flows. The simulation analysis will also assess the welfare impacts of a gradual reduction in remittance transaction costs to the target levels envisaged in the current draft proposal for the post-2015 sustainable development goals. This latter simulation scenario will take account of recent empirical estimates of the elasticity of remittances with respect to remittance costs reviewed in McKenzie and Yang (2014). The existing previous global CGE-model-based studies of gains from further international labor migration (e.g. Walmsley and Winters, 2005; World Bank, 2006; Walmsley et al., 2007) focus predominantly on South-North migration impacts. However, in terms of absolute headcount figures, the present observed extent of South-South migration is nearly as large as that of South-North migration (UNESA, 2012; Ratha and Shaw, 2007; Bakewell, 2009). Heterogeneity in demographic trends, as well as wage differentials across regions within the “Global South,” suggests non-trivial potential gains from further South-South migration. Thus, the present study includes a quantification of the potential gains from an incremental increase in South-South migration flows, starting from observed South-South migration patterns. The analytical framework is a modified version of the recursive dynamic global CGE model LINKAGE. An earlier version of this model has been used in an assessment of potential gains from further international migration reported in the World Bank Global Economic Prospects Report 2006. The new extended version of the model will be calibrated to the recent GMig2 extension of the GTAP 8.1 database, which contains the latest available model-consistent estimates of bilateral migration stocks, labor earnings and remittance flows at GTAP 8.1 regional aggregation level as described in Walmsley et al (2013). The construction of a dynamic baseline up to 2030 under the assumption of no changes in the stance of international migration policies will be based on the latest World Bank global economic projections including UNDESA population and labour force growth projections. As the latter already contain assumptions about the evolution of migration flows over the simulation horizon, it is important to back out these assumptions at the dynamic model calibration stage to arrive at a methodologically clean separation of changes in migration implicitly built into the baseline and changes in migration due to deviations from the baseline migration policy path. Attention to this important detail appears to have been neglected in respective previous modelling work. The existing CGE studies capture remittance effects and direct wage effects on origin countries, but largely ignore other sending country impact channels identified in the literature. These channels include in particular potential productivity impacts associated with return migration and potential brain gain effects arising from incentives to invest in human capital formation in the presence of expected future migration opportunities. As Kerr and Kerr (2011) emphasize, proper accounting for return migration is essential for determining the economic impacts for both origin and host countries, given the available evidence on the extent of return migration from the main host regions and existing empirical estimates of possible associated benefits for the home country (e.g. Mayr and Peri, 2008; Dustmann and Weiss, 2007, De Vreyer, Gubert and Robilliard, 2010). With respect to brain gain effects, recent econometric evidence seems to point to a “robust, positive and sizeable effect of skilled migration prospects on human capital formation in developing countries” (Beine, Docquier and Rapoport, 2010; see Docquier and Rapoport, 2012 for qualifications). The present study aims to incorporate these additional impact channels in a stylized form. In each case, the calibration of the respective new model parameters that govern the size orders of these effects will be based on a review of the pertinent recent empirical literature, so that the model-based simulation results can credibly inform ongoing controversial debates about the relative importance of these impact channels. Back-of-the-envelope calculations as well as previous model-based simulation studies suggest that the potential net benefits from reducing barriers to international labor mobility are large. As Clemens (2011) has put it, “(r)esearch on this question has been distinguished by its rarity and obscurity, but the few estimates we have should make economists’ jaws hit their desks”., as these benefits “may be much larger than those available through any other shift in a single class of global economic policy”. We do not expect that our new results - which will be based on more recent and better data and incorporates a wider range of impact channels – will overturn this broad conclusion. However, the attention to factors that could qualify the development impact of migration, such as the cost of remittances, will help marry the literature on the overall gains of migration to specific interventions.