Remittances, poverty and human capital: evidence from developing countries

Abstract

Purpose The purpose of this paper is to analyze the relationship between remittances and poverty through the human capital channel in developing countries, which has received less attention in the literature. Design/methodology/approach The paper applied the system GMM developed by Arellano and Bond (1991) and Arellano and Bover (1995) containing 54 developing countries. This estimator is appropriate compared to a cross-section technique because it controls for the endogeneity of all explanatory variables, includes unobserved country-specific effects and allows for the inclusion of lagged dependent variables. Findings The results suggest that, while remittances reduced poverty, the effect is moderated via education. A 1 percent increase in remittances reduces the poverty headcount by 0.47 percent, while the reduction is 0.33 percent via education. The marginal effect of remittances is negatively related to the level of education, indicating that education mitigates the effect of remittances on poverty. Practical implications This paper includes the implications for the policymakers to justify the need for more effective approaches. It is useful to identify whether and how remittances and human capital interact in their effect on poverty when deciding the most desirable allocation of available resources between these two priorities. Originality/value This paper takes a step forward filling the limited evidence on the role of human capital in remittances–poverty relationship in developing countries. Different from the existing studies which have used the traditional panel estimators, this study utilizes the dynamic panel estimators such as system GMM to tackle the specification issues of endogeneity, measurement errors and heterogeneity.