Mobile money and multidimensional energy poverty: a cross-national study of Burkina Faso and Togo

Abstract

Using data from the FinScop survey (2016), this study aims to analyse the effect of mobile money on multidimensional energy poverty (MEPI) in Burkina Faso and Togo. MEPI was calculated using the method of Alkire and Foster in 2011, and a linear regression based on the instrumental variable strategy was applied to control for endogeneity bias arising from the dual causality between energy poverty and mobile money. To test the robustness of our results, we used the endogenous switching regression (ESR) model to resolve the problems of self-selection bias and endogeneity. The average effects of treatment on the treated (ATT) and the untreated (ATU) were calculated from the coefficients of the ESR models. The incidence of multidimensional energy poverty was estimated at 90.7 and 91.1% in Burkina Faso and Togo, respectively. In addition, we found substantial differences between the subgroups in terms of multidimensional energy poverty in each of the countries. The results also robustly indicate that an increase in mobile money adoption per standard deviation is associated with a reduction in multidimensional energy poverty of −0.402 standard deviations in the case of Burkina Faso and −0.628 standard deviations in the case of Togo. We argue that mobile money can be an effective policy tool in the fight against energy poverty in developing countries.