The Nature and Impact of Repeated Migration Within Households in Rural Ghana

In this paper, we explore repeated migration within a household and consequent welfare outcomes. Specifically, we use a household panel survey collected in 2013 and again in 2015 in rural areas of Ghana. We exploit the rich information about migration experience as well as the panel nature of the data to overcome sources of selection, omitted variable and reverse causality bias. Wealth is measured with an index including indicators of housing quality. We apply Multiple Correspondence Analysis (MCA) to construct the welfare index and entropy-balancing weights (Hainmueller, 2012) to obtain a valid comparison group for our analysis. We provide evidence that households often have more than one migrant member and that they have different characteristics depending on who moved first. New migrants are more likely to be from a younger generation, they face lower migration costs, and only few of them remit. We find no effect of sending a new migrant on the change in the asset index. We conclude that the different nature of migration of new migrants implies neither an economic gain for the household nor a loss. The reason for the former is that the migrants remit less or not at all and the reason for the latter is that migration becomes less costly with prior experience