Abstract:
In this article we re-examine the relationship between remittances and economic growth placing special attention on the nonlinearity of this relationship. Previous studies have ignored the non-linearity of the relationship between remittances and economic growth or have used a quadratic term to capture nonlinearity. We show that the relationship between remittances and growth is neither linear nor quadratic and propose the use of a semiparametric model to avoid the risk of misspecification bias from imposing an arbitrary functional form. We find evidence of a positive relationship between remittances and growth in parametric estimations; however, such a relationship disappears when nonlinearity is taken into account using non-parametric techniques.