Forecasting sectoral trade growth under flexible exchange rates

A huge body of empirical and theoretical literature has emerged on the relationship between exchange rate uncertainty and international trade. In empirical studies the estimated impacts of exchange rate uncertainty on trade figures are at most weak and often ambiguous with respect to their directions. Almost all empirical contributions on the topic start from the assumption of some linear relationship, the potentional of nonlinearity or state dependence of causal links between volatility and trade has been ignored yet. In addition, widely used regression models have not been evaluated in terms of ex-ante forecasting. In this paper we analyze the impact of exchange rate uncertainty on specific categories of exports and imports for 13 industrialized economies towards the rest of the world. Our results support the view that the relationship of interests might be nonlinear and, moreover, lacks of homogeneity across countries, economic sectors and when contrasting imports vs. exports. Parametric threshold models are found to outperform linear regression models in terms of fitting and ex-ante forecasting. In addition, semiparametric models deliver sequences of forecast errors with less dynamic structure than parametric specifications and help to uncover the nature of the nonlinear relation.