Causal links between bilateral aid and exports: The Swiss case

This paper investigates the direction of Granger causality, if Granger causality there is, between Swiss bilateral official development aid and Swiss exports for a panel of 50 recipient countries over the period 1974-2013. To account for possible geographical effects, three specific group of recipient countries across the same period are considered, that is "Africa", "Latin America" and "Asia". To shed on light potential level effects, four other group of countries are considered according to the relative amount of aid, respectively exports, Switzerland has exchanged with a given recipient country across the all period. The former panel plus the seven specific group of countries and the fact that we test for Granger causality in both directions give us a total of 16 models. As a necessary first step, four different panel unit root tests are conducted on each series of all eight panels and results systematically indicate stationary series in levels. Then and to test for Granger causality, Dumitrescu-Hurlin Granger non-causality tests are conducted. Results indicate bidirectional Granger causality in all panels except for the "African" one where Granger causality seems to run only from exports to bilateral aid. This result is important in terms of policy implications, particularly for a donor country as Switzerland, since it implies that if aid towards a recipient country is cut, it is likely to cause Swiss export reductions to that same recipient country.