This paper investigates the short-run and long-run causal relationships that may exist between a set of variables that are selected to proxy for components of expenditure based GDP for eight European countries, namely Cyprus, France, Germany, Greece, Ireland, Italy, Portugal, and Spain. Due to the identification of I(1) cointegrated variables, the analysis is performed within a VECM framework, that models each country individually as a closed economy, and then as an open economy. The estimated variables are then used to provide out-of-sample short horizon forecasts of GDP, which are compared to actual GDP data. The results indicate that the estimated open economy VECM outperforms the closed economy VECM, but only for open economies within the sample.