The Causal Relationship Between Saving and Growth : Case of Colombia, Mexico, Sweden and the UK

This paper uses a vector error correction model (VECM) and the unrestricted vector autoregressive (VAR) model combined with Granger causality in order to determine the casual direction between saving and growth. The countries being studied are Colombia, Mexico, Sweden and the UK. First, the stationarity of the variables are tested by employing an Augmented Dickey-Fuller test. Secondly, the long-run relationship is tested through a Johansen cointegrating system. For those countries that have cointegrated variables the VECM model is used. For those lacking such a long-run relationship a standard VAR model is used and further analysis is done through Granger non-causality testing. The results from the cointegration analysis suggests a long-run relationship for Sweden and the UK. This was further attested to by the VECM analysis. For Colombia saving only Granger-caused growth in the short-run and for Mexico no relationship was found as the Granger-causality test showed no causal link in any direction