Causal Relationship between Air Transpor (1)

The aim of this study is twofold: to describe the overtime changes in trends and to investigate the causal relationship between air transport, tourism and economic growth for South Africa between 1995 and 2015. Using an ecological design, Joinpoint regressions tailored at p<0.05 were computed for each variable to determine annual percentage changes. Causality and co-integration were inferred through the Granger causality and the Johansen co-integration tests. Additionally, a vector auto-regressive model (VAR) was computed to test for linear inter-dependencies among the variables. Significant increases were observed for all variables between 1995 and 2015. The results of the stationary tests showed that air transport variables were stationary at the first differences while the tourism variables and GDP variables were stationary at second difference. Cointegration tests can be applied on series that are stationary at the same level. Therefore no further inference was made on the relationship between air transport variables and GDP. Both the cointegration and causality tests did not provide evidence of causality and long run relationships between GDP and tourism variables. However, a proportion of the variance in tourism expenditure and tourism receipts was shown to be explained by GDP through the VAR model. The results imply potential associations between tourism and GDP in South Africa.