Capital Inflows-National Saving Dynamics in Tunisia: Evidence from Cointegration, Weak Exogeneity and Simultaneous Error Correction Modelling

The negative relationship between capital inflows and savings in less developed countries is an accepted fact in the existing literature. However, this result is based essentially on standard econometrics which ignores the nonstationarity of these two variables. This study investigates the ‘direct’ and ‘indirect’ effect of capital inflows on savings in Tunisia using Johansen's multivariate cointegration technique, weak exogeneity test and simultaneous error correction modelling. In the short and long run, the econometric estimates show that capital inflows have a negative effect on domestic savings, which invalidates the Chenery-Strout thesis. Nevertheless, the direction of causality between the aggregates being dealt with still remains a subject of debate. Relying on time-series data of the Tunisian economy, Granger's causality test shows a causal relationship in the long term running from domestic savings to capital inflows. In the short term, however, this paper reveals a two-way causation. [E20, F35, C32, C51]