Islamic banking and real performances in a dual banking system Evidence from Saudi Arabia

Purpose - There is a rich debate on the nature of Islamic banking (IB)-growth nexus and the direction of causality governing this nexus. This study aims to focus on this issue in the case of Saudi Arabia, the largest country-holder of Islamic Banks (IBs)' assets worldwide. It assesses empirically the nature of dynamic interactions between IBs' financing and the real performances in the non-oil private sector (investment and GDP) in the context of a dual banking system where IBs operate alongside their conventional counterparts. Design/methodology/approach - This study employs the Bounds test in the context of reparametrized autoregression distribution lags (ARDL) models to analyse both long-run and short-run dynamics governing Islamic and conventional banks' (CBs) financings on one hand and real investment and GDP in the private sector on the other hand over the 2007q1-2016q4 period. It also uses the Toda and Yamamoto ( 1995) augmented Granger-causality test to assess the direction of causality governing these dynamics. Findings - The more important results are: there is a stable and significant long-run relationship between IBs' financing and real performances in the private sector. This nexus is governed by the "feed-back hypothesis", implying the validity of both the "supply-leading" and the "demand-following" hypotheses. In a dual banking system context, IBs exert two effects on the financing of their conventional counterparts: a negative "crowding-out" effect and a positive and "stimulating" effect which transmits through the "competition" channel. Finally, in the long-run, steady-state, real GDP is dissociated from CBs' financing. Originality/value - This paper highlights an issue that has not received the needed attention in the case of Saudi Arabia. It has also found novel results with important policy implications.