FDI and financial intermediation in Sub-Saharan African countries: Evidence from panel data

This study aims to examine the relationship between financial development and foreign direct investment (FDI) in Sub-Saharan African (SSA) countries. Some previous studies have argued that an increase in FDI net inflows increases the funds available in the economy and leads to an increase in financial intermediation through financial markets or banking systems (Desai et al., 2006; Henry, 2000). Others have argued that a well‐functioning financial market is likely to channel foreign investments more efficiently into productive sectors, thereby creating more value for investors and making the country more attractive to FDI (Otchere et al., 2016). Despite this undeniable link between FDI and financial development, very little research has been conducted on the causal relationship between FDI and financial development. Even where such studies have been conducted, the findings have been inconclusive. The focus of most of the previous studies has been largely on Asia and Latin America, thereby leaving many SSA countries with little or no coverage. In addition, some of the previous studies based their inferences on a bivariate Granger-causality test, which is known to suffer from the omission-of-variable bias. As has been previously reported, the inclusion of an additional variable in a bivariate causality setting has the potential not only to alter the magnitude of the estimates, but also to change the direction of causality. It is against this background that the current study aims to examine the causal relationship between FDI and financial development using panel data from Sub-Saharan African countries. Considering the fact the financial markets in many SSA countries are still at their development stage, the study employs bank-based financial indicators to measure the level of financial development in the studied countries. Specifically, the study uses four proxies of bank-based financial development against FDI/GDP. In order to address the omission-of-variable bias, which is inherent in a bivariate Granger-causality model, the study incorporates economic growth as an intermittent variable between FDI and financial development, thereby leading to a trivariate Granger-causality model. The study uses Pedroni (1999) panel co-integration test to investigate the long-run relationship between the various proxies of bank-based financial development and FDI in a multivariate setting; and the panel Granger-causality test to examine the causal link between these variables. Our results show that the causal relationship between FDI and financial development varies significantly, depending on the variable used as a proxy for financial development. It is also sensitive to time lag.