The institutional determinants of financial development : evidence from the Southern African Development Community (SADC)

Abstract

This study provides evidence on the role of democratic institutions in fostering financial development in ten economies in the Southern African Development Community classified into three income groups from 1975 to 2013. Polity IV variables, considered as measure of democracy are applied to quantify institutions, while bank deposits, private credit and liquid liabilities proxy financial development. Initially, panel regressions are estimated using Ordinary Least Square and Instrumental Variable estimators and find evidence of a linear and non-linear association between democratic institutions and financial development. The instrumental variable regression allows for the joint endogeneity of regressors through the use of internal instruments. Evidence from panel regressions suggests that democratic institutions are positively associated with financial development for the upper middle income countries. A negative relationship is found for the lower middle and low-income countries. Applying Bayesian Vector Auto-regressions and variance decomposition of annual proxies for financial development, the study determines the most important institutional variables that account for variations in financial development. The use of Bayesian inferences mitigates estimation risk, sample bias, over-parameterisation and provides better forecasting. The results show that shocks to democratic variables positively affect financial development in the upper income countries, with substantive democracy and human capital development contributing the most towards variations in financial development while the effect is negative for the other income groups but, however, improved slightly after 1990. The Markov switching model is applied to show evidence of fragility of financial development in the Southern African Development Community region. Results from the Markov Switching Model find that the countries switch between regimes of high and low financial development following political regime changes. Countries in the lower middle and low-income groups show a greater probability of being in a regime of low financial development while the upper income countries show a higher probability of being in a state of high financial development. These probabilities improved for the lower middle and low-income groups after 1990, a period associated with increasing democratisation. The Markov findings suggest that democratic institutions may have helped promote financial development despite the inherent instability of the financial system. Therefore, it seems that appropriate government policies in strengthening democratic institutions may mitigate instability and foster financial development with the likely potential of producing long-term gains in the financial sectors for all the income groups.