Bank regulation and bank lending in selected sub-Saharan African countries
Loading...
Authors
Thamae, Retselisitsoe I.
Issue Date
2023-07
Type
Thesis
Language
en
Keywords
Bank regulation , bank regulatory measures , bank lending , bank credit , linear and nonlinear panel autoregressive distributed lag (ARDL) , dynamic common correlated effects (CCE) , dynamic panel threshold regression (PTR) , generalised method of moments (GMM) , symmetric and asymmetric panel Granger causality , error correction , sub-Saharan Africa
Alternative Title
Abstract
Most countries, including the ones in sub-Saharan Africa (SSA), have been facing
pressure to increase the stringency of bank regulation since the aftermath of the
2007-2008 global financial crisis. Therefore, this study analyses the relationship
between bank regulation and bank lending in selected SSA countries and/or their
low-income and middle-income groups from 1995 to 2017. First, the empirical results
from the linear panel autoregressive distributed lag (ARDL) model estimated through
the dynamic common correlated effects (CCE) method showed that bank entry
barriers impacted bank lending negatively in the long run in all selected SSA
economies and low-income SSA countries, while macroprudential policies had a
negative long-run impact on bank credit in middle-income SSA economies, but
supervisory power mitigated these effects. Alternatively, the findings from the
nonlinear panel ARDL model indicated that various shocks to bank regulatory
measures affected bank lending differently. Second, the dynamic panel threshold
regression model results estimated through the generalised method of moments
approach revealed that the threshold values for the stringency of bank entry barriers
and capital regulations in the case of all selected SSA economies were 62.8% and
76.5%, respectively. The effect of bank entry barriers stringency on bank credit was
found to be positive below its threshold value but negative above it, while that of
capital regulation stringency was insignificant either below or above its threshold
level. Lastly, the empirical findings from the panel error correction-based Granger
causality models generally highlighted that long-run causality existed between bank
regulatory measures and bank lending in the context of selected SSA countries and
their income groups, while various shocks to bank regulatory measures and bank
credit had different causal effects. Thus, the study recommends that policymakers
should not introduce bank regulatory and supervisory reforms for their own sake,
since regulations that are too stringent could hamper bank credit. Furthermore,
regulatory authorities should take into consideration the existence of nonlinear and
threshold effects in the relationship between bank regulation and bank lending as not
doing so could lead to biased estimates and result in wrong policy conclusions.