Purpose The purposes of the paper are as follows: (1) Analysing the effect of financial inclusion on financial stability. (2) Examining whether financial inclusion non-linearily impacts financial stability. (3) Analysing whether the effect of financial inclusion varies across quantiles of financial stability. (4) Investigating whether dimensions of financial inclusion affect financial stability differently. (5) Examining whether the effect of financial inclusion on financial stability depends on competitiveness of the banking industry.

Design/methodology/approach Using panel data for 19 African countries for the period 2006–2022, we first developed multidimensional index of financial inclusion using two-stage indexing approach. Then employing panel semiparametric regression, we analyse the non-linear nexus between financial stability and financial inclusion. We further employ panel quantile regression to investigate the differential effect of financial inclusion at different quantiles of financial stability. We also employed two-stage least squires, and alternative measurement of financial stability as robustness checks.

Findings Employing panel semiparametric regression, we demonstrate that the financial inclusion-stability nexus exhibits non-linearity: below (above) threshold level financial inclusion promotes (reduces) financial stability. Employing panel quantile regression, we find that the effect of financial inclusion increases at higher quantiles of financial stability. We further demonstrate that the effect of financial inclusion on financial stability is pronounced in a more competitive bank industry. The findings are robust to two-stage least squares estimation, and alternative measurement of financial stability. The results suggest that keeping a balance between achieving stable and inclusive financial system, and ensuring a competitive banking industry are essential to achieve bank soundness while promoting financial inclusion.

Originality/value The study incrementally contributes to the literature related to the financial inclusion – stability nexus in four-fold. First, unlike studies that relied on some indicators of financial inclusion, we employed the effect of multidimensional financial inclusion on financial stability and further examined whether or not the effect varies across financial inclusion dimensions. Second, unlike studies that assumed a linear nexus between financial inclusion and stability, employing panel semiparametric regression, we investigated for non-linear relationship between the two. Employing a novel panel quantile estimation approach, we further scrutinised whether the effect of financial inclusion varies across quantiles of financial stability. Third, to our knowledge, our study is the first to examine the effect of multidimensional financial inclusion on bank soundness in Africa.