Dividend policy, agency cost and bank performance in sub-Saharan Africa.

Abstract

This study explored the relationship between dividend policy, agency costs and bank performance among sub-Saharan African (SSA) commercial banks. More specifically, it examined the factors that determine the dividend payout ratio of commercial banks in this region; established the direction of causality between the dividend policy and financial performance of commercial banks in sub-Saharan Africa; determined the effects of operational diversification on these banks’ financial performance and evaluated the relationship among dividend policy, agency costs, market risk and performance in SSA commercial banks. The study was motivated by the desire to assist banks to formulate a balanced dividend policy that will enable them to contribute to economic growth and thus ameliorate the poverty, underdevelopment and poor financial depth that characterise the region. Data were collected from 250 commercial banks from 30 SSA countries using BankScope for the period 2006 to 2015. The data were analysed using descriptive statistics and inferential statistics with different econometric techniques, namely, static regression via pooled, fixed and random estimations and dynamic regression analysis via Panel Generalised Method of Moments (GMM), Panel Vector Error Correction Model (P-VECM), the Panel Granger Causality test, Impulse response function and Variable decomposition. Using both Differenced and System GMM, the study found that the lagged dividend payout ratio, after tax income, size and leverage are the key determinants of the dividend payout ratio in SSA commercial banks. The analysis of the causal relationship between dividend policy (dividend payout and retention ratio) and bank performance revealed unidirectional causality between the retention ratio and Return on Assets as well as between Return on Equity and the dividend payout ratio. The study also found that none of the dimensions of operational diversification have a significant effect on financial performance in the static regression analysis, while the GMM analysis (dynamic analysis) showed that, past year performance (lagged Return on Average Assets), asset diversification, deposit diversification, loan diversification and income diversification have a significant effect on banks’ financial performance (Return on Average Assets). In addition, a long-run relationship was identified between dividend policy, agency costs, and market risk and bank performance. The disequilibrium from the long run estimate will take about 39.5% annual speed of adjustment to return to a steady state. In terms of the two proxies of market risk, the interest rate risk has a negative effect, while the foreign exchange risk has a significant positive effect on variations in bank performance in sub-Saharan Africa. The evidence from the impulse response function and variable decomposition shows that all the variables in the series responded to shocks in performance (ROA) directly or indirectly during the investigated period, with dividend policy and agency costs the most significant. These findings imply that SSA banks should curtail payment of dividends as the current situation warrants re-investment of earnings to boost their assets and make a meaningful contribution to the region’s economic growth. Among others, this study recommends policies to improve dividend policy formulation in such a way that the agency costs of debt and equity will be minimised, and all the banks’ stakeholders’ interests will be protected to promote the future growth of the sector. It contributes to the extant literature by examining dividend policy with a regional focus using data from 30 SSA countries and identifying the major bank-specific determinants of the dividend payout ratio that can serve as a uniform formula for dividend payments across the region. Furthermore, this is the first study to establish that only dividend retention policy can cause bank performance in this region. Finally, the study used the Herfindahl-Hirschmann Index to measure the diversification of four major dimensions of banking operations and is the first of its kind in the SSA region to evaluate the relationship between dividend policy, agency costs, and market risk and bank performance using long-run analysis.