A Note on the Profitability of African Banks: Islamic versus Conventional

Transcription

1. A Note on the Profitability of African Banks : Islamic versus Conventional Whelsy Boungou* 04 May 2020 Abstract This paper aims to examine and compare the profitability of Islamic and conventional banks located in 20 African countries over the period 2009-2018. Based on a sample of 21 Islamic banks and 297 conventional banks, this study shows that Islamic banks perform better than conventional banks in Africa. This result is driven by large banks. JEL codes: E44, G21, G32, L10, O16 Keywords: Profitability, Islamic Banks, Business cycle, Panel data, Africa. * University of Bordeaux – LAREFI. E-mail address: whelsy.boungou@u-bordeaux.fr. 1
2. 1 . Introduction What are the determinants of the profitability of Islamic and conventional banks in Africa? The Global Financial Crisis of 2007-2009 has reminded policy makers and academics of the importance of understanding the environment in which banks operate. To this end, this paper aims to analyze and compare the determinants of profitability of African Islamic and conventional banks. Following Goddard et al. (2004), we also investigate whether these determinants differ according to bank size. We contribute to the extensive literature on the determinants of bank profitability in several ways. First, while the previous literature has focused mainly on individual countries in Africa (see, Bennaceur and Goaied, 2008, Chinoda, 2014, Kiganda, 2014) and the rest of the world (see, Molyneux and Thornton, 1992, Goddard et al. 2004, Athanasoglou et al. 2008, Garcia-Herrero et al. 2009, Dietrich and Wanzenried, 2011, Vera-Gilces et al. forthcoming), we complement this literature by conducting a crosscountry analysis of the determinants of profitability of banks operating in 20 African countries. Second, we compare the determinants of profitability of Islamic and conventional African banks, which at our level of knowledge has not been explored so far. The latest novelty of this paper is the use of the most recent data from 318 African banks over the period 2009-2018. This goes beyond previous work that has mainly focused on periods before the crisis (among others, Athanasoglou et al. 2008, Bennaceur and Goaied, 2008, Garcia-Herrero et al. 2009, Dietrich and Wanzenried, 2011). Using data from 21 Islamic and 297 conventional banks, our results indicate that Islamic banks perform better than conventional banks in Africa. We find that African banks (Islamic and conventional) with better management efficiency are more profitable. We also show that the determinants of profitability differ between Islamic banks and conventional banks. Indeed, while large Islamic banks with fewer liquid assets tend to be more profitable, the most profitable conventional banks are better capitalized and more dependent on deposits. Finally, we highlight that these results are driven by the large banks. This paper is organized as follows. Section 2 gives descriptions of our data and empirical model. Section 3 summarizes our results. Section 4 concludes this study. 2. Data and empirical model 2.1. Data Our database corresponds to an unbalanced panel dataset of 318 banks operating in 20 African countries1 from 2009 to 2018. Our bank sample consists of 21 Islamic banks and Algeria, Burkina Faso, Chad, Côte d’Ivoire, Djibouti, Egypt, Gambia, Ghana, Guinea, Libya, Mali, Mauritania, Morocco, Niger, Nigeria, Senegal, Sierra Leone, Sudan, Tanzania, Tunisia. 1 2
3. 297 conventional banks . While the balance sheet data of the banks was extracted from the Fitch Connect database, the macroeconomic indicators were obtained from Datastream. We have sorted our database by winsorizing the data at the 1st and 99th percentile level to ensure that outliers do not bias our estimates. The profitability measures. In order to measure the profitability of banks, we use two measures that are commonly used in the literature, namely return on assets (ROA) and return on equity (ROE)2. While ROA is measured as net income to total assets, ROE is measured as net income to shareholders' equity. These two profitability measures reflect how banking management uses the bank's actual investment resources (assets or equity) to generate profits. The evolution of these profitability measures is shown in Figure 1. Figure 1. Average ROA and ROE among Islamic banks (red line) and conventional banks (blue line) from 2009-2018. The bank-specific controls. According to the previous banking literature, we include in our analysis six bank-specific variables that influence bank profitability. Among these variables, we use the natural logarithm of total assets as a proxy for bank size. To measure the capitalization of banks, we use the equity to assets ratio. Cost to income ratio is used as a proxy for bank efficiency. Liquid assets to total assets measure the liquidity of banks. Bank funding refers to the ratio of customer deposits to total assets. To determine the banks' overall credit supply, we use net loans to total assets. The country-specific controls. Among the country-specific controls present in our regressions is the Herfindahl-Hirschman index (HHI), which is a measure of market structure. To account for the business cycle, we include the annual growth rate of the consumer price index (Inflation) and the real GDP growth rate (GDP). We use three other proxies for bank profitability as robustness checks: net interest income to total assets; net income to customer deposits and profit before tax to total assets. To preserve space, the results are not tabulated but available upon request. 2 3
4. 2 .2. Model In order to investigate the profitability determinants of Islamic and conventional banks located in Africa, we use the standard OLS model with fixed effects3. Equation (1) summarizes our baseline model: where is the banks’ profitability measures, alternatively ROA and ROE, for bank i in country j at year t. Using the Variance Inflation Factor (VIF), we test the control variables for multicollinearity. A mean VIF of 1.54 suggests that our control variables are not highly correlated. While refers to bank-specific controls, refers to country-specific controls. The selection of these control variables are based on the literature on the determinants of bank performance (e.g., Athanasoglou et al. 2008, Dietrich and Wanzenried, 2011, Goddard et al. 2004, Molyneux and Thornton, 1992). , and are respectively time fixed-effect, time-invariant bank fixed-effects, and idiosyncratic error. In all our regressions, we use robust standard errors clustered at bank-level to control for heteroscedasticity and dependence. 3. Empirical Findings In this section, we analyze and compare the determinants of profitability of Islamic and conventional African banks. Based on the analysis of Goddard et al. (2004), we then examine whether the determinants of profitability of African banks differ by size. The results of these analyses are reported in Tables 1 and 2. Related to the literature on the determinants of banks profitability (see, Goddard et al. 2004, Athanasoglou et al. 2008, Dietrich and Wanzenried, 2011), we perform dynamic panel GMM estimations to control for potential endogeneity bias. We find similar results to our baseline (not reported but available on request). 3 4
5. Table 1 . Determinants of ROA and ROE ROA Size Capitalization ROE Islamic banks Conventional banks 1.217 0.175 (1.19) (0.16) *** 0.097 0.065 (0.01) (0.09) *** Conventional banks 9.775*** (3.18) (2.04) 0.016 -0.151 (0.24) (0.10) *** 1.458 Efficiency -0.053 (0.02) -0.052 (0.01) -0.485 (0.07) -0.392*** (0.04) Liquidity -0.095 0.003 -0.028 (0.07) (0.00) -0.302* (0.15) 1.521 * Funding ** Islamic banks (0.03) 0.737 (0.43) -4.841 3.019 (2.44) (11.17) (3.57) Loans -0.051 -0.002 -0.166 -0.045 (0.06) (0.00) (0.15) (0.03) HHI 27.230 -0.375 216.485 10.939 (24.13) (1.13) (144.90) (12.08) GDP -0.015 -0.002 (0.04) (0.00) Inflation 0.107* (0.06) 0.028 Constant -4.646 * 0.002 0.211 (0.11) (0.01) (0.02) 0.485*** (0.14) 0.092* (0.05) 1.404 -35.028 23.615 (10.24) (2.02) (28.23) (25.82) Observations 135 1751 135 1749 Number of banks 21 297 21 297 R-squared (within) 0.412 0.342 0.644 0.306 Bank FE Yes Yes Yes Yes Year FE Yes Yes Yes Yes Note: Robust standard errors clustered by banks in parenthesis. ∗∗∗, ∗∗ and ∗ indicate statistical significance at 1%, 5%, and 10%, respectively. Table 1 presents our main findings. All our estimates include bank and year fixed effects. According to Table 1, Islamic banks lead in most of the determinants of profitability. Our findings highlight that Islamic and conventional banks do not have the same determinants of profitability. Moreover, these determinants of profitability are different depending on the profitability measure used (ROA or ROE). We show that ROE is a better indicator of profitability for Islamic banks while ROA is a better indicator for conventional banks. To summarize the results of Table 1, the most profitable Islamic banks are those that are large, with better cost management, and hold fewer liquid assets. For conventional banks, our results indicate that the best performing conventional banks hold more capital and deposits, and are more efficient4. Using annual data of 2013 on 44 Islamic banks from Asian and African regions, Chowdhury and Rasid (2015) find similar results. 4 5
6. We also find a positive and significant relationship between inflation and profitability of African banks , which implies a better expectation of inflation on the part of both Islamic and conventional banks. Indeed, as Chowdhury and Rasid (2015) point out, when banks anticipate a change in inflation they adjust their balance sheet in order to increase their profits. Table 2 reports the results of the analysis of the determinants of bank profitability as a function of size. We divide our sample into subsets around the median of the natural logarithm of assets. A bank is considered small if its size is below the median and large if its size is above the median. Consistent with the results presented in the previous paragraph, our results indicate that Islamic banks perform better than conventional banks in Africa. We highlight that the performance of Islamic (or conventional) banks in Africa is mainly due to large banks. Table 2. Determinants of ROA and ROE according to bank size ROA Islamic banks Size Capitalization ROE Conventional banks Islamic banks Conventional banks Small Large Small Large Small Large Small Large 0.167 0.163 0.251 0.148 3.045 1.170 1.173 (0.54) (0.28) (0.33) (0.17) (2.66) 12.143*** (3.53) (4.61) (1.96) 0.011 0.052 -0.172** (0.06) -0.064 (0.03) 0.075*** (0.02) -0.012 (0.02) 0.065*** (0.01) (0.28) (0.13) -0.491*** (0.14) *** *** *** Efficiency -0.065 (0.02) -0.088 (0.02) -0.052 (0.01) -0.052 (0.01) -0.368 (0.09) -0.693 (0.13) -0.297*** (0.06) -0.522*** (0.04) Liquidity -0.018 -0.008 -0.002 0.018 (0.01) (0.00) (0.11) -0.299** (0.12) 0.049 (0.02) 0.010* (0.01) -0.117** (0.05) Funding Loans *** *** -1.071 -1.820 0.779 0.413 -4.098 (0.73) (2.66) (0.86) (0.43) (4.35) 0.000 0.002 -0.003 -0.001 0.073 (0.07) (0.01) (0.02) *** * (0.04) 4.885 3.592 (6.34) (4.82) 0.014 (0.03) -0.115** (0.05) 899.158 (271.86) 27.199** (12.22) 19.772** (9.73) 0.451*** (0.10) -0.015 (0.01) 0.173** (0.09) 0.086 -0.007 -33.693 (15.78) -0.244* (0.12) (0.00) (0.00) 84.364 (32.77) -0.183 0.441 (1.78) (0.71) -139.199 (75.43) -0.043 0.021 -0.003 0.005 0.035 (0.05) (0.02) (0.00) (0.01) (0.22) -0.008 0.003 0.030 0.001 0.061 (0.01) (0.02) (0.03) (0.01) (0.05) 0.224* (0.11) (0.09) (0.11) 4.976 0.861 1.837 8.069 -36.188 16.161 42.337 (4.21) 5.964** (2.63) (3.10) (2.61) (18.18) (34.57) (43.87) (29.19) Observations 73 62 695 1056 73 62 693 1056 Number of banks 14 12 137 187 14 12 137 187 R-squared (within) ** HHI -26.837 (12.05) GDP Inflation Constant ** * *** 0.617 0.737 0.351 0.389 0.715 0.809 0.333 0.372 Bank FE Yes Yes Yes Yes Yes Yes Yes Yes Year FE Yes Yes Yes Yes Yes Yes Yes Yes Note: Robust standard errors clustered by banks in parenthesis. ∗∗∗, ∗∗ and ∗ indicate statistical significance at 1%, 5%, and 10%, respectively. 6
7. 4 . Conclusion The Global Financial Crisis (GFC) of 2007-2009 has reminded policy makers and academics of the importance of understanding the environment (internal and external) in which banks operate. Previous work has shown that the behavior of banks before and/or during the crisis varied according to their status (Islamic or conventional) and location (low-, middle-, and high-income countries). This paper contributes to the extensive literature on the determinants of bank profitability by analyzing and comparing the determinants of profitability of 21 Islamic banks and 297 conventional banks operating in 20 African countries in the post-crisis period. Our results highlight that bank-specific characteristics explain a significant part of the evolution of the profitability of African Islamic and conventional banks. We find that Islamic banks perform better than conventional banks. Moreover, our results show that large banks, whether Islamic or conventional, do better than small banks. 7
8. References Athanasoglou P ., Brissimis S., Delis M., 2008. Journal of International Financial Markets, Institutions & Money, Vol. 18, pp. 121-136. Bennaceur S., Goaied M., 2008. The determinants of commercial bank interest margin and profitability: Evidence from Tunisia. Frontiers in Finance and Economics, Vol. 5, No. 1, pp. 106130. Chinoda T., 2014. The determinants of commercial banks profitability in Zimbabwe (2009-2014). Journal of Economics and Finance, Vol. 5, Issue 6, pp. 69-80. Chowdhury M., Rasid M., 2015. The determinants of the profitability of Islamic banks: a crosssectional study from Asia and Africa. International Journal of Business and Globalisation, Vol. 15, No. 3, pp. 375-388. Dietrich A., Wanzenried G., 2011. Determinants of bank profitability before and during the crisis: Evidence from Switzerland. Journal of International Financial Markets, Institutions & Money, Vol. 21, pp. 307-327. Garcia-Herrero A., Gavila S., Santabarbara D., 2009. What explains the low profitability of Chinese banks? Journal of Banking & Finance, Vol. 33, pp. 2080-2092. Goddard J., Molyneux P., Wilson J., 2004. The profitability of European banks: A cross-sectional and dynamic panel analysis. The Manchester School, Vol. 73, No. 3, pp. 363-381. Kiganda E., 2014. Effect of macroeconomic factors on commercial banks profitability in Kenya: Case of equity bank limited. Journal of Economics and Sustainable Development, Vol. 5, No. 2, pp. 46-56. Molyneux P., Thornton J., 1992. Determinants of European bank profitability: A note. Journal of Banking and Finance, Vol. 16, pp. 1173-1178. Vera-Gilces P., Camino-Mogro S., Ordenana-Rodriguez X., forthcoming. A look inside banking profitability: Evidence from a dollarized emerging country. The Quarterly Review of Economics and Finance, forthcoming. In Press. 8 View publication stats