A Note on the Profitability of African Banks: Islamic versus

Conventional

Whelsy Boungou*

June 2022

Abstract

This paper aims to examine and compare the profitability of Islamic and conventional banks
located in 20 African countries over the period 2009-2017. Based on a sample of 21 Islamic
banks and 143 conventional banks, this paper shows that the determinants of performance
are almost the same between Islamic and conventional banks in Africa. Our findings
highlight that the specific characteristics (such as capitalization and efficiency) of
conventional and Islamic banks influence their performance in Africa. Precisely, we find that
large and deposit-dependent banks perform better.

JEL codes: E44, G21, G32, L10, O16

Keywords: Profitability, Islamic Banks, Business cycle, Africa.

* Paris School of Business. E-mail address: w.boungou@psbedu.paris.

Electronic copy available at: https://ssrn.com/abstract=3603046

1. Introduction
What are the determinants of the profitability of Islamic and conventional banks in
Africa? The Global Financial Crisis (GFC) of 2007-2009 has reminded policy makers
and academics of the importance of understanding the environment in which banks
operate. Moreover, Islamic banks have been more resilient to the GFC compared to
other banks (Pappas et al. 2017). The post-GFC period also corresponds to the
internalization of Islamic finance (Causse-Broquet, 2012). Consequently, emerging
from the crisis, Islamic banking has become an important part of the global financial
system and is increasingly expanding in Africa (IMF, 2015).

Although Islamic and conventional banks are by definition different, they participate
in the economic and financial development of countries (Alexakis et al. 2019). On the
one hand, we have Islamic banks whose primary vocation is not to get rich (or to
generate profit), in accordance with Islamic law (Shariah), as cash is not considered a
commodity (Hamdan, 2009). Conversely, conventional banks, being considered as
companies, will rather follow logic of performance and try to generate the maximum
profit (Pappas et al. 2017). Against this background, 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 and reliance on deposits.

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,
Pappas et al. 2017, Vera-Gilces et al. forthcoming), we complement this literature by
conducting a cross-country 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 164 African banks over the period 2009-2017. 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, Pappas et al. 2017).

Using data from 21 Islamic and 143 conventional banks, our results highlight that the
determinants of profitability are almost the same between Islamic and conventional
banks in Africa. We show that the most capitalized and efficient banks (Islamic and
conventional) perform best in Africa. Furthermore, the results of our size and deposit
dependence analysis highlight that these performance are influenced by the size and
deposit holding of banks. Specifically, we show that large and deposit-dependent
banks perform better than other banks.

Electronic copy available at: https://ssrn.com/abstract=3603046

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 164 banks operating in 20

African countries1 from 2009 to 2017. Similar to Beck et al. (2013), we use a sample that
includes only countries with both conventional and Islamic banks. Our bank sample
consists of 21 Islamic banks and 143 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. Following the previous literature, we consider a set of variables for
our estimations (among others, Beck et al. 2013, Pappas et al. 2018, Alexakis et al. 2019).

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
net interest income (NII)2. While ROA is measured as net income to total assets, NII is
measured as net interest income to total assets. 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. As we can see in Figure 1, the profitability of Islamic and conventional banks
follow a similar trend over our study period.

1 Algeria, Burkina Faso, Chad, Côte d’Ivoire, Djibouti, Egypt, Gambia, Ghana, Guinea, Libya, Mali,
Mauritania, Morocco, Niger, Nigeria, Senegal, Sierra Leone, Sudan, Tanzania, Tunisia.
2 We use two other proxies for bank profitability as robustness checks: net interest margins to total assets

and profit before tax to total assets. To preserve space, the results are not tabulated but available upon
request.

Electronic copy available at: https://ssrn.com/abstract=3603046

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 unemployment rate, the annual
growth rate of the consumer price index (Inflation) and the real GDP growth rate
(GDP).

The descriptive statistics and correlation matrix of the variables used for our
estimations are displayed in Table A and Table B in the Appendix, respectively.

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:

𝑃𝑟𝑜𝑓𝑖𝑡𝑖,𝑗,𝑡 = 𝑐 + 𝛼1 𝑋𝑖,𝑗,𝑡 + 𝛼2 𝑌𝑗,𝑡 + 𝜃𝑡 + 𝜆𝑖 + Ɛ𝑖,𝑗,𝑡 (1)

where 𝑃𝑟𝑜𝑓𝑖𝑡𝑖,𝑗,𝑡 is the banks’ profitability measures, alternatively ROA and NII, 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 1.50 suggests that our control
variables are not highly correlated. While 𝑋𝑖,𝑗,𝑡 refers to bank-specific controls, 𝑌𝑗,𝑡
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).
𝜃𝑡 , 𝜆𝑖 and Ɛ𝑖,𝑗,𝑡 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

3 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).

Electronic copy available at: https://ssrn.com/abstract=3603046

examine whether the determinants of profitability of African banks differ by size and
reliance on deposits. The results of these analyses are reported in Tables 1, Table 2 and
Table 3, respectively. All our estimates include- and year- fixed effects.

Table 1 presents our main findings. Our results are divided into three columns, the first
of which "All sample" refers to the whole sample (including all banks), "Conventional"
considers only conventional banks and "Islamic" for Islamic banks. Considering only
the first column, we observe that bank-specific factors (notably size, capitalization and
efficiency) are the main determinants of the profitability of African banks. While size is
positively correlated with ROA, it is negatively and significantly correlated with NII.
This result is mainly influenced by conventional banks. Concerning capitalization, it is
positively related with both ROA and NII for all banks. While the relationship is
statistically stronger for conventional banks, it is not significant for Islamic banks.

Table 1. Determinants of ROA and NII

ROA NII

All sample Conventional Islamic All sample Conventional Islamic

Size 0.054 0.086 0.79 -0.505*** -0.427** -0.351

[0.31] [0.22] [1.21] [0.17] [0.20] [0.49]
*** *** ** *
Capitalization 0.073 0.080 0.105 0.028 0.034 0.034**
[0.02] [0.01] [0.10] [0.01] [0.02] [0.01]
Efficiency -0.065*** -0.071*** -0.057** -0.019*** -0.016*** -0.044**
[0.01] [0.01] [0.02] [0.00] [0.01] [0.02]
*** ***
Liquidity -0.003 0.007 -0.097 -0.016 -0.016 -0.015
[0.01] [0.01] [0.07] [0.00] [0.00] [0.02]
***
Funding 1.089 1.275 1.253 0.903 1.03 2.057
[0.74] [0.46] [2.25] [0.70] [0.98] [1.18]
Loans -0.006 -0.005 -0.05 0.009 0.012 0.018
[0.01] [0.01] [0.06] [0.01] [0.01] [0.02]
*
HHI 0.878 5.572 33.079 -1.322 0.43 -7.904
[3.87] [3.20] [28.56] [2.59] [2.60] [15.07]
GDP 0 -0.007 0.03 -0.004 -0.004 0.014
[0.02] [0.03] [0.03] [0.02] [0.02] [0.04]
*
Inflation 0.048 0.021 0.131 -0.003 0.001 -0.014
[0.03] [0.03] [0.06] [0.01] [0.01] [0.02]
Unemployment -0.003 -0.01 0.048 -0.028 -0.027 -0.035
[0.01] [0.02] [0.07] [0.02] [0.02] [0.05]
Observations 1076 880 112 1133 930 111
Number banks 160 131 16 160 132 15
R2(within) 0.355 0.397 0.425 0.203 0.198 0.398
Note: ROA refers to return on assets to total assets. NII is net interest income to total assets. All our estimates
include bank-specifc (natural logarithm of total assets, equity to assets ratio, liquid assets to total assets, and cost-to-
income ratio, liquid assets to total assets, deposits to assets ratio), country-specific controls (Herfindahl-Hirschman
index, real GDP growth rate, the percent change in average consumer price, and unemployment rate), bank- and time-
fixed effets. Robust standard errors clustered by bank are in brackets. ***, ** and * indicate statistical significance at
1%, 5% and 10 % respectively.

Conversely, considering NII the impact is the same for all banks. Concerning
Efficiency, the relationship is negative for all banks, with a stronger effect for
conventional banks considering the ROA variable. Conversely, for the NII the effect is

Electronic copy available at: https://ssrn.com/abstract=3603046

stronger for Islamic banks. As for liquidity, it is negatively associated and statistically
significant only with conventional banks. On the other hand, Funding and HHI are
positively and statistically significant for conventional banks. Conversely, inflation is
statistically significant for Islamic 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.

Overall, our results highlight that the determinants of profitability are almost the same
across banks. In other words, the best performing conventional banks are those that are
more efficient, hold more capital and deposits and operate in a more competitive
environment. Conversely, the best performing Islamic banks are more efficient and
sensitive to inflationary developments4.

Table 2. Determinants of ROA according to bank size and deposits

ROA
Si ze Depos i ts

Islamic banks Conventional Islamic Conventional

Low- High- Low- High-

Small Large Small Large
deposits deposits deposits deposits
Size 0.234 0.247 0.26 0.146 -1.284 0.304 0.704 0.11
[0.68] [0.28] [0.34] [0.17] [2.87] [0.31] [0.43] [0.15]
Capitalization 0.01 0.05 0.065*** 0.075*** 0.001 0.109* 0.064*** 0.073***
[0.02] [0.03] [0.01] [0.02] [0.03] [0.06] [0.01] [0.02]
Efficiency -0.068*** -0.093*** -0.052*** -0.052*** -0.012 -0.056*** -0.042*** -0.061***
[0.02] [0.02] [0.01] [0.01] [0.03] [0.01] [0.01] [0.01]
***
Liquidity -0.02 -0.006 0.01 -0.002 -0.048 -0.007 0.007 0.002
[0.02] [0.01] [0.01] [0.00] [0.01] [0.01] [0.01] [0.00]
Funding -1.475 -2.295 0.759 0.336 -3.582 0.049 1.142** -0.232
[1.37] [2.40] [0.89] [0.43] [6.59] [2.81] [0.49] [0.62]
Loans -0.007 0.006 -0.003 -0.001 -0.204* -0.019 0.003 -0.004
[0.02] [0.02] [0.00] [0.01] [0.09] [0.01] [0.00] [0.01]
HHI -29.284* 83.393** -0.086 0.454 -13.428 73.859* -0.722 -0.676
[13.95] [32.84] [1.79] [0.70] [37.49] [41.39] [2.17] [1.49]
***
GDP -0.043 0.024 -0.003 0.006 0.151 0.022 -0.002 0
[0.04] [0.02] [0.00] [0.01] [0.02] [0.01] [0.00] [0.00]
Inflation -0.006 0.008 0.03 0.004 0.143*** 0.018 0.063 0.008
[0.02] [0.02] [0.03] [0.01] [0.02] [0.02] [0.04] [0.01]
Unemployment -0.003 -0.01 0.048 -0.027 -0.01 0.048 -0.027 -0.035
[0.01] [0.02] [0.07] [0.02] [0.02] [0.07] [0.02] [0.05]
Observations 68 67 687 1064 44 91 640 1111
Number banks 13 12 137 189 12 17 165 223
R2(within) 0.632 0.757 0.351 0.388 0.953 0.691 0.341 0.425
Note: ROA refers to return on assets to total assets. All our estimates include bank-specifc (natural logarithm of total assets, equity
to assets ratio, liquid assets to total assets, and cost-to-income ratio, liquid assets to total assets, deposits to assets ratio),
country-specific controls (Herfindahl-Hirschman index, real GDP growth rate, the percent change in average consumer price, and
unemployment rate), bank- and time-fixed effets. Robust standard errors clustered by bank are in brackets. ***, ** and * indicate
statistical significance at 1%, 5% and 10 % respectively.

In Table 2, we present the results of the determinants of bank profitability as a function

of bank size and deposit dependency. To analyze the determinants by size and deposit
holding, we use the median. For size a bank (Islamic or conventional) is considered
small if its natural logarithm of total assets is below the median (8.62 or 10.91,

4 Using annual data of 2013 on 44 Islamic banks from Asian and African regions, Chowdhury and Rasid
(2015) find similar results.

Electronic copy available at: https://ssrn.com/abstract=3603046

respectively) and conversely it is considered large. For dependence on deposits, we
also proceed in the same way. A bank (Islamic or conventional) is deposit dependent if
it has a deposits to assets ratio above the median (0.60 or 0.65) and conversely it is low
dependent. This allows us to further document these determinants for African banks.

For ROA, we find that increasing capital leads to an improvement in the profitability of
conventional banks, with a stronger effect for large banks. Our results also show that
larger Islamic banks are more efficient, which allows them to perform better. For
deposit dependence, we find that banks (both Islamic and conventional) holding more
deposits and being more efficient perform better. For NII, we also find similar results
reported in Table 3. In sum, we highlight that bank characteristics (such as size and
deposit dependence) influence bank performance. Our results seem to show that large
and highly deposit-dependent banks perform better than other banks, both Islamic and
conventional. This result is also in line with previous analyses such as Beck et al. (2013)
who conclude that banks with more capital perform better.

Table 3. Determinants of NII according to bank size and deposits

NII
Size Deposits

Islamic Conventional Islamic Conventional

Low- High- Low- High-

Small Large Small Large
deposits deposits deposits deposits
Size -1.083 -0.139 -0.749* -0.404 -2.839** -0.288 -0.560* -0.086
[0.67] [0.48] [0.39] [0.27] [1.15] [0.26] [0.30] [0.25]
** * *** ** ** ***
Capitalization 0.038 0.087 0.023 0.151 0.050 0 0.049 0.131
[0.02] [0.05] [0.02] [0.02] [0.02] [0.04] [0.02] [0.02]
Efficiency -0.008 -0.040*** -0.012** -0.013** -0.06 -0.029*** -0.004 -0.024***
[0.01] [0.01] [0.01] [0.01] [0.03] [0.01] [0.01] [0.01]
Liquidity -0.032 -0.006 -0.015*** 0.002 0.009 0.012 -0.006 -0.007*
[0.02] [0.02] [0.01] [0.01] [0.03] [0.01] [0.01] [0.00]
Funding 1.475 5.801* 2.251** 0.729 8.478** 2.351* 1.992** 0.595
[2.27] [2.83] [1.12] [0.53] [3.25] [1.25] [0.93] [0.88]
*** * ***
Loans 0.007 0.039 0.003 0.039 0.049 0.012 0.017 0.028
[0.02] [0.02] [0.01] [0.01] [0.05] [0.01] [0.01] [0.01]
HHI -6.267 -23.24 1.178 -0.48 12.681 6.526 2.171 1.537
[24.61] [32.63] [1.81] [1.11] [18.86] [24.18] [3.48] [2.01]
GDP -0.005 -0.021 -0.001 0.01 -0.026 -0.032 0.001 0.003
[0.06] [0.05] [0.00] [0.02] [0.06] [0.04] [0.00] [0.00]
*** * *
Inflation 0.016 -0.023 0.01 -0.052 -0.009 -0.030 0 -0.018
[0.01] [0.02] [0.01] [0.02] [0.02] [0.02] [0.01] [0.01]
Unemployment -0.023 -0.01 0.048 -0.027 -0.003 -0.01 -0.009 -0.027
[0.02] [0.02] [0.07] [0.02] [0.01] [0.02] [0.02] [0.02]
Observations 66 68 733 1177 38 96 696 1214
Number banks 12 11 137 193 9 17 172 228
R2(within) 0.474 0.684 0.134 0.325 0.883 0.378 0.156 0.265
Note: NII is net interest income to total assets. All our estimates include bank-specifc (natural logarithm of total assets, equity to
assets ratio, liquid assets to total assets, and cost-to-income ratio, liquid assets to total assets, deposits to assets ratio), country-
specific controls (Herfindahl-Hirschman index, real GDP growth rate, the percent change in average consumer price, and
unemployment rate), bank- and time-fixed effets. Robust standard errors clustered by bank are in brackets. ***, ** and * indicate
statistical significance at 1%, 5% and 10 % respectively.

Electronic copy available at: https://ssrn.com/abstract=3603046

4. Conclusion
The Global Financial Crisis 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 143 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 and conventional banks have similar determinants. Moreover, our results
show that large and deposit-dependent banks, whether Islamic or conventional,
generally perform better than other banks.

Electronic copy available at: https://ssrn.com/abstract=3603046

References
Alexakis C., Izzeldin M., Johnes J., Pappas V., 2019. Performance and productivity in Islamic and
conventional banks: Evidence from the global financial crisis. Economic Modelling, Vol. 79, pp.
1-14.

Athanasoglou P., Brissimis S., Delis M., 2008. Bank-specific, industry-specific and macroeconomic
determinants of bank profitability. Journal of International Financial Markets, Institutions &
Money, Vol. 18, pp. 121-136.

Beck T., Demirgüç-Kunt A., Merrouche O., 2013. Islamic vs conventional banking: Business model,
efficiency and stability. Journal of Banking and Finance, Vol. 37, pp. 433–447

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. 106-
130.

Causse-Broquet G., 2012. La finance islamique. Ed. Revue Banque. Marchés Finances.

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 cross-
sectional 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.

IMF, 2015. Islamic finance and the role of the IMF, 15 September 2015.

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.

Pappas V., Ongena S., Izzeldin M., Fuertes A.-M., 2017. A survival analysis of Islamic and
Conventional banks. Journal of Financial Services Research, Vol. 51, pp. 221-256.

Electronic copy available at: https://ssrn.com/abstract=3603046

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.

10

Electronic copy available at: https://ssrn.com/abstract=3603046

 Online Appendix

Table A. Descriptive statistics

Conventional banks Islamic banks
Variables Obs Mean Std. Dev. Min Max Obs Mean Std. Dev. Min Max
Profitability measures
ROA 1,883 1.909 2.002 0 33.990 137 2.201 2.290 0.040 23.520
NII 2,213 4.217 4.034 0 48.250 145 3.279 1.452 0.098 6.504
Bank-specific controls
Size 2,302 10.914 2.441 2.432 15.607 155 8.625 1.844 4.631 13.721
Capitalization 2,274 16.101 15.100 0.800 99.090 155 18.372 15.190 3.710 99.930
Efficiency 2,050 58.314 19.351 3.090 100 142 56.018 19.551 15.300 96.310
Liquidity 2,298 26.685 20.609 0 99.990 155 30.906 19.383 0.470 90.990
Funding 2,239 0.651 0.198 0 1.612 154 0.609 0.233 0 0.920
Loans 2,288 47.078 20.730 0 98.970 154 43.947 21.722 0.260 93.930
Country-specific controls
HHI 4,819 0.017 0.065 0 0.685 240 0.015 0.055 0 0.552
GDP 4,687 4.268 10.882 -66.657 124.709 230 1.843 12.034 -66.657 124.709
Inflation 4,725 7.845 7.890 -2.786 63.293 230 13.720 14.488 -2.786 63.293
Unemployment 2,230 12.038 4.349 5.092 22.562 177 15.083 3.645 5.092 22.562
Note: ROA refers to return on assets to total assets. NII is net interest income to total assets. Size is the natural logarithm of total assets. Capitalization is
equity to assets ratio. Efficiency is cost to income ratio. Liquidity refers to liquid assets to total assets. Funding is deposits to total assets. Loans is net
loans to total assets. HHI is the Herfindahl-Hirschman index. GDP is the real GDP growth rate. Inflation refers to the annual growth rate of the consumer
price index. Unemployment is unemployment rate.

11

Electronic copy available at: https://ssrn.com/abstract=3603046

 Table B. Correlation matrix
L1. L2. L3. L4. L5. L6. L7. L8. L9. L10.
L1. Size 1
L2. Capitalization -0.314 1
L3. Efficiency 0.078 -0.123 1
L4. Liquidity -0.264 0.185 -0.059 1
L5. Funding 0.307 -0.553 0.147 0.053 1
L6. Loans 0.129 -0.159 -0.019 -0.669 -0.058 1
L7. HHI -0.055 -0.020 0.020 0.104 0.038 -0.048 1
L8. GDP 0.101 -0.038 0.031 -0.007 0.048 0.016 -0.022 1
L9. Inflation -0.156 0.125 -0.144 0.146 0.049 -0.338 -0.024 -0.058 1
L10. Unemployment -0.335 0.080 0.063 -0.032 -0.014 -0.045 -0.078 -0.496 0.376 1
Note: ROA refers to return on assets to total assets. NII is net interest income to total assets. Size is the natural logarithm of total assets.
Capitalization is equity to assets ratio. Efficiency is cost to income ratio. Liquidity refers to liquid assets to total assets. Funding is deposits to total
assets. Loans is net loans to total assets. HHI is the Herfindahl-Hirschman index. GDP is the real GDP growth rate. Inflation refers to the annual
growth rate of the consumer price index. Unemployment is unemployment rate.

12

Electronic copy available at: https://ssrn.com/abstract=3603046