The Interactional Relationships Between Credit Risk,Liquidity Risk and Bank Profitability in MENA Region

Abstract

The purpose of this article is to investigate the relationship between credit risk, liquidity risks and bank profitability within the Middle East and North African (MENA) countries. We selected data related to a sample of conventional banks observed during the period 2004–2015 and we performed the Seemingly Unrelated Regression (SUR) method in the empirical section. The overall results suggest that profitability of MENA banks is negatively and significantly sensitive to an increase in credit and/or liquidity risks. This negative effect was confirmed for either the separate or the interaction effects of these two risks. Furthermore, the findings indicate that bank profitability decreases significantly the level of credit and liquidity risks. We also found that the law and order as institutional quality increases the profitability of MENA banks and decreases both credit and liquidity risks.

Keywords

Credit risk liquidity risk profitability MENA banks SUR method

Introduction

Following the international financial crisis of 2008, the liquidity risk has become one of the most important issues for both policymakers and academics. As reaction to avoid the recurrence of such crisis, policymakers have implemented several measures to reinforce capital and bank liquidity. These new reforms are taking in order to keep the overall banking system sound and safe (Hamdi & Hakimi, 2019). In both developing and developed economies, the bad quality of asset has been at the core of costly banking crises (Nikolaidou & Vogiazas, 2017). Credit risk usually proxied by the level of nonperforming loans (NPLs) was considered as the main determinant of bank failures that leads to banking crisis (Reinhart & Rogoff, 2011). Banks that experienced a high level of NPLs may threaten the stability of the banking industry and the whole financial system. More generally, banks often face or take distinct risks, such as credit and/or liquidity risks. Despite that each of these risks may exist independently, they are often dependent and reciprocal in reality. Liquidity is considered as a fundamental part of banking operations (Cornett et al., 2011), and credits are one of the main assets generating profit for the bank (Greuning & Bratanovic, 2004).

According to the modern theory of financial intermediation (Bhattacharya & Thakor 1993), banks exist because they ensure two crucial roles. First, they are considered as liquidity provider and second they transform risk. Classic theories of the financial intermediation support the view that liquidity and credit risks are closely linked. However, several studies have looked into bank credit or liquidity risks separately (Chen et al., 2018; Hakimi & Zaghdoudi, 2017; Hamdi & Hakimi, 2019; Partovi & Matousek, 2019).

There is a little research effort devoted to investigate the relationship between credit risk, liquidity risk and bank profitability in MENA region (Ghenimi, Chaibi, & Omri, 2017). Most empirical studies are focussed on the American context (De Nicolo, 2000; Imbierowicz & Rauch, 2014), the European context (Chortareas, Girardone, & Ventouri, 2011; Kim, 2015; Thorsten, Heiko, Thomas, & Natalja, 2009) and Asian countries (Arif & Anees, 2012; Sohaimi, 2013; Tan, 2016; Zolkifli, Abdul Hamid, & Hawati Janor, 2015). The collapse of the global financial system, following the late-2000 crisis, has raised concern that bank risk-taking has negatively affected the financial stability in various countries. For this reason, the policy agenda for financial sector reforms in developing regions requires, among other issues, an accurate analysis of the impact of bank risk-taking, in order to take into account, the specific features and peculiarities of regions and to identify the proper policy measures to be implemented.

This article aims to investigate the relationship between credit risk, liquidity risks and bank profitability within the Middle East and North African (MENA) countries. To this end, we used data from 38 conventional banks located in 10 MENA countries and observed during the period of 2004–2015. Taking into the reciprocal relationship between credit and liquidity risks and considering these two risks as the main factors that affect bank profitability, our empirical approach is based on simultaneous equations. More precisely, we use the Seemingly Unrelated Regression (SUR) method of Zellner (1962). Empirical results of the SUR method indicate that bank profitability is negatively and significantly sensitive to an increase in credit and liquidity risks. Furthermore, the findings indicate that bank profitability decreases significantly the level of credit and liquidity risks. We also found that the law and order as institutional quality increases the profitability of MENA banks and decreases both credit and liquidity risks. As surprising results, we have found that capital adequacy ratio is negatively and significantly correlated with the bank profitability and significantly increases the liquidity risk.

This article extends the existing literature by focussing on the MENA region, where the banking sector is a crucial driver for economic development and growth. In some MENA countries, the economic stability depends on the stability of their banking system. For example, bank assets account for 60 per cent to over 100 per cent of gross domestic product (GDP) across MENA countries (Ghosh, 2017). On the other hand, banking system is still characterized by the dominance of state-owned banks, high level of NPLs and weak level of liquidity. For all these reasons, we grant more attention to study the factors that affect bank profitability, pillar that affect the stability of the whole banking system through which economic stability may be affected. This study differs from previous studies and contributes to the existing literature in several ways. Frist, the MENA region is considered as an appropriate case of study as the banking sector is considered as the main source of financing the economy. Hence, it is of great importance for policymakers to focus on how ensuring profitability and stability of this banking sector. Second, unlike earlier studies that investigated either the impact of credit or liquidity risks on bank profitability, in the current study, we tested the interaction of these two risks on the bank profitability. Third, we introduce in our econometric model institutional variables to capture the bank risk–profitability relationship. Finally, most of studies focussed on the impact of bank risks on bank profitability have used ordinary least square (OLS), fixed and random effect and/or dynamic panel data analysis. In this article, we performed a simultaneous equation, based on the SUR.

The rest of this article is structured as follows. Section 2 gives literature review on the reciprocal relation between credit and liquidity risks and their effects on bank profitability. Section 3 describes briefly the MENA banking sector. While, section 4 presents and discusses the empirical analysis. In section 5, we conclude.

Review of Literature

We start this section by reviewing the reciprocal relationship between credit and liquidity risks, then we present recent development on the relationships between credit risk and bank profitability and finally, we summarize studies on the effect of liquidity risk on bank profitability.

The Reciprocal Relationship Between Credit and Liquidity Risks

Theoretically, credit risk is related to liquidity risk through borrower defaults and fund withdrawals (Diamond & Dybvig, 1983). Banks’ mix of illiquid (long maturity) assets and liquid (short-term) liabilities may lead to panic among depositors which may be displayed through borrower defaults and fund withdrawals. Banks’ asset and liability structures are closely connected, especially in terms of borrower defaults and deposit outflows (Bryant, 1980). Banks, mostly because of their maturity transformation role, are exposed to the liquidity risk. A loan default can increase liquidity risk by leading to a decrease in cash flow and depreciations in loan assets (Dermine, 1986; Diamond and Rajan, 2001). Thus, according to the classical financial intermediation theory, a relationship exists between credit and liquidity risks. Moreover, when banks have a high level of liquidity, their managers tend to take more risk by lowering the lending standards for the intention of increasing the volume of loans (Acharya & Naqvi, 2012.

Generally speaking, literature on the relationship between credit and liquidity risks has revealed two main views. The first view is articulated around the financial intermediation theory. Several studies have concluded that credit risks are positively related in banks. They stated that the financing of risky or distressed project leads to an increase in NPLs that decreases bank liquidity and struggle banks to meet depositors’ demands for funds (Acharya & Viswanathan, 2011; Gorton & Metrick, 2011; He & Xiong, 2012). For the Ukrainian context, Cai and Zhang (2017) over the period from Q1 2009 to Q4 2015 have reported that with high level of non-performing loans, banks are unable to respond partial or integral withdrawal demands. This situation lowers cash flow and triggers depreciations in loan assets and leads to an increase liquidity risk. Cai and Zhang (2017) have found that for foreign and large banks, the positive association between credit and liquidity risks is more pronounced. Using a sample of banks in 43 countries over the period of 2002–2010, Chen and Lin (2016) show that credit, interest rate and liquidity risks are related to one another, and that the interactions among them can be reduced by corporate governance and regulations.

The second view supports the opposite view. Less abundant empirical studies document a negative relation or no relation between liquidity and credit risks (Cai & Thakor, 2008; Wagner, 2007). For example, Imbierowicz and Rauch (2014) investigate US commercial banks during the period of 1998–2010 and analyze the relationship between these the credit and liquidity risks sources on the bank institutional level, and how this relationship influences banks’ probabilities of default. They show that the two risks do not have an economically meaningful reciprocal contemporaneous or time-lagged relationship. However, they do influence banks’ probability of default.

Credit Risk and Bank Profitability

Literature on the effect of credit risk and bank profitability provides mixed and inconclusive results. Some empirical studies have supported the negative association (Athanasoglou et al. (2008); Berríos, 2013; Cucinelli, 2015; Laryea, tow-Gyamfi, & Azumah Alu, 2016). However, few studies found positive evidence (Flamini, Valentina, McDonald, & Liliana, 2009; Hakimi, Hamdi, & Djelassi, 2011).

In banking literature, several measures have been used as proxy of credit risk. However, the standard and the most used indicator is the NPLs ratio. Broadly, a loan is classified as non-performing when payments of principal and interest are 90 days or more past due. An increase in bad loans in turn increases the non-payments of credits, which lower the profitability of MENA banks and could even lead to bank failure.

Most of the studies focussed on the determinants of bank profitability have used bank specifics, industry specifics, institutional quality, macroeconomic and financial environment to explain changes in bank profitability. Some others have carried out on the effect of non-traditional activities based on the non-interest incomes. In this line of idea, Isshaq et al. (2019) used a bank-level financial statement related to the Ghanaian economy over the period of 1999–2015 to explore the link between NII, risk and performance. The authors found that NII increases profitability for large banks. However, no significant effect of non-interest income (NII) on bank risk was found. Sufian (2009) has analyzed the determinants of bank profitability in the Malaysian context over the period of 2000–2004. The main findings of this study indicate that credit risk and loan concentration decrease significantly the level of profitability. However, well-capitalized banks with more banking activities diversification registered higher profitability level.

Some recent studies have reported that credit risk could be influenced by corporate governance. For example, to investigate the link between internal governance mechanism and credit risk, Ben Moussa (2019) used a sample of listed banks observed during the 2000–2014 period. The empirical results show that the higher board size and the duality are associated with lower quality of credit that increases credit risk. However, the author found that the presence of independent and foreign investors enhances credit quality and hence decreases the level of NPLs.

Using a sample of 488 Italian banks over the period of 2007–2013, Cucinelli (2015) found that credit risk exerts a negative impact on bank lending behaviour. In this study, NPLs and loan loss provision ratio are used as measures of credit risk. Similarly, Athanasoglou et al. (2008) have used a sample of a panel of Greek banks over the period of 1985–2001 and reported that credit risk decreases significantly the bank profitability. The authors explained this result by the risk-averse strategy adopted by Greek banks in order to maximize their profits. Berríos (2013) has used a sample of 40 banks observed during the period of 2005–2009 to analyze the linkage between credit risk and profitability and liquidity. Empirical findings show a negative association between less prudent lending and net interest margin.

Abu Hanifa, Pervin, Chowdhury, and Banna, (2015) investigated the effect of credit risk on profitability in Bangladesh. To this end, they used a sample of 18 banks over the period of 2003–2013. Empirical results indicate that credit risk decreases significantly the bank profitability. Focussing on an emerging market, Laryea et al. (2016) have used a sample of 22 Ghanaian banks over the period of 2005–2010 to investigate the effect of NPLs on bank profitability. The empirical results confirm the negative relationship.

However, there are less abundant studies that support the positive relationship between bank profitability and credit risk. This positive association is due to the measure of credit risk. Some studies that used loan to assets as proxy of credit risk, generally have found positive relation. An increase in the loan to assets’ ratio leads to an increase in bank profitability. Banks with high loan to assets received more interest revenue that increases the interest margin and consequently the bank profitability. For example, for the Sub-Saharan African context, Flamini et al. (2009) have used a sample of 389 banks in 41 Sub-Saharan African countries over the period of 1998–2006. The authors found a positive relation between credit risk and bank profitability. In this study, credit risk was measured by asset quality based on standard asset pricing. The same result was found by Hakimi et al. (2011) for the Tunisian context basis on a sample of nine banks over the period of 1980–2009. Based on this development above, we can put the following hypothesis:

H1: Credit risk decreases the level of bank profitability.

Liquidity Risk and Bank Profitability

Literature on the effect of liquidity risk and bank profitability is ambiguous. In fact, some empirical studies reported that liquidity affects positively bank profitability (Bourke, 1989; Kosmidou, Tanna, & Pasiouras, 2005; Olagunju, David, & Samuel, 2012), while many others concluded that liquidity exerts a negative effect on bank profitability under the misallocation of resources.

The negative effect of liquidity risk was confirmed by several studies. For example, the study of Cuong Ly (2015) focussed on the effect of liquidity risk on the profitability of European banks during the period of 2001–2011 indicates that liquidity risk decreases significantly the bank profitability. However, for the same European context, Cucinelli (2015) has found that there is no significant association between liquidity and probability of default in the long term. Based on a sample of 97 banks from G7 and the Switzerland, Mamatzakis and Bermpei (2014) have reported that liquidity exerts a negative impact on bank profitability. More recently, Adelopo, Lloydking, and Tauringana (2018) have used a sample of 123 banks over the period of 1999–2013. Empirical findings indicate that liquidity risk decreases significantly the bank profitability during the three sub-periods of study.

The same negative association between liquidity risk and bank profitability was confirmed for the Iranian example, which was investigated by Tabari, Ahmadi, and Emami (2013) for a sample of Iranian commercial banks over the period of 2003–2010. The authors reported that both credit and liquidity risks exert a negative and significative effect on the profitability of Iranian banks. Within the same Asian context, Arif and Anees (2012) have used a sample of 22 Pakistani banks during the period of 2004–2009 to check the relationship between liquidity risk and bank profitability. They found that there is a negative and significant association between liquidity risk and bank profitability. Furthermore, results indicate that the liquidity gap and the level of NPLs are the main factors that increased liquidity risk.

The effect of liquidity risk on bank profitability was recently explored by Hamdi and Hakimi (2019). The authors found that this relation is nonlinear and the effect of liquidity risk on bank profitability depends on a certain optimal threshold. The used a large sample of 127 countries observed during the period of 2005–2015. The whole sample was divided into 46 high-income countries and 81 low- and middle-income countries. Results of the panel smooth transition regression (PSTR) model indicate that the optimal level of liquidity and its effect on bank profitability differ from one group of countries to another. The Tunisian context was explored by Hakimi and Zaghdoudi (2017). The authors have used a sample of 10 Tunisian banks for the period of 1990–2013. The results of random effect regression show that liquidity risk decreases significantly Tunisian bank profitability. For the South African context, Marozva (2015) has used a sample of banks over the period of 1998–2014. Bank profitability was measured by the net interest margin. Empirical findings of the autoregressive distributed lag (ARDL)-bound approach show that liquidity risk negatively affects bank profitability. With reference to the findings of these studies, we formulate the following hypothesis:

H2: Liquidity risk exerts a negative effect on bank profitability.

Since credit and liquidity risks are qualified as reciprocal risks, and based on the two hypotheses formulated above that supposed the negative effect of both credit and liquidity risks on bank profitability, we can propose the following third hypothesis:

H3: The interaction between credit and liquidity risks decreases bank profitability.

Banking Sector in the MENA Region

The MENA region contains both ends of the spectrum; it ranges from the wealthiest and biggest oil-producing countries to the politically unstable and oil-importing countries. During the past two decades, some countries in the MENA region have recorded continuous strong economic growth, while some others have been stuck in political conflict and civil war that inhibit their growth.

In general, banking system in the MENA region presents not only some similarities but also some differences. For example, in the North African countries, banking sector is dominated by state-owned banks, and it plays a crucial role in the economy as it is recognized as the principal source of funding. According to Ghosh (2017), bank assets account for 60 per cent to over 100 per cent of GDP across MENA countries. Compared the North African countries, banking sector in the Gulf Cooperation Council (GCC) region is more developed, more stable and less exposed to bank risks. Furthermore, banking sector in the Middle East in general is more profitable and more efficient while North Africa’s banking sector recorded low profitability, insufficient liquidity and high level of NPLs.

Table 1 presents the average annual evolution of some indicators related to banking system in MENA region. For example, we analyze the annual evolution of profitability measured by the ROA and the ROE. Also, we analyze the evolution of bank risks in this region, more precisely the credit and the liquidity risks. Data used in this table are collected from the Global Financial Development Database.

Table 1.

Annual Evolution of Profitability and Bank Risks in the MENA Region

Years	Bank Profitability (%)		Bank Risks (%)
Years	ROA	ROE	LIQR	CRISK
2004	1.58	14.29	87.30	16.44
2005	1.99	15.69	90.60	13.76
2006	1.78	14.54	93.36	9.23
2007	1.71	15.22	92.52	7.94
2008	1.42	12.92	97.48	5.46
2009	1.26	12.03	98.33	10.16
2010	1.46	12.41	91.72	8.25
2011	1.23	10.61	89.98	6.12
2012	1.52	12.73	90.19	5.92
2013	1.58	12.90	88.07	6.00
2014	1.51	13.14	86.00	5.58
2015	1.43	12.55	87.09	5.46

Source: The authors.

From Table 1, it was shown that the profitability of MENA banks has followed a downward trend during the period of 2004–2015. The ratio of ROA was 1.58 per cent in 2004 and became 1.43 per cent in 2015. The same trend was registered by the ROE which crossed from 14.29 per cent in 2004 to reach only 12.55 per cent.

Regarding the evolution of the credit risk, Table 1 indicates that the level of NPLs in the MENA region has significantly decreased moving from 16.44 in percentage of total loans to be only 5.46 per cent in 2015. This downturn was explained by the several actions and steps that have been taken to respect international Basel standard in terms of increasing capital adequacy ratio and reducing NPLs.

As measured by the ratio of loans to deposits, an increase in this ratio means that banks are more exposed to liquidity risk. However, any decrease in this ratio indicates on a sufficient level of liquidity. Contrary to the downward trend registered for the credit risk, the MENA banks have recorded an increase in the liquidity risk. The ratio of loans to deposits crossed from 87.30 per cent in 2004 to reach 97.48 per cent in 2008 and 98.33 per cent in 2009. This increase was explained by the decrease in the level of deposit during the period of the international financial crisis of 2008. During the period of 2010–2015, the level of liquidity risk returned to record a downward trend with a ratio of 87.09 per cent in 2015.

Theoretical Framework

Methodology—Data Source, Sample Frame and Empirical Model

To investigate the effect of credit risk, liquidity risk and their interaction with bank profitability, we used an initial sample made by 112 banks. However, due to the availability and the continuity of bank’s information, several banks have been excluded. For example, we exclude Islamic banks, investment banks and non-commercial banks to ensure the homogeneity of the sample. Hence, only 65 banks have been retained. Moreover, we eliminate 27 commercial banks for which data on NPLs were missing. The final sample was then reduced to only 38 conventional banks located in 10 MENA countries¹ over the period of 2004–2015. Bank-level data are obtained from the Bankscope database and the annual reports of each bank. However, macroeconomic conditions are collected from World Bank Indicators (WDI) database.

Since credit and liquidity risks are considered as reciprocal risks and also recognized as the main determinants of bank profitability, our empirical approach is based on simultaneous equations. More precisely, we use the SUR method of Zellner (1962). The SUR model is system of multiple equations (M) with single dependent variable for each equation and (K) independent or exogenous variables. In econometrics literature, observed variables are considered either endogenous (dependent) or exogenous (independent) variables.

For these M equations, there is no relationship between them except that their disturbances are correlated. Based on the previous studies (Felmlee & Hargens, 1988; Kim & Cho, 2019; Kmenta, 1971; Zellner, 1962, 1963), the general specification of SUR simultaneous equation systems can be represented as follows:

Y 1 = β_{11} + β_{12} X_{12} + β_{13} X_{13} + \dots \dots \dots \dots \dots + β_{1 K} X_{1 K 1} + ε_{1}

(1)

Y 2 = β_{21} + β_{22} X_{12} + β_{23} X_{23} + \dots \dots \dots \dots \dots + β_{2 K} X_{2 K 2} + ε_{2}

(2)

Y M = β_{M1} + β_{M2} X_{12} + β_{M3} X_{M3} + \dots \dots \dots + β_{MK} X_{MKM} + ε_{M}

(3)

Y_{m} = X_{m} β_{m} + ε_{m} (m = 1, 2 \dots \dots \dots \dots m)

where Ym is an (N × 1) vector of dependent variable, Xm is an (N × K) of independent variables (with Xmi = 1 for all m), βm is a (Km 1) vector of coefficient regressions and εm is an (N × 1) vector of disturbances.

The use of SUR method was firstly motivated by the gain efficiency in estimation, since it results information combination from different equations. Second, using this method, it will be able to test restrictions that involve parameters in different equations. Compared with OLS estimators, the two-stage general least square (GLS) and ML estimators of SUR model are considered as most efficient. These two methods provide smaller standard errors especially for large sample. However, with reference to Hanushek and Jackson (1977) dealing with small sample and having covariances of disturbances for the multiples equations equal to zero, the OLS estimators are the best linear unbiased estimates. Taking into consideration, the advantage of this estimator and since the sample used in this study is made by only 38 banks; we performed the OLS estimators of the SUR method.

The empirical analysis of this study is based on three steps. The first step consists to test separately the effect of credit and liquidity risks on bank profitability. In Equation (1), bank profitability (PROF) is the dependent variable. In Equation (2), credit risk (CRISK) is the dependent variable. Liquidity risk (LIQR) is the endogenous variable in Equation (3). The SUR model can be written as follows:

\begin{array}{l} P R O F = β_{10} + β_{11} C R I S K_{11} + β_{12} L I Q R_{12} + β_{13} S I Z E_{13} + \\ β_{14} C A P_{14} + β_{15} C R I S I S_{15} + β_{16} G D P_{16} + β_{17} I N F_{17} + ε_{1} \end{array}

(1)

\begin{array}{l} C R I S K = β_{20} + β_{21} P R O F_{21} + β_{22} L I Q R_{22} + β_{23} S I Z E_{23} + \\ β_{24} C A P_{24} + β_{25} C R I S I S_{25} + β_{26} G D P_{26} + β_{27} I N F_{27} + ε_{2} \end{array}

(2)

\begin{array}{l} L I Q R = β_{30} + β_{31} P R O F_{31} + β_{32} C R I S K_{32} + \\ β_{33} S I Z E_{33} + β_{34} C A P_{34} + β_{35} C R I S I S_{35} + β_{36} G D P_{36} + β_{37} I N F_{37} + ε_{3} \end{array}

(3)

In the second step, the effect of the interaction between credit and liquidity risks on the bank profitability measured by the ROA and the ROE. In Equation (4), PROF is the dependent variable. In Equation (5), the interaction between the two risks (CRISK × LIQR) is considered as the endogenous variable. Hence, the SUR model can be written as follows:

\begin{array}{l} P R O F_{i, t} = β_{10} + β_{11} C R I S K x L I Q R_{11} + β_{12} S I Z E_{12} + \\ β_{13} C A P_{13} + β_{14} C R I S I S_{14} + β_{15} G D P_{15} + β_{16} I N F_{16} + ε_{1} \end{array}

(4)

\begin{array}{l} C R I S K x L I Q R_{i, t} = β_{20} + β_{21} P R O F_{21} + β_{22} S I Z E_{12} + \\ β_{23} C A P_{23} + β_{24} C R I S I S_{24} + β_{25} G D P_{25} + β_{26} I N F_{26} + ε_{1} \end{array}

(5)

Besides bank specifics and macroeconomic conditions, we believe that institutional quality could play a crucial role in the bank risk–profitability relationship. Hence, we investigate in a third step whether institutional quality enhances regulatory mechanisms and improves bank profitability or not. Therefore, we estimate Equations (1)–(3) by introducing two institutional variables: the government stability (GOVS) and the law and order of law (LAW). To the best of our knowledge, no article has studied yet the interaction between credit and liquidity risks on bank profitability using institutional quality data. The empirical model can be written as follows:

D e p e n d v a r = β_{10} + \sum_{i = 1}^{n} β_{i} X_{i,t} + \sum_{i = 1}^{n} β_{i} Y_{i,t} \sum_{i = 1}^{n} β_{i} Z_{i,t} + ε_{i}

(6)

where dependent variables are profitability, credit and liquidity risks, X is the matrix of bank-specific variables (PROF, CRISK, LIQR, SIZE and CAP), Y is the matrix of financial environment and macroeconomic variables (international financial crisis, inflation and GDP growth) and Z is the matrix of institutional variables (government stability and law and order).

Table 2 gives the definition and measurement of all variables used in this study. We classified variables into three categories: dependent variables, bank-specific variables and variables that reflected macroeconomics conditions and institutional context.

Table 2.

Definition and Measurement of Variables

Variables	Definition	Measurement
Dependent variables
PROF	Bank profitability	Return on assets (ROA) and return on equity (ROE)
CRISK	Credit risk	Bank NPLs to gross loans (%)
LQIR	Liquidity risk	Loans to deposit ratio (%)
CRISK × LIQR	Interaction term	The interaction between credit and liquidity risks
Bank specifics
SIZE	Bank size	Napierian logarithm of total assets
CAP	Capital adequacy ratio	Bank capital to total assets (%)
Macroeconomic conditions
GDPG	Growth of gross domestic product	Economic growth measured by annual growth rate of GDP (%)
INF	Inflation rate	Inflation rate measured by consumer price index (%)
Institutional quality
GOVS		Government stability: score of the International Country Risk Guide (ICRG) that ranges between 0 (weak government stability) and 12 (strong government stability).
LAW		Law and order: score of the ICRG that ranges between 0 (weak strong law and order) and 5 (strong law and order).

Source: The authors.

Analysis

Descriptive Statistics and Correlation Matrix

Table 3 presents information about descriptive statistics of all the variables used in the empirical model. These statistics give more details about characteristics of banking sector and macroeconomic specifics in the MENA region. Descriptive statistics are displayed in Table 3.

Table 3.

Descriptive Statistics

Variable	Obs.	Mean	Std. Dev.	Min.	Max.
ROA	452	0.037	0.199	-0.103	2.090
ROE	452	0.129	0.120	-1.119	0.797
CRISK	451	0.091	0.086	0.002	0.479
LIQR	441	0.716	0.269	0.047	1.456
CRISK × LIQR	456	0.064	0.079	0.000	0.500
SIZE	456	16.123	1.822	11.581	20.105
CAP	456	0.112	0.040	-0.016	0.256
LAW	456	4.535	0.582	2.875	5.000
GOVS	456	9.001	1.868	5.208	11.500
CRISIS	456	0.667	0.472	0	1
GDPG	456	0.049	0.039	-0.019	0.208
INF	428	0.043	0.037	-0.049	0.183

Source: The authors.

From Table 3, we notice that the mean value of bank profitability for the selected banks in MENA region is 3.7 per cent for ROA and 12.8 per cent for ROE. These two indicators register a maximum value of 209 per cent for ROA and 79.7 per cent for ROE, while their minimum values are –10.3 per cent for ROA and –111.9 per cent for ROE, respectively.

With regard to bank size, banks in the MENA region record on average a value of 16.123. The minimum and the maximum sizes are 11.581 and 20.105, respectively. From Table 3, we also notice that on average banks in MENA region are moderately capitalized with a value of 11.2 per cent. However, it does not prevent the existence of some others less capitalized with a minimum value of –1.6 per cent and well capitalized with a maximum value of 25.6 per cent. In this study, NPLs are used as proxy of credit risk. MENA banks record on average a value of 9.1 per cent. However, the maximum value of NPLs to total loans represents 47.9 per cent. As reciprocal risk, the mean value of liquidity risk is 71.6 per cent with a maximum of 145.6 per cent and a minimum of 4.7 per cent.

For institutional quality, statistics indicate that the average value of law and order is 4.535, while its maximum value is 5. For the mean value of government stability, Table 3 reveals 9.001 as the mean value and 11.5 as the maximum value.

Since banks operate in a macroeconomic environment, we introduce two variables in our empirical model: the real GDP growth and the inflation rate. The highest level of growth recoded by MENA region during the period of 2004–2015 was 20.8 per cent, while the weakest level was –1.9 per cent. Descriptive statistics also indicate that during the same period, the mean value of inflation was 4.3 per cent with 18.3 per cent as maximum value and –4.9 per cent as minimum value.

Table 4 presents the levels of correlation between independent variables. To this end, we have used the Pearson correlation to check the nature (positive or negative) and the level (high or weak) of correlation.

Table 4.

Correlation Matrix

	CRISK	LIQR	CRISK × LIQR	SIZE	CAP	LAW	GOVS	CRISIS	GDPG	INF
CRISK	1.0000
LIQR	0.0985	1.0000
CRISK × LIQR	0.8569	0.4488	1.0000
SIZE	-0.4203	0.0286	-0.3097	1.0000
CAP	-0.1835	0.1227	-0.1493	-0.1314	1.0000
LAW	0.0116	0.5134	0.2227	0.0562	-0.0036	1.0000
Govs	-0.2186	0.2869	-0.0629	0.2186	0.2711	0.3581	1.0000
CRISIS	-0.1975	0.0353	-0.0950	0.2093	0.0511	0.1204	0.5631	1.0000
GDPG	-0.1839	0.0181	-0.1612	0.0718	0.2699	0.0377	-0.0983	-0.2863	1.0000
INF	-0.0153	-0.1607	-0.0884	-0.0252	0.0719	-0.3866	-0.1351	-0.0186	0.1680	1.0000

Source: The authors.

From Table 4, we conclude that the level of correlation between all independent variables is very weak. The highest level of correlation is between CRISK and the interaction between credit and liquidity risks (CRISK × LIQR). Since the second variable is an interaction variable, we expected a high level of correlation between the two variables. To avoid multicollinearity problem, these two variables are not introduced jointly in the same econometric model. As independent variables, the interaction variable (CRISK × LIQR) was tested only in model (4). Thus, we can conclude that there is no significant problem of multicollinearity.

Discussion

The Separate Effect of Credit and Liquidity Risks on Bank Profitability

The use of the SUR method requires the disturbances correlation as necessary condition. It is for this reason that we firstly check the residuals correlation of three equations that should be different to zero. Table 5 gives the results of correlation matrix of residuals and Breusch–Pagan test of independence.

Table 5.

Correlation Matrix of Residuals

	Prof is Measured by ROA			Prof is Measured by ROE
	Prof	Crisk	Liqr	Prof	Crisk	Liqr
PROF	1.0000			1.0000
CRISK	0.1042	1.0000		0.4246	1.0000
LIQR	0.0602	-0.1546	1.0000	0.0933	-0.1064	1.0000
Breusch–Pagan test of independence
χ²(2)	15.697			81.922
Prob.	(0.0013)			(0.0000)

Source: The authors.

From Table 5, we notice that, for the correlation of the residuals in the PROF, CRISK and LIQR equations are different from zero that we can reject the hypothesis that this correlation is zero. Therefore, the residuals of the three equations are correlated. Also, the result of the test of Breusch–Pagan indicates that there is a residual correlation. The probabilities of this test are equal to 0.0013 when profitability is measured by ROA and 0.000 for ROE which are lower than 5 per cent and confirm the correlation between residuals of the three equations.

Once, we have checked and confirmed the disturbances correlation of the three equations, we can apply the SUR method. Table 6 summarizes the results of the separate effect of credit and liquidity risks on bank profitability in the MENA region.

Table 6.

Results of Separate Effect on Bank Profitability: SUR Model

			Profitability is ROA			Profitability is ROE
Equations			Coef.		Z	Coef.			Z
Profitability	CRISK		-0.557		-3.970***	-1.041			-17.050***
(Equation [1])	Liqr		-0.083		-2.080***	-0.050			-2.710***
	SIZE		-0.023		-3.590***	-0.001			-0.440
	CAP		-0.424		-1.490	-0.277			-2.080**
	CRISIS		-0.020		-0.790	-0.050			-4.160***
	GDPG		0.278		0.930	-0.177			-1.250
	INF		0.195		0.670	0.442			3.250***
	_cons		0.556		4.700***	0.330			6.040
Credit risk	PROF		-0.068		-3.970***	-0.474			-17.050***
(Equation [2])	LIQR		0.083		6.080***	0.041			3.290***
	SIZE		-0.019		-9.210***	-0.010			-5.130***
	CAP		-0.443		-4.570***	-0.361			-4.080***
	CRISIS		0.027		3.010***	0.037			4.550***
	GDPG		-0.309		-2.980***	-0.257			-1.530***
	INF		0.161		1.600	0.310			3.360***
	_cons		0.412		11.380***	0.341			10.400***
Liquidity risk	PROF		-0.127		-2.08***	-0.348			-2.710***
(Equation [3])	CRISK		1.044		6.08***	0.625			3.290***
	SIZE		0.021		2.68***	0.021			2.720***
	CAP		0.960		2.74***	0.865			2.480**
	CRISIS		0.011		0.36	-0.007			-0.210
	GDPG		0.463		1.24	0.323			0.870
	INF		-1.206		-3.41***	-1.062			-2.970***
	_cons		0.213		1.44	0.310			2.120**
Equations	Obs.	RMSE	R ²	χ²	p-Value	Obs.	RMSE	R²	χ²	p-Value
Profitability	409	0.2059	0.0283	33.1400	0.0000	409	0.101	0.207	376.390	0.000
Credit risk	409	0.0725	0.2508	194.5400	0.0000	409	0.068	0.334	489.990	0.000
Liquidity risk	409	0.2569	0.0404	59.3400	0.0000	409	0.254	0.060	46.510	0.000

Source: The authors.

Note: *** and ** indicate level of significance at 1% and 5%, respectively.

With regards to Equation (1), we have found that credit risk; liquidity risk and bank size decreases significantly the bank profitability for both ROA and ROE. As surprising result, findings also indicate that capital adequacy ratio is negatively and significantly correlated with the bank profitability only for ROE. For the macroeconomic effect, results show that only the inflation exerts a positive and significant effect on ROE; however, the GDP growth do not exert any significant effect.

Measured by the NPLs, credit risk is shown to be negatively and significantly correlated with bank profitability for both ROA and ROE. As we expect this association, an increase in bad loans increases the non-payments of credits, which lower the profitability of MENA banks and could even lead to bank failure. Also, when borrowers are unable to fulfil their commitments, banks become more rigid, stickier and more restrictive towards credit distribution, which lower interest revenues and consequently banking profitability. This result is in line with Hamdi, Rachdi, Hakimi, and Guesmi (2018).

As reciprocal risk and similarly to the effect of credit risk, liquidity risk decreases significantly the bank profitability measured by ROA and ROE. In banking literature, liquidity was recognized as the necessary pillar of banking activity. It was considered as the main input to support bank profitability. Less-liquid banks tend to have lower profitability. Under the traditional bank activities, banks as financial intermediaries operate based on liquidity. Furthermore, insufficient liquidity is one of the factors that negatively affect income revenues derived from loans’ activity which lower bank profitability and reduce bank reputation, customer trust. This result corroborates the findings of Kosmidou (2008), Marozva (2015) and Hakimi and Zaghdoudi (2017).

From Table 6, we notice that an increase in bank size decreases significantly the level of bank profitability. This result is confirmed for both ROA and ROE. Despite that large banks have the advantage of economies of scale that leads to lower costs; we found that bank size is negatively and significantly correlated with bank profitability. Big banks can be engaged in more diversified activities and be faced to high level of conflict of interest, governance problems and asymmetric information with higher cost of information. In such a case, it results bad loans’ decisions that lead to an increase in NPLs, which negatively impacts bank profitability. This finding is convergent to the study of Barros, Ferreira, and Willians (2007) and divergent to Pasiouras & Kosmidou (2007); Berger Hanweck, & Humphrey (1987).

The most surprising for these results is the negative and significant association between capital adequacy ratio and the return on equity. Theoretically, it was admitted that a better-capitalized bank should be more profitable. However, in this study, we have found that CAP exerts a negative and significant effect on bank profitability for both ROA and ROE. We can explain this surprising result as following: in some cases, an increase in capital leads to an increase in bank risk-taking. High-capitalized banks can extend credit with insufficient guarantees, which increases the level of NPLs and lower bank profitability. This finding is similar to Aggarwal and Jacques (1998) and contradicts the works of Abreu & Mendes (2002); Goddard et al. (2010); Garcia-Herrero et al. (2009).

With regard to the effect of the international financial crisis of 2008 (CRISIS), the results show a negative and significant association between this variable and bank profitability measured by ROE. These results show that MENA banking sector was not sheltered from the subprime crisis, especially when we remember the crisis that hit Dubai in 2009 where most local and foreign banks have recorded colossal losses. Our results are in line with most of the researches that studied the impact of the 2008 crisis on bank profitability (Hamdi et al., 2018).

The results of Equation (2) indicate that bank profitability, bank size, capital adequacy ratio and GDP growth are negatively and significantly correlated with credit risk. However, the results show that only liquidity risk, financial crisis and inflation increase significantly the level of NPLs.

Table 6 reveals that an increase in bank profitability decreases significantly the level of credit risk measured by NPLs. Recording satisfactory level of profitability, banks can be engaged in an improvement process of credit risk management. Banks are motivated by enhancing their experiences and their skills in the process of credit risk management in order to cover this risk and to reduce the probability that will be exposed to this risk. With high level of profitability, banks are able to spend and to finance the improvement of credit risk management process.

The level of growth proxied by GDP growth is found negatively and significantly correlated with credit risk. An increase in the level of growth in the MENA region decreases significantly the level of NPLs in this region. Macroeconomic conditions allow for controlling for business cycle fluctuations. Under a stable and favour economic condition, the probability of solvency increases and borrowers are able to fulfil their commitments. Hence, the banking profitability increases. This result is in line with the works of Athanasoglou et al. (2008) and Calza, Gartner, and Sousa (2003).

Empirical findings also indicate that only liquidity risk, crisis and inflation increase significantly the level of NPLs. Considered as reciprocal risk, an increase in liquidity risk leads to an increase in credit risk.

In the period of crisis, the level of NPLs has witnessed a sudden increase. Several studies show that NPLs are one of the crucial determinants of banking and financial crises and hence, during the turmoil, the level of NPLs gets worse, which in turn adds more vulnerability and risk.

The results of Equation (3) show that an increase in the level of banks profitability decreases significantly the level of liquidity risk. However, it was also found that liquidity risk increases when credit risk, bank size and capital ratio increase.

The more profitable banks are the less exposed to liquidity risk. Recording a certain level of profitability, banks become more capitalized and try to reinforce their own equity in order to improve customer reputation and to lower the exposition to liquidity risk. We also find that the inflation rate acts negatively and significantly on the liquidity risk. Economically, inflation leads to a redistribution of income in favour of borrowers than of lenders. It is for this reason that banks become more rigid and restrict their lending activities in inflationary context.

Giving the reciprocal relationship between credit and liquidity risks, the results indicate that an increase in credit risk leads to the drying up of bank liquidity and the increase in liquidity risk. The high level of NPLs causes losses of principal and interest, which decreases banking liquidity, and more exposes banks in MENA region to a problem of insufficient liquidity.

The results of SUR model show that the capital adequacy ratio (CAP) is positively and significantly correlated with liquidity risk. This result seems surprising since the most capitalized banks are normally less exposed to liquidity risk. However, in the context of high competition, weak prudential regulation and inefficient governance practices, banks can adopt risk-taking behaviour with granting credits sometimes without sufficient guarantees, which will be finished in some case by bad loans with high level of NPLs. The high level of credit risk leads to an increase in the liquidity risk.

The Interaction Effect of Credit Risk, Liquidity Risk on Bank Profitability

Like the first step, we start our empirical analysis by checking the disturbances correlation. The results displayed in Table 7 indicate that the correlation of the two equations is different from zero, which confirms the correlation among residuals.

Table 7.

Correlation Matrix of Residuals

	Prof is Measured by ROA		Prof is Measured by ROE
	PROF	CRISK × LIQR	PROF	CRISK × LIQR
PROF	1.0000		1.0000
CRISK × LIQR	0.1101	1.0000	0.4421	1.0000
Breusch–Pagan test of independence
χ² (1)	5.139		82.871
Prob.	(0.0234)		(0.0000)

Source: The authors.

From Table 7, we also conclude that the probability of χ² of the Breusch–Pagan test of independence is lower than 5 per cent that we can reject the hypothesis that residuals are independent. This test confirms once again that the residuals of the two equations are correlated between them.

In the second step, we aim to test the interaction effect of credit and liquidity risks (CRISK ´ LIQR) on the MENA bank profitability. The empirical results are displayed in Table 8.

Table 8.

Results of the Interaction Effect (CRISK × LIQR) on Bank Profitability

			Profitability is ROA			Profitability is ROE
			Coef.		Z	Coef.				Z
Profitability (Equation [4])	CRISK × LIQR		-0.631		-4.530***	-1.226				-18.560***
	SIZE		-0.022		-3.700***	0.002				0.530
	CAP		-0.498		-1.810*	-0.437				-3.100***
	CRISIS		-0.011		-0.440	-0.053				-4.270***
	GDPG		0.330		1.130	-0.128				-0.850
	INF		0.159		0.590	0.306				2.190**
	_cons		0.475		4.360***	0.261				4.710***
CRISK × LIQR (Equation [5])	PROF		-0.075		-4.530***	-0.446				-18.560***
	SIZE		-0.014		-7.400***	-0.005				-2.970***
	CAP		-0.411		-4.420***	-0.372				-4.450***
	CRISIS		0.014		1.720*	0.030				4.000***
	GDPG		-0.170		-1.540	-0.149				-1.560
	INF		-0.113		-1.210	0.077				0.910
	_cons		0.371		10.960***	0.276				9.020***
Equations	Obs.	RMSE	R ²	χ²	p-Value	Obs.	RMSE	R ²	χ²	p-Value
Profitability	424	0.203	0.023	30.430	0.000	424	0.110	0.199	405.170	0.000
Crisk × Liqr	424	0.070	0.171	109.070	0.000	424	0.066	0.260	453.330	0.000

Source: The authors.

Note: *** and ** indicate level of significance at 1% and 5%, respectively.

The results displayed in Table 8 indicate that the bank profitability is negatively sensitive to the interaction between the two risks. The results indicate that the interaction between credit and liquidity risks is negatively and significantly at 1 per cent level with bank profitability measured by ROA and ROE. Nevertheless, it was shown that an increase in bank profitability decreases jointly credit and liquidity risks.

The Role of Institutional Quality in the Bank Risk–Profitability Relationship

In the third step, we test whether institutional quality strengths or threats bank profitability through an indirect effect on credit and liquidity risks. Theoretically, it was confirmed that good institutional quality leads to an improvement of regulatory enforcement and reinforce the lending terms that positively affects the probability of loan repayment and decreases the liquidity risk (Akins et al., 2017; Francis et al., 2014).

Table 9 presents the results of correlation matrix of residuals and Breusch–Pagan test of independence.

Table 9.

Correlation of Disturbances:

	PROF is Measured by ROA			PROF is Measured by ROE
	PROF	CRISK	LIQR	PROF	CRISK	LIQR
PROF	1.0000			1.0000
CRISK	0.1073	1.0000		0.4335	1.0000
LIQR	0.1102	−0.1638	1.0000	0.0172	−0.1454	1.0000
Breusch–Pagan test of independence
χ²(2)	15.690			85.538
Prob.	(0.0013)			(0.0000)

Source: The authors.

Table 9 indicates that the probability of χ² of the Breusch–Pagan test of independence is lower than 5 per cent that we can reject the hypothesis that residuals are independent. Once again, it confirms that the residuals of the three equations are correlated between them.

The results of the third step that consist to test the direct effect of institutional quality on credit and liquidity risks and the indirect effect on bank profitability are given in Table 10.

Table 10.

Results of the Effects of Institutional Quality

				Profitability is ROA		Profitability is ROE
Equations	Variables			Coef.	z	Coef.				z
Profitability	CRISK			-0.571	-4.100***	-1.051				-17.640***
(Equation [1])	LIQR			-0.031	-0.680	0.016				0.780
	SIZE			-0.024	-3.750***	-0.002				-0.820
	CAP			-0.649	-2.170***	-0.482				-3.480***
	LAW			0.063	2.670***	0.047				4.280***
	GOVS			0.009	1.750*	0.009				2.360**
	CRISIS			-0.034	-1.140	-0.064				-4.600***
	GDPG			0.423	1.690*	0.072				0.510
	INF			-0.034	-0.110	0.317				2.260**
	_cons			0.783	5.200***	0.471				6.840***
Credit risk	Prof			-0.070	-4.100***	-0.495				-17.640***
(Equation [2])	LIQR			0.093	5.980***	0.076				5.360***
	SIZE			-0.019	-9.040***	-0.010				-5.020***
	CAP			-0.448	-4.330***	-0.460				-4.900***
	RLAW			-0.009	-2.110**	-0.036				-4.670***
	GOVS			-0.001	-1.740*	0.003				1.030
	CRISIS			0.023	2.180**	0.039				4.130***
	GDPG			0.286	2.710***	-0.170				-1.380
	INF			0.115	1.090	0.182				1.490
	_cons			0.456	9.380***	0.470				10.770***
Liquidity risk	PROF			0.018	0.330	0.090				0.780
(Equation [3])	CRISK			0.982	6.590***	0.886				5.360***
	SIZE			0.018	2.650***	0.015				2.180**
	CAP			1.163	3.600***	1.130				3.490***
	LAW			-0.230	-9.870***	-0.234				-9.890***
	GOVS			0.014	1.620	0.013				1.500
	CRISIS			-0.035	-1.060	-0.034				-1.020
	GDPG			-0.189	-0.570	-0.238				-0.720
	INF			0.103	0.310	0.089				0.270
	_cons			-0.941	-5.770***	-0.893				-5.480***
Equations	Obs.	RMSE	R ²	χ²	p-Value	Obs.	RMSE	R ²	χ²	p-Value
Profitability	409	0.204	0.044	41.220	0.000	409	0.099	0.243	403.610	0.000
Credit risk	409	0.072	0.254	197.640	0.000	409	0.068	0.333	525.220	0.000
Liquidity risk	409	0.257	0.041	59.070	0.000	409	0.222	0.283	189.080	0.000

Source: The authors.

Note: *** and ** indicate level of significance at 1% and 5%, respectively

As for the effect of institutional quality, the results in Table 10 indicate that the coefficient of law and order is positive and significant at 1 per cent level for both ROA and ROE. However, the effect of government stability is found to be only significant at 10 per cent when the ROE is the dependent variable. This means that with higher levels of the law and judicial efficiency, it results an improvement of regulatory enforcement that improves bank-lending activity through the protection of creditors rights that increase bank profitability and ensure the stability of the whole banking system. Our results are similar to the work of Daher (2017) and Olken (2007).

The findings indicate also that the law and order decreases significantly the level of credit and liquidity risks. Coefficient of this variable is negative and significant at 1 per cent level. Under a strong and strict law and order, it results high probability of loan repayment due to strong creditor rights and the rule of law. Also, the high level of judicial efficiency improves the lending terms necessary condition to reduce borrower moral hazard and decrease loan defaults. When the probability of loans repayment increases, the level of credit and liquidity risks decreases. Our findings are in line with and Bushman et al. (2004).

Conclusion

The lack of studies that combine the effect of credit risk, liquidity risk and the interaction effect between each other has motivated us to investigate this topic within the MENA region context. Since credit and liquidity risks are considered as reciprocal risks and recognized as the main determinants of bank profitability, our empirical approach is based on simultaneous equation. More precisely, we use the SUR method of Zellner (1962) and we use a sample of 38 conventional banks observed during the period of 2004–2015. The empirical analysis of this study is based on two steps. The first step consists of testing separately the effect of credit and liquidity risks on bank profitability measured by the ROA and the ROE, and the second step studies the effect of the interaction between both risks on bank profitability.

The empirical results indicate that the profitability of the MENA banks is negatively and significantly sensitive to an increase in credit and liquidity risks. This negative effect is confirmed either when we test separately the impact of these two risks in Equations (2) and (3) or jointly in Equation (4). Qualified as reciprocal risks, an increase in credit risk leads to the increase in liquidity risk for MENA banks and vice versa. Furthermore, we have found that bank profitability decreases significantly the level of credit and liquidity risks. Also, this negative effect is confirmed for either the separate or the interaction effects of these two risks.

By introducing institutional quality variables, we also found that the law and order increases the profitability of MENA banks and decreases both credit and liquidity risks. Surprisingly, we have found that capital adequacy is negatively and significantly correlated with the MENA bank profitability. However, it significantly increases the liquidity risk. This implies more attention to the bank capital allocation in this region.

Implications

The results of this article have some important policy implications for MENA banks. Banks in this region should continue to reinforce the capital adequacy ratio but with more attention on how it will be allocated. There is a strong need to further improve banking supervision, governance practices, credit decision-making and credit risk management. Also, MENA banks are requested to improve their profitability since it reduces both credit and liquidity risks. Since it was confirmed for MENA banks that credit and liquidity risks are strongly reciprocal, we recommend to MENA banks to improve their loans quality and to develop the process of credit risk management in order to reduce the level of NPLs and consequently the level of liquidity risk.

Limitations/Future Research

This study has some limitations. First, the sample used in this study is based only on 38 banks and seems very small to generalize findings. Second, to measure liquidity risk, we have only used loans to deposit ratio (%) and we have not applied for the liquidity coverage ratio (LCR) and the net stable funding ration (NSFR). Third, this current study was limited only to conventional banks.

The empirical findings of this article can be improved by taking into all these limitations. Hence, using a large sample of MENA banks, introducing some Islamic banks and making comparison and using others proxies of liquidity risk certainly improved the results of this article.

Footnotes

Acknowledgements

The authors are grateful to the anonymous referees of the journal for their extremely useful suggestions to improve the quality of the article. Usual disclaimers apply.

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.

Note

ORCID iD

Hakimi Abdelaziz

References

Abreu

Mendes

(2002). Commercial bank interest margins and profitability: Evidence from E.U. countries (University of Porto Working Paper no. 245). Porto: University of Porto.

Abu Hanifa

M-N.

Pervin

Chowdhury

M-M.

Banna

(2015). The effect of credit risk on the banking profitability: A case on Bangladesh. Global Journal of Management and Business Research: C Finance , 15(3), 1–9.

Acharya

Naqvi

(2012). The seeds of a crisis: A theory of bank liquidity and risk taking over the business cycle. Journal of Financial Economics , 106(2), 349–366.

Acharya

Viswanathan

(2011). Leverage, moral hazard, and liquidity. Journal of Finance , 66(1), 99–138.

Adelopo

Lloydking

Tauringana

(2018). Determinants of bank profitability before, during, and after the financial crisis. International Journal of Managerial Finance , 14(4), 378–398.

Aggarwal

Jacques

K. T.

(1998). Assessing the impact of prompt corrective action on bank capital and risk. Economic Policy Review , 4(3), 23–32.

Akins

Dou

(2017). Corruption in bank lending: The role of timely loan loss recognition. Journal of Accounting and Economics , 63, 454–487.

Arif

Anees

A. N.

(2012). Liquidity risk and performance of banking system. Journal of Financial Regulation and Compliance , 20(2), 182–195. Retrieved from https://doi.org/10.1108/13581981211218342

Athanasoglou

Brissimis

Delis

(2008). Bank-specific industry-specific and macroeconomic determinants of bank profitability. Journal of International Financial Markets. Institutions and Money , 18(2), 121–136.

10.

Barros

C. P.

Ferreira

Willians

(2007). Analysing the determinants of performance of best and worst European banks: A mixed logit approach. Journal of Banking and Finance , 31, 2189–2203.

11.

Ben Moussa

(2019). The influence of internal corporate governance on bank credit risk: An empirical analysis for Tunisia. Global Business Review , 20(3), 640–667.

12.

Berger

A. N.

Hanweck

Humphrey

(1987). Competitive viability in banking: Scale, scope and product mix economies. Journal of Monetary Economics , 20, 501–520.

13.

Berríos

M-R.

(2013). The relationship between bank credit risk and profitability and liquidity. The International Journal of Business and Finance Research , 7(3), 105–118.

14.

Bhattacharya

Thakor

(1993). Contemporary banking theory, Journal of Financial Intermediation , 3, 2–50.

15.

Bourke

(1989). Concentration and other determinants of bank profitability in Europe. North America and Australia. Journal of Banking and Finance , 13, 65–79. Retrieved from https://doi.org/10.1016/0378-4266(89)90020-4

16.

Bryant

(1980). A model of reserves, bank runs, and deposit insurance. Journal of Banking & Finance , 4(4), 335–344.

17.

Bushman

Chen

Engel

Smith

, (2004). Financial accounting information, organizational complexity and corporate governance systems. Journal of Accounting and Economics , 37, 167–201.

18.

Cai

Thakor

(2008). Liquidity Risk, Credit Risk and Interbank Competition (SSRN Working Paper).

19.

Cai

Zhang

(2017). How does credit risk influence liquidity risk? Evidence from Ukrainian banks. Visnyk of the National Bank of Ukraine , 241, 21–33.

20.

Calza

Gartner

Sousa

(2003). Modelling the demand for loans to the private sector in the euro area. Applied Economics , 35, 107–117.

21.

Chen

Y. K.

Shen

C. H.

Kao

Yeh

C. Y.

2018. Bank Liquidity Risk and Performance. Review of Pacific Basin Financial Markets and Policies , 21, 1–40.

22.

Chen

H. J.

Lin

K. T.

(2016). How Do Banks Make the Trade-Offs Among Risks? The Role of Corporate Governance. Journal of Banking & Finance , 72, 39–69.

23.

Chortareas

Girardone

Ventouri

(2011). Bank supervision, regulation, and efficiency: Evidence from the European Union. Journal of Financial Stability , 8(4), 292–302.

24.

Cornett

M.M.

McNutt

J.J.

Strahan

P.E.

Tehranian

(2011). Liquidity risk management and credit supply in the financial crisis. Journal of Financial Economics , 101, 297–312.

25.

Cucinelli

(2015). The impact of non-performing loans on bank lending behavior: Evidence from the Italian banking sector. Eurasian Journal of Business and Economics , 8(16), 59–71.

26.

Cuong

Ly. K.

(2015). Liquidity risk, regulation and bank performance: Evidence from European banks. Global Economy and Finance Journal , 8(1), 11–33.

27.

Daher

(2017). Creditor control rights, capital structure, and legal enforcement. Journal of Corporate Finance , 44, 308–330.

28.

Dermine

(1986). Deposit rates, credit rates and bank capital: The Klein-Monti Model Revisited. Journal of Banking & Finance , 10(1), 99–114.

29.

De Nicolo

(2000). Size, charter value and risk in banking: An international perspective . International Finance Discussion N° 689, Board of Governors of the Federal Reserve System.

30.

Diamond

D. W.

Dybvig

P. H.

(1983). Bank Runs, Liquidity and Deposit Insurance. Journal of Political Economy , 91(3), 401–419.

31.

Diamond

Rajan

(2001). Liquidity Risk, Liquidity Creation, and Financial Fragility: A Theory of Banking. Journal of Political Economy , 109(2), 287–327.

32.

Felmlee

Hargens

(1988). Estimation and hypothesis testing for seemingly unrelated faris and felmlee 71 regressions: A sociological application. Social Science Research , 17, 384–399.

33.

Flamini

Valentina

McDonald

Liliana

(2009). The determinants of commercial bank profitability in sub-saharan Africa (IMF Working Paper). Washington, DC: IMF.

34.

Francis

B. B.

Hasan

Zhu

(2014). Political uncertainty and bank loan contracting. Journal of Empirical Finance , 29, 281–286

35.

García-Herrero

Gavila´

Santaba´rbara

(2009). What explains the low profitability of Chinese banks? Journal of Banking and Finance , 33, 2080–2092.

36.

Ghenimi

A.;

Chaibi

Omri

M. A.

(2017). The effects of liquidity risk and credit risk on bank stability: Evidence from the MENA region. Borsa Istanbul Review , 17(4), 238–248.

37.

Ghosh

(2017). Corporate governance reforms and bank performance: Evidence from the Middle East and North Africa. Corporate Governance , 17(5), 822–844.

38.

Goddard

Liu

Molyneux

Wilson

(2010). Do bank profits converge? European Financial Management , 19(2), 345–365.

39.

Gorton

Metrick

(2011). Securitized Banking and the Run on Repo. Journal of Financial Economics , 104(3), 425–451

40.

Greuning

H.V.

Bratanovic

S.B.

(2004). Analysing and Managing Banking Risk: A Framework for Assessing Corporate Governance and Financial Risk (2nd ed.). Washington, DC: The World Bank.

41.

Hakimi

Hamdi

Djelassi

(2011). Financial liberalization and banking profitability: A panel data analysis for Tunisian banks. International Journal of Economics and Financial Issue , 1(2), 19–32.

42.

Hakimi

Zaghdoudi

(2017). Liquidity risk and bank performance: An empirical test for Tunisian banks. Business and Economic Research , 7(1), 46–57.

43.

Hamdi

Hakimi

(2019). Does liquidity matter on bank profitability? Evidence from a nonlinear framework for a large sample. Business and Economics Research Journal , 10(1), 13–26.

44.

Hamdi

Rachdi

Hakimi

Guesmi

(2018). What drives banking profitability after the international financial crisis of 2008? Evidence from Eurozone banks. Bankers, Markets & Investors , 152–153, 21–33.

45.

Hanushek

E. A.

Jackson

J-E.

(1977). Statistical methods for social scientists . New York, NY: Academic Press.

46.

Xiong

(2012). Dynamic Debt Runs. Review of Financial Studies , 25(6), 1799–1843.

47.

Imbierowicz

Rauch

(2014). The relationship between liquidity risk and credit risk in banks. Journal of Banking & Finance , 40, 242–256.

48.

Isshaq

Amoah

Appiah-Gyamerah

(2019). Non-interest income, risk and bank performance. Global Business Review , 20(3), 595–612.

49.

Kim

(2015). Liquidity risk regulation and bank performance. Evidence from European bank. Global Economy and Finance Journal , 8(1), 11–33.

50.

Kim

Cho

(2019). Seemingly unrelated regression tree. Journal of Applied Statistics , 46(7), 1177–1195.

51.

Kmenta

(1971). Elements of econometrics . New York, NY: Macmillan.

52.

Kosmidou

(2008). The determinants of banks’ profits in Greece during the period of EU financial integration. Journal of Managerial Finance , 34(3), 146–159.

53.

Kosmidou

Tanna

Pasiouras

(2005). Determinants of profitability of domestic UK commercial banks: Panel evidence from the period 1995–2002. In Proceedings of the 37th Annual Conference of the Money Macro and Finance (MMF) Research Group, Rethymno, Greece, September .

54.

Laryea

tow-Gyamfi

Azumah Alu

(2016). Nonperforming loans and bank profitability: Evidence from an emerging market. African Journal of Economic and Management Studies , 7(4), 1–37.

55.

Mamatzakis

Bermpei

(2014). What drives investment bank performance? The role of risk, liquidity and fees prior to and during the crisis. International Review of Financial Analysis , 35, 102–117.

56.

Marozva

(2015). Liquidity and bank performance. International Business & Economics Research Journal , 14(3), 453–462.

57.

Nikolaidou

Vogiazas

(2017). Credit risk determinants in sub-saharan banking systems: Evidence from five countries and lessons learnt from Central East and South East European countries. Review of Development Finance , 7, 52–63.

58.

Olagunju

David

A. O.

Samuel

(2012). Liquidity management and commercial banks’ profitability in Nigeria. Research Journal of Finance and Accounting , 2(7–8), 24–38.

59.

Olken

(2007). Monitoring Corruption: Evidence from a Field Experiment in Indonesia. Journal of Political Economy , 115, 200–249.

60.

Partovi

Matousek

(2019). Bank efficiency and non-performing loans: Evidence from Turkey. Research in International Business and Finance , 48, 287–309.

61.

Pasiouras

Kosmidou

(2007). Factors influencing the profitability of domestic and foreign commercial banks in the European Union. Research in International Business and Finance , 21(2), 222–237.

62.

Reinhart

C M.

Rogoff

K. S.

(2011). From Financial Crash to Debt Crisis. American Economic Review , 101(5), 1676–706.

63.

Sohaimi

A. N. A.

(2013). Liquidity risk and performance system in Malaysia (SSRN Working Paper).

64.

Sufian

(2009). Factors influencing bank profitability in a developing economy: Empirical evidence from Malaysia. Global Business Review , 10(2), 225–241.

65.

Tabari

Ahmadi

Emami

(2013). The effect of liquidity risk on the performance of commercial Banks. International Research Journal of Applied and Basic Sciences , 4(6), 1624–1631.

66.

Tan

(2016). The impacts of risk and competition on bank profitability in China. Journal of International Financial Markets, Institutions and Money , 40, 85–110.

67.

Thorsten

Heiko

Thomas

Natalja

(2009). Bank ownership and stability: Evidence from Germany (IMF Working Paper 09/200). Washington, DC: IMF.

68.

Wagner

(2007). Aggregate liquidity shortages, idiosyncratic liquidity smoothing and banking regulation. Journal of Financial Stability , 3, 18–32.

69.

Zellner

(1962). An efficient method of estimating seemingly unrelated regression equations and tests of aggregation bias. Journal of the American Statistical Association , 57, 500–509.

70.

Zellner

(1963). Estimators for seemingly unrelated regression equations: Some finite sample results. Journal of the American Statistical Association , 58, 977–992.

71.

Zolkifli

N-A.

Abdul Hamid

Hawati Janor

(2015). Liquidity risk and performance: The case of Bahrain and Malaysian banks. Global Economy and Finance Journal , 8(2), 95–111.