Keywords

Performance, Emerging markets, Financial structure.

Introduction

In the wake of high and rising oil prices since 2003, the member states of the GCC, have seen dynamic economic development, enhancing their role in the global economy as investors and trade partners. The International financial crisis of 2008 produced a negative impact on the economic performance of The MENA region and the GCC counties are of no exception. The GCC countries' strong orientation towards oil and gas Implies that the diversification of their economies is a key challenge. The fall in the oil prices remain the new challenges to the GCC economies in the light of prevailing inflation rate, Interest rate and instability of the neighboring countries. The UAE has the second largest economy in the Arab world (after Saudi Arabia) with a gross domestic product (GDP) of USD 414 billion in 2019. The UAE economic policy at both the federal and emirate levels will continue to focus on reducing the economy's dependence on the Hydrocarbons sector (Dev. plan, 2021). The UAE hydrocarbon sector is estimated to have exhibited a growth of 3.4% in 2019. However, non-oil activities advanced at a softer pace growing by 1%. The Spread a convid-19 is expected to impact trade and supply chain movements in 2020 and 2021. The insight into the UAE banking sector shows the hottest topics of mergers and acquisitions in 2019 and 2020. The output of the financial sector activities in current prices achieved an increase from US$ 36.50 billion in 2017 to USD 38. 1 billion in 2018 with a growth rate of 4.6%. The number of listed companies in the UAE securities market rose from 130 in 2017 to 137 in 2018 and to 140 in 2019 (72 in Abu Dhabi and 68 in Dubai). The contribution of the financial sector to the GDP declined from 9.6% in 2017 to about 9.2% in 2018 and 9.1% in 2019.

The total number of banks operating in the UAE's financial market reached 59 banks by the end of Dec. 2019, including 21 national banks and 38 foreign banks. The total bank branches of national banks are around 981 in 2020 and 170 branches of foreign banks, coming to the total of 1151 bank branches. According to the World Bank database, The UAE population grew to about 9631000 in 2019 and the workforce is about 7384000 in 2019 with a ratio of 77%. Based on Goldsmith model of 1965, we could state that the banking density ratio in the UAE is 1.20 with a positive deviation of 20%. The total assets of banks operating in the UAE reached US$ 827 billion by the end of Dec 2019 with an annual growth rate of 6%. The total deposits of banks grew strongly in 2018 with a rate of 7.9% as compared to the year 2019, with a growth rate of 3.6%. The high level of deposits (495 USD Billion in 2019) and capital solvency and capital adequacy of the banking sector (17.3% in 2019), improve the capability and ability of banks to continue their banking activities, with a total banking credit growth rate of 4.8% in 2018 and 3.2% in 2019. The total banking credit in the UAE reached USD 465 billion by the end of Dec 2019. It is apparent that the ratio of "loans to deposits" continued to decline gradually to reach 93.9% in 2019 compared to 94.3% in 2018 and 97.1% in 2017.

Deposits by type of national banks (conventional VS Islamic), represents 78.5% and 21.5% of the total deposits at the end of 2019, respectively. Meanwhile, the share of national and foreign banks' deposits represents 88.6% and 11.4% respectively in 2019.

Despite the decline in Lending to individuals (less than 30% of total credit to the private sector), the increase in credit to private corporates led to a rise in overall credit growth to the private sector. In this connection, it is worth to document the UAE banking lending to the small, medium enterprise. Micro, small, medium enterprises (MSMES) play vital role in diversifying the UAE economy and boosting job creation. The survey conducted by The Central Bank of the UAE (CBUAE) on MSMES, shows 22% micro, 21% small and 57% are medium-sized enterprises. The CBUAE Survey (2019) revealed that the primary challenge MSMES face is related to financial statements which at least 25% of respondents (620 firms) fail to provide to raise institutional finance. 69% of the respondents consider themselves financially constrained with high interest rate (32% reported paying more than 10%.

The performance highlights of the UAE banking sector as an average of the last five years (2015-2019) could be reflected into following Table 1.

Table 1 The Performance of the UAE Banking Sector (2015-2019)
Average Net profit margin percentage	11.5 %
Average cost to income ratio	1.6%
Average capital adequacy ratio	17.33%
Average Return on Equity (ROE)	13.7%
Average Return on Total Assets (ROTA)	1.71%
Average liquidity Ratio	33.52%
Average Non-performing loan ratio	3.13%

In addressing the association and the degree of interaction between the financial structure, lending capacity and the performance of banks operating in a competitive financial market, several questions are expected to be raised and answered:

1. Is there a relationship between financial structure, lending capacity and efficiency of banks regardless of size and nature in the UAE?

2. In what sense the components of the financial structure determine the lending capacity of banks and consequently influence their efficiency and effectiveness?

Literature Review

This part of the paper will shed light on the related three topics of financial structure of the banking sector in the emerging market, lending behavior and determination of lending capacity and the of performance of banks.

The financial structure of bank is playing an important role in determining its performance. In addressing the relationship between bank's capital structure and the performance EL-Chaarani & EL-Abiad (2019) found that the performance is influential by the capital structure. In addressing the issue in the Middle East at post crisis era (2011-2016), the authors found that short term debt and total debt affect ROTA negatively, but they have a positive effect on the ROE. In a study conducted by Taani (2013), to examine the impact of capital structure on the performance of banks in the Jordanian economy, the research findings reveal that both ROE and net profit are affected significantly by the total debt. Saeed, Gull and Rasheed; 2013 investigated the impact of financial structure on performance of banks in Pakistan during the period (2007-2011). The study was built around many independent variables such as total debt, and size. ROTA and ROE were used to determine the performance as well as EPS. The study provides an evidence on a positive relationship between the size of the bank and the three dependent variables ROTA, ROE and EPS. Swai1 et al (2016) studied 14 banks operating in Tanzania for the period (1998-2010). He found that size, profitably and growth of the bank were the major determinants of the bank's financial structure. In the West African context, Musah (2017) examined how can the financial structure of 23 banks operating in Ghana affects the profitability. The research results provided an evidence that there is negative relationship between profitability and short- and long-term debt ratios. As the size of the bank was considered in Musah (2017) study as of one of control variables, the study showed a positive relationship between the bank size and the profitability indicators.

Ebenezer (2015) has conducted a study on relationship between financial structure and performance of banks in sub-Saharan Africa (2000-2006). The performance was measured by the dependent variables (ROTA, ROE, NPM) whereas total debt ratio was used as independent variable. Growth, tax, and inflation rates were used as control variables. Based on Granger causality test, the study revealed that the performance of bank affects its financial structure and not via versa. On The same trend, Akhtar et al (2016) had studied to influence of the financial structure of five banks in Pakistan for the period 2005-2015. The study employed five independent variables such as profitability, tangibility, liquidity ratios, growth rate and interest rate to determine the performance. By applying the pooled analysis, the study results indicated that all the employed Independent variables associated positively with the performance of the selected banks. It is apparent that most scholars used ROTA, ROE and NPM as measures of the performance of the banking sector.

Lending behavior of banks vary from bank to bank and from economy to economy. Ladime et al. (2013), studied the banking lending behavior in Ghana in pre-financial crisis (1997-2006). They used three characteristics, namely financial structure, bank size, macro-economic characteristics, which include the Central Bank lending rate and exchange rate and finally the industry characteristics. The study provides an evidence that on a positive impact of both bank's characteristics on lending behavior of the bank. On the other hand, the study indicated that the relationship between the lending behavior and macro-economic characteristics was negative. In a very comprehensive study conducted on 89 commercial banks based in Nigeria, under long-run analysis (1980-2005) Olokoyo (2011) found that the lending behavior of the banks was influenced by their deposits. Following the same methodology, Ayieyo(2016) on examining the lending behavior of banks in Nigeria, he found that the impact on the lending behavior was positive in relation to the volume of deposits and negative in relation to the interest rate.

Under co- integration analysis, study conducted by Olusanya et al. (2012), to investigate the factors that influence the lending behavior of commercial banks in Nigeria during the period 1975-2010, using the same variables of Ayieyo (2016). The study results give evidence on positive impacts of volumes of deposits and GDP at current market price on the lending behavior expressed by loans and advances. However, a negative impact of interest rate on lending behavior was found. By using descriptive statistical tools and regression analysis, Khangalah (2016), investigated the impacts of factors such as liquidity, capital adequacy and interest rate on the lending behavior of Commercial banks operating in the Kenyan financial market. The research findings provided indication that the lending behavior of these banks positively associated with capital adequacy and liquidity. On the other hand, interest rate and asset quality negatively associated with the lending behavior.

Performance of banks can be measured through many ratios (ROE, ROA, and NPM). A study to measure the performance during the 2008 financial crisis of two sets of banks 4 Islamic banks and 7 conventional banks based in Saudi Arabia, was conducted by Mbarek and Zehri (2016) for the period 2005 to 2014. The different nine ratios used by the authors were divided equally, three ratios for profitability, three for efficiency and three to assess the risk. By using the logit regression, the results revealed that Islamic banks have low degree of efficiency, but it is still more profitable than conventional banks. Banerjee (2018) investigated the performance of 21 National Commercial banks operating in the UAE between 2014 and 2017. The Study divided performance into three categories: internal based performance (ROA); market-based performance (Tobin's Q) and economic based performance (EVA). By applying the multiple regression and correlation, the results showed that all independent variables except size of the bank have a positive impact on ROTA. On the other hand, there was a positive correlation between Tobin’s Q ratio and all independent variables, but it was very weak.

The relevant literature showed that the size of the bank is one of the independent variables that affect the performance of bank in term of its profitability (Aladwan, 2015; Samad, 2015)

The conceptual framework of this investigation is built around the interaction between the financial structure, lending behavior and the UAE banking performance from 2008 to 2018. Based on the critical review of the relevant literature, we end up with two main hypotheses:

a. Hypothesis 1: ROA, ROE and NIM are influenced by the financial structure of the bank.

b. Hypothesis 2: There is an association between ROA, ROE and NIM and lending capacity of the bank.

Research Methodology

Theoretical Model

The financial system in the UAE is entirely regulated and managed by the Central Bank considering the liberalization and coexistence of conventional and Islamic financial institutions. The review of the literature asserted that the conventional and non-conventional banks were criticized by the demand side for the lack of their response to the financial needs of the productive sectors (Ahmed, 2020). This debate raises the question on the degree of association between the financial resources of banks as exemplified into equity, deposits, and bank borrowing and the lending capacity and in turn the overall performance of the banking system.

To determine the effect of total financial resources components and lending capacity on bank's performance of UAE banks, eight commercial banks, namely Abu Dhabi Commercial Bank, Emirates NBD, Abu Dhabi Islamic Bank, National Bank of Umm AlQaiwain, Dubai Islamic Bank, Sharjah Islamic Bank, National Bank of Abu Dhabi and Mashreq Bank were selected. Annual panel data covers the period from 2008 to 2018 was collected from the financial statements of the selected commercial banks. A description of the characteristics of the variables used in the study is given in Table 2 below which reports their statistical means, median, and standard deviation.

Table 2 Descriptive Statistics of Study Variables
	LC	NIM	%ΔLC	ROA	ROE	TD	SIZE	TDR
Mean	155.488	2.26590	0.11964	1.64931	11.2085	0.64872	170.431	84.5236
Median	117.000	2.09500	0.07080	1.55500	11.3050	0.63050	122.500	87.1250
Maximum	734.000	5.41000	4.71428	4.06000	21.0300	0.81100	763.000	91.2800
Minimum	11.0000	-0.39000	-0.93457	-0.32000	-2.69000	0.50400	12.0000	69.5600
Std. Dev.	144.132	0.92838	0.60934	0.63184	3.91319	0.07104	162.579	5.88260
Observations	88	88	88	88	88	88	88	88

In this paper, we tried to examine empirically the impacts of the financial structure components and lending capacity on the performance of the bank. Thus, performance is the dependent variable in the empirical investigation. A frequently used measures of bank performance are Return on Assets (ROA), Return on Equities (ROE), and Net Interest -Margin (NIM).

Since the study targeted to investigate the effect of the financial structure components and lending capacity on the performance of banks, financial structure, and lending capacity variables are taken as the explanatory variables. Thus, we take total debt ratio which is total debt over total assets, change in lending capacity, lending capacity and total deposits, as explanatory variables. It’s expected that an increase in all explanatory variables will improve bank's performance as they increase the net income of the bank.

In order to separate the effects of financial structure and lending capacity on bank’s performance, a control variable is applied in this study, and it is about one of the bank-specific variables that’s expected to influence the performance of the bank and therefore are controlled. Bank-specific control variable employed in this study is bank size. According to recent banking theories, the banking efficiency resulting from economies of scale is related to the bank size (Flamini, et al 2009). Bank size is generally considered a relevant determinant of bank performance. Smirlock (1985) finds a positive and significant relationship between size and bank profitability. To capture the relationship between size and bank profitability, we use the log total assets of the banks as a proxy for bank size. Accordingly, Bank performance and bank size should be positively related.

In a summary, Table 3 below provides a recap of the used variables and their corresponding measurements.

Table 3 Variables of Study
Variable	Legend	Measurement
Dependent:
Return on Total assets	ROA	Net income over total assets
Return on Equity	ROE	Net income over total equity
Net-Interest Margin	NIM	Net Income over total finance
Independent:
Total Debt Ratio	TDR	Total debt over total assets
Percentage Change in lending capacity	%ΔLC	lending capacity- previous lending capacity over previous lending capacity
Lending capacity	LC	Total financial resources- reserves
Total deposits	TD	Demand deposits+ quasi deposits
Control:
Size	SIZE	Natural logarithm of total assets

To investigate the above-mentioned relationship, we form the following estimable models:

ROA= f (LC, %ΔLC, TD, TDR, SIZE) (1)

ROE= f (LC, %ΔLC, TD, TDR, SIZE) (2)

NIM = f (LC, %ΔLC, TD, TDR, SIZE) (3)

Analytical Methods

A common panel data regression model looks like

Y_it = a + b X_it + μ_it (4)

where Y is the dependent variable, X is the explanatory variable, a and b are coefficients, "i" denoting cross sections and “t” denoting the time dimension of panel data while “μit” denotes the unobservable factors affect in the panel data modelling stated above.

The fixed versus random effects issue has generated debate in the biometrics and statistics literature which has spilled over into the panel data econometrics literature (Baltagi, 2005). The debate is about which is better fixed effect model or random effect model.

Fixed effect model assumes that there is one true effect size, which underlies all the studies in the analysis, and that all differences in observed effects are due to sampling error. In a fixed effect model, the unobserved variables can have any associations whatsoever with the observed variables. Fixed effects model control for the effects of time-invariant variables with time-invariant effects. This is true whether the variable is explicitly measured or not. (Baltagi, 2005) stated that when our inference is limited to the individual behavior of a group of firms, the fixed effect model is an appropriate choice for prediction of the relationship between dependent and independent variables in a panel data model.

Accordingly, the econometric model for fixed effect is as follow:

Y_it = (a + μ_i) + X '_it β + v_it (5)

where a and β are constant parameters; “μi” is also a parameter but assumed to be fixed and is estimated for the purpose of inference in the panel data. The remaining unobservable factors are stochastic with “νit” which are distributed identically as IID (0, σ_ν²) and changes with individual banks and time invariants. Also, it is assumed that under the fixed effect model and for all “i” and “t”, the explanatory variable (“X_it”) does not depend on “ν_it” for the purpose of inference. Our model can be converted into fixed effect as follows:

ROA_it = (β_o+ μ_i) + β₁ LC_it + β₂ %ΔLC_it + β₃ TD_it + β₄ SIZE_it +β₅ TDR_it + ν_it (6)

ROE_it = (b_o+ μ_i) + b₁ LC_it + b₂ %ΔLC _it + b₃ TD_it + b₄ SIZE_it +b₅ TDR_it + ν_it (7)

NIM_it = (c_o+ μ_i) + c₁ LC_it + c₂ %ΔLC _it + c₃ TD_it + c₄ SIZE_it +c₅ TDR_it + ν_it (8)

Under random-effects model we allow that the true effect could vary from study to study. For example, the effect size might be higher or lower in studies. Because studies will differ in the mixes of participants and in the implementations of interventions, among other reasons, there may be different effect sizes underlying different studies. In a random effect model, the unobserved variables are assumed to be uncorrelated with all the observed variables. Random effect model can be estimated via Generalized Least Squares. Generalized Least Squares (GLS) method is used in case of heteroskedasticity or auto-correlation problems. This method gives more better results and high significance most of time. According to Baltagi (2005) the random effect model is appropriate in a situation where the researcher intends to draw individuals randomly from a large population. In this case the term μ_it assumed to be random.

Then the random effect model looks like:

Y_it =a + X '_it β + (μ_it + v_it) (9)

Where μit is distributed identically as IID (0, σ_μ²), νit also distributed identically as IID(0, σ_ν²) and that the values of μ_it and ν_it are independent. Also, there is not any relationship between the values of X_it and the values of μ_it and ν_it.

Our model can be converted into random effect as follows:

ROA_it = β_o + β₁ LC_it + β₂ %ΔLC _it + β₃ TD_it + β₄ SIZE_it +β₅ TDR_it + (μ_it+ ν_it ) (10)

ROE_it = b_o + b₁ LC_it + b₂ %ΔLC _it + b₃ TD_it + b₄ SIZE_it +b₅ TDR_it + (μ_it+ ν_it ) (11)

NIM_it = c_o + c₁ LC_it + c₂ %ΔLC _it + c₃ TD_it + c₄ SIZE_it +c₅ TDR_it + (μ_it+ ν_it ) (12)

In panel data analysis, the Hausman specification test (1978) is run for differentiating between fixed effects model and random effects model. The Hausman test examines the consistency of the Generalized Least Square (GLS) (random effects) estimates. The null hypothesis is that the random effect estimates are consistent, that is, that the disturbances and X's are independent. In this case, random effects model is preferred under the null hypothesis due to high efficiency. While under the alternative hypothesis fixed effect is at least as consistent and thus preferred.

After estimating the above fixed effect models in (6,7,8) and random effect model in (10,11,12), we shall have to decide which model is good to accept using the Hausman test. If we get statistically significant p-value we shall use fixed effect model, otherwise random effect model will be used.

Empirical Results

The choice of model in panel data must be based on information about the individual-specific components and the exogeneity of the independent variables. In this study the Hausman test is used for testing whether fixed or random effects model is appropriate, by identifying the presence of endogeneity in the explanatory variables. In Table 4 we report the results of Hausman specification test for model selections.

Table 4 Hausman Specification Test for Model  Selections
Dependent variable	Chi-Sq. Statistic	Chi-Sq. d.f.	Significance Level
ROA	12.15	5	0.0 33
ROE	9.47	5	0.091
NIM	23.35	5	0.00 0

From Table 4 above its obvious that that the p-values of the three equations are statistically significant which mean that the fixed effect models are appropriate and accepted in this current study. Fixed effects assume that differences between individuals (cross section) can be accommodated from different intercept. To estimate the Fixed Effects Model with different intercept between individuals, the dummy variable technique is used. Such estimation models are often referred to as the Least Squares Dummy Variable technique or abbreviated LSDV.

Estimation results of fixed effect models in (6,7,8) are reported in Table 5 below.

Table 5 Estimation of Fixed Effect Models for ROA, ROE, and NIM
Dependent variable	Regressor	Coefficient	t-Ratio	Significance Level	R2	Calculated-F	Significance Level
ROA	Constant	-1.236	-1.140	0.258	0.90	58.19	0.000
	LC	6.186	3.457	0.001
	%ΔLC	0.091	4.347	0.000
	TD	1.335	2.878	0.005
	SIZE	0.146	2.449	0.017
	TDR	-0.054	-8.821	0.000
ROE	Constant	-36.36	-3.974	0.000	0.95	118.31	0.000
	LC	8.118	0.725	0.471
	%ΔLC	0.884	6.432	0.000
	TD	19.155	5.584	0.000
	SIZE	0.869	2.049	0.044
	TDR	0.275	7.905	0.000
NIM	Constant	-4.263	-1.896	0.062	0.96	49.84	0.000
	LC	8.461	4.038	0.000
	%ΔLC	0.099	6.336	0.000
	TD	3.115	6.003	0.000
	SIZE	0.352	5.267	0.000
	TDR	-0.062	-5.664	0.000

The results in Table 5 above suggest that the estimated relationships are statistically significant at 1% significant level, as indicated by the high values of the F- statistic which are equal to 58.19, 118.31, and 49.84 for the equation of ROA, ROE, and NIM, respectively. The values of R² which measures the overall goodness of fit of the estimated models, are high, where 90%, 95%, and 96% of variations in ROA,ROE, and NIM of selected banks in UAE, respectively, have been explained by variations in lending capacity (LC), percentage change in lending capacity (%ΔLC), total deposits (TD), total debt ratio (TDR) and bank size (SIZE). Furthermore, the coefficients of the most explanatory variables of the three models are highly statistically significant as indicated by the high values of t-ratios except the coefficient of lending capacity (LC) in the ROE model is statistically insignificant. As expected lending capacity, percentage change in lending capacity, total deposits, and bank size exert positive effect on the performance of the banks in UAE as represented by ROA, ROE, and NIM, i.e. an increase in lending capacity, percentage change in lending capacity, and total deposits improve bank's performance in UAE. Also, banks with greater market share obtain higher profits by pricing above competitive levels.

Although the coefficient of total debt ratio (TDR) is positive as expected in ROE model, they have wrong negative signs in other two models (ROA and NIM models). In this regard there are many empirical results about the relationships between total debt to total assets ratio and bank's performance. However, some have found out positive impacts as others negative impacts. Nkegbe and Ustarz (2015) found negative relationship between Bank debt and bank's performance in Ghana. Also, Ramadan and Ramadan (2015) used pooled ordinary least squares and realized a negative impact of short?term debt to total assets, long?term debt to total assets and total debts to total assets on the performance of Jordanian firms during the period 2005 – 2013. On the other hand, Nikoo (2015) identified a positive effect of capital structure on banks’ performance using data of seventeen commercial banks from 2009 through 2014. Thus, our findings about negative impact of total debts to total assets on the performance of UAE banks (ROA, NIM) are consistent with the studies of Nkegbe and Ustarz (2015) and Ramadan and Ramadan (2015).

Findings and Discussion

The purpose of this study is to determine the effect of total financial resources components and lending capacity on bank's performance of UAE banks, eight commercial banks, namely Abu Dhabi Commercial Bank, Emirates NBD, Abu Dhabi Islamic Bank, National Bank of Umm AlQaiwain, Dubai Islamic Bank, Sharjah Islamic Bank, National Bank of Abu Dhabi and Mashreq Bank were selected. Annual panel data covers the period from 2008 to 2018 was collected from the financial statements of the selected commercial banks. The findings demonstrate that most coefficients of the variables exert positive and significant effect on the performance of the bank except the coefficients of the total debt ratios which exert negative effect in ROA and NIM models although they are statistically significant.

Conclusions and Policy Implications

The purpose of this study is to determine the effect of total financial resources components and lending capacity on bank's performance of UAE banks, eight commercial banks, namely Abu Dhabi Commercial Bank, Emirates NBD, Abu Dhabi Islamic Bank, National Bank of Umm AlQaiwain, Dubai Islamic Bank, Sharjah Islamic Bank, National Bank of Abu Dhabi and Mashreq Bank were selected. Annual panel data covers the period from 2008 to 2018 was collected from the financial statements of the selected commercial banks. The findings demonstrate that most coefficients of the variables exert positive and significant effect on the performance of the bank except the coefficients of the total debt ratios which exert negative effect in ROA and NIM models although they are statistically significant.

The findings of the research have policy implications for the development of the banking sector in the UAE and the GCC at large. This is a very natural process of empirical research, starting with the formulation of research questions from a critical review of the literature and our background. This part of the study is addressed primarily to those who oversee policy formulation in the financial sector. The policy implications of this research will be structured around the following:

1. The variation between ROE and the ROA of all banks regardless of size and nature provides evidence on the variation between equity and total assets. Moreover, it shows that the operational cost and branching are not in line with the banking technology invested, which accounts for 10-15% of the total asset. This calls for policy measures on operational performance and typology of banking technology investment and banking approach to reach its target in an efficient and effective way.

2. The study shows a high degree of similarity between conventional and Islamic banks in connection with deposit structure and the growth rate. This question the policy for Islamic banks established by the Central Bank.

3. Because there is negative relationship between total debt ratio and some performance indicators (ROA, NIM), we recommend that banking managers plus policymakers should design prudent decisions aimed at reducing overreliance on debts to yield optimal capital structure levels to enable banks to remain at the top of the profitability game competitively in the banking sector.

References

1. Ahmed, S.A., Ahmed, I.E., & Khan, K.M. (2020). Banking behaviour ss business finance in the UAE: Supply Side Assessment based on Importance Performance Map Analysis. Seminar paper, Ajman university.
2. Abdullah, A.M., & Naser, K. (2015). Determinants of capital stucture of Banking sector in GCC: An empirical investgation. Asian Economic and Financial Review, 5(7), 959.
3. Akhtar, N., Bano, M.M., Bano,S., Zia, H., & Jameel, N. (2016). Capital structure impact on banking sector performance in Pakistan. International Review of Management and Business Research, 5(2), 519-535.
4. Al Shamsi, F., Aly, H., & El-Bassiouni, M. (2009). Measuring and explaining the efficincies of the United Arabb Emirates banking system. Applied Economics, 41(27), 2505-3519.
5. Aladwan, M. (2015). The impact of bank size on profitability" an empirical study on listed Jordanian commercial banks. European Scientific Journal, 11(34).
6. Al-Tamimi, H.H. (2010). Factors influencing performance of the uae islamic and conventional national banks. Global Journal Of Business Research, 4(2), 1-9.
7. Ayieyo, J. (2016). Determinants of lending behavior in selected commercial banks in Kenya. International Journal of Economics, Commerce and Management, 14(9), 767-782.
8. Banerjee, A. (2018). Can ratios predict the financial performance in banks: A case of national banks in United Arab Emirates. International Journal of Accounting and Financial Reporting, 8(4), 2162-3082.
9. Baltagi, B.H. (2005). Econometric Analysis of Panel Data. 3rd Edition, John Wiley & Sons Inc., New York.
10. El-Chaarani, H., & El-Abiad, Z. (2019). Analysis of capital structure and performance of banking sector in Middle East countries. International Journal of Economics and Financial Issues, 9(2), 1.
11. Ebenezer, B. (2015). Determinants of capital structure of banks: Evidence from Sub-Sahara Africa. Asian Economic and Financial Review, 5(4), 624-640.
12. Flamini, V., McDonald,C., & Schumacher, L. (2009). The Determinants of Commercial Bank Profitability in Sub-Saharan Africa, IMF working paper WP/09/15.
13. Guisse, M.L. (2012). Financial performance of the malaysian banking industry: Domestic vs foreign banks. (Doctoral dissertation) Eastern Mediterranean University.
14. Hausman, J.A. (1978). Specification tests in econometrics.  Econometrica, 46(6), 1251-1271. 
15. Khangalah, M.O. (2016). Determinants of commercial banks’ lending behavior in Kenya: Case of state owned banks in Kenya Unpublished MBA Project, University of Nairobi.
16. Kothari, C. (2004). Research Methodology: Methods and Techniques. New Age International Publishers.
17. Ladime, J., Osei, K.A., & Sarpong-Kumankoma, E. (2013). Determinants of bank lending behaviour in Ghana. Journal of Economics and Sustainable Development, 4(17), 42-47.
18. Mbarek, N., & Zehri, F. (2016). Banks' performance in KSA during financial distress: A comparative study islamic and conventional banks. Arabian J. Bus Manag Review S, 1, 1-6.
19. Mehta, A. (2012). Financial performance of UAE banking sector-A comparison of before and during crisis ratios. International Journal of trade, economics and finance, 3(5), 381-387.
20. Musah, A. (2017). The impact of capital structure on profitablity of commercail banks in Ghana. Asian Journal of Economic Modelling, 6(1), 21-36.
21. Nikoo, S.F. (2015). Impact of capital structure on banking performance, evidence from Tehran stock exchange. International Research Journal of Applied and Basic Sciences, 9(6), 923?–?927.
22. Nkegbe, P.K., & Ustarz, Y. (2015). Banks performance in Ghana: Trends and determinant, GJDS, 12(1).
23. Olokoyo, F.O. (2011). Determinants of commercial banks’ lending behavior in Nigeria. International Journal of Financial Research, 2(2), 61-72.
24. Olusanya, S., Oyebo, A., & Ohadebere, E. (2012). Determinants of lending behaviour of commercial banks: Evidence from Nigeria, a co-integration analysis (1975-2010).  Journal of Humanities and Social science, 5(5), 71-80.
25. Ramadan, Z.S., & Ramadan, I.Z. (2015). Capital structure and firm’s performance of Jordanian manufacturing sector. International Journal of Economics and Finance, 7(6), 279?–?84.
26. Saeed, M.M., Gull, A.A., & Rasheed, M.Y. (2013). Impact of capital structure on banking performance. Interdisciplinary Journal Of Contemporary Research In Business, 4(10), 393-403.
27. Samad, A. (2015). Determinants bank profitability: Empirical evidence from Bangladesh commercial banks. International Journal of financial reseach, 6(3), 173-176.
28. Shoaib, A., & Siddiqui, M. (2011). Measuring performance through capital structure. African Journal of Business Management, 5(5), 1871-1879.
29. Smirlock, M. (1985). Evidence on the (Non) relationship between concentration and profitability in Banking. Journal of Money Credit and Banking, 17, (1), 69-83.
30. Swai1, A., Lwiza, D.V., & Ndanshau, M.O. (2016). determinants of bank capital structure-does bank ownership matter? evidence from Tanzania. Business Management Review,19(1)
31. Taani, K. (2013). Capital structure effects on banking performance: A case study of Jordan. International Journal of Economics. Finance and Management Sciences, 1(5), 227-233.