An Empirical Study into Comparing Conventional and Islamic Banks in the UAE

Banking system constitutes the fundamental pillar of every economy. Banks acts financial intermediaries between sectors that have excess funds and those that are in deficit. Islamic banks operate under Sharia principles of risk sharing and interest prohibition as contrasted with conventional banks that buy capital to pool funds and sell capital to generate interest income or profit. This paper applies banks’ internal factors related to their balance sheet and income statement and using a total of 23 financial ratios pertaining to the internal factors, it attempts to compare and contrast between conventional and Islamic banks. This research explains the structure, operation and management of banks in the UAE coupled with the functioning of Islamic banks. The paper also aims to determine the profitable and efficient banks among the chosen sample. The sample includes 12 banks, equally distributed between Islamic and conventional banks using data between the periods of 2014 2018. The sample is broadly categorized based on profitability ratio, efficiency ratio, asset indicator ratio and risk ratios. Correlation and Regression analysis is used to determine a substantial ratio analysis between conventional and Islamic banks. Results from the study reveal indicators of financial characteristics such as profitability ratios, efficiency ratios, asset quality indicators and risk/ risk management ratios. The results clarify that Islamic banks are operationally efficient and profitable because of risks sharing and greater dependency on deposits capital. However, on an overall basis, the ratios indicate conventional banks have higher scores than their counterparts.


I. Introduction
The history of banking in the GCC region dates back to 1918 with the establishment of the first bank in Bahrain. The regional banking sector is unique in nature owing to oil wealth and lending business that focuses on building, real estate, customer loans and heavy external from competition protection. Dubai Islamic Bank was the first standalone Islamic Bank established in 1975 in the UAE. Over the years, several international banks have also established Islamic banking division that upholds traditional Islamic values and offers products and services compliant with Sharia principles. Today, more than 200 Islamic banks operate in 70 countries with $ 2.6 billion in assets under management. In its initial years, Islamic banks witnessed its growth in South Asia and GCC countries. Islamic banks receive funds from deposits instead of shareholders. In contrast to the conventional banks, the principle of risk sharing lead to a better return on equity for Islamic banks. Statistical evidence shows that Islamic banks tend to achieve a higher profit margin compared to conventional banks.
Over the years, the banking system globally has evolved in its offerings to suit the changing consumer demands. One of the primary determinants of this change resulted from the religious beliefs of the people resulting in the phenomenal growth of Islamic Banking System. The predominance of these banks is in countries with significant Muslim population such as Iran, Pakistan, and Sudan but not restricted to them. The difference between the Islamic Bank and the Conventional Bank is generally the framework and principle that governs them. Sharia or Islamic law governs Islamic banking institutions. The holy book of combination of internal and external arrangements and that relies heavily on transparency and disclosure of information that is market relevant.
Al-Ajmi, Hussain, and Al-Saleh (2009) investigated into the intentions of Bahraini customers in choosing a bank, their understanding of bank products and their relative benefits. Considered to be the first study that samples three distinctive set of clients, namely Islamic, conventional and both. Using questionnaire method to 1000 clients and applying Mann-Whitney and Kruskal-Wallis tests and factor analysis the study concluded that Islamic religious belief, social responsibility and cost benefit are the primary determinants in bank selection among Bahraini bank clients. The study also revealed that though the clients of conventional and Islamic banks share common requirements, they do differ in their product understandings. Islamic bank clients are more familiar with sharia compliant products/services. Al-Muharrami and Matthews (2009) evaluated the performance of the GCC banking industry in the pretext of the "Structure-Conduct-Performance" hypothesis using sample data between the period of 1993-2002 and applied panel estimation distinguishing between bank fixed effects and country fixed effects. The paper examined the "Relative-Market-Power and the Efficient-Structure hypotheses" distinguishing between the two by applying a non-parametric measure of technical efficiency, and found that the banking industry in the GCC countries is best explained by the mainstream SCP hypothesis.
Despite minor differences, UAE shares a common economic, cultural and political similarity with other GCC countries with an efficient, stable and profitable collective banking system (Al-Muharrami & Matthews, 2009). The banking sector's contributes to GDP is next to oil and gas sector. The region grew rapidly during the period 2000-2008 primarily from oil exports. The Islamic banking and financial services industry have shown rapid growth over the past ten years, reaching a milestone of over $ 2 trillion in value by 2015. GCC banks have grown strong, well-capitalized and modern banking services have been adopted. Islamic banks accounted for 18% of the financial system in 2007 that accounted for around 36% of global Islamic financial assets. Over the last decade, Islamic Banks has grown at a rate of 20 -30 percent per year, which is three times higher than conventional banks. The Islamic banking system in comparison with traditional banks was less affected by the economic and financial crisis since investments in toxic assets and derivatives are strictly prohibited (Iqbal, 2006). Merchant (2012) studied the performance of Islamic and conventional banks during different time periods including crises and post crises periods. While there were many researches conducted to uphold that Islamic banks have much better viability of generating profits and are considered stable, the research objectives included the measure taken by these banks to curb the ill effects of the crises. The methodology involved sampling 17 banks in the GCC during the period 2008 to 2011 and used CAMEL test factors. The results of the 2-tailed test showed that Islamic banks were able to significantly increase their loan loss reserves and equity to total asset ratio. Islamic banks also had adequate capital structures with lower return on average equity. Islamic banks and conventional banks were comparable in terms of asset quality and liquidity. Hanif, Tariq, and Tahir (2012) analyzed and compared the performance of Islamic and conventional banks in Pakistan on the basis of their internal and external factors. They used nine ratios to ascertain profitability, liquidity and credit risk which were part of the internal factors. The external factors were specific to analyzing the customer preferences and perception while banking with Islamic or conventional banks. The results of the research indicated that conventional banks were better in managing profitability and liquidity while Islamic banks had a head-start in capital maintenance and liquidity risk management. Key findings included the perception of sharia compliance for Islamic banks variety of products for conventional banks which were considered as motivating factors for each type of banks.
Beck, Demirgüç-Kunt and Merrouche (2010) stated that products offered by the conventional banks can be converted to sharia compliant products by making minor changes. The research utilized anecdotal evidences to prove that Islamic banks can be cost effective than conventional banks with insignificant differences in business orientation, efficiency, asset quality and stability. The study also revealed that countries where conventional banks enjoyed higher market share, it was found that Islamic banks had lower level of stability but were cost effective. Additionally, the research highlighted that Islamic banks had performed better in the 2008 financial crises as they enjoyed higher capitalization and liquidity coverage as compared to the conventional banks.
Based on the above literature review, it is clearly evident that in order to judge the banks image, one has to look beyond regular normal ratios like ROA, ROE, etc. The approach has to be more granular that can consider efficiency, profitability, risk and risk management into more detail. Considering these issues, the research is done using 23 ratios covering several areas of the banks financial statement. These ratios may not be touched upon using traditional ratio analysis approach. The main headers would include the profitability ratios, efficiency ratios, asset quality ratios, risk ratios and risk management ratios which would be sub divided into 23 further ratios.
The objective of the study is to ascertain if these ratios significantly impact the dependent variables of the conventional and Islamic banks in the UAE.

Hypotheses of the Study
In order to examine the research objectives, it is imperative to compare the financial ratios of conventional and Islamic banks. For these purposes we have defined the following research hypothesis: Hypotheses 1:There is no significant difference between profitability ratios of conventional banks and Islamic banks in the UAE.
Hypothesis 2: There is no significant difference between efficiency ratios of the conventional banks and Islamic banks in the UAE.
Hypothesis 3: There is no significant difference between asset quality of the conventional banks and Islamic banks in the UAE.
Hypothesis 4: There is no significant difference between Liquidity of the conventional and Islamic banks in the UAE Hypothesis 5: There is no significant difference between liquidity risk management of the conventional and Islamic banks in the UAE.
The alternative hypothesis for all the above would include a significant difference in the financial ratios being tested at each stage.

Data and Methodology
The purpose of the study is to compare the conventional and Islamic banks that have been licensed in the UAE. The study will aid to channelize the future development of both these banking sectors in terms of deposits, profitability and efficiency. The proposed study will analytical in nature and will use secondary data from the annual published financial statements of the conventional and Islamic banks in the UAE. The authors have chosen quantitative analysis using excel to arrive at the conclusion. Data relating to profitability, efficiency, asset quality, liquidity and risk management has been taken from the annual reports since 2014 to 2018. The sample size consists of top 6 conventional banks and Islamic banks, respectively, operating in the UAE. These include:

Independent
The loan-to-deposit ratio (LDR) is used to assess a bank's liquidity by comparing a bank's total loans to its total deposits for the same period.

Independent
The retained earnings to total assets ratio measures the banks' ability to accumulate earnings using its assets

Descriptive Analysis
The study is proposed to include regression analysis to ascertain underlying factors that contribute to the performance of conventional and Islamic banks in the UAE. Based on the formation of hypothesis, five different models have been established. These models will encompass one independent variable and several dependent variables within each hypothesis. This has been described in Table 2 below. Mehta and Bhavani (2017) stated that there are several internal and external variables that are determinants of the profitability of the profitability of the banks in the UAE. Most notable among these were the cost efficiency ratio, adequate capital adequacy ratio and improvement in the asset quality ratio. However, it is also important to keep in mind Return on Assets which could also enhance profitability by diversifying income sources.

Accounting Ratios
Across the UAE, conventional banks have adopted accounting policies based on International Accounting Standards Board (IASB)and IFRS standards (Hussain et al., 2002). In contrast, Islamic banks follow accounting policies established by the Accounting and Auditing Organization for Islamic Financial Institutions (AAOFI). Each Islamic bank establishes a Sharia committee that monitors and examines all bank transactions in compliance with Islamic principles. The ratios under study are categorized based on profit, efficiency, quality of assets and risk. Existing research focused solely on profitability and efficiency. This research uses ratios such as Bank Return on Asset (ROA), Profit Margin (PM), Deposit Return (ROA), Equity Return (ROE), Return on Shareholder Capital (ROSC) and various other ratios for analysis purposes. Based on the previous study, that Islamic banks are more moneymaking than conventional banks. Provision for earning profit (PEA), show how a bank manages its assets. High PEA implies higher reserves for bad loans and unforeseen emergencies and to lower risk. For the dependent variables, the author has chosen ROA, NIM, PEA, CA and EM. The dependent variables correspond to studies made by Pradhan and Shrestha (2016) which stated that management efficiency has a positive strong correlation to the bank performance. While macro-economic factors are important, they do not play a vital role on the impact of the individual performance of the bank. The study further revealed that ROA and NIM could be decisive factors affecting banks performance. Menicucci and Paolucci (2016) further investigated the relation between bank specific factors with profitability in the European Banking Sector to ascertain the role of internal factors achieving higher profitability. His analysis used regression model for 35 banks based on capital ratios and loan loss provision ratios on the efficiency of the banks over the period 2009 to 2013. His study suggested that banks with higher deposits and loan ratios are more profitable but their effect was statistically insignificant to compare efficiency of these banks. Abobaker (2018) study stated that high profitability can be achieved by increasing the bank assets, capital ratio and operating income. Consequently, the profitability is reduced as non-interest income increases over time. Following the same concepts for conventional and Islamic banks as provided in Table 2 above, there is nominal deviation in ROA, NIM and Capital Adequacyof the conventional and Islamic banks in the UAE. The reason could include that banks in UAE apply similar interest rates and have sufficient liquid assets within UAE without considering macro-economic factors. There is significant variation in PEA and the reason could be the shariah requirements which need to be followed by the Islamic banks as these are more conservative in their lending practices. For independent variables, the table reveals range of variation. Most notable variation is seen in ROE, ED, TLSC and RETA. The reason for large difference in the means is due to size and composition of the banks' balance sheet. Also, some banks have been in existence for a longer time than others viz conventional banks in the UAE. While other independent variables indicate insignificant variation, it signifies consistency of the business model between Islamic and conventional banks. It has to be known that UAE is an overbanked market with 22 local national banks, 27 branches of foreign banks and 11 wholesale foreign bank branches as at 30 June 2019. Further, even though there is demand for Islamic products, UAE Islamic banks enjoy a relatively small market share as compared to its conventional counterparts as stated by study conducted by Kapur (2020). The alternative hypothesis will be that there is significant difference between profitability ratios of conventional and Islamic banks in the UAE.  Table 3: 1) There is positive correlation between ROA and ROE for conventional and Islamic banks 2) There is positive correlation between ROA and PM for conventional and Islamic banks. While both show a positive relation, it is significantly stronger in Islamic banks 3) There is positive relation between ROA and ROD for conventional and Islamic banks 4) The correlation is mixed for ROA and NOM for conventional and Islamic banks. It is positive for conventional banks and negative for Islamic banks

for both conventional and Islamic banks which means there is 97% variation in the dependent variables Return on Assets (ROA) is explained by independent variables Return on Equity (ROE), Profit Margin (PM), Return on Deposits (RD) and Net Operating Margin (NOM).
The value of F-stat is 255.01 for conventional banks and 229.45 for Islamic banks with df values (4,25) and is significant as the level of significance is less than 5% or 0.05, hence we can conclude that there is overall significant relationship between the predictor for independent variable Return on Equity (ROE), Profit Margin (PM), Return on Deposits (RD), Net Operating Margin (NOM) as a group and they predict the independent variable ROA significantly. So, we reject the null hypothesis and accept the alternative hypothesis which indicate that significant difference between profitability ratios of conventional and Islamic banks in the UAE.
To assess the significance of each independent variable on the dependent variable ROA, it is established that ROE, PM and ROD have significant impact on ROA for conventional banks as their P value is less than 0.05 and similarly for Islamic banks ROE and ROD have significant impact. The NOM for conventional banks and PM, NOM for Islamic banks do not have any significance on ROA as its P value is greater than 5%. The alternative hypothesis will be that there is significant difference between efficiency ratios of conventional and Islamic banks in the UAE. Correlation is significant at p value of 0.05 level Analysis of Table 5: 1) There is positive correlation between NIM and OEA, OIA, ATO and IEE for conventional banks as well as Islamic banks. The correlation is positively strong for conventional banks as compared to Islamic banks. 2) There is negative correlation for NIM and NNIM for conventional and Islamic banks but it is strong negative in case of conventional banks. The correlation is weak for Islamic banks. 3) NIM is positively correlated with OER for conventional banks and negatively correlated for Islamic banks.
On an overall level there is positive correlation except for Net Non-Interest Margin Ratio for conventional and Islamic banks in the UAE. Referring to Table 6, the adjusted R-Square for conventional banks is marginally higher than Islamic banks indicating higher variation of dependent variable Net Interest Margin (NIM) on independent variables Operating expense to Assets (OEA), Operating income to Assets (OIA), Operating Expenses to Revenue (OER), Asset Turnover (ATO), Interest Income to Expenses (IEE), Net Non-Interest Margin (NNIM). The variation for conventional banks is 99.9% whereas for Islamic banks it is at 95.59%.
The value of F-stat is 11,532.6 for conventional banks and 105.67 for Islamic banks with df values (6,23) and is significant as the level of significance is less than 5% or 0.05, hence we can conclude that there is overall significant relationship between the predictor for independent variable Operating expense to Assets (OEA), Operating income to Assets (OIA), www.cribfb.com/journal/index.php/ijafr International Journal of Accounting & Finance Review Vol. 5, No. 2;2020 Operating Expenses to Revenue (OER), Asset Turnover (ATO), Interest Income to Expenses (IEE), Net Non-Interest Margin (NNIM) as a group and they predict the dependent variable NIM significantly. So, we reject the null hypothesis and accept the alternative hypothesis which indicate that there is significant difference between efficiency ratios of conventional and Islamic banks in the UAE.
To assess the significance of each independent variable on the dependent variable NIM, it is established that OEA, OIA and NNIM have significant impact on NIM for conventional banks as their P value is less than 0.05 and similarly for Islamic banks ATO and IEE have significant impact. The OER, ATO and IEE for conventional banks and OEA, OIA, OER, NNIM for Islamic banks do not have any significance on ROA as its P value is greater than 5%.

Model 3
Hypothesis 3: There is no significant difference between asset quality of the conventional banks and Islamic banks in the UAE.

Dependent Variable:Provision to Earning Assets (PEA) Independent Variable:Loan Ratio (LR), Loans to Deposits (LTD)
The alternative hypothesis will be that there is significant difference between asset quality of conventional and Islamic banks in the UAE.  Table 7: 1) There is weak positive correlation between Provision to earning Asset (PEA) and Loan Ratio (LR) for conventional banks and weak negative correlation for Islamic banks. 2) However, for loan to deposit ratio, both conventional and Islamic banks have a negative correlation between Provision to earning Asset (PEA) and Loan to Deposit (LTD) The reason for mixed variation could be due to the lending restrictions under the sharia principles and the stringent loan criteria. Referring to Table 8, the adjusted R-Square for conventional banks is significantly higher than Islamic banks indicating higher variation of dependent variable Provision to Earning Assets (PEA) on independent variables Loan Ratio (LR), Loans to Deposits (LTD). Infact the adjusted R Square for Islamic banks is negative indicating insignificance of exploratory values. The variation for conventional banks is 5.54% whereas for Islamic banks it is negative 3.68%.
The value of F-stat is 1.85 for conventional banks and 0.485 for Islamic banks with dovalues (2,27) and is significant as the level of significance is less than 5% or 0.05, hence we can conclude that there is no significant relationship between the predictor for independent variable Loan Ratio (LR), Loans to Deposits (LTD) as a group and they do not predict the dependent variable PEA significantly. So, we accept the null hypothesis and reject the alternative hypothesis which indicate that there is no significant difference between asset quality of the conventional banks and Islamic banks in the UAE.
To assess the significance of each independent variable on the dependent variable PEA, it is established that LR and LTDdo not have significant impact on PEA for conventional banks as well as Islamic banks as their P value is greater than 5%. This proves that while the banks differ in profitability and efficiency, the asset quality structure and provisioning requirements have almost been similar since 2014 to 2018. The following model can be created for conventional banks: PEA = 0.0390+ 0.1505 LR-0.1164LTD + Error The following model can be created for Islamic banks: PEA = 0.0274 1 0.0272 LR + 0.0025 LTD + Error

Model 4
Hypothesis 4: There is no significant difference between Liquidity of the conventional and Islamic banks in the UAE Dependent Variable:Cash to Assets (CTA) Independent Variable:Cash to Deposits (CTD) The alternative hypothesis will be that there is significant difference between liquidity of conventional and Islamic banks in the UAE. The Correlation is significant at p value of 0.05 level. Accordingly, liquidity shows a strong positive correlation for conventional and Islamic banks in the UAE. The shows that both have managed to gather sufficient liquidity since 2014 to 2018 to fund the assets from the deposits by simultaneously maintaining liquid funds to meet unforeseen contingencies. Referring to Table 10, the adjusted R-Square for conventional banks is marginally higher than Islamic banks indicating higher variation of dependent variable Cash to Asset (CTA) on independent variable Cash to Deposit (CTD). The variation for conventional banks is 95.8% whereas for Islamic banks it is at 91.51%.
The value of F-stat is 673.2 for conventional banks and 313.737 for Islamic banks with do values (1,28) and is significant as the level of significance is less than 5% or 0.05, hence we can conclude that there is overall significant relationship between the predictor for independent variable Cash To Deposits (CTD) and it predicts the dependent variable Cash To Assets (CTA) significantly.
So, we reject the null hypothesis and accept the alternative hypothesis which indicate that there is significant difference between liquidity of conventional and Islamic banks in the UAE.
To assess the significance of each independent variable on the dependent variable CTA, it is established that CTD has significant impact on CTA for conventional and Islamic banks as their P value is less than 5%.
The following model can be created for conventional banks: CTA = 0.0186+ 0.5928CTD + Error The following model can be created for Islamic banks: CTA = 0.0419 + 0.5745 CTD + Error 3.5 Model 5 Hypothesis 5: There is no significant difference between liquidity risk management of the conventional and Islamic banks in the UAE. The alternative hypothesis will be that there is significant difference between liquidity risk management of conventional and Islamic banks in the UAE.  Table 11: 1) correlation is weak. There is perfect positive correlation between EM and TLE indicating change in Total Liabilities to Equity in same proportion to Equity Multiplier. There is negative weak correlation for RETA and negative strong correlation for ETD. 2) For Islamic banks, the results are very similar to conventional banks, indicating that movement of ratios for risk management is almost identical.
The results prove that conventional and Islamic banks utilize similar strategies for risk management. Referring to Table 12, the adjusted R-Square for conventional banks and Islamic banks similar at 96% indicating related variation of dependent variable equity multiplier (EM) on independent variables Deposits To Assets (DTA), Equity To Deposits (ETD), Total Liabilities to Equity (TLE), Total Liabilities to Shareholder Capital (TLSC), Retained Earnings to Total Assets (RETA).
The value of F-stat is 7,066 for conventional banks and 1,063,4E for Islamic banks with do values (5,25) and is significant as the level of significance is less than 5% or 0.05, hence we can conclude that there is overall significant relationship between the predictor for independent variable Deposits To Assets (DTA), Equity To Deposits (ETD), Total Liabilities to Equity (TLE), Total Liabilities to Shareholder Capital (TLSC), Retained Earnings to Total Assets (RETA) as a group and they predict the dependent variable Equity Multiplier (EM) significantly. So, we reject the null hypothesis and accept the alternative hypothesis which indicate that there is significant difference between liquidity risk management of conventional and Islamic banks in the UAE.
To assess the significance of each independent variable on the dependent variable EM, it is established that DTA, ETD and TLE have significant impact on EM for conventional banks as their P value is less than 0.05 and similarly for Islamic banks DTA, ETD and TLE have significant impact. The TLSC and RETA for conventional banks and Islamic banks do not have any significance on EM as its P value is greater than 5%.

Summary and Conclusion
This paper examines the comparative performance between conventional and Islamic banks in the banks in UAE by using ratio analysis. The study uses several statistical tools to ensure their reliability, including mean testing, correlation and regression analysis. Results from the study indicate financial characteristics of profitability ratios, efficiency ratios, asset quality indicators and risk/ risk management ratios. The results clarify that Islamic banks are operationally efficient and profitable because of risks sharing and greater dependency on deposits capital. Overall, the ratios indicate, the conventional banks to have higher scores than the counterparts. Further, the statistical analysis of the comparative performance showed that the conventional and Islamic banks are significantly different for profitability, efficiency, liquidity, risk and risk management. However, there is no significant difference for asset quality ratios in the conventional and Islamic banks in the UAE. Islamic banks