The Determinants of Bank Profitability: How Malaysian Islamic Banks Response to the Financing Risk

The purpose of this article is to investigate the relationship between financing risk and profitability of Malaysian Islamic banks to find the impact of financing risk on bank profitabily in achieving high level of stability. A regression analysis is built on an unbalanced panel data set comprising 175 observations of 17 top Malaysian Islamic banks over the period 1994-2015. To this end, the empirical data are collected from individual Islamic banks’ financial statements as submitted to the Central Bank of Malaysia for supervisory purposes. Complementary sources include the Global Market Information Database, the World Bank's Annual Report through the International Monetary Fund (IMF), Statistical, Economic and Social Research and Training Center for Islamic Countries (SESRTCIC) and Asian Development Bank (ADB). All the determinant variables included in the model have statistically significant impacts on Malaysian Islamic banks’ profitability. However, the effects are not uniform across profitability measures. Regression findings reveal that capital ratio, financing loss provision and liquidity ratio significantly affect the profitability of Islamic banks. Some variables of monetary policy variables such as interest rate and interbank rate have a significant role on the Islamic banks’profitability. Interestingly, there is a difference of the two variables where interest rate shown negative sign while positive sign from interbank rate. Findings also suggest the economic conditions factors such as GDP growth and unemployment also affects the profitability of Islamic banks despite the sign is different.


INTRODUCTION
In general, banking experts characterize banks as profit-oriented institutions. Profitability doesn't only contribute to the success of the banking sector (Bobáková, 2003), but also serves as an indicator for efficiency of bank management) as well as acting as a forecast for the financial crisis (Fraser & Kalor, 2001. The importance of profitability can be seen from the micro and macro aspects (Aslı & Detragiache, 1999). From a micro point of view, profits are the key indicator of competitiveness and fundraising at the lowest cost. While from the macro aspect, the lucrative banking sector provides a buffer against negative shocks and acts as insurance against any economic shock and financial instability (Aburime, 2009).
Islamic banks also put profitability as a major target even though it also strives to fulfil the Shariah rules and goals. Islamic banks are business institution where profits generated are the major determinants of long-term sustainability of institutions as well as allowing depositors, bankers and shareholders to enjoy returns from their invested funds (Hassoune, 2008). In addition, profitability becomes important due to the function of Islamic banks that not only serves as an intermediary, but also as a monetary policy transmission channels. If something undesirable happens to this industry, the country's economy will also be affected by its huge and significant impacts on the smooth of economic system. However, to achieve the profitability, Islamic banks as well as conventional banks whereas financing activities are the main sources of revenue and affected bank performance positively (Fekri, 2015). However, in carrying out these activities, Islamic bank is exposed to various form of risks related to the financial especially credit risk. An increase in this type of risk is often driven by financing activities to generate high profits for the bank. Nonetheless, if such activities are made more than the absence of a sound and prudent financing policy, it exposes Islamic banks to an increase in troubled financing that ultimately affects overall financial stability following the chain effects borne out of this form of risk.
The rest of this article is organized as follows. Section 2 reviews the literature. Section 3 describes the methodology consist model spesifications, empirical variables, methodology and data. Result are reported in section 4 which is consist descriptive analysis, model slection test, independent variable matrix analysis abd empirical result. Finally, section 5 will concludes the paper.

II.
Literature review This section provides an overview of the studies related to the determinants of bank profitability. This issue has received considerable attention in academic literature and has been widely investigated theoretically and empirically. There have been several studies about the effects of firm characteristics on profitability, and following early works edited by Short (1979) and Bourke (1989), a few more recent studies have attempted to identify some of the major determinants of bank profitability in many countries. Some studies are country-specific, while few of them consider panel ofcountries.
Sudin Haron (1996) examines whether the impact of competition and other factors affecting the conventional banking profits have also affected the profitability of Islamic banking. The study adopted the least-squares estimation method (OLS) on a combination ISSN: 2349-7300

IJIRMPS1811003
Website : www.ijirmps.org Email : editor@ijirmps.org 20 of time series data and cross-section for fourteen banks representing several countries for the period 1982 to 1994. The study found that the rate of inflation and bank size had a significant positive impact on the profitability of Islamic banking. Whereas in terms of market share and money supply, these two variables have the opposite impact. Bashir (1999) reviews the impact of bank scales (total assets) on the performance of Sudanese Islamic banks. Using the data of two banks in Sudan, empirical studies support the theory that the relationship between size and profitability of Islamic banking is significant and this shows that Islamic banking sample is more profitable. However, the negative relationship between the size and the equity ratio on bank capital shows a large-sized bank systematically stacked. Additionally, the negative and significant relationship between the size of Islamic banks and the risk index indicates that large banks are efficient. Conversely, a significant negative relationship between bank size and market value is contrary to the theory's expectations.  at the following year investigated the factors that influenced the profitability of Islamic banking in the Middle East countries for the period 1993-1998. This study applies regression analysis methods of 14 banks in eight countries involving internal and external internal variables. 1 . The findings show that leverage ratio and high asset lending ratio and economic conditions contribute to the level of profitability of Islamic banking in the Middle East. Additionally, factors such as implicit and explicit taxes affect the profitability of Islamic banking as the assumptions of other factors remain unchanged.
Samad & Kabir (2000) assess Bank Islam Malaysia Berhad's (BIMB) time and bank performance in terms of luck, liquidity, risk, solvency and community involvement. The study used a financial ratio to measure BIMB's performance for the period 1984-1997 and found BIMB's return on assets (ROA) and equity returns (ROE) had significant progress throughout the review period. From the liquidity aspect, BIMB was more liquid compared to the other 8 conventional banks where DER BIMB showed a value of 0.021 compared to conventional banks with a value of only 0.012. In terms of risk measurement and solvency, the study shows that BIMB's risk has increased significantly in DER (debt-equity) and EM (equity multiplier). The study also shows that BIMB is still less risky and has higher solvency than 8 other conventional banks.
Pasiouras & Kosmidou (2007) examine the factors affecting the profits of domestic and foreign commercial banks in the European Union (EU) countries. Using the accumulated bank data set, this study looks at how the bank's specification and conditions affect the profits of domestic and foreign commercial banks operating in fifteen EU countries during the period 1995-2001. Finally, studies have found that the profits of domestic and foreign banks in union member countries are not only affected by bank specification factors, but also the financial market structure and macroeconomic conditions. Kosmidou (2008) individually reviews the performance of Greek bank institutions throughout the financial integration period by using the unbalanced time series data set for twenty-three commercial banks in Greek. The result of the study found that banks with good capitalization and low cost to income ratio were associated with high return on assets (ROA). While bank size is always positive but only significant when the macroeconomic variables and financial structures are included in the model. Gross Domestic Product (GDP) variable is positively and significantly influenced by the return of assets of sample banks, while the inflation variables are the opposite.
Hassoune (2008) reviews the profitability of Islamic banking in the Gulf countries in the interest rate cycle over the 10-year period from 1992 to 2001. The study adopts a estimation method that includes some basic variables and divides the study into three main divisions namely theoretical model, Measurement of theoretical model and empirical evidence. Finally, the study found that Islamic banks in the Gulf countries was more profitable than conventional banking in terms of balance sheet structure. However, studies found that gains earned by Islamic banking were not the ultimate solution to the problems faced as Islamic banking was burdened with weaknesses in liquidity structures, risk management and operational management efficiency.

IJIRMPS1811003
Website : www.ijirmps.org Email : editor@ijirmps.org 21 Albertazzi and Gambacorta (2009) analyse the relationship between business cycle changes and banking sector profits, and how these relationships are influenced by structural and institutional features. The study uses the estimation method of the set of equation of net interest income, non-interest income, operating cost, allowance and profit before tax for banks in major industrial countries and subsequently assessing the impact of macroeconomic and financial shock on bank's profitability. The findings show that banks' profits are pro-cycles where GDP affects net interest income (through lending activity) and loan loss provision (LLP).
Finally, Sufian and Habibullah (2010) analyse the impact of economic freedom on the performance of the banking sector in Malaysia. In this study, the performance of the Malaysian banking sector is measured through bank profit variables such as return on assets (ROA) and return on equity (ROE) using panel data from 1999 to 2007. The researches put banks specification, macroeconomic variables and also economic freedom variables as independent variables. The findings show that low levels of economic freedom and business in doing business have a negative impact on bank profitability. Empirical finding also found that corruption had a profound effect on the profitability of the Malaysian banking sector.

Model Specification
To investigate the impact of financing risk on Malaysian Islamic banks' profitability, the cross-section and time series data have been examined by applying a panel data multiple regression. In this paper, we follow the framework suggested by Bashir (1999), Donsyah Yudistira (2004), Anouar Haasoune (2008), Ramlall (2009). The specific model we use in this paper is specified in equation (1): The baseline model is specified as follows: where is the Islamic banks profitability, B is bank-specific and financing risk determinants, M is monetary policy determinants and I is macroeconomic determinants, is the residual term and i and t refer to bank and time respectively.

Empirical Variables
The dependent variable used in this research is the Islamic banks profitability. The explanatory variables are divided into bankspecific determinants, financing risk, monetary policy and macroeconomic determinants. The explanatory variables are selected as suggested in the literature. The descriptions of the variables, the data sources as well as the expected signs are presented in Table I;  (Brooks , 2008). t=1, 2……., T of each i=1, 2,…., N cross-section observations in the sample. The panel data model fits this study because it can analyse changes at the bank level which cannot be done in either cross-section or time series models. Furthermore, using this technique can also reduce the multicolinearity problem and provide a larger degree of freedom. These two can be achieved because the technique increase the number of data points (Nerlove, 2002).

IV. Data
The study makes use of secondary data collected from various sources. The principal data source is the individual Islamic banks' financial statements as submitted to the Central Bank of Malaysia for supervisory purposes.  Table II. V. Table II provides the detailed summary of Islamic banks profitability and all its determinants, endogenous and exogenous, used in the empirical analysis. This Table shows   To identify whether the data in the sample normal or not, this study adapts general least square (GLS) methodology and provides better estimation results. This is because the data used is not normally distributed because there is no statistical characteristics required. Among the features is the mean value and the median for all variables are not the same while the skewness value does not equal to zero. In the meantime, the kurtosis value likewise does not equivalent the value of three. Jarque-Bera values indicate that only liquidity ratio is insignificant, while the rest of the data is significant at the one percent significance level. This shows the hypothesis that the data used normal scattered is rejected.

Model Selection Test
Since the data used in this study are based on panel data, it is worthwhile choosing the appropriate regression test to be used as an analytical tool and to interpret the data based on. Initially, pooled OLS is applied and based on the results, a further test is conducted to decide on the suitability of pooled OLS against random effects. The results of random effects test is presented in Table III. A comparison test between the pooled OLS and random effects result is conducted using the Lagrange Muliplier (LM) test (Breusch, T.S. and Pagan, A.R., 1980). The LM test concludes that random efftects is more appropriate than pooled OLS. Therefore, the findings will be interpreted based on random effects [5]. The model is run with Rogers (1993) standard errors correction for both heteroscedasticity and autocorrelation. As the results of LM test shows that random effects are more appropriate than Pooled OLS in early estimation shown in Table III, further tests are conducted to choose the appropriate regression based on random or fixed effects (feasible GLS) whichever is appropriate. The Hausman Specification test (Hausman 1978), which compares a random effect model to its fixed counterpart, is undertaken to specify the appropriatness of each method. The Hausman Specification test suggests that fixed effects outperformed random effects. Therefore, the results of this section are presented and explained based on fixed effects.

Independent Variable Matrix Analysis
In this article, a collinearity test was conducted using the Pearson correlation matrix as shown in Table IV. The results confirm that no collinearity problem occurs between the independent variables, as multicollinearity can be considered a problem when the correlation is above 0.80 (Kennedy, 1992). In this regard, the correlation between each of the variables is not elevated and the highest degree of correlation found is very satisfactory. The matrix shows that multicollinearity problems do not exist, confirming that the model used is valid and reliable. The empirical analysis shows that all the independent variables have a statistically significant relationship with Islamic banks profitability measures included in the model. The regression results on the relationship between bank profitability and the explanatory variables are displayed in Table IV.   Table V describes the empirical results of Islamic banks profitability with bank-specific, monetary policy and macroeconomci variables for 17 Malaysian Islamic banks. As shown in the baseline models in Table IV, the level of financing risk is found to be negative and significant at the conventional levels in all the estimations. This result proves that high financing risk is associated with low efficiency, as more bad financings will reduce the profitability of the banks. These results also indicate the excessive credit granting activities has negatively impact the profitability of Islamic banks. It happened due to the excessive financing and at the same time placing various categories of problematic financing and further affecting bank revenue. According to Miller and Noulas (1996), a decline in the quality of the financing portfolio would lead to lower profitability of bank. This finding is consistent with Miller and Noulas (1997), Donsyah (2004) and Ramlall (2009) who found that banks vulnerable to risky loans typically collect more non-performing loans and bad debts which further affect the profitability of banks (Misman, 2011

Bank Spesification
For bank specification aspect, the results of the estimation show that Capital ratio, financing loss provision and Liquidity ratio significantly affect the profitability of Islamic banks. Capital ratio ratio shows a positive and statistically significant influence on bank profitability at the 0.0398 percent level. The findings are consistent with Demirguc-Kunt and Huizinga (1999) and Pasiouras & Kosmidou (2007) who provide guidance on raising the capital made by Islamic banks through core capital (tier I) and support capital (tier II) increasing the ability of banks to diversify profit generation activities. In addition, an increase in capital also reduces the cost of financial pressures (Mathuva, 2011) apart from reducing the dependence on Islamic banking on sources of external funding.
Financing loss provision has consistently exerted a positive and significant impact on the Islamic banks profitability with a value of -6.06E-08 percent. This result shows that Islamic banks have a high level of problematic financing and banks are forced to increase this provision to deal with the problem. Thus, bank's capability to generate profits affected where the funds that were supposed to be used for other financial activities had to be channeled for lubricating bank revenue. On the other hand, if Islamic banks managed to reduce this provision, their capability to increase profit margins will increase and this is supported by the discovery of Miller and Noulas (1997) who also find such a decision (Sufian & Habibullah, 2010).
The impact of liquidity ratio on Islamic banks profitability is positifve and statistically significant in all estimations with coefficient of 0.0139 percent. The findings indicate that the available liquidity may also be channeled into other financial instruments and activities for generating higher profits. The findings are consistent with Suffian & Habibullah (2010), Pasiouras and Kosmidou (2007) and Molyneux and Thornton and Eichengreen and Heather who found that a decision in which there is a significant relationship between liquidity holdings and bank profitability (Molyneux & Thornton, 1992).

Monetary Policy
The results in Table III  The results also explain that variables of Islamic Interbank rates show significant relationship with the profitability of Islamic banks. These results are consistent with Demirgüc Kunt and Huizinga's findings (Asli & Huizinga, 1999) and shows that returns from Islamic short-term investments can contribute to bank's profitability as well as provide access to instant liquidity requirements. The findings also pointed out that the development of the Islamic money market by the government had been successful in promoting Malaysian Islamic banks to provide their own funding and practice indirectly stimulating financing growth which in turn encouraged increased institutional profit (Fathin & Ghafar, 2007).

Economic Conditons
In relation to the impact of economic conditions on the Islamic banks profitability in Malaysia, the result in Table VI show that GDP influenced the efficiency positively at 1 per cent level of significance only when it is included in the model with micro variables. This is supported by Sufian (2009) and Johnes et al. (2014) on the grounds that the healthier the economic situation, the better the performance of the banks. This is so because the demand for banking services increases as the economy grows, reflecting the increased wealth. A lower demand for financial services, accompanied by high probability of loans default, would lead to a decrease in national output.
The economic conditions also affects the profitability of Islamic banks. It can be proven through GDP Growth, and unemployment which are significantly affect the profitability of banks despite the sign of different coefficients. The GDP Growth for example, is significant. This illustrates that good economic conditions will encourage household and firm demand for the various financial  Calza et al. (2003) and  on the grounds that the healthier the economic situation, the better the performance of the banks. Good economic condition also affecting banking business by increasing levels of savings, financing and investments that directly affected bank's profitability   5 . Hence, in this favorable economic conditions, Islamic banks should make every effort to exploit opportunities to generating higher profits as a preparation for bank in facing instability and shocks for the next financial years  Furthermore, unemployemnt variable shows a negative and significant coefficient with the profitability of Islamic banking with value of -1.52E-07 per cent. This finding shows that Islamic banking practice "selective financing policies" where funding is focused solely on customers with a stable and sound career background to secure a good repayment record. Such financing policy however, if accompanied by an increase in unemployment will reduce public demand for financing. This situation slightly affected the profitability of Islamic banking as long as the government did not try to overcome it drastically. Similar findings were also obtained by Taqiuddin, Joni and Azam (2008) where they found that contraction of this funding driven by bank concerns over the ability of customers to repay financing on unstable and strong background backgrounds.

VI. Summary and recommendations
The study investigates the effect of financing risk on the profitability of Islamic banks in Malaysia, based on a survey of 17 fullfledged Islamic banks in Malaysia over the period 1994-2015. The main findings of the study are as follows:


The high level of financing risk is found to be negative impact on the profitability of Islamic banks.  Government policies related to the monetary such as interest rate and Islamic Interbank rates have a significant impact on Islamic banks profitability.  Economic conditions of Malaysia also play an important role to affect the profitability of Islamic banks in Malaysia. The good and steble economic conditions is expected to positively affect the efficiency of the bank. Economic measurement through Gross Domestic Product (GDP) may influence the demand for financial services which is affected by GDP growth which would influence the efficiency of bank.
The study therefore recommends that Islamic banks in Malaysia needs to operate under clear financing procedures, in line with the "financing policies" and take into account the different capital considerations according to the risk categories faced. The established financing procedures should outline legitimate financing limits that can impede the problem of credit creation arbitrarily or based on speculation that exposes banks to unnecessary risks. This financing procedure requires Islamic banks to determine in advance the appropriate candidates for financing by identifying the goals of financing, how and where funds will be used and the ability to repay financing. The move is to ensure that funding is always conducive and supportive of economic activities besides avoiding problems in Islamic banking in the future.