Determinants of Islamic banking growth: an empirical analysis

Tamsir Cham (Islamic Research and Training Institute, Islamic Development Bank, Jeddah, Saudi Arabia)

International Journal of Islamic and Middle Eastern Finance and Management

ISSN: 1753-8394

Publication date: 16 April 2018

Abstract

Purpose

This paper aims to examine the determinants of growth rate in Islamic banking using annual time series data.

Design/methodology/approach

The author applied several econometrics methods including generalized linear model and survey-based indicators. The author uses the World Bank Enterprise Survey data to supplement the answers.

Findings

The results support the view that high oil prices, stable domestic prices, higher educated populace and greater presence of capital resources have positive effects on growth in Islamic banking. The findings, however, revealed that instability adversely affects Islamic banking growth. The author found no clear conclusion on the impact of economic growth, greater presence of Muslim population and presence of sharia in the legal system of the country on growth in Islamic banking. The major constraints impeding Islamic banking growth include regulations, tax rates and skilled labor force.

Originality/value

There is no empirical work that has been done on the determinants of Islamic banking growth by taking into account the following factors: oil price dynamics, sharia compliant, macroeconomic variables, instability and World Bank Enterprise survey. This paper attempts to search for the push and pull factors of Islamic banking growth to fill the gap in determining the Islamic banking growth.

Keywords

Citation

Cham, T. (2018), "Determinants of Islamic banking growth: an empirical analysis", International Journal of Islamic and Middle Eastern Finance and Management, Vol. 11 No. 1, pp. 18-39. https://doi.org/10.1108/IMEFM-01-2017-0023

Download as .RIS

Publisher

:

Emerald Publishing Limited

Copyright © 2018, Emerald Publishing Limited


1. Introduction

There is growing interest in Islamic banking recently in the midst of a surge in asset size of the industry, its resilience to the recent global financial crisis and the increasing need for non-conventional financing in meeting sustainable development goals. However, the recent volatility in oil prices and the presence of larger share of Islamic banks in oil exporting countries exerted further interest in the study. Hence, the increasing asset size, the uncertainty in oil prices, the presence of sharia in the legislation system and the instability in the Middle East and North Africa (MENA) region served as a motivation to search for the determinants of Islamic banking growth.

Recently, a number of studies have been done on Islamic banking, Islamic finance and economic growth. Imam and Kpodar (2016) investigated whether development of Islamic banking is conductive to economic growth using a sample of 52 countries with data covering the period 1990-2010. Their findings revealed that despite the small size of the Islamic financial system relative to the economy and financial system as a whole, Islamic banking is positively associated with growth after controlling other determinants of growth. They concluded that Islamic countries that suffer from low growth should develop their banking sector through modernizing the legislative, regulatory and infrastructural environment.

Lone and Ahmad (2017) proposed that Islamic banks are intermediary institutions that take care of investors’ expectations to keep the value and return to their investments intact with the market fluctuations. Their findings offered better insight about Islamic finance to further improve the industry and serve the society better. Lone and Rehman (2017) examined customer satisfaction in full-fledged Islamic banks and Islamic banking windows with reference to their service quality using SERVQUAL model. Their findings revealed a better image about full-fledge Islamic banks as perceived by their customers.

Rashid and Jabeen (2016) examine the bank-specific, financial and macroeconomic determinants of performance of Islamic and conventional banks in Pakistan using financial performance index based on CAMELS’ ratios, unbalanced annual panel data and applied GLS regression. They found operating efficiency, reserves and overheads are significant determinants of conventional banks; performance, operating efficiency, deposits and market concentration are significant in explaining performance of Islamic banks. However, gross domestic product (GDP) and lending interest rate on performance are negative for both types of banks. Ramlan and Adnan (2016) study the profitability in Islamic and conventional banks in Malaysia using annual data from 2006 to 2011. They found Islamic banks are more profitable than conventional banks, whereas total loan to total asset for Islamic bank is higher than conventional bank.

Chowdhury et al. (2016) conducted a comparative analysis of the internal and external determinants of the Islamic banks’ profitability in the Gulf Cooperation Council (GCC) region applying dynamic generalized method of moments (GMM), quantile regression and wavelet coherence approaches. According to their study, dynamic GMM tends to indicate that equity financing and operating efficiency and macroeconomic variables such as money supply and inflation are significantly related to Islamic banks’ performance. However, bank-specific variables such as credit risk and equity ratio are not significant at different percentiles.

Imam and Kpodar (2013) investigated the determinants of the pattern of Islamic banking expansion worldwide using country-level data spanning from 1992 to 2006. The findings revealed that income per capita, share of Muslims population and economic integration with the Middle Eastern countries are linked to the development of Islamic banking. While interest rates have a negative impact, the quality of institutions and the September 11 incident are found not to be significant. The findings also revealed that Islamic banks complement conventional banks.

Naceur et al. (2015) examined the relationship between Islamic banking and financial inclusion using two types of indicators of Islamic banking:

  1. the number of Islamic banks operating in the country; and

  2. the size of the assets of these banks.

These indicators can be scaled by one or more of the following: total population, number of adults, total assets of the banking system and total number of banks. They found out that although physical access to financial services has grown more rapidly in the Organization of Islamic Cooperation (OIC) countries, its usage has not increased so quickly. In addition, their regression analysis shows a positive link to credit to households and firms for financing investment, but the empirical findings remain relatively weak. Furthermore, OIC countries exhibit lower financial inclusion, and income per capita is a significant determinant of financial inclusion.

Abedifar et al. (2015) examined the recent empirical literature in Islamic banking and finance. The empirical results suggested no major difference between Islamic and conventional banks in terms of their efficiency, competition and risk features. However, Islamic finance enhances inclusion and financial development. In addition, in the areas of risk and return features of mutual funds, Islamic funds out performed conventional funds.

Gheeraert and Weill (2015) evaluated whether Islamic banking influence macroeconomic efficiency by applying stochastic estimation technics using a sample of 70 countries covering Islamic banks worldwide from 2000-2005. Their findings support the view that Islamic banking enhances macroeconomic efficiency to a certain point. However, beyond a certain point, Islamic banking expansion adversely affects macroeconomic efficiency. Bougatef (2015) investigated the impact of corruption on the soundness of Islamic banks using a panel of 69 Islamic banks over the period 2008-2010. His findings revealed that corruption level aggravates the problems of decreased financing. He further argued that condition of corruption could prevent Islamic banking from being a better effective and meaningful path toward poverty reduction and economic growth.

Zeitum (2012) examined the impact of foreign ownership, bank-specific variables and macroeconomic indicators on the performance of Islamic and conventional banks in the GCC region using annual data from 2002 to 2009. His findings revealed that GDP and inflation significantly impact bank performance; cost to income ratio negatively impacts the performance of both Islamic and conventional banks and foreign ownership does not have an impact on bank performance.

Some studies looked into the determinants of the performance of Islamic banks vs conventional banks. For example, Tamimi (2010) studied the performance of Islamic and conventional banks in the United Arab Emirates using 1996-2008 data series, measuring bank performance using return on asset and return on equity with independent variables such as GDP, liquidity, bank size, financial development indicators and cost. His findings revealed that while liquidity and concentration have an impact on bank performance for conventional banks, Islamic banks’ performance are impacted significantly by operating cost and number of branches.

Hassan and Bashir (2003) looked into the performance of Islamic banks using 1994-2001 data taking into account both internal (overhead cost, liquidity, earnings ratios, etc.) and external factors such as GDP per capita, tax, interest rate and financial indicators, to determine whether these have significant impact on bank performance. In their study, they found that while macroeconomics variables have a positive impact on banks performance, taxes negatively impact banks performance. Similarly, Bougatef (2015) studied the impact of correction on health of Islamic banks using GMM estimation method for 69 Islamic banks with data series ranging from 2008 to 2010. His finding revealed that corruption level significantly impacts financial soundness indicators.

Song and Oosthuizen (2014) examined the results of the survey conducted by the International Monetary Fund to document international experiences and country practices related to legal and prudential frameworks governing Islamic banking activities. They found that progress has been made in creating legal, regulatory and supervisory frameworks in Islamic banking. However, a number of challenges faced by regulatory and supervisory agencies remained.

Gheeraert (2014) examined the empirical impact of Islamic banking on banking sector development. Using IFIRSTdatabase covering period 2000-2005, the found that the development of Islamic banking in Muslim countries leads to a higher banking sector development, which is measured by the ratio of private credit or bank deposits to GDP. The findings also support the view that Islamic banking sector acts as a complement to the conventional banking in Muslim countries.

Beck et al. (2013) examined the Islamic vs conventional banking in the areas of business model, efficiency and stability. Controlling for time-variant country-fixed effects, they found few significant differences in business orientation. In addition, their findings revealed that Islamic banks are less cost-effective but have a higher intermediation ratio, higher asset quality and are better capitalized. They also found that Islamic banks are better capitalized, have higher asset quality and are less likely to disintermediate during crisis. The better stock performance of listed Islamic banks during the recent crisis is also due to their higher capitalization and better asset quality. In another study, Beck et al. (2007) examined reaching out, access to and use of banking services across countries using new indicators of banking sector penetration across 99 countries based on a survey of bank regulatory authorities. They found greater outreach correlated with standard measures of financial development and with economic activity. Similarly, they found better communication and transport infrastructure and better governance associated with greater outreach. In addition, they found greater banking sector outreach minimizes firms’ financial constraints.

King and Levine (1993) found positive relationship between financial development and growth, and those components of economic growth are contemptuously correlated with components of financial growth. According to King and Levine (1993), indicators of the level of financial development, namely, the size of the formal financial intermediary sector relative to GDP, are robustly correlated with economic growth. They also found that the predetermined or predictable components of these financial development indicators are significantly related with subsequent values of the growth indicators.

Bhattacharya and Wolde (2010) examined growth in the MENA region using World Bank Enterprise survey (WBES) data and applied augmented Solow growth model. Their finding revealed that the key constraints impeding economic growth in the MENA region include lack of access to finance, shortage of labor skill and insufficient electricity supply.

An empirical analysis on financial stability of Islamic banks was recently undertaken by Cihak and Hesse (2010) covering 18 banking systems with a substantial presence of Islamic banking. According to their findings, small Islamic banks tend to be financially stronger than small commercial banks; large commercial banks tend to be financially stronger than large Islamic banks; and small Islamic banks tend to be financially stronger than large Islamic banks.

The analysis is interesting for the following reasons:

  • the presence of sharia compliance in the legal system as one of the determinants; and

  • the use of WBES which examines the major impediments confronting businesses.

The justification for having WBES is that like any business entity, any factor(s) that inhibits business affects Islamic banking as well. According to Imam and Kpodar (2016), Islamic banking responds to the specific needs of households and firms. Their findings revealed that for any given financial developments, Islamic banking is found to stimulate growth. The coefficients for all the indicators of Islamic banking that enter the regressions remain statistically significant, confirming the theoretical predication. Third, we incorporate macroeconomic variable such as real GDP and domestic prices, as these factors are essential for sustainable macroeconomic stability. Finally, we consider high oil prices and presence of Muslim population as possible determinants of Islamic banking.

The main goal of this study is to investigate the determinants of Islamic banking growth. The uniqueness of the study consists of its employment of standard growth model and other determinant variables such as the presence of sharia in the legal system, real GDP, inflation and WBES data. To the best of our knowledge, no empirical work has been done to assess the determinants of Islamic banking growth by taking into account the following factors oil price dynamics, sharia compliance, inflation, real GDP, instability and WBES. This paper attempts to search for the push and pull factors of Islamic banking growth to fill the gap in determining the Islamic banking growth.

The rest of the paper is organized as follows: Section 2 reviews the stylized facts of Islamic banking. Section 3 discusses recent developments, trends and performance in Islamic banking. Section 4 specifies the empirical model and determinants and discusses findings on measures of Islamic banking growth. Section 5 concludes and gives policy recommendations.

2. Stylized facts on Islamic banking

In this section, we present some stylized facts on Islamic banking. Islamic banking is a system of banking which abides by Islamic sharia law and its activities are bounded by the principles of Islam. One of the fundamental differences between Islamic banking and conventional banking is the prohibition of riba, which refers to interest. According to Mirakhor and Iqbal, in sharia, riba refers to the premium that the borrower must pay to the lender on top of the principal, and it is sometimes put in simple terms as a “return of money on money” to emphasize the point between interest rate and rate of return on a financial investment. Islam forbids activities such as maysir, gharar and haram which are, respectively, gambling, speculation and venturing into businesses that are considered impermissible such as production or trade in alcohol, pork or any businesses that produce product(s) that are considered illegal in Islam. Zakat, one of the pillars of Islam, is the distribution of the proportion of one’s wealth to the needy.

Mudaraba is a financing arrangement between two or more parties where one partner provides the financing of the project, and the other manages the business. While the financer provides all the necessary capital, profit retained from the investment is distributed between the two in ratios agreed between the two parties. Musharaka is a form of a loan arrangement based on profit and loss sharing. The partners involved contribute capital to the project and share the risks and rewards associated with it. While in Murabaha contract agreement, the bank buys an asset or a commodity from a third party and later resells it to another party at a price which usually includes a mark-up, in Ijara arrangement, the sale of the goods is done at a specific period. In Ijara, rental fees are paid to the buyer of the asset which is usually the bank. Sukuk is one of the most popular financing instruments in Islamic bank. It is to a large extent similar to bonds in conventional banks. While in a conventional bond issuance interest rate is the main determinant, sukuk issuance is backed mainly with asset.

3. Recent developments and trends in Islamic banking

Islamic bank-like conventional banks are profit-making entities although the constraints confronting Islamic banking differ to certain extents with those of conventional banks. Islamic banking has been growing not only in the Muslim countries but also in non-Muslim countries. As of end 2013, there are 147 Islamic banks globally, which constitutes about 28 per cent increase since 2010. The total number of Islamic banks in 2010 stood at 115. On average, total assets, net income and equity have increased by 44.9 and 6 per cent, respectively, from 2010 to 2013 (Figure 1).

It has been argued that increase in number of Islamic banks could be related to the increase in revenue due to a surge in oil prices in the past couple of years. It has been argued that the sector is immune to the recent financial crisis, which hit hard the conventional financial institutions. Furthermore, Islamic banking practice is gaining momentum and is being covered in the legal and regulatory frameworks.

As indicated in Table I, with the exception of Malaysia, the majority of Islamic banking assets is concentrated in the Middle East region including Iran. In countries where Islamic banking exists, Iran, Saudi Arabia and United Arab Emirates have the largest asset size. Suffice it to state that the Middle East dominated the Islamic banking at global level.

There is a large potential for Islamic banking to continue to grow due to large population of Muslims and economic growth rates in these countries. The following countries in Asia and MENA region demonstrate growth in Islamic banking: Bangladesh, Indonesia, Malaysia, Iran, Egypt and Turkey.

In addition, there are quite a number of countries with substantial large Muslim populations without Islamic banking or less presence of Islamic banks. It has been argued that two potential sources that could contribute to the expansion of Islamic banking are entry of new Islamic banking institutions in countries where Islamic banking already exists and introduction of Islamic banking in countries where Islamic banking does not currently exist but the substantial share of the Muslim population can potentially mobilize establishment of Islamic banks. Nigeria and Senegal in West Africa are candidates in good demand for Islamic banking in the very near future[1].

Although data availability is an issue in most of the member countries of the OIC, there is some evidence that Islamic banks are contributing to shared prosperity through the mode of financing they are offering as well as the sectors they are supporting. The distribution of financing by sector (Table II) shows that about 25 per cent of total financing is allocate to the acquisition of consumer durables while real estate and construction, manufacturing and trading point at 16.1, 11.7 and 8.3 per cent respectively. Another key item in this distribution is listed as “Others”, which included government financing through sovereign Sukuk used for liquidity management, one of the biggest challenges facing Islamic banks.

4. Data and model

The data set consists of 19 countries for which data are available. Out of these countries, 14 are from the MENA region. We recognized that some of these countries such as Syria and Yemen are involved in conflict for years, but the instability dummy will take account of this. We also considered countries that not part of the MENA region with greater presence of Islamic banking. These countries are Brunei, Bangladesh, Malaysia and Pakistan. We also use enterprise survey data. The World Bank has accompanied series of Enterprise survey with individual countries’ data ranging from 2007-2010. We were able to access data for the following countries Bangladesh, Egypt, Iraq, Jordan, Malaysia, Pakistan, Syrian and Yemen. While asset and loan data are sourced from Bank Scope, the rest of the macroeconomic data were sourced from World Bank World Development Indicators database. Apart from enterprise survey data, all the series considered are from 2000 to 2013. We acknowledged lack of enterprise survey data for most of the countries under consideration.

We begin our analysis by estimating the standard growth model as a point of departure. We borrowed from the work of Solow, Bhattacharya and Wolde (2010) and Hesse (2008)s and used the augmented Solow growth model as a framework for our analysis. As noted by Hesse (2008) taking into account that the empirical growth literature has been blame for its approach of propelling in all kinds of possible explanatory causes, the explanatory variables in our regression analysis focus on expectations of the augmented slow growth model. According to Hesse (2008), in the Solow growth model, the growth in output per worker, among others is a function of initial output per worker, the saving rate, initial level of technology, rate of technological progress, the rate of depreciation and the growth rate of the workforce. In the model, higher savings will cause a higher growth of output per worker where as an increasing growth rate of the labor force has the opposite effect on growth. In the augmented Solow growth model, a measure for human capital is added as an additional determinant of growth.

Our emphases are on determining Islamic banking growth which largely mimicking growth. We however include additional variables such as inflation, oil prices and instability index, which are not included in the structural determinants of long-term growth. A dummy variable for MENA is also included to gauge whether Islamic banking growth in the region has reach its potential level. Muslim population growth, school enrollment rate and fixed capital formation are considered as inputs just as in the case of standard growth model. Secondary school enrolment rate is a proxy for human capital, one of the variables needed to drive growth in Islamic banking due to ease in promoting awareness of Islamic banking.

We begin with the Solow model given as:

(1) Y=F(K,L)
(2) Y=α+(Kt,AtLt)
where Y is output, K is capital input, L is labor input and A is level of technology, knowledge, efficiency of work and AL effective labor. Using the above model as a framework, our baseline regression model looks as follows:
(3) IBjt=α+β1GDPjt+β2Schooljt+β3GFCFjt+β4Mpopjt+β5Infljt+β6Oiljt+β7Sharijt+β8InStabjt+β9MENA+ejt

The box below provides a summary explanation of the variables used to estimate the determinants of growth in Islamic banks. Subscripts (jt) used in the dependent variables represent country and time, respectively, and ejt is the residual term.

IB

= Growth rate of Islamic bank in terms of assets;

GDP

= GDP growth;

MPop

= Percentage of Muslim population;

School

= Percentage of population with secondary education;

GFCF

= Gross fixed capital formation per cent of GDP;

Infl

= Inflation rate prevailing during the period;

Oil

= Oil prices;

Instab

= Dummy variable taking the value of one in period of conflict/instability and zero otherwise;

Shari

= Dummy variable taking the value of one for countries where sharia is a significant source of the legislation/legal system and zero otherwise;

MENA

= Dummy variable taking the value of one if the country is in the Middle East and North Africa region, zero otherwise;

REG

= Percentage of firms identifying licensing and permits as a major constraint;

FIN

= Percentage of firms identifying access to finance as a major constraint;

LAB

= Percentage of firms identifying skilled labor as a major constraint;

TAX

= Percentage of firm identifying tax rates and administration as a major constraint;

COR

= Percentage of firm identifying corruption as a major constraint; and

INF

= Percentage of firm identifying infrastructure (electricity) as a major constraint.

The variables of interest consist of:

  • GDP: Increase in economic activity and economic growth income will induce the society into more saving and investment hence the need for more banking activities. We therefore expect an increase in Islamic banking growth to correlate positively with economic growth.

  • Muslim population: We expect that where there is more Muslim presence, the tendency to engage in Islamic banking increases. We expect Muslims to shy away from interest-earning activities and may likely be involved in some Islamic banking or advocate for its presence. Consequently, we expect greater presence of Islamic banking activities in the MENA and Southeast Asia region as compared to North America or Europe.

  • School: Human capital accumulation proxied by school enrollment rate is expected to enhance Islamic banking due to the increase in sensitization and financial inclusion. A larger educated population will increase the labor force that increase sensitization of Islamic banking within the community and the country as whole. We expect positive association between Islamic Banking and quality of human capital.

  • Inflation: For a financial system to have a firm footing in any region or country, there is a need for macroeconomic stability. In this exercise, we restrict ourselves to using inflation as one of the macroeconomic variables. As Islamic banking is more of risk sharing, where a country demonstrates price stability, there is relatively lower risk in investment and hence we expect higher presence of financial activities. It also been argued that lower inflation increases real output growth. It suffices to say that price stability increases Islamic banking growth.

  • Oil prices: Capture oil prices, higher oil prices normally translate into higher oil revenue particularly for oil exporting countries, many of which are in the MENA region. The increase in revenue could increase Islamic banking activities.

  • Sharia: Is a dummy, which takes the value of 1 if the significant sources of the legislation apply sharia and zero otherwise[2]. We expect countries that apply sharia will be more sharia compliant and will be more familiar with the concept of Islamic banking.

  • Instability: Is a dummy, which takes the value of 1 if a country is in a period of instability and zero otherwise. Increase in instability in a country could disrupt business activities. Therefore, we expect a decline in Islamic banking in countries, which are relatively unstable[3].

  • MENA: Is a dummy, which takes the value of 1 if a country is part of the MENA region and zero otherwise. We expect MENA dummy to have positive sign, as there is larger presence of Islamic banks in the region.

4.1 Empirical results

In this section, we start our empirical investigation by presenting the descriptive statistics. Table A1 presents mean values, standard deviation, maximum and minimum values of the variables used in the estimations. These statistics provide information regarding the distribution of the variables. Mean value is the measure of the average of the variable over the examined period. Standard deviation shows how much a variable is diversified from the average value. While the minimum indicates the lowest value, the maximum value indicates the highest value in the sample. All the variables have positive mean ranging from 0.10 to 4.13. The mean value of instability index is 0.10, and its standard deviation is 0.30. The maximum and minimum values of the Islamic bank growth are 0.36 and 18.45, respectively. The mean value of loan growth is 13.03, and the standard deviation is 3.11. The maximum and minimum value of Muslim population is 634 and 16, respectively. The value of the standard deviation of 167 indicates that this variable is more volatile as compared to other variables considered in the study.

Table AII presents the correlation matrix of the variables. The correlation matrix does not reveal strong association between the tested determinants. However, there is a negative correlation between the component of the law with greater presence of sharia and gross fixed capital formation per cent of GDP. The correlation between the two variables is −0.47. The absence of strong correlation between the determinant eliminates the possibility of multicollinearity in the regression analysis. However, there is a strong positive correlation between the two measures of Islamic banking growth (asset growth and loan growth). The correlation between the two variables is 0.58. We also assessed the level of correlation between major constraints presented in Table AIII. All the constraints identified and used showed positive correlation. The highest correlation occurred between tax and regulatory constraints and corruption and financial constraints with a correlation value of 0.68 and 0.63, respectively. As we found positive correlation among the constraints variables, it justifies the introduction of few constraints variables one at a time in the regression analysis. Therefore, we introduce pair of constraints that are correlated weakly.

We used augmented Dickey–Fuller unit root test to examine whether the data is stationary (Table AIV). The unit root test revealed that all the series are stationary at level and intercept. Taking into account that the explanatory variables are not correlated, the absent of multicollinearity, the non-linearity between the predictors and the dependents variables, stationary of the series and the estimation involves routines of integer valued, ordered and censored data, we used generalized linear model (GLM) estimation technics to estimate the determinants of growth in Islamic banking. The GLM encompasses a broad and empirically useful range of specifications that includes linear regression, logistic and probit analysis and poison models. Our model exhibits some of these characteristics.

We discuss the regression results of the determinants of Islamic banking growth for the selected countries using 2000-2013 panel data with fixed effect. Our empirical results from estimation of the equation is in Tables III to VII. As expected, oil prices positively influence Islamic banking. The oil price coefficient is positive and statistically significant in both measures used for Islamic banking growth. Our findings are in line with the results of Imam and Kpodar (2016) that oil prices significantly influence Islamic banking. Both secondary school enrolment rate and gross fixed capital formation coefficients have positive signs and are statistically significant. Coefficient for inflation rate is negative in all regression measures and statistically significant. This finding is in line with proponents of Islamic finance in that Islamic mode of financing go toward the real sector and is less inflationary. The results also support the view that price stability enhances growth in Islamic banking Our results is in line with the findings by Imam and Kpodar (2016) that low inflation uncertainty is good for growth. Our empirical results are in line with the work of Zeitum (2012) that low inflation positively impacts Islamic banking.

The coefficient of real GDP in most of the regression analysis is statistically insignificant. Our results do not corroborate the findings by Naceur et al. (2015) that income per capita is a significant determinant of Islamic banking. One possible reason for the different results could be due to structural break because of the 2008 global financial crisis. The coefficients for the presence of Muslim population and larger presence of sharia in the judiciary system of the country showed positive sign but are statistically insignificant. These findings support the view that countries with smaller share of Muslim population can benefit from Islamic banking. Our results is similar to that of Imam and Kpodar (2016).

We also assessed whether instability impacts Islamic banks. Our findings revealed that instability negatively affects growth in Islamic banking. The coefficient is negative and statistically significant in all regressions. When the MENA dummy enters the regression, the coefficient is negative and statistically significant. The interpretation for this is that there is room for Islamic banking to grow further in the MENA region. We examined some of the constraints that might have caused slowed growth in Islamic banking.

4.2 World Bank Enterprise Survey

To examine the determinants of Islamic banking growth more broadly, we supplement our empirical findings with the WBES data. We include WBES as additional explanatory variables for Islamic banking growth. Empirical evidence revealed that improved transport and commination infrastructure enhances banking outreach. Similarly, greater banking outreach minimizes firms constraints (Beck et al., 2007). We contribute to the literature on Islamic banking with the introduction of these additional constraints variables to our regression. These variables could help to explain to some extent the impediments to Islamic banking growth.

As Islamic banking has similar characteristics of any business, we consider the following constraints that are commonly looked into in doing business: corruption, tax rates and administration, access to finance, skilled labor, regulations and infrastructure. The highest constraints considered are regulations, corruption, tax rates and administration and labor. Table AVI presents percentage of firms identified these constraints as major impediments to business.

Adding these constraints[4] to the augmented growth model and applying the GLM, we found regulations (business licensing and permits and customs trade regulations), taxes (tax rates and tax administration), labor, corruption and financial as the major constraints that affect Islamic banking growth. The coefficient for regulations constraints is negative and statistically significant at the 99 per cent level. Reducing the burden on business licensing and permits, customs and trade regulations could increase Islamic banking by 9 per cent. Similarly, labor constraint has a coefficient that is negative and statistically significant at 90 per cent significance level, and reducing this constraint could increase Islamic banking by 2 per cent annually. Similarly, tax rates constraint has a coefficient that is negative and statistically significant at the 95 per cent significance level. Reducing this constraint could increase Islamic banking by 6 per cent annually. Furthermore, corruption constraint has a coefficient that is negatively and statistically significant at 90 per cent level. Eliminating this constraint could increase Islamic banking by 1.6 per cent yearly. From the analysis, a percent improvement in regulations, taxes, educated workforce and reduction in corruption will results in 9, 6, 2 and 1.6 per cent Islamic banking growth, respectively.

To test the robustness of our results, we test for two-way causality between Islamic banking growth and GDP growth (Table AV). The results of the granger causality reject the null hypothesis that real GDP growth does not granger cause Islamic banking growth using loan growth. However, the granger causality failed to reject the null hypothesis that Islamic banking growth (using loan growth as a measure of Islamic banking growth) does not granger cause GDP growth. When we applied asset growth as a measure for Islamic banking growth, the granger causality failed to reject the null hypothesis that Islamic banking growth does not granger cause real GDP growth and GDP does not granger cause Islamic banking growth. We therefore conclude that there is no two-way causality between Islamic banking growth[5] and real GDP growth. Due to the presence of strong correlations in certain constraint-variables (Table AIII), we introduce the constraints in pairs where the correlation between the two is not strong. First, we introduce tax and finance constraints, second, we introduce tax and corruption constraints and finally we add finance and regulation constraints. This does not change the trust of our results. We performed several diagnostic tests to check the validity of our results. Wald tests was performed on the variables in the model. The test indicated that all coefficients are significantly different from zero. We used Ramsey Regression Equation Specification Error Test for omitted variables. The test revealed no evidence of nonlinearity. In addition, the residual looks well behaved, with a mean around zero as shown in Appendix 4. We examine the possibility of endogeneity, as macroeconomic variables are interdependent. The correlation matrix, however, does not reveal strong association between the tested determinants. The constraint variables that show strong association, we introduce these constraints in the regression one at time.

5. Conclusions and policy recommendations

Our empirical results support the view that Islamic banking is influenced positively by high oil prices as indicated in earlier studies (Imam and Kpodar, 2016). Our findings revealed that a more educated work force and greater fixed capita formation enhance Islamic banking growth. On a macro level, our findings revealed that price stability promotes Islamic banking. However, the findings do not give a definitive conclusion that economic growth and presence of greater Muslim population enhance Islamic banking. Although the coefficients for real GDP and greater presence of Muslim population are positive in most of the regressions, they are but not statistically significant. The findings reveal that Islamic banking has the potential to grow in both developed and developing economies. It also reveals that Islamic banking can still grow in countries with few Muslim populations. Our finding indicates that instability inhibits Islamic banking. Therefore, to promote Islamic banking growth further, individual countries need to advocate for stability and peace within the region. Therefore, concerted effort is required to promote regional cooperation and integration. At the macro level, economies need to stabilize their domestic prices to boost Islamic banking, as stable inflation is good for Islamic banking growth.

The introduction of the WBES revealed more interesting results. As far as the doing business constraints are concerned, regulations, taxes and lack of educated work force are the major impediments to growth in Islamic banking. Our results are in line with earlier work by Beck et al. (2007) that improvement in firm constraints enhance banking outreach. Our results to a certain extent tally with the finding of Bougatef (2015) that corruption negatively affects Islamic banking. Our findings offer some insight and policy recommendations to new field of research. From the analysis, three major constraints impeding Islamic banking growth include regulations, tax rates, labor and corruption, which requires attention. Given that most of the economies considered under the study are heavily dependent on oil as a source of revenue, there is a need to revisit the current tax administrative systems to implement tax reforms and make it much more viable. There is a need to revisit the existing regulations regarding business licensing, permit, customs and trade regulations.

These empirical findings call for urgent policy measures in the medium to the long-term. To address skilled labor constraint, there is a need to revamp the education system. Secondary school education should be encouraged. A more educated populace will to a larger extent enhance financial inclusion which will expand banking. Women participation in the labor force needs to be encouraged as this will enhance greater sensitization about Islamic banking growth. Authorities in the MENA region need to take advantage of the large size of its young population to address skilled labor need. To minimize the level of corruption, there is a need to introduce and re-enforce checks and balances and reduce informal processes. This may require greater transparency and accountability in major government institutions. To ensure stability in the banking system and increase access to finance, there is a need to establish credit reference bureaus and increase acceptable collateral securities to make access to credit from banks much easier.

However, major limitations of this paper include inadequate and poor data quality and failure to include other determinants of Islamic banking growth such as the level of financial inclusion, level of conventional banking development and quality of institutions such as the rule of law. Future research in this area can include the implications of Islamic banking products or instruments on macroeconomic variables such as inflation and economic growth.

Figures

Average increased (2010-2013 in per cent)

Figure 1.

Average increased (2010-2013 in per cent)

GLM regressions residuals and fitted values

Figure A1.

GLM regressions residuals and fitted values

Islamic banking assets by regions 2013

Region/Country Total assets in billion US dollars
Middle East including Iran
Iran 94.3
Saudi Arabia 90.4
Kuwait 80.6
UAE 96.7
Bahrain 46.2
Qatar 59.0
South and Southeast Asia
Malaysia 156.7
Indonesia 13.0
Bangladesh 17.0
Pakistan 6.2
Africa
Sudan 6.5
Egypt 5.0
Tunis 0.76
Gambia 0.02
Other
UK 3.3
Turkey 44.8

Source: Bank Scope

Global financing of Islamic banking by sector (2012)

Element (%)
Consumer durables 26.9
Agriculture 0.7
Manufacturing 11.7
Trading 8.3
Transportation 1.7
Real estate and construction 16.1
Banks and financial institutions 14.7
Service 2.7
Others 17.2

Source: Ibisonline.net, Islamic Research and Training Institute (IRTI)

The Islamic bank growth regression results, GLM

Variable Coefficient z-statistics Probability
Real GDP growth 0.001 0.026 0.980
Oil prices 1.744*** 3.908 0.000
Muslim population −0.001 −0.353 0.724
School 0.018 1.544 0.123
Gross fixed capital form 0.327*** 7.027 0.000
Instability −2.300*** −3.396 0.001
Sharia law 0.275 0.446 0.656
Inflation −0.139*** −5.574 0.000
Mean-dependent variation 13.03 SD-dependent var
Log likelihood
Schwarz criterion
Deviance
Pearson SSR
Dispersion
3.11
Sum squared residue 417.80 −232.80
Akaike info criterion 4.30 4.49
Hannan–Quinn criteria 4.38 417.80
Deviance statistics 4.02 417.80
Pearson statistics 4.02 4.02
Number of observation 112
Notes:

The dependent variable is loan growth. The coefficients, z-stat, standard errors and probability values;

***

1% significance level; **5% significance level; and *10% significance level

The Islamic bank growth regression results, GLM

Variable Coefficient z-statistics Probability
Real GDP growth −0.001 −0.027 0.980
Oil prices 1.791*** 4.13 0.000
Muslim population 0.004* 1.786 0.074
School 0.047*** 3.176 0.002
Gross fixed capital form 0.306*** 6.721 0.000
Instability −2.455*** −3.717 0.000
Sharia law 0.621 1.016 0.310
Inflation −0.123*** −4.999 0.000
Tax rates −0.066*** −2.923 0.004
Fin Access −0.036* −1.804 0.071
Mean-dependent variation 13.03 SD-dependent var
Log likelihood
Schwarz criterion
Deviance
Pearson SSR
Dispersion
3.11
Sum squared residue 377.94 −227.26
Akaike info criterion 4.24 4.48
Hannan–Quinn criteria 4.34 377.94
Deviance statistics 3.71 377.94
Pearson statistics 3.71 3.71
Number of observation 112

Notes: The dependent variable is loan growth. The coefficients, z-stat, standard errors and probability values;

***, **

, and

*

denotes significant levels of 1%, 5% and 10% respectively

The Islamic bank growth regression results, GLM

Variable Coefficient z-statistics Probability
Real GDP growth 0.005 0.193 0.847
Oil prices 1.518*** 3.900 0.000
Muslim population 0.002 1.246 0.213
School 0.045*** 4.149 0.000
Gross fixed capital form 0.302*** 7.500 0.000
Instability −0.774 −1.188 0.235
Sharia law 1.017* 1.860 0.063
Inflation −0.125*** −5.708 0.000
Regulations −0.091*** −6.075 0.000
Fin Access 0.0132 0.694 0.487
Mean-dependent variation 13.03 SD-dependent var
Log likelihood
Schwarz criterion
Deviance
Pearson SSR
Dispersion
3.11
Sum squared residue 300.78 −214.48
Akaike info criterion 4.01 4.25
Hannan–Quinn criteria 4.11 300.78
Deviance statistics 2.95 300.78
Pearson statistics 2.95 2.95
Number of observation 112

Notes: The dependent variable is loan growth. The coefficients, z-statistic, standard errors and probability values;

***

, **, and

*

denotes significant levels of 1%, 5% and 10% respectively

The Islamic bank growth regression results, GLM

Variable Coefficient z-statistics Probability
Real GDP growth −0.004 −0.1445 0.885
Oil prices 1.691*** 3.774 0.000
Muslim population 0.003 1.157` 0.247
School 0.043** 2.328 0.02
Gross fixed capital form 0.305*** 6.289 0.000
Instability −2.250*** −3.400 0.001
Sharia law 0.674 1.077 0.281
Inflation −0.133*** −5.402 0.000
Corruption −0.002 −0.108 0.914
Tax rates −0.059* −2.054 0.040
Mean-dependent variation 13.03 SD-dependent var
Log likelihood
Schwarz criterion
Deviance
Pearson SSR
Dispersion
3.11
Sum squared residue 389.95 −229.02
Akaike info criterion 4.27 4.51
Hannan–Quinn criteria 4.36 389.95
Deviance statistics 3.82 389.95
Pearson statistics 3.82 3.82
Number of observation 112

Notes: The dependent variable is loan growth. The coefficients, z-stat, standard errors and probability values;

***

,

**

, and

*

denotes significant levels of 1%, 5% and 10% respectively

The Islamic bank growth regression results, GLM

Variable Coefficient z-statistics Probability
Real GDP growth 0.003 0.112 0.911
Oil prices 1.808*** 4.084 0.000
Muslim population 0.000 0.054 0.957
School 0.032** 2.325 0.020
Gross fixed capital form 0.308*** 6.543 0.000
Instability −1.930*** −2.756 0.001
Sharia law 0.055 0.089 0.930
Inflation −0.134*** −5.367 0.000
Lab skill −0.031* −1.823 0.068
Mean-dependent variation 13.03 SD-dependent var
Log likelihood
Schwarz criterion
Deviance
Pearson SSR
Dispersion
3.11
Sum squared residue 404.74 −231.06
Akaike info criterion 4.29 4.51
Hannan–Quinn criteria 4.38 404.74
Deviance statistics 3.93 404.74
Pearson statistics 3.93 3.93
Number of observation 112

Notes: The dependent variable is loan growth. The coefficients, z-stat, standard errors and probability values;

***

,

**

, and

*

denotes significant levels of 1%, 5% and 10% respectively

Descriptive statistics

Variables Observations Mean Median Maximum Minimum SD
Real GDP 112 4.73 5.17 54.16 −33.10 6.63
Muslim population 112 413.83 449.81 634.57 16.48 167.56
Gross fixed capital formation per cent of GDP 112 19.68 20.30 30.63 2.92 5.24
Population with Secondary education 112 59.38 50.00 93.89 21.78 20.50
Inflation 112 8.03 5.95 53.23 −10.07 8.34
Oil price 112 4.01 4.14 4.65 3.19 0.54
Sharia law 112 0.38 0.00 1.00 0.00 0.49
Instability 112 0.10 0.00 1.00 0.00 0.30
Asset growth 111 13.98 14.15 18.45 4.71 2.04
Loan growth 112 13.03 13.32 18.45 0.36 3.11

Correlation among Islamic banking growth and its determinants

Variables Real GDP Muslim population Gross fixed capital formation per cent of GDP Population with Secondary education Inflation Oil price Sharia law Instability Asset growth Loan growth
Real GDP 1.00 −0.02 0.05 0.06 −0.03 0.04 0.02 0.00 0.04 0.05
Muslim population −0.02 1.00 −0.19 −0.20 0.30 0.35 −0.39 0.17 −0.31 −0.25
Gross fixed capital formation per cent of GDP 0.05 −0.19 1.00 0.43 −0.25 0.18 −0.47 0.20 0.48 0.65
Population with secondary education 0.06 −0.20 0.43 1.00 −0.23 0.08 −0.47 0.25 0.11 0.37
Inflation −0.03 0.30 −0.25 −0.23 1.00 0.18 0.07 0.06 −0.31 −0.51
Oil price 0.04 0.35 0.18 0.08 0.18 1.00 −0.01 0.29 0.22 0.20
Sharia law 0.02 −0.39 −0.47 −0.47 0.07 −0.01 1.00 −0.26 0.01 −0.23
Instability 0.00 0.17 0.20 0.25 0.06 0.29 −0.26 1.00 0.03 −0.04
Asset growth 0.04 −0.31 0.48 0.11 −0.31 0.22 0.01 0.03 1.00 0.58
Loan growth 0.05 −0.25 0.65 0.37 −0.51 0.20 −0.23 −0.04 0.58 1.00

Correlation among Islamic banking growth and major constraints

Variables Asset growth Regulatory constraint Corruption constraint Financial constraint Labor constraint Tax constraint
Loan growth 0.58 −0.56 −0.58 −0.40 −0.15 −0.40
Asset growth 1.00 −0.22 −0.45 −0.25 −0.22 −0.44
Regulatory constraint −0.22 1.00 0.42 0.29 0.48 0.68
Corruption constraint −0.45 0.42 1.00 0.63 0.21 0.37
Financial constraint −0.25 0.29 0.63 1.00 0.38 0.01
Labor constraint −0.22 0.48 0.21 0.38 1.00 0.40
Tax constraint −0.44 0.68 0.37 0.01 0.40 1.00

Augmented dicky-fuller unit root resta,b,c

Variables Level First difference Second difference
Real GDP growth 0.000 Not Needed Not needed
Muslim population 0.102 0.000 Not needed
Gross fixed capital formation 0.128 0.000 Not needed
School 0.142 0.000 Not needed
Inflation 0.000 Not needed Not needed
Oil prices 0.000 Not needed Not needed
Asset growth 0.007 0.000 Not needed
Loan growth 0.034 0.000 Not needed

Notes:

a

All variables are in natural logarithm;

b

p-values are reported for null hypothesis: H0: series have unit root;

c

All tests include intercept and number of lags is based on Schwartz Information Criterion

Pairwise granger causality tests

Null hypothesis Observation F-statistics Probability
Asset growth does not granger cause real GDP growth 110 0.16171 0.8509
Real GDP growth does not granger cause asset growth 0.36639 0.6941
Loan growth does not granger cause real GDP growth 110 0.26058 0.7711
Real GDP growth does not granger cause loan growth 2.98352 0.0549

Descriptive statistics

Country and % of firms identifying as a major constraint
Country/Constraint Bangladesh Egypt Iraq Jordan Malaysia Pakistan Syria Yemen MENA (Average)
Inadequate education workforce 25 50 34 33 20 8 60 24 40
Finance 43 31 46 25 15 18 34 35 34
Infrastructure 6 15 40 12 11 14 18 14 20
Electricity 78 14 54 24 16 75 58 56 41
Tax rates 20 46 40 53 21 40 43 43 45
Tax administration 30 29 30 36 17 23 51 42 37
Business license and permits 9 14 40 42 10 17 42 25 32
Custom and trade 11 23 23 14 15 0 6 14 18
Corruption 55 45 62 41 16 59 67 68 57

Source: World Bank Business Enterprise Survey and Author’s calculations

Notes

1.

Other good demand candidate countries are Morocco, Mali, Algeria, Cameroon, Libya, Mozambique, Kazakhstan, Uzbekistan, Afghanistan and Azerbaijan.

2.

We take into consideration countries where sharia is a significant source of legislation and is applied such as Saudi Arabia, Sudan, Iran, Iraq, Afghanistan, Pakistan, Brunei, United Arab Emirates, Qatar, Yemen and Mauritania.

3.

We recognized that some of these countries such as Syria and Yemen are involved in conflict for years, but the instability dummy will take account of this.

4.

The constraints we covered include the following regulations: business licensing and permits and customs and trade regulations, tax and tax administration, inadequate educated workforce, corruption and financial.

5.

When asset growth is used as a measure of Islamic banking growth.

Appendix 2

Table AIV

Table AV

Appendix 3

Table AVI

Appendix 4

Figure A1

References

Abedifar, P., Ebrahim, S.M., Molyneux, P. and Tarazi, A. (2015), “Islamic banking and finance: recent empirical literature and directions for future research”, Journal of Economic Surveys, Vol. 29 No. 4, pp. 637-670.

Beck, T., Demirguc-Kunt, A. and Merrouche, O. (2013), “Islamic vs. conventional banking: business model, efficiency and stability”, Journal of Banking and Finance, Vol. 37 No. 2, pp. 433-447.

Beck, T., Demirguc-Kunt, A. and Peria, M.S.M. (2007), “Reaching out: access to and use of banking services across countries”, Journal of Financial Economics, Vol. 85 No. 1, pp. 234-266.

Bhattacharya, R. and Wolde, H. (2010), “Constraints on growth in the MENA region”, IMF Working Paper 2010.

Bougatef, K. (2015), “The impact of corruption on the soundness of Islamic banks”, Borsa Istanbul Review, Vol. 15 No. 4, pp. 283-295.

Cihak, M. and Hesse, H. (2010), “Islamic banks and financial stability: an empirical analysis”, Journal of Financial Services Research, Vol. 38 No. 2, pp. 95-113.

Chowdhury, M.A.F., Haque, M. and Masih, M. (2016), “Re-examining the determinants of islamic bank performance: new evidence from dynamic GMM, quantile egression, and wavelet coherence approaches”, Emerging Markets Finance and Trade, Vol. 53 No. 7, pp. 1519-1534.

Gheeraert, L. (2014), “Does islamic finance spurs banking sector development?”, Journal of Economic Behavior and Organization, Vol. 103, pp. S4-S20.

Gheeraert, L. and Weill, L. (2015), “Does islamic banking development favors macroeconomic efficiency: evidence on the islamic finance-growth nexus”, Economic Modelling, Vol. 47, pp. 32-39.

Hassan, M.K. and Bashir, A.M. (2003), “Determinants of Islamic banking profitability”, paper presented at the Proceedings of the Economic Research Forum (ERF) 10th Annual Conference, Marrakesh-Morocco, 16-18 December.

Hesse, H. (2008), “Export diversification and economic growth”, The Commission on Growth and Development Working Paper Series No. 21.

Imam, P. and Kpodar, K. (2013), “Islamic banking: how has it expanded?”, Emerging Markets Finance and Trade, Vol. 49 No. 6, pp. 112-137.

Imam, P. and Kpodar, K. (2016), “Islamic banking: good for growth?”, Economic Modelling, Vol. 59, pp. 387-401.

King, R.G. and Levine, R. (1993), “Finance and growth: schumpeter might be right”, World Bank Financial Policy Systems Working Paper Series No. 1083.

Lone, F.A. and Ahmad, S. (2017), “Islamic finance: more expectations and less disappointment”, Investment Management and Financial Innovations, Vol. 14 No. 1, pp. 134-141.

Lone, F.A. and Rehman, A.U. (2017), “Customer satisfaction in full-fledged islamic banks and Islamic banking windows: a comparative study”, Journal of Internet Banking and Commerce, Vol. 22 No. S7.

Naceur, S.B., Barajas, A. and Massara, A. (2015), “Can Islamic banking increase financial inclusion?”, IMF Working Paper 2015.

Ramlan, H. and Adnan, M.S. (2016), “The profitability of Islamic and conventional bank: case study in Malaysia”, Procedia Economics and Finance, Vol. 35, pp. 359-367.

Rashid, A. and Jabeen, S. (2016), “Analyzing performance determinants: conventional versus islamic banks in Pakistan”, Borsa Istanbul Review, Vol. 16 No. 2, pp. 92-107.

Song, I. and Oosthuizen, C. (2014), “Islamic banking regulation and supervision: survey results and challenges”, IMF Working Paper 2014.

Tamimi, H.A.H. (2010), “Factors influencing performance of UEA, Islamic and conventional banks”, Global Journal of Business Research, Vol. 4 No. 2, pp. 1-9.

Zeitum, R. (2012), “Determinants of islamic and conventional banks performance in GCC countries using panel data analysis”, Global Economy and Finance Journal, Vol. 5 No. 1, pp. 53-72.

Further reading

DCIBF Annual Report (2014), “Islamic banking growth, efficiency and stability”, Dubai Center for Islamic Banking and Finance Annual Report 2014.

Gelbard, E., Hussain, M., Maino, R., Mu, Y. and Yehoue, E.B. (2014), “Islamic finance in Sub-Saharan Africa: status and prospects”, IMF Working Paper 2014.

Hassan, M.K. and Lewis, M.K. (2007), Handbook of Islamic Banking, Edward Elgar, Cheltenham, Northampton, MA.

Islamic Research and Training Institute (2017), Ibisonline.net, IRTI, ISDB.

World Bank (2015a), World Development Indicators (WDI) 2015, World Bank.

World Bank (2015b), World Bank Enterprise Survey 2015, World Bank.

Corresponding author

Tamsir Cham can be contacted at: tamsirc@hotmail.com