Bulletin of Monetary Economics and Banking, Vol. 22, No. 4 (2019), pp. 529 - 550
FINANCIAL INTERMEDIATION COSTS IN A DUAL BANKING
SYSTEM: THE ROLE OF ISLAMIC BANKING
Mansor H. Ibrahim*, Siong Hook Law**
*International Centre for Education in Islamic Finance, Lorong University, Kuala Lumpur, Malaysia.
Email: mansorhi@inceif.org
**Department of Economics, Universiti Putra, Malaysia.
ABSTRACT
This paper empirically analyses the role of Islamic banking in financial intermediation costs as measured by net interest margins for a leading dual banking country, Malaysia. Controlling for theoretically motivated determinants of the margins, the paper compares the interest/financing margins of conventional and Islamic banks and examines the impacts of Islamic banking presence on bank margins. The analysis provides evidence of the higher margins of Islamic banks compared to those of conventional banks. Further, the difference in bank margins between the two types of banks can be attributed to differences in market power, operating costs, and diversification. Finally, Islamic banking presence or penetration, as represented by the ratio of Islamic financing to aggregate bank credit/financing and, alternatively, the share of Islamic banking assets, is robustly associated with lower bank margins, on average. These results bear important implications for the development of the Islamic banking industry and in fostering the efficient allocation of financial resources by the banking system.
Keywords: Interest margins; Dual banking system; Malaysia.
JEL Classifications: C33; G21; G32.
Article history: |
|
|
Received |
: September 02, 2019 |
|
Revised |
: November 18, 2019 |
|
Accepted |
: December 15, |
2019 |
Available online : December 31, |
2019 |
https://doi.org/10.21098/bemp.v22i4.1236
530 |
Bulletin of Monetary Economics and Banking, Volume 22, Number 4, 2019 |
|
|
I. INTRODUCTION
Islamic banking has demonstrated remarkable development, even amid the global financial crisis, and has risen into prominence in several dual banking jurisdictions. According to Ernst & Young’s World Islamic Banking Competitiveness Report 2016, Islamic banking recorded an asset growth rate of 16% per annum from 2010 to 2014. While the growth of Islamic banking assets has slowed down in recent years, with a compound annual growth rate of 8.8% from the last quarter of 2013 to 2017 (Islamic Financial Services Board, or IFSB, 2018), they continue to outpace their conventional counterparts. Islamic banking is now systematically important in 10 dual banking countries, according to the IFSB, an international
This paper aims to answer the question by addressing a core aspect of financial intermediaries, that is, financial intermediation costs. More specifically, focusing on Malaysia, we examine the extent to which the presence of Islamic banking affects financial intermediation costs as measured by net interest margins.2 Among systematically important dual banking countries, Malaysia is at the forefront in the development of Islamic banking, with many countries attempting to emulate its model and framework (Lassoued, 2018; Solarin et al., 2018). Starting with only one
Our contribution to the literature is twofold. First, we contribute to a growing list of studies on Islamic banking and its financial and economic roles. Predominantly, studies have focused on comparative analyses of Islamic and conventional banks based on various performance metrics, most notably efficiency (Beck et al., 2013; Johnes et al., 2014;
1The IFSB categorizes a country as having a systematically important Islamic banking sector based on an asset share of 15% or higher. Excluding the
Saudi Arabia, the United Arab Emirates, and Yemen.
2Islamic banking strictly prohibits interest, and, hence, the margins for Islamic banks are normally referred to as net financing margins. To avoid repeating the term interest/financing in reference to conventional and Islamic banks, we use the terms net margins, bank margins, and simply margins throughout.
3The figures are based on Monthly Highlights and Statistics (January 2018), Table 1.7 (Banking System:
Statement of Assets) and Table 1.71 (Islamic Banking System: Statement of Assets), published by Bank Negara Malaysia (i.e. the Central Bank of Malaysia).
Financial Intermediation Costs in a Dual Banking System: The Role of Islamic Banking |
531 |
|
|
(Beck et al., 2013; Boukhis and Nabi, 2013; Kabir et al., 2015; Sorwar et al., 2016; Kabir and Worthington, 2017; Zins and Weill, 2017; Alqahtani and Mayes, 2018; Ibrahim and Rizvi, 2018), profitability (Olson and Zoubi, 2017; Trad et al., 2017; Yanikkaya et al., 2018), and financing/deposit behavior (Abdul Karim et al., 2014; Ibrahim, 2016; Ibrahim and Rizvi, 2018). The underlying premise of these studies is that, if Islamic banks are demonstrated to be more efficient, more stable, and more profitable and their financing behavior less procyclical, they would contribute more positively to the
Our second contribution stems from the fact that, compared to other dimensions of bank performance, the implications of Islamic banking on financial intermediation costs remain understudied. Arguably, Islamic banking presence can result in higher or lower bank margins. In the provision of banking services, Islamic banks face additional risk and requirements, such as Shari’ah non- compliant risk, Shari’ah governance requirement, and the complexity of contracts. This means that, at the outset, Islamic banking services involve higher costs. Accordingly, to account for the additional costs from adherence to Islamic law, or Shari’ah, Islamic banks can set larger margins. The increasing presence of Islamic banking, however, induces competitive pressure and can thus favorably influence pricing in the banking industry. In other words, indirectly, through bank competition, the presence of Islamic banks would lead to lower net margins in the banking sector. An analysis is therefore necessary to ascertain whether Islamic banking entails higher costs and whether Islamic banking presence shapes the intermediation costs of the banking sector. Such insight would be important for drawing policy initiatives to foster the efficient allocation of resources by the Islamic banking sector and the overall banking system.
After controlling for their differences in market power, capitalization or risk aversion, scale of lending activity, operating costs, and diversification, our analysis provides evidence that Islamic banks have higher net margins. Further, the difference in the margins of Islamic and conventional banks is attributed to the differences in their market power, operating costs, and diversification. Finally, Islamic banking presence, as represented by the ratio of Islamic financing to aggregate bank credit/financing and, alternatively, the share of Islamic banking assets, is robustly associated with lower bank margins, on average. Thus, to optimize the benefits of Islamic banking presence, the development of Islamic banking should be encouraged, however, measures are needed to reduce the incremental costs incurred by Islamic banks in their adherence to Shari’ah principles, as well as to reduce cost inefficiency and to expand into
The paper is structured as follows. Section II reviews the literature. Section
IIIdescribes the empirical approach. Section IV presents the data and estimation results. Finally, Section V concludes the paper with a summary of the main findings and policy implications.
532 |
Bulletin of Monetary Economics and Banking, Volume 22, Number 4, 2019 |
|
|
II. RELATED LITERATURE
The theoretical basis for bank margins is the dealership model developed by Ho and Saunders (1981) and later augmentations by Allen (1988), Angbanzo (1997), and Maudos and Fernandez de Guevara (2004). According to the (augmented) dealership model, variations in margins across banks are accounted for by the following factors: market structure, transaction size, managerial risk aversion, interest rate uncertainty, operating costs, diversification, and credit risk. Taking advantage of the model’s flexibility in terms of factors to be included, some studies have also considered regulations, institutional quality, and bank ownership to explain variations in bank margins (Poghosyan, 2010, 2013; Fungacova and Poghosyan, 2011; Birchwood et al., 2017). Beck and Hesse (2009) neatly categorize these factors under four views: a
Empirically, research on bank margins comprises
Noting that inefficiencies of financial intermediation as manifested by high margins are more a feature of developing countries, several
Financial Intermediation Costs in a Dual Banking System: The Role of Islamic Banking |
533 |
|
|
Over recent years, Islamic banking has become a key financial development feature in various markets, particularly in Malaysia and the Middle East. Its ever- increasing presence in these countries notwithstanding, the public generally views Islamic banking as more expensive than its conventional counterpart. Corroborating this popular view, Ernst & Young’s World Islamic Banking Competitiveness Report
Along with Poghosyan’s (2010) evaluation of foreign bank penetration, we can argue that Islamic banking can exert both direct and indirect effects on bank margins. Being financial intermediaries, Islamic banks face similar risks as those of conventional banks, including credit risk, liquidity risk, and interest rate risk. However, distinct from conventional banks in their adherence to the principles of Shari’ah, Islamic banks’ operations entail additional unique risks, including fiduciary risk, displaced commercial risk, and rate of return risk (Archer and Karim, 2006). Islamic banking transactions also run the risk of being Shari’ah non- compliant after contracts are concluded, also known as Shari’ah
Indirectly, increasing the presence of Islamic banking should lead to greater competition in the banking sector. As hypothesized by Abedifar et al., (2016), this would likely have a spillover effect on conventional banks, such that, in the presence of Islamic banking, conventional banks become more cost efficient. Utilizing
Although empirical literature on Islamic banking is expanding, studies on the relations between Islamic banking and bank margins or intermediation costs remain limited. Further, such studies focus mainly on identifying the determinants of Islamic banks’ net margins or, at most, comparing the determinants of bank margins for Islamic and conventional banks (e.g., Abdul Kader Malim et al.,
534 |
Bulletin of Monetary Economics and Banking, Volume 22, Number 4, 2019 |
|
|
2017; Sun et al., 2017; Trinugroho et al., 2018). Accordingly, no inferences can be made about whether Islamic banking presence results in higher or lower financial intermediation costs. This paper fills this gap in the Islamic banking literature.
III. EMPIRICAL APPROACH
The model specification for bank margins takes the form of either a static panel model (Beck and Hesse, 2009; Poghosyan, 2010) or a dynamic panel model (Birchwood et al., 2017; Claessens et al., 2018). We opt for the static panel model to investigate the role of Islamic banking presence in financial intermediation costs. The key reason is our panel sample size does not fit the dynamic panel setting. For consistency, dynamic panel regressions are normally estimated using generalized method of moments (GMM) estimators
Building on the work of Poghosyan (2010), which is based on the dealership model of Ho and Saunders (1981) and its later extensions, we specify the following general specification to address the extent to which Islamic banking presence affects net margins:
(1)
where BM is the net margin as a measure of financial intermediation costs; IB is an Islamic bank dummy; IBShare is the market share of the Islamic banking sector; Bank is a vector of
A more direct measure of net margins is the difference between lending/ financing and deposit rates, which can be based on the implicit lending rate (interest/financing income on loans over total loans) and the implicit deposit rate (interest/financing expenses on customer deposits to total deposits), following Birchwood et al. (2017). However, the information on interest/financing income on loans and interest/financing expenses on customer deposits is only available for a short time. Accordingly, we use the ratio of gross interest/financing and dividend income net of total interest/financing expenses to average earning assets, which is another standard measure of BM in the literature (Beck and Hesse, 2009; Poghosyan, 2010; Claessens et al., 2018).
Financial Intermediation Costs in a Dual Banking System: The Role of Islamic Banking |
535 |
|
|
The key variables in Equation (1) are IB and IBShare, where IB is a dummy variable taking the value of one if the bank is Islamic, and zero otherwise. The inclusion of IB is to test whether the margins of Islamic banks are, on average, significantly different from the margins of conventional banks, which can be considered a direct effect of Islamic banking presence. We expect Islamic banks to have higher margins due to their specific, unique
In line with the literature, we include the following macroeconomic variables (Macro) in the analysis: gross domestic product (GDP) growth (∆Y), inflation (INF), the interest rate (INTR), and interest rate uncertainty (σ(INTR)). The effect of GDP growth on bank margins cannot be signed a priori. On one hand, as noted by Poghosyan (2013) and Birchwood et al. (2017), economic growth expands investment activities and improves borrowers’ creditworthiness. Accordingly, bank margins will be lower. On the other hand, the increases in credit demand and deposit supply from economic expansion can widen bank margins (Birchwood et al., 2017). Capturing macroeconomic uncertainty, inflation can increase bank margins. We employ the overnight interbank rate to represent the interest rate.
536 |
Bulletin of Monetary Economics and Banking, Volume 22, Number 4, 2019 |
|
|
Representing monetary policy as well as the marginal costs of funds, higher interest rates can lead to larger margins. Interest rate uncertainty is measured by the standard deviation of the monthly overnight interbank rate. It captures the uncertainty that banks face in their provision of intermediation services. As the uncertainty heightens, banks are more likely to increase net margins. We also include a foreign bank dummy (Poghosyan, 2010) and two crisis dummies (for the Asian and global financial crises) to control for the potential pricing differences of foreign banks and the effects of financial crises, respectively.
The static panel model, such as Equation (1), is typically estimated using traditional panel estimators. However, in our case, two features in our model specification require attention: the inclusion of a
IV. DATA AND RESULTS
We employ unbalanced panel data for 21 conventional banks and 16 Islamic banks over the period from 1997 to 2015.4 The
4The time span is dictated by the availability of all the relevant data at the time this study began. Our look at recent data does not reveal many changes in the Islamic banking share, which is our key variable, and the data are available only up to 2017, and then only for some banks. Accordingly, we do not believe that the addition of one or two more years of observations is likely to have a material impact on the results.
Financial Intermediation Costs in a Dual Banking System: The Role of Islamic Banking |
537 |
|
|
Table 1.
Descriptive Statistics
This table reports descriptive statistics of the data used in the paper for a sample of all banks and
Variable |
All Banks |
Islamic Banks |
Conventional Banks |
||||
Mean |
Std. Dev. |
Mean |
Std. Dev. |
Mean |
Std. Dev. |
||
|
|||||||
BM |
2.335 |
0.772 |
2.445 |
0.830 |
2.290 |
0.743 |
|
LERNER |
0.403 |
0.110 |
0.383 |
0.124 |
0.411 |
0.104 |
|
EQA |
10.588 |
6.110 |
8.444 |
3.874 |
11.466 |
6.625 |
|
NPL |
5.657 |
7.458 |
4.392 |
5.213 |
6.175 |
8.152 |
|
Ln(LOAN) |
7.657 |
1.857 |
7. 707 |
1.009 |
7.636 |
2.109 |
|
OPCOST |
1.368 |
0.523 |
1.389 |
0.547 |
1.359 |
0.513 |
|
NII |
16.738 |
11.168 |
8.813 |
7.235 |
19.973 |
10.877 |
|
IBShare |
13.455 |
7.794 |
|
|
|
|
|
σ(INTR) |
0.322 |
0.563 |
|
|
|
|
|
INTR |
3.468 |
1.654 |
|
|
|
|
|
∆Y |
4.632 |
3.949 |
|
|
|
|
|
INF |
2.446 |
1.294 |
|
|
|
|
Several observations are worth noting in the tables. First, as anticipated, the intermediation costs or margins of Islamic banks are, on average, higher than those of conventional bank margins. This finding indicates that Islamic banking products are more costly. Second, in line with the international sample of Beck et al. (2013), Islamic banks have better loan quality, lend more, and have higher operating costs per unit of assets. However, Islamic banks in Malaysia tend to be different, in that they have lower capitalization. Moreover, their income depends heavily on financing. Interestingly, they have less market power. This could be due to the consolidation of domestic conventional banks since the Asian crisis, which has allowed them to set higher prices for their products. Third, Table 2 gives a first indication that Islamic banking participation likely brings down intermediation costs, as reflected in the negative correlation between bank margins and Islamic financing share.
Table 2.
Correlation Coefficients
This table reports the pairwise correlation coefficients between variables: BM is bank margins measured by net interest margins; LERNER is the Lerner index; EQA is the ratio of equity to total assets; NPL is ratio of
Variable |
BM |
LERNER |
EQA |
NPL |
Ln(LOAN) |
OPCOST |
NII |
IBShare |
σ(INTR) |
INTR |
∆Y |
INF |
|
|
|
|
|
|
|
|
|
|
|
|
|
BM |
1.000 |
|
|
|
|
|
|
|
|
|
|
|
LERNER |
0.418 |
1.000 |
|
|
|
|
|
|
|
|
|
|
EQA |
0.156 |
1.000 |
|
|
|
|
|
|
|
|
|
|
NPL |
0.118 |
0.095 |
0.072 |
1.000 |
|
|
|
|
|
|
|
|
Ln(LOAN) |
0.239 |
0.099 |
1.000 |
|
|
|
|
|
|
|
||
OPCOST |
0.293 |
0.299 |
1.000 |
|
|
|
|
|
|
|||
NII |
0.112 |
0.295 |
0.091 |
0.271 |
1.000 |
|
|
|
|
|
||
IBShare |
0.230 |
1.000 |
|
|
|
|
||||||
σ(INTR) |
0.353 |
0.201 |
0.006 |
0.134 |
0.059 |
1.000 |
|
|
|
|||
INTR |
0.287 |
0.111 |
0.054 |
0.011 |
0.814 |
1.000 |
|
|
||||
∆Y |
0.030 |
0.002 |
0.126 |
1.000 |
|
|||||||
INF |
0.098 |
0.010 |
0.010 |
0.279 |
0.507 |
1.000 |
||||||
|
|
|
|
|
|
|
|
|
|
|
|
|
538
2019 4, Number 22, Volume Banking, and Economics Monetary of Bulletin
Financial Intermediation Costs in a Dual Banking System: The Role of Islamic Banking |
539 |
|
|
We further note the potential importance of market power/competition and
A. Baseline Results
Table 3 reports the results of Equation (1) using the
(1)to (3), we include only
Table 3.
Estimation Results – Net Interest Margin
This table reports the baseline results. The estimations are performed using the Panel Corrected Standard Errors (PCSE) estimator allowing for autocorrelation in the data (the Prais - Winston estimator).
Independent |
|
|
Regression |
|
|
|
Variables |
(1) |
(2) |
(3) |
(4) |
(5) |
(6) |
IB |
|
0.1206* |
|
0.1144* |
||
|
(0.236) |
|
(0.057) |
(0.381) |
|
(0.067) |
IBSHARE |
|
|
||||
|
|
(0.000) |
(0.000) |
|
(0.000) |
(0.000) |
FB |
0.0282 |
0.0703 |
0.0080 |
0.0316 |
0.0715 |
|
|
(0.885) |
(0.576) |
(0.114) |
(0.868) |
(0.525) |
(0.106) |
LERNER |
2.7089*** |
2.2435*** |
2.1301*** |
2.5888*** |
2.2266*** |
2.1232*** |
|
(0.000) |
(0.000) |
(0.000) |
(0.000) |
(0.000) |
(0.000) |
EQA |
0.0033 |
0.0076** |
0.0102** |
0.0042 |
0.0075* |
0.0100** |
|
(0.405) |
(0.050) |
(0.014) |
(0.285) |
(0.051) |
(0.015) |
NPL |
0.0019 |
0.0018 |
||||
|
(0.539) |
(0.238) |
(0.277) |
(0.556) |
(0.273) |
(0.309) |
Ln(LOAN) |
0.0461*** |
0.0800*** |
0.0988*** |
0.0550*** |
0.0807*** |
0.0984*** |
|
(0.006) |
(0.000) |
(0.000) |
(0.001) |
(0.000) |
(0.000) |
OPCOST |
0.6740*** |
0.6170*** |
0.5934*** |
0.6580*** |
0.6183*** |
0.5964*** |
|
(0.000) |
(0.000) |
(0.000) |
(0.000) |
(0.000) |
(0.000) |
NII |
||||||
|
(0.000) |
(0.000) |
(0.000) |
(0.000) |
(0.000) |
(0.000) |
5The assumption of a common correlation coefficient across banks does not materially affect the results.
540Bulletin of Monetary Economics and Banking, Volume 22, Number 4, 2019
Table 3.
Estimation Results – Net Interest Margin (Continued)
Independent |
|
|
Regression |
|
|
|
Variables |
(1) |
(2) |
(3) |
(4) |
(5) |
(6) |
σ(INTR) |
|
|
|
0.1710*** |
0.0547 |
0.0555 |
|
|
|
|
(0.000) |
(0.221) |
(0.262) |
INTR |
|
|
|
|||
|
|
|
|
(0.478) |
(0.449) |
(0.590) |
ΔY |
|
|
|
0.0283*** |
0.0223*** |
0.0220*** |
|
|
|
|
(0.000) |
(0.000) |
(0.000) |
INF |
|
|
|
|||
|
|
|
|
(0.375) |
(0.238) |
(0.244) |
AFC |
1.1307*** |
0.9059*** |
0.8947*** |
1.4891*** |
1.3067*** |
1.2673*** |
|
(0.000) |
(0.000) |
(0.000) |
(0.000) |
(0.000) |
(0.000) |
GFC |
0.0221 |
0.0007 |
0.0022 |
0.2028*** |
0.1509*** |
0.1513*** |
|
(0.680) |
(0.984) |
(0.950) |
(0.002) |
(0.001) |
(0.002) |
Constant |
0.3200* |
0.6268*** |
0.5072*** |
0.2157 |
0.5432*** |
0.4178** |
|
(0.061) |
(0.000) |
(0.007) |
(0.267) |
(0.003) |
(0.040) |
N |
470 |
470 |
470 |
470 |
470 |
470 |
# of Banks |
37 |
37 |
37 |
37 |
37 |
37 |
R2 |
0.5597 |
0.5756 |
0.5770 |
0.5669 |
0.5792 |
0.5803 |
Chi2 |
1025.2799 |
981.4577 |
873.4208 |
3441.0917 |
1516.8851 |
1237.8291 |
0.0000 |
0.0000 |
0.0000 |
0.0000 |
0.0000 |
0.0000 |
The estimation results generally conform to expectations and lend themselves to intuitive interpretation. As Table 3 shows, market power, the
The size of banking operations as measured by the size of loans is significantly and positively related to bank margins. This result conforms to the prediction of the dealership model. Since the size of lending activities is closely related to bank size, it likely reflects the dominance of “too big to fail” or moral hazard effects over the effect of economies of scale as banks grow. The result contradicts findings
Financial Intermediation Costs in a Dual Banking System: The Role of Islamic Banking |
541 |
|
|
for Islamic banks by Abdul Kadir Malim et al. (2017) and Trinugroho et al. (2018). Using total assets as a measure of bank size, these authors find, respectively, negative and nonsignificant coefficients of bank size. Compared to the banking literature in general, our results are in line with those of Poghosyan (2010) for 11 Central and Eastern European countries and those of Were and Wambua (2014) for Kenya; however, they contradict those of Kasman et al. (2010) for European countries, Poghosyan (2013) for
We also find operating costs to be consistently positive and significant in all regressions. In a similar vein probably indicating
Turning to our main theme, we find that the Islamic bank dummy (IB) enters significantly, albeit at the 10% significance level, when it is included together with the Islamic bank market share (IBShare) in the regressions. Based on regression
(3)or (6), Islamic banks with comparable
Although, comparatively, Malaysia’s Islamic banks tend to have higher intermediation costs, their increasing presence lowers the average margins of the banking sector. This is reflected in the negative coefficient of IBShare, which is distinguishable from zero at the 1% significance level in all regressions. Based on regression (6), the estimated coefficient suggests that a
542 |
Bulletin of Monetary Economics and Banking, Volume 22, Number 4, 2019 |
|
|
of Islamic banking has economic significance as far as the financial intermediation costs are concerned. Although Islamic banks tend to be more expensive, the competitive pressure that they bring to the banking system makes banks more efficient in the allocation of financial resources.
B. Robustness Checks
We provide robustness checks of our baseline results in several directions. First, instead of the financing share of Islamic banks, we use their asset share as a measure of Islamic banking presence or penetration (Gheeraert, 2014; Abedifar et al., 2016; Imam and Kpodar, 2016). Second, acknowledging that the banking sector in Malaysia underwent a tumultuous period during the
Table 4.
Estimation Results – Robustness
This table reports the results from robustness analysis using asset share as an alternative measure of Islamic banking share, shortening the sample, and excluding Islamic banks from the sample. The estimations are performed using the Panel Corrected Standard Errors (PCSE) estimator allowing for autocorrelation in the data (the Prais - Winston estimator).
Independent |
Asset Share |
Shortened Sample |
Islamic Bank Excluded |
|||
Variables |
(1) |
(2) |
(3) |
(4) |
(5) |
(6) |
IB |
0.1012 |
0.1002 |
0.1955** |
0.1859** |
|
|
|
(0.1047) |
(0.1074) |
(0.0275) |
(0.0360) |
|
|
IBSHARE |
|
|
|
|
||
(Financing) |
|
|
(0.0000) |
|
(0.0000) |
|
IBSHARE |
|
|
||||
(Assets) |
(0.0000) |
(0.0000) |
|
(0.0000) |
|
(0.0000) |
FB |
0.0561 |
0.0604 |
0.0651* |
0.0591 |
||
|
(0.2221) |
(0.1810) |
(0.4063) |
(0.3132) |
(0.0797) |
(0.1083) |
LERNER |
2.2423*** |
2.2049*** |
1.6736*** |
1.7593*** |
2.5516*** |
2.5952*** |
|
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
EQA |
0.0084** |
0.0085** |
0.0145*** |
0.0131*** |
0.0137*** |
0.0126** |
|
(0.0359) |
(0.0343) |
(0.0044) |
(0.0087) |
(0.0086) |
(0.0143) |
NPL |
||||||
|
(0.2886) |
(0.3468) |
(0.1046) |
(0.1223) |
(0.8580) |
(0.8364) |
Ln(LOAN) |
0.0887*** |
0.0904*** |
0.1056*** |
0.0970*** |
0.1140*** |
0.1098*** |
|
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
OPCOST |
0.6083*** |
0.6057*** |
0.6563*** |
0.6679*** |
0.5708*** |
0.5740*** |
|
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
Financial Intermediation Costs in a Dual Banking System: The Role of Islamic Banking |
543 |
|
|
Table 4.
Estimation Results – Robustness (Continued)
Independent |
Asset Share |
Shortened Sample |
Islamic Bank Excluded |
|||
Variables |
(1) |
(2) |
(3) |
(4) |
(5) |
(6) |
NII |
||||||
|
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
σ(INTR) |
|
0.0939** |
0.4277** |
0.5652** |
0.0660 |
0.0959** |
|
|
(0.0253) |
(0.0285) |
(0.0116) |
(0.2354) |
(0.0462) |
INTR |
|
0.0865 |
0.0978 |
0.0130 |
0.0224 |
|
|
|
(0.6628) |
(0.3208) |
(0.2784) |
(0.7835) |
(0.5681) |
∆Y |
|
0.0242*** |
0.0131 |
0.0242** |
0.0163*** |
0.0177*** |
|
|
(0.0000) |
(0.2169) |
(0.0449) |
(0.0065) |
(0.0003) |
INF |
|
0.0068 |
0.0047 |
|||
|
|
(0.2398) |
(0.2678) |
(0.2148) |
(0.6139) |
(0.6805) |
AFC |
0.9308*** |
1.2735*** |
|
|
1.1087*** |
1.0650*** |
|
(0.0000) |
(0.0000) |
|
|
(0.0000) |
(0.0000) |
GFC |
0.0289 |
0.1899*** |
0.0570 |
0.1765* |
0.0120 |
|
|
(0.4055) |
(0.0000) |
(0.5512) |
(0.0725) |
(0.6660) |
(0.7937) |
Constant |
0.5121*** |
0.3922** |
0.4845* |
0.3035 |
||
|
(0.0059) |
(0.0430) |
(0.0967) |
(0.2956) |
(0.8216) |
(0.7337) |
N |
470 |
470 |
389 |
389 |
339 |
339 |
# of Banks |
37 |
37 |
37 |
37 |
21 |
21 |
R2 |
0.5728 |
0.5773 |
0.6163 |
0.6105 |
0.6440 |
0.6426 |
Chi2 |
965.9754 |
1431.4006 |
870.4162 |
576.8230 |
1192.0317 |
1451.5450 |
0.0000 |
0.0000 |
0.0000 |
0.0000 |
0.0000 |
0.0000 |
The results in Table 4 provide further support for our conclusions that
(i)Islamic banking services are more costly and (ii) Islamic banking penetration lowers banking intermediation costs. The coefficients of IB remain positive and significant at better than the 5% significance level in the reduced sample, though their
544 |
Bulletin of Monetary Economics and Banking, Volume 22, Number 4, 2019 |
|
|
Table 5.
Estimation Results using Alternative Estimators – Robustness
This table reports estimation results using alternative estimators. The RE estimator is implemented with robust standard errors. The HT estimator takes interest rate, real GDP growth and inflation to be endogenous. Finally, the
Independent |
Full Sample |
Shortened Sample |
Islamic Bank Excluded |
|||
Variables |
Financing |
Assets |
Financing |
Assets |
Financing |
Assets |
|
Panel A: |
|
|
|||
IB |
0.0583 |
0.0406 |
|
|
||
|
(0.8235) |
(0.7582) |
(0.6584) |
(0.7581) |
|
|
IBSHARE |
|
|
|
|||
(Financing) |
(0.1876) |
|
(0.0001) |
|
(0.0035) |
|
IBSHARE |
|
|
|
|||
(Assets) |
|
(0.2650) |
|
(0.0026) |
|
(0.0097) |
FB |
0.0101 |
0.0026 |
||||
|
(0.4766) |
(0.3986) |
(0.0787) |
(0.0522) |
(0.9153) |
(0.9784) |
LERNER |
3.2503*** |
3.3180*** |
2.4894*** |
2.5917*** |
3.0167*** |
3.0737*** |
|
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
NPL |
||||||
|
(0.2732) |
(0.3104) |
(0.0564) |
(0.0827) |
(0.5941) |
(0.5772) |
Ln(LOAN) |
0.0213 |
0.0117 |
0.0125 |
0.0908** |
0.0859** |
|
|
(0.5731) |
(0.7534) |
(0.7276) |
(0.9732) |
(0.0248) |
(0.0361) |
OPCOST |
0.7871*** |
0.7953*** |
0.7796*** |
0.7940*** |
0.6536*** |
0.6599*** |
|
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
NII |
||||||
|
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
∆Y |
0.0255** |
0.0267** |
0.0133 |
0.0231* |
0.0173 |
0.0186 |
|
(0.0259) |
(0.0204) |
(0.2522) |
(0.0518) |
(0.1293) |
(0.1020) |
|
Panel B: |
|
|
|||
IB |
0.0238 |
|
|
|||
|
(0.7890) |
(0.7251) |
(0.8947) |
(0.9663) |
|
|
IBSHARE |
|
|
|
|||
(Financing) |
(0.0603) |
|
(0.0148) |
|
(0.0039) |
|
IBSHARE |
|
|
|
|||
(Assets) |
|
(0.1316) |
|
(0.0791) |
|
(0.0068) |
FB |
||||||
|
(0.5727) |
(0.5112) |
(0.0731) |
(0.0483) |
(0.9828) |
(0.9114) |
LERNER |
3.2975*** |
3.3720*** |
2.5652*** |
2.6569*** |
2.9117*** |
2.9609*** |
|
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
EQA |
0.0033 |
0.0024 |
0.0080 |
0.0066 |
0.0118* |
0.0107* |
|
(0.5912) |
(0.6910) |
(0.1790) |
(0.2633) |
(0.0519) |
(0.0744) |
Ln(LOAN) |
0.0189 |
0.0095 |
0.0902*** |
0.0845** |
||
|
(0.6203) |
(0.8021) |
(0.2838) |
(0.0973) |
(0.0069) |
(0.0112) |
OPCOST |
0.7897*** |
0.7988*** |
0.7652*** |
0.7738*** |
0.6342*** |
0.6394*** |
|
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
Financial Intermediation Costs in a Dual Banking System: The Role of Islamic Banking |
545 |
|
|
Table 5.
Estimation Results using Alternative Estimators – Robustness (Continued)
Independent |
Full Sample |
Shortened Sample |
Islamic Bank Excluded |
|||
Variables |
Financing |
Assets |
Financing |
Assets |
Financing |
Assets |
NII |
||||||
|
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
(0.0000) |
∆Y |
0.0250** |
0.0262** |
0.0119 |
0.0186 |
0.0171 |
0.0184 |
|
(0.0416) |
(0.0321) |
(0.4589) |
(0.2309) |
(0.1804) |
(0.1471) |
|
|
Panel C: |
|
|
||
BM |
0.5158*** |
0.3950*** |
0.5416** |
0.4635** |
0.2226 |
0.2090 |
(0.0008) |
(0.0088) |
(0.0297) |
(0.0161) |
(0.5672) |
(0.7316) |
|
|
||||||
IB |
|
|
||||
|
(0.7333) |
(0.6265) |
(0.5884) |
(0.7166) |
|
|
IBSHARE |
|
|
|
|||
(Financing) |
(0.2237) |
|
(0.0323) |
|
(0.0030) |
|
IBSHARE |
|
|
|
|||
(Assets) |
|
(0.0409) |
|
(0.0403) |
|
(0.6702) |
NPL |
0.0095 |
0.0082 |
0.0033 |
0.0049 |
0.0123* |
0.0065 |
|
(0.1343) |
(0.2543) |
(0.7353) |
(0.6256) |
(0.0822) |
(0.3433) |
Ln(LOAN) |
0.5906*** |
0.0377 |
||||
|
(0.7447) |
(0.7404) |
(0.9150) |
(0.7933) |
(0.0019) |
(0.8868) |
σ(INTR) |
0.0184 |
0.3981 |
0.4691* |
0.1343 |
0.1587 |
|
|
(0.5970) |
(0.8615) |
(0.1950) |
(0.0751) |
(0.4206) |
(0.3527) |
∆Y |
0.0247** |
0.0242** |
0.0232** |
0.0271** |
0.0157 |
0.0215 |
|
(0.0199) |
(0.0137) |
(0.0200) |
(0.0378) |
(0.2644) |
(0.1015) |
In Table 5, we report the results using alternative estimators, namely, the random effect (RE), the
In the estimation, we include all explanatory variables, as in column (6) of Table
3.To conserve space, we do not tabulate the results for the two crisis dummies or the results for other variables that are nonsignificant in all the regressions. The results suggesting the benefit of Islamic banking penetration in reducing banking intermediation costs and hence improving the efficient intermediation
546 |
Bulletin of Monetary Economics and Banking, Volume 22, Number 4, 2019 |
|
|
of financial resources are quite robust to the alternative estimators, as well as to the employment of the Islamic banking asset share, to the reduced sample, and to the exclusion of Islamic banks. By contrast, the IB dummy becomes nonsignificant throughout. The RE and HT estimators further reaffirm the significance of market power, operating costs, and diversification in affecting bank margins, in line with the previously documented findings. Use of the system GMM, however, makes most of the variables become nonsignificant, which could be attributed to the small sample size.
In a nutshell, our robustness checks further support the role of Islamic banking presence in reducing financial intermediation costs. Additionally, our robustness exercise shows some support for the greater cost of Islamic banks. Apart from these results, we could attribute the differences in the margins between Islamic and conventional banks to market power, operating costs, and diversification. The higher operating costs and lower
V. CONCLUSION
The rapid penetration of Islamic banking in the global financial scene has captivated greater research interest, especially in its contribution to financial and economic development. In this paper, we address the extent to which Islamic banking presence affects financial intermediation costs, by comparing the net margins of Islamic and conventional banks (direct effect) and assessing the relation between Islamic banking penetration and the intermediation costs of the banking sector (indirect effect). The analysis provides evidence that Islamic banks have higher net margins. Islamic banking presence, as represented by the ratio of Islamic financing to aggregate bank credit/financing and, alternatively, the share of Islamic banking assets, is, however, robustly associated with lower bank margins, on average. Apart from these key findings, we note that market power, operating costs, and diversification appear to be the most robust determinants of bank margins in Malaysia. Thus, the higher cost inefficiency and lower diversification of Islamic banks seem to partly contribute to higher intermediation costs of Islamic banks.
These results provide firm grounds for further expansion of Islamic banking, since it improves efficiency in the banking system’s allocation of financial resources. At the same time, several recommendations are made to lower the intermediation costs of Islamic banks. Namely, Islamic banks must look for ways to reduce the incremental costs related to their adherence to Shari’ah principles, for example, by improving risk management, simplifying financial contracts, and becoming more cost efficient. These could require a more competitive Islamic banking environment. Finally, Islamic banks need to diversify their
Financial Intermediation Costs in a Dual Banking System: The Role of Islamic Banking |
547 |
|
|
REFERENCES
Abdul Kader Malim , N., Ibrahim, M. H., Mohd Rasid, M.E.S. (2017). Explaining Intermediation Costs of Islamic Banks in OIC Countries. In Hassan, M. K. (ed), Handbook of Empirical Research on Islamic and Economic Life, Cheltenham: Edward Elgar,
Abdul Karim, M., Hassan, M.K., Hassan, T., Mohamed, S. (2014). Capital Adequacy and Lending and Deposit Behaviors of Conventional And Islamic Banks.
Abdul Majid, M., Falahaty, M., Jusoh, M. (2017). Performance of Islamic and Conventional Banks: A
Abedifar, P., Hasan, I., Tarazi, A. (2016).
Ahmad, R. (2007). A Note on the
Allen, L. (1988). The Determinants of Bank Interest Margins: A Note. Journal of Financial and Quantitative Analysis, 23,
Alqahtani, F., Mayes, D.G., Brown, K. (2017). Islamic Bank Efficiency Compared to Conventional Banks During the Global Crisis in the GCC Region. Journal of International Financial Markets, Institutions and Money, 51,
Alqahtani, F., Mayes, D.G. (2018). Financial Stability of Islamic Banking and the Global Financial Crisis: Evidence from the Gulf Cooperation Council. Economic Systems, 42,
Angbanzo, L. (1997). Commercial Bank Net Interest Margins, Default Risk, Interest- Rate Risk, and
Archer, S., Karim, R.A.A. (2006). On Capital Structure, Risk Sharing and Capital Adequacy in Islamic Banks. International Journal of Theoretical and Applied Finance, 9,
Beck, T.,
Beck, T., Hesse, H. (2009). Why Are Spreads so High in Uganda? Journal of Development Economics, 88,
Beck, N., Katz, J.N. (1995). What to Do (and not to Do) with
Birchwood, A., Brei, M., Noel, D. M. (2017). Interest Margins and Bank Regulation in Central America and the Caribbean. Journal of Banking and Finance, 85,
Boukhis, K., Nabi, M.S. (2013). Islamic and Conventional Banks’ Soundness During the
Chortareas, G..E.,
548 |
Bulletin of Monetary Economics and Banking, Volume 22, Number 4, 2019 |
|
|
Claessens, S., Coleman, N., Donnelly, M. (2018).
Fungacova, Z., Poghosyan, T. (2011). Determinants of Bank Interest Margins in Russia: Does Bank Ownership Matter? Economic Systems, 35,
Gheeraert, L. (2014). Does Islamic Finance Spur Banking Sector Development? Journal of Economic Behaviour and Organization, 103,
Gheeraert, L., Weill, L. (2016). Does Islamic Banking Development Favour Macroeconomic Efficiency? Economic Modelling, 47,
Ho, T., Saunders, A. (1981). The Determinants of Bank Interest Margins: Theory and Empirical Evidence. Journal of Financial and Quantitative Analysis, 16, 581- 600.
Hossain, M. (2012). Financial Reforms and Persistently High Bank Interest Spreads in Bangladesh: Pitfalls in Institutional Development? Journal of Asian Economics, 23,
Ibrahim, M.H. (2016). Business Cycle and Bank Lending Procyclicality in a Dual Banking System. Economic Modelling, 55,
Ibrahim, M.H., Rizvi, S.A.R. (2018). Bank Lending, Deposits and
Imam, P., Kpodar, K. (2016). Islamic Banking: Good for Growth? Economic Modelling, 59,
Islamic Financial Services Board (IFSB). (2018). Islamic Financial Services Industry Stability Report 2018, Kuala Lumpur: IFSB.
Johnes, J., Izzeldin, M., Pappas, V. (2014). A Comparison of Performance of Islamic and Conventional Banks
Kabir, M.N., Wothington, A., Gupta, R. (2015). Comparative Credit Risk in Islamic and Conventional Bank.
Kabir, M.N, Worthington, A.C. (2017). The
Kafle, K., Jollife, D.,
Kasman, A., Tunc, G., Vardar, G., Okan, B. (2010). Consolidation and Commercial Bank Net Interest Margins: Evidence from the Old and New European Union Members and Candidate Countries. Economic Modelling, 27,
Lassoued, M. (2018). Comparative Study on Credit Risk in Islamic Banking Institutions: The Case of Malaysia. The Quarterly Review of Economics and Finance, 70,
Financial Intermediation Costs in a Dual Banking System: The Role of Islamic Banking |
549 |
|
|
Lebdaoui, H., Wild, J. (2016). Islamic Banking and Financial Development. Review of Middle Eastern Economics and Finance, 12,
Lepatit, L., Meslier, C., Strobel, F., Wardhana, L. (2018). Bank Dividends, Agency Costs and Shareholder and Creditor Rights. International Review of Financial Analysis, 56,
Maudos, J., Fernandez de Guevara, J. (2004). Factors Explaining the Interest Margin in the Banking Sectors of the European Union. Journal of Banking and Finance, 28,
Meslier, C., Risfandy, T., Tarazi, A. (2017). Dual Market Competition and Deposit Setting in Islamic and Conventional Banks. Economic Modelling, 63,
Olson, D., Zoubi, T., (2017). Convergence in Bank Performance for Commercial and Islamic Banks During and After the Global Financial Crisis. The Quarterly Review of Economics and Finance, 66,
Poghosyan, T. (2013). Financial Intermediation Costs in Low Income Countries: The Role of Regulatory, Institutional, and Macroeconomic Factors. Economic Systems, 37,
Poghosyan, T. (2010).
Roodman, D. (2009). A Note on the Theme of too many Instruments. Oxford Bulletin of Economics and Statistics, 71,
Solarin, S.A., Hammoudeh, S., Shahbaz, M. (2018). Influence of Economic Factors on Disaggregated Islamic Banking Deposits: Evidence with Structural Breaks in Malaysia. Journal of International Financial Markets, Institutions and Money, 55,
Sorwar, G., Pappas, V., Pereira, J., Nurullah, M. (2016). To Debt or not to Debt: Are Islamic Banks Less Risky than Conventional Banks? Journal of Economic Behaviour & Organization, 132,
Sufian, F., Habibullah, M. S. (2013). Financial Sector Consolidation and Competition in Malaysia: An Application of the
Sun, P. H., Mohamad, S., Ariff, M. (2017). Determinants Driving Bank Performance: A Comparison of Two Types of Banks in the OIC.
Trad, N., Trabelsi, M.A., Goux, J. F. (2017). Risk and Profitability of Islamic Banks: A Religious Deception or an Alternative Solution? European Research on Management and Business Economics, 23,
Trinugroho, I., Agusman, A., Tarazi, A. (2014). Why Have Bank Interest Margins Been so High in Indonesia since the 1997/1998 Financial Crisis? Research in International Business and Finance, 32,
Trinugroho, I., Risfandy, T., Ariefianto, M.D. (2018). Competition, Diversification, and Bank Margins: Evidence from Indonesian Rural Banks. Borsa Istanbul Review, in press.
Were, M., Wambua, J. (2014). What Factors Drive Interest Rate Spread of Commercial Banks? Empirical Evidence from Kenya. Review of Development Finance, 4,
550 |
Bulletin of Monetary Economics and Banking, Volume 22, Number 4, 2019 |
|
|
Williams, B. (2007). Factors Determining Net Interest Margins in Australia: Domestic and Foreign Banks. Financial Markets Institutions and Instruments, 16,
Yanikkaya, H., Gumus, N., Pabuccu, Y.U. (2018). How Profitability Differs Between Conventional and Islamic Banks: A Dynamic Panel Data Approach. Pacific- Basin Finance Journal, 48,
Zins, A.., Weill, L. (2017). Islamic Banking and Risk: The Impact of Basel II. Economic Modelling, 64,