Page |0

Factors affecting
profitability:
Islamic Vs
Conventional banks
A comparative analysis with
evidence from Bangladesh

Prepared by:
Ahadul Islam
ID: 2022151059
Department of Finance & Banking
 Page |I

Bangladesh University of Professionals

Submission of Term Paper
Course: FIN4204 – Bank Management

Submitted to
Md. Nahid Alam
Assistant Professor
Department of Finance and Banking
Bangladesh University of Professionals

Submitted by
Ahadul Islam
Section: A
Department of Finance and Banking
Batch: 2020

Submission Date: 20th Oct 2023

 Page |I

ACKNOWLEDGEMENT

This term paper and other activities behind it for our Bank Management (FIN-4204) course would
not have been possible without the exceptional support of our course teacher, Md. Nahid Alam,
Assistant Professor, Department of Business Administration in Finance and Banking under Faculty
of Business Studies, Bangladesh University of Professionals. His enthusiasm, knowledge and
exacting attention to detail have been an inspiration and kept our work on track. We acknowledge
with thanks the sincere and prompt facilitation and useful advice of his. Without his timely
guidance, suggestion, and constructive criticism, we could never craft a project of this scale. We
have gained vast knowledge in the field of Banking sector and its management through this course
and the term paper. This will prove to be helpful in our future career development.
 P a g e | II

EXECUTIVE SUMMARY
In the evolving landscape of global finance, the coexistence of Islamic and conventional banking in
countries like Bangladesh has garnered attention. With a significant portion of its population
abiding by Islamic principles, understanding the nuances and outcomes of these two distinct
banking practices becomes imperative.

Utilizing data sourced from annual reports of 6 Islamic and 6 conventional banks in Bangladesh
over a five-year span, this research adopted a comparative approach to understand banking
profitability determinants. The key profitability metric employed was the Return on Assets (ROA).
Various bank-specific indicators like liquidity, solvency, and asset quality, alongside
macroeconomic variables such as inflation and exchange rates, were integrated into the study.
Statistical software STATA was employed for data analysis, allowing for the identification and
comparison of factors influencing the profitability trajectories of these two distinct banking systems
within the Bangladeshi financial landscape.

The research illuminated intriguing dynamics in the financial performances of Islamic and
conventional banks in Bangladesh. Both banking models shared a positive correlation between their
Loan-to-Asset ratio and Return on Assets (ROA), indicating that lending activities, irrespective of
their nature, remain central to profitability. Furthermore, an increased Liabilities-to-Asset ratio was
found to negatively impact ROA for both, highlighting the challenges associated with effective
liability management. However, divergences emerged in other areas. For instance, conventional
banks showed a decline in ROA with higher loan loss provisions, while this relationship was less
evident in Islamic banks, hinting at unique risk management practices rooted in Sharia principles.
The role of Tier 1 capital in influencing profitability also differed. Conventional banks benefited
from strong capital adequacy, but for Islamic banks, this relationship was more nuanced.
Macroeconomic challenges, particularly inflation, were found to be more detrimental for Islamic
banks compared to their conventional counterparts. Lastly, while conventional banks saw
profitability benefits from favorable exchange rate movements, Islamic banks showed a more muted
relationship, suggesting varied operational focuses between the two.

The study reveals distinct financial dynamics between Islamic and conventional banks in
Bangladesh, driven by their foundational principles, risk management practices, and responsiveness
to macroeconomic factors.
 P a g e | III

TABLE OF CONTENTS

ACKNOWLEDGEMENT...................................................................................................................I
EXECUTIVE SUMMARY.................................................................................................................II
TABLE OF CONTENTS...................................................................................................................III
LIST OF TABLES..............................................................................................................................III
1. INTRODUCTION...........................................................................................................................1
1.1 Problem Statement....................................................................................................................1
1.2 Research Objective....................................................................................................................2
2. LITERATURE REVIEW...............................................................................................................2
3. RESEARCH DESIGN.....................................................................................................................4
3.1 Data.............................................................................................................................................4
3.2 Variable Measurement..............................................................................................................4
3.3 Research Methodology..............................................................................................................7
4. RESULTS.........................................................................................................................................8
4.1 Descriptive Statistics..................................................................................................................8
4.2 Empirical Results.......................................................................................................................9
5. FINDINGS & CONCLUSION......................................................................................................15
5.1 Similarities...............................................................................................................................15
5.2 Differences................................................................................................................................15
REFERENCES..................................................................................................................................17

LIST OF TABLES

Table 01: Selected Variables .................................................................................................. 5

Table 02: Descriptive statistics – Conventional banks ........................................................... 8
Table 03: Descriptive statistics – Islamic banks...................................................................... 9
Table 04: Regression Estimates – Conventional banks .......................................................... 9
Table 05: Regression Estimates – Islamic banks .................................................................. 12
 Page |1

1. INTRODUCTION

In the landscape of global finance, the coexistence of both Islamic and conventional banking
systems has led to intriguing discussions regarding their operational dynamics and financial
outcomes. This dichotomy is particularly pronounced in countries like Bangladesh, where a
significant portion of the population adheres to Islamic principles and values. The banking sector in
Bangladesh has experienced substantial growth, with both Islamic and conventional banks playing
vital roles in the country's economic development. Against this backdrop, a comprehensive analysis
of the factors influencing the profitability of Islamic and conventional banks in Bangladesh
becomes a pertinent subject of research.

The juxtaposition of Islamic and conventional banking systems within the same economic
environment raises compelling questions regarding the intricacies of their operational models and
the variances in their financial performance. This dynamic is inherently interesting within the
context of Bangladesh, a country where the Islamic way of life profoundly influences socio-
economic decisions. Islamic banks operate on the tenets of Sharia, emphasizing ethical transactions,
profit-sharing, and the prohibition of interest (riba). In contrast, conventional banks adhere to more
conventional interest-based financial practices. The objective of this research is to conduct an in-
depth and comparative analysis of the myriad factors that shape the profitability trajectories of
Islamic and conventional banks within the unique financial landscape of Bangladesh.

1.1 Problem Statement

The coexistence of Islamic and conventional banks within a single financial ecosystem raises
intriguing questions about the differences in their profitability and the underlying factors that drive
these disparities. While Islamic banks operate in accordance with Sharia principles, emphasizing
ethical and interest-free transactions, conventional banks follow traditional interest-based financial
practices. This study aims to delve into the comparative profitability of Islamic and conventional
banks in Bangladesh and unravel the intricate interplay of factors that contribute to their financial
performance.

1.2 Research Objective

The primary objective of this research is to conduct a thorough comparative analysis of the factors
influencing the profitability of Islamic and conventional banks in Bangladesh. This objective can be
further broken down into the following specific goals:
 Page |2

Specific objectives:
1. Assess and compare the financial performance of Islamic and conventional banks over a
specific period in Bangladesh.
2. Identify and analyze the key factors that impact the profitability of Islamic and conventional
banks.
3. Investigate the relationship between Sharia compliance and profitability. It will explore
whether Islamic banks' unique operational framework affects their financial performance
positively or negatively.
4. Consider the broader macroeconomic environment in Bangladesh and its potential influence
on the profitability of both banking systems.

2. LITERATURE REVIEW
In a study by Bashir (2000), an investigation was conducted into the factors influencing the
profitability of 14 Islamic banks operating in eight Middle Eastern nations during the period of
1993 to 1998. The study utilized cross-country bank-level data to analyze these determinants. The
findings of this research revealed a notable correlation between the profitability indicators of the
Islamic banks and the level of bank capital. Specifically, the study demonstrated a positive and
statistically significant relationship between bank profitability and the amount of capital held by
these Islamic financial institutions.

Cˇihák and Hesse (2008), in an examination encompassing 20 banking systems, presented findings
that underscored the significant impact of certain ratios on the return on assets (ROA). Particularly,
the study highlighted the influence of the total equity-to-total assets (Eq/TA) ratio and the total
loans-to-total assets ratio on ROA. Notably, their analysis indicated that as the loans-to-assets ratio
(LOAN/TA) increases, there is a corresponding increase in the profitability of Islamic banks.

According to a study comparing the profitability of Islamic and conventional banks in Bangladesh,
Islamic banks exhibited higher profitability, which they attributed to the application of risk-sharing
principles, ethical practices, and a focus on Shariah-compliant financial activities (Islam et al.
2014). Karim and Bashar (2017) demonstrated a similar view. Islamic banks followed a more
conservative liquidity management approach. This approach not only mitigated liquidity risks but
also contributed to enhanced profitability, particularly during periods of financial turmoil.
 Page |3

Moreover, the interplay between macroeconomic variables and Islamic bank profitability has also
been a subject of investigation. Wasiuzzaman and Tarmizi (2010) expound that certain
macroeconomic factors bear a connection to the profitability of Islamic banks. In their study, they
observed that the Gross Domestic Product (GDP) and inflation exerted a positive influence on bank
profitability. This positive correlation between GDP and bank profitability aligns with the earlier
findings of Bashir (2000).

Numerous studies have scrutinized the influence of bank size on the profitability of Islamic banks.
Analyzing 78 Islamic banks across 25 nations from 1992 to 2009, Mohamad et al. (2011) discerned
that among various factors, solely bank size exhibited a significant positive effect on Islamic bank
profitability. Similarly, Al-Qudah and Jaradat (2013) affirmed that bank size significantly correlates
with enhanced return on assets (ROA) and return on equity (ROE), accentuating its role in
bolstering profitability.

Several studies have presented evidence suggesting that Islamic banks outperform conventional
banks in terms of profitability. In an assessment conducted by Kader and Asarpota (2007), the
performance of three Islamic banks in the UAE was juxtaposed against that of five conventional
banks spanning the years 2000 to 2004. The findings of this study revealed that Islamic banks
exhibited relatively higher levels of profitability, lower risk profiles, and diminished liquidity
compared to their conventional counterparts.

Contrarily, a subset of studies has presented an alternative viewpoint, indicating that Islamic banks
might exhibit lower profitability when compared to conventional banks. A case in point is the
analysis conducted on Meezan Bank Limited in Pakistan during the period of 2003-2008, in
comparison to a set of five conventional banks. Cˇihák and Hesse (2008) found that Meezan Bank
Limited showcased comparatively lower profitability and efficiency levels. However, it was
observed that the bank demonstrated greater solvency, resulting in reduced risk exposure.

Furthermore, a comprehensive examination spanning the years 1991 to 2000, as conducted by

Samad (2004), yielded empirical results that suggest a lack of substantial disparity in terms of both
profitability and liquidity performance between the two categories of banks. Similarly, Akhter et al.
(2011) highlighted that, based on their investigation, the disparity in profitability between interest-
based and interest-free banking appears to be minimal. However, their study indicated a divergence
in liquidity and credit performance, implying that while profitability might not exhibit pronounced
differences, other aspects of performance can manifest variation between the two banking
paradigms.
 Page |4

These studies collectively emphasize the nuanced nature of the relationship between Islamic and
conventional banks, warranting a multifaceted understanding of their comparative performance
metrics.

3. RESEARCH DESIGN
3.1 Data
The data used in this study has been obtained from the annual reports published by the banks on
their websites. The financial statements were analyzed to form ratios and interpret the firm’s
performance. To accomplish the objective of this study a panel data was formed by collecting data
from 5 Shari’ah based Islamic Banks and 5 conventional banks over a period of 5 years. The data
has been analyzed using the statistical software STATA.

3.2 Variable Measurement

Profitability indicator

Return on Assets (ROA) has been used as a representative of profitability of banks in the study. The
ROA is quantified as a ratio that expresses total profitability as a percentage of total assets. It
indicates how efficient the assets of a bank are in generating return. This metric is obtained by
dividing a bank's net after-tax income by its total assets.

Net profit after tax

ROA=
Total assets

Determinant factors

A combination of bank-specific factors and macroeconomic factors has been used. The bank-
specific factors can further be broken down into Liquidity factors, Solvency factors, Asset quality
factors, Efficiency factors, and Capital factors. The details of the variables and their expected
relationship are provided below.
 Page |5

Table 01. Selected Variables

Expected
Category Factors Notation Description
relationship
Loans total Asset Ratio =
Bank-specific Liquidity LOANTA +/-
Loans/Total Assets

Liabilities to Asset Ratio =

Solvency LTA +/-
Total liabilities/Total sales

Asset Utilization =
Efficiency AU +
Revenue/Total assets

Asset Asset Quality =

LLP -
quality Loan Loss Provision/Net interest revenue

Tier 1 ratio =
Capital T1 +
Tier 1 Capital/RWA

Size =
Size SIZ +/-
Ln(Total Assets)

Macroeconomi Inflation =
Inflation INF +
c Change in CPI

Exchange Exchange rate movement =

EXR +/-
rate Change in exchange rates
Source: Authors' calculations

Note: (+) indicates positive impact; (-) indicates negative impact; (+/-) either positive or negative impact

To assess the potential impact of a bank's asset composition on its profitability, the liquidity aspect
is examined using the LOAN/TA ratio. LOAN/TA serves as a gauge of liquidity and signifies the
extent to which a bank has allocated its total assets to financing activities. A higher LOAN/TA ratio
suggests that the bank has engaged in more extensive lending, potentially exposing itself to elevated
financial stress. This ratio is particularly crucial in analyzing Islamic bank profitability, as
highlighted by research such as Bashir (2000), where a higher LOAN/TA ratio can signify
increased profitability (due to heightened profit income), but also entails elevated risk.

In our analysis, the total liabilities to total assets ratio (L/TA) is employed as an indicator of risk.
This ratio provides insights into the extent of leverage, or the level of capital held by the bank. We
anticipate a positive correlation between L/TA and profitability metrics. However, the scenario of
heightened risk-taking associated with greater leverage implies a potential negative association
between L/TA and profitability. The reason being that excessive risk exposure increases the
 Page |6

vulnerability of the bank to insolvency. Consequently, the coefficient of L/TA may exhibit a
negative sign in our analysis.

Efficiency ratios are instrumental in evaluating a bank's competence in managing and overseeing its
assets. AU specifically assesses the bank's efficacy in harnessing its assets. A low AU implies that
the bank is not optimizing its asset potential, thereby signaling the need for a more effective
utilization strategy.

The evaluation of banks' asset quality involves assessing several key ratios: the loan loss provision
to net interest revenue ratio (LLP), the loan-loss-reserves-to-gross-loans ratio (LLR), and the Tier 1
capital ratio (TOR). These ratios collectively provide insights into the quality of a bank's loan
portfolio. When loans are of high quality, aligning with the risk-return hypothesis, a notably high
ratio could potentially indicate a positive association between risk and profit. This connection
implies that a bank is strategically managing its assets to balance risk and reward.

The Tier 1 capital ratio (TOR) stands as a fundamental metric that regulators employ to gauge a
bank's financial robustness. This ratio is regarded as a pivotal indicator of a bank's strength from a
regulatory perspective. It offers a safeguard against unforeseen losses, in contrast to expected losses
that are addressed through provisions, reserves, and profits accrued in the current year.

The size of a bank, measured by the natural logarithm of its total assets, can influence profitability.
Larger banks might have economies of scale, potentially leading to enhanced cost efficiency and
better risk management. However, excessive size might hinder agility, leading to complex
operations that impact profitability negatively.

Anticipated inflationary increases tend to curb spending and borrowing for firms and households.
Elevated inflation is often linked to higher loan interest rates. This alignment can yield a positive
impact on bank profitability, as the correlation between inflation and interest rates enhances lending
revenue, potentially boosting a bank's profitability.

Changes in foreign exchange rates can wield substantial influence on banks' profitability. When a
domestic currency strengthens, the value of foreign assets diminishes, leading to translation losses.
Conversely, a weaker domestic currency can elevate profits by increasing the value of foreign
earnings upon conversion. These fluctuations necessitate strategic risk management and hedging to
mitigate potential adverse impacts on banks' bottom lines.
 Page |7

3.3 Research Methodology

Panel data models can be categorized into Balanced and Unbalanced. Balanced data consists of
observation for every variable throughout the period. If the data has any missing values, it is called
unbalanced data. The data used in this is strongly balanced i.e., there is no missing data.

To analyze the relationship between the dependent and independent variables mentioned in the
previous subsection, statistical regression analysis has been undertaken. Since the study does not
constitute any dummy variables, the POLS (Pooled Ordinary least Squared) model has been used
for the regression analysis.

The POLS model works best when individual effect (cross-sectional or time specific effect) does
not exist. According to Woolridge (2010), this model is employed when a different sample is
selected for each time period in the panel data. The POLS model can be mathematically represented
as follows where, ‘y’ is the dependent variables, ‘α’ is the intercept, ‘X’ is the independent variable,
‘β’ is the coefficient and ‘ε ’ is the residual.

y=α + β 1 X 1+ β2 X 2 + ε

The Random Effects (RE) model is another widely used method of panel data regression analysis. It
assumes that individual effect is not correlated with any regressor and then estimates error variance
specific to groups. The parameter estimate (u) of a dummy variable is a part of the error component
in the Random Effects model (Park, 2011) The model can be mathematically expressed as shown
below.

y=α + β 1 X 1+ β2 X 2 +(ε +u)

A fixed group effect (FE) model examines individual differences in intercepts, assuming the same
slopes and constant variance across individual groups and entity. Since an individual specific effect
is time invariant and considered a part of the intercept is allowed to be correlated with other
regressors. In other words, the parameter estimate is a part of the intercept in the FE model (Park,
2011).

y=(α +u)+ β1 X 1 + β 2 X 2+ ε

The following is the POLS regression model to examine the determinants of bank profitability:

ROA=α + β 1 LOANTA + β 2 LTA+ β3 AU + β 4 LLP+ β 5 T 1+ β 6 SIZ + β 7 INF + β 8 EXR + ε

 Page |8

4. RESULTS
4.1 Descriptive Statistics
Table 02 illustrates the descriptive statistics of the variables selected for conventional banks. The
table offers insights into central tendencies, dispersions, skewness, and kurtosis for each variable.
The LLP has the highest standard deviation among the variables indicating the variation in
provisioning across the banking industry. Based on the Jarque-Bera test, LOANTA, LTA, INF, and
LLP are not normally distributed (p-values < 0.05), while the other variables do not show strong
evidence against normality. Therefore, ROA, AU, T1, SIZ, and EXR may be considered
approximately normally distributed.

Table 02: Descriptive statistics - Conventional banks

ROA LOANTA LTA AU LLP T1 SIZ INF EXR
Mean 0.008 0.671 0.931 0.056 0.389 0.090 26.741 0.060 0.027
Median 0.007 0.684 0.934 0.056 0.233 0.089 26.769 0.056 0.012
Maximum 0.013 0.751 1.127 0.075 1.800 0.113 27.166 0.077 0.078
Minimum 0.004 0.334 0.515 0.034 0.002 0.070 26.128 0.055 0.002
Std. Dev. 0.003 0.076 0.098 0.012 0.436 0.013 0.271 0.009 0.029
Skewness 0.336 -3.553 -2.775 0.090 1.779 0.270 -0.439 1.472 0.923
Kurtosis 1.824 16.582 14.635 1.936 5.634 2.051 2.632 3.212 2.295

Jarque-Bera 1.911 244.767 173.087 1.212 20.419 1.242 0.944 9.076 4.068
Probability 0.385 0.000 0.000 0.546 0.000 0.537 0.624 0.011 0.131

Sum 0.189 16.773 23.267 1.394 9.726 2.259 668.527 1.500 0.676
Sum Sq. Dev. 0.000 0.140 0.232 0.003 4.554 0.004 1.759 0.002 0.020

Observations 25.000 25.000 25.000 25.000 25.000 25.000 25.000 25.000 25.000

Table 03 on the other hand shows the descriptive statistics of the variables selected for conventional
banks. The SIZ of banks has the highest standard deviation among the variables showing
differences in the size of Islamic banks in the country. Based on the Jarque-Bera test, LOANTA,
LTA, and INF are not normally distributed (p-values < 0.05), while the other variables do not show
strong evidence against normality. Therefore, ROA, AU, LLP, T1, SIZ, and EXR may be
considered approximately normally distributed.
 Page |9

Table 03: Descriptive statistics – Islamic Banks

ROA LOANTA LTA AU LLP T1 SIZ INF EXR
Mean 0.005 0.787 0.981 0.068 0.153 0.085 26.887 0.060 0.027
Median 0.005 0.739 0.954 0.066 0.188 0.085 26.805 0.056 0.012
Maximum 0.007 1.250 1.481 0.118 0.349 0.109 28.240 0.077 0.078
Minimum 0.003 0.688 0.923 0.045 0.005 0.060 26.007 0.055 0.002
Std. Dev. 0.001 0.116 0.112 0.017 0.112 0.012 0.641 0.009 0.029
Skewness -0.144 2.632 3.774 0.928 -0.109 -0.162 0.637 1.472 0.923
Kurtosis 1.833 11.189 17.174 3.942 1.653 2.873 2.622 3.212 2.295

Jarque-Bera 1.506 98.702 268.622 4.510 1.940 0.125 1.842 9.076 4.068
Probability 0.471 0.000 0.000 0.105 0.379 0.939 0.398 0.011 0.131

Sum 0.129 19.674 24.526 1.703 3.830 2.122 672.186 1.500 0.676
Sum Sq. Dev. 0.000 0.323 0.303 0.007 0.303 0.004 9.846 0.002 0.020

Observations 25.000 25.000 25.000 25.000 25.000 25.000 25.000 25.000 25.000

4.2 Empirical Results

To analyze the impact on the profitability of Conventional and Islamic banks, regression has been
conducted based on POLS (Pooled Ordinary Least Squared), Random effects and Fixed effects
model. The results obtained are shown below.

Table 04: Regression estimates – Conventional banks

(1) (2) (3)
VARIABLES POLS RE FE
ROA ROA ROA

LOANTA 0.00196 0.00196 -0.0118

(0.00890) (0.00890) (0.0173)
LTA -0.00653 -0.00653 -0.000773
(0.00561) (0.00561) (0.00980)
AU 0.0374 0.0374 -0.00976
(0.0343) (0.0343) (0.0568)
LLP -0.000248 -0.000248 -0.000267
(0.000807) (0.000807) (0.00121)
T1 0.276* 0.276* 0.250*
(0.0404) (0.0404) (0.0481)
SIZ -0.00216 -0.00216 -0.00684
(0.00142) (0.00142) (0.00625)
INF -0.397* -0.397* -0.298*
(0.110) (0.110) (0.147)
EXR 0.112* 0.112* 0.0933
(0.0279) (0.0279) (0.0332)
Constant 0.0641* 0.0641* 0.193
(0.0388) (0.0388) (0.168)
 P a g e | 10

Observations 25 25 25
R-squared 0.733
Number of id 5 5 5
Standard errors in parentheses
*** p<0.01, ** p<0.05, * p<0.1

Table 04 shows the impact different variables have on the ROA of a conventional bank. The results
show that T1, INF, and EXR have statistically significant impact on the profitability of conventional
banks. This is consistent across the POLS, RE and FE models.

The Hausman test shown in Fig 01 returned a chi2 p-value 0.869. This means that the individual
effects are not correlated. Random Effects model shows a better picture than the Fixed Effects
model.

Fig 01: Hausman statistics – Conventional banks

Coefficients
(b) (B) (b-B) sqrt(diag(V_b-V_B))
FE RE Difference S.E.

loanta -.011839 .0019561 -.0137952 .0147918

lta -.0007733 -.0065333 .00576 .0080424
au -.0097612 .0373882 -.0471494 .0452356
llp -.0002669 -.0002485 -.0000184 .0009062
t1 .2499888 .2758027 -.0258139 .0261389
siz -.0068437 -.0021638 -.0046799 .0060872
inf -.2981062 -.3974826 .0993763 .0984097
exr .0932984 .1124019 -.0191035 .0180175

b = consistent under Ho and Ha; obtained from xtreg

B = inconsistent under Ha, efficient under Ho; obtained from xtreg

Test: Ho: difference in coefficients not systematic

chi2(8) = (b-B)'[(V_b-V_B)^(-1)](b-B)
= 4.31
Prob>chi2 = 0.8286
(V_b-V_B is not positive definite)

Source: STATA output

Since the Hausman statistics say that the RE model shows better results, we consider the RE model
to explain the variables.
 P a g e | 11

The LONTA ratio is positively correlated with the ROA under both POLS and RE models. This
suggests that the provision of loans is positively correlated with the earnings of conventional banks.
The interest income generated from the loans contribute to the high earnings of banks. This is
consistent with the findings of Bashir (2003).

The LTA ratio is consistently negatively correlated with the ROA under all three models. The
liabilities of banks include deposits and debt capital with is associated with interest expenses.
Interest payments to the depositors and debt holders adversely affect the banks’ profitability.

AU positively affects the banks’ ROA under POLS and RE models. There could be several
explanations behind this relationship. Banks that are more efficient at generating revenue from their
assets tend to have higher ROA. When banks generate more revenue relative to their asset base,
they are better positioned to cover their expenses and generate profits, leading to a higher ROA.
Furthermore, banks that embrace technological advancements and innovative solutions can enhance
their revenue generation capabilities without significantly increasing their asset base.

LLP being negatively correlated to ROA can be attributed to the prudent risk management
practices. When banks increase their LLP, it indicates that they are setting aside more funds to
cover potential loan losses, which can reduce their reported profits, leading to a lower ROA. This
conservative approach is designed to safeguard the bank's financial stability by preparing for
uncertain economic conditions or a rise in loan defaults.

The positive relationship between T1 and ROA signifies the importance of strong capital adequacy
in bank profitability. When T1, representing a bank's core capital, is higher as a proportion of risk-
weighted assets, it indicates greater financial strength and resilience. Banks with ample Tier 1
capital can absorb losses, have lower default risk, and are often perceived as safer by investors and
regulators. This trust can lead to lower borrowing costs and increased business opportunities,
ultimately boosting profitability and ROA.

SIZ has shown to have a negative relationship with the ROA. Larger banks often face challenges
related to operational complexity and increased competition, which can erode profit margins.
Additionally, managing larger asset bases may require more resources and carry higher risks, such
as exposure to macroeconomic fluctuations. Smaller banks, on the other hand, may benefit from
nimbleness and lower overhead costs.

Higher expected INF has an adverse economic impact on the banking sector. Rising inflation
reduces the disposable income of households leaving them with lower funds to set aside for
 P a g e | 12

investment. This adversely affects the deposits in banks reducing the lending ability of banks.
Moreover, due to the fisher effect, high inflation contributes to higher cost of capital for the banks.

The positive relationship of EXR and ROA suggests that some banks may benefit from exchange
rate movements, potentially through activities such as foreign currency trading or international
lending. When exchange rates fluctuate significantly, banks with effective strategies for managing
currency risk can capitalize on these movements to generate additional income, which contributes to
higher profitability, as reflected in the positive correlation with ROA. However, it's important to
note that exchange rate dynamics can also introduce volatility and risk, so effective risk
management is crucial for success in this area.

Table 05: Regression estimates – Islamic banks

(1) (2) (3)
VARIABLES POLS RE FE
ROA ROA ROA

LOANTA 0.00678 0.00678 0.0429*

(0.00813) (0.00813) (0.0126)
LTA -0.0114 -0.0114 -0.0374*
(0.00737) (0.00737) (0.00880)
AU 0.0605* 0.0605* 0.101
(0.0344) (0.0344) (0.0370)
LLP 0.00401 0.00401 -0.00358
(0.00265) (0.00265) (0.00285)
T1 0.0259 0.0259 -0.0129
(0.0256) (0.0256) (0.0278)
SIZ -0.000276 -0.000276 0.00866
(0.000412) (0.000412) (0.00309)
INF -0.0200 -0.0200 -0.192
(0.0749) (0.0749) (0.0828)
EXR 0.00913 0.00913 0.0321
(0.0190) (0.0190) (0.0190)
Constant 0.0124 0.0124 -0.219
(0.0133) (0.0133) (0.0823)

Observations 25 25 25
R-squared 0.785
Number of id 5 5 5
Standard errors in parentheses
*** p<0.01, ** p<0.05, * p<0.1
 P a g e | 13

Table 05 shows the impact different variables have on the ROA of an Islamic bank. The results
from the FE model show that LOANTA, LTA, AU, SIZ, and INF have statistically significant
impact on the profitability of Islamic banks.

The Hausman test shown in Fig 02 returned a chi2 p-value 0.00. This means that the individual
effects are correlated. Fixed Effects model shows a better picture than the Random Effects model.

Fig 02: Hausman statistics – Islamic banks

Coefficients
(b) (B) (b-B) sqrt(diag(V_b-V_B))
FE RE Difference S.E.

loanta .0428989 .0067784 .0361205 .0095787

lta -.0373827 -.0114194 -.0259633 .0048078
au .1011064 .0604771 .0406292 .0134037
llp -.0035766 .0040113 -.0075879 .0010494
t1 -.0129449 .025931 -.0388759 .0108413
siz .0086636 -.0002757 .0089392 .0030582
inf -.1923669 -.0199588 -.1724081 .0351118
exr .0321494 .0091316 .0230178 .

b = consistent under Ho and Ha; obtained from xtreg

B = inconsistent under Ha, efficient under Ho; obtained from xtreg

Test: Ho: difference in coefficients not systematic

chi2(8) = (b-B)'[(V_b-V_B)^(-1)](b-B)
= 312.29
Prob>chi2 = 0.0000
(V_b-V_B is not positive definite)

Source: STATA output

Since the Hausman statistics say that the FE model shows better results, we consider the FE model
to explain the variables.

Similar to conventional banks the LONTA ratio is positively correlated with the ROA under both
POLS, RE, and FE models. Greater investment generates higher returns of the Islamic banks in the
form of shared profits.

LTA adversely affects the profitability of Islamic banks as well. When the proportion of liabilities
becomes substantial relative to total assets, it may lead to increased funding costs and reduced profit
margins. This can lower ROA, as a significant share of assets is allocated to servicing liabilities.

Islamic banks generate revenue primarily from compliant financial services and products. When
AU is higher, it suggests that a bank is effectively utilizing its assets to generate income, which can
 P a g e | 14

boost its ROA. This positive correlation reflects efficient operational management, successful
product offerings, and potentially a diverse range of revenue sources, contributing to higher profits.

While the coefficient for LLP is negative but not statistically significant, its suggests that when
Islamic banks allocate a larger proportion of their net interest revenue to cover potential loan losses,
it negatively impacts their profitability, as reflected in a lower ROA.

While T1 capital represents a bank's core capital and financial strength, its negative correlation with
ROA may indicate that Islamic banks with substantial capital buffers may face challenges in
efficiently deploying their capital to generate returns. Although not statistically significant, it
suggests that maintaining high capital adequacy levels, while essential for stability, may potentially
hinder profitability due to underutilization or conservative investment strategies, impacting the
bank's ROA negatively.

The positive relationship between SIZ and ROA in Islamic banks suggests that, on average, larger
Islamic banks, as measured by the natural logarithm of total assets, tend to have higher profitability,
as indicated by a higher ROA. This positive correlation implies that larger Islamic banks, with more
extensive asset bases, may be more efficient at generating returns on those assets, or they might
have advantages in terms of economies of scale, diversified operations, or access to a larger
customer base.

Inflation erodes the real value of assets and can lead to higher borrowing costs, both of which can
compress profit margins. For Islamic banks, which adhere to Sharia principles, inflation can pose
challenges in maintaining the purchasing power of assets while complying with Islamic financial
principles. Therefore, the negative correlation between inflation and ROA suggests that managing
the effects of inflation is crucial for sustaining profitability in the context of Islamic banking.

Islamic banks may benefit from exchange rate movements, potentially through activities such as
foreign currency trading or international lending. When exchange rates ch ange significantly, banks
with effective strategies for managing currency risk can capitalize on these movements to generate
additional income, contributing to higher profitability as reflected in the positive correlation with
ROA.

5. FINDINGS & CONCLUSION

 P a g e | 15

The analysis of the impact of various variables on the profitability of conventional and Islamic
banks reveals some interesting similarities and differences between these two types of financial
institutions. It's important to note that the results are based on regression models, and while they
provide valuable insights, they should be interpreted cautiously.

5.1 Similarities
1. Loan-to-Asset Ratio (LOANTA): In both conventional and Islamic banks, a positive correlation
exists between LOANTA and ROA. This suggests that both types of banks benefit from providing
loans as it contributes positively to their earnings. Interest income generated from loans appears to
be a key driver of profitability for both.

2. Liabilities-to-Asset Ratio (LTA): Conversely, LTA has a negative impact on the ROA for both
conventional and Islamic banks. This finding underscores the importance of managing liabilities
effectively, as higher liabilities can lead to increased interest expenses, which can erode
profitability.

3. Asset Utilization (AU): AU positively affects ROA in both types of banks. This implies that
banks that efficiently generate revenue from their assets tend to have higher profitability. Effective
asset utilization is a common driver of profitability in the banking sector, irrespective of the banking
model.

5.2 Differences
1. Loan Loss Provisions (LLP): In conventional banks, LLP is negatively correlated with ROA,
indicating that setting aside more funds to cover potential loan losses can reduce reported profits
and ROA. In contrast, for Islamic banks, LLP does not show a statistically significant impact on
profitability. This suggests that Islamic banks might have different risk management practices or
that the relationship is less pronounced in this context.

2. Tier 1 Capital (T1): The impact of T1 capital on ROA differs between the two types of banks.
In conventional banks, a positive relationship exists, highlighting the importance of strong capital
adequacy. However, in Islamic banks, T1 capital is not statistically significant. This suggests that
while capital strength is crucial for conventional banks, it may not have the same impact on the
profitability of Islamic banks, potentially due to their unique operational and financial structures.

3. Size (SIZ): Size, measured by total assets, has a negative correlation with ROA in conventional
banks, indicating that larger banks may face challenges related to operational complexity and
 P a g e | 16

competition. In contrast, in Islamic banks, size has a positive impact on ROA, suggesting that larger
Islamic banks tend to be more efficient at generating returns, possibly due to economies of scale
and a broader customer base.

4. Inflation (INF): The effect of inflation on profitability differs significantly between the two
banking models. In conventional banks, high inflation is negatively correlated with ROA, as it
reduces disposable income and increases the cost of capital. For Islamic banks, the negative
relationship is even stronger, emphasizing the importance of managing inflation's impact on the real
value of assets and compliance with Sharia principles.

5. Exchange Rate (EXR): Exchange rate movements have a positive impact on ROA in
conventional banks, indicating potential benefits from activities like foreign currency trading. In
Islamic banks, the relationship is positive but not statistically significant, suggesting that these
banks may not rely as heavily on such activities for profitability.

While there are common factors that drive profitability in both conventional and Islamic banks,
such as efficient asset utilization and the impact of liabilities, there are notable differences. These
differences stem from the unique characteristics and principles governing Islamic banking, such as
the treatment of risk provisions, the role of capital adequacy, and the influence of inflation.
Understanding these distinctions is vital for policymakers, regulators, and banking institutions to
tailor strategies that enhance the financial health and sustainability of each banking model.
Moreover, it highlights the importance of recognizing the diverse nature of the global banking
industry, where different models coexist, each influenced by its unique principles and economic
conditions.
 P a g e | 17

REFERENCES
Bashir, A. M. (2000). Assessing the performance of Islamic banks: Some evidence from the Middle
East. Journal of Economic Behavior & Organization, 51(4), 463-481.

Čihák, M., & Hesse, H. (2008). Islamic banks and financial stability: An empirical analysis.
Journal of Financial Services Research, 34(2-3), 151-177.

Wasiuzzaman, S., & Tarmizi, H. A. (2010). Determinants of Islamic and conventional bank
profitability in Malaysia. Journal of Financial Reporting and Accounting, 8(2), 128-146.

Mohamad, S., Saad, N., & Hanafi, N. (2011). Bank-specific and macroeconomic determinants of
Islamic bank profitability: The case of Malaysia. Global Economy and Finance Journal, 4(1),
103-122.

Al-Qudah, O., & Jaradat, M. (2013). The effect of bank size on profitability: Evidence from Jordan.
International Journal of Business and Management, 8(19), 63-69.

Kader, H. A., & Asarpota, H. A. (2007). Comparative performance of Islamic versus conventional
banks: Empirical evidence from the UAE. Global Review of Islamic Economics and Business,
6(2), 63-79.

Samad, A. (2004). Performance of interest-free Islamic banks vis-a-vis interest-based conventional

banks of Bahrain. IIUM Journal of Economics and Management, 12(1), 41-57.

Akhter, W., Ali, A., & Akhtar, S. (2011). A comparative performance analysis of conventional and
Islamic banks: Empirical evidence from Pakistan. European Journal of Economics, Finance
and Administrative Sciences, 31, 108-115.