Abstract

The objective of this study is to examine total factor productivity changes (TFPCH) in Islamic and conventional banks to determine whether they exhibit progression or regression. As earlier studies have focused mainly on productivity in conventional banks rather than Islamic banks, the current study aims to bridge the gap in the literature by investigating both types of bank in the Middle East, Southeast Asia, and South Asia. A total of 385 Islamic and conventional banks from 18 countries were selected, with data acquired for the period from 2008 to 2017. Panel data analysis was undertaken using DEA-based MPI to investigate the impact of selected determinants of banks’ productivity, as indicated by TFPCH. The results from both the t-test and nonparametric tests revealed that Islamic banks are more productive than conventional banks, which can be attributed to the increase in efficiency changes. However, no statistically significant difference in productivity exists between the types of bank. The main contribution of this study is that it provides not only corroboration for previous studies but also additional insight into bank productivity in Islamic and conventional banks, which will be important to banks, regulators, investors, and researchers.

Keywords: Total factor productivity change, Malmquist productivity index, Islamic banks, Conventional banks, Middle East, Southeast Asia, South Asia.

JEL Classifications: G21; G28.

Received: 9 June 2020 / Revised: 13 July 2020 / Accepted: 18 August 2020/ Published: 7 September 2020

Contribution/ Originality

This study is one of very few that have investigated the level of productivity in Islamic and conventional banks sector. It specifically focuses on countries in the Middle East, Southeast Asia, and South Asia, which are representative of global Islamic banking and finance.

1. INTRODUCTION

As a result of the global financial crisis (GFC),which occurred between mid-2007 and early 2009, Islamic banking has attracted significant interest and attention as an alternative to conventional banking–especially after investment banks collapsed(Rosman, Wahab, & Zainol, 2014). Likewise, the deterioration in banks’ performance during the more recent financial crises has encouraged academia, financial markets, and banks to investigate the factors associated with performance to avoid the adverse effects that threaten and contribute to potential instability in the financial markets.

Nonetheless, the banking sector continues to grow, at least until another form of banking emerges, and both Islamic and contemporary economists are becoming more interested in Islamic banking. Islamic banksare able to not only provide Muslims with institutions that follow the Islamic legal code,Shariah(Rosman et al., 2014)but also reduce the risks in financial transactions, which affects economic growth(Hassan & Aliyu, 2018).
In principle, the Islamic financial system prohibitspaying or charging interest, speculation, uncertainty (gharar), and transactions related to alcohol, tobacco, pornography, and any activity considered detrimental to society(Hassan & Aliyu, 2018). The theoretical differences between Islamic banks offeringShariah-compliant financeandconventional banksappear in the levels of complexity, agency costs, and maturity and development(Beck, Demirgüç-Kunt, & Merrouche, 2013);other differences include risk-taking, interest rates, income streams, and size (Habib, 2018). However, both Islamic and conventional banks prioritize profitability by focusing on productivity.

Consequently, several studies have analyzed the efficiency of Islamic bankstoassess their performance(Kamarudin, Sufian, Loong, & Anwar, 2017a; Rosman et al., 2014; Said, 2013; Sufian & Kamarudin, 2017; Sufian, Kamarudin, & Md. Nassir, 2017; Wanke, Azad, Kalam, Barros, & Hassan, 2016) , but few have examined productivity of neither Islamic norconventional banksacting as intermediaries (Kamarudin et al., 2017a). Thus, this studyaims tocontribute a better understanding of banks’ productivity to the existing body of literature.

As Siddiqi (2006)asserted that Islamic economic and financial theories were still underdeveloped, this study uses real-life data to validate foundational theories of productivity,which,along with profitability and growth, is a crucial dimension in assessing the broad concept of financial performance(Bottazzi, Secchi, & Tamagni, 2008). Profitability, which is required to maximize the shareholder wealth,reflects overall efficiency;however, to generating increased profits, productivity is essential.Indeed, Bottazzi et al. (2008)revealed that high productivity can lead to high profitability.

On a global scale, Islamic banking occupies a small share of the financial market, but this is rapidly expanding in many regions, particularly Asia and the Middle East(International Monetary Fund, 2015). According to Houben (2003)and Kamarudin et al. (2017a), though, SoutheastAsia is neglected by researchers across the world, despite its rising Muslim population. However, as Islamic finance becomes a greater part of the global capital market, it has the distinct potential to contribute to economic growth(Imam & Kpodar, 2016);hence, it is important that Islamic banks remain productive to be competitive. As such, the current study benefits research in this fieldby comparing the productivity of Islamic and conventional banks, focusingon three regions: South Asia (SA), Southeast Asia (SEA), and Middle East (ME). It will also continue the ongoing debate on which are more productive, Islamic orconventional banks. The research question is thus whether the productivity of Islamic banks differ from that of conventional banks?

This paper begins with a brief review of related studies, followed by a description of the sources of data and methodology, a discussion of the empirical results, and finally the conclusion.

2. LITERATURE REVIEW

The role of the conventional banking sector as a financial intermediary cannot be overlooked considering itsinfluence on stable economic growth and development. Islamic banks plays a similar but slightly different role,and are therefore considered a replacement or an alternative source of banking, albeit Shariah-compliant, products and services.

To date, no definite decisionhas been reached on whether Islamic banks should be more productive, or efficient,than theirconventional counterparts(Beck et al., 2013). Islamic banksbase their financial decisions on the productivity of the project in which it invests, meaning productivity is extremely important to ensure high profitability. Moreover, the Shariah Advisory Council (SAC) plays a key part in this respect byconfirming stakeholders’ Shariah-compliant behavior and being responsible for minimizing information asymmetry and agency costs within Islamic banks.

As, according toJensen and Meckling (1976),a conflict of interest between the principals (shareholders) and agents (bank management) can influence organizational performance, information asymmetry and agency conflicts should occur less often in Islamic banks(Hussain, Kamarudin, Thaket and Salem, 2019; Toumi, Louhichi, and Vivian, 2012). In fact, the SAC’sexternal monitoring can prevent agency conflicts and reduce agency costs,thereby increasing the efficiency, andso productivity, as demonstrated by Ang, Cole, and Lin (2000). However, the opposite may occur given that the productivity dimensions—such as complexity, and maturity and development—exert distinctly different effects on Islamic andconventional banks.

Kopleman (1986) defined productivity as the relationship between the amount of physical output(s)produced by a certain amount of physical input(s): total production (output) is influenced by the amount of capital invested and labor involved.Fare, Grosskopf, Norris, and Zhang (1994)asserted that productivitycould be furtherdecomposedinto changes in efficiency, or the catching-up effect, and changes in technology, or innovation, assuming that the outputs are equivalent to the inputs.The total factor productivity (TFP) growth indexmeasures the changes, or innovation, in technology, which can be considered as a change in performance that can be adjusted by altering a chosen input. Basically, higher productivity means higher profitability (Alaeddin et al., 2018;Kamarudin, Hue, Sufian, & Anwar, 2017b; Kamarudin et al., 2017a; Sufian, 2012; Sufian & Kamarudin, 2014; Sufian & Kamarudin, 2015) : when banks increase their productivity, they generate additional output from a given amount of input. The Cobb–Douglas production function is thusused in this study to compare productivity levels between Islamic and conventional banks in SA, SEA, and the ME.

There have been previous comparative studies with varying findings:some show that Islamic banks are significantly more productive thanconventional banks, othersshow the opposite,while a few show no difference.More recently, Alexakis, Izzeldin, Johnes, and Pappas (2018) reported that both Islamic and conventional banks experienced a decline in productivity, thoughto a greater extent in the latter, during 2008/09. Maredza and Ikhide (2013)stated that this was probably due to GFC in the Gulf Cooperation Council (GCC) banking sector. The results also indicated that there were differences in technological changes and efficiency between GCC Islamic banks, possibly because a number of mature banks do exist in a developing banking sector, although itmay be owing to the various financial products, bank status, client base, and innovation.

On the other hand,Rodoni, Salim, Amalia, and Rakhmadi (2017)conducted a comparative study of productivity and efficiency in31 Islamic banks across Pakistan, Indonesia, and Malaysia between 2009 and 2013. Using the Malmquist productivity index (MPI) and data envelopment analysis (DEA)to measureproductivity and efficiency, respectively, they found that the Malaysian banking sector was far more efficientthan in Indonesia, while Pakistan was close to 100% efficient.Kamarudin et al. (2017a)undertook a similar studyof productivity in 29 Islamic banks in Malaysia, Indonesia, and Brunei between 2006 and 2014. Using a nonparametric DEA-based MPI toestimate TFP, they found that no statistical difference in productivity and efficiency between locally and internationally managed banks with similar technology and client base.

In another study,Doumpos, Hasan, and Pasiouras (2017) investigated the financial robustness of 347 conventional banks, 101 Islamic banks, and 52 Islamic windows within conventional banksacross 57 member countries of the Organisation of Islamic Cooperation (OIC)between 2000 and 2011. They found that the individual financial ratios of differedbetween banks, but no statistically significant difference in overall financial strength was evident. Furthermore,Mobarek and Kalonov (2014)compared the performance of 101 Islamic and 307conventional banks in 18 member countries during the pre-GFC period(2004–2006) and actual GFC (2007–2009). DEA and stochastic frontier analysis (SFA) of cross-sectional data indicated that the efficiency of conventional banks between 2006 and 2009 was higher than Islamic banks;however, this was an unfair comparison because the mean value of the efficiency score was larger forconventional banks.

Finally, Kamarudin, Nordin, Muhammad, and Hamid (2014)examined the efficiency—in terms of the profit, revenue, and costs—of 47 conventional and 27 Islamic banks between 2007 and 2011 in the GCC region. Taking anintermediation approach,DEArevealed that conventional banks exhibited higher levels of efficiency in all threeareas. Moreover, the results suggested that the primary determinant for the level of profit efficiency was the level of revenue efficiency.

Thus, most of the earlier studies have reported disparate findings on the level of efficiency in Islamic and conventional banks worldwide,while studies onproductivity levels inthose banks are less common, particular in Asian regionswhere Islamic banks are prevalent(Kamarudin et al., 2017a). Hence, this studyintends to offer empirical evidence for the productivity levelsof Islamic and conventional banks.

3. METHODS

3.1 Data Sources

The data source for this study was the Fitch Connect online database, which comprises financial reports, accounting ratios, and credit ratings of over 30,000 Islamic and conventional banks worldwide. Data were extracted for Islamic and conventional banks in SA, SEA, and the ME between 2008 and 2017 (Khan & Bhatti, 2008). To facilitate the comparison, all currencies were expressed in US dollars, while to prevent bias, a dummy variable representing the 2008–2009 GFC was applied.

A total of 385 banks (66 Islamic and 319 conventional) were selected from 18 countries (3 in SA, 4 in SEA, and 11 in the ME) with dual banking systemswere selected, as represented in Table 1. All investment banks, and insurance and finance companies were excluded to maintain homogeneity.

Table-1. Bank data.

No.	Country	Income Group*	Region	No. of Islamic Banks	No. of Conventional Banks
1	Bahrain	High	Middle East	8	12
2	Egypt	Lower Middle	Middle East	1	23
3	Iran	Upper Middle	Middle East	1	8
4	Iraq	Upper Middle	Middle East	1	3
5	Jordan	Upper Middle	Middle East	2	11
6	Kuwait	High	Middle East	1	4
7	Lebanon	Upper Middle	Middle East	2	31
8	Oman	High	Middle East	2	7
9	Qatar	High	Middle East	3	5
10	Saudi Arabia	High	Middle East	3	8
11	UAE	High	Middle East	7	14
12	Brunei	High	South East Asia	1	1
13	Indonesia	Lower Middle	South East Asia	8	92
14	Malaysia	Upper Middle	South East Asia	13	31
15	Singapore	High	South East Asia	1	8
16	Bangladesh	Lower Middle	South Asia	4	37
17	Maldives	Upper Middle	South Asia	1	1
18	Pakistan	Lower Middle	South Asia	7	23
	Total			66	319

Note: Income levelsextracted from World Bank Open Data.
Source: Fitch connect.

3.2. Data Envelopment Analysis-Based Malmquist Productivity Index

DEA was developed byCharnes, Cooper, and Rhodes (1978),who posited that the greater the output generated by the inputs, the greater the efficiency of the production process.The method has since become a recognized performance measurement tool across all fields of management science, as revealed by.

Efficiency and productivity are interrelated in the current study: changes in the former are affected by alterations in the latter. The input–output ratio can be used to determine productivity, but it is important to remember that efficiency fails to take account of the time taken by the production process. Therefore, MPI,sometimes referred to as TFP, has been widely adopted for DEA in a range of countries and sectors due to its ability to assess any change in efficiency or technology in terms of progress or regress over time. Output-based MPIis used to not only measure and understand the change in productivity of banks but also to determine the change in TFP (TFPCH), which can be decomposedinto technical change (TCH) and efficiency change (EFCH). In addition,EFCH can be further decomposed into changes in scale efficiency (SECH) and pure technicalefficiency (PTECH). Figure 1 illustrates these interactive relationships.

Figure-1. Interactive relationship between MPI efficiency indices.

Source:Fare et al. (1994).

In the analysis, 2007 was set as the benchmark year, with the MPI and all its components starting with a value of 1; the efficiency scores were constrained within the lower bound of0 and upper bound of 1. Hence, banks with an efficiency score lower/higher than 1after 2007 perform below/above the efficiency frontier (i.e., when the decision-making unit (DMU) operates at the optimal efficiency). Furthermore, efficiency scores represent the radial distance between the efficiency frontier and DMU under consideration.

3.3. Specification for Input and Output of Banks

To explore the productivity of banks, DEA has been adopted for this study because of its widespread use and sustained relevance and effectiveness over 40 years,appearing in more than 1000 published studies per year (Emrouznejad & Yang, 2018). In addition, an intermediation approach has been takento classify the input and output of banks, once more owing to its use in many studies(Bhatia, Basu, Mitra, & Dash, 2018; Kamarudin, Sufian, & Nassir, 2016) as the initial stage of DEA, as well as the significant roleplayed by banks as financial intermediaries.

Inputs and outputs were selected for the current study by following the process described in several studies (Alexakis et al., 2018; Colwell & Davis, 1992; Kamarudin et al., 2017a; Sufian & Habibullah, 2014) . Table 2 presents all the variables, derived from the MPI model,that were examined through nonparametric DEA in the initial stages of the analysis.

Table-2.Bankinput and output variables.

Variable	Symbol	Variable Name	Definition
Outputs	y1	Loans	Net loans
	y2	Investments	Total securities
Inputs	x1	Deposits	Total deposits, money market, and short-term funding
	x2	Labor	Personnel expenses
	x3	Physical capital	Book value of fixed assets

Note:According to Casu and Girardone (2006) andAriss (2010), loans are identified as financing activities by Islamic banks.

According to Banker and Datar (1989) and Cooper, Seiford, and Tone (2002), the number of inputs and outputs selected must meet a predetermined assumption prior to performing DEA:

4. RESULTS AND DISCUSSIONS

This study investigates the TFP levels of Islamic and conventional banks in SA, SEA, and the ME with DEA-based MPI. To determine the variation in productivity (y-axis) between Islamic and conventional banks, a parametric (t-test) and nonparametric (Mann–Whitney [Wilcoxon] and Kruskal–Wallis) testswere performed. Table 3presents the summary statistics for eachvariable, in US $m, which were used to construct the efficiency frontiers for banks’ productivity.

Table-3. Summary statistics forinputs and output variables in the DEA model (US$m).

4.1. Productivity Decomposition of Islamic and Conventional Banks

Table 4shows the geometric mean scores of both TFPCH and its components for all banks (Panel A), conventional banks (Panel B), and Islamic banks (Panel C). The performance of the banks can thus be assessed for each year between 2007 and 2017.

Table-4. MPI decompositions.

Year	Indices
Year	TFPCH	TCH	EFCH	PTECH	SECH
*Panel A: All banks*
2007	1.000	1.000	1.000	1.000	1.000
2008	1.038	0.808	1.286	1.289	0.997
2009	0.936	1.149	0.814	0.856	0.952
2010	0.949	1.041	0.911	0.991	0.920
2011	1.058	0.898	1.178	1.091	1.080
2012	0.995	0.978	1.018	1.013	1.004
2013	1.080	1.056	1.023	1.074	0.952
2014	0.916	0.999	0.917	0.912	1.005
2015	0.954	0.981	0.973	0.953	1.021
2016	1.052	0.990	1.063	1.062	1.001
2017	0.662	0.599	1.101	1.109	0.993
Geometric mean	0.919	0.908	1.012	1.019	0.994

*Panel B: Conventional banks*
2007	1.000	1.000	1.000	1.000	1.000
2008	1.025	0.790	1.297	1.294	1.002
2009	0.928	1.163	0.798	0.842	0.948
2010	0.997	1.054	0.945	1.021	0.925
2011	1.066	0.897	1.189	1.102	1.079
2012	0.949	0.984	0.965	0.973	0.992
2013	1.093	1.045	1.045	1.103	0.948
2014	0.926	0.983	0.941	0.921	1.023
2015	0.991	0.995	0.995	0.992	1.004
2016	0.961	0.983	0.977	0.976	1.001
2017	0.679	0.592	1.148	1.154	0.995
Geometric mean	1.010	0.915	1.016	0.994	0.924

*Panel C: Islamic banks*
2007	1.000	1.000	1.000	1.000	1.000
2008	1.140	0.943	1.209	1.255	0.963
2009	0.988	1.059	0.934	0.954	0.978
2010	0.716	0.967	0.739	0.833	0.888
2011	1.017	0.903	1.124	1.037	1.086
2012	1.259	0.947	1.329	1.245	1.068
2013	1.018	1.106	0.921	0.947	0.972
2014	0.869	1.078	0.806	0.872	0.924
2015	0.797	0.917	0.870	0.789	1.104
2016	1.648	1.023	1.611	1.609	1.001
2017	0.581	0.634	0.898	0.913	0.984
Geometric mean	0.959	0.945	1.013	1.016	0.997
Note: The table presents the geometric means for total factor productivity change (TFPCH), and its mutually exhaustive components of technical change (TCH) and efficiency change (EFCH),which is further decomposed into pure technical efficiency change (PTECH) and scale efficiency change (SECH).

FromTable 4 Panel A, it is evident that, on average, all banks exhibited TFPCH regression of –8.1% (0.919) over the whole study period,witha low in 2017 when regression was–33.8% (0.662) and a high in 2013 when progressionwas 8.0% (1.080). The –8.1% (0.919) average regression could be mainly due to the –9.1% (0.908) decrease in TCH, as there was a 1.2% (1.012) increase in EFCH, which appears to have been causedmore by PTECH than SECH. Therefore, all banks were efficient in managing cost control, though operating at a non-optimal scale.The results for conventional banks are shown in Table 4 Panel B, in which TFPCH was a 1.0% (1.010) progression on average;the highest progression occurred in 2013 with a TFPCH of 9.3% (1.093). The1.6% (1.016) increase in EFCHled to the progression in conventional banks, as TCH decreased by–8.5% (0.915). Moreover, the reason for the increase in EFCH was mainly managerial (PTECH) rather than scale of operation (SECH). Likewise, Table 4 Panel Cshows the results for Islamic banks. On average, the TFPCH reflectsa–4.1%(0.959) regression, indicating a lower level of productivity thanconventional banks. The results reveal that the lowest level of productivity occurred in 2017 with a TFPCH regression of 41.9% (0.581),while the highest level was reached in 2016 with a progression of 64.8% (1.648). The cause of the overall TFPCH regression appears to be the decrease of–5.5% (0.945) in TCH, whereas the EFCH seems to have increasedby 1.3% (1.013). As in conventional banks, the increase in EFCH of Islamic banks was due to management rather than scale of operation.

Therefore, productivity progress in conventional banks stems primarily from improvement in efficiency (EFCH), while productivity regress in Islamic banks results mainly from technological stagnation (TCH). Overall, it is evident that both conventional and Islamic banks are more productive in cost control, but are operating on the wrong scale.

4.2. Progressive and Regressive Productivity in Islamic and Conventional Banks

To control for possible outliers, Table 5reports the trend in the number and percentage of all banks that experienced a productivity progress or regress during the study period. It can be seen fromTable 5 Panel Athat only 121 (35.80%) of all banks experienced productivity progress in 2008,but then increased substantially to a maximum of 206 (56.13%) banks by2013, before declining to 142 (38.38%) banks in 2017. Technical progress also increased from 157 (46.45%) banks in 2008 to 232 (63.22%) in 2013,and then drastically declined to 76 (20.54%) banks in 2017. These figures were validated by the banks that experienced technical regress, which declined from 174 (51.48%) banks in 2008 to 134 (36.51%) 2013,then radically rose to 294 (79.46) in 2017.

The trend for justconventional banks over the study period is shown in Table 5 Panel B. The results follow a similar trend to that in Panel A: productivity progress increased from 99 (33.45%) of banks in 2008 to 175 (55.73%) in 2014, then decreased gradually to 118 (33.44%) in 2017; technical regress decreased from 159 (53.72%) banks year 2008 to 114 (37.50%) in 2013,followed by an increase to 248 (80.78%) in 2017, although there was a sudden, sharp rise between 2010 and 2012, up to 210 (70.47%).As can be seen inTable 5 Panel C,of the trend in Islamic banks over the study period fluctuates. Between 2008 and 2017, there were two periods of productivity progress, albeit unstable: after a decline from 22 (52.38”) of banks in 2008, a substantial increase to 31–35 (52.4–55.56%) occurred between 2010 and 2013, followed by another slight decline in 2014 to 27 (41.54%) and gradual rise to a peak of 45 (71.43%) in 2016,ending in a sharp decline to 24 (38.10%) in 2017. Likewise, there were significant increases and declinesin technical progress during the first half of the study period, rising from 25 (59.52%) of banks in 2008 to its peak in 2013 of 43 (68.25%), followed by smaller peaks and troughs before drastically declining to 17 (26.98%) by 2017. In contrast, the fluctuations in technical regress consisted of a series of steady increases and decreases over the study period, starting with15 (37.71%) of banks in 2008,experiencing a peak at 47 (79.66%) in 2012, and endingwith 46 (73.02%) in 2017.A more comprehensive analysis of productivity, by not only type of bank but also income level ofcountry,was performed. Figure 2illustrates the trend in productivity levelsof Islamic and conventional Banks from 2008 to 2017,revealing that Islamic banks outperformed conventional banks for most of the study periodafter the 2008–2009 GFC. Nevertheless, on average, both Islamic and conventional banks had productivity indices above 1.00,demonstrating that all banks experienced annual productivity progress. However, this trend was uneven between 2008 and 2017, particularly in Islamic banks where sharp peaks were reached in 2012 and 2016.Thistrend can be explained by the growing Muslim population, particularly in SEA,leading to the ethical character and financial stability of Islamic finance becoming popular as an alternative to conventional banks(Komijani & Hesary, 2018). Moreover, the global sukuk(Shariah-compliant bonds) market reached its peak in 2012,due to its growing popularity in the corporate sector and among sovereigns in SEA for raising funds, mainly in Malaysia and Indonesia,which enabled theregion to dominate over 70% of the world’s sukuk issuances. Furthermore, in 2016, sukuk issuances played an important role in financing infrastructure development inSA, which was essential for economic growth (Asian Development Bank, 2017;Komijani & Hesary, 2018).

Table-5. Number and percentage of banks experiencing progressive and regressive productivity.

Notes: Productivity growth: TFPCH > 1; Productivity Loss: TFPCH< 1; Productivity Stagnation: TFPCH = 1.

Figure-2. Trend in productivity levels for conventional and Islamic banks from 2008 to 2017.

Source: Banks’ annual reports.

Figure-3. TFPCH for Islamic banks in different income groups from 2008 to 2017.

Source: Banks’ annual reports and authors’ calculations.

Figure 3 represents the TFPCH of Islamic banks in high-, upper middle-, and lower middle-income countries from 2008 to 2017. Overall, high-income countries have the lowest TFPCH, TCH, EFCH, and PECH,while lower middle-income countries have the highest TFPCH, EFCH, PECH, and SECH, andthe upper middle-income countries have slightly higher TCH than the other countries.

This can be explainedby the increasing number of Islamic banks in upper middle-and lower middle-income countries, such as Malaysia, Indonesia, Pakistan, and Bangladesh, where there are alsolarge Muslim populations whose income levels positivelyaffect the development of the Islamic banking sector(Boukhatem & Moussa, 2018). In addition, according to Boukhatem and Moussa (2018), those countries that adopted a hybrid legal system based on both common and Islamic (Shariah) law have been able to respond flexibly to changing macroeconomic conditions, which has contributed to the development of Islamic banks.

Theseconditions have created a highly competitive market between Islamic and conventional banks in these countries, which,according toAbedifar, Hasan, and Tarazi (2016), can motivate banks to be more innovative and increase the efficiency of the whole banking system.It is evident from Figure 3, TFPCH in all countries stems from EFCH,revealing managerial changes in the banks.

Figure-4. TFPCH for conventional banks in different income group from year 2008 to 2017.

Source: Bank annual report and authors own calculation.

Figure 4 represents the TFPCH of conventional banks for the same income groups over the study period. The upper middle-income countries have a slightly higher TFPCH than the others,much of whichis again due to EFCH,or in other words, managerial efficiency. This finding corresponds to that of Aluko and Ajayi (2018),who alsodiscovered that lower-more than high-income countries tend to have more efficient banks. This was explained by Ghosh (2016) as owing to high-income countries having a larger banking sectorin whichgreater competitive pressures result inhigher agency and overhead costs, and consequently, lower productivity.

4.3. Robustness Tests

Table 6 presents the results of the parametric (t-test) and nonparametric (Mann-Whitney [Wilcoxon] and Kruskal–Wallis) testsand reveals the significant difference between the productivity levels of Islamic and conventional banks in specific years and regions. As already observed,it appears that Islamic banks are slightly more productive than conventional banks across all regions;however, the difference is only statistically significant during 2016 (Panel I). On the other hand, the t-test results, confirmed by the nonparametric tests, in Panels A, C, F, and H suggest that conventional banks were relatively more productive, though only statistically significant in 2010 (Panel C).

In the ME, there wasgreater productivity progress in Islamic than conventional banks in seven years of the study period, andthe results are significant at the 1% level in 2016 (Panel I);however, when the reverse occurred in 2010, 2014, and 2015 (Panels C, G, and H), the greater progression in conventional banks is statistically significant in each year. In addition, Islamic banks experienced relatively more productivity progress thanconventional banks: in 2008 (at 1% significance level), 2012, and 2015, 2016 (at 10% significance level), and 2017 (Panels A, E, H, I, and J) in SEA; and in 2011(at 1% significance level), 2012, 2014, and 2016(Panels D, E, G, and I) in SA. Nonetheless, in SA,conventional banks exhibited more productivity progress in 2008, (at 10% significance level), 2009, 2010, 2013, 2015, and 2017 (Panels A, B, C, F, H, and J).
Further analysisof the other components of MPI are also shown in Table 6. Greater productivity progresswas generated mainly from TCH in both Islamic banks—statistically significant in 2008, 2011. 2013, and 2017 (Panels A, D, F, G, and J)—and conventional banks—statistically significant in 2010, 2012, and, 2015(Panels C, E, and H). Furthermore, EFCH in Islamic banks was higher in 2009, 2011, 2012, and 2016 (Panels B, D, E, and I), with PECH showing statistical significance in 2016, and lower in 2008, 2010, 2013, 2014, 2015, and 2017(Panels A, C, F, G, H, and J), where PECH was statistically significant in 2013 and 2017, and SECH in 2014, at the 5% level.

The overall results for all years and regions are provided in Panel K of Table 6, from which it can be inferred that Islamic banks are more productive than conventional banks (mean difference = 0.653) according to not only thet-test butalso the nonparametric tests,which is attributable to the progress in EFCH (mean difference = 0.589). However, the mean difference in TFPCH between Islamic and conventional banks is not statistically significant in any of the three regions studied. Furthermore, the test for equality of populations rejects the null hypothesis that the difference between Islamic and conventional banks is equal. The findingsof the current study thus corroborate those ofYahya, Muhammad, and Hadi (2012) and Doumpos et al. (2017): there is no statistically significant difference between the total factor productivity of Islamic and conventional banks.

There are several reasons for Islamic banks being more productive than conventional banks. First, the risk-sharing paradigm and higher asset quality in Islamic finance is more resilientto financial shocks(Beck et al., 2013;Darrat, 1988); Khan, 1986). Second, Islamic banks carry lower credit and insolvency risksbecause their bank charges and loan quality areless affected by fluctuations in interest rates(Abedifar, Molyneux, & Tarazi, 2013),whichwhy Islamic banks are more stable thanconventional banks.Third, the moral principles underpinning the Islamic financial system facilitate the sustainability of Islamic banks as well as enhance socioeconomic well-being through financial outreach(Aliyu, Yusof, & Naiimi, 2017). Finally, Islamic banks are risk averse in terms of capital investment in the real economy(Abedifar et al., 2016).

5. CONCLUSION

This study aims to contribute to the body of literature on banks’ performance. With the rapid increase in Islamic banks, it is imperative to study their productivity;thus,the total factor productivity change in Islamic and conventional banks across 18 countries in the Middle East, South Asia, and South Asiawhere dual-banking systems exist was analyzed. The data were analyzed using nonparametric DEA-based MPI, and the results tested through parametric (t-test) and nonparametric (Mann–Whitney and Kruskal–Wallis) tests. Theoretically, it has beenposited that the effect of various productivity determinants (e.g., levels of complexity, agency costs, and maturity and development)are distinctly different between Islamic andconventional banks; however, empirical estimation suggests that there is no statistically significant difference between their total factor productivity.

Following detailed analysis of specific years and individual regions, Islamic banks exhibited slightly greater productivity progress than conventional banks in almost every year in each region. In addition, the productivity progress of all banks could be attributed solely to the increase in efficiency changes, whichindicates that both Islamic and conventional banks are managerially efficient.

Table-6. Summary of parametric and nonparametric tests for conventional and Islamic banks.

Notes:Obs.: observations; ME: Middle East; SEA: Southeast Asia; SA: South Asia.
a, b, c indicate 1%, 5%, and 10% significance levels, respectively.
The figures in brackets indicate that the productivity means of conventional banks are higher than those of Islamic banks.

Table-6.(cont.)

Table-6.(cont.)

6. ABBREVIATIONS

CRS Constant Returns to Scale.
DEA Data Envelopment Analysis.
DMU Decision-Making Unit.
EFCH Efficiency Change.
GCC Gulf Cooperation Council.
GFC Global Financial Crisis.
ME Middle East.
MPI Malmquist Productivity Index.
OIC Organisation of Islamic Cooperation.
PTECH Pure Technical Efficiency Change.
SA South Asia.
SAC Shariah Advisory Council.
SEA Southeast Asia.
SECH Scale Efficiency Change.
SFA Stochastic Frontier Analysis.
TCH Technical Change.
TFP Total Factor Productivity.
TFPCH TFP Change.
VRS Variable Returns to Scale.

Funding: Authors offer special thanks to the following organizations that funded our research: 1) Fundamental Research Grant Scheme (FRGS), RMI File No: 600-IRMI/FRGS 5/3 (141/2019), sponsored by Ministry of Higher Education, Malaysia; 2) MARA University of Technology; 3) Research University Grant Scheme (RUGS)/Putra Grant, Vote No: 9632100, sponsored by University Putra, Malaysia; 4) Fundamental Research Grant Scheme (FRGS), Vote No: FRGS/1/2015/SS01/UPM/02/1 [5524716], sponsored by Ministry of Higher Education, Malaysia; 5) Putra Grant-Putra Graduate Initiative (GP-IPS), Vote No: 9651500, sponsored by University Putra, Malaysia; 6) University Grant Phase 2/2017, research code (PPP/FEM/0217/051000/10218), sponsored by Islamic Science University of Malaysia (USIM); and 7) USIM/YTI/FEM052002/41118, sponsored by Yayasan Tun Ismail Mohamed Ali Berdaftar (YTI)–Permodalan Nasional Berhad (PNB).

Competing Interests: The authors declare that they have no competing interests.

Acknowledgement: Authors would like to thank the editors and anonymous referees of this journal for their constructive comments and suggestions, which were a considerable help in improving the contents of this paper.

REFERENCES

Abedifar, P., Hasan, I., & Tarazi, A. (2016). Finance-growth nexus and dual-banking systems: Relative importance of Islamic banks. Journal of Economic Behavior & Organization, 132, 198-215. Available at: https://doi.org/10.1016/j.jebo.2016.03.005.

Abedifar, P., Molyneux, P., & Tarazi, A. (2013). Risk in Islamic banking. Review of Finance, 17(6), 2035-2096.

Alaeddin O., Altounjy, R., Zainudin Z. & Kamarudin, F. (2018). From physical to digital: Investigating consumer behaviour of switching to mobile wallet. Polish Journal of Management Studies, 17(2), 18-30. Available at:https://doi.org/10.17512/pjms.2018.17.2.02

Alexakis, C., Izzeldin, M., Johnes, J., & Pappas, V. (2018). Performance and productivity in Islamic and conventional banks: Evidence from the global financial crisis. Economic Modelling, 79, 1-14. Available at: https://doi.org/10.1016/j.econmod.2018.09.030.

Aliyu, S., Yusof, R. M., & Naiimi, N. (2017). The role of moral transaction mode for sustainability of banking business: A proposed conceptual model for Islamic microfinance banks in Nigeria. International Journal of Social Economics, 44(12), 2238-2256. Available at: https://doi.org/10.1108/ijse-07-2016-0205.

Aluko, O. A., & Ajayi, M. A. (2018). Determinants of banking sector development: Evidence from Sub-Saharan African countries. Borsa Istanbul Review, 18(2), 122-139. Available at: https://doi.org/10.1016/j.bir.2017.11.002.

Ang, J., Cole, R., & Lin, J. (2000). Agency costs and ownership structure. The Journal of Finance, 55(1), 1-39.

Ariss, R. T. (2010). Competitive conditions in Islamic and conventional banking: A global perspective. Review of Financial Economics, 19(3), 101-108. Available at: https://doi.org/10.1016/j.rfe.2010.03.002.

Asian Development Bank. (2017). Meeting Asia's infrastructure needs. Manila: Asian Development Bank.

Banker, R. D., & Datar, S. M. (1989). Sensitivity, precision, and linear aggregation of signals for performance evaluation. Journal of Accounting Research, 27(1), 21-39. Available at: https://doi.org/10.2307/2491205.

Beck, T., Demirgüç-Kunt, A., & Merrouche, O. (2013). Islamic vs. Conventional banking: Business model, efficiency and stability. Journal of Banking and Finance, 37(2), 433-447. Available at: https://doi.org/10.1016/j.jbankfin.2012.09.016.

Bhatia, V., Basu, S., Mitra, S. K., & Dash, P. (2018). A review of bank efficiency and productivity. Opsearch, 55(3-4), 557-600.

Bottazzi, G., Secchi, A., & Tamagni, F. (2008). Productivity, profitability and financial performance. Industrial and Corporate Change, 17(4), 711-751. Available at: https://doi.org/10.1093/icc/dtn027.

Boukhatem, J., & Moussa, F. B. (2018). The effect of Islamic banks on GDP growth: Some evidence from selected MENA countries. Borsa Istanbul Review, 18(3), 231-247. Available at: https://doi.org/10.1016/j.bir.2017.11.004.

Casu, B., & Girardone, C. (2006). Bank competition, concentration and efficiency in the single European market. The Manchester School, 74(4), 441-468. Available at: https://doi.org/10.1111/j.1467-9957.2006.00503.x.

Charnes, A., Cooper, W. W., & Rhodes, E. (1978). Measuring the efficiency of decision making units. European Journal of Operational Research, 2(6), 429-444. Available at: https://doi.org/10.1016/0377-2217(78)90138-8.

Colwell, R., & Davis, E. (1992). Output and productivity in banking. The Acandinavian Journal of Economics, 94, S111-S129. Available at: https://doi.org/10.2307/3440252.

Cooper, W., Seiford, L., & Tone, K. (2002). Data envelopment analysis: A comprehensive text with models, applications, references and DEA-solver software. Boston: Kluwer Academic Publishers.

Darrat, A. F. (1988). The Islamic interest-free banking system: Some empirical evidence. Applied Economics, 20(3), 417-425. Available at: https://doi.org/10.1080/00036848800000054.

Doumpos, M., Hasan, I., & Pasiouras, F. (2017). Bank overall financial strength: Islamic versus conventional banks. Economic Modelling, 64(C), 513-523. Available at: https://doi.org/10.1016/j.econmod.2017.03.026.

Emrouznejad, A., & Yang, G.-l. (2018). A survey and analysis of the first 40 years of scholarly literature in DEA: 1978-2016. Socio- Economic Planning Sciences, 61, 4-8. Available at: https://doi.org/10.1016/j.seps.2017.01.008.

Fare, R., Grosskopf, S., Norris, M., & Zhang, Z. (1994). Productivity growth, technical progress, and efficiency change in industrialized countries. American Economic Review, 84(1), 66-83.

Ghosh, A. (2016). Banking sector globalization and bank performance: A comparative analysis of low income countries with emerging markets and advanced economies. Review of Development Finance, 6(1), 58-70. Available at: https://doi.org/10.1016/j.rdf.2016.05.003.

Habib, S. F. (2018). Chapter 3: Islamic banking versus conventional banking. In S. F. Habib, Fundamentals of Islamic Finance and Banking (pp. 47-78): John Wiley & Sons.

Hassan, M. K., & Aliyu, S. (2018). A contemporary survey of Islamic banking literature. Journal of Financial Stability, 34(C), 12-43. Available at: https://doi.org/10.1016/j.jfs.2017.11.006.

Houben, V. J. (2003). Southeast Asia and Islam. The ANNALS of the American Academy of Political and Social Science, 588(1), 149-170.

Hussain, H.I., Kamarudin, F., Thaker, H.M.T., & Salem, M.A. (2019). Artificial Neural Network to model managerial timing decision: Non-linear evidence of deviation from target leverage. International Journal of Computational Intelligence Systems, 12(2), 1282-1294. Available at: https://doi.org/10.2991/ijcis.d.191101.002

Imam, P., & Kpodar, K. (2016). Islamic banking: Good for growth? Economic Modelling, 100(59), 387-401. Available at: https://doi.org/10.1016/j.econmod.2016.08.004.

International Monetary Fund. (2015). IMF annual Report 2015: Tackling challenges together. International Monetary Fund.

Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3(4), 305-360. Available at: https://doi.org/10.1016/0304-405x(76)90026-x.

Kamarudin, F., Hue, C. Z., Sufian, F., & Anwar, N. A. M. (2017b). Does productivity of Islamic banks endure progress or regress?: Empirical evidence using data envelopment analysis based Malmquist Productivity Index. Humanomics: The International Journal of Systems and Ethics, 33(1), 84-118. Available at: https://doi.org/10.1108/h-08-2016-0059.

Kamarudin, F., Nordin, B. A. A., Muhammad, J., & Hamid, M. A. A. (2014). Cost, revenue and profit efficiency of Islamic and conventional banking sector: Empirical evidence from gulf cooperative council countries. Global Business Review, 15(1), 1-24. Available at: https://doi.org/10.1177/0972150913515579.

Kamarudin, F., Sufian, F., Loong, F. W., & Anwar, N. A. M. (2017a). Assessing the domestic and foreign Islamic banks efficiency: Insights from selected Southeast Asian countries. Future Business Journal, 3(1), 33-46. Available at: https://doi.org/10.1016/j.fbj.2017.01.005.

Kamarudin, F., Sufian, F., & Nassir, A. M. (2016). Does country governance foster revenue efficiency of Islamic and conventional banks in GCC countries? EuroMed Journal of Business, 11(2), 181 - 211. Available at: https://doi.org/10.1108/emjb-06-2015-0026.

Khan, M. M., & Bhatti, M. I. (2008). Islamic banking and finance: On its way to globalization. Managerial Finance, 34(10), 708-725. Available at: https://doi.org/10.1108/03074350810891029.

Khan, M. S. (1986). Islamic interest-free banking: A theoretical analysis. Staff Papers, 33(1), 1-27.

Komijani, A., & Hesary, F. T. (2018). An overview of Islamic banking and finance in Asia. ADBI Working Paper Series, 853, 1-14.

Kopleman, R. (1986). Managing productivity in organizations: A practical, people-oriented perspective. New Delhi: McGraw Hill Book Company.

Maredza, A., & Ikhide, P. (2013). Measuring the impact of the global financial crisis on efficiency and productivity of the banking system in South Africa. Mediterranean Journal of Social Sciences, 4(6), 553-568.

Mobarek, A., & Kalonov, A. (2014). Comparative performance analysis between conventional and Islamic banks: Empirical evidence from OIC countries. Applied Economics, 46(3), 253-270. Available at: https://doi.org/10.1080/00036846.2013.839863.

Rodoni, A., Salim, M. A., Amalia, E., & Rakhmadi, R. S. (2017). Comparing efficiency and productivity in Islamic banking: Case study Indonesia, Malaysia and Pakistan. Al-Iqtishad: Jurnal Ilmu Ekonomi Syariah, 9(2), 227-242. Available at: https://doi.org/10.15408/aiq.v9i2.5153.

Rosman, R., Wahab, N., & Zainol, Z. (2014). Efficiency of Islamic banks during the financial crisis: An analysis of Middle Eastern and Asian countries. Pacific-Basin Finance Journal, 28, 76-90. Available at: https://doi.org/10.1016/j.pacfin.2013.11.001.

Said, A. (2013). Risks and efficiency in the Islamic banking systems: The case of selected Islamic banks in MENA region. International Journal of Economics and Financial Issues, 3(1), 66-73.

Siddiqi, M. (2006). Islamic banking and finance in theory and practice: A survey ofstate of the art. Islamic Economic Studies, 13(2), 2-48.

Sufian, F. (2012). For which option is credit risk more representative on China banks' total factor productivity: Efficiency change or technological progress? China Finance Review International, 2(2), 180-202.

Sufian, F., & Habibullah, M. S. (2014). Banks' total factor productivity growth in a developing economy: Does globalisation matter? Journal of International Development, 26(6), 821-852. Available at: https://doi.org/10.1002/jid.2897.

Sufian, F., & Kamarudin, F. (2014). The impact of ownership structure on bank productivity and efficiency: Evidence from semi-parametric malmquist productivity index. Cogent Economics and Finance, 2(1), 1-27. Available at: http://dx.doi.org/10.1080/23322039.2014.932700.

Sufian, F., & Kamarudin, F. (2015). Determinants of revenue efficiency of Islamic banks: Empirical evidence from the Southeast Asian countries. International Journal of Islamic and Middle Eastern Finance and Management, 8(1), 36-63. Available at: https://doi.org/10.1108/imefm-12-2012-0114.

Sufian, F., & Kamarudin, F. (2017). Forced mergers on bank efficiency and productivity: Evidence from semi-parametric Malmquist productivity index. Global Business Review, 18(1), 19-44. Available at: https://doi.org/10.1177/0972150916666850.

Sufian, F., Kamarudin, F., & Md. Nassir, A. (2017). Globalization and bank efficiency nexus: Empirical evidence from the Malaysian banking sector. Benchmarking: An International Journal, 24(5), 1-8.

Toumi, K., Louhichi, W., & Vivian, J.-L. (2012). Alternative financial decision principles: Theoretical foundations of Islamic banks’ capital structure. In W. Barnett, & F. Jawadi, recent developments in alternative finance: Empirical assessments and economic implications (international symposia in economic theory and econometrics: Emerald Group Publishing Limited.

Wanke, P., Azad, M., Kalam, A., Barros, C., & Hassan, M. K. (2016). Predicting efficiency in Islamic banks: An integrated multi criteria decision making approach. Journal of International Financial Markets, Institutions and Money, 45(C), 126-141. Available at: https://doi.org/10.1016/j.intfin.2016.07.004.

Yahya, M. H., Muhammad, J., & Hadi, A. R. A. (2012). A comparative study on the level of efficiency between Islamic and conventional banking systems in Malaysia. International Journal of Islamic and Middle Eastern Finance and Management, 5(1), 48-62. Available at: https://doi.org/10.1108/17538391211216820.

Views and opinions expressed in this article are the views and opinions of the author(s), International Journal of Management and Sustainability shall not be responsible or answerable for any loss, damage or liability, etc. caused in relation to/arising out of the use of the content.