Comparative Analysis of Islamic Bank’s Productivity and
Conventional Bank’s in Indonesia Period 2008-2016
Lina Nugraha Rani
1
, Aam Slamet Rusydiana
2
and Tika Widiastuti
3
1
Faculty of Economics and Business,Airlangga University, AirlanggaStreet, Surabaya, Indonesia
2
SMART Consulting,AchmadAdnawijaya Street, Bogor, Indonesia
3
Faculty of Economics and Business,Airlangga University, Surabaya, Indonesia
Keywords: Productivity Measurement, Islamic Bank’s, and Conventional Bank’s.
Abstract: The purpose of this research is to analyze the productivity differences between Islamic bank’s and
conventional banks in Indonesia during the period 2008 - 2016. In the first stage, bootstrapped Malmquist
index of Islamic bank’s and five conventional banks operates in Indonesia during the period 2008-2016. In
the second stage, the data panel models are used to Investigate the determinants of productivity change. The
results of the first stage show both Islamic bank’s and conventional banks are experiencing decreasing
productivity from 2008 to 2016. The results of the second stage show that Conventional banks are not
influenced by specific banking factors compared to the Islamic bank. This paper provides relevant
recommendations for improving the Islamic bank’s productivity in Indonesia. This research is aim to
expanding the literature on the productivity measurement in Islamic banks to conventional banks. The
productivity measurement analysis technique using Malmquist Index is still limited in Indonesian banking
studies.
1 INTRODUCTION
In Indonesia, the economic development of Islamic
finance began in 1992 and pioneered with the
establishment of the first Islamic bank, Bank
Muamalat Indonesia.
At this time, based on statistical data of Islamic
banking of the Financial Services Authority (FSA)
(2017) per May 2017, the number of Islamic banking
has reached 13 Islamic Bank’s, 21 Islamic Office
Channelling and 167 Bank of Islamic Financing with
whole office network as much as 458 offices
throughout Indonesia.
DEA (Data Envelopment Analysis) indicates the
inefficiency specifications of the service unit. Since
the DEA method was first introduced by Charnes,
Cooper, and Rhodes in 1978, researchers in some
areas recognize that DEA is an excellent
methodology and relatively easy to use in the
operational modeling process for performance
evaluation (Charnes, et. al, 1978). In this study, DEA
is used as a tool to measure and compare the
performance of Islamic and conventional banking, in
this case, all Islamic bank in Indonesia period 2008 –
2016.
Besides, to measure the productivity of Islamic
and conventional banks that was observed, this study
uses analysis of Malmquist Productivity Index (MPI).
The Malmquist index is a part of the DEA method that
specifically looks at the level of productivity of each
business unit so that it will see a change in the
efficiency and technology levels used based on
predetermined inputs and outputs. The Malmquist
index is also used to analyze performance changes
over time.
Determination of the limiting factor into a
benchmark whether a company has worked
efficiently and productively is a separate problem.
Not necessarily the factor is chosen as a variable to
measure the level of efficiency it represents the whole
aspect of the company, in this case, the bank. For that,
we need a measurement formulation of the level of
efficiency and productivity that can involve multi-
variable.
2 LITERATURE REVIEW
Efficiency and productivity are the concept that
shows the ratio of the result of comparison between
118
Rani, L., Widiastuti, T. and Rusydiana, A.
Comparative Analysis of Islamic Bank’s Productivity and Conventional Bank’s in Indonesia Period 2008-2016.
In Proceedings of the 1st International Conference on Islamic Economics, Business, and Philanthropy (ICIEBP 2017) - Transforming Islamic Economy and Societies, pages 118-123
ISBN: 978-989-758-315-5
Copyright © 2018 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved

input and output. An activity might be called efficient
if the effort has done to provide maximum output,
both quantity, and quality. An activity might also be
said to be efficient if the minimum effort can achieve
a specific output. Ozcan (2008) divided into several
parts namely efficiency: technical efficiency, scale
efficiency, cost efficiency and allocative efficiency.
The concept of productivity is fundamentally a
relationship between output and input in a production
process. Productivity measurement is the most widely
used method of Total Factor Productivity (TFP). This
method is used to overcome the weakness of
efficiency calculation more than one input and one
output. TFP is measured using index numbers that
could measure changes in price and quantity over
time. Also, TFP also measures comparisons and
differences between entities.
The Malmquist Index has become a standard
approach in measuring productivity levels, especially
when using nonparametric specs on microdata. This
index first introduced by Caves, Christensen and
Diewert (1982). The first generation model developed
by Caves et.al (1982), there are 2 (two) Malmquist
productivity index models (Bjurek, 1996). The first is
'Malmquist input quantity index' and the second is
'Malmquist output quantity index'.
Some research that applied banking productivity
measurement with TFP change value, for example,
was done by Yaumidin (2007), Saad et al. (2010),
Raphael (2013), Bahrini (2015) and Yildirim (2015).
Yaumidin (2007) attempted to compare the efficiency
of Islamic banks in the Middle East and Southeast
Asia. That research based on the failure of the bank
which then affects the occurrence of financial crisis,
both domestic and international. Overall, the result
shows that Islamic banks in Southeast Asia are
slightly more efficient than Islamic banks in the
Middle East. One of the causes was tragedy 9/11 in
2001 and the Iraq war of 2002. Likewise, the value of
TFP change.
Saad et al. (2010) examined the efficiency of the
selected company conventional and Islamic unit
trusts in Malaysia during the period 2002-2005.
Overall efficiency Islamic unit trust company
comparable to a conventional unit trust, and at any
given time some Islamic unit trust was found to be
above average in TFP. During the analysis period, the
average unit trust Malaysia suffered a setback TFP
and the main thing caused by a decrease in technical
efficiency. However, the change in efficiency
contributes positively to TFP. The change in
efficiency is mostly due to pure efficiency, not scale
efficiency. That shows the larger the size of the unit
trust will hurt the performance of TFP. The
substantial setback in the technical components and
the efficiency of positive growth, implying that the
decline of TFP in the unit trust industry in Malaysia
caused by a lack of innovation in technical
components.
3 METHODOLOGY
TFP growth estimation, as well as the components of
this study, refers to the Malmquist Index and DEA
method application-Dual Programming. Malmquist
Index This productivity is measured by DEAP 2.1
software developed by Coelli (1996). Nevertheless, to
see the effect of several variables both micro and
macro banking to the level of TFP change is done by
panel data regression.
The data used in this study was 5 Islamic Bank
and 5 Conventional Bank from 2008 to 2016. The
input and output variables obtained from the balance
sheet and profit and loss of each bank. Meanwhile for
phase two, the variables used to determine the TFP
effect of Islamic and Conventional banking is; the
Capital Adequacy Ratio (CAR), Bank Size, Bank
Management Quality, Business Diversification,
Credit Risk (NPF/NPL), Return on Equity (ROE),
Loan to Deposit Ratio (LDR), and Cash Ratio
compared to Total Bank Assets.
4 RESULTS AND DISCUSSION
4.1 Data and Variables
This study using the intermediation approach. The
input and output variables used listed in the below
Table. The first input is labor costs (X1). The second
input is fixed assets (X2). The third input is Total
Third Parties Funds (X3). Then the first output
variable used is total loan/financing provided by
conventional and Islamic bank’s (Y1). This variable
is the primary output in the intermediation approach.
Then the second output is the bank's investment
portfolio (Y2), and the third output is the net
operating income (Y3). Input and output variables
have represented the intermediation of a commercial
bank.
Comparative Analysis of Islamic Bank’s Productivity and Conventional Bank’s in Indonesia Period 2008-2016
119

Table 1: Statistics of Input-Output Variables.
4.2 Productivity Change of Indonesian
Bank’s
Analysis of growth rate of productivity Commercial
Banks in Indonesia using Malmquist Total Factor
Productivity Index (MTFPI) approach. Malmquist
Index can be decomposed into two components,
namely the Technical Efficiency Change (EFFCH)
and Technological Change (TECHCH). According to
Avenzora (2008), this is very useful because the
analysis can be done more specifically by component.
EFFCH positive (positive efficiency change) is
evidence that changes in efficiency been approaching
the frontier, while TECHCH positive (positive
technological change) note that changes in
technology as innovation (innovation). Then EFFCH
can be decomposed into two components, namely the
Pure Technical Efficiency Change (PECH) and Scale
Efficiency Change (Sech) (Fare et al., 1994).
On table 2 the estimated value or Malmquist
Productivity Index Malmquist Productivity Index
(MPI) of the overall banks in Indonesia included in
the observation:
Table 2: Results of MPI Overall Banks.
Overall Banks
EFFCH
TECHCH
PECH
SECH
TFPCH
2008-2009
1.008
0.753
0.995
1.013
0.759
2009-2010
0.957
0.966
1.001
0.956
0.924
2010-2011
1.030
1.001
1.003
1.026
1.031
2011-2012
0.998
0.953
0.989
1.008
0.950
2012-2013
1.012
0.754
1.007
1.004
0.763
2013-2014
0.983
0.950
1.003
0.980
0.934
Year/
Statistics
Total Loans
(Y1)
Investment
Portofolio (Y2)
Net Operating
Income (Y3)
Labor (X1)
Fixed Asset
(X2)
Total Deposits
(X3)
2008
Mean 58,248,737 25,161,863 7,542,465 1,782,606 1,357,759 92,859,404
SD 64,334,197 36,446,069 9,516,178 2,259,419 1,643,426 110,475,248
2009
Mean 69,246,417 30,470,427 8,276,847 1,925,670 1,447,973 107,705,759
SD 76,932,114 45,790,956 10,537,050 2,368,365 1,651,086 124,790,944
2010
Mean 85,244,680 34,749,007 14,527,246 2,235,668 1,551,390 114,742,903
SD 94,757,905 56,280,572 23,021,176 2,784,458 1,796,465 132,545,495
2011
Mean 105,287,758 40,026,850 13,370,536 2,470,822 1,765,733 136,194,466
SD 113,460,229 59,773,345 18,312,052 2,869,359 2,010,805 151,916,050
2012
Mean 132,139,043 45,723,451 15,141,007 2,828,442 2,230,811 167,034,256
SD 139,373,691 58,400,975 20,045,770 3,254,269 2,545,973 185,362,823
2013
Mean 162,761,724 48,022,560 17,944,682 3,312,096 2,675,966 188,517,140
SD 171,205,889 59,999,151 23,583,550 3,927,108 2,909,030 208,232,329
2014
Mean 182,443,439 58,334,955 21,028,719 3,836,230 3,344,325 216,252,876
SD 194,626,377 75,616,342 28,636,704 4,524,523 3,422,294 243,043,578
2015
Mean 206,297,111 50,416,417 24,123,640 4,313,027 5,198,562 233,616,643
SD 221,766,342 62,665,995 33,151,898 5,320,659 6,504,022 260,698,911
2016
Mean 231,247,543 78,596,738 27,167,525 4,792,993 10,519,982 263,883,725
SD 249,859,621 108,601,266 37,317,203 5,926,098 12,577,413 294,404,626
ICIEBP 2017 - 1st International Conference on Islamic Economics, Business and Philanthropy
120

2014-2015
1.005
0.756
0.982
1.023
0.759
2015-2016
1.010
1.093
1.007
1.003
1.104
Geometric
Mean
1.000
0.895
0.999
1.001
0.895
On table 2, it appears that from 2008 through 2016
study, commercial banks in Indonesia shows a
decline in productivity growth. The lowest
productivity decrease occurred in 2008-2009 and
2014-2015 with the value of TFPCH 0.759. As we
know, in 2008 and early 2009 there was a financial
crisis in Europe that bring impacts to Indonesia.
4.3 Productivity Change
ofConventional vs Islamic Banks
Table 3: Results of MPI Conventional Bank’s Observed.
ConventionalBanks
EFFCH
TECHCH
PECH
SECH
TFPCH
2008-2009
1.027
1.076
1.000
1.027
1.106
2009-2010
0.997
0.908
1.000
0.997
0.905
2010-2011
0.999
0.913
1.000
0.998
0.912
2011-2012
0.998
1.097
0.994
1.004
1.095
2012-2013
0.985
0.792
1.006
0.979
0.780
2013-2014
1.022
1.119
1.000
1.022
1.143
2014-2015
0.996
0.734
1.000
0.996
0.732
2015-2016
1.001
1.111
1.000
1.001
1.113
Geometric Mean
1.003
0.958
1.000
1.003
0.961
Table 4: Results of MPI of Islamic Bank’s Observed.
IslamicBanks
EFFCH
TECHCH
PECH
SECH
TFPCH
2008-2009
1.014
1.019
1.000
1.014
1.033
2009-2010
0.981
0.915
1.000
0.981
0.897
2010-2011
0.970
0.702
1.000
0.970
0.681
2011-2012
1.051
0.683
1.000
1.051
0.717
2012-2013
0.987
0.974
1.000
0.987
0.962
2013-2014
0.977
0.703
1.000
0.977
0.687
2014-2015
1.037
0.842
1.000
1.037
0.874
2015-2016
0.991
1.878
0.995
0.996
1.861
GeometricMean
1.001
0.913
0.999
1.001
0.914
Based on the value of table 3 and 4 TFPCH,
Conventional Bank and Islamic Bank’s during period
2008-2016 has declined. Impairment TFPCH Islamic
Banks were higher than conventional commercial
bank occurred because the components that affect the
Islamic Bank TFPCH more downward than
conventional commercial bank rather component
stagnant.
Based on the value of table 3 and 4 EFFCH,
Conventional Bank and Islamic Bank’s between 2008
and 2016 has increased. Impairment TECHCH
higher in Islamic Bank’s caused by low technological
innovation in creating new quality products.
Based on the value of table 3 and 4 PECH,
Conventional Bank’s during the period 2008 to 2016
has stagnated while Islamic Bank’s has declined
4.4 The Determinants of Indonesian
Banks’ Productivity Changes
For examining the relationship between financial and
non-financial response to the rate of productivity
growth of commercial banks in Indonesia, using
ordinary least squares (OLS) regressions simple
method. OLS is used as the dependent variable (Y) in
this case the level of productivity is a rational number,
considering the value of MPI is less than or greater
than 1. Here is a regression model for the variable
relationship of financial and non-financial to
productivity growth for commercial banks in
Indonesia:
Comparative Analysis of Islamic Bank’s Productivity and Conventional Bank’s in Indonesia Period 2008-2016
121

MPIi = β1 + β2CARi + β3 BDi +β4BMQi + β5CRi
+ β6CRISKi + β7LDRi + β8ROEi +
β9SIZEi + εi (4)
Information:
MPI (Level Productivity Growth Commercial
Bank’s); CAR (Capital Adequacy Ratio); BD
(Business Diversification); BMQ (Bank Management
Quality); CR ( Cash ratio); CRISK (Credit
risk); LDR (Loan to Deposit Ratio); ROE ( Return on
Equity); SIZE (Bank size)
In table 5 are the estimated results of
independent variables of the growth productivity as
(MPI) of overall banks in Indonesia:
Table 5: Results of OLS for Two Stages MPI Overall Bank’s.
Variable
Coefficient
Std. Error
t-Statistic
Prob.
C
2.276952
0.806302
2.823944
0.0062
CAR
0.018828
0.017973
1.047614
0.2984
BD
0.985789
4.469228
0.220572
0.8261
BMQ
-0.183430
0.227332
-0.806882
0.4224
CR
-0.005321
0.086796
-0.061300
0.9513
CRISK
-0.037874
0.058534
-0.647040
0.5197
LDR
-0.001446
0.004624
-0.312715
0.7554
ROE
-0.001495
0.003599
-0.415401
0.6791
SIZE
-0.075695
0.037041
-2.043546
0.0447
R-squared
0.086215
Mean dependent var
0.959375
Adjusted R-squared
-0.016746
S.D. dependent var
0.371107
S.E. of regression
0.374201
Akaike info criterion
0.977608
Sum squared resid
9.941896
Schwarz criterion
1.245586
Log likelihood
-30.10433
Hannan-Quinn criter.
1.085048
F-statistic
0.837354
Durbin-Watson stat
2.356349
Prob(F-statistic)
0.572973
Results were processing OLS model with Eviews
6 shown in Table 5. Processing results showed that
among all independent variables in the model, only
the bank size variable (SIZE) that significantly
influence the level of productivity of commercial
banks in Indonesia.
Table 6: Results of Two Stages MPI OLS for Islamic Bank’s.
Variable
Coefficient
Std. Error
t-Statistic
Prob.
BD
2.245865
4.908790
0.457519
0.6504
BMQ
0.055926
0.229465
0.243726
0.8090
CAR
0.022708
0.023548
0.964357
0.3421
CR
-0.153624
0.138096
-1.112445
0.2742
CRISK
-0.097788
0.074025
-1.321002
0.1959
LDR
-0.006091
0.006551
-0.929801
0.3594
ROE
-0.003753
0.004347
-0.863189
0.3945
SIZE
0.108939
0.051408
2.119106
0.0419
R-squared
0.123149
Mean dependent var
1.056725
Adjusted R-squared
-0.068662
S.D. dependent var
0.354579
S.E. of regression
0.366550
Akaike info criterion
1.007492
Sum squared resid
4.299482
Schwarz criterion
1.345268
Log likelihood
-12.14984
Hannan-Quinn criter.
1.129621
Durbin-Watson stat
2.229184
Table 7: OLS Results for Two Stages MPI Conventional Bank’s.
Variable
Coefficient
Std. Error
t-Statistic
Prob.
BD
-11.41301
14.62457
-0.780400
0.4409
BMQ
-1.314508
1.008639
-1.303249
0.2018
CAR
0.017308
0.030365
0.570003
0.5727
CR
0.052357
0.169496
0.308898
0.7594
CRISK
-0.031536
0.182023
-0.173250
0.8635
ICIEBP 2017 - 1st International Conference on Islamic Economics, Business and Philanthropy
122

LDR
0.004082
0.008136
0.501734
0.6193
ROE
0.006542
0.012932
0.505865
0.6164
SIZE
0.004021
0.054004
0.074452
0.9411
R-squared
0.082181
Mean dependent var
0.862025
Adjusted R-squared
-0.118592
S.D. dependent var
0.365795
S.E. of regression
0.386877
Akaike info criterion
1.115438
Sum squared resid
4.789568
Schwarz criterion
1.453214
Log likelihood
-14.30876
Hannan-Quinn criter.
1.237567
Durbin-Watson stat
2.188528
The results of OLS model processing with Eviews
6 shown in table 6 and 7 are those factors tested for
the effect on the productivity value of Islamic banks
and conventional banks. The processing result shows
that the size of bank’s (SIZE) has a significant effect
on the productivity of Islamic Banks in Indonesia.
While in Conventional Bank’s SIZE variable does not
affect.
5 CONCLUSION
The results of the test in the first stage are the general
level of MPI of commercial banks in Indonesia has
decreased productivity level which marked by the
value of changes in Total Factor Productivity
(TFPCH) below than 1. The external factors that
cause is a financial crisis that occurred in the interval
of research. The internal factors that cause this to
happen is the low level of technological innovation in
banking and stagnation of changes in the level of
efficiency. On the other hand, the results of the MPI
of Islamic Commercial Banks in Indonesia also
showed a decline in productivity growth, the reason
for the decline was also caused by the level of
technological innovation of banking and stagnation of
changes in the level of efficiency.
The second stage test result is to measure the
effect of the whole variable to the bank indicates that
only variable size of the bank (SIZE) has a significant
adverse effect on productivity level of the whole
commercial bank and Islamic bank in Indonesia. It is
because the bigger the size of a bank tends to make
become less productive. The bank is not flexible in
facing the challenges of competition, so it relatively
become less agile in determining strategic decisions.
REFERENCES
Avenzora Ahmad dan Jossy P. Moeis. “Analisis
Produktivitas dan Efisiensi Industri Tekstil dan Produk
Tekstil di Indonesia tahun 2002-2004. Jakarta 2008.
Bahrini, Raef. 2015. Productivity of MENA Islamic Banks:
a bootstrapped and Malmquist index appoarch.
International Journal of Islamic and Middle Eastern
Finance and Management. Vol. 8 No. 4, 2015, pp. 508
-528. Emerald Group Publishing Limited 1753-8394.
Bjurek, H. 1996, The malmquist total factor productivity
index, The Scandinavian Journal of Economics, Vol.
98, pp. 303-313.
Caves et.al. 1982. The Economic Theory of Index Number
and The Measurement of Input, Output and
Productivity. Econometrica, 50(6):1393-1414.
Charnes, A., Cooper, W.W., and Rhodes, E. 1978,
Measuring the Efficiency of Decision Making Units,
European Journal of Operation Research, 2, 6, 429-44.
Coelli, T. 1996, A Guide to DEAP Version 2.1: A Data
envelopment analysis program, CEPA Working Paper
96/08.
Charnes, A., Cooper, W.W., and Rhodes, E. (1978),
“Measuring the Efficiency of Decision Making Units”,
European Journal of Operation Research, 2, 6, 429-44.
Fare, R., S. Grosskopf, B. Lindgren and P. Roos, (1994),
Productivity Developments in Swedish Hospital: A
Malmquist Output Index Approach, in A. Charnes,
W.W. Cooper, A. Lewin and L. Seiford (eds.), Data
Envelopment Analysis: Theory, Methodology and
Applications, Boston: Kluwer Academic Publishers.
Financial Service Authority. 2017. Sharia Banking Statictic
in Indonesia 2017.
Ozcan YA. 2008. Health Care Benchmarking and
Performance Evaluation: An Assessment using Data
Envelopment Analysis. Springer; Norwell, MA: 2008.
Raphael, G. (2013). A DEA based malmquist productivity
index approach in assessing performance of
commercial banks: Evidence from Tanzania”,
European Journal of Business and Management, Vol.
5, No. 6, pp. 25-34.
Saad, Norma Md.Et al. 2010. A comparative analysis of the
Performance of conventional and Islamic unit trust
companies in Malaysia. Emerald, Vol. 6, No. 1, pp. 24-
47.
Yaumidin, U.K. 2007, Efficiency in Islamic banking, A
non-parametric approach, BuletinEkonomiMoneter dan
Perbankan, April 2007.
Yildirim, I. 2015, Financial efficiency analysis in Islamic
banks: Turkey and Malaysia models.Journal of
Economics, Finance and Accounting, Vol. 2, Issue 3,
pp. 289-300.
Comparative Analysis of Islamic Bank’s Productivity and Conventional Bank’s in Indonesia Period 2008-2016
123