Vous êtes sur la page 1sur 25

Industrial Management & Data Systems

Benchmarking top Arab banks' efficiency through efficient frontier analysis


Mohamed Mostafa
Article information:
To cite this document:
Mohamed Mostafa, (2007),"Benchmarking top Arab banks' efficiency through efficient frontier analysis",
Industrial Management & Data Systems, Vol. 107 Iss 6 pp. 802 - 823
Permanent link to this document:
http://dx.doi.org/10.1108/02635570710758734
Downloaded on: 27 June 2016, At: 15:56 (PT)
References: this document contains references to 76 other documents.
Downloaded by University of Ghana At 15:56 27 June 2016 (PT)

To copy this document: permissions@emeraldinsight.com


The fulltext of this document has been downloaded 1955 times since 2007*
Users who downloaded this article also downloaded:
(1998),"A study of efficiency evaluation in Taiwans banks", International Journal of Service Industry
Management, Vol. 9 Iss 5 pp. 402-415 http://dx.doi.org/10.1108/09564239810238820
(2007),"The efficiency of Islamic banking industry in Malaysia: Foreign vs domestic banks", Humanomics,
Vol. 23 Iss 3 pp. 174-192 http://dx.doi.org/10.1108/08288660710779399
(2007),"Modeling the efficiency of GCC banks: a data envelopment analysis approach",
International Journal of Productivity and Performance Management, Vol. 56 Iss 7 pp. 623-643 http://
dx.doi.org/10.1108/17410400710823651

Access to this document was granted through an Emerald subscription provided by emerald-srm:534301 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.

*Related content and download information correct at time of download.


The current issue and full text archive of this journal is available at
www.emeraldinsight.com/0263-5577.htm

IMDS
107,6 Benchmarking top Arab banks
efficiency through efficient
frontier analysis
802
Mohamed Mostafa
Gulf University for Science and Technology, Hawally, Kuwait

Abstract
Purpose The major aim of this research is to measure the relative efficiency of the top 100 Arab
banks. The sensitivity of the results is also investigated.
Design/methodology/approach Data envelopment analysis DEA method was used to evaluate
Downloaded by University of Ghana At 15:56 27 June 2016 (PT)

the relative efficiency of Arab banks. Cross-sectional data for the year 2005 were used to conduct the
analysis.
Findings The results indicate that the performance of several banks is sub-optimal, suggesting the
potential for significant improvements. Separate benchmarks were derived for possible reductions in
resources used, and significant savings are possible on this account.
Originality/value From a policy perspective, this study highlights the importance of encouraging
increased efficiency throughout the banking industry in the Arab world.
Keywords Data analysis, Benchmarking, Banks, Resource efficiency
Paper type Research paper

Introduction
Sustained high oil prices over the past two years have continued to drive growth in
banks based in the Arab worlds main oil producing states. A certain degree of banking
sector liberalization in some states, plus some economic diversification, have also
boosted the success of the financial industry in this part of the world. Saudi Arabian
banks dominated the upper reaches of the table of the top 100 Arab banks with four out
of the top five financial banks and eight out of the top 20. A further six of the top 20 are
registered in the United Arab Emirates (UAE), while other Gulf banks make up the
remainder. Despite the overall strength and success of their economies, North African
banks are minor players in the region overall. Attijariwafabank is the top-ranked
North African Bank, followed by National Bank of Egypt one position lower at 25th.
Indeed, the absence of an Egyptian bank in the higher echelons of the table (see
Appendix) may indicate the level of competition in that country, as a total of 12
Egyptian banks are among the regions top 100 The Egyptian government has made
significant progress in recent years in transferring control and ownership of the
countrys banking services to the private sector and it is hoped that several of the
countrys banks will be able to grow over the next few years on the back of their
greater freedom (Ford, 2006).
Industrial Management & Data Despite the unprecedented growth in the banking industry in the Arab world,
Systems research on performance and efficiency of the Arab banking industry is virtually
Vol. 107 No. 6, 2007
pp. 802-823 nonexistent. Measuring efficiency levels at the Arab banks is an important issue for
q Emerald Group Publishing Limited
0263-5577
managers and investors alike. Consumers also benefit from efficient resource usage
DOI 10.1108/02635570710758734 and allocation because this may mean lower prices and more professional service
(Anderson et al., 1998; Prajogo, 2007). Although widely employed to evaluate bank Benchmarking
efficiency in the West (Rickards, 2003), Data envelopment analysis (DEA) is less well top Arab banks
known within the banking sector in developing countries, and the Arab countries are
no exception. In this research we fill this research gap through some initial analysis on efficiency
top commercial banks in the Arab world, an area where there has been virtually no
previous research. The paper also contributes methodologically through the use of
various non-parametric DEA methods that provide considerable information for 803
business analysis. More specifically, the purpose of this research is twofold to:
(1) assess the efficiency of the top Arab banks using 2005 operating data; and
(2) develop separate benchmarks for possible reductions in resources and possible
increase in outputs used by top Arab banks.

The rest of the paper is organized as follows. The next section reviews relevant
literature. The following section summarizes the methodology used to conduct the
Downloaded by University of Ghana At 15:56 27 June 2016 (PT)

analysis. The subsequent section presents empirical results of the efficiency levels of
Arab banks. After a brief preliminary data analysis, this section first sets out efficiency
scores derived from estimating the basic DEA models; it also presents sensitivity
analysis of DEA-derived efficiency scores as a rough validity check on the results.
Next, the paper sets out some managerial and policy implications of the analysis.
The final section deals with research limitations and explores avenues for future
research.

Literature review
During the late 1980s and particularly in the 1990s, the DEA method has been used
extensively to evaluate banking institutions. In their review, Berger and Humphrey
(1997) count 130 studies on the efficiency of the banking industry in 21 countries; 116 of
them were published between 1992 and 1997.
Miller and Noulas (1996) examined the efficiency of large US banks. They found
overall technical efficiency (TE) of around 97 per cent. However, the majority of banks
were found to be too large and experiencing decreasing returns to scale. A second stage
regression analysis showed that pure TE is positively related to bank size and bank
profitability.
Bhattacharya et al. (1997) used a two-stage DEA approach to examine the impact of
liberalization on the efficiency of the Indian banking industry. In the first stage a TE
score was calculated, whereas in the second stage a stochastic frontier analysis was
used to attribute variation in efficiency scores to three sources: temporal, ownership
and noise component.
Alirezaee et al. (1998) utilised data from 1,282 bank branches in Canada to conduct
numerical experiments relating to DEA results to sample size. They found that the
average branch efficiency score varied inversely with the number of branches in the
sample and directly with the total number of inputs and outputs. They also cautioned
that using relatively small sample sizes in a model with as few as three inputs and
three outputs could lead to a substantial upward bias in efficiency scores.
Seiford and Zhu (1999) examined the performance of the top 55 US banks using a
two-stage DEA approach. Results indicated that relatively large banks exhibit better
performance on profitability, whereas smaller banks tend to perform better with
respect to marketability.
IMDS Cook and Hababou (2001) studied both the sales and service efficiencies of bank
107,6 branches. They extended the standard additive DEA modelling approach using goal
programming to derive optimal efficiency scores while accounting for shared branch
resource inputs.
Drake and Howcroft (2002) assessed the relative efficiency of UK clearing bank
branches using DEA method. This paper utilised the basic efficiency indices and
804 extended the analysis by examining the relationship between size and efficiency.
Yildirim (2002) evaluated the efficiency of Turkish commercial banks between 1988
and 1999 using DEA method. Results suggest that over the sample period both pure
technical and scale efficiency measures showed a great variation and the sector did not
achieve sustained efficiency gains.
Using a bootstrapping DEA technique, Casu and Molyneux (2003) investigated
efficiency across European banking systems. Results suggest that there has been a
slight improvement in bank efficiency levels since the implementation of the EUs
Downloaded by University of Ghana At 15:56 27 June 2016 (PT)

Single Market Programme.


Krishnasamy (2003) used both DEA and Malmquist total factor productivity
index (MPI) to evaluate bank efficiency and productivity changes in Malaysia over
the period 2000-2001. The results from the analysis indicated that total MPI
increased in all the bans studied. The growth of productivity in these banks was
attributed to technological change rather than TE change.
Lo and Lu (2006) employed a two-stage DEA approach including profitability
and marketability to explore the efficiency of financial holding companies (FHCs)
in Taiwan. Factor-specific measures and BCC (Banker-Charnes-Cooper) model were
combined together to identify the inputs/outputs that are most important and to
distinguish those FHCs which can be treated as benchmarks. Results show that
big-sized FHCs are generally more efficient than small-sized ones.
Wu et al. (2006) integrated DEA and neural networks (NNs) to examine the relative
branch efficiency of a large Canadian bank. Findings suggest that the predicted
efficiency using the DEA-NN model has good correlation with that calculated by DEA,
which indicates that the predicted efficiency using the DEA-NN approach is a good
proxy to classical DEA approach.
Studies on performance evaluation and efficiency of the banking industry in the
Arab world are rare in extant literature. Notable among them are Al-Faraj et al. (1993),
Chaffai (1997) and Al-Shammari and Salimi (1998). Al-Faraj et al. (1993) used DEA to
evaluate 15 bank branches of one of the largest commercial banks in Saudi Arabia.
The results indicated that 12 branches were efficient. Chaffai (1997) evaluated the TE
in the Tunisian banking industry. Results of this study showed that TE of labour and
capital inputs are decreasing over time and labour input is more inefficiently used than
capital input. Al-Shammari and Salimi (1998) evaluated the comparative operating
efficiency of banks in Jordan using a modified version of DEA. The results suggest that
the majority of banks investigated are fairly inefficient over the period 1991-1994.
Table I summarises previous research conducted to evaluate bank efficiency using
DEA.
From this brief review we find that although numerous studies have attempted to
assess banks efficiency in the West and other parts of the world, very few studies have
focused on measuring efficiency in the Arab world. The lack of empirical research in
this area confirms Atiyyas (1992) findings concerning the lack of empirical research
Downloaded by University of Ghana At 15:56 27 June 2016 (PT)

No. of
Study Country banks Inputs Outputs

Sherman and Gold (1985) USA 14 Employees, expenses, space Number of transactions
Parkan (1997) Canada 35 Employees, expenses, space, rent, terminals Number of transactions, customer response,
error corrections
Oral and Yolalan (1990) Turkey 20 Employees, terminals, number of accounts, Number of transactions
credit applications
Vassiloglou and Giokas (1990) Greece 20 Employees, suppliers, space, Computer Number of transactions
terminals
Giokas (1991) Greece 17 Employees, expenses, rent Number of transactions
Al-Faraj et al. (1993) Saudi 15 Employees, location, expenses, acquired Net profit, balance of current accounts,
Arabia equipment savings account, loans, number of accounts
Fukuyama (1993) Japan 143 Employees, capital, funds from customers Loan revenue, other revenues
Sherman and Ladino (1995) USA 33 Employees, expenses, rent Number of transactions
Favero and Papi (1995) Italy 174 Employees, capital, loanable funds, deposits Loans, investment in securities, non-interest
income
Athanassopoulos and Curram UK 250 ATMs, employees, counter transactions, Loans sales, liability sales, investments and
(1996) potential market insurance policies sold
Athanassopoulos (1997) Greece 68 Employees, ATMs, terminals, interest costs, Non-interest income
non-interest costs, location
Resti (1997) Italy 270 Employees, capital Loans, deposits, non-interest income
Bhattacharya et al. (1997) India 74 Interest expense, operating expense Advances, deposits, investments
Schaffnit et al. (1997) Canada 291 Employees Transactions, maintenance
Ayadi et al. (1998) Nigeria 10 Interest on deposits, expenses on personnel, Total loans, interest income, non-interest
total deposits income
Al-Shammari and Salimi (1998) Jordan 16 Selected financial ratios Selected financial ratios
Chen and Yeh (1998) Taiwan 34 Employees, assets, number of branches, Loans, investments interest income,
operating costs, interest expenses non-interest income
Seiford and Zhu (1999) USA 55 Employees, assets, capital stock Revenue, profits
Golany and Storbeck (1999) USA 182 Employees, space, marketing Loans, deposits, accounts per customer,
satisfaction
(continued)

research in banks
top Arab banks
Benchmarking

efficiency

805

A survey of DEA
Table I.
Downloaded by University of Ghana At 15:56 27 June 2016 (PT)

806
107,6
IMDS

Table I.
No. of
Study Country banks Inputs Outputs

Drake and Howcroft (1999) UK 250 Number of loan accounts, number of Personal loans, new cheque accounts,
mortgage accounts, number of cheque mortgage loans, insurance commission,
accounts change in marketed balances
Zenios et al. (1999) Cyprus 144 Employees, terminals, space, current Number of transactions
accounts, savings accounts, credit
applications
Mukherjee et al. (2002) India 68 Networth, borrowings, operating expenses, Deposits, net profit, advances, non-interest
employees, number of branches income, interest income
Ho and Zhu (2004) Taiwan 41 Capital stocks, assets, number of branches, Sales, deposits
employees
Sakar (2006) Turkey 11 Branch numbers, employees per branch, ROA, ROE, interest income/assets, interest
assets, loans, deposits income/operating income, non-interest
income/assets
Wu et al. (2006) Canada 142 Employees, expenses Deposits, revenues, loans
Howland and Rowse (2006) Canada 162 Non-sales FTE, sales FTE, size, city Loans, deposits, average number of
employment rate products/customer, customer loyalty
into Arab management practices in general. In this investigation we aim to fill this Benchmarking
research gap by empirically evaluating banks efficiency in the Arab world.
top Arab banks
Methodology efficiency
Data envelopment analysis
Efficiency is one of the most important performance measures of a business (Rao and
Miller, 2004; Phusavat and Photaraon, 2006). Gandjour et al. (2002) concluded that 807
many quality and efficiency indicators used by executives are lacking in general
validity. Using a recognized and valid measure of efficiency is critical for managers
seeking to increase the effectiveness of their organizations.
Introduced in 1978 by Charnes et al. (1978), DEA assigns an efficiency score to each
unit by comparing the efficiency score of ach unit with that of its peers. It identifies a
frontier comprising best performers. Those units that lie on the frontier are recognized
as efficient, and those that do not, as inefficient. DEA involves the solution of a linear
Downloaded by University of Ghana At 15:56 27 June 2016 (PT)

programming (LP) problem to fit a non-stochastic, non-parametric production frontier


based on the actual input-output observations in the sample. In the basic DEA model
(CCR), the objective is to maximize the efficiency value of a test firm k from among a
reference set of s firms, by selecting the optimal weights associated with the input and
output measures. The maximum efficiencies are constrained to 1. The formulation is
represented by model (1):
P
Oky V ky
maximize E kk PI
y

kx U kx
x
1
subject to : E ks # 1 ; firms s
uks ; vky $ 0

where Eks is the efficiency score of firm s, using the weights of test firm k; Osy is the
value of output y for firm s; Isx is the value for input x of firm s; vky is the weight
assigned to firm k for output y; and ukx is the weight assigned to firm k for input x.
This non-linear programming is the equivalent to the LP problem represented by
model (2).
P
maximize E kk Oky V ky
y

subject to : E ks # 1 ; firms s
P 2
I kx U kx 1
x

uks ; vky $ 0

The transformation is completed by constraining the efficiency P ratio denominator from


equation (1) to a value of 1, represented by the constraint x I kx U kx 1:
The result of formulation (2) is an optimal simple or TE value (E *kk) that is at most
equal to 1. If E *kk 1, then no other firm is more efficient than firm k for its selected
weights. That is, E kk* 1 has firm k on the optimal frontier and is not dominated by
any other firm. If E *kk , 1, then firm k does not lie on the optimal frontier and there is at
IMDS least one other firm that is more efficient for the optimal set of weights determined by
107,6 equation (2). Formulation (2) is executed s times for each firm.
The dual of the CCR model is represented by model (3):
minimize u
subject to :
808 P
ls I sx 2 usx # 0 ; inputs I
s 3
P
ls Osy 2 Oky $ 0 ; outputs O
s

ls $ 0 ; firms s

where u is the efficiency score.


Downloaded by University of Ghana At 15:56 27 June 2016 (PT)

The CCR model has an assumption of constant returns to scale (CRS) for the inputs
and outputs. To take into consideration variable returns to scale (VRS), a model
introduced by Banker et al. (1984) (BCC) is utilized. The BCC model aids in determining
the scale efficiency of a set of units (which is a technically efficient unit for the VRS
model). This model has an additional convexity constraint defined by limiting the
summation of the multiplier weights (l) equal to 1, or:
X
ls 1 4
s

The BCC model evaluates whether increasing, constant, or decreasing returns to scale
would boost the efficiency observed. In the case of CRS, the output changes
proportionally to input, as it also does in the CCR model. But with VRS, a change in the
input leads to a disproportional change in the output. The use of the CCR and BCC
models together helps determine the overall technical and scale efficiencies of the firm
and whether the data exhibits varying returns to scale (Sarkis, 2000).

Data
To estimate the production frontier, we used cross-sectional data for the year 2005,
obtained from one of the leading business magazines in the Arab world (Middle East,
2006). This magazine publishes annually a list of the top 100 Arab banks. To be
included in the data set used in this study, banks had to meet two conditions: first, that
financial information is available; and, second, that they do not have negative financial
data. DEA requires that data set to be non-negative for the outputs and strictly positive
for the inputs (Sarkis and Weinrach, 2001). Unfortunately, there is no DEA model to
date that can be used with negative data directly without any need to transform them
(Portela et al., 2004). About 15 banks did not meet these two conditions and were
excluded from analysis.

Inputs and outputs


The first and very crucial step in conducting a DEA is the determination of inputs and
outputs. The main important point in this process is that the input-output variables
should be chosen in accordance with the type of efficiency being assessed (Sherman
and Rupert, 2006). Efficiency in DEA is not confined to a traditional sense of operating
efficiency; it can be generalized to represent relative evaluation of performance in any Benchmarking
performance dimension if the inputs and outputs are specified according to the top Arab banks
performance dimension considered (Manandhar and Tang, 2002).
At present, there is no agreement on the explicit definition and measurement of efficiency
banks inputs and outputs (Casu and Molyneux, 2003). However, in the literature, two
approaches have been used to measure banks efficiency: production approach and
intermediation approach. In the production approach the bank is considered as a firm 809
delivering services to its clients in the form of transactions. Benchmarking models
examine how well different banks combine their resources to support the largest
possible number of transactions. Under the intermediation approach, banks are
thought of as primarily intermediating funds between savers and investors. The inputs
and outputs are measured in monetary units. Based on similar studies (Seiford and
Zhu, 1999), inputs in this study were assets and capital. Outputs were chosen to be net
profit, rate on assets (ROA), and rate on equity (ROE). It should be noted that the
Downloaded by University of Ghana At 15:56 27 June 2016 (PT)

availability of the data imposed strong restrictions on the type of variables one is able to
use. However, we think that the variables selected are sufficient to illustrate the
approach used to measure efficiency in the Arab banks.
It is well known that DEA is sensitive to variable selection. As the number of
variables increases, the ability to discriminate between the DMUs decreases. The more
variables are added the greater becomes the chance that some inefficient unit
dominates in the added dimension and becomes efficient (Smith, 1997). Thus, to
preserve the discriminatory power of DEA the number of inputs and outputs should be
kept at a reasonable level. There are no diagnostic checks for model misspecification in
DEA that could result due to wrong choices in variable selection (Galagedera and
Silvapulle, 2003). However, Raab and Lichty (2002) suggest a general rule of thumb
the minimum number of DMUs is greater than three times the number of inputs plus
outputs. In our study, with a total of two inputs and three outputs, a good minimum set
is 15 data points; we have 85 data points.

Results
Preliminary data analysis
The simple DEA model is based on CRS, implying that the size of a bank is not relevant
when assessing efficiency. However, it is likely that the size of the bank will influence its
ability to produce services more efficiently. As the CRS totally ignores the scale of
operations and will possibly lead to an identification of very unrealistic benchmarks
(Munksgaard et al., 2005), a VRS model is used in this study. A VRS frontier allows best
practice level of outputs to inputs to vary with size of bank. A DEA model can be
analysed in two ways, an input orientation or an output orientation. An input orientation
provides information as how much proportional reduction of inputs is necessary while
maintaining the current levels of outputs for an inefficient bank to become
DEA-efficient. On the other hand, an output orientation analysis provides information
on how much augmentation to the levels of outputs of an inefficient bank is necessary
while maintaining current input levels for it to become DEA-efficient. Since, it is well
known that, in competitive markets, the DMUs are output-oriented (Barros and
Athanassiou, 2004), we use the output maximisation assumption in this study.
To ensure the validity of the DEA model specification, an isotonicity test (Avkiran,
1999) was conducted. An isotonicity test involves the calculation of all inter-correlations
IMDS between inputs and outputs for identifying whether increasing amounts of inputs lead to
107,6 greater outputs. As positive inter-correlations were found, the isotonicity test was
passed and the inclusion of the inputs and outputs was justified.

Efficiency scores
While standard optimisation software packages can be used for estimating efficiency
810 scores, here we used a commercial package called Frontier Analyst Professional
Version 3.0 (Banxia Holdings Ltd, 2001). In this software LP models outlined above are
solved 85-times once for each of the banks in the data set. For each bank, the
software searches for a linear combination of banks in the sample that produces a
greater level of output with fewer inputs. The model is searching for a comparison that
identifies output slacks or excess input usage of the bank under analysis. In solving the
LP problem three characteristics of the model must be specified by the user: the returns
to scale, the valuation system, and the orientation system. Returns to scale may be
Downloaded by University of Ghana At 15:56 27 June 2016 (PT)

either CCR or VRS. The evaluation system refers to weights placed on the inputs and
outputs in the objective function, subject to the inequality constraints. The orientation
system, which defines the objective function, can be designated as input-orientation or
output-orientation. In this study, we use the VRS output-orientation model with the
default weights suggested by the software.
The VRS scores measure pure TE only. However, for comparative purposes, we also
present the CRS scores, which are composed of a non-additive combination of pure TE
and scale efficiencies. A ratio of the overall efficiency scores to pure TE scores provides
a scale efficiency measurement. The relative efficiency scores of the banks analysed are
presented in Table II. The results indicate that scores range from 0 to 100 per cent for
the banks in the sample, with an average of 31 per cent when using the CCR model
(CRS) with a standard deviation of 21.6, and from 0 to 100 per cent, with an average of
43 per cent and a standard deviation of 27.2 for the banks in the sample when using
the BCC model (VRS). This means that, if the average bank in the sample was to
achieve the level of its most efficient counterpart then the average bank could realize
a 57 per cent cost saving (i.e., 1 2 [43/100]). A Spearmans rank order correlation
coefficient between the efficiency rankings derived from CCR and BCC analyses is 0.98.
The positive and strong correlation indicates that the rank of each bank derived from
applying the two approaches is similar. This implies that the choice of methodology
has no apparent impact on the estimated average efficiency scores.
These results are not surprising as it has been shown that DEA scores computed
with the CRS assumption are less than or equal to the corresponding VRS efficiency
scores (Banker et al., 1984). However, Because of difference in bank-specific operating
environments, it may be difficult to interpret overall banks efficiency by comparison
with other countries. For example, in a study of the large banks in the USA, Miller and
Noulas (1996) found 0.97 mean TE scores, which may mean that banks in the Arab
world are less competitive and/or less efficient than the US banking industry. Studies
comparing the German and Austrian banking systems efficiency found a relative
efficiency scores in the order of 75-90 per cent (Hauner, 2005). Our results corroborate
other empirical results which generally found that efficiency scores for developing
countries are lower than those of developed countries. For example, in Turkey cost
efficiency was 68.5 per cent in 1996 (Isik and Hassan, 2002). Hardy and Bonaccorsi d
Patti (2001) report efficiency in the range of 48.5 to 72.8 per cent in Pakistan.
Benchmarking
Bank Score BCC Score CCR RTS Scale efficient Peer
top Arab banks
1 Banque de Caire, Egypt 100 100 0 1 efficiency
2 National Bank of UAQ, UAE 100 100 0 1
3 United Arab Bank, UAE 100 100 0 1
4 Egyptian American Bank, Egypt 100 100 0 1
5 Suez Canal Bank, Egypt 78.3 100 0 0.78 811
6 Commercil Bank of Syria, Syria 76.5 100 0 0.77
7 Intl Banking Corp, Bahrain 67.4 100 0 0.67
8 Al Rajhi Bank, Saudi Arabia 61.6 100 0 0.62
9 Libya Arab Foreign Bank, Libya 53.1 93.4 1 0.57 7,4
10 National Comm Bank, Saudi Arabia 50 88.9 1 0.56 8
11 Bank of Sharjah, UAE 47.7 86.4 21 0.55 7,2
12 National Bank of Abu Dhabi, UAE 46.8 76.3 1 0.61 8,7
13 Samba, Saudi Arabia 45.5 73.8 1 0.62 8,7
Downloaded by University of Ghana At 15:56 27 June 2016 (PT)

14 National Bank of Kuwait, Kuwait 43.6 73.2 1 0.6 8,7


15 Bank of Dhofar, Oman 41.6 69.8 21 0.6 7,3,4
16 Saudi British Bank, Saudi Arabia 40.3 69 1 0.58 8,7
17 Fransabank, Lebanon 39.6 66.4 1 0.6 7,5
18 Mashreqbank, UAE 39.4 62.9 1 0.63 8,7
19 Riyadh Bank, Saudi Arabia 38.4 59.6 1 0.64 8,7
20 Arab National Bank, Saudi Arabia 37.3 56.4 1 0.66 8,7
21 Invest Bank, UAE 36.4 55.7 21 0.65 7,3,4
22 Doha Bank, Qatar 35.9 53.7 1 0.67 7,4
23 Audi Sardar Bank, Lebanon 35.7 53.5 1 0.68 7
24 Qatar National Bank, Qatar 35.6 53.2 1 0.67 8,7
25 Banque Saudi Fransi, Saudi Arabia 35.4 51.8 1 0.68 8,7
26 Commercial Bank of Kuwait, Kuwait 35.2 50.8 1 0.69 8,7
27 Dubai Islamic Bank, UAE 35 49.2 1 0.71 8,7
28 CIB, Egypt 34.9 49 1 0.71 7,4
29 Gulf Bank, Kuwait 33.7 48.9 1 0.69 8,7
30 Abu Dhabi Comm Bank, UAE 33.3 48.9 1 0.68 8,7
31 Emirates Bank Intl, UAE 31.8 46.9 1 0.68 8,7
32 Bank Al Jazeera, Saudi Arabia 31.7 46.3 1 0.68 8,7
33 Saudi Hollandi Bank, Saudi Arabia 31.5 44.9 1 0.7 8,7
34 Kuwait Finance House, Kuwait 31.2 44.4 1 0.7 8,7
35 Bank of Bahrain and Kuw, Bahrain 30.3 43.4 1 0.69 7,4
36 Arab African Intl Bank, Egypt 29.6 42.8 21 0.69 7,3,4
37 Intl Bank of Qatar, Qatar 28.6 41.9 21 0.68 7,3,4
38 First Gulf Bank, UAE 27.3 40.8 1 0.67 8,7
39 Union National Bank, UAE 26.8 40.5 1 0.66 8,7
40 Naional Bank of Dubai, UAE 26.7 39.9 1 0.67 8,7
41 United Gulf Bank, Bahrain 26.5 39.6 21 0.67 7,2,3,5
42 Bank of Kuwait & ME, Kuwait 26.1 38.7 1 0.67 7,4
43 Qatar Islamic Bank, Qatar 25.9 38.2 1 0.68 7,4
44 Oman Intl Bank, Oman 25.8 37.3 1 0.69 7,4
45 Saudi Invest. Bank, Saudi Arabia 25.3 36.7 1 0.69 8,7
46 Credit Populaire de Maroc, Morocco 25.3 36.6 1 0.69 8,7
47 Commercial Bank of Dubai, UAE 24.6 35.6 1 0.69 7,4
48 Ahli United Bank, Bahrain 24.3 34.2 1 0.71 8,7
49 Arab Bank Plc, Jordan 24.2 33.5 1 0.72 8 Table II.
50 Sharjah Islamic Bank, UAE 24 32.4 21 0.74 7,2 Arab banks efficiency
(continued) scores
IMDS Bank Score BCC Score CCR RTS Scale efficient Peer
107,6
51 RAK Bank, UAE 23.9 31.9 1 0.75 7,4
52 Credit de Maroc, Morocco 23.7 31.6 1 0.75 7,4
53 National Bank of Fuj., UAE 23.2 31 21 0.75 7,3,4
54 Burgan Bank, Kuwait 23.2 29.6 1 0.78 8,7
55 Commercial Bank of Qatar, Qatar 23 29.6 1 0.78 8,7
812 56 Al Ahli Bank of Kuwait, Kuwait 22.7 29.3 1 0.78 8,7
57 Soc. Gen. Mar. de Banques, Morocco 22 29.2 1 0.75 7,4
58 Blom Bank, Lebanon 21 28.2 1 0.74 8,7
59 Shamil Bank of Bahrain, Bahrain 21 27.8 21 0.76 7,2,3
60 National Bank of Bahrain, Bahrain 20.5 27.4 1 0.75 7,4
61 The Housing Bank, Jordan 19.5 27.3 1 0.71 7,4
62 Bank Muscat, Oman 19 26.3 1 0.72 8,7
63 Gulf Intl Bank, Bahrain 18.2 24.7 1 0.74 8,7
64 Banque Maroc. du Com. Ext., Morocco 17.6 23.2 1 0.76 8,7
Downloaded by University of Ghana At 15:56 27 June 2016 (PT)

65 HSBC Bank Egypt, Egypt 16.4 22.5 1 0.72 7,5,4


66 National Bank of Oman, Oman 16.4 21.6 1 0.76 7,4
67 Banque de LHabitat, Tunisia 16.3 21.2 1 0.77 7,4
68 Arab Bank for Invest & FT, UAE 15.3 19.5 21 0.78 7,2,3
69 Gulf Intl Corp. GSC, Kuwait 14.2 19 1 0.75 8,7
70 Byblos Bank, Lebanon 13.8 19 1 0.73 7,4
71 Attijariwafabank, Morocco 13.7 18.7 1 0.73 8,7
72 Banque Lib-Franc., Lebanon 13.4 17.6 1 0.76 7,4
73 Wahda Bank, Libya 12.1 17.2 1 0.7 7,4
74 Arab Banking Corp, Bahrain 11.9 16.9 1 0.7 8,7
75 Gumhouria Bank, Libya 11.9 15.4 1 0.77 7,4
76 Banque Intl Arabe de Tun., Tunisia 10 14.6 1 0.68 7,4
77 Bank of Alexandria, Egypt 9.3 13.2 1 0.7 7,4
78 Arab Intl Bank, Egypt 8.6 9.5 1 0.91 7,4
79 Banque National Agricole, Tunisia 7.8 9.4 1 0.83 7,4
80 Banque de la Med., Lebanon 7.1 7.7 1 0.92 7,4
81 Banque Misr, Egypt 6 7.3 1 0.82 7,4
82 National Bank of Egypt, Egypt 3.9 5.5 1 0.71 8,7
83 Banque Ext. dAlger, Algeria 3.2 5 1 0.64 7,4
84 Soc. Tunis. de Banques, Tunisia 1.7 2 1 0.85 7,4
85 Egyptian Arab Land Bank, Egypt 0.38 0.4 1 0.95 7,6,4
Table II. Notes: RTS Return to scale; 0 constant; 2 1 decreasing; 1 increasing

Bergendahl (1998) states that DEA technique is an adequate tool for benchmarking,
since it allows the identification of a group of efficient firms for each non-efficient
one. This identified group may be used in the definition of operational goals for
their non-efficient counterpart, considering its various input and output variables.
Table II provides the linear combination of banks on the efficiency frontier closest to a
particular bank. The linear combination is also referred to in the literature as the peer
group or the reference set for this bank and indicates to which of the efficient banks an
inefficient bank is closest in its combination of inputs and outputs. A bank, which
appears frequently in the reference set is likely to be a bank which is efficient with
respect to a large number of factors, and is probably a good example of an exemplary
operating performer. Efficient banks that appear seldom in the reference set of other
banks are likely to possess a very uncommon input/output mix and are thus not
suitable examples for other inefficient banks. Figure 1 shows the reference frequencies Benchmarking
of the efficient banks. top Arab banks
It can be seen from Figure 1 that out of the 85 banks in the data set only eight are
efficient. Of these, the one that appears more frequently as peer (i.e. benchmark) is efficiency
International Banking Corporation, Bahrain (80 times) followed by Al-Rajhi Bank,
Saudi Arabia (36 times) followed by Egyptian American Bank, Egypt (35 times). In
other words, the peer count number can be considered a measure of the extent to which 813
the performance of an efficient bank can be a useful for the non-efficient ones.

Jackknifing and detecting outliers


To test whether there were extreme outliers, which affected the frontier and efficiency
scores, we ran DEA analyses that dropped out each efficient bank one at a time from
analysis This is a procedure known as Jackknifing. And it tests the robustness of the
DEA results in regard to outliers. Bonesronning and Rattso (1994) used Jackknifing in
Downloaded by University of Ghana At 15:56 27 June 2016 (PT)

their analysis. The difference in our analysis is that they dropped each unit once at a
time whereas we dropped only the efficient units that construct the frontier.
For example, for the VRS, we ran ten additional DEA analyses. We then tested the
similarity of efficiency rankings between the model with all the banks included and
those based on dropping out each efficient unit one at a time by the Spearman
rank correlation coefficient. Correlations ranged from 0.96 to 0.99. The high rank
correlation coefficients show that the rankings are stable in regard to outlier
banks determining the efficiency frontier. Similar results were obtained using the CRS
model.

Sensitivity analysis
At the individual bank level, DEA also provides rich diagnostic information through
sensitivity analysis. Table III includes the sensitivity analysis results for each bank in
the data set. This table shows the amount of slack in each of the controllable input and

Int'l Banking Corp, Bahrain 80

Al Rajhi Bank, Saudi Arabia 36

Egyptian American Bank, Egypt 35

United Arab Bank, UAE 9

National Bank of UAQ, UAE 6

Suez Canal Bank, Egypt 4

Commercil Bank of Syria, Syria 1

Banque de Caire, Egypt 0


Figure 1.
Reference set frequencies
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80
Downloaded by University of Ghana At 15:56 27 June 2016 (PT)

banks
814
107,6
IMDS

Table III.
Percentage of

variable of inefficient
improvement for each
Bank Percentage of capital Percentage of assets Percentage of profits Percentage of ROE Percentage of ROA

1 Banque de Caire, Egypt 0 0 0 0 0


2 National Bank of UAQ, UAE 0 0 0 0 0
3 United Arab Bank, UAE 0 0 0 0 0
4 Egyptian American Bank, Egypt 0 0 0 0 0
5 Suez Canal Bank, Egypt 0 0 0 0 0
6 Commercil Bank of Syria, Syria 0 0 0 0 0
7 Intl Banking Corp, Bahrain 0 0 0 0 0
8 Al Rajhi Bank, Saudi Arabia 0 0 0 0 0
9 Libya Arab Foreign Bank, Libya 0 2 78.4 43.7 7.1 471.4
10 National Comm Bank, Saudi Arabia 237.7 2 34.8 12.4 81 71.5
11 Bank of Sharjah, UAE 22.1 0 43.4 26.2 15.7
12 National Bank of Abu Dhabi, UAE 0 2 44.9 31.1 75.2 347
13 Samba, Saudi Arabia 0 2 16.2 35.4 41.3 78.8
14 National Bank of Kuwait, Kuwait 0 2 33.9 36.6 80.1 279.7
15 Bank of Dhofar, Oman 0 0 60.3 43.3 44.6
16 Saudi British Bank, Saudi Arabia 0 2 25 44.8 92.6 259.5
17 Fransabank, Lebanon 0 2 52.9 345.8 178.4 50.5
18 Mashreqbank, UAE 0 2 11.3 58.9 113.1 241
19 Riyadh Bank, Saudi Arabia 0 2 5.9 67.9 97.8 161.3
20 Arab National Bank, Saudi Arabia 0 2 39.8 77.3 138.4 459.6
21 Invest Bank, UAE 0 0 140.3 79.6 88.2
22 Doha Bank, Qatar 0 2 44.8 98 86.1 246.7
23 Audi Sardar Bank, Lebanon 232.7 2 79.1 289 478 86.8
24 Qatar National Bank, Qatar 0 2 33 88 151.5 427.9
25 Banque Saudi Fransi, Saudi Arabia 0 2 42.4 93.1 158.2 531.6
26 Commercial Bank of Kuwait, Kuwait 0 2 49.2 97 127.1 442.8
27 Dubai Islamic Bank, UAE 0 2 60.8 103.4 141.4 675.7
28 CIB, Egypt 0 2 55.1 161.1 103.9 411.3
29 Gulf Bank, Kuwait 0 2 41.5 104.7 150.5 454.7
30 Abu Dhabi Comm Bank, UAE 20.9 0 105.8 168.3 262.3
31 Emirates Bank Intl, UAE 0 2 13.2 113.5 181.6 351.3
32 Bank Al Jazeera, Saudi Arabia 0 2 29.3 116.1 122.4 228.7
(continued)
Downloaded by University of Ghana At 15:56 27 June 2016 (PT)

Bank Percentage of capital Percentage of assets Percentage of profits Percentage of ROE Percentage of ROA

33 Saudi Hollandi Bank, Saudi Arabia 0 2 49.2 122.9 175 608.3


34 Kuwait Finance House, Kuwait 0 2 25.2 125.5 202.4 470.6
35 Bank of Bahrain and Kuw, Bahrain 0 2 49.1 196 130.5 408.9
36 Arab African Intl Bank, Egypt 0 0 191 133.7 152.4
37 Intl Bank of Qatar, Qatar 0 0 187.8 138.7 140.5
38 First Gulf Bank, UAE 242 0 144.9 449.5 336.9
39 Union National Bank, UAE 0 2 7.4 147 229.3 398.9
40 Naional Bank of Dubai, UAE 0 2 40.8 150.8 232.4 685
41 United Gulf Bank, Bahrain 0 0 212.3 152.3 152.3
42 Bank of Kuwait & ME, Kuwait 0 2 57.9 173.2 158.7 529.7
43 Qatar Islamic Bank, Qatar 0 2 13.5 186.3 161.6 215.1
44 Oman Intl Bank, Oman 0 2 5.5 246.9 168.3 220.3
45 Saudi Invest. Bank, Saudi Arabia 0 2 17.2 172.2 262.6 515.8
46 Credit Populaire de Maroc, Morocco 0 2 64 173 222.8 1023
47 Commercial Bank of Dubai, UAE 0 2 42.2 195.1 181 396.9
48 Ahli United Bank, Bahrain 0 2 69.8 192.1 239 1276.7
49 Arab Bank Plc, Jordan 24.1 2 7.7 198.9 211.5 222.4
50 Sharjah Islamic Bank, UAE 221.2 0 287.2 319.5 208.8
51 RAK Bank, UAE 0 2 0.5 311.8 213.7 258
52 Credit de Maroc, Morocco 0 2 27.7 333.5 216.6 406.3
53 National Bank of Fuj., UAE 0 0 329.3 222.7 241.8
54 Burgan Bank, Kuwait 0 2 58.3 238.1 249.1 781.4
55 Commercial Bank of Qatar, Qatar 217.5 0 238.3 417.3 468.6
56 Al Ahli Bank of Kuwait, Kuwait 0 2 40 241.2 294.6 703.8
57 Soc. Gen. Mar. de Banques, Morocco 0 2 40.6 344.9 242.1 550.9
58 Blom Bank, Lebanon 0 2 65 255.2 312 1338.6
59 Shamil Bank of Bahrain, Bahrain 0 0 373.4 260.1 262.7
60 National Bank of Bahrain, Bahrain 0 2 45.1 323.4 265.5 612.7
61 The Housing Bank, Jordan 0 2 40.3 349.7 266.2 574.1
62 Bank Muscat, Oman 0 2 41.4 280.9 306.2 663.1
63 Gulf Intl Bank, Bahrain 0 2 55.2 305.2 444.7 1627.8
64 Banque Maroc. du Com. Ext., Morocco 0 2 59.5 330.6 370.2 1222.6
(continued)
top Arab banks
Benchmarking

efficiency

Table III.
815
Downloaded by University of Ghana At 15:56 27 June 2016 (PT)

816
107,6
IMDS

Table III.
Bank Percentage of capital Percentage of assets Percentage of profits Percentage of ROE Percentage of ROA

65 HSBC Bank Egypt, Egypt 0 2 23.2 386.9 344.9 344.9


66 National Bank of Oman, Oman 0 2 3.8 441.1 363.5 417.1
67 Banque de LHabitat, Tunisia 0 2 27.9 514.3 371.4 640.7
68 Arab Bank for Invest & FT, UAE 0 0 575.7 412 429.2
69 Gulf Intl Corp. GSC, Kuwait 27.6 0 424 638.1 838.5
70 Byblos Bank, Lebanon 0 2 67.1 486.9 425.8 1579
71 Attijariwafabank, Morocco 0 2 44.3 434.3 594.9 1605.2
72 Banque Lib-Franc., Lebanon 0 2 44.6 662.1 467.9 1076.3
73 Wahda Bank, Libya 0 2 5.9 644.4 483.5 594.2
74 Arab Banking Corp, Bahrain 0 2 31.8 490.6 692.4 1504
75 Gumhouria Bank, Libya 0 2 48.2 667.2 551 1261.3
76 Banque Intl Arabe de Tun., Tunisia 0 2 30 824.6 586.4 1029.1
77 Bank of Alexandria, Egypt 0 2 70 893.8 659.2 2823.4
78 Arab Intl Bank, Egypt 0 2 38.8 1,076.6 949.9 1677.5
79 Banque National Agricole, Tunisia 0 2 30.6 1,297.5 968.7 1650
80 Banque de la Med., Lebanon 0 2 54.1 1,268.9 1196.3 2804.4
81 Banque Misr, Egypt 0 2 84.1 1,343.2 1278.8 8791.9
82 National Bank of Egypt, Egypt 0 2 72.9 1,729.6 2220 11251.1
83 Banque Ext. dAlger, Algeria 0 2 72 2,365.1 1883.6 8169.8
84 Soc. Tunis. De Banques, Tunisia 0 2 42.6 6,096.3 4806.6 9572
85 Egyptian Arab Land Bank, Egypt 0 2 25.7 26,423.6 26423.6 55133.7
output observations for each bank. Slack variables are used to identify sources of Benchmarking
inefficiency. If a variable is slack, it is not acting as a constraint on production and top Arab banks
hence the bank is not fully efficient as inputs could be reduced without reducing
outputs. The slack is computed by comparing the input and output of each bank with efficiency
the inputs and outputs of its efficient reference set of banks. These efficient reference
banks are banks which operate under circumstances similar to that of the bank being
compared, but have 100 per cent efficiency. For example, bank 28 can become efficient 817
(increasing efficiency from 49 to 100) by increasing all outputs by the corresponding
slack amounts or decreasing all controllable inputs by corresponding slacks. The
possible savings indicated in Table III reveals the extent of inefficiencies prevalent
with respect to asset mismanagement in Arab banks. In certain banks, there is
considerable scope for reduction of assets.

Implications
Downloaded by University of Ghana At 15:56 27 June 2016 (PT)

Systematic benchmarking through efficiency measurement is one method managers


can use to benchmark the efficiency of their banks. The process of benchmarking may
be used to identify useful best business practices, innovative ideas, effective operating
procedures, and winning strategies that can be adopted by a bank to accelerate its own
progress by ensuring quality, productivity, and cost improvements (Chang and Lo,
2005; Seydel, 2006; Eisenberg and Hsieh, 2007). In contrast with piecemeal examination
of single performance indicators, global efficiency techniques used in this study can
offer Arab bank managers a rounded assessment of their banks performance. Unlike
targets that are based on individual performance measures, global efficiency measures
can offer local managers the freedom to set their own priorities, and to seek out
improvements along dimensions of performance where they believe that gains are
most readily secured. DEA results can also be used by Arab bank managers to support
other objectives, such as allocating finance or identifying the priorities for inspection
and improvement of performance.
As shown in Table II, several banks in our sample have been found to be operating at
a decreasing return to scale, and it is better for the management of these banks to refrain
from further investments in these banks, as their outputs do not seem to be proportional
to the inputs invested in them. On the other hand, banks found to be operating under an
increasing return to scale may prefer to expand their operations in the future. However, it
should be noted that because of the different production mix, the return to scale
relationships of the larger banks are likely to be nonlinear (Ramanathan, 2005), and this
may warrant further research.
One of the important implications of this study is that efficiency measures facilitate
the publication of league tables or rankings of entire Arab banks. Some authors
believe that such rankings nurture public interests in the performance of organizations,
promote accountability and stimulate a search for improvement (Hibbard et al., 2003).
Finally, it is hoped that Arab bank managers have the possibility to analyze
organizational practices of the peer groups and that they are able to improve their
future efficiency by adapting these practices for their inefficient banks.

Limitations and future research


Like any other study, the present study has several limitations that warrant more
research. First, it may not always be possible for a bank to ever become efficient
IMDS because several of the inputs may not be under the full control of management.
107,6 Therefore, it must be clear that some DEA targets might be impossible to be achieved
in practice. DEA results are obtained from the application of a mathematical algorithm,
without considering specific conditions and restrictions of a bank. It is in the hands of
managers to skilfully use these results as a support for decision-making. Second, the
selected variables in the present study might not be exhaustive, and the data set is
818 short. Staat (2001) has showed formally how DEA efficiency scores are affected by
sample size. Future studies may use larger sample size and panel data with different
sets of inputs and outputs to test the robustness of the results. Third, this study used a
cross-sectional data set to evaluate the efficiency of the top Arab banks. However,
Sengupta (1995) suggests that competitiveness or efficiency can better be evaluated
through analysis of average efficiencies across time. Future studies may use
longitudinal designs to assess time-varying efficiency. Fourth, as suggested by Bauer
et al. (1998), for the frontier-based efficiency scores to be useful, the estimated scores
Downloaded by University of Ghana At 15:56 27 June 2016 (PT)

should be positively correlated with traditional non-frontier based measures of


performance. Avkiran (1999) also suggests that assessment of banks operational
performance through DEA should be complemented by ratio analysis that measures
financial performance of the bank. He recommends that as a sound managerial
practice, profitability measures should be compared with DEA results and significant
disagreements investigated. Therefore, future studies should test the existence of
positive rank-order correlations between efficiency scores obtained from DEA analysis
and traditional efficiency measures such as financial ratios. This test would give
assurance that frontier measures are not simply artificial products of the assumptions
made regarding the underlying optimisation techniques used. Finally, it should be
noted that previous studies (Weill, 2004) have found a lack of robustness between
parametric and non-parametric approaches used to evaluate efficiency. Therefore,
future studies of Arab banking industry should compare efficiency scores based on
parametric and nonparametric techniques.

References
Al-Faraj, T., Alidi, A. and Bu-Bshait, K. (1993), Evaluation of bank branches by means of data
envelopment analysis, International Journal of Operations & Production Management,
Vol. 13, pp. 45-52.
Al-Shammari, M. and Salimi, A. (1998), Modeling the operating efficiency of banks:
a nonparametric methodology, Logistics Information Management, Vol. 11, pp. 5-12.
Alirezaee, M., Howland, M. and van de Panne, C. (1998), Sampling size and efficiency bias in
data envelopment analysis, Journal of Applied Mathematics and Decision Sciences, Vol. 2,
pp. 51-64.
Anderson, R., Fok, R., Zumpano, L. and Elder, H. (1998), The efficiency of franchising in the
residential real estate brokerage market, Journal of Consumer Marketing, Vol. 15,
pp. 386-96.
Athanassopoulos, A. (1997), Service quality and operating efficiency synergies for management
control in the provision of financial services: evidence from Greek bank branches,
European Journal of Operational Research, Vol. 98, pp. 300-12.
Athanassopoulos, A. and Curram, S. (1996), A comparison of data envelopment analysis and
artificial neural networks as tools for assessing the efficiency of decision making units,
Journal of the Operational Research Society, Vol. 47, pp. 1000-16.
Atiyya, H. (1992), Research in Arab countries, Organization Studies, Vol. 13, pp. 105-12. Benchmarking
Avkiran, N. (1999), An application reference for data envelopment analysis in branch banking: top Arab banks
helping the novice researcher, International Journal of Bank Marketing, Vol. 17, pp. 206-20.
Ayadi, F., Adebayo, A. and Omolehinwa, E. (1998), Bank performance measurement in a
efficiency
developing economy: an application of data envelopment analysis, Managerial Finance,
Vol. 24, pp. 5-16.
Banker, R., Charnes, A. and Cooper, W. (1984), Some models for estimating technical and scale 819
inefficiencies in data envelopment analysis, Management Science, Vol. 30, pp. 1078-92.
Banxia Holdings Ltd (2001), Frontier Analyst Professional (Version 3.0), Banxia Frontier Analyst
User Guide, Banxia Holdings Ltd, Glasgow.
Barros, C. and Athanassiou, M. (2004), Efficiency in European seaports with DEA: evidence
from Greece and Portugal, Maritime Economics & Logistics, Vol. 6, pp. 122-40.
Bauer, P., Berger, A., Ferrier, G. and Humphrey, D. (1998), Consistency conditions for regulator
analysis of financial institutions: a comparison of frontier efficiency methods, Journal of
Downloaded by University of Ghana At 15:56 27 June 2016 (PT)

Economics and Business, Vol. 50, pp. 85-114.


Bergendahl, G. (1998), DEA and benchmarks an application to Nordic banks, Annals of
Operations Research, Vol. 82, pp. 233-49.
Berger, A. and Humphrey, D. (1997), Efficiency of financial institutions: international survey
and directions for future research, European Journal of Operational Research, Vol. 98,
pp. 175-212.
Bhattacharya, A., Lovell, C. and Sahay, P. (1997), The impact of liberalization on the productive
efficiency of Indian commercial banks, European Journal of Operational Research, Vol. 98,
pp. 332-45.
Bonesronning, H. and Rattso, J. (1994), Efficiency variation among the Norwegian high schools:
consequences of equalization policy, Economics of Education Review, Vol. 13, pp. 289-304.
Casu, B. and Molyneux, P. (2003), A comparative study of efficiency in European banking,
Applied Economics, Vol. 35, pp. 1865-76.
Chaffai, M. (1997), Estimating input-specific technical inefficiency: the case of the Tunisian
banking industry, European Journal of Operational Research, Vol. 98, pp. 314-31.
Chang, D. and Lo, L. (2005), Measuring the relative efficiency of a firms ability to achieve
organizational benefits after ISO certification, Total Quality Management, Vol. 16,
pp. 57-69.
Charnes, A., Cooper, W. and Rhodes, E. (1978), Measuring the efficiency of decision making
units, European Journal of Operational Research, Vol. 2, pp. 429-44.
Chen, T. and Yeh, T. (1998), A study of efficiency evaluation in Taiwans banks, International
Journal of Service Industry Management, Vol. 9, pp. 402-15.
Cook, W. and Hababou, M. (2001), Sales performance measurement in bank branches,
Omega-International Journal of Management Science, Vol. 29, pp. 299-307.
Drake, L. and Howcroft, B. (1999), Measuring the relative efficiency of the selling function:
an application of data envelopment analysis to UK bank branches, Journal of Financial
Services Marketing, Vol. 3, pp. 297-315.
Drake, L. and Howcroft, B. (2002), An insight into the size efficiency of a UK bank branch
network, Managerial Finance, Vol. 28, pp. 24-36.
Eisenberg, L. and Hsieh, C.T. (2007), Implementing risk management systems with a
benchmark: a web-based DSS approach, International Journal of Electronic Finance, Vol. 1
No. 3, pp. 293-303.
IMDS Favero, C. and Papi, L. (1995), Technical efficiency and scale efficiency in the Italian banking
sector: a nonparametric approach, Applied Economics, Vol. 27, pp. 385-95.
107,6
Ford, N. (2006), Oil boom fuels banking growth, The Middle East, pp. 36-8, October.
Fukuyama, H. (1993), Technical efficiency and scale efficiency of Japanese commercial banks:
a nonparametric approach, Applied Economics, Vol. 25, pp. 1101-12.
Galagedera, D. and Silvapulle, P. (2003), Experimental evidence on robustness of data
820 envelopment analysis, Journal of the Operational Research Society, Vol. 54, pp. 654-60.
Gandjour, A., Kleinschmit, F., Littmann, V. and Lauterbach, K. (2002), An evidence-based
evaluation of quality and efficiency indicators, Quality Management in Health Care,
Vol. 10, pp. 41-52.
Giokas, D. (1991), Bank branches operating efficiency: a comparative application of DEA and
loglinear model, Omega-International Journal of Management Science, Vol. 19, pp. 549-57.
Golany, B. and Storbeck, J. (1999), A data envelopment analysis of the operational efficiency of
bank branches, Interfaces, Vol. 29, pp. 14-26.
Downloaded by University of Ghana At 15:56 27 June 2016 (PT)

Hardy, D. and Bonaccorsi d Patti, E. (2001), Bank reform and bank efficiency in Pakistan,
IMF Working Paper No. 01/38.
Hauner, D. (2005), Explaining efficiency differences among large German and Austrian banks,
Applied Economics, Vol. 37, pp. 969-80.
Hibbard, J., Stockard, J. and Tusler, M. (2003), Does publicizing hospital performance stimulate
quality improvement efforts?, Health Affairs, Vol. 22, pp. 84-94.
Ho, C. and Zhu, D. (2004), Performance measurement of Taiwans commercial banks,
International Journal of Productivity and Performance Management, Vol. 53, pp. 425-34.
Howland, M. and Rowse, J. (2006), Measuring bank branch efficiency using data envelopment
analysis: managerial and implementation issues, INFOR, Vol. 44, pp. 49-63.
Isik, I. and Hassan, M. (2002), Technical, scale and allocative efficiencies of Turkish banking
industry, Journal of Banking & Finance, Vol. 26, pp. 719-66.
Krishnasamy, G. (2003), Malaysian post merger banks productivity: application of Malmquist
productivity index, Managerial Finance, Vol. 30, pp. 63-74.
Lo, S. and Lu, W. (2006), Does size matter? Finding the profitability and marketability
benchmark of financial holding companies, Asia-Pacific Journal of Operational Research,
Vol. 23, pp. 229-46.
Manandhar, R. and Tang, C. (2002), The evaluation of bank branch performance using data
envelopment analysis: a framework, Journal of High Technology Management Research,
Vol. 13, pp. 1-17.
Middle East (2006), Middle East top 100 Arab banks, Middle East, pp. 33-5.
Miller, S. and Noulas, A. (1996), The technical efficiency of large bank production, Journal of
Banking & Finance, Vol. 20, pp. 495-509.
Mukherjee, A., Nath, P. and Pal, M. (2002), Performance benchmarking and strategic
homogeneity of Indian banks, International Journal of Bank Marketing, Vol. 20,
pp. 122-39.
Munksgaard, J., Pade, L. and Fristrup, P. (2005), Efficiency gains in Danish district heating: is
there anything to learn from benchmarking?, Energy Policy, Vol. 33, pp. 1986-97.
Oral, M. and Yolalan, R. (1990), An empirical study on measuring operating efficiency and
profitability of bank branches, European Journal of Operational Research, Vol. 46,
pp. 282-94.
Parkan, C. (1997), Measuring the efficiency of service operations: an application to bank Benchmarking
branches, Engineering Costs and Production Economics, Vol. 12, pp. 237-42.
top Arab banks
Phusavat, K. and Photaraon, W. (2006), Productivity/performance measurement: case
application at the government pharmaceutical organization, Industrial Management & efficiency
Data Systems, Vol. 106, pp. 1272-87.
Portela, M., Thanassoulis, E. and Simpson, G. (2004), Negative data in DEA: a directional
distance approach applied to bank branches, Journal of the Operational Research Society, 821
Vol. 55, pp. 1111-21.
Prajogo, D. (2007), The relationship between competitive strategies and product quality,
Industrial Management & Data Systems, Vol. 107, pp. 69-83.
Raab, R. and Lichty, R. (2002), Identifying sub-areas that comprise a greater metropolitan area:
the criterion of county relative efficiency, Journal of Regional Science, Vol. 42, pp. 579-94.
Ramanathan, R. (2005), Operations assessment of hospitals in the Sultanate of Oman,
International Journal of Operations & Production Management, Vol. 25, pp. 39-54.
Downloaded by University of Ghana At 15:56 27 June 2016 (PT)

Rao, M. and Miller, D. (2004), Expert systems applications for productivity analysis, Industrial
Management & Data Systems, Vol. 104, pp. 776-85.
Resti, A. (1997), Evaluating the cost-efficiency of the Italian banking system: what can be
learned from the joint application of parametric and nonparametric techniques, Journal of
Banking & Finance, Vol. 21, pp. 221-50.
Rickards, R. (2003), Setting benchmarks and evaluating balanced scorecards with data
envelopment analysis, Benchmarking: An International Journal, Vol. 10, pp. 226-45.
Sakar, B. (2006), A study on efficiency and productivity of Turkish banks in Istanbul Stock
Exchange using Malmquist DEA, Journal of American Academy of Business, Vol. 8,
pp. 145-55.
Sarkis, J. (2000), An analysis of the operational efficiency of major airports in the United States,
Journal of Operations Management, Vol. 18, pp. 335-51.
Sarkis, J. and Weinrach, J. (2001), Using data envelopment analysis to evaluate environmentally
conscious waste treatment technology, Journal of Cleaner Production, Vol. 9, pp. 417-27.
Schaffnit, C., Rosen, D. and Paradi, J. (1997), Best practice analysis of bank branches:
an application of DEA in a large Canadian bank, European Journal of Operational
Research, Vol. 98, pp. 269-89.
Seiford, L. and Zhu, J. (1999), Profitability and marketability of the top 55 US commercial
banks, Management Science, Vol. 45, pp. 1270-88.
Sengupta, J. (1995), Dynamics of Data Envelopment Analysis: Theory of Systems Efficiency,
Kluwer Academic Publishers, Boston, MA.
Seydel, J. (2006), Data envelopment analysis for decision support, Industrial Management &
Data Systems, Vol. 106, pp. 81-95.
Sherman, H. and Gold, F. (1985), Bank branch operating efficiency: evaluation with data
envelopment analysis, Journal of Banking & Finance, Vol. 9, pp. 297-315.
Sherman, H. and Ladino, G. (1995), Managing bank productivity using data envelopment
analysis (DEA), Interfaces, Vol. 9, pp. 297-315.
Sherman, H. and Rupert, T. (2006), Do bank mergers have hidden or foregone value? Realized
operating synergies in one bank merger, European Journal of Operational Research,
Vol. 168, pp. 253-68.
Smith, P. (1997), Model misspecification in data envelopment analysis, Annals of Operations
Research, Vol. 73, pp. 233-52.
IMDS Staat, M. (2001), The effect of sample size on the mean efficiency in DEA: a comment, Journal of
Productivity Analysis, Vol. 15, pp. 129-37.
107,6
Vassiloglou, M. and Giokas, D. (1990), A study of the relative efficiency of bank branches:
an application of data envelopment analysis, Journal of the Operational Research Society,
Vol. 41, pp. 591-9.
Weill, L. (2004), Measuring cost efficiency in European banking: a comparison of frontier
822 techniques, Journal of Productivity Analysis, Vol. 21, pp. 133-52.
Wu, D., Yang, Z. and Liang, L. (2006), Using DEA-neural network approach to evaluate branch
efficiency of a large Canadian bank, Expert Systems with Applications, Vol. 31, pp. 108-15.
Yildirim, C. (2002), Evolution of banking efficiency within an unstable macroeconomic environment:
the case of Turkish commercial banks, Applied Economics, Vol. 34, pp. 2289-301.
Zenios, C., Zenios, S., Agathocleous, K. and Soteriou, A. (1999), Benchmarks of the efficiency of
bank branches, Interfaces, Vol. 2, pp. 37-51.
Downloaded by University of Ghana At 15:56 27 June 2016 (PT)

Further reading
Avkiran, N. (1997), Productivity Analysis in the Services Sector with Data Envelopment Analysis,
North-Holland, Amsterdam.
Barr, R., Killgo, K. and Zimmet, S. (2002), Evaluating the productive efficiency and performance
of US commercial banks, Managerial Finance, Vol. 28, pp. 3-25.
Sun, S. and Lu, W. (2005), Evaluating the performance of the Taiwanese hotel industry using a
weight slacks-based measure, Asia-Pacific Journal of Operational Research, Vol. 22,
pp. 487-512.

Appendix. Partial list of the data used in the analysis

Bank Capital Assets Profits ROE ROA

National Comm Bank, Saudi Arabia 5,777 38,929 1,338 23.1 3.44
Arab Bank Plc, Jordan 3,749 27,484 503.2 13.42 1.83
Al Rajhi Bank, Saudi Arabia 3,597 25,377 1,504 41.8 5.9
Samba, Saudi Arabia 3,446 28,920 1,073 31.1 3.71
Riyadh Bank, Saudi Arabia 2,908 21,383 758 26.06 3.54
Abu Dhabi Comm Bank, UAE 2,344 15,653 523 22.31 3.34
Emirates Bank Intl, UAE 2,110 16,172 471 22.32 2.91
National Bank of Kuwait, Kuwait 2,109 21,234 736 34.9 3.46
First Gulf Bank, UAE 2,085 7,157 287 13.76 4.01
Saudi British Bank, Saudi Arabia 2,001 17,604 669 33.43 3.8
National Bank of Abu Dhabi, UAE 1,915 22,780 717 37.44 3.15
Kuwait Finance House, Kuwait 1,842 16,031 406 22.04 2.53
Arab Banking Corp, Bahrain 1,842 17,588 155 8.41 0.9
Mashreqbank, UAE 1,718 12,455 550 32.1 4.41
Arab National Bank, Saudi Arabia 1,692 18,022 488 28.84 2.71
Gulf Intl Bank, Bahrain 1,613 22,857 207 12.83 0.9
Qatar National Bank, Qatar 1,478 13,753 422 28.55 3.07
Banque Saudi Fransi, Saudi Arabi 1,473 15,934 410 27.83 2.57
Union National Bank, UAE 1,425 9,512 314 22.03 3.3
Table AI.
(continued)
Bank Capital Assets Profits ROE ROA
Benchmarking
top Arab banks
Saudi Invest. Bank, Saudi Arabia 1,417 10,569 284 20.04 2.68
Naional Bank of Dubai, UAE 1,356 13,998 300 22.12 2.14 efficiency
Gulf Intl Corp. GSC, Kuwait 1,321 7,256 135 10.22 1.86
Commercial Bank of Qatar, Qatar 1,295 6,094 194 15 3.2
Attijariwafabank, Morocco 1,199 12,725 131 10.9 1.03
Banque Intl Arabe de Tun., Tuni 194 2,688 14 7.21 0.52 823
Saudi Hollandi Bank, Saudi Arabi 980 10,670 281 28.67 2.63
Gulf Bank, Kuwait 955 8,932 302 31.62 3.38
Dubai Islamic Bank, UAE 871 11,708 290 33.3 2.47
Audi Sardar Bank, Lebanon 869 11,478 127 14.61 11
Credit Populaire de Maroc, Moroc 862 12,555 215 24.94 1.71
Ahli United Bank, Bahrain 819 13,872 196 23.93 1.41
Blom Bank, Lebanon 817 1,1918 161 19.7 1.35
Al Ahli Bank of Kuwait, Kuwait 811 6,876 167 20.59 2.42
Downloaded by University of Ghana At 15:56 27 June 2016 (PT)

Commercial Bank of Kuwait, Kuwai 804 8,023 288 35.82 3.59


Banque Maroc. du Comm. Ex., Moro 704 8,174 124 17.6 1.51
Bank Muscat, Oman 668 5,185 137 20.5 2.64
Burgan Bank, Kuwait 624 6,472 150 24.03 2.31
Bank Al Jazeera, Saudi Arabia 620 3,778 234 37.74 6.2
Intl Banking Corp, Bahrain 585 2,403 494 84.44 20.55
Sharjah Islamic Bank, UAE 559 1,443 51 9.12 3.53
Banque Misr, Egypt 535 14,719 31 5.8 0.21
Commercial Bank of Dubai, UAE 530 4,033 150 28.3 3.72
Soc. Tunis. de Banques, Tunisia 321 3,573 4 1.24 0.11
Bank of Kuwait & ME, Kuwait 524 5,520 160 30.53 2.9
Doha Bank, Qatar 516 4,184 217 42.05 5.18
Banque de Caire, Egypt 506 7,40 18 3.55 0.23
Qatar Islamic Bank, Qatar 485 2,624 140 28.86 5.33
Bank of Sharjah, UAE 469 1,565 164 34.96 10.48
Arab Intl Bank, Egypt 459 3,652 32 6.97 0.89
Byblos Bank, Lebanon 458 6,784 64 13.9 0.94 Table AI.

Corresponding author
Mohamed Mostafa can be contacted at: mostafa@usa.com

To purchase reprints of this article please e-mail: reprints@emeraldinsight.com


Or visit our web site for further details: www.emeraldinsight.com/reprints
This article has been cited by:

1. Anwar Al-Gasaymeh. 2016. Bank efficiency determinant: Evidence from the gulf cooperation council
countries. Research in International Business and Finance 38, 214-223. [CrossRef]
2. Mini Kundi Department of Management Studies, Indian Institute of Technology, Delhi, New Delhi,
India Seema Sharma Department of Management Studies, Indian Institute of Technology, Delhi, New
Delhi, India . 2016. Efficiency of glass firms in India: an application of data envelopment analysis. Journal
of Advances in Management Research 13:1, 59-74. [Abstract] [Full Text] [PDF]
3. Mohammad Nourani, Evelyn Shyamala Devadason, Qian Long Kweh, Wen-Min Lu. 2016. Business
excellence: the managerial and value-creation efficiencies of the insurance companies. Total Quality
Management & Business Excellence 1-18. [CrossRef]
4. Congcong Yang, Alfred Taudes, Guozhi Dong. 2015. Efficiency analysis of European Freight Villages:
three peers for benchmarking. Central European Journal of Operations Research . [CrossRef]
5. Jonchi Shyu, Pang-Tien Lieu, Wei Chang. 2015. How the environment determines banking efficiency:
Downloaded by University of Ghana At 15:56 27 June 2016 (PT)

a comparison of banking firms in Taiwan, Hong Kong, and Mainland China. International Transactions
in Operational Research 22:4, 757-770. [CrossRef]
6. Varun Mahajan Department of Humanities and Social Sciences, Indian Institute of Technology, Roorkee,
India D.K. Nauriyal Department of Humanities and Social Sciences, Indian Institute of Technology,
Roorkee, India S.P. Singh Department of Humanities and Social Sciences, Indian Institute of Technology,
Roorkee, India . 2014. Technical efficiency analysis of the Indian drug and pharmaceutical industry.
Benchmarking: An International Journal 21:5, 734-755. [Abstract] [Full Text] [PDF]
7. Muhammet Enis Bulak Department of Industrial Engineering, Fatih University, Istanbul, Turkey Ali
Turkyilmaz Department of Industrial Engineering, Fatih University, Istanbul, Turkey . 2014. Performance
assessment of manufacturing SMEs: a frontier approach. Industrial Management & Data Systems 114:5,
797-816. [Abstract] [Full Text] [PDF]
8. Yusuf Tansel . 2014. A TOPSIS based design of experiment approach to assess company ranking. Applied
Mathematics and Computation 227, 630-647. [CrossRef]
9. Ming-Hung Shu Department of Industrial Engineering and Management, National Kaohsiung University
of Applied Sciences, Kaohsiung, Taiwan Jan-Yee Kung Department of Industrial Engineering and
Management, National Kaohsiung University of Applied Sciences, Kaohsiung, Taiwan Thanh-Lam
Nguyen Graduate Institute of Mechanical and Precision Engineering, National Kaohsiung University
of Applied Sciences, Kaohsiung, Taiwan Bi-Min Hsu Department of Industrial Engineering and
Management, Cheng Shiu University, Kaohsiung, Taiwan . 2013. Dynamically monitoring the service
efficiency with tracking signals. Industrial Management & Data Systems 113:9, 1334-1350. [Abstract]
[Full Text] [PDF]
10. Ying-Chen Hsu, Cheng-Chi Chung, Hsuan-Shih Lee, H. David Sherman. 2013. Evaluating and
Managing Tramp Shipping Lines Performances: A New Methodology Combining Balanced Scorecard
and Network DEA. INFOR: Information Systems and Operational Research 51:3, 130-141. [CrossRef]
11. Vidyashankar GourishankarDepartment of Management Studies, Indian Institute of Technology Madras,
Chennai, India Prakash Sai LokachariDepartment of Management Studies, Indian Institute of Technology
Madras, Chennai, India. 2012. Benchmarking educational development efficiencies of the Indian states:
a DEA approach. International Journal of Educational Management 26:1, 99-130. [Abstract] [Full Text]
[PDF]
12. Sunil KumarFaculty of Economics, South Asian University, New Delhi, India. 2011. State road transport
undertakings in India: technical efficiency and its determinants. Benchmarking: An International Journal
18:5, 616-643. [Abstract] [Full Text] [PDF]
13. Darwish Abdulrahman Yousef. 2011. Operations research/management science in the Arab world:
historical development. International Transactions in Operational Research 18:1, 53-69. [CrossRef]
14. Jos Humberto Ablanedo-Rosas, Hongman Gao, Xiaochuan Zheng, Bahram Alidaee, Haibo Wang. 2010.
A study of the relative efficiency of Chinese ports: a financial ratio-based data envelopment analysis
approach. Expert Systems 27:5, 349-362. [CrossRef]
15. Ali Emrouznejad, Abdel Latef Anouze. 2009. A note on the modeling the efficiency of top Arab banks.
Expert Systems with Applications 36:3, 5741-5744. [CrossRef]
16. D.M. Mathuva. 2009. Capital Adequacy, Cost Income Ratio and the Performance of Commercial Banks:
The Kenyan Scenario. The International Journal of Applied Economics and Finance 3:2, 35-47. [CrossRef]
17. HsinHung ChenDepartment of Business Administration, Cheng Shiu University, FengShan, Taiwan.
2008. Stock selection using data envelopment analysis. Industrial Management & Data Systems 108:9,
Downloaded by University of Ghana At 15:56 27 June 2016 (PT)

1255-1268. [Abstract] [Full Text] [PDF]


18. Darwish Abdulrahman YousefDepartment of Business Administration, College of Business and
Economics, United Arab Emirates University, AlAin, United Arab Emirates. 2008. Operations research/
management science in the United Arab Emirates. Education, Business and Society: Contemporary Middle
Eastern Issues 1:4, 255-266. [Abstract] [Full Text] [PDF]
19. Sunil KumarPunjab School of Economics, Guru Nanak Dev University, Amritsar, India Rachita
GulatiPunjab School of Economics, Guru Nanak Dev University, Amritsar, India. 2008. Evaluation of
technical efficiency and ranking of public sector banks in India. International Journal of Productivity and
Performance Management 57:7, 540-568. [Abstract] [Full Text] [PDF]
20. Reggie Davidrajuh. 2008. Building a fuzzy logic based tool for e-readiness measurement. Electronic
Government, an International Journal 5:2, 120. [CrossRef]