29
NAŠE GOSPODARSTVO / OUR ECONOMY 69 (1) 2023 Fotova Čiković, K., Cvetkoska, V., Keček, D.
The Impact of COVID-19 on Bank Efficiency in the Western 
Balkans: A DEA Approach
Katerina Fotova Čiković
a*
, Violeta Cvetkoska
b
, Damira Keček
a
a
University North, Trg dr. Žarka Dolinara 1, 48000 Koprivnica, Croatia
b
Ss. Cyril and Methodius University in Skopje, Faculty of Economics Skopje, Blvd. Goce Delchev 9 V, 1000 Skopje, North 
Macedonia
kcikovic@unin.hr, vcvetkoska@eccf.ukim.edu.mk, dkecek@unin.hr
                        
ARTICLE INFO
         
Original Scientific Article
Article History:
Received December 2022
Revised February 2023
Accepted February 2023
         
JEL Classification:
G21, C67
         
Keywords:
Banking sector
Data envelopment analysis
COVID-19
Western Balkans
         
UDK: 336.71:616-036.22
DOI: 10.2478/ngoe-2023-0004
         
Cite this article as: Fotova Čiković, K., 
Cvetkoska, V., & Keček, D. (2023). The 
Impact of COVID-19 on Bank Efficiency 
in the Western Balkans: A DEA Approach. 
Naše Gospodarstvo/Our Economy, 69(1), 
29-42. DOI: 10.2478/ngoe-2023-0004.
         
©2022 The Authors. Published by 
Sciendo on behalf of University of 
Maribor, Faculty of Economics and 
Business, Slovenia. This is an open access 
article under the CC BY-NC-ND license 
(http://creativecommons.org/licenses/
by-nc-nd/4.0/). 
         
Abstract
This paper aims to assess and compare the relative efficiency of commercial 
banks in six Western Balkan developing countries (North Macedonia, Serbia, 
Montenegro, Bosnia and Herzegovina (B&H), Kosovo
1
, and Albania), using the 
leading nonparametric methodology Data Envelopment Analysis (DEA) for the 
period 2016-2020, and investigating the impact of the COVID-19 pandemics 
on the performance of banking systems in these countries. The output-
oriented DEA model has been implemented with interest expenses and non-
interest expenses as inputs and interest revenue and non-interest revenue 
as outputs. Our findings show that the average efficiency of the six Western 
Balkan banking systems differs, and Kosovo’s banking system has noted the 
highest average efficiency in the whole observed period (2016–2020). In 
contrast, the banking system in Bosnia and Herzegovina noted the lowest 
average efficiency. The COVID-19 pandemic decreased the relative efficiency 
of the banking sectors in all six of the Western Balkan countries observed, 
except Kosovo. However, additional research that includes all pandemic 
years is recommended to assess the impact of the COVID-19 pandemic on 
the banking systems in the Western Balkans. This study provides invaluable 
insights for academic members, banking management and regulatory bodies, 
governments, and the interested public. It is the first empirical study that 
includes a glimpse of the first impact of the COVID-19 pandemic on banking 
systems in the developing countries of the Western Balkans.
Introduction
With the disintegration of the Socialist Federal Republic of Yugoslavia (SFRY) 
and the Union of Soviet Socialist Republics (USSR) at the end of the 1980s 
and the beginning of the 1990s, the countries of Central and Eastern Europe 
abandoned the system of "planned economies". They began to rapidly build 
economic systems based on market principles and rules that were compatible 
1
 All references to Kosovo in this document should be understood to be in the context of United Nations' 
Security Council Resolution 1244 (1999).
* 
Corresponding author

30
NAŠE GOSPODARSTVO / OUR ECONOMY 69 (1) 2023 Slijepčević, S.
and comparable with those of developed countries 
(Cvetkoska et al., 2020). Except Albania, the six Western 
Balkan countries studied were all part of SFR Yugoslavia 
until 1991. Therefore, these countries underwent a transition 
process that included "the transition of ownership, of the 
market, and of the political structure" (Vojnić, 1995). This, of 
course, implied a fundamental restructuring of their banking 
sectors as well. Namely, in the past economic system, 
banks were "financial organizations that in most cases 
were designed to support the central planning apparatus" 
(Cvetkoska et al., 2021). However, other authors (Radzic & 
Yuce, 2008) claim that "Yugoslavia’s banking market differed 
from other planned economies in the sense that its banks 
participated in commercial as well as financial activities for 
enterprises."
In this transition process, the Western Balkan countries 
experienced quite some challenges in restructuring 
their banking sectors, such as numerous reorganizations 
(mergers, acquisitions, changes in management, etc.) (Toci, 
2009). The main characteristic of these countries’ financial 
sectors is that, unlike in developed countries, they are 
"largely dominated by banks rather than the capital market." 
The financial systems of the Western Balkan countries are 
dominated by banks (banks hold 83–98% of financial sector 
assets, except for Kosovo with 65%), which are mostly foreign-
owned. Banks operate on a traditional business model 
characterized by the taking of deposits and the provision 
of loans. Their banking systems are "well capitalized and 
liquid," but challenges regarding asset quality and indirect 
credit risk remain. Furthermore, they "seem less prone to 
financial stress from maturity mismatches" (ECB, 2019).
This paper aims to assess the efficiency of the banking 
sectors in six Western Balkan countries and then analyze 
and comment on the impact of the COVID-19 pandemic on 
their efficiency. This study addresses the inevitable question: 
How has the COVID-19 pandemic affected the efficiency and 
performance of banks in the Western Balkan?
The efficiency of any business is crucial for its sustainability 
and growth. It refers to "the minimization of inputs used 
by a firm to produce a given level of outputs or the 
maximization of outputs produced by a given set of inputs 
under a given state of technology" (Toci, 2009). The leading 
non-parametric methodology, DEA, is "the most widely 
used efficiency evaluation method" (Radojicic et al., 2018). 
As a result, the DEA-window technique with two inputs and 
two outputs was used in this study from 2016 to 2020 (5 
consecutive years). The findings reveal that the average 
efficiency of the six Western Balkan banking systems differs 
over the observed period, and Kosovo’s banking system has 
noted the highest average efficiency in the whole observed 
period (2016-2020). In contrast, the banking system 
in Bosnia and Herzegovina noted the lowest average 
efficiency. The COVID-19 pandemic decreased the relative 
efficiency of the banking sectors in all six of the Western 
Balkan countries observed, except Kosovo. These findings 
will be important to bank management, regulatory bodies, 
investors, and the public.
The rest of the paper is structured as follows: After the 
introduction, Section 2 presents a literature review of 
studies concerning bank efficiency with DEA for each of 
the six Western Balkan countries. Section 3 presents the 
methodology used, variables, and data. Section 4 presents 
and analyses the findings. Section 5 discusses the results, 
while Section 6 concludes the paper.
Literature Review
The literature on the efficiency of banks in the Western Balkans 
is still "relatively scarce, though growing, with the parametric 
approach dominating the non-parametric approach" (Toci & 
Hashi, 2013). Data Envelopment Analysis is one of the leading 
non-parametric approaches to measuring efficiency. Since its 
introduction in 1978 by Charnes, Cooper, and Rhodes, the DEA 
methodology has gained popularity and spread its application 
in many industries. Moreover, DEA has been implemented in 
numerous studies regarding banking worldwide. According to 
Emrouznejad & Yang's (2018) study, banking is one of the top 
five areas in which DEA is most commonly applied, along with 
agriculture, supply chain, transportation, and public policy.
This study conducts an in-depth review of the literature on 
bank efficiency studies with DEA approach in each of the 
Western Balkan countries examined (eight in Serbia, six in 
North Macedonia, none in Montenegro, three in Albania, three 
in Kosovo, five in Bosnia and Herzegovina, and six cross-country 
studies that include any of these Western Balkan countries). 
Table 1 provides an overview of the scholarly literature on 
bank efficiency in each of the countries studied.
Methodology
In this paper, we assess and compare the relative efficiency of 
commercial banks in each of the six Western Balkan developing 
countries (North Macedonia, Serbia, Kosovo, Montenegro, 
Albania, and Bosnia and Herzegovina) using the DEA window 
analysis technique for the period 2016-2020, in order to assess 
the impact of the COVID-19 pandemic on the efficiency and 
performance of the six countries' banking sectors.

31
NAŠE GOSPODARSTVO / OUR ECONOMY 69 (1) 2023 Fotova Čiković, K., Cvetkoska, V., Keček, D.
Data Envelopment Analysis is one of the most prominent 
nonparametric methodologies extensively used in different 
industries and "applied in more than 6500 publications" 
(Radojicic et al., 2018) since its introduction in 1978. DEA is a 
technique of operations research that is used in the efficiency 
evaluation of homogeneous units called decision-making units 
(DMUs) (Braimllari & Benga, 2019). The DMUs can be banks, 
hospitals, police stations, tax offices, defense bases, schools, 
insurance companies, libraries, university departments, etc. DEA 
is a "productive efficiency measurement method" (Rabar, 2013) 
with the use of the "same multiple inputs and multiple outputs" 
(Jemrić & Vujčić, 2002). The basic DEA model introduced by 
Charnes, Cooper, and Rhodes (1978) assumed a constant return 
to scale (CRS). This model was extended in the study of Banker, 
Charnes, and Cooper (1984), who introduced the BCC DEA model 
with a variable return to scale (VRS) (Anayiotos et al., 2010). 
Furthermore, a DEA model can be input- or output-oriented. The 
input orientation focuses on the minimization of inputs while 
satisfying at least the given output levels. In contrast, the output 
orientation attempts to maximize outputs without requiring 
more of any of the observed input values (Cooper et al., 2007).
DEA has several advantages over parametric methodologies, 
including "a single unambiguous measure of performance, 
the handling of multiple input and output variables in 
different measurement units, and without any assumptions 
on data distribution; a focus on DMU best practices and 
recommendations by calculating a virtual DMU efficiency for 
each DMU under evaluation" (Mihailović et al., 2009); "the 
ability to include more than one output and not set assuages" 
(Cvetkoska & Fotova Čiković, 2021). The DEA methodology 
provides a separate evaluation of the efficiency of each DMU 
in the sample relative to its reference set, thus "providing a 
relative measurement of efficiency for every single DMU" 
(Anayiotos et al., 2010). 
In this study, the leading nonparametric DEA technique 
Window analysis under the VRS assumption, based on the 
Banker-Charnes-Cooper (BCC) model, has been used, and 
more specifically, the operating (income-based) approach 
has been applied following Cvetkoska & Fotova Čiković 
(2020). The input and output variables were chosen using 
the model in Banker et al. (2010) and Cvetkoska & Fotova 
Čiković (2020). Therefore, in this study, two input and two 
output variables have been selected: interest expenses and 
non-interest expenses as inputs and interest revenue and 
non-interest revenue as outputs. The data for those variables 
have been drawn from the official financial reports of each 
commercial bank. The detailed specification of the variables 
is presented in Table 2.
Table 1 
Scholarly literature regarding Western Balkan bank efficiency
Cross-country studies North Macedonia
Bosnia and 
Herzegovina
Serbia Albania Kosovo Montenegro
Toci (2009)
Giustianiani & Ross 
(2008)
Efendić & Avdić 
(2011)
Mihailović et al. 
(2009)
Spaho (2015)
Zogjani & 
Kelmendi (2015)
none
Anayiotos et al. (2010) Cvetkoska (2010)
Abu-Alkheil et al. 
(2012)
Marković et al. 
(2015)
Spaho, Mitre & 
Shehu (2015)
Zogjani, Mazelliu 
& Humolli (2018)
Toci & Hashi (2013)
Micajkova & Poposka 
(2013)
Memić & Škaljić-
Memić (2013)
Radojicic et al. 
(2015)
Braimllari & 
Benga (2019)
Sahiti & Sahiti 
(2021)
Varesi (2015)
Naumovska & 
Cvetkoska (2014)
Efendić (2017)
Popović, Stank-
ović & Marjano-
vić (2017)
Cvetkoska, Fotova 
Čiković & Tasheva 
(2021)
Naumovska & 
Cvetkoska (2016)
Husejinović (2019)
Lukić, Sokić & 
Kljenak (2017)
Milenković et al. 
(2022)
Fotova Čiković & 
Cvetkoska (2017)
Savić et al. 
(2017)
Cvetkoska and Fotova 
Čiković (2020)
Radojicic et al. 
(2018)
Bošković & 
Krstić (2020)
Source: Authors’ research

32
NAŠE GOSPODARSTVO / OUR ECONOMY 69 (1) 2023 Fotova Čiković, K., Cvetkoska, V., Keček, D.
The envelopment form of the output-oriented BCC DEA model 
is given in (1) – (5):
()max
,
BCCO o
B
B



(1)
subject to Xx  
0 (2)

B
yY
0
0 
(3)
e 1
(4)
 0
(5)
where 
η
B is a scalar. The input data for DMUj (j = 1, …, n) are 
(x1j, x2j, …, xnj), and the output data are (y1j, y2j, …, ynj); the 
data set is given by two matrices X and Y, where X is the input 
data matrix, and Y is the output data matrix, λ is a column 
vector and all its elements are non-negative, while e is a row 
vector and all its elements are equal to 1 (Cooper et al., 2007, 
pp. 22, pp. 91–92), (Cvetkoska & Barišić, 2017, pp. 33–34). The 
BCC efficiency requires the fulfilment of two conditions: (1) the 
result of the BCC efficiency to be equal to 1 (100%), and (2) all 
slacks to have a zero value (Cvetkoska et al., 2021).
The "window analysis" technique is defined as "subdividing 
the entire period of data into smaller windows to be analyzed 
separately" and is often used to "analyze the evolution of 
efficiency trends over time." With the window analysis, every 
window is separately analyzed with DEA (Paradi et al., 2018). 
Moreover, the window analysis is also treated as "a method for 
studying the stability of DEA results" because it involves the 
removal of entire sets of observations and their replacement 
by other (previously not considered) observations (Cooper et 
al., 2007). This technique is one of the methods used "to verify 
productivity change over time," which makes it very applicable 
in "detecting performance trends of a decision-making unit 
over time" (Repkova, 2014).
The DEA analysis's most significant advantage is that it is 
advantageous with small samples and works well when 
multiple inputs and outputs are selected. This method increases 
the number of observed DMUs and introduces dynamicity and 
a time dimension into the DEA model (Cvetkoska & Fotova 
Čiković, 2020). According to Repkova (2014), the DEA window 
analysis technique can be perceived as "a special case of a 
sequential analysis." Its advantage is reflected in the possible 
comparison and contrast of the performance and efficiency 
of each DMU against itself and other banks, as well as its 
performance in other periods in addition to the performances 
of other units.
This method's most frequently mentioned limitation is 
that entities in the first and last periods are not tested as 
frequently as others (Cvetkoska & Fotova Čiković, 2020). Due 
to the notion that efficiency changes over time, we found that 
the DEA window analysis technique is the most suitable and 
appropriate technique for this study.
Research Analysis and Results
The six samples consist of a total of 92 banks that include 
13 Macedonian banks (excluding one commercial bank 
that ceased to operate in August 2020), 24 Serbian banks, 8 
Kosovo banks (out of a total of 11, since Komercijalna Banka 
AD Beograd is a branch of the Serbian Komercijalna Banka AD 
Beograd; Turkiye Cumhuriyeti Ziraat Bankasi does not publish 
its statements on its website; and Credins Bank is a newly 
founded bank that started operating on March 17, 2020), 12 
Montenegrin banks, 12 Albanian banks, and 23 Bosnia and 
Herzegovina banks. The banks’ efficiency and performance 
over the past five years (2016–2020) are analyzed, and then 
a three-year window is selected. The length of the window 
was determined using Sun's (1988) formula, which is n = 92, 
Table 2 
Inputs and outputs of the DEA model
Character of variable Variables Elements of the variables
INPUT interest expenses (I1)
non-interest expenses (I2)
Expenses on fees and commissions
General administrative expenses and depreciation
Expenses on value adjustments and provisions
Other operating expenses
OUTPUT
interest revenues (O1)
non-interest revenues (O2)
Income from fees and commissions
Other operating income
Source: Authors’ research

33
NAŠE GOSPODARSTVO / OUR ECONOMY 69 (1) 2023 Fotova Čiković, K., Cvetkoska, V., Keček, D.
five years from 2016 to 2020 (k = 5), the length of the window 
is three years (p = 3), and the number of windows is three 
(w = k + p + 1 = 5 + 3 + 1 = 3). In each window, there are 
276 banks, and the number of "different" banks is 828. The 
national development banks have been excluded from the 
analysis due to their specific development and support roles 
in the financial market. The panel data used for all six samples 
is strongly balanced and is drawn from the audited financial 
reports for 2016–2020.
Every window includes three years (for example, window 1 
includes three years: 2016, 2017, and 2018; whereas, in the 
next window (window 2), the data for 2016 is removed and 
the data for 2019 is added; and the final window (window 3) 
begins in 2018 and ends in 2020, as shown in Table 3.
Table 3 
Inputs and outputs of the DEA model
Window 1 2016 2017 2018
Window 2 2017 2018 2019
Window 3 2018 2019 2020
Source: Authors’ research
Table 4 shows the average relative efficiency of the banking 
system in each analyzed country by year, with yellow 
highlighting the average efficiency greater than 0.9000 
(90.00%). These results show that Kosovo had the highest 
average relative efficiency in 2018 (0.9385, i.e., 93.85%). 
Additionally, the average efficiency of the banking system of 
each country for the whole period is presented, based on which 
we can state that Kosovo is the leader with 90.01%. Also, the 
column sparkline (in green) for the average efficiency for the 
whole period for each country's banking system is presented. 
The previous column has a line sparkline with red markers at 
the lowest and the highest points. Furthermore, we show the 
average efficiency of all systems for banking efficiency by year 
(the highest is achieved in 2018 (83.61%)) as well as a column 
sparkline (orange colour) for those efficiencies.
The results show that the Kosovo banking system is the most 
efficient (90,01%), while the B&H banking system is the 
least efficient (71.84%) over the entire 5-year period studied. 
However, the crucial analysis explores whether and how the 
COVID-19 pandemic has impacted these Western Balkan 
economies and, thus, their banking systems.
The obtained results for the Serbian banking sector show 
that the lowest efficiency (62.15%) was observed in the first 
examined year (i.e., in 2016), while the highest efficiency 
(78.04%) was observed in 2018. An evident decrease in 
efficiency after 2018 resulted in an efficiency score of 71.47% 
in 2020. The results regarding each bank indicate that large 
banks in Serbia note the highest efficiency (from 89.71% for 
Komercijalna Banka AD Beograd to 99.55% for Banca Intesa AD 
Beograd), whereas the small-sized banks (e.g., Mobi Banka AD 
Beograd and Api Bank AD Beograd) note the lowest efficiency 
(0.3206 and 0.3650, respectively). The study's findings are 
also consistent with those of Savic et al. (2012), who claimed 
that the majority of Serbian commercial banks have efficiency 
scores ranging from 60% to 70%. 
Table 4
The average efficiency of the Western Balkan banking systems (2016-2020)
Western Balkan Country
Average efficiency of the whole banking system
Line sparkline
Average efficiency for the 
whole period for each 
country's banking system
Year
2016 2017 2018 2019 2020
North Macedonia 0,8041 0,8112 0,8160 0,7643 0,7771 0,7945
Serbia 0,6215 0,7566 0,7804 0,7491 0,7147 0,7245
Montenegro 0,8475 0,9170 0,9250 0,9125 0,8969 0,8998
Albania 0,7495 0,8361 0,8306 0,8143 0,7753 0,8012
Kosovo 0,9242 0,8977 0,9385 0,8396 0,9008 0,9001
Bosnia & Herzegovina 0,5932 0,6939 0,7260 0,7998 0,7792 0,7184
Average efficiency for the 
whole period of all analyzed 
banking systems
0,7567 0,8188 0,8361 0,8133 0,8073
Source: Authors’ research

34
NAŠE GOSPODARSTVO / OUR ECONOMY 69 (1) 2023 Fotova Čiković, K., Cvetkoska, V., Keček, D.
The results for North Macedonia’s commercial banks show that 
efficiency has been slowly increasing in the period 2016–2018, 
reaching its highest score of 81.60% in 2018, and after that 
decreasing in the period 2019–2020 (the slight increase from 
76.43% in 2019 to 77.70% in 2020 is insignificant). When 
analyzing the average efficiency per term per bank (Appendix 1), 
it is evident that the group of large banks seems to outperform 
regarding efficiency (except for HalkBank AD Skopje, which 
notes an efficiency score of 69.98%). Moreover, small-sized 
banks note the lowest efficiency in the sample. Ohridska Banka 
AD Skopje is the least efficient bank in the sample, with an 
efficiency score of 63.15%, which is in line with the previous 
findings of Cvetkoska et al. (2021). Namely, this bank has noted 
"high impairment costs of financial assets in 2016," including 
the portfolio's growth provisions, as well as additional 
reservations at the discretion of the relevant bodies of the 
bank. It was acquired by Steiermaerkische Sparkasse Group in 
2019 (Cvetkoska et al., 2021). In contrast, Stopanska Banka AD 
Skopje is the most efficient bank, with an efficiency score of 1 in 
4 out of 5 years and an average efficiency of 99.89%.
The Montenegrin banking system showed its lowest efficiency 
in 2016 (84.75%) and its highest in 2018 (92.50%). It has 
noted a decrease in efficiency in 2019 and 2020. However, 
Montenegro’s banking system is the second most efficient 
in the observed period, with 89.98%. The results for each 
commercial bank show that the largest banks are the most 
efficient. Crnogorska Komercijalna Banka AD Podgorica is the 
most efficient Montenegrin bank, with four years, a relative 
efficiency score of one, and an average efficiency of 98.80%.
The DEA window analysis technique results for the Albanian 
banking system show that 2016 was the least efficient year 
(with an efficiency score of 74.95%), whereas 2018 was the 
most efficient year (83.06%). What is captivating is that this is 
the only sample in which a bank has been relatively efficient 
in all of the observed years (Raiffeisen Bank S.A.), and this is a 
bank that belongs to the large banks’ group. The large banks 
note the highest efficiency in the Albanian banking system 
(ranging from 89.38% to 100.0%).
The banking system in Kosovo shows inconsistent results, and 
it is the only Western Balkan country that noted a significant 
increase in efficiency from 83.95% in 2019 to 90.08% in 2020. 
Furthermore, it is the most efficient banking system in the 
sample, with an average efficiency of 90.01% for the whole 
observed period. The most efficient banks are large banks (in 
particular, ProCredit Bank, which notes a relative efficiency 
score of 1 in 3 out of 5 years). The least efficient bank is Banka 
Per Biznis, with an average efficiency score of 0.78462. These 
findings do not correspond with Varesi (2015) findings, where 
the Kosovar banking system was the least efficient (together 
with the Albanian).
Interestingly, the DEA window analysis results for Bosnia and 
Herzegovina show that no bank in this sample is relatively 
efficient in all of the examined windows and/or years. 
The average efficiency result for the entire period under 
consideration is 71.84%. The lowest efficiency has been 
recorded in 2016 (59.32%), whereas the highest was in 2019 
(79.97%), followed by a slight fall to 77.92% in 2020. The most 
efficient banks are the large ones, except Intesa Sanpaolo Bank, 
which noted an average efficiency of 76.60%. This result is due 
to the overall low efficiency in the first studied year, 2016. The 
findings are consistent with Efendic and Avdic (2011)'s finding 
that B&H banks are "on average significantly inefficient," but 
not with Memić & Škaljić-Memić (2013)'s claim that small 
banks are more efficient than large banks. 
Except for North Macedonia and Kosovo, which saw a slight 
increase in efficiency (from 76.42% in 2019 to 77.70% in 2020) 
and a significant increase in efficiency (from 83.95% in 2019 
to 90.08% in 2020), all of the Western Balkan countries studied 
saw a decrease in efficiency in 2019 and 2020.The obtained 
results also show that the highest efficiency scores mainly 
were in 2018 (for North Macedonia, Serbia, Montenegro, 
and Kosovo; 2019 for Bosnia and Herzegovina, and 2017 for 
Albania), i.e., before the COVID-19 pandemic, whereas the 
lowest efficiency was noted in the first studied year: 2016 (and 
2019 for North Macedonia and Kosovo).
Furthermore, the findings indicate that, in general, small-sized 
banks are least efficient. In contrast, large banks are most 
efficient in the Western Balkan countries, which is in line 
with the findings of Guistininani & Ross (2008), Micajkova 
& Poposka (2013), Fotova Čiković & Cvetkoska (2017), and 
Cvetkoska et al. (2021).
Discussion
In this paper, the relative efficiency of the banking sectors of six 
developing Western Balkan countries - North Macedonia, Serbia, 
Montenegro, Kosovo, Albania, and Bosnia and Herzegovina 
- has been evaluated by using the window DEA analysis 
technique for the period from 2016 to 2020. A balanced panel 
of data has been used for the six samples. The output-oriented 
BCC DEA window model results show a decrease in efficiency 
in most (four out of six) of the examined countries in 2020, 
which could potentially prove that the COVID-19 pandemic 
influenced these countries' banking systems. Even though 
it took a time lag for the spillover effects of the GFC (Global 
Financial Crisis) in 2008/09 in small and developing countries 
with underdeveloped financial systems (such as the sample 
countries of the Western Balkans), the COVID-19 pandemics 
and thus the crisis year (2020) affected all countries regardless 

35
NAŠE GOSPODARSTVO / OUR ECONOMY 69 (1) 2023 Fotova Čiković, K., Cvetkoska, V., Keček, D.
of their economic development and geopolitical position. The 
COVID-19 pandemic greatly impacted economies and societies 
globally because countries needed to face the new situation 
with numerous challenges for private, public, and business 
entities "while minimizing the negative social and economic 
implications" (Fotova Čiković, 2021). More importantly, it 
impacted every industry and every business sector globally. 
This empirical study supports this fact. However, if we want to 
confirm the impact of the COVID-19 pandemic on the Western 
Balkans' banking systems’ efficiency, additional analyses, 
including those for 2021 and 2022, need to be conducted.
According to the findings, small banks are the least efficient, 
and large banks are the most efficient in the Western Balkan 
countries, as previously reported by Guistininani and Ross 
(2008), Micajkova and Poposka (2013), Fotova Čiković and 
Cvetkoska (2017), and Cvetkoska et al. (2021).
Except for North Macedonia and Kosovo, which saw a slight 
increase in efficiency (from 76.42% in 2019 to 77.70% in 2020) 
and a significant increase in efficiency (from 83.95% in 2019 to 
90.08% in 2020), the other Western Balkan countries recorded 
a decrease in efficiency in 2019 and 2020. The obtained results 
also show that the highest efficiency scores mainly were in 
2018 (for North Macedonia, Serbia, Montenegro, and Kosovo; 
2019 for Bosnia and Herzegovina, and 2017 for Albania), i.e. 
before the COVID-19 pandemic, whereas the lowest efficiency 
was noted in the first studied year: 2016 (and 2019 for North 
Macedonia and Kosovo). 
This is the first study that addresses and empirically measures 
the efficiency of the banking systems of Western Balkan 
countries, including the first pandemic year (2020), and thus 
the first study that addresses the COVID-19 pandemic impact 
on banks' efficiency and performance. For these reasons, this 
research represents a gain for the scientific community but 
also provides invaluable insights for banking management, 
regulatory bodies, governments, potential investors, the 
interested public, and other bank stakeholders.
Conclusion
This paper aimed to assess and compare the relative 
efficiency of commercial banks in each of the six developing 
countries in the Western Balkans - North Macedonia, Serbia, 
Kosovo, Montenegro, Albania, and Bosnia and Herzegovina 
- in the observed period from 2016 to 2020 and to explore 
the impact of the COVID-19 pandemic on the efficiency and 
performance of the banking sectors of the six Western Balkan 
developing countries in the year 2020. The DEA Window 
analysis technique has been employed with two inputs and 
two outputs: interest and non-interest expenses as inputs and 
interest and non-interest revenues as outputs.
According to the DEA Window O-V model results, the average 
efficiency of the six Western Balkan banking systems 
significantly differs. The highest average efficiency of the 
whole banking system in the whole observed period (2016–
2020) has been noted in Kosovo, whereas B&H noted the 
lowest average efficiency, followed by the Serbian banking 
system.
There are no studies on the efficiency of the banking sectors 
of the six studied Western Balkan countries and no studies 
on the impact of COVID-19 pandemics on the efficiency and 
performance of banking systems in these countries.
This study, however, is not without limitations. Even though it 
is among the first studies to evaluate and assess the impact 
of the COVID-19 pandemic on the efficiency and performance 
of six Western Balkan economies, it only analyzes the first year 
of the pandemic (i.e., 2020). Furthermore, while 2020 was the 
most difficult pandemic year to date (due to the surprise effect 
and the countries' lack of preparedness for a crisis of this 
magnitude), more research should be conducted to gain new 
insights into the total impact and consequences COVID-19 
left on these economies. Moreover, the DEA methodology 
itself has a few limitations. As Stolp (1990) claimed, "the 
enthusiasm for DEA has not been universal, and many 
detractors have criticized it on several methodological and 
substantive grounds." According to Jemrić and Vujčić (2002), 
"the main disadvantage of DEA is that the frontier is sensitive 
to extreme observations and measurement errors, since the 
basic assumption is that random errors do not exist and that 
all deviations from the frontier indicate inefficiency." Another 
caveat of DEA is that the classification of the relative efficient 
DMUs means those DMUs are "only efficient compared to 
the others in the sample" (Repkova, 2014). DEA, on the other 
hand, provides a "complete unbiased numerical score, ranking, 
and efficiency potential improvement targets for each of the 
inefficient units" (Popović et al., 2017) and should be viewed 
as "more conventional statistical methods, as an informative 
and useful tool for systematic sensitivity analysis" (Stolp, 
1990) and "a significant diagnostic tool" that provides new 
insights to policymakers, investors, and other stakeholders" 
(Fotova Čiković & Lozić, 2022).
In future research, the ability of the banking management to 
cope with the crisis in the analyzed Western Balkan countries 
could be investigated.

36
NAŠE GOSPODARSTVO / OUR ECONOMY 69 (1) 2023 Fotova Čiković, K., Cvetkoska, V., Keček, D.
Appendix 1
Table 5
Average efficiency by term for the Serbian commercial banks (2016-2020)
BANK
YEAR
Average score
2016 2017 2018 2019 2020
Addiko Bank AD Beograd 0.5550 0.5995 0.6255 0.6584 0.6500 0.6177
Agroindustrijsko Komercijalna Banka AD Skopje 1.0000 1.0000 1.0000 0.9729 1.0000 0.9946
Alta Banka AD Beograd 0.6297 0.6749 0.9120 0.9735 0.7337 0.7848
API Bank AD Beograd 0.2842 0.4228 0.4800 0.3374 0.3005 0.3650
Banca Intesa AD Beograd 0.9775 1.0000 1.0000 1.0000 1.0000 0.9955
Banka Postanska Stedionica AD Beograd 0.4854 0.7446 0.6943 0.6625 0.6878 0.6549
Bank of China Srbija AD Beograd n.a. 1.0000 1.0000 0.6869 0.3534 0.6081
Credit Agricole Banka Srbija AD Novi Sad 0.6271 0.6213 0.6823 0.7105 0.7193 0.6721
Direktna Banka AD Kragujevac 0.3902 0.9663 1.0000 0.6827 0.6938 0.7466
Expobank AD Beograd 0.3666 1.0000 0.7938 0.5744 0.5371 0.6544
Erste Bank AD Novi Sad 0.7493 0.7918 0.7776 0.7324 0.7039 0.7510
Eurobank AD Beograd 0.8072 0.7694 0.8419 0.7962 0.7557 0.7941
Halkbank AD Beograd 0.6459 0.6616 0.7113 0.7246 0.7016 0.6890
Komercijalna Banka AD Beograd 0.7674 0.9324 0.9930 0.9704 0.8223 0.8971
Mirabank AD Beograd - Novi Beograd 0.2245 0.2545 0.4019 0.4564 0.9996 0.4674
Mobi Banka AD Beograd 0.2121 0.2777 0.3497 0.3677 0.3957 0.3206
NLB Banka AD Beograd 0.7068 0.6903 0.7845 0.6394 0.6618 0.6966
Opportunity Banka AD Novi Sad 0.9142 0.9347 0.9553 0.9709 0.8920 0.9334
OTP Banka Srbija AD Novi Sad 0.5957 0.5348 0.6202 0.9901 0.6204 0.6722
ProCredit Bank AD Beograd 0.8143 0.7533 0.6846 0.6290 0.6192 0.7001
Raiffeisen Banka AD Beograd 0.9311 1.0000 1.0000 1.0000 0.9114 0.9685
Sperbank Srbija AD Beograd 0.6259 0.6472 0.6662 0.6212 0.5575 0.6236
Srpska Banka AD Beograd 0.6398 1.0000 0.7560 0.8204 1.0000 0.8432
Unicredit Bank Srbija AD Beograd 0.9669 0.8810 1.0000 1.0000 0.8359 0.9368
Source: Results from the window DEA model

37
NAŠE GOSPODARSTVO / OUR ECONOMY 69 (1) 2023 Fotova Čiković, K., Cvetkoska, V., Keček, D.
Table 6
Average efficiency by term for the North Macedonian commercial banks (2016-2020)
BANK
YEAR
Average score
2016 2017 2018 2019 2020
NLB (Tutunska) Banka AD Skopje 0.8610 0.9849 0.9287 1.0000 0.8890 0.9327
Stopanska Banka AD Skopje 1.0000 1.0000 1.0000 0.9949 1.0000 0.9990
Komercijalna Banka AD Skopje 0.9350 1.0000 0.9907 0.9708 1.0000 0.9793
Halk Banka AD Skopje 0.6975 0.7024 0.7348 0.7154 0.6489 0.6998
Uni Banka AD Skopje 0.6832 0.8371 0.7327 0.6510 0.7431 0.7294
ProKredit Banka AD Skopje 0.6841 0.7169 0.6577 0.6555 0.6765 0.6781
TTK Banka AD Skopje 1.0000 0.9899 0.9571 0.9321 0.8528 0.9464
Kapital Banka AD Skopje 0.8367 0.4585 1.0000 0.9187 1.0000 0.8428
Stopanska Banka AD Bitola 0.6640 0.6443 0.6390 0.5839 0.7287 0.6520
Sparkasse Banka AD Skopje 0.8878 0.9042 0.7857 0.7427 0.6531 0.7947
Ohridska Banka AD Ohrid 0.6824 0.6791 0.7030 0.5506 0.5427 0.6316
Centralna Kooperativna Banka AD Skopje 0.6774 0.6284 0.6520 0.6530 0.6762 0.6574
Silk Road Banka AD Skopje 0.8439 1.0000 0.8260 0.5669 0.6906 0.7855
Source: Results from the window DEA model
Table 7
Average efficiency by term for the Montenegrin commercial banks (2016-2020)
BANK
YEAR
Average score
2016 2017 2018 2019 2020
Crnogorska komercijalna banka AD Podgorica  
member of OTP Group
1.0000 0.9404 1.0000 1.0000 1.0000 0.9881
Hipotekarna banka AD Podgorica 1.0000 0.9746 0.9940 1.0000 0.9299 0.9797
Prva banka Crne Gore AD Podgorica 0.9477 0.8890 0.7457 0.7796 0.7753 0.8275
ERSTE Bank AD Podgorica 1.0000 0.9445 0.9460 1.0000 1.0000 0.9781
NLB Banka AD Podgorica 0.9459 0.9133 0.9260 1.0000 0.8568 0.9284
Komercijalna banka AD Podgorica 0.2926 0.9198 0.9118 0.8556 0.7422 0.7444
Addiko Bank AD Podgorica 0.4359 0.9169 0.8867 0.8040 0.9374 0.7962
Universal Capital Bank AD Podgorica 0.8843 0.8459 1.0000 1.0000 0.9086 0.9278
Lovćen banka AD Podgorica 0.9135 1.0000 1.0000 0.9700 0.9642 0.9695
Zapad banka AD Podgorica 1.0000 0.8874 0.9776 0.6382 0.6483 0.8303
ZIRAAT Bank Montenegro AD Podgorica 0.7505 1.0000 0.9915 1.0000 1.0000 0.9484
Adriatic Bank AD Podgorica 1.0000 0.7726 0.7212 0.9030 1.0000 0.8794
Source: Results from the window DEA model

38
NAŠE GOSPODARSTVO / OUR ECONOMY 69 (1) 2023 Fotova Čiković, K., Cvetkoska, V., Keček, D.
Table 8
Average efficiency by term for the Albanian commercial banks (2016-2020)
BANK
YEAR
Average score
2016 2017 2018 2019 2020
Alpha Bank - Albania S.A. 0.5178 0.6690 0.5989 0.6008 0.6504 0.6074
American Bank of Investment S.A. 0.5610 0.7182 0.6720 0.6534 0.8899 0.6989
Credins Bank S.A. 0.6641 0.7058 1.0000 0.8210 1.0000 0.8382
First Investment Bank, Albania S.A. 0.6777 1.0000 0.9804 0.6621 0.6902 0.8021
Intesa Sanpaolo Bank Albania S.A. 0.9099 0.9139 0.7643 0.9980 0.8831 0.8938
National Commercial Bank S.A.- BKT 0.8191 1.0000 1.0000 0.9417 1.0000 0.9522
OTP Bank Albania S.A 0.7440 0.8128 0.9444 1.0000 0.9494 0.8901
Procredit Bank S.A. 0.6706 0.5352 0.4886 0.3937 0.3729 0.4922
Raiffeisen Bank S.A. 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000
Tirana Bank S.A. 0.5967 0.7826 0.7140 0.9397 0.8356 0.7737
Union Bank S.A. 0.8334 0.8962 0.9067 0.7615 0.6487 0.8093
United Bank of Albania S.A. 1.0000 1.0000 0.8984 1.0000 0.3838 0.8564
Source: Results from the window DEA model
Table 9
Average efficiency by term for the Kosovar commercial banks (2016-2020)
BANK
YEAR
Average score
2016 2017 2018 2019 2020
NLB Bank 0.9310 0.9288 0.9286 1.0000 1.0000 0.9577
Banka për Biznes 0.6683 0.8561 0.9245 0.7759 0.6983 0.7846
Banka Ekonomike 0.8349 0.8339 0.8139 0.7831 0.6830 0.7898
Raiffeisen Bank Kosovo 0.9591 1.0000 0.9932 1.0000 1.0000 0.9905
ProCredit Bank 1.0000 1.0000 1.0000 0.9770 0.9996 0.9953
TEB J.S.C 1.0000 0.9695 0.9950 0.9216 0.9892 0.9750
Banka Kombëtare Tregtare 1.0000 0.8358 0.8526 0.5873 0.8365 0.8224
Turkiye Is Bankasi 1.0000 0.7572 1.0000 0.6716 1.0000 0.8858
Source: Results from the window DEA model

39
NAŠE GOSPODARSTVO / OUR ECONOMY 69 (1) 2023 Fotova Čiković, K., Cvetkoska, V., Keček, D.
Table 10
Average efficiency by term for the B&H commercial banks (2016-2020)
BANK
YEAR
Average score
2016 2017 2018 2019 2020
Addiko Bank Banja Luka 0.5856 0.6127 0.6145 0.6759 0.6267 0.6231
Addiko Bank FBIH 0.4099 0.6406 0.7124 1.0000 1.0000 0.7526
ASA Banka 0.4379 0.4479 0.5936 0.7205 0.6759 0.5752
Bosna Bank International 0.5365 0.6994 0.7437 0.7968 0.6757 0.6904
Intesa Sanpaolo Banka 0.4957 0.7086 0.8778 0.9239 0.8240 0.7660
Komercijalna Banka Banja Luka 0.4423 0.5882 0.6807 0.7489 0.7508 0.6422
Komercijalno investiciona banka 0.9921 1.0000 1.0000 1.0000 0.8325 0.9649
MF banka 0.3459 0.6269 0.7120 0.8798 0.7484 0.6626
Naša banka 0.3936 0.5964 0.5193 0.6306 0.8004 0.5881
NLB Banka AD Banja Luka 0.8367 0.9836 0.8922 1.0000 1.0000 0.9425
NLB BANKA d.d. Sarajevo 0.5778 0.7151 0.7666 0.7872 0.7483 0.7190
Nova banka 0.7302 0.7068 0.6838 0.7161 0.6842 0.7042
PrivrednaBanka Sarajevo 0.8132 0.5910 0.6902 0.6988 0.7374 0.7061
ProCredit Bank 0.4199 0.4245 0.5497 0.5753 0.6175 0.5174
Reiffeisen Bank 0.9035 0.9740 1.0000 1.0000 0.9578 0.9671
Sberbank AD Banka Luka 0.4778 0.5567 0.7110 0.8679 0.9705 0.7168
Sberbank Sarajevo 1.0000 0.5842 0.6273 0.6929 0.6157 0.7040
Sparkasse Bank Sarajevo 0.6287 0.7448 0.7749 0.8139 0.8107 0.7546
UniCredit Bank Mostar 0.8006 0.9165 1.0000 1.0000 0.9703 0.9375
UniCredit Bank Banja Luka 0.4919 0.7663 0.8442 1.0000 0.8227 0.7850
Union Banka 0.4023 1.0000 0.5788 0.6590 0.7187 0.6718
Vakufska banka 0.5504 0.5610 0.6253 0.6041 0.7568 0.6195
ZiraatBank BH 0.3707 0.5144 0.4993 0.6027 0.5766 0.5127
Source: Results from the window DEA model
References
Abu-Alkheil, A.M., Burghof, H.- P., Khan, W.A. (2012). Comparison of efficiency and productivity changes of Islamic and 
conventional banks: Evidence from Europe and Muslim-Majority countries? Journal of Applied Business Research, 28(6), 
1385-1412. DOI: https://doi.org/10.19030/jabr.v28i6.7351
Anayiotos, G., Toroyan, H., Vamvakidis, A. (2010). The efficiency of emerging Europe’s banking sector before and after the 
recent economic crisis. Financial Theory and Practice 34, 247–267. 
Bank of Albania (2021) Financial Stability Report 2020. Retrieved from: https://www.bankofalbania.org/Financial_Stability/
Financial_Stability_Reports/Financial_Stability_Reports/ 
Bošković, A., Krstić, A. (2020). The Combined Use of Balanced Scorecard and Data Envelopment Analysis in the Banking 
Industry. Business Systems Research, 11(1), 4-15. DOI: https://doi.org/10.2478/bsrj-2020-0001
Braimllari, A., Benga, A. (2019). Cost efficiency of banks in Albania: A Data Envelopment Analysis for the period 2015-2017. 
Proceedings from the International Conference on Applied Analysis and Mathematical Modelling ICAAMM19, March 10-13, 
2019, Istanbul-Turkey

40
NAŠE GOSPODARSTVO / OUR ECONOMY 69 (1) 2023 Fotova Čiković, K., Cvetkoska, V., Keček, D.
Central Bank of Bosnia and Herzegovina (2021). Financial Stability Report 2020. Retrieved from: https://www.cbbh.ba/Content/
Archive/575?lang=en 
Central Bank of Montenegro (2021). Financial system of Montenegro. Retrieved from https://www.cbcg.me/en/core-functions/
financial-stability/financial-system-of-montenegro 
Central Bank of the Republic of Kosovo (2021). Financial Stability Report 2020, Number 17. Retrieved from https://bqk-kos.org/
wp-content/uploads/2021/06/CBK_FSR_17.pdf 
Cooper, W. W., Seiford, L. M., Tone, K. (2007). Data envelopment analysis: A comprehensive text with models, applications, 
references and DEA-solver software. New York, NY: Springer Science + Business Media, LLC. DOI: https://doi.
org/10.1007/978-0-387-45283-8
Cvetkoska, V., Barišić, P. (2017). The Efficiency of The Tourism Industry in The Balkans. Proceedings of the Faculty of Economics 
in East Sarajevo, 14, 31-41. DOI: https://doi.org/10.7251/ZREFIS1714031C
Cvetkoska, V. (2010). A Two-Stage Approach for Measuring the Efficiency Applying the Methods DEA and AHP. Master’s Thesis, The 
University of Belgrade, Belgrade, Serbia.
Cvetkoska, V., Fotova Čiković, K. (2020). Assessing the relative efficiency of commercial banks in the Republic of North 
Macedonia: DEA window analysis. Croatian operational research review, 11(2), 217-227. DOI: https://doi.org/10.17535/
crorr.2020.0017
Cvetkoska, V., Fotova Čiković, K., Tasheva, M. (2021). Efficiency of Commercial Banking in Developing Countries. Mathematics, 
9(14), 1-21. DOI: https://doi.org/10.3390/math9141597
Cvetkoska, V., Savić, G. (2021). DEA in banking: Analysis and visualization of bibliometric data. Data Envel. Anal. J., 5(2), 
455–485. DOI: https://doi.org/10.1561/103.00000044
Durguti, E.A., Krasniqi, E.H., Krasniqi, D. (2020). Assessing the performance of factors affecting the profitability of the banking 
system: Evidence from Kosovo. European Journal of Sustainable Development, 9(2), 304-314. DOI: https://doi.org/10.14207/
ejsd.2020.v9n2p304
Durguti, E.A., Kryeziu, N. (2021). Importance of Corporate Governance: Evidence from Kosovo’s Banking Sector. Croatian 
Economic Survey, 23(2), 5-32. DOI: https://doi.org/10.15179/ces.23.2.1
ECB. European Central Bank. (2019). Occasional Paper Series. Financial stability assessment for EU candidate countries and 
potential candidates. Developments since 2016. https://www.ecb.europa.eu/pub/pdf/scpops/ecb.op233~ae2d552448.
en.pdf 
Efendić, V., Avdić, A. (2011). An analysis on the efficiency of banks in Bosnia and Herzegovina using DEA method. Technics 
Technologies Education Management, 6(1), 147-158. 
Emrouznejad, A., Yang, G. (2018). A survey and analysis of the first 40 years of scholarly literature in DEA: 1978–2016. Socio 
Econ. Plan. Sci., 61, 4–8. DOI: https://doi.org/10.1016/j.seps.2017.01.008 
FIPA (2022). Banking and Financial Services. Retrieved from http://www.fipa.gov.ba/atraktivni_sektori/bankarski_sektor/
default.aspx?id=46&langTag=en-US.
Fotova Čiković, K., Cvetkoska, V. (2017). Efficiency of the Macedonian banking sector: A non-parametric approach. CEA Journal 
of Economics, 12(2), 17-26.
Fotova Čikovič, K., Lozić, J. (2022). Application of Data Envelopment Analysis (DEA) in Information and Communication 
Technologies. Tehnički glasnik 16(1), 129-134. DOI: https://doi.org/10.31803/tg-20210906103816 
Giustianiani, A., Ross, K. (2008). Bank Competition and Efficiency in the FYR Macedonia. South-Eastern Journal of Economics, 
2, 145-167.
Husejinović, A. (2019). Efficiency of commercial banks operating in Federation of Bosnia and Herzegovina using DEA method. 
Sustainable Engineering and Innovation, 1(2), 106-111. DOI: https://doi.org/10.37868/sei.v1i2.83 
Jemrić, I., Vujčić, B. (2002). Efficiency of Banks in Croatia: A DEA Approach. Comparative Economic Studies, 44(2-3), 169-193, 
DOI: https://doi.org/10.1057/ces.2002.13
Lukić, R., Sokić, M., Kljenak, D.V. (2017). Efficiency analysis of banking sector in Republic of Serbia. Business Excellence and 
Management; Faculty of Management, 7(4), 5–17.
Marković, M., Knežević, S., Brown, A., Dmitrović, V. (2015). Measuring the productivity of Serbian banks using Malmquist index., 
Manag. J. Sustain. Bus. Manag. Solut. Emerg. Econ., 20, 1–10. DOI: https://doi.org/10.7595/management.fon.2015.0022 
Memić, D., Škaljić-Memić, S. (2013). Performance Analysis and Benchmarking of Commercial Banks Operating in Bosnia and 
Herzegovina: a DEA Approach. Business Systems Research, 4(2), 17–37. DOI: https://doi.org/10.2478/bsrj-2013-0010 
Micajkova, V., Poposka, K., (2013). Efficiency of Macedonian Banks: A DEA Approach. Research Journal of Finance and Accounting, 
4(12), 141-149.

41
NAŠE GOSPODARSTVO / OUR ECONOMY 69 (1) 2023 Fotova Čiković, K., Cvetkoska, V., Keček, D.
Mihailović, N., Bulajić, M., Savić, G. (2009). Ranking of banks in Serbia. Yugoslav Journal of Operations Research 19(2), 323-334. 
DOI: https://doi.org/10.2298/YJOR0902323M
Milenković, N., Radovanov, B., Kalaš, B., Horvat, A.M. (2022). External Two Stage DEA Analysis of Bank Efficiency in West 
Balkan Countries. Sustainability, 14, 978. DOI: https://doi.org/10.3390/su14020978
National Bank of Serbia (2021). Annual Financial Stability Report 2020, Belgrade. Retrieved from https://www.nbs.rs/export/
sites/NBS_site/documents-eng/publikacije/fs/fsr_2020.pdf.
Naumovska, E., Cvetkoska, V. (2014). Measuring the Efficiency of the Banking Sector in the Republic of Macedonia. Annual of 
the Faculty of Economics – Skopje, 49, 221-238.
Naumovska, E., Cvetkoska, V., (2016). Efficiency of the Macedonian banking sector. Yugoslav Journal of Operations Research, 
26(3), 317-329. DOI: https://doi.org/10.2298/YJOR150228019N
NBRM (2021). Financial Stability Report 2020. 
Paradi, J.C., Sherman, D. H., Tam, F.K. (2018). Data Envelopment Analysis in the Financial Services Industry. A guide for practitioners 
and analysts working in Operations Research using DEA, 266. Springer. DOI: https://doi.org/10.1007/978-3-319-69725-3
Popović, Ž., Stanković, J., Marjanović, I. (2017). Analysis of bank efficiency in the Republic of Serbia: DEA approach. In Proceedings 
of the 48
th
 International Scientific Conference: Contemporary Approaches in the Analysis of Economic Performances, Niš, Serbia, 
11-12 October 2017, 233–242.
Rabar, D. (2013). Assessment of regional efficiency in Croatia using data envelopment analysis. Croatian Operational Research 
Review, 4(1), 76-88. https://hrcak.srce.hr/97382 
Radojicic, M., Savić, G., Jeremic, V., (2018). Measuring the Efficiency of Banks: The Bootstrapped I-Distance GAR DEA Approach. 
Technological and Economic Development of Economy, 24(4), 1581-1605. DOI: https://doi.org/10.3846/tede.2018.3699
Radojicic, M., Savic, G., Radovanovic, S., Jeremic, V. (2015). A Novel Bootstrap Dba-DEA Approach in Evaluating Efficiency of 
Banks. Mircea cel Batran Naval Academy Scientific Bulletin, 18(2), 375-384. DOI: https://doi.org/10.3846/tede.2018.3699 
Radzic, J., Yuce, A. (2008). Banking Development in The Former Yugoslavian Republics. International Business & Economics 
Research Journal, 7(9), 35-48, DOI: https://doi.org/10.19030/iber.v7i9.3291
Repkova, I. (2014). Efficiency of the Czech banking sector employing the DEA window analysis approach. Procedia Economics 
and Finance, 12, 587-596. DOI: https://doi.org/10.1016/S2212-5671(14)00383-9
Sahiti, A., Sahiti, A. (2021). The Commercial Banks’ Credit Risk Efficiency: Empirical Evidence from Kosovo. Journal of Eastern 
European And Central Asian Research, 8(2), 240-254. DOI: http://dx.doi.org/10.15549/jeecar.v8i2.635 
Savic, G., Radosavljevic, M., Ilijevski, D. (2012). DEA Window Analysis Approach for Measuring Efficiency of Serbian Banks Based 
on Panel Data. Management J. Theory Pract. Manag, 17, 5-14. DOI: https://doi.org/10.7595/management.fon.2012.0028
Shkodra, J., Ismajli, H. (2017). Determinants of the credit risk in developing countries: a case of Kosovo banking sector. Banks 
and Bank Systems, 12(4), 90-97. DOI: https://doi.org/10.21511/bbs.12(4).2017.08
Spaho, A. (2015). Determinants of Banks’ Efficiency in Albania: Data Envelopment Analysis and Tobit Regression Analysis. In: 
Proceedings of 6
th
 International Conference on Social Sciences, 11-12 September 2015, Istanbul, Turkey, 364-374. 
Spaho, A., Mitre, T., Shehu, V. (2015). Technical Efficiency and Super-Efficiency of Commercial Banks in Albania. In Proceeding 
of 5
th
 International Conference on Humanities and Social Sciences, 25-26 September, 2015, Rome, Italy, 686-692.
Stolp, C. (1990). Strengths and weaknesses of data envelopment analysis: An urban and regional perspective. Computers, 
Environment and Urban Systems, 14(2), 103-116. DOI: https://doi.org/10.1016/0198-9715(90)90016-M.
Sun, D.B. (1988). Evaluation of Managerial Performance in Large Commercial Banks by Data Envelopment Analysis, Ph.D. Thesis, 
Austin, Texas: The University of Texas, Graduate School of Business.
Toçi, V. Z. (2009). Efficiency of Banks in South-East Europe: with Special Reference to Kosovo. CBK Working paper, No. 4. 
Toci, V. Z., Hashi, I. (2013). Intermediation Efficiency of Banks in South-East Europe: An Empirical Assessment Using DEA and 
Malmquist Index. International Journal of Finance & Banking Studies IJFBS, 2(3), 1-20. DOI: https://doi.org/10.20525/ijfbs.
v2i3.151 
Varesi, L. (2015). Measuring Banking Efficiency during Crisis Period Using Data Envelopment Analysis: Western Balkan 
Countries Case. Academic Journal of Interdisciplinary Studies 4(1), 261-274. DOI: https://doi.org/10.5901/mjss.2015.
v4n1p261
Vojnić, D. (1995). European Integrational Processes and The Countries in Transition - with Special References to Croatia and 
Former Yugoslavia. Croatian Economic Survey, 2, 203-239.
Zogjani, J., Kelmendi, M. (2015). Banking Efficiency in Kosovo during the Global Financial Crisis. Academic Journal of 
Interdisciplinary Studies, 4(2), 367. DOI: https://doi.org/10.5901/ajis.2015.v4n2p367
Zogjani, J., Mazelliu, M., Humolli, B. (2018). The efficiency and impact of banking electronic services: evidence for Kosovo. Int. 
J. Financial Innovation in Banking, 2(1), 82-97. DOI: https://doi.org/10.1504/IJFIB.2018.092421

42
NAŠE GOSPODARSTVO / OUR ECONOMY 69 (1) 2023 Fotova Čiković, K., Cvetkoska, V., Keček, D.
Vpliv COVID-19 na učinkovitost bank na Zahodnem Balkanu: 
pristop z metodo analize ovojnice podatkov
Izvleček
Namen tega prispevka je oceniti in primerjati relativno učinkovitost poslovnih bank v šestih državah v razvoju na 
Zahodnem Balkanu - Severni Makedoniji, Srbiji, Črni gori, Bosni in Hercegovini (B&H), Kosovu in Albaniji - z uporabo vodilne 
neparametrične metodologije Data Envelopment Analysis (DEA) za obdobje 2016-2020 ter preučiti vpliv pandemije COVID-19 
na uspešnost njihovih bančnih sistemov. Implementiran je bil izhodno usmerjen model DEA z obrestnimi in neobrestnimi 
odhodki kot vhodnimi podatki ter obrestnimi in neobrestnimi prihodki kot izhodnimi podatki. Naše ugotovitve kažejo, da 
se povprečna učinkovitost šestih bančnih sistemov Zahodnega Balkana razlikuje, pri čemer je bila v celotnem opazovanem 
obdobju (2016-2020) ugotovljena najvišja povprečna učinkovitost bančnega sistema Kosova, medtem ko je bila najnižja 
povprečna učinkovitost ugotovljena pri bančnem sistemu Bosne in Hercegovine. Pandemija COVID-19 je zmanjšala relativno 
učinkovitost bančnih sektorjev v vseh šestih opazovanih državah Zahodnega Balkana, razen na Kosovu. Vendar je za oceno 
vpliva pandemije COVID-19 na bančne sisteme na Zahodnem Balkanu priporočljiva dodatna raziskava, ki bi vključevala vsa 
leta pandemije. Ta študija zagotavlja neprecenljiva spoznanja za akademike, bančne upravne in regulativne organe, vlade 
ter zainteresirano javnost. Gre za prvo empirično študijo, ki vključuje vpogled v prvi vpliv pandemije COVID-19 na bančne 
sisteme v državah v razvoju na Zahodnem Balkanu.
Ključne besede: bančni sektor, analiza ovojnice podatkov, COVID-19, Zahodni Balkan