Efficiency and Productivity of Commercial Banks: Evidence from Bangladesh

Introduction

Since the early 1970s, the banking sector in Bangladesh is one of the most important instruments of countries financial system. Since commercial banks is an integral part of the financial system in Bangladesh, an efficient and productive banking system strengthen the economy of the country. Therefore, all financial institutions, including commercial banks, are required to accomplish economic objectives set by the government of a country. For long-term investments and the base of economic escalation, the banking segment is the key source of funds (Schumpeter, 1934). The financial performance of banks is often stated in terms of efficiency. In any nation, the effectiveness of the keeping money segment ensures sheltered, compelling and productive monetary framework. Levine (1998) recognized that the proficiency of money related establishments impacts a nation's monetary extension and in the meantime, bank bankruptcies could result in fundamental emergencies significant in cynical ramifications on the economy.

Wheelock and Wilson, (1999), revealed that the calculated efficiency is treated as dissimilarity between observed input and output levels as well as the corresponding optimal values. According to (Yilmaz, 2013), in the financial market, the effectiveness of the banking structure is the most significant issues because it directly affects the constancy of the banking sectors and effectiveness of the country’s economic policy. For that reason, efficiency scores of banks are pointers of the success of banking industry as an entire and these indicators also assist to verify the probable impact of government rules and regulations on efficiency (Wheelock and Wilson, 1999).

As indicated by (Shameem, 1995), since the creation of the National Commission on Money, Banking and Credit in 1986, the viability of the dealing with a record territory has changed into a fundamental issue in Bangladesh. The essential point of view in the establishment of the commission among others is to discover answers for capable organization and assignments of the keeping cash system (Shameem, 1995). Additionally, in 1991, the World Bank likewise guided Bangladesh Bank (BB) the Central Bank of Bangladesh to fortify the country's saving money division rules direction and control. Hence, the productivity of the managing an account segment is basic to keep the dauntlessness of the saving money framework.

However, the majority of the conventional evaluators used ratios analysis for measuring banks efficiency. According to (Rao and Tekeste, 2012), ratios analysis can be confusing and misleading as it computes incomplete efficiency of banks. To prevail over this dilemma in the past few decades, researchers are trying to use frontier analysis methods. Sufian, F. what's more, (Kamarudin, F. 2013), attempted to look at the level of benefit proficiency in the Bangladesh managing an account division over the periods 2004– 2011 utilizing non-parametric Data Envelopment Analysis (DEA) strategy. they inspected the Bangladesh managing an account area has uncovered the least and most elevated amount of benefit productivity amid the years 2009 and 2004 separately and they additionally uncovered that there are four banks are delegated benefit wasteful, while another eight banks which have been benefit proficient. Moreover, that benefit productivity proposes valuable data on the effectiveness of bank administration (Berger and Mester, 2003) and (Maudos and Pastor 2003).

Thus, the objective of this study is to investigate the relative technical efficiency and productivity change of the commercial banks in Bangladesh over the years 2013–2017. To do as such, we use the non-parametric Data Envelopment Analysis (DEA) technique. From that point onward, we bless with a brief survey of related examinations, trailed by blueprints the information and system. The accompanying piece reports the observational outcomes and the last segment finishes up and gives strategy suggestions.

Related Literature

According to Fare, et al. (1994), the essential idea of efficiency is that it gauges how well firms change their input into profit in line with their detectable intentions. A firm is said to be productive on the off chance that it can accomplish its objectives and unproductive in the event that it fails. In ordinary conditions, a company's objective is thought to be cost belittlement of production. According to the production theory, it is usually accepted that organizations are behaving proficiently in a monetary sense. The effectiveness of a bank can be assessed using non-parametric, parametric and financial ratios approach.

 (Bauer, et al. 1998) examined that, financial ratios similar to return on assets or return on equity from firm’s income statements and balance sheets are habitually used by financial institutions like banks to determine the efficiency of a firm. On the other hand, frontier efficiency measures deviations in execution from that of firms with the best execution on the proficient frontier, holding consistent various exogenous market factors like costs in local markets. Frontier efficiency is a more predominant estimation than financial ratios since it utilizes a statistical strategy to take off the impact of contrasts in costs and different exogenous market factors influencing the standard accomplishment ratio so as to get better gauges (Bauer, et al. 1998). In a study on the Chinese banking sectors, (Chen, et al. 2017) finds that the overall efficiency level of the Chinese banks remains low, and this is notably determined by  two important variables including the ownership structure and cost structure of the Chinese banks.

(Muvingi, J and Hoteral, S, 2015; Bader, et al. 2008)  pointed out that most of the examinations on bank proficiency are focusing on the developed nations financial aspects and there are a reasonable number of studies which have investigated the effectiveness of the saving money ventures of growing nations. Nevertheless, earlier learning have mostly determined on the scale, technical and PTE (pure technical efficiency) concept (Yudistira, 2004; Tahir and Haron, 2008; Isik and Hassan, 2002). Alternatively, studies which explore the cost, profit, and revenue efficiency are relatively insufficient (Maudos, et al. 2002; Berger and Mester, 1999; Ariff and Can, 2008; Adongo, et al. 2005).

Darrat, et al. (2002) examined the efficiency of banks in Kuwait, by using the Data Envelopment Analysis (DEA). The results indicate that about 47% of the assets of the banks were not used appropriately to produce profit. The small size banks were found to be more proficient than the larger ones. Muvingi, J. & Hoteral, S, (2015) investigated the efficiency and productivity analysis of commercials banks in Zimbabwe through DEA approach, the average gains of technical efficiency under the VRS and CRS reports for the commercial banking segment as an entire are 81.5% and  70.95% jointly. Thus, the average scale efficiency for commercial banks working in Zimbabwe was an initiate to be 73.7%..(B. Joseph, et al, 2015) shows that the foreign sector banks have gotten an amazing mean Total Factor Productivity (TFP) of 1.06 and nine foreign sector banks has acquired a qualified score of one which portrays that foreign sector banks are doing admirably in India. (Garamu, G. 2016) examined the technical efficiency and productivity of Ethiopian commercial banks under Data Envelopment Analysis (DEA) method. The results show that, on average, Ethiopian commercial banks were comparatively technically inefficient while scale inefficiency takes the most important contribution for the basis of inefficiency. The study also found that the average Total Factor Productivity (TFP) transforms is 0.965 throughout the study period.

The exceeding literature obviously unveils the accompanying research holes. First, the larger parts of these examinations have concentrated on the continuing Banking regions of the developed and western countries. Furthermore, considers examining the proficiency of the banking part have for the most part centered around the scale, technical as well as PTE, whereas test realities on the productivity and TE (specialized productivity) are similarly deficient under DEA) approach and MPI. In line with these information gaps, the present study seeks to make available new factual proof on the productivity and technical efficiency in the Bangladesh banking segment.

Data and Methodology

Method of Data Analysis

This study employed secondary data that were collected from the annual reports of listed private commercial banks in Dhaka Stock Exchange (DSE) of Bangladesh for the period under investigation from 2012 to 2017). To determine banks’ efficiencies Banker, Charnes and Cooper (BCC) and Charnes, Cooper and Rhodes (CCR) input-oriented DEA models are used. Further, the study uses the DEA-based Output-orientated MPI to evaluate both technical efficiency and productivity growth of the bank. According to (Sharma et al. 2013), there are two methods existing in assessing productivity—the Stochastic Frontier Approach (SFA) and DEA, but DEA has obtained the most important in recent years. For the aim of the study, we comprise 19 private commercials banks of which information are accessible from 2012 to 2017. The whole rundown of banks incorporated into the investigation is given in Table 1.

Table 1: Sample of the Study

Bank                                                                                                                         Abbr.

AB Bank Ltd.

ABBL

Al-Arafah Islami Bank Ltd.

AIBL

Bank Asia Ltd.

BAL

BRAC Bank Ltd.

BBL

Dutch-Bangla Bank Ltd.

DBBL

Dhaka Bank Ltd.

DBL

Eastern Bank Ltd.

EBL

Export-Import Bank of Bangladesh Ltd.

EXIMBL

Islami Bank Bangladesh Ltd.

IBBL

Mercantile Bank Ltd.

MBL

National Bank Ltd.

NBL

One Bank Ltd.

OBL

Shahjalal Islami Bank Ltd.

SJIBL

Southeast Bank Ltd.

SEBL

Standard Bank Ltd.

SBL

City Bank Ltd.

CBL

United Commercial  Bank Ltd.

UCBL

Uttara Bank Ltd.

UBL

Social Islami Bank Ltd.

SIBL

 

Data Envelopment Analysis (DEA) Approach

DEA is the non-parametric methodology; it is utilized for the estimation of effective boondocks and to decide the execution of DMUs, for example, banks. As per Charnes, et al. (1978) characterized productivity as a weighted entirety of yields to a weighted aggregate of data sources. This examination utilizes the two most often utilized models, among various Data Envelopment Analysis models: the BCC show - (Banker et al. 1984) and the CCR demonstrate (Charnes et al. 1978) to examine the relative proficiency.

……………………. (3.1)

 

The Malmquist productivity index (MPI)

The Malmquist Productivity Index is often used in determining official and non- official banking institutions’ productivity. MPI was firstly used by (Malmquist, (1953). this method has three primary favorable circumstances of MPI which define it better to different procedures. In the first place, it doesn't require the cost minimization or benefit amplification presumption. Second, there is no expecting of info and yield costs. Third, if there is board information, the strategy allows the deterioration of profitability into two. The current article utilizes the yield arranged Malmquist profitability record for DMU between the period's t and t + 1 as received by Färe et al. (1994), as pursues.

 

 (  =   …………… (3.2)       

 

=       …………………

 

This means the profitability of the generation point (𝑥𝑡+1, 𝑦𝑡+1) relative to the creation point (𝑥𝑡 , 𝑦𝑡 ). An esteem more prominent than 1 signifys positive TFP extension from period t to t+1.

= TEC ( , )  TC ( , ). ………………………………….(3.4)

The TEC can be additional broken down into unadulterated specialized effectiveness change (PTE) and scale productivity (SE). Following Färe et al. (1994), disintegration of TEC in Eq. (3) is according to the accompanying,

TEC = ……………………(3.5)

 

where DVRS is the yield separate limit with regards to variable return to-scale (VRS) and DCRS is the yield separate limit with regards to predictable return to-scale (CRS). The hidden segment of Eq. (3.5) addresses the PTE while the second part tends to SE. The PTE shows the utmost of DMU to keep up a key division from wastage by utilizing responsibilities to pass on the best yield while SE is the capacity to work at an ideal scale (Bassem, 2014). The multiplication of TC and TEC produces TFP whose respect more than 1 is productivity progress, less than one is regress and 1 is stagnant.

Selection of Inputs, Outputs and Approaches

Data Envelopment Analysis (DEA) techniques preserve either be input oriented or output oriented. The requisites output and input-oriented transmit to the way in which unproductive DMUs are anticipated against the efficient frontier. Input-oriented techniques go for lessening the input sums by however much as could reasonably be expected without decreasing present output levels and Output-oriented techniques expand output levels without expanding input utilization ( Muvingi, J. and  Hoteral, S, 2015). According to Sufian, (2007), the clarification and the measurement of bank's data sources and yields in the keeping money reason buildup far from being obviously true among the analysts. To choose what makes information sources and yields out of banks, one needs to initially decide the character and nature of keeping banking technology. Thus, the choice of variables in efficiency and productivity studies extensively influences the attained outcomes (Das and Ghosh, 2006).

In the literature relating to banking fields, there are three chief approaches that are extensively used for defining inputs and outputs in the study of the banks efficiency were developed, namely: the intermediation approach, the production and value-added approaches (Sealey and Lindley, 1977; Benston, 1965; Frexias and Rochet, 1997).

The first approach is the intermediation approach which defines a bank act as an intermediary that transforms assets from the excess units to shortage units. To scrutinize the effectiveness of banking areas in evolving countries, the intermediation approach is the chosen approach amongst researchers applying the DEA technique (e.g., Sufian, et al. 2012; Bader, et al. 2008; Sufian, et al. 2012). This approach was used by the earlier examination on the effectiveness of the banking industries in developing nations (e.g., Bhattacharyya, et al. 1997; Sufian 2009; Sathye 2001 and Charnes, et al. 1990).

The second approach is the production approach which indicates that monetary organizations like banks act as creators of services for account holders, and therefore the inputs are the physical elements such as labour and capital and all other assets as well as liabilities are outputs. Earlier investigations which implement a similar approach are DeYoung (1997), Ferrier and Lovell (1990) and Fried, et al. (1993). The third approach, the esteem included approach, perceives bookkeeping report classes (assets or liabilities) as yields which add to the regard alternative of a bank, for example, business related with the usage of certifiable resources (Berger, et al. 1987).

Therefore, two input variables and two output variables are being selected by utilizing an intermediate approach. The decision of the info and yield factors depends on Garamu, G. (2016) and other most important an on the efficiency of the banking industries in rising countries (e.g., Muvingi,J. & Hoteral, S, 2015; Sufian, Muhamad, et al., 2012; Bader et al., 2008; Sufian, Kamarudin, et al., 2012; Isik and Hassan, 2002; ; Hassan, 2005; Sufian and Habibullah, 2009). Table 2 demonstrates the inputs and outputs adopted in the current study.

Inputs

Outputs

No. of Employee

Total Deposits

Fixed Assets

Total Loan and Advances

 

Consequently, in light of this data sources and yields, the efficiency benchmarks by MPI examination are given as pursues:

Table 3: Malmquist Productivity Index (MPI)

Malmquist Productivity

Index Productivity Level

M less than 1

Decline in Productivity

M equal to 1

Stable in Productivity

M greater than1

Increase in Productivity

 

Empirical Findings

Technical Efficiency and Productivity of the Bangladesh Banking Sector: Data from Particular periods

From table 5 below illustrates that, the listed commercial banks in Bangladesh are around 98% technically efficient on an average. The findings also imply that the banks are depleting only 2% of their inputs on an average. The empirical results also suggest that six banks, namely, BBL, DBBL, EXIMBL, NBL, SBL and UBL, have presented the utmost technical efficiency level. Conversely, rests of the commercial banks have showed the technical inefficiency over the study period.

Therefore, the greater technically efficient bank is BRAC Bank Ltd. (122%) and the minimum efficient bank is Eastern Bank (92%). This indicates that, on an average, BRAC Bank applies 122% of its inputs to produce outputs and Eastern Bank misuses 8% of its inputs to produce output. BRAC Bank Ltd. attained entire efficiency (a score of 1) for 2 years whereas Eastern Bank could not attain entire efficiency for a single year during the study period under analysis.

Table 4: Descriptive Statistics

Variable

Mean

Std. Dev.

Min

Max

Total Deposit

182746.5

109901.6

76077

755022.3

Total Loan and Advances

163013.4

112930.7

61328.6

748672.5

No of Employee

3392.404

2760.316

1270

13760

Fixed Assets

4816.748

3102.756

433.09

15926.36

Source:  Author’s own computation based on annual reports of sample Banks

 

Table 5: Technical Efficiency of the Sample Banks Based on CCR model

Bank Name

2013

2014

2015

2016

2017

Banks Mean

AB Bank Ltd.

1.000

1.000

1.000

0.927

0.897

0.965

Al-Arafah Islami Bank Ltd.

1.000

0.973

1.027

0.866

0.898

0.953

Bank Asia Ltd.

0.995

0.988

0.904

1.079

0.975

0.988

BRAC Bank Ltd.

1.336

0.945

1.021

0.608

0.644

0.911

Dutch-Bangla Bank Ltd.

1.184

1.001

0.986

0.722

0.701

0.919

Dhaka Bank Ltd.

1.016

1.006

0.973

1.129

0.946

1.014

Eastern Bank Ltd.

1.000

0.896

1.025

0.945

1.091

0.991

EXIM Bank Ltd.

0.927

0.972

1.150

0.996

1.004

1.010

Islami Bank Bangladesh Ltd.

1.045

1.005

0.880

0.897

0.804

0.926

Mercantile Bank Ltd.

0.870

1.004

0.966

1.021

0.981

0.968

National Bank Ltd.

1.000

1.000

1.000

0.698

0.766

0.893

One Bank Ltd.

0.901

0.956

0.893

1.029

1.467

1.049

Shahjalal Islami Bank Ltd.

0.953

0.938

0.900

1.104

1.257

1.030

Southeast Bank Ltd.

1.000

1.000

1.000

1.000

1.000

1.000

Standard Bank Ltd.

0.945

1.143

0.996

1.451

0.804

1.068

City Bank Ltd.

1.018

0.895

0.819

0.879

0.794

0.881

United Commercial  Bank Ltd.

0.996

0.981

1.055

1.009

1.009

1.010

Uttara Bank Ltd.

1.052

0.904

0.955

0.802

0.808

0.904

Social Islami Bank Ltd.

1.047

0.930

0.889

0.932

0.993

0.958

 Year Mean

1.011

0.974

0.968

0.936

0.922

0.962

Source:  Author’s own computation based on annual reports of sample Banks

 

 

        Figure 1: Level of Technical Efficiency for Sample Banks from (2013-2017)

The mean level of technical efficiency for sample banks during the study periods 2013–2017 is given in figure 1. From Figure 1, it can be noticed that the level of technical efficiency change is continuously declining during the years 2013 to 2017. This indicates the insignificant sign for improving productivity among the sample banks over the time period. In 2013, technical efficiency change stretches its climax phase and in 2017, it stretches the minimum phase.

Table 6: Annual Average TE and Its Elements, PTE and SE

 

2013

2014

2015

2016

2017

Average

Technical Efficiency (TE)

1.011

0.974

0.968

0.936

0.922

0.962

Pure Technical Efficiency (PTE)

1.028

0.995

0.982

0.979

1.000

0.997

Scale Efficiency (SE)

0.983

0.979

0.985

0.956

0.922

0.965

Source:  Author’s own computation based on annual reports of sample Banks

 

As exposed by table 6 above, the annual average technical efficiency of listed commercial banks over the study period was .962 which illustrates the sample banks could have decreased their input approximately by 3.8% according to generate the similar point of output. This finding signifies that on an average sample banks are comparatively technically inefficient. Additionally, the annual mean of PTE and SE were 0.997 and 0.965 together. When the pure technical efficiency is larger than scale efficiency, then ineffectiveness is caused by scale ineffectiveness.

Therefore, the basis of technological inefficiency among listed commercial banks working in Bangladesh on an average is owing to scale inefficiency. Thus, the current study is agreeing with Garamu, G. (2016), Muvingi, J. & Hoteral, S, (2015), Soetanto and Ricky (2011), Darwin (2012), Usman et al., (2010) those disputed that the basis of inefficiency is scale ineffectiveness instead of  PTE while it opposes the results of Miller and Noulas (1996) and Drake and Hall (2003). According to Miller and Noulas (1996) investigated that PTE of bulky United States banks is double as large as scale inefficiency. On the other hand, Drake and Hall (2003) found that the mostly inefficiency of the Japanese commercial banks is due to pure technical incompetence instead of scale incompetence.

Table 7: Malmquist TFP Index Summary of Sample Banks Average

No

Bank

Effch

Techch

Pech

Sech

Tfpch

1

AB Bank Ltd.

0.964

0.931

0.981

0.982

0.897

2

Al-Arafah Islami Bank Ltd.

0.951

0.975

0.966

0.984

0.928

3

Bank Asia Ltd.

0.986

0.942

1.001

0.986

0.929

4

BRAC Bank Ltd.

0.872

1.158

1.046

0.834

1.01

5

Dutch-Bangla Bank Ltd.

0.900

1.019

0.972

0.926

0.918

6

Dhaka Bank Ltd.

1.012

0.955

1.000

1.012

0.967

7

Eastern Bank Ltd.

0.989

0.922

1.000

0.989

0.912

8

Export-Import Bank of Bangladesh Ltd.

1.007

1.049

1.000

1.007

1.057

9

Islami Bank Bangladesh Ltd.

0.922

0.956

1.000

0.922

0.882

10

Mercantile Bank Ltd.

0.967

0.94

0.977

0.99

0.909

11

National Bank Ltd.

0.882

1.038

0.977

0.903

0.916

12

One Bank Ltd.

1.03

0.946

1.041

0.99

0.975

13

Shahjalal Islami Bank Ltd.

1.022

0.952

1.024

0.999

0.974

14

Southeast Bank Ltd.

1.000

0.941

1.000

1.000

0.941

15

Standard Bank Ltd.

1.046

1.061

1.09

0.96

1.11

16

City Bank Ltd.

0.878

0.95

0.921

0.953

0.834

17

United Commercial  Bank Ltd.

1.010

0.95

1.03

0.98

0.96

18

Uttara Bank Ltd.

0.899

1.02

0.969

0.928

0.918

19

Social Islami Bank Ltd.

0.957

0.948

0.956

1.001

0.907

 

Average

0.961

0.98

0.997

0.965

0.942

 

The MPI assesses the productivity changes after some time which is estimated as the result of technical efficiency change and technological change. The fundamental decide that is pursued assessment is if the TFP index is greater than 1, it illustrates banks are functioning efficiently. However, in the event that the estimation of TFP is less than 1 that implies efficiency is deteriorating during the investigation time frame.

Table 7 reveals about TFP performance of listed commercial sector banks in Bangladesh. The responsibility of commercial banks is growing constantly in Bangladesh banking sector. According to TFP change, 3(16%) banks improved their average annual TFP, while 16 (84%) banks lessening in total factor productivity described. It also discovered that in 6(around 32%) banks technological improvement has been noticed. On the other hand, 13(around 68%) banks illustrate the decline in technology during the study period. Hence, the empirical findings show, on average, the total factor productivity variation is 0.942 which is less than 1; it signifies a 5.8% deteriorate over the study period (2013-2017). The decline in total factor productivity has been particularly as results of technical efficiency deteriorate in listed commercial banks.

From Figure 2 demonstrate that Standard Bank Ltd. recorded the highest mean positive augment in TFP of 11%. The 11% productivity achieve for Standard Bank Ltd. contains an efficiency grow of 4.6 % and technological increases of 6.1%. Export-Import Bank of Bangladesh Ltd. reported the lowest average TFP transform with an average deterioration of around 17% in total factor productivity index. This 17% deterioration comprises of an efficiency and technological decline of 12.2% and 5% respectively. The empirical results also illustrate that only 3 among 19 banks employed in this paper displayed an overall improvement in productivity and sixteen banks presented an overall decline in productivity.

The results indicate that the unproductive banks utilize surplus inputs and under composed output which exposes management and scale inadequacy of the banks.

 

 

Figure 2:  Level of Average Total Factor Productivity (TFP) change in the Bangladeshi listed Commercial Banking Sectors.

Table 8: Average Changes in TFP and Its Elements by Year

year

Effch

Techch

Pech

Sech

Tfpch

2013

1.011

0.886

1.028

0.983

0.895

2014

0.974

0.945

0.995

0.979

0.921

2015

0.968

0.962

0.982

0.985

0.931

2016

0.936

1.046

0.979

0.956

0.979

2017

0.922

1.073

1.000

0.922

0.989

Average

0.961

0.980

0.997

0.965

0.942

Source:  Author’s own computation based on annual reports of sample Banks

 

Table 8 portrays of the average change in TFP and its elements by year. The findings expose that the annual average technical efficiency (TE) is equal to 0.961.  This is the outcome of a decrease in technological, PTE and SE change averagely by 2%, .3% and 3.5% respectively. Technical efficiency change influences the climax phase in the period 2013 and the lowest phase in 2017. However, the annual average technological transform of Bangladeshi commercial banks is .980. This indicates that the technological decline of 2% during the study period 2013 to 2017. In 2017, technological transform stretches its peak phase and in 2013, it stretches the minimum phase.

Thus, the factual result signifies that TE transform is constantly reducing, on the other hand, the technological transform is constantly growing. This implies that Bangladeshi commercial banks are concentrating on technological progress than executive superiority in the distribution of properties into the input to output mix up. The average total factor productivity change during the years (2013-2017) is 0.942 which indicates a decline in TFP. This is the outcomes of average annual TE change of 0.961 and technological transformation of .980. This illustrates that TE deterioration for decreasing TFP. Further, the findings also expose that the annual average PTE change is 0.997, while SE change is .965. This indicates TE decline is owing to managing ineffectiveness rather than scale efficiency.

Finally, we terminate that findings of the study are agreeing with Garamu, G. (2016), Muvingi, J. & Hoteral, S, (2015),Soetanto and Ricky (2011), Darwin (2012), Usman et al., (2010) those disputed that the basis of inefficiency is scale ineffectiveness instead of  PTE, and also in line with Sturm and Williams (2004) which indicates that the prime basis of efficiency progress was technological change instead of  TE change.

Conclusions and Managerial Implications

To date, studies on bank technical efficiency and productivity are abundant. Nonetheless, the majority of these papers have focused on the banking areas of developed nations. The current study endeavours to fill in this exacting gap and supports recent experimental information on the technical efficiency and productivity of the Bangladesh commercial banking area during the periods 2013–2017 using DEA and MPI as well as a strongly balanced dataset of 114 observations from nineteen banks.

The empirical results from this paper show that, listed commercial banks are on average comparatively technically inefficient. The empirical results also suggest that six banks, namely, BBL, DBBL, EXIMBL, NBL, SBL and UBL, have presented the utmost technical efficiency level. Conversely, rests of the commercial banks have showed the technical inefficiency over the study period. However, the empirical results also illustrate that only 3 among the 19 banks employed in this paper displayed an overall improvement in productivity and sixteen banks presented an overall decline in productivity.

This study obviously calls for decision-makers and regulators to assess the technical efficiency and productivity improvement trend in Bangladesh banking sector. Therefore, the inefficient bank's administration should take care about scale and technical efficiency while assigning their inputs. To achieve the proficiency frontier, the administrators of inefficient banks have to centre on the overabundance utilized inputs and lessen the excess used amount. Bank supervisors should be aware of benefit and proficiency levels. This infers banks working in the Bangladesh continuing keeping banking fragment need to consider each conceivable technologies that could improve their benefit proficiency levels since the principle intention of banks is to intensify financial investors' esteem or riches through profit expansion. Finally, the movements of productivity development throughout the study period show slightly increase. For keeping this trend, banks need to arrange training and education program expected to progress managerial capability, or enlargement programs planned to boost-up the ratification of new technologies.

Therefore, the future study may look at the extra variables. In addition, in the nonexistence of correct compute of the input prices, the study concentrated on technical efficiency rather than allocate and economic efficiencies. Whereas endorsing these limitations, next we expect that they do not notably diminish the significance of a comprehensive investigation of technical efficiency and productivity change in Bangladeshi banking sectors.

Acknowledgement

We would like to grateful Dr. Aslam Mia for his helpful guidance in the process of completing this manuscript and the anonymous referee for the constructive comments and feedbacks, which have notably improved the contents of the article. All remaining errors are our own responsibility.