IMPACT OF AGENT BANKING ON BANK PROFITABILITY IN BANGLADESH

A Thesis Report submitted to

Department of Finance, University of Dhaka
In partial fulfillment for the award of the degree of

MASTER OF BUSINESS ADMINISTRATION (MBA)

By
Bishwanath Das, 40001

Under the Supervision of

Dr. Muhammad Saifuddin Khan, Associate Professor

MBA Program (Evening)

Department of Finance, Faculty of Business Studies
University of Dhaka

December 30, 2021

Summer 2021
Impact of Agent Banking on Bank Profitability in Bangladesh

……………………………………..
Dr. Muhammad Saifuddin Khan
Associate Professor & Placement Director
Department of Finance
University of Dhaka

Supervisor’s Remarks

Signature

……………….………..
Bishwanath Das
Batch: 40th
ID: 40001
 Student’s Declaration

I declare that the submitted project paper is original and solely produced by me. This is not a
copy in any form from any sources. If this paper is found to be a copy work which falls under
the plagiarism norms, this paper will be deemed canceled and can be reported to the appropriate
authority of the University.

Signature

……………….………..
Bishwanath Das
Batch: 40th ID: 40001

Consultation Schedule with the Supervisor

Consultation # Date Remarks Signature of the
Supervisor
1 September 03, 2021 1st Meeting with the Supervisor

2 September 20, 2021 2nd Meeting with the Supervisor

3 October 01, 2021 3rd Meeting with the Supervisor

4 October 29, 2021 4th Meeting with the Supervisor

5 November 05, 2021 5th Meeting with the Supervisor

6 November 12, 2021 6th Meeting with the Supervisor

7 November 26, 2021 7th Meeting with the Supervisor

8 December 03, 2021 8th Meeting with the Supervisor

9 December 10, 2021 9th Meeting with the Supervisor

10 December 17, 2021 10th Meeting with the Supervisor

 Acknowledgment

Firstly, and for most, I would like to express my deepest gratitude to God Almighty who has
blessed me and enabled me to do this work. Secondly, a particular expression of gratefulness
must go to my supervisor, Dr. Muhammad Saifuddin Khan, under whose well-organized
direction and academic guidance the research study conducted. He offered me not only
directions but also help a lot for the development of theoretical concepts, active suggestions,
and practical feedback. Thirdly, I would also like to appreciate and recognize recommendations
from friends and colleagues for refining the insight of this study. Lastly, I would like to express
my special gratitude to my family mostly my father and mother for their great support and
guidance during study time. In addition, I would heartily thank to Authority of Business faculty,
Dhaka University for providing such a nice environment for doing this research study.
 Abstract

Commercial banks always search for a cost effective approach for providing financial service
to “unbanked” people instead of using bank branches and their own work force. The concept
of agent banking is getting much consideration in our economic and financial system in a very
respectable way for its related benefits and cost effectiveness. Using agent-banking service,
people can send and receive funds even in very tiny amounts easily and safely. The purpose of
the study was to investigate the impact of agent banks on the profitability of banks in
Bangladesh. I have taken ten commercial banks in Bangladesh as the sample for the period of
October 2016 to July 2021 and data samples have been taken on a quarterly basis. For
conducting the study, I have taken secondary data from Bangladesh Bank website, commercial
bank website and annual reports. To represent the profitability, I have taken two proxies that
are Return on Equity (ROE) and Return on Assets (ROA) and Independent variables are- no.
of agents, no. of outlets, and no. of accounts, deposit, remittance and loan. For conducting the
regression analysis, the panel data methodology has been used. For analyzing the data
Microsoft Excel and SATA software has been used. To represent the profitability of the bank,
I have taken Return of Assets (ROA), Return of Equity (ROE) and Net interest Margin (NIM)
as dependent variables and Independent variables are- Number of agent, Deposit, Loan and
Remittance. The research study found that there is significant positive correlation between
number of accounts and profitability of bank and it has a significant positive impact on ROA,
ROE and NIM. The research study also found that there is significant positive correlation
between volume of remittance distribution and profitability of bank and it has a significant
positive impact on ROA, ROE and NIM. In addition, study also found that there is a negative
correlation between number of agents, volume of deposit and bank profitability. There is also
a positive relation between age and profitability. There is a recommendation that management
should more concentration on expanding agent banking service for increasing the profitability
of the bank. In Bangladesh, Agent banking is the new concept and so far, very little study has
been conducted in this area. The study I have done cannot cover all sides of agent banking and
I expect future researchers would have lots of interest in conducting research in this area.

Keywords: Agent banking, Bank Profitability, Panel Data

 Table of Contents

Page Number
1 Introduction ...................................................................................................... 1
1.1 Statement of the Problem ............................................................................ 1
1.2 Objective of the Study ................................................................................ 1
1.3 Research Hypothesis................................................................................... 2
1.4 Significance of the study............................................................................. 2
1.5 Limitation of the study................................................................................ 2
1.6 Scope of the study ...................................................................................... 3
1.7 Outline of the study .................................................................................... 3
2 Theoretical Background..................................................................................... 4
2.1 Theoretical Background .............................................................................. 4
3 Literature Review.............................................................................................. 5
3.1 Theoretical review ...................................................................................... 5
3.1.1 Bank-focused theory .............................................................................. 5
3.1.2 Bank Led theory .................................................................................... 5
3.1.3 Non-Bank Led Theory ........................................................................... 6
3.2 Review of past research .............................................................................. 6
3.3 Conceptual Framework ............................................................................... 8
4 Synopsis on Agent Banking in Bangladesh........................................................10
4.1 Brief of Agent banking coverage ................................................................10
4.1.1 Growth of Agents .................................................................................11
4.1.2 Growth of Accounts..............................................................................13
4.1.3 Growth of Deposit acquisition...............................................................13
4.1.4 Growth of Loan Disbursement ..............................................................14
4.1.5 Growth of Remittance Distribution........................................................14
4.2 Pioneer of agent banking in Bangladesh: Bank Asia Ltd .............................15
5 Data and Methodology .....................................................................................17
5.1 Sources of Data and Study Period ..............................................................17
5.2 Sampling Technique and Sample Size ........................................................17
5.3 Measurement of Variables .........................................................................18
5.3.1 Return on Assets (ROA) .......................................................................18
5.3.2 Return on Equity (ROE) .......................................................................18
5.3.3 Number of Agents ................................................................................18

Page | i
 5.3.4 Number of Accounts.............................................................................18
5.3.5 Volume of Deposit ...............................................................................18
5.3.6 Volume of Loan Disbursement..............................................................19
5.3.7 Volume of remittance Distribution ........................................................19
5.4 Analytical Techniques ...............................................................................19
5.5 Econometric Model ...................................................................................19
6 Data Analysis and Findings ..............................................................................20
6.1 Summary Statistics ....................................................................................20
6.2 Correlation Analysis ..................................................................................21
6.3 Regression Analysis ..................................................................................22
6.3.1 Regression Result ROA ........................................................................23
6.3.2 Regression Result ROE.........................................................................24
6.3.3 Regression Result NIM.........................................................................25
6.4 Other Analysis...........................................................................................26
6.4.1 Profitability trend .................................................................................26
6.5 Findings ....................................................................................................28
7 Conclusion .......................................................................................................29
7.1 Conclusion ................................................................................................29
7.2 Policy Implication .....................................................................................30
7.3 Scope of Future Research ..........................................................................30
References .............................................................................................................31
Appendices

List of Tables

Table 1: Agent banking activities at a glance ...........................................................10

Table 2: List of commercial bank in agent banking operation ...................................11
Table 3: At a glance Bank Asia agent operation.......................................................15
Table 4: List of commercial banks as sampled .........................................................17
Table 5: Summary Statistic .....................................................................................20
Table 6: Correlation Matrix ....................................................................................21
Table 7: Regression Result ROA.............................................................................23
Table 8: Regression Result ROE .............................................................................24
Table 9: Regression Result NIM .............................................................................25
Table 10: Trend Analysis of ROA, ROE & NIM .....................................................26

Page | ii
 List of Figure

Figure 1: Conceptual Model .................................................................................... 9

Figure 2: Bank wise number of agents .....................................................................12
Figure 3: Number of Agent growth .........................................................................12
Figure 4: Number of accounts growth .....................................................................13
Figure 5: Volume of Deposit growth .......................................................................13
Figure 6: Volume of Remittance Growth.................................................................14
Figure 7: Agent growth of Bank Asia ......................................................................16
Figure 8: Accounts growth of Bank Asia .................................................................16
Figure 9: Average ROA Trend ................................................................................27
Figure 10: Average ROE & NIM trend....................................................................27

Page | iii
 1 Introduction

1.1 Statement of the Problem

Bank financial institutes are always searching for a cost effective new approach for providing
financial service to “unbanked” people instead of using bank branches and their own work
force. The concept of agent banking is getting much consideration in our economic and
financial system in a very respectable way for its related benefits and cost effectiveness. The
implementation of agent banking is primarily operating to develop the portion of overall
industry share by absorbing and holding their customers, refining financial profitability and
providing diversity of services. However, the degree to which agent banking is regularly
utilized as a tool to develop financial areas remains unclear as well as the genuine advantages
of agent banking model to clients, banks and agents remain unstudied. In spite of the fact that
clients have helped a great deal over agent banking, it is unclear whether the financial
profitability of banks has enhanced because of operating agent banking. Therefore, in this
study, I have tried to address the research question- Does agent banking have any connection
for the financial profitability improvement of commercial banks in Bangladesh?

1.2 Objective of the Study

The main objective of this research study is to determine the impact of agent banking on
profitability of Bank in Bangladesh.

Specific Objective
a. To find out the significant positive correlation between number of agent growth and
profitability of Bank.
b. To determine positive relationship between agents’ deposit and Bank profitability.
c. To find out positive correlation between growth of account and Bank profitability.
d. To determine the correlation between agent’s remittance facility and Bank profitability.
e. To determine significant positive correlation between agents’ credit disbursement volume
and profitability of Bank.

Page | 1
1.3 Research Hypothesis
The research hypotheses of the study are:
H1: There is a positive correlation between number of agent and bank profitability.
H1: There is a positive correlation between deposit acquisition and bank profitability.
H1: There is a positive correlation between number of account and bank profitability
H1: There is a positive correlation between remittance disbursement and Bank Profitability.
H1: There is a positive correlation between loan disbursement and Profitability of Bank.

Research Questions
a. To what extent does growth of agent impact on bank profitability?
b. To what extent does agent deposit acquisition impact on bank profitability?
c. What are the impact of agent customers' account opening on the profitability of bank?
d. Does agent remittance disbursement service impact on bank profitability?
e. To what level does growth of agent credit disbursement impact on bank profitability?

1.4 Significance of the study

This research will be importance to bank for taking decision regarding their connection in agent
banking. Although agent banking emerging issue in the financial sector is the principal element
of bank success and there is necessary to know the impact of agent banking on the financial
profitability of bank. This research will be significant in policymaking as well as it will help to
policy creators for formulating the efficient policy in agent banking. It will also be significance
to government for adding or modifying policy in agent banking for promoting the business.
Academicians and different scientists will get review material that will be valuable in their
investigations since they will have a prepared source of literature analysis.

1.5 Limitation of the study

For conducting the research, I have taken 10 commercial banks as a sample out of 28 banks
that are presently functioning agent banking services in Bangladesh. This study concentrated
mostly on secondary data that is not as inclusive and lack rationality. To mitigate the secondary
data issue that has been collected from Bangladesh Bank that is highly trustworthy as it endures
rigorous scrutinizing by qualified financial analyst and auditors. For conducting research study,
I have to face so much time constraint issue due to full professional as well as student.

Page | 2
1.6 Scope of the study
This research study emphases on the impacts of agent banking on profitability of bank in
Bangladesh. The research focuses mostly on 10-listed commercial banks out of 28 bank that
have been providing agent banking service in Bangladesh.

1.7 Outline of the study

This study is structured in 6 chapters. Followed by chapter introduction that contains statement
of the problem, objective of the study, research hypothesis, significance of the study, limitation
of study. Chapter 2 contains theoretical background of the study. Chapter 3 contains a brief
literature review of agent banking. Chapter 4 contains sources of data and study period,
Sampling Technique and Sample Size, Measurement of variables, Analytical Techniques and
Econometric Model. Chapter 5 contains summary statistics, correlation matrix, multiple
Regression Analysis, Findings. Chapter 8 includes conclusion and policy recommendations.

Page | 3
 2 Theoretical Background

2.1 Theoretical Background

For quite a while since the beginning of financial services in Bangladesh, financial
organizations have used an old-style method which is to set up bank branches in different areas
for providing banking services. This old-style method of banking made numerous regions in
Bangladesh particularly rural regions to encounter absence of banking services for such a long
time while in metropolitan regions the conventional method of banking came about to long
chains of client lining in banks branches. As indicated by Veniard and Melinda (2010), a crucial
obstacle for giving financial services by a method of conventional branches were intrinsic
extreme expenses of old-style financial techniques particularly in regions where there are very
few transactions. For maximizing the financial profitability, banks open branches in diverse
locations and provide services but in some cases banks financial profitability reduces to less
transactions made by customers which result in closed branches.To resolve that issue, agent
banking has come to focus. Agent banking is changing financial business which has a
significant impact on banking business and that is presently not generally restricted to the
branches where one needs to move toward the branch for deposit and withdraw cash. Agent
banking has also empowered rural clients to get fundamental financial services, for example,
deposits, withdrawals, remittance disbursement facility, utility bills payment facilities, balance
enquiry, insurance payment, credit and debit card applications and other services. Branchless
banking envelops the arrangement of an expansive scope of banking service outside traditional
banking model and agent banking advancements. Agent banking has become especially broad
over the years and it is booming over the year in Bangladesh. The central bank of Bangladesh
has taken a few drives to widen the agent banking services in Bangladesh for which
introduction of Agent Banking in Bangladesh in 2013 was one of the dynamic drives to bring
the unbanked particularly rural individuals under banking service in a savvy way. The Central
bank of Bangladesh also created several standard guidelines and instructions to guarantee
security of customer deposit and other issues in agent banking model.

Page | 4
 3 Literature Review

The literature review is a significant portion of investigation that assists me to recognize the
gap of research topic and provide me with a theoretical background. Besides, the analysis of
past research expands my understanding on my topic in various viewpoints that include markets
and regions. This chapter presents a review of past study of agent banking research and
empirical data that conduct my study eloquently. It additionally checks out the summary of
literature review and scrutinizes literature as well as gaps of previous research findings. It also
incorporates the conceptual framework of variable relation.

3.1 Theoretical review

Bank without branch signifies an innovative delivery channel that allows commercial banks to
offer banking services outside the old-style bank premises. As per theory, there are three
categories to explain branchless banking which are- Bank-focused theory, Bank-led and
Nonbank-led theory.

3.1.1 Bank-focused theory

Bank focused theory arises when there is a requirement that banks are required to mobilize low
cost deposits as well as deliver financial service to existing customers through non-
conventional ways which are automatic teller machines (ATM) to internet banking (IB) or
mobile banking to agent banking, Voice banking, SMS banking etc. (Vutsengwa & Ngugi,
2013). This model highlights the safety, consistency and effectiveness of financial institutes
for providing financial service to clients (Kengere, 2014). Commercial banks can minimize
risk (Yakub, Bello, & Adengua, 2013) by providing banking service through internet banking,
agent banking and mobile banking.

3.1.2 Bank Led theory

Under this model, banks create an agreement with agents for providing service to customers as
delivery channels and agents’ mother accounts maintained by the bank itself. Banks take
approval from regulators for providing service to customers by agent outlet as well as bank use
agent for new product promotion, new production lunch and new service delivery channel.
Onwei (2015) stated that using this model banks preserve customers' detailed information

Page | 5
through agent outlet use as a delivery channel that significantly assists banks to largely
improvise their financial services. Under this model, customers can take services from various
agent points, postal office points, and merchant points that relieve them from long queues in
traditional branch banking.

3.1.3 Non-Bank Led Theory

Under non-bank-led theory, customers do not maintain bank accounts; rather they deal with
nonbank firms using either mobile network or agent banking services. In theory, banks do not
perform a primary rather than a mobile operator or third party agent perform as a principal for
providing services and holding the customers KYC and other details information. Third party
nonbank organizations get approval for providing the banking service and dealing with
customers in agent banking platforms through mobile network or internet (Yakub, Bello, &
Adengua, 2013). As per regulatory perspective, this is model is more risky because third party
dot not take seriously on customer identification which may led anti-money laundering and
counter-terrorism financing issues. Under this model, banks do not perform directly rather they
just work with service providers by providing facilities for smooth operation.

3.2 Review of past research

Kitali, Chepkulei, & Shibairo (2015) performed research for finding the effect of agency
banking on customer satisfaction in Kenyan banks. That research performed for establishing
the relationship between agency banking and numerous categories of financial services i.e. on
time service, high quality of service, low cost service delivery to agent customers and fulfill
customer’s necessity. Preliminary data gathered from an agency banking outlet by a random
sample of 250 customers and research found that majority customer satisfaction was higher for
agency banking services.

Rahman (2016) performed research studies for finding the impact of agent banking to the
development of the SME sector in Bangladesh. In that research, data gathering has been done
from primary sources and qualitative research methods have been followed. There are some
questionnaires which are given to SME banking customers as well as agent banking for
collecting data and those data analysis done through SPSS. For conducting this study,
dependent variables were Satisfaction of SME customers and independent variables were agent
network, efficiency of agent, customer dealing capacity, frequency of small business owner

Page | 6
and transaction level security. This research found that agent banking service volume has
significant positive correlation with profitability of banks as well as customer satisfaction level
of SME customers positively impacted by all independent variables.

Caitlin (2013) performed research study for finding the impact of agent banking on financial
inclusion in Brazilian Banks. This study found that there is no effect on banking service if
service availability is far distance, study also found that most of the agent banking customers
are poor, less educate and women and they want to visit regularly in agent banking outlets for
taking banking services.

Karangwa & Mulyungi (2018) conducted a study on the effect of agency banking service on
the financial condition of Rwandaian bank. In their study, they used primary data and their
target population was 60 officials of Equity bank. They used a descriptive research design to
find out the study. Their dependent variable was ROA and the independent variable was the
number of agents, cash deposit, cash withdrawal, cash collection. They found that deposit, bill
payment, and fund transfer had a significant positive impact on the profitability of the bank
whereas the volume of fund withdrawn had a significant negative impact on the bank’s
profitability. They recommend that banks should continue to invest in the agent banking and
should concentrate on innovation.

Hasan (2019) performed research studies for finding the impact of agent banking on
financial profitability of banks in Bangladesh. For conducting the study, six commercial bank
quarterly basis data year 2017 to 2019 used and data taken from commercial bank websites and
Bangladesh Bank. This research control variable was net profit margin after tax and
independent variables were account, deposit, loan and agent outlets and study found that there
is significant positive correlation between agent outlets and profitability of banks.

Karangwa & Mulyungi (2018) performed research studies for finding the impact of agent
banking on the financial condition of Rwandan banks. Preliminary data gathered from sixty
officials of the equity bank. Dependent variables of the study are Return of assets and
independent variables of this study are number of agents, deposit and loan. Study found that
there is a significant positive correlation between deposit and profitability of banks and
recommended that banks should focus more on new innovation and invest on agent banking
for better financial profitability.

Page | 7
Amit (2018) performed research study for investing the perspective of agent banking on female
RMG Staff in Bangladesh. The aim of that research was to discover the supporting and
executing factors of agent banking for RMG staff. For conducting the study, qualitative
research methods and semi-structured questionnaire data were used in association with BANK
ASIA and NRBC Bank. This research recommends some important policies for developing
agent banking in Bangladesh as well as some comments for developing the new products for
fortifying agent banking mobilization.

Mwando (2013) performed a research study for finding the impact of agent banking on
financial profitability of Kenyan commercial banks. For conducting the study, Nine
commercial banks of Kenya chose a descriptive survey research method that has been followed
as some questionnaires have been prepared for doing the survey. Data analysis has been
performed through SPSS and the dependent variables were financial profitability and
independent variables were agent bank regulation, lower transaction cost and availability of
services. Study found that there is a significant positive correlation between financial
profitability and bank regulation, lower transaction and availability of services.

Mukuria & Memba ( 2016) performed research studies for finding the connection between
liquidity and profitability of agent banking in Kenya. For collection data, a survey has been
conducted to managers, executive and agent banking personnel. For analyzing the data both
quantitative and qualitative research used and found that liquidity is the immense challenge for
agent banking profitability in Kenya and recommended that principal banks confirm liquidity
flow for retaining the profitability of commercial banks.

3.3 Conceptual Framework

Conceptual framework is the representation of dependent variables and idependent variables
relatinship. In this research study, dependent variables are- Return on Assets (ROA), Return of
Equity (ROE), Net Profit Margin (NIM) and Independent variables are- Number of agents,
Number of accounts, Volume of Deposit, Volume of Loan Disbursement, Volume of
Remittance Disbursement. The below model is showing the relatioship between dependent
variables and independent variables:

Page | 8
Figure 1: Conceptual Model

Page | 9
 4 Synopsis on Agent Banking in Bangladesh

Agent banking was introduced in 2013 under the guidelines of Bangladesh Bank with bunches
of policies and regulation. The fundamental intention is to provide a convenient alternate
delivery channel of financial service to rural people and unbanked people who are in
geographically diverse locations.By the help of agent banking, commercial banks enlarge
financial services through searching for rural people with minimal expenditure and low cost
deposits that expand the country economy. From the very beginning, agent banking is so much
popular to rural unbanked people, through agent banking service rural unbanked people can
deposit and withdraw money from agent outlets very easily and conveniently. Agent banking
also helps people for remittance disbursement as well as loan disbursement. At present, there
are 28 banks operating agent banking services in Bangladesh.

4.1 Brief of Agent banking coverage

Currently, agent banking become rewarding for customers and commercial banks. Many
commercial banks come forward for operating the agent banking services that rises the volume
of agent banking deposit, volume of loan disbursement and remittance facility.

Table 1: Agent banking activities at a glance

June 2020 June 2021 Change
Year -to- Year
No. of Banks with License 28 28 0
No. of Banks in Agent 23 28 5
Banking
No. of Agents 8,764 12,912 47.33%
No. of Outlets 12,449 17,145 37.72%
No. of Accounts 73,58,190 1,22,05,358 65.87%
Amount of Deposits 10,22,021.00 20,37,928.34 99.40%
Amount of Loan Disbursed 72,054.00 3,18,628.60 342.21%
Amount of Inward Remittance 26,65,059.00 67,95,404.46 154.98%
Source: Financial inclusion department, Bangladesh Bank

The information shows that no. of agents, no. of outlets, no. of accounts, amount of deposits
expanding very significantly. Agent banking which we can find in the information inspires
more individuals. There is 47.33% changes of no. of agent from previous year that means
47.33% no. of agent increases. No. of accounts increase by 65.87% from previous year, amount

Page | 10
of deposit acquisition increase by 99.40% from previous year, amount of loan disbursement
increase by 342.21% from previous year and amount of inward remittance increase by 154.98%
which demonstrate the agent banking potentiality in Bangladesh.

4.1.1 Growth of Agents

Table 2: List of commercial bank in agent banking operation

Sl Bank Name No. of Agents No. of Outlets
1 Bank Asia Limited 4538 4667
2 NRB Commercial Bank Limited 581 594
3 Dutch-Bangla Bank Limited 913 4621
4 Al-Arafah Islami Bank Limited 345 512
5 Modhumoti Bank Limited 421 421
6 Social Islami Bank Limited 167 178
7 Standard Bank Limited 26 26
8 First Security Islami Bank Limited 58 58
9 Agrani Bank Limited 400 400
10 Mutual Trust Bank Limited 174 185
11 Midland Bank Limited 66 80
12 United Commercial Bank Limited 185 185
13 The City Bank Limited 1278 1296
14 AB Bank Limited 124 151
15 Islami Bank Bangladesh Limited 2411 2411
16 The Premier Bank Limited 37 100
17 NRB Bank Limited 318 320
18 BRAC Bank Limited 493 552
19 Eastern Bank Limited 28 29
20 One Bank Limited 99 102
21 Mercantile Bank Limited 111 111
22 Shahjalal Islami Bank Limited 69 69
23 Exim Bank Limited 12 13
24 Padma Bank Limited 1 7
25 Jamuna Bank Limited 10 10
26 Prime Bank Limited 42 42
27 Global Islami Bank Limited 3 3
28 Meghna Bank Limited 2 2
Source: Financial inclusion department, Bangladesh Bank

Page | 11
 Bank wise number of agent

Bank Asia Limited NRB Commercial Bank Limited Dutch-Bangla Bank Limited
Al-Arafah Islami Bank Limited Modhumoti Bank Limited Social Islami Bank Limited
Standard Bank Limited First Security Islami Bank Limited Agrani Bank Limited
Mutual Trust Bank Limited Midland Bank Limited United Commercial Bank Limited
The City Bank Limited AB Bank Limited Islami Bank Bangladesh Limited
The Premier Bank Limited NRB Bank Limited BRAC Bank Limited
Eastern Bank Limited One Bank Limited Mercantile Bank Limited
Shahjalal Islami Bank Limited Exim Bank Limited Padma Bank Limited
Jamuna Bank Limited Prime Bank Limited Global Islami Bank Limited
Meghna Bank Limited

Figure 2: Bank wise number of agents

The primary objective of agent banking is to spread financial service to unbanked people.In
figure4-2 show quarterly basis growth of agent banking for last one year.

Growth of Agents

15000

10000

5000

0
Mar'20
Jun'20
Sep'20
Dec'20
Mar'21
Jun'21

Figure 3: Number of Agent growth

Page | 12
4.1.2 Growth of Accounts
Last one year there was so much increasing trend of agent banking accounts. Below figure4-3
showing the graphical representation of agent banking accounts growth.

Growth of Accounts

15000000

10000000

5000000

0
Mar'20
Jun'20
Sep'20
Dec'20
Mar'21
Jun'21

Figure 4: Number of accounts growth

4.1.3 Growth of Deposit acquisition

Agent banking deposit acquisition increasing trend is higher. Below figure4-4 showing the
graphical representation of agent banking deposit growth.

Growth of Deposit

2500000
2000000
1500000
1000000
500000
0
Mar'20
Jun'20
Sep'20
Dec'20
Mar'21
Jun'21

Figure 5: Volume of Deposit growth

Page | 13
4.1.4 Growth of Loan Disbursement
Agent banking loan disbursement trend is higher. Below figure4-5 showing the graphical
representation of agent banking loan disbursement growth.

Growth of loan

400000

300000

200000
100000
0
Mar'20
Jun'20
Sep'20 Series1
Dec'20
Mar'21
Jun'21

Figure4-5: Growth of loan disbursement

4.1.5 Growth of Remittance Distribution

Agent banking remittance distribution trend is higher. Below figure4-6 showing the graphical
representation of agent banking loan disbursement growth.

Remittance distribution Growth

8000000

6000000

4000000
2000000
0
Mar'20
Jun'20
Sep'20
Dec'20
Mar'21
Jun'21

Figure 6: Volume of Remittance Growth

Page | 14
4.2 Pioneer of agent banking in Bangladesh: Bank Asia Ltd
Bank Asia is the pioneer of agent banking in Bangladesh and it started agent banking journey
in 2014 with the vision of providing financial service to rural unbanked people and within a
short span it became so popular to rural unbanked people by providing remittance distribution
services, deposit collection and bill payments. Currently BANK Asia has 4538 agents that
cover 64 district, 492 upazila and 3689 districts. The service of the bank is under:
 Cash deposit, withdrawal and Transfer
 Inward Remittance Distribution
 Loan Disbursement
 Utility bill payment & Insurance Premium payment

Table 3: At a glance Bank Asia agent operation

No.of agent No. of Accounts Volume of remittance
31-12-16 1110 130047 204.51
31-03-17 1140 162378 269.84
30-06-17 1192 201111 359.21
30-09-17 1223 237278 456.02
31-12-17 1440 312247 862.26
31-03-18 1919 380936 1061.09
30-06-18 2171 514402 1249.96
30-09-18 2296 585205 1544.31
31-12-18 2487 749165 1851.70
31-03-19 2667 875169 2253.55
30-06-19 2877 1064488 2690.65
30-09-19 3094 1207790 3146.98
31-12-19 3422 1915756 3656.99
31-03-20 3422 2457175 4202.22
30-06-20 3576 3007612 4729.21
30-09-20 3889 3243675 5400.02
31-12-20 4257 3606421 6141.28
31-03-21 4276 3968997 6821.92
30-06-21 4538 4374983 7662.78

Page | 15
 500
1000
1500
2000
2500
3000
3500
4000
4500
5000

1000000
1500000
2000000
2500000
3000000
3500000
4000000
4500000
5000000

0
500000
01-12-16
01-12-16 01-02-17
01-02-17 01-04-17
01-04-17 01-06-17
01-06-17 01-08-17
01-08-17 01-10-17
01-10-17 01-12-17
01-12-17 01-02-18
01-02-18 01-04-18
01-04-18 01-06-18
01-06-18
01-08-18
01-08-18
01-10-18
01-10-18
01-12-18
01-12-18
01-02-19
01-02-19
01-04-19
01-04-19
01-06-19
01-06-19
01-08-19
01-08-19
01-10-19
01-10-19
Below are the graphical representation of agent increasing trend:

Bank Asia Agent Increasing trend

01-12-19

Bank Asia Accounts Increasing Trend

01-12-19
01-02-20
Figure 7: Agent growth of Bank Asia

01-02-20

Figure 8: Accounts growth of Bank Asia

01-04-20
Below are the figure of Bank Asia Number of accounts increasing trend:
01-04-20
01-06-20
01-06-20
01-08-20
01-08-20
01-10-20
01-10-20
01-12-20
01-12-20
01-02-21
01-02-21
01-04-21
01-04-21

Page | 16
01-06-21
01-06-21
 5 Data and Methodology

5.1 Sources of Data and Study Period

The main purpose of this research study is to find the impact of agent banking on Bank
profitability in Bangladesh. I have gather the data from secondary sources that are Bangladesh
Bank publications, Bangladesh Bank websites, and Annual reports of Commercial banks. It is
mention that collected data are quarterly basis. Study period was 5 years and study samples are
10 commercial banks in Bangladesh.

5.2 Sampling Technique and Sample Size

This research review is quantitative in nature and the complete population of this research
review is 59 commercial banks in Bangladesh. To get the outcome 10 commercial banks have
been chosen through random sampling method as a sample of this study. So, sample size of
this study is 10 commercial banks in Bangladesh and Sampling technique is random sampling.

Table 4: List of commercial banks as sampled

SL Name Type
1 AB Bank Ltd Conventional
2 Al-Arafah Islami Bank Ltd Islami
3 Bank Asia Ltd Conventional
4 The City Bank Ltd Conventional
5 Dutch-Bangla Bank Ltd Conventional
6 First Security Islami Bank Islami
7 Islami Bank Bangladesh Ltd Islami
8 Mutual Trust Bank Ltd Conventional
9 Social Islami Bank Ltd Islami
10 Standard Bank Limited Islami

Page | 17
5.3 Measurement of Variables
I have cogitate ROA, ROE and NIM for defining the profitability of commercial bank. In this
study, Return on Assets (ROA) and Return on Equity (ROE) has been used as bank’s
profitability indicator and it is clarified by diverse independent variables i.e. number of agents,
number of outlets, number of accounts, volume of deposits, volume of loan distribution and
volume of remittance distribution.

5.3.1 Return on Assets (ROA)

Return on assets is one the most important control variable of this research study. ROA
specifies how competently management makes revenue utilizing the bank’s assets.
Equation: ROA= (Net Income)/ (Total Assets)

5.3.2 Return on Equity (ROE)

Return on equity is another important proxy of this research study. ROE specifies how
competently manager makes revenue utilizing the shareholders’ equity.
Equation: ROE= (Net Income)/ (Shareholders’ Equity)

5.3.3 Number of Agents

Commercial banks provide licenses to agents for providing the service on behalf of the bank.
Agents provide financial services as per instruction and guidelines of commercial banks as well
as Bangladesh Bank. They are key performer for running agent banking activities. No. of agents
is and independent variable that has significant impact on commercial bank profitability.

5.3.4 Number of Accounts

In this research study, Number of accounts another important independent variable that has
significant impact on commercial bank profitability.

5.3.5 Volume of Deposit

In this research study, volume of deposit is a very important independent variable and that has
significant positive impact on commercial bank profitability.

Page | 18
5.3.6 Volume of Loan Disbursement
In this research study, volume of loan disbursement is an independent variable and that has
significant positive impact on commercial bank profitability.

5.3.7 Volume of remittance Distribution

In this research study, volume of remittance distribution is an independent variable and that
has significant positive impact on commercial bank profitability.

5.4 Analytical Techniques

This research study has been conducted through Regression analytical techniques to scrutinize
multifaceted associations between variables. Regression analysis assumes that dependent
variables are significantly impacted by one or more independent variables

5.5 Econometric Model

For conducting the research study, Five years ten commercial banks’ quarterly data has been
used as well as panel data methodology has been used for regression. Functional model of this
research study stated below:

𝑃𝑟𝑜𝑓𝑖𝑡𝑎𝑏𝑖𝑙𝑖𝑡𝑦 = 𝑓 (𝐴𝐺𝐸𝑁, 𝐴𝐶𝑇𝑁, 𝐷𝐸𝑃𝑂𝑆𝐼𝑇, 𝐿𝐷𝑅, 𝑅𝐸𝑀, 𝐴𝐺𝐸 )

Econometrically, the model accepted for this research study is stated below:

𝑅𝑂𝐴ᵢ ̦ₜ = 𝛽0 + 𝛽1 𝐴𝐺𝐸𝑁 + 𝛽2 𝐴𝐶𝑇𝑁 + 𝛽3 𝐷𝐸𝑃𝑂𝑆𝐼𝑇 + 𝛽4 𝐿𝐷𝑅 + 𝛽5 𝑅𝐸𝑀 + 𝛽6 𝐴𝐺𝐸 + 𝜀ᵢ ̦ₜ

𝑅𝑂𝐸ᵢ ̦ₜ = 𝛽0 + 𝛽1 𝐴𝐺𝐸𝑁 + 𝛽2 𝐴𝐶𝑇𝑁 + 𝛽3 𝐷𝐸𝑃𝑂𝑆𝐼𝑇 + 𝛽4 𝐿𝐷𝑅 + 𝛽5 𝑅𝐸𝑀 + 𝛽6 𝐴𝐺𝐸 + 𝜀ᵢ ̦ₜ

𝑁𝐼𝑀ᵢ ̦ₜ = 𝛽0 + 𝛽1 𝐴𝐺𝐸𝑁 + 𝛽2 𝐴𝐶𝑇𝑁 + 𝛽3 𝐷𝐸𝑃𝑂𝑆𝐼𝑇 + 𝛽4 𝐿𝐷𝑅 + 𝛽5 𝑅𝐸𝑀 + 𝛽6 𝐴𝐺𝐸 + 𝜀ᵢ ̦ₜ

Where,
𝑅ᵢₜ= Net Profit for Banks i Deposit= Volume of Deposit
β0 = Constant Value LDR= Volume of Loan Disbursement
β1 to 𝛽6 : = Coefficient of the variables REM= Volume of Remittance Distribution
ε= Error term AGE= Age
AGEN = No. of Agent network ACTN = No. of Accounts

Page | 19
 6 Data Analysis and Findings

6.1 Summary Statistics

Summary analysis is performed for finding the statistical properties of variables. Table-5
provide a summary of measures that contain mean which measures how closely joined the
variables are. Summary statistic also measure spread and variation such as standard deviation.
Data sequences are distorted into a regular log-form to provide an additional accurate
explanation and it shows that maximum variables have changed very rapidly over the period
of time.
Table 5: Summary Statistic
Std.
Variable Mean Dev. Min Max Observations
roa 0.0017 0.0017 -0.0019 0.0114 179
roe 0.0274 0.0262 -0.0303 0.1684 179
nim 0.0063 0.0032 -0.0005 0.0299 179
ln_NoAgent 4.6493 1.9046 0.0000 8.4202 179
ln_NoAccount 10.9825 2.2292 4.6913 15.2914 179
ln_Deposit 4.6756 2.3125 -1.0217 10.2526 179
ln_Remittance 4.5608 2.9810 -2.6593 10.5309 175
ln_Asset 10.4497 0.4497 9.6331 11.9142 179
ln_Age 3.1704 0.2674 2.8332 3.6636 179

The summary statistic describes minimum values of the return on assets(roa), return on
equity(roe), net interest maring(nim), number of agents (In_NoAgent), number of
accounts(ln_NoAccounts), volume of deposits (ln_Deposit), volume of remittance
(ln_Remittance), Assets (ln_Asset), Age(ln_Age), are approximately -0.0019, -0.0303, -
0.0005, 0.0000, 4.6913, -1.0217, -2.6593, 9.6331, 2.8332 while their maximum values are
0.0114, 0.1684, 0.0299, 8.4202, 15.2914, 10.2526, 10.5309, 11.9142, 3.6636 correspondingly.
The mean values are 0.0017 for return on asset, 0.0274 for return on equity, 0.0063 for net
interest margin, 4.6493 number of agents (In_NoAgent), 10.9825 number of accounts
(ln_NoAccounts), 4.6756 volume of deposits (ln_Deposit), 4.5608 volume of remittance
(ln_Remittance), 10.4497 Assets (ln_Asset), 3.1704 Age (ln_Age). The table specify that the
mean values of all dependent and independent variables are in the range between minimum and
maximum values. The standard deviation of return on assets(roa), return on equity(roe), net
interest maring(nim), number of agents (In_NoAgent), number of accounts(ln_NoAccounts),

Page | 20
volume of deposits (ln_Deposit), volume of remittance (ln_Remittance), Assets (ln_Asset),
Age(ln_Age), are approximately 0.0017, 0.0262, 0.0032, 1.9046, 2.2292, 2.3125, 2.9810,
0.4497, 0.2674. The research study also point out that net interest margin has lowest standard
deviation and volume of deposit has height standard deviation.

6.2 Correlation Analysis

The research study is used Pearson correlation for scrutinizing the robustness of connection
between bank profitabilitys as measured by ROA, ROE, NIM and number of agents, number
of accounts, volume of deposits, volume of remittance distribution, assets, age. Correlation
analysis in research is a statistical technique used to quantify the strength of the direct
connection between two variables and process their affiliation. Correlation analysis computes
the degree of progress in one variable because of the change in the other. A high correlation
focuses on a solid connection between the two variables, while a low relationship implies that
the variables are feebly related. There is a positive connection between two variables while an
expansion in one variable prompts expansion in the other variable. Then again, a negative
connection implies that when one variable expands, different reductions as well as the other
way around. The outcome of an analysis of the correlation are as provided in Table-6.

Table 6: Correlation Matrix

1 2 3 4 5 6 7 8 9
1 roa 1.00
2 roe 0.96 1.00
3 nim 0.57 0.57 1.00
4 ln_NoAgent 0.10 0.06 0.11 1.00
5 ln_NoAccount 0.14 0.14 0.24 0.91 1.00
6 ln_Deposit 0.07 0.07 0.12 0.87 0.89 1.00
7 ln_Remittance 0.13 0.10 0.14 0.90 0.92 0.94 1.00
8 ln_Asset -0.09 -0.01 0.11 0.42 0.40 0.48 0.42 1.00
9 ln_Age -0.11 -0.15 0.02 0.09 -0.01 0.11 0.06 0.62 1.00

The correlation matrix Table 6 show that financial profitability Return on Asset (ROA), Return
on Equity (ROE), Net interest margin (NIM) is positively correlated with number of agents,
number of accounts, volume of deposits, volume of remittance distribution as specified by the
correlation coefficient of 0.10, 0.14,0.07 and 0.13 correspondingly. The correlation matrix also

Page | 21
show that the financial profitability of commercial banks Return on Asset (ROA), Return on
Equity (ROE), Net interest margin (NIM) has a negative correlation with assets and age that is
portrayed by correlations coefficients of -0.09, and -0.11 correspondingly. The highest
correlation (0.14) is between Return on Asset (ROA), Return on Equity (ROE), net interest
margin (NIM) and number of accounts. The weakest correlation (0.07) is between Return on
Asset (ROA), Return on Equity (ROE), net interest margin (NIM) and volume of deposits.

6.3 Regression Analysis

Apart from summary statistics and correlation analysis, the research study is also performed
multiple regression model. For performing the regression analysis, I have used panel data
methodology. Under the panel data methodology, random effect model of regression has
performed with generalized least squares (GLS). Random effects regression model is best
suitable for panel data. The convenience of recurring observations on the same elements
permits the researcher to develop the model by providing additional term in regression.
Generalized Least Squares (GLS) estimators of the constraints model are more efficient than
simpler model that are overlooking unobserved factors.

Page | 22
6.3.1 Regression Result ROA

. xtreg roa ln_NoAgent ln_NoAccount ln_Deposit ln_Remittance ln_Asset ln_Age

Random-effects GLS regression Number of obs = 175
Group variable: Code Number of groups = 10
R-sq: within = 0.0134 Obs per group: min = 12
between = 0.2712 avg = 17.5
overall = 0.0466 max = 19
Wald chi2(6) = 4.28
corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.6385

roa Coef. Std. Err. z P>|z| [95% Conf. Interval]

ln_NoAgent -.0000863 .0002632 -0.33 0.743 -.0006023 .0004296
ln_NoAccount .0000998 .0002641 0.38 0.705 -.0004178 .0006175
ln_Deposit -.0003496 .0002354 -1.49 0.137 -.000811 .0001117
ln_Remittance .0002612 .0002187 1.19 0.232 -.0001675 .0006899
ln_Asset -.00025 .0007955 -0.31 0.753 -.0018091 .001309
ln_Age -.0005561 .0012275 -0.45 0.651 -.0029619 .0018497
_cons .0058704 .0064892 0.90 0.366 -.0068482 .018589
sigma_u .00071524
sigma_e .00166597
rho .15563296 (fraction of variance due to u_i)

Table 7: Regression Result ROA

Table 7 shows a model summary of regression analysis between dependent variable Return on
Asset (ROA) and independent variables including number of agents, number of Accounts,
volume of deposits, volume of remittance distribution, assets, age. The Coefficient of
determination R-sq overall value is 0.0466 means about 4.6 percent of variation of Return on
Asset (ROA) that is clarify by independent variables number of agents, number of accounts,
volume of deposits, volume of remittance, assets, age. On the other hand, 95.4% is clarify by
other variables outside the model and error term. Number of observation is 175 and number of
groups are 10 which define this study. Regression result shows that number of agent banking
accounts and volume of remittance distribution has a positive impact on the Return on assets
(ROA). On the other hand, number of agents, volume of deposit, assets, and age has a negative
impact on the ROA of commercial banks.

Page | 23
6.3.2 Regression Result ROE

. xtreg roe ln_NoAgent ln_NoAccount ln_Deposit ln_Remittance ln_Asset ln_Age

Random-effects GLS regression Number of obs = 175

Group variable: Code Number of groups = 10

R-sq: within = 0.0117 Obs per group: min = 12

between = 0.4238 avg = 17.5
overall = 0.0562 max = 19

Wald chi2(6) = 5.09

corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.5329

roe Coef. Std. Err. z P>|z| [95% Conf. Interval]

ln_NoAgent -.0026676 .0038693 -0.69 0.491 -.0102514 .0049161

ln_NoAccount .0032985 .0038827 0.85 0.396 -.0043113 .0109084
ln_Deposit -.0043684 .0034962 -1.25 0.211 -.0112208 .002484
ln_Remittance .0026557 .0032229 0.82 0.410 -.003661 .0089725
ln_Asset .0063994 .0112175 0.57 0.568 -.0155865 .0283852
ln_Age -.0205258 .0173266 -1.18 0.236 -.0544854 .0134338
_cons .0104209 .0917742 0.11 0.910 -.1694532 .190295

sigma_u .00976622
sigma_e .0254518
rho .12834035 (fraction of variance due to u_i)

Table 8: Regression Result ROE

Table 8 shows a model summary of regression analysis between dependent variable Return on
Asset (ROE) and independent variables including number of agents, number of Accounts,
volume of deposits, volume of remittance distribution, assets, age. The Coefficient of
determination R-sq overall value is 0.0562 means about 5.6 percent of variation of Return on
Asset (ROA) that is clarify by independent variables number of agents, number of accounts,
volume of deposits, volume of remittance, assets, age. On the other hand, 94.4% is clarify by
other variables outside the model and error term. Number of observation is 175 and number of
groups are 10 which define this study. Regression result shows that number of agent banking
accounts, volume of remittance distribution and assets has a positive impact on the Return on
assets (ROE). On the other hand, number of agents, volume of deposit, and age has a negative
impact on the Return on Equity (ROE) of commercial banks.

Page | 24
6.3.3 Regression Result NIM

. xtreg nim ln_NoAgent ln_NoAccount ln_Deposit ln_Remittance ln_Asset ln_Age

Random-effects GLS regression Number of obs = 175

Group variable: Code Number of groups = 10

R-sq: within = 0.0292 Obs per group: min = 12

between = 0.0983 avg = 17.5
overall = 0.0422 max = 19

Wald chi2(6) = 5.86

corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.4391

nim Coef. Std. Err. z P>|z| [95% Conf. Interval]

ln_NoAgent -.0002722 .0004562 -0.60 0.551 -.0011662 .0006219

ln_NoAccount .0004589 .0004553 1.01 0.313 -.0004334 .0013513
ln_Deposit -.0006734 .0003936 -1.71 0.087 -.0014448 .0000981
ln_Remittance .0002705 .0003753 0.72 0.471 -.000465 .0010061
ln_Asset .0005826 .0016648 0.35 0.726 -.0026804 .0038455
ln_Age -.0015834 .0025856 -0.61 0.540 -.006651 .0034842
_cons .0034539 .0133651 0.26 0.796 -.0227412 .029649

sigma_u .00165826
sigma_e .00254277
rho .29839231 (fraction of variance due to u_i)

Table 9: Regression Result NIM

Table 9 shows a model summary of regression analysis between dependent variable net interest
margin (NIM) and independent variables including number of agents, number of Accounts,
volume of deposits, volume of remittance distribution, assets, age. The Coefficient of
determination R-sq overall value is 0.0422 means about 4.2 percent of variation of net interest
margin (NIM) that is clarify by independent variables number of agents, number of accounts,
volume of deposits, volume of remittance, assets, age. On the other hand, 94.8% is clarify by
other variables outside the model and error term. Number of observation is 175 and number of
groups are 10 which define this study. Regression result shows that number of agent banking
accounts, volume of remittance distribution and assets has a positive impact on the net interest
margin (NIM). On the other hand, number of agents, volume of deposit, and age has a negative
impact on the net interest margin (NIM) of commercial banks.

Page | 25
6.4 Other Analysis
6.4.1 Profitability trend
In this research study, return on Asset (ROA), Return on Equity (ROE) and net interest margin
(NIM) as profitability proxy. In table-10, have exposed average ROA, ROE and NIM sampled
10 commercial banks from December 31, 2016 to June 30, 2021. Table-10 shows that the ROA,
ROE and NIM do not have any specific trend rather it is changing from quarter to quarter.

Avg. ROA Avg. ROE Avg. NIM

December 31, 2016 0.2861 0.0429 0.0066
March 31, 2017 0.1863 0.0260 0.0060
June 30, 2017 0.1493 0.0229 0.0065
September 30, 2017 0.1608 0.0224 0.0062
December 31, 2017 0.2587 0.0419 0.0067
March 31, 2018 0.1301 0.0202 0.0060
June 30, 2018 0.1243 0.0196 0.0063
September 30, 2018 0.1355 0.0205 0.0064
December 31, 2018 0.2917 0.0487 0.0076
March 31, 2019 0.0951 0.0163 0.0066
June 30, 2019 0.1366 0.0223 0.0068
September 30, 2019 0.1406 0.0215 0.0060
December 31, 2019 0.2058 0.0352 0.0084
March 31, 2020 0.1366 0.0221 0.0059
June 30, 2020 0.1036 0.0173 0.0046
September 30, 2020 0.0742 0.0131 0.0036
December 31, 2020 0.2406 0.0413 0.0071
March 31, 2021 0.1938 0.0290 0.0051
June 30, 2021 0.1641 0.0269 0.0059
Table 10: Trend Analysis of ROA, ROE & NIM

The highest ROA was in the fourth quarter of 2016 and it was 0.2861. The highest ROE was
in the fourth quarter of 2018 and it was 0.0487. The highest NIM was in the fourth quarter of
2019 and it was 0.0084.

Page | 26
 0.0000
0.0500
0.0600
0.0000
0.1000
0.1500
0.2000
0.3000
0.3500

0.0100
0.0200
0.0300
0.0400
0.0500
0.2500
December 1, 2016 December 1, 2016
February 1, 2017 February 1, 2017
April 1, 2017 April 1, 2017
June 1, 2017 June 1, 2017
August 1, 2017 August 1, 2017
October 1, 2017 October 1, 2017
December 1, 2017 December 1, 2017
February 1, 2018 February 1, 2018
April 1, 2018 April 1, 2018
June 1, 2018 June 1, 2018
August 1, 2018 August 1, 2018
October 1, 2018 October 1, 2018

Avg. ROE
December 1, 2018 December 1, 2018
February 1, 2019 February 1, 2019
April 1, 2019 April 1, 2019
Avg. ROA Trend

June 1, 2019 June 1, 2019

Avg. ROE and NIM

August 1, 2019 August 1, 2019

Avg. NIM
October 1, 2019 October 1, 2019
December 1, 2019 December 1, 2019
February 1, 2020
Figure 9: Average ROA Trend

February 1, 2020
April 1, 2020 April 1, 2020

Figure 10: Average ROE & NIM trend

June 1, 2020 June 1, 2020

August 1, 2020 August 1, 2020

October 1, 2020 October 1, 2020

December 1, 2020 December 1, 2020

February 1, 2021 February 1, 2021

April 1, 2021 April 1, 2021

June 1, 2021
June 1, 2021

Page | 27
6.5 Findings
The coefficient and p-value of ln_NoAgent are -0.0000863 and 0.0743 for Return on Asset
(ROA), -0.0026676 and 0.491 for Return on Equity (ROE), -0.0002722 and 0.551 for NIM. It
indicates that the number of agents has a significant negative impact on the profitability of the
bank. The result show that number of agents has a significant negative impact on the
profitability that does not support research hypothesis. The coefficient and p-value of
ln_NoAccount are 0.0000998 and 0.0705 for Return on Asset (ROA), 0.0032985 and 0.396 for
Return on Equity (ROE), -0.0004589 and 0.313 for NIM. It indicates that the number of
accounts has a significant positive impact on the profitability of the bank that support research
hypothesis. The coefficient and p-value of ln_Deposit are -0.0003496 and 0.137 for Return on
Asset (ROA), -0.0043684 and 0.211 for Return on Equity (ROE), -0.0006734 and 0.087 for
NIM. It indicates that the volume of deposit has a significant negative impact on the
profitability of the bank that does not support with research hypothesis. The coefficient and p-
value of ln_Remittance are 0.0002612 and 0.232 for Return on Asset (ROA), 0.00265576 and
0.410 for Return on Equity (ROE), 0.0002705 and 0.471 for NIM. It indicates that the volume
of remittance distribution has a significant positive impact on the profitability of the bank that
support research hypothesis. The coefficient and p-value of ln_Asset are -0.00025 and 0.753
for Return on Asset (ROA), 0.0063994 and 0.568 for Return on Equity (ROE), 0.0005826 and
0.726 for NIM. It indicates that assets has a significant positive impact on the profitability of
the bank. The coefficient and p-value of ln_Age are -0.0005561 and 0.651 for Return on Asset
(ROA), -0.0205258 and 0.236 for Return on Equity (ROE), -0.0015834 and 0.540 for NIM. It
indicates that the age has a significant negative impact on the profitability of the bank.
The equation (ROA, ROE,NIM) of model look under:

𝑅𝑂𝐴ᵢ ̦ₜ = 0.0058704 − 0.0000863𝐴𝐺𝐸𝑁 + 0.0000998𝐴𝐶𝑇𝑁 − 0.0003496𝐷𝐸𝑃𝑂𝑆𝐼𝑇 + 0.0002612𝑅𝐸𝑀

− 0.0005561𝐴𝐺𝐸 + 𝜀ᵢ ̦ₜ
𝑅𝑂𝐸ᵢ ̦ₜ = 0.0104209− 0.0026676𝐴𝐺𝐸𝑁 + 0.0032985𝐴𝐶𝑇𝑁 − 0.0043684𝐷𝐸𝑃𝑂𝑆𝐼𝑇 + 0.0026557𝑅𝐸𝑀
− 0.0205258𝐴𝐺𝐸 + 𝜀ᵢ ̦ₜ

𝑁𝐼𝑀ᵢ ̦ₜ = 0.0034539 − 0.0002722𝐴𝐺𝐸𝑁 + 0.0004589𝐴𝐶𝑇𝑁 − 0.0006734𝐷𝐸𝑃𝑂𝑆𝐼𝑇 + 0.0002705𝑅𝐸𝑀

− 0.0015834𝐴𝐺𝐸 + 𝜀ᵢ ̦ₜ

Page | 28
 7 Conclusion

7.1 Conclusion
Agent banking is one of the most noticeable non-traditional banking channel that are providing
financial service to rural and unbanked people, outside bank premise. It has been getting
exceptionally famous in various nations across the world, particularly to the rural areas. The
principal objective of this study was to determine the impact of agent banking on profitability
of bank in Bangladesh. The research study concentration on number of agents, number of
accounts, volume of deposit and volume of remittance distribution as independent variables
and return on asset (ROA), return on equity (ROE) and net interest margin (NIM) as dependent
variables. Study has performed through secondary data that has gather from Bangladesh Bank
and data study period was December 31, 2016 to June 30, 2021. Data has explained through
summary statistic, correlation matrix and liner regression model. In Summary statistic find that
mean values are 0.0017 for return on asset, 0.0274 for return on equity, 0.0063 for net interest
margin, 4.6493 number of agents (In_NoAgent), 10.9825 number of accounts
(ln_NoAccounts), 4.6756 volume of deposits (ln_Deposit), 4.5608 volume of remittance
(ln_Remittance), 10.4497 Assets (ln_Asset), 3.1704 Age (ln_Age). Summary statistic specify
that the mean values of all dependent and independent variables are in the range between
minimum and maximum values. In Correlation matrix find that the highest correlation (0.14)
is between Return on Asset (ROA), Return on Equity (ROE), net interest margin (NIM) and
number of accounts. The weakest correlation (0.07) is between Return on Asset (ROA), Return
on Equity (ROE), net interest margin (NIM) and volume of deposits. In regression analysis find
that the coefficient and p-value of ln_NoAccount are 0.0000998 and 0.0705 for Return on Asset
(ROA), 0.0032985 and 0.396 for Return on Equity (ROE), -0.0004589 and 0.313 for NIM. It
indicates that the number of accounts has a significant positive impact on the profitability of
the bank that support research hypothesis. Analysis also find that the coefficient and p-value
of ln_Remittance are 0.0002612 and 0.232 for Return on Asset (ROA), 0.00265576 and 0.410
for Return on Equity (ROE), 0.0002705 and 0.471 for NIM. It indicates that the volume of
remittance distribution has a significant positive impact on the profitability of the bank that
support research hypothesis. From this research, I would recommend that management of
commercial bank would focus on expanding the agent banking networking for the profitability
of the bank as well as it will help to unbanked people to get banking service easily.

Page | 29
7.2 Policy Implication
The review results have some approach suggestions on the impact of agent banking on the bank
profitability in Bangladesh. The policy implication depends on regression analysis:

 Commercial banks ought to contribute more resources towards expanding their number of
accounts to increase the bank profitability particularly in rural regions. Banks ought to
likewise be careful in choosing agent for providing services as well as Bangladesh Bank
can organize more seminars, provide accurate guideline to relevant stakeholders on agent
banking services.
 Commercial banks ought to develop strong process for remittance distribution through
agent banking channel by ensuring that agent banking customers’ got proper services. Bank
management have to more concern remittance distribution and provide accurate policy,
guidelines and instruction to agent who providing remittance services.
 Commercial banks ought to contribute more resources towards expanding their number of
agent to increase the bank profitability particularly in rural regions. Banks ought to likewise
be careful in choosing agent for providing services as well as Bangladesh Bank can
organize more seminars, provide accurate guideline to relevant stakeholders on agent
banking services.

7.3 Scope of Future Research

 Evaluating customer's level of satisfactions through agent banking services in Bangladesh.
 Challenging factors for adopting the agent banking of commercial bank in Bangladesh.
 Impact of Agent Banking on the development of rural and unbanked people in Bangladesh.
 Influence of deposit acquisition to financial profitability of commercial bank in
Bangladesh.

Page | 30
 References

Caitlin, S. (2013). Do Agents Improve Financial Inclusion? Evidence from a National Survey

in Brazil. Bankable Frontier Associates.

Mwando, S. (2013). Contribution Of Agency Banking On Financial Profitability Of

Commercial Banks in Kenya. Journal of Economics and Sustainable Development,

4(20).

Kengere, D. (2014). The Effects of Agency Banking on the Non-Funded Income of

Commercial Banks in Kenya. Unpublished MBA Project. University of Nairobi.

Kitali, M. C., Chepkulei, B., & Shibairo, P. M. (2015, June). The impact of agency banking on

customer satisfaction: A survey on agent banks in Kenya. International Journal of

Economics, Commerce and Management, III(6), 691-714. Retrieved from

http://ijecm.co.uk

Amit, S. (2018). Prospects and Challenges of Agent Banking in Bangladesh: A Case Study of

Female RMG Workers. Review of Interative Business & Economics Research, 7(4), 75-

89. Retrieved from http://buscompress.com/journal-home.html

CGAP. (2006). Use of agents in branchless banking for the poor: Rewards, Risks, and

Regulation. Consultative group to assist the poor, Focus Note 38.

Chatterjee, S., & Price, B. (1977). Regression Analysis by Example. New York: John Wiley

and Sons, Inc., 19-22.

Kambua, D. (2014). The Effect of Agency Banking on Financial Profitability of Commercial

Banks in Kenya. Unpublished MBA Project. University of Nairobi.

Page | 31
Karangwa, F., & Mulyungi, D. (2018, April-September). Effect Of Agency Banking Services

On Financial Profitability Of Commercial Banks: A Case Of Equity Bank. 6(1), 758-

765. Retrieved from www.researchpublish.com

Kengere, D. (2014). The Effects of Agency Banking on the Non-Funded Income of

Commercial Banks in Kenya. Unpublished MBA Project. University of Nairobi.

Financial Inclusion Department, B. B. (2020). Agent Banking Activities in Bangladesh,

Quarterly Report , Jan-March 2020. Dhaka: Bangladesh Bank.

Gibson, E., Pasini, F., & Buckley, R. (2014). Regulating Digital Financial Services Agents in

Developing Countries to promote Financial inclusion. UNSW Australia.

Hasan, Z. (2019). The Effects of Agent Banking on the Profitability of Commercial Banks in

Bangladesh. Unpublished MBA thesis.

Kitali, M. C., Chepkulei, B., & Shibairo, P. M. (2015, June). The impact of agency banking on

customer satisfaction: A survey on agent banks in Kenya. International Journal of

Economics, Commerce and Management, III(6), 691-714. Retrieved from

http://ijecm.co.uk

Lyman, T., Ivatury, G., & Stachen, S. (2006). Use of agency banking for poor; Rewards, risk

and regulation. Focus Note 38.

Mukuria, P. N., & Memba, F. (2016, July). The Relationship Between Challenge Liquidity Of

Agency Banking And Profitability Of Agency Banks In Kisii County, Kenya. OSR

Journal of Business and Management (IOSR-JBM), 18(7), 01-08. Retrieved May 26,

2020, from www.iosrjournals.org

Onwei, B. (2015). Operationability of mobile banking in Nigeria: Evidence from Ebonyi State.

International Journal of Economics and Financial Reserach, 1 (6), 85-89.

Page | 32
Pyle, D., & Leland, H. (1977). Informational Asymmetries, Financial Structure, and Financial

Intermediation. The Journal of Finance, 32(2), 371-387.

Rahman, B. (2016). The Impact of Agency Banking to the Development of SME Sector: The

Case of Bangladesh. World Journal of Social Sciences, 6(3), 59-75.

Vutsengwa, R., & Ngugi, K. (2013). An assesment of the challenges facing commercial banks

in sustainability of agency banking in Kenya: A case of commercial banks.

International Journal of social science and entrepreneurship, 1 (2), 613-620.

Wambari, A. (2009). Mobile Banking in Developing Countries: ( A case study on Kenya).

University of applied science.

Yakub, J., Bello, H., & Adengua, I. (2013). Mobile Money Services in Nigeria: An inquiry of

existing models. International Journal of Economics and Management Sciences, 2(9),

94-105.

Page | 33
 Appendix

Appendix A: Construction of variables are:

roa=NetProfitAfterTaxesCrore/TotalAssetCrore
roe=NetProfitAfterTaxesCrore/TotalEquityCrore
nim=NetInterestIncomeCrore/TotalAssetCrore
ln_NoAgent=log(NumberofAgent)
ln_NoOutlet=log(NumberofOutlet)
ln_NoAccount=log(NumberofAccount)
ln_Deposit=log(DepositCrore)
ln_Loan=log(LoanCrore)
ln_Remittance=log(RemittanceCrore)
ln_Asset=log(TotalAssetCrore)
ln_Age=log(Age)

Page | 34