Thesis
Thesis
Thesis
Page
Table of Contents ii
List of Tables vii
List of Figures vii
List of Appendices vii
Abstract viii
Table of Contents
Introduction............................................................................................................................................................5
1.0 Background and Motivation...................................................................................................................6
1.2 The Research Objectives........................................................................................................................9
1.3 Contribution and Importance of the Study...........................................................................................11
1.4 Organization of the Thesis....................................................................................................................12
Chapter 2...............................................................................................................................................................15
2.0 Introduction..........................................................................................................................................15
2.1 Accountability and Reporting of MFIs.................................................................................................15
2.2 Accountability and Reporting of MFIs in Bangladesh.........................................................................20
2.3 Importance of Financial Statement in Discharging MFI’s Accountability..........................................23
2.4 Summary...............................................................................................................................................25
Chapter 3...............................................................................................................................................................26
3.0 Introduction..........................................................................................................................................26
1
3.1 The History of Microfinance in Bangladesh........................................................................................26
3.2 The Socio-economic Environment.......................................................................................................27
3.3 Microfinance Providers........................................................................................................................28
Table 3.1 Development of Microfinance Sector in Bangladesh (Tk. in million).......................................29
3.3.1 Direct Fund Providers..................................................................................................................29
ASA (The Association for Social Advancement).......................................................................................30
The Grameen Bank.....................................................................................................................................31
Palli Daridra Bimochon Foundation (PDBF).............................................................................................32
Rural Development Scheme (RDS) of Islami Bank (IBBL)......................................................................32
3.3.2 Apex Fund Provider....................................................................................................................32
Table 3.2 A Component-wise Disbursement Breakup of PKSF’s Loan...................................................33
3.4 Sources of MFI Funds..........................................................................................................................34
Table 3.3 Sources of Microfinance and Revolving Loan Funds of MFIs..................................................35
3.5 Outreach of MFIs:................................................................................................................................36
3.6 Microfinance and Different Types of Collateral..................................................................................37
Table 3.4 Credit Delivery and Savings Mechanisms of MFIs...................................................................38
3.7 The Social Impact of Microfinance in Bangladesh..............................................................................39
Table 3.5 Impact of Microfinance (compared to non-participants)............................................................41
Chapter 4...............................................................................................................................................................43
4.0 Introduction..........................................................................................................................................43
4.1 Regulatory Framework of Microfinance Institutions...........................................................................43
4.2 Legal and Financial Reporting Requirements of MFIs in Emerging Countries...................................46
4.3. Legal Provisions for Formation and Reporting of MFIs in Bangladesh.....................................50
The Societies Registration Act 1860 (Act XXI of 1860)...........................................................................50
The Trust Act 1882.....................................................................................................................................51
Voluntary Social Welfare Agencies (VSWA) (Registration and Control) Ordinance 1961......................51
The Companies Act 1994...........................................................................................................................51
Cooperative Societies Act 2001.................................................................................................................52
4.3.1 Regulatory Bodies.......................................................................................................................52
PKSF (Palli Karma-Sahayak Foundation)..................................................................................................53
The NGO Affairs Bureau (NGOAB)..........................................................................................................54
2
Microcredit Regulatory Authority..............................................................................................................54
Registration requirements:..........................................................................................................................55
License requirements for MFIs:.................................................................................................................55
Reporting Requirements for MFIs:.............................................................................................................55
4.3.2 Financial Reporting Environment of MFIs in Bangladesh..........................................................56
4.4 Summary...............................................................................................................................................57
Chapter 6...............................................................................................................................................................69
6.0 Introduction..........................................................................................................................................69
6.1 Hypotheses Development.....................................................................................................................69
6.2 Attributes of MFI Financial Statement.................................................................................................70
MFI’s Size..................................................................................................................................................72
MFI’s Debt level.........................................................................................................................................72
Rate Surplus of Revenue over Expenditure................................................................................................73
Size of Audit Firms.....................................................................................................................................73
Nature of Registration................................................................................................................................74
Hypothesized Relationships between Dependent and Independent variables............................................76
6.3 Hypotheses:..........................................................................................................................................76
6.4 Summary...............................................................................................................................................77
Chapter 7...............................................................................................................................................................79
7.0 Introduction..........................................................................................................................................79
7.1 Research Methodology.........................................................................................................................79
7.1.1 Questionnaire Survey Method.....................................................................................................79
7.1.2 Disclosure index Method.............................................................................................................81
3
7.1.3 Construction of disclosure index.................................................................................................84
7.1.4 Scoring for the Disclosure Index.................................................................................................85
7.2 Data Collection.....................................................................................................................................88
7.2.1 Survey sample..............................................................................................................................88
7.2.2 Annual report (Financial Statement) Source...............................................................................89
7.3 Data Analysis tool................................................................................................................................90
7.3.1 Descriptive Analysis....................................................................................................................90
7.3.2 Preliminary Tests on Regression Assumptions...........................................................................91
7.3.2.1 Multicollinearity..........................................................................................................................91
7.3.2.2 Homoscedasticity.........................................................................................................................91
7.3.2.3 Linearity.......................................................................................................................................92
7.4 Hypothesis testing.................................................................................................................................92
7.4.1 Hypothesis 1:...............................................................................................................................92
7.4.2 Hypothesis 2:...............................................................................................................................93
7.4.3 Hypothesis 3:...............................................................................................................................93
7.4.4 Hypothesis 4:...............................................................................................................................94
7.4.5 Hypothesis 5:...............................................................................................................................94
Regression Model.......................................................................................................................................97
7.4 Summary...............................................................................................................................................97
8.0 Introduction..........................................................................................................................................99
8.1 Descriptive Statistics and Correlation Analysis...................................................................................99
Table 8.1: Descriptive Statistics of the Dependent and Independent Variables.......................................100
Correlation Analysis.................................................................................................................................101
Table 8.2 Pearson Correlation Coefficients between Variables...............................................................101
8.2 Analysis and Interpretation of the Results..........................................................................................102
8.3 The Degree of Consensus between Users and Preparers of MFIs......................................................102
• H1: There is no consensus between users and preparers regarding the importance attached to the
information items disclosed in MFI financial statement....................................................................102
Demographic Profile of the Respondents (Users and Preparers).............................................................103
Table 8.3 Users and Preparers’ Background............................................................................................103
Table 8.4 Users’ Information Needs and Preparers’ Perceptions.............................................................104
Table 8.5 Importance of Information items by Users and Preparers........................................................105
Table 8.6 Users and Preparers’ Information Needs/ Perceptions (Sectional)..........................................107
4
8.4 Disclosure practice and Users’ information needs.............................................................................108
• H2: There is no significant gap between the information needs of users and actual disclosure in
MFI financial statement......................................................................................................................108
Table 8.7 Distribution of Overall Disclosure Indices According to Number of MFIs in 2006...........109
Table 8.8 Disclosure Practice and Users’ Information Needs..................................................................110
8.5 Disclosure Practice and Preparers’ Perceptions.................................................................................111
H3: There is no significant gap between the perceptions of preparers and actual disclosure in MFI
financial statement..............................................................................................................................111
Table 8.9 Disclosure Practice and Preparers’ Perceptions.......................................................................112
H4: There has been no improvement in the quality of MFI disclosure since 2004..................................112
Table 8.10 Distribution of Average Sectional and Total Disclosure Index in 2004, 2005 and 2006.......113
Table 8.11 T value Showing the Significance of Differences in Disclosure Indices...............................114
8.7 MFI Characteristics and Disclosure Quality......................................................................................114
H5: There is no association between the total disclosure quality and the MFI’s characteristics such size,
debts levels, surplus of revenue over expenditure, size of audit firms, nature of registration, MFI’s
age, governance structure, frequency of board meetings and qualification of board members.........114
Table 8.12: Regression Results of Disclosure Index and MFI Attributes................................................116
8.8 Further Robustness Test.....................................................................................................................119
Table 8.13 Results of Random Effect Regression Model........................................................................119
8.9 Summary.............................................................................................................................................119
References............................................................................................................................................................127
5
Introduction
In recent years, the world has witnessed a significant growth in Microfinance Institutions
(MFIs). These are specialised entities that provide a small amount of business credit
people who lack steady employment and a verifiable credit history. MFIs are also called Non-
similar nature of their operations and social objectives. In this thesis, NGOs, NFPOs and
MFIs are used interchangeably. In Bangladesh, MFIs are given not-for-profit status and are
not required to pay taxes on their income. They generate income by charging interests for
their services, however, their main purpose is still being social upliftment rather than profit
maximisation. MFIs are required to charge interests to operate their organizations efficiently,
Microcredit is not a new concept and has been practised at various times in modern history
(Cons & Paprocki, 2008). The provision of microcredit, initially operationalised by Dr.
Akhther Hamid Khan during the late sixties in Comilla, Bangladesh. Later through the
Muhammad Younus (a Nobel laureate for his contribution to poverty alleviation) popularised
the concept, not only in emerging countries but also in many developed countries, including
the U.S.A where Grameen Bank started their operation in New York in 1999.
6
opportunities and assisting them with training and education to sustain entrepreneurial
activities. It is estimated that the microfinance industry accounts for roughly USD. 15 billion-
17 billion in microfinance loans outstanding at the end of 2006, with a total of 3,552 MFIs of
which 1,727 are located in Asia, 935 are in Sub-Saharan Africa, 613 are in Latin America
and the Caribbean and the remainder 85 in the Middle East and North Africa (Harris-Daley,
2009). Because of significant growth, the United Nations declared 2005 as the ‘international
Year of Microfinance’. The Forbes (2009) ranked the world's top fifty microfinance
institutions based on scale, efficiency, portfolio risk and profitability, and found that most of
the MFI’s are located in Bangladesh, India and some Latin American countries.
Along with other emerging nations Bangladesh also witnessed exponential growth in MFIs in
terms of number, size, membership, and finance (Devine, 2006). Increasing availability of
donor funds and supportive policy environment under successive regimes since independence
of the country has further propelled the pace of their expansion (Lam, 2006). These
organizations are well known, more than anything else, for pioneering microcredit programs
as these now reach as many as 37 percent of all Bangladeshi households and around 60
percent of poor households (Siddiquee & Faroqi, 2009). For example, the World Bank alone
reached more than 6 million poor people in Bangladesh through microfinance projects worth
Bangladesh is credited with microfinance innovation and has taken a leading role in
improving the reporting standards of MFIs. For example, the Consultative Group to Assist
the Poor (CGAP) of the World Bank in association with Palli Karma-Sahayak Foundation
7
(PKSF)1, has issued a set of disclosure guidelines on MFI financial reporting requirements
disclosure guidelines were developed in consultation with microfinance practitioners and the
donor members of CGAP (the Consultative Group to Assist the Poor). CGAP is a consortium
of bilateral foreign aid agencies from 16 countries, 12 multi-lateral agencies, and 2 private
foundations. This final version (July 2003) is based on the results of field testing and input
from the Financial Services Working Group (FSWG) of the Small Enterprise and Education
‘Microcredit Regulatory Authority Act 2006’ and established the Microcredit Regulatory
the country and this reporting regime is being adopted by many other emerging countries
where microcredit is getting momentum (MIX Market Report, 2009). However, the process
mostly under-regulated and users and regulators get limited amount of information from
MFIs in Bangladesh despite the existence of legislation and regulatory bodies (Tazi, 2006).
With the significant growth in microfinance, the problem of accountability and governance of
these institutions have become very important issues and pose several challenges. These
challenges also include financial reporting of their performance not only to local depositors
(members) but also to regulators and international funding agencies such as the World Bank,
1
Palli Karma-Sahayak Foundation (PKSF) means Rural (Palli) Employment (Karma) Support
(Sahayak) Foundation (Foundation) in English.
2
These disclosure guidelines were developed in consultation with microfinance practitioners and the member
donors of CGAP (the Consultative Group to Assist the Poor). CGAP is a consortium of bilateral foreign aid
agencies from 16 countries, 12 multi-lateral agencies, and 2 private foundations. This final version (July 2003)
is based on the results of field testing and input from the Financial Services Working Group (FSWG) of the
Small Enterprise and Education and Promotion Network (SEEP).
Richard Rosenberg, Patricia Mwangi, Robert Peck Christen, and Mohamed Nasr were the principal drafters of
this document.
8
and the United Nations Capital Development Fund (UNCDF) (Ebrahim, 2003a&b; Khan,
2003; and Bala & Mir, 2006). However, to date academic research with respect to MFIs’
empowerment of women (Hashemi et al., 1996; Khandaker, 1998 & 2003; Pitt, 1999;
Murdoch, 1999; BIDS, 2001; Khandaker & Cartwright, 2003). Some have examined the role
of their governance structure in outreach to the poor and their financial performance
Despite the recognition that quality financial reporting is necessary to improve MFIs
accountability to stakeholders, there has been no substantive empirical study to date to assess
how these organizations report to various parties using a general purpose financial reporting
framework. This study, in this regard, is the first substantive attempt to redress this gap in
our knowledge in the financial reporting quality of MFIs by using a large sample 435 firm-
year reports over a period of three years in Bangladesh, a country which is credited with
microfinance innovation and taken a leading role to improve reporting standards of MFIs.
In this study, three important disclosure issues have been focused: (1) a cross-sectional study
of the level of information disclosure by MFIs, (2) how MFIs governance, auditors and
internal factors are associated with the variation in disclosure and (3) users’ information
needs and preparers’ perception about disclosed items in MFI financial statement. The
82 items appropriate for MFIs based on the recommendations of the World Bank, IASB, and
9
government and non-government bodies engaged in monitoring reporting performance of
needs of users and perceptions of preparers about the disclosed information and compared it
to the current disclosure practice of MFIs to identify if there is any expectation gap. Data
were collected by two sources: questionnaire survey and archival records such as annual
report (financial Statement). The perceptions of users and preparers about disclosed
information items in financial statement were collected through mail survey and annual
(1) What are the institutional and legal disclosure requirements or guidelines affecting
(2) What are the current accounting and disclosure practices of microfinance institutions
(3) Is there any consensus between the preparers and users with respect to their perceived
information needs?
(4) What is the expectation gap between the level of disclosure and the users’ information
needs?
(5) What is the expectation gap between the level of disclosure and the preparers’
(6) Has there been any improvement in disclosure quality over three years since 2004?
(7) Is there any association between the quality or extent of actual disclosure and some
selected organization characteristics such as size, debts level, rate surplus of revenue
10
governance structure, frequency of board meetings and qualification of board
members.
This study will contribute significantly to our understanding of disclosure practices of MFIs
standards. The study may also help the other emerging countries which have similar
The results of this study will assist accounting standard setters and regulators by determining
whether there is a gap between the expectations of users and MFI disclosure practices and if
so, the extent of the gap. They will also be benefited by knowing the perceptions of preparers
about disclosed items in MFI financial statement. The study will help them to understand the
degree of MFI disclosure compliance with the provisions of the existing guidelines and the
practice will enhance our knowledge by identifying the significant factors explaining the
variability of disclosure practice of MFIs in Bangladesh. This information will also assist the
11
1.4 Organization of the Thesis
The reminder of the thesis is organized as follows: chapter two presents the nature and
pattern of accountability of microfinance institutions. The chapter also focuses on how these
Section one focuses on the accountability and reporting of MFIs along with the theories
underlined the framework. Second section outlines the accountability and reporting
statement of MFIs in discharging accountability along with the perceived information needs
organizations (as mentioned earlier MFIs are one form of NGOs) in Bangladesh and beyond.
Section one to six outlines the microfinance activities in Bangladesh. Section seven discusses
the social impact of microfinance activities in Bangladesh. Section eight outlines the
expansion of microcredit organizations in the world along with the some criticisms of
microfinance activities. The last section discusses the international bodies which are involved
Chapter four outlines the regulatory framework of microfinance institutions. Section one
presents the regulatory framework of microfinance institutions along with the legal and
financial reporting requirements of MFIs in emerging countries. The last section of the
chapter highlights the major provisions of the various Acts and institutional guidelines
12
Chapter five presents previous studies on the accountability, reporting and disclosure,
corporations. Sections one to five outline the previous studies on reporting and disclosure,
governance and regulations of micro finance and not-for-profit sector. Section six discusses
about the disclosure and governance studies of profit making organizations. The last section
In chapter six, the hypotheses are presented for empirical investigation in this study. Total
five hypotheses have been developed for testing: (1) There is no consensus between users and
preparers regarding the importance attached to the information items disclosed in MFI
financial statements (2)There is no significant gap between the information needs of users
and actual disclosure in MFI financial statements (3) There is no significant gap between the
information needs of preparers and actual disclosure in MFI financial statement (4) There has
been no improvement in the overall disclosure practice since 2004 and (5) There is no
association between the total disclosure quality and some selected organization
characteristics such as size, debts level, surplus of revenue over expenditure, size of audit
Chapter seven outlines the data collection methods and sources followed by the research
Chapter eight presents the findings of this study. This chapter is organized into four sections:
Section one presents descriptive statistics and correlation analysis. Section two analyses and
13
interprets the results of univariate test and multiple regression models followed by discussion
The final chapter summarises and concludes this thesis. Additionally, the policies on relevant
14
Chapter 2
2.0 Introduction
The purpose of this chapter is to present the nature and pattern of accountability of
microfinance institutions along with how these organizations are discharging their
accountability via disclosure in financial statement. Section one focuses on the accountability
and reporting of MFIs along with the theories underlined the framework. The second section
outlines the accountability and reporting framework of MFIs in Bangladesh. The third section
presents the importance of MFI financial statement along with the perceived information
recognized authority or authorities and are held responsible for their actions’ (Edwards &
Hulme, 1996). In the context of MFIs, it connotes an obligation on the part of MFIs to
provide accounts to a set of legitimate authorities and those mechanisms and strategies
through which MFI organizations are kept responsible for their actions. However, Lewis
(2001) and Clerk (1991) argue that the accountability framework is rarely so simple for MFIs
because they have multiple stakeholders and are accountable in different and complex ways
to a variety of different groups and interests. Edwards and Hulmes (1996) and Lewis (2001)
also argue that MFIs face demands for two principle types of accountability. They face the
functional accountability such as accounting for resources, resource use and its impact. They
15
also have to meet strategic accountability such as accounting for impacts that MFI actions
have more widely and on other organizations and society. However, reconciling functional
and social accountability may produce tensions, which might affect their upward
accountability with funders and regulators (Ebrahim, 2003a). This makes accountability a
Leat (1988) and Lewis (2001) also outline another three levels of NGO accountability, which
are: (a) full accountability, (b) explanatory accountability and (c) responsive accountability.
In the case of full accountability, there is a right to demand an account from an NGO and
then impose a penalty if is not forthcoming, such as withdrawal of funding. This type of
accountability has become a dominant feature of the NGO world because it has been
associated with the growing concerns of fund providers to ensure that NGOs use resources
efficiently. In the case of explanatory accountability, an NGO needs to respond to a call for
account with information and explanation. Sanctions for failure to meet this accountability
come in the way of disapproval but not as a penalty. In the case of responsive accountability,
there is no formal sanction at all but the system runs on trust and good faith.
MFI accountability has also been classified in many ways- micro level and macro-level,
formal and informal, short term and long-term, functional and strategic. Micro-level
accountability is concerned with the availability, reliability, cost and quality of services
provided while macro-level accountability deals with how public expenditures decisions are
taken, controlled and monitored through accounting systems, external audit and review
procedures (Robinson, 1994). Formal accountability takes into cognizance whether agreed
objectives in a program have been met while informal accountability includes ongoing
discussions between partners (Edwards & Hulme, 1996). Short term and long term
16
accountability are not merely determined by the time dimension but also by the nature of
organizational functioning and goals. In all these frameworks, the role of financial reporting
Although microfinance organizations do not have residual claims and the separation of
ownership and managers are not severe because of funding arrangements and organizational
structure, they nonetheless face agency conflicts (Behn et al., 2007). For example, donors
would like their funds to be used wisely and efficiently on the programs and missions, but the
managers could spend the fund on bigger bonuses, better offices, conferences and travel, or
agent relationships whereby the principles e.g. donors, funders provide resources to agents
17
(MFI) to deliver specialized services and manage resources on their behalf. Such a separation
of ownership and control requires that a set of checks and balances are put in place that can
ensure that the agents discharge their roles and functions faithfully. It also ensures that agents
manage funds in a mechanism like auditing, reporting, monitoring and evaluation seeking to
keep close tabs on not only whether the fund have spent but also on how these have been
spent. This allows funders to assert both “financial control” (seeking accountability of
Within this framework, Najam (1996) argues, like publicly available financial reports, not-
for-profit financial reports play an important informational role by mitigating the inherent
Some researchers use agency theory to define and measure accountability for NGOs on the
basis of contractual relationships between principal and agent (Laughlin, 1996; Yetman &
Yetman, 2004; Desai & Yetman, 2004). The focus of the theory is the contract governing the
relationship between principal and agent and determining how the contract can be made as
efficient as possible (Eisenhardt, 1989). Using this principal-agent model, Laughlin (1996)
where expectations are formally recorded, which Roberts (1991) calls the relationship
relationship is also known as “Hard accountability” (Leat, 1988; Stewart, 1984; Laughlin,
1996) and information disclosure about how resources being used via financial reporting
Ebrahim (2003b) conceptualize accountability for NGOs in a broader and more inclusive
manner through the stakeholder framework. He identifies three primary groups which have
18
explicit or implicit contacts with NGOs: (1) Funders (2) Sectors Regulators and (3) Clients
requires the support of all its stakeholders, and that in order to engender this support, it needs
to be accountable to all its stakeholder groups. This framework supports the traditional
from funds providers (Dhanani, 2009). This decision usefulness framework for reporting has
also been accepted as one way of discharging accountability not only in the private sector but
participations, self-regulation and social audits. While the first two refer to accountability that
are prepared regularly, participation and self-regulation are generally more and multi-faceted
than tools emphasising certain courses of action rather than a distinct end result. Social
auditing overlaps the border of tools and process. Disclosure of statements and reports are
among the most widely used tools of accountability and frequently required by funding
bodies and regulatory authority. Such reports and legal disclosures are significant tools of
accountability in that they make available basic data on organizational operation. However,
the bulk of this reporting emphasizes upward reporting of financial data, with only limited
indication of the quality of MFIs work and almost no attention to downward accountability to
stakeholders.
Notwithstanding all this, academics remain doubtful about MFI’s ability to maintain
appropriate and effective level of accountability to a range of stakeholders. But all of them
19
mentioned about one type of accountability-conventional accountability for utilization of
resources (Edward & Hulme, 1996 and O’Dwyer & Unerman, 2005 (functional
broader understanding of the term, for it gives rise to the questions such as, why is reporting
necessary, who are legitimate authorities, what actions can be reported to and how? As a
result, like corporate financial report, MFI financial statements also play a very important
role for discharging their accountability. Various stakeholders especially funders and
regulators can also make more informed investing and regulating decisions by utilising
information in the MFI financial statement about the financial condition and performance of
The MFI accountability and reporting in Bangladesh is centred round three groups of
stakeholders- the funders/donors, the regulators and the beneficiaries. However, the
towards the demands of the two influential groups- the donors and the regulators. Khan
(2003) and Mir and Bala (2006) also find that NGOs in Bangladesh are more interested to
discharge their upward functional accountability to their funders and regulators The MFIs
receiving funding are subject to the tight accountability and reporting requirements of the
respective donors. The Palli Karma Shayak Foundation (PKSF) sets some pre conditions for
funds are subject to the Foreign Donations Regulation Ordinance, 1978 and Rules under the
20
ordinance, administered by the NGO Affairs Bureau. The Government’s regulatory
framework is mainly concerned with ensuring that MFIs activities are lawful and do not
conflict with government policies. On the other hand, the two apex bodies-the Association of
(FNB) set some self-regulatory code of ethics for their member NGOs. The code intends to
The Rules under the some of the Acts have specified the books of accounts to be kept by
MFIs. Annual accounts are required and audit is done by a qualified chartered accountant
from a list approved under the NGO Bureau (for foreign donations). The Ordinance
empowers the Government to inspect NGO books and to carry out its own audits if
necessary. Annual reports must also be submitted to the NGO Bureau. Although NGOs are
required to submit annual reports, the NGO Bureau is rarely concerned about the real
outcome of the project. On the other hand, donors demand project-based accounting but they
are more interested in knowing how their funds have been utilized (World Bank, 2006). Only
exception is the PKSF (Apex finance body). PKSF set some standards and reporting
guidelines for its partner NGOs. It strictly follows these guidelines before disbursing any
fund. All partner organizations are required to submit annual report and audited financial
Principle-Agent Framework. Ebrahim (2003b) identifies three primary groups which have
explicit or implicit contacts with NGOs: (1) Funders, (2) Sector Regulators and (3) clients
and communities.
21
Financial Reporting Framework for MFIs in Bangladesh
FUNDERS
International Donors
Government
PKSF
World Bank
MFIs
Figure 2.1
In Bangladesh, funders mainly PKSF, government and other donor agencies including World
Bank provide money to MFIs in exchange for regular reports and evaluations that confirm the
legitimate use of those funds. These reports and evaluations (which include financial
NGOs accountable to their funders. In other words, the accountability mechanisms position
NGOs are linked to sector regulators by accountability mechanisms that position them as
both agents of the public and of governments. Sector regulators (PKSF, Central Bank,
22
CGAP, different Authorities) impose some of the regulations and reporting requirements on
MFIs to follow. Laws governing not-for-profit status and their attendant requirements for
The third set of relationships involves clients and communities or beneficiaries. The service
providing MFIs has the powerful option of threatening to withdraw from current or future
funds in the event of non cooperation or non performance. At the same time, members
(depositors) are also concerned about their fund and demand financial reports from MFIs to
evaluate about their performances. Under such conditions, MFIs may be viewed both as
In all these contractual relations, financial report can assist these groups to monitor their
contracts. However, the literature suggests that, in most of the cases, Bangladeshi MFIs are
discharging their upward functional accountability to their funders and regulators through
the most important sources due to the availability of other sources, but it is not so for
such as analysts and management forecasts, and financial news papers for such entities.
statements remains the only sources though fund providers get some amount of information
23
The Statement of Financial Accounting Standards (SFAS No. 117) in USA states that the
to meet the common interest of donors, members, creditors and others who provide resources
to not-for-profit organizations. Like corporation, the external parties use MFI financial
statements in assessing (a) an organization’s services and its ability to provide those services
and (b) how managers discharge their stewardship responsibilities and other aspects of their
performance (FASB 1993). So the MFI annual report or financial statement should disclose
SFAS 117 identified donor/funder as the main user of non profit financial statement. PKSF is
the main fund provider for MFIs in Bangladesh as well as regulator However, in this study
auditors are treated as the main users as they are the most important parties to use the MFIs’
financial statements more frequently. They prepare management reports based on these
MFIs’ financial statements, which later on, utilized by PKSF. They also work as agents for
fund providers and regulators. PKSF appoint auditors every year to audit the MFIs’ financial
Auditors examine different aspects of MFI’s performances i.e., debt collection, liquidity,
debt-coverage and rate of return by calculating some of the ratios (Prescribed by PKSF)
based on the financial statements. In this regard, their perceptions about disclosed
auditors can also help to identify disclosure issues and related norms and opportunities. On
the other hand, views of MFI accountants as preparers can also be considered important for
the purpose of determining the adequacy of disclosures in MFI financial statements. MFI
accountants prepare financial statements based on some guidelines (CGAP 2003, PKSF and
Bangladesh bank) and some International Accounting Standards. They are well aware about
24
the disclosure need of which items are more important than others. The regulatory bodies
may also take into consideration the preparers’ perceptions for standard setting or
formulating regulations.
In the light of the above framework, this thesis will seek to understand and explain disclosure
pattern of informational items and the perceived gaps between users and preparers.
2.4 Summary
The chapter outlines the nature and pattern of accountability and reporting of microfinance
institutions along with how these organizations discharge their accountability via disclosure
in financial statements. The main finding of this chapter is that MFIs accountability and
reporting framework is skewed to two influential groups –funders and regulators. The
regulators are PKSF, central bank, government and CGAP. MFIs discharge their upward
functional accountability through financial reports in most of the cases. In Bangladesh, the
main fund provider, PKSF, also uses MFI financial statements to evaluate its member MFIs’
performances. The chapter also discusses the importance of users and preparers’ perceptions
25
Chapter 3
3.0 Introduction
The purpose of this chapter is to discuss the microfinance activities in Bangladesh and other
emerging countries. Section one to six outlines the microfinance activities in Bangladesh.
Section seven and eight discuss the social impact of microfinance activities in Bangladesh
along with some of the criticisms of microfinance institutions. Section nine outlines
expansion of microcredit in other emerging countries. The last section discusses the
It has been claimed that Bangladesh is the leader for microfinance activities in the world.
Though microcredit is not a new concept, Dr. Muhammad Yunus is credited with
economics at the University of Chittagong, Dr. Yunus initiated a research program to design
a framework that would bring the poor within a viable banking network. (Gibbons,1999).
Through the Grameen Bank Project, he demonstrated that poor, asset-less and landless people
were creditworthy and that lending to the poor could be an economically viable activity. In
1983, the Grameen Bank Project was transformed into an independent commercial bank,
known as the Grameen Bank. In 2007, cumulative loan disbursement of Grameen Bank was
Tk. 356.89 (USD.5.10) billion, occupying 31.69 percent of the market share of the sector
26
Following the success of the Grameen Bank, a number of existing Bangladeshi non-
Committee (BRAC) and the Association for Social Advancement (ASA), developed their
own microfinance programs based on the Grameen model. According to MRA statistics
(2008), there are more than 4,000 MFIs in Bangladesh with outstanding loan of more than
Tk.115 (USD.1.65) billion in 2007. However, there are only 402 MFIs who procured the
license from Microcredit Regulatory Authority (MRA) and are required to prepare the
Bangladesh is one of the most densely populated and poorest countries in the world. In 2008,
the population of Bangladesh was over 140 million people and the population density was
1083 people per square kilometre. The average income per capita was approx. USD. 599 per
year and close to 40 percent (nearly 56 million) of the population lived below the poverty
line. The number of hard core poor was 35 million. The adult literacy rate was 56 percent and
The above mentioned socio-economic factors that have contributed to the problem of poverty
have also encouraged microfinance to thrive in Bangladesh. Poverty creates demand for
access to credit remains severely restricted for the asset-less poor (Squillance, 2004). Despite
the immense poverty in Bangladesh, the poor have managed to find many profitable income
tea stalls, embroidery works, handicrafts etc. These businesses are generally easy to start and
manage, provide services that are in always in demand (even by the poor) and generate a
27
profit in a short amount of time. These factors ensure the success of the business, full
repayment of the loan and the tendency of the borrower to renew or increase the size of
his/her next loan. High population density increases the demand for these business services,
which also contribute to the creation of a large microfinance market in Bangladesh. (Geeta et
al.,1998).
Institutionally microfinance is provided at two stages: direct providers and apex lenders. At
present, the main direct providers are microfinance institutions (MFIs), Grameen Bank (GB),
Palli Daridra Bimochon Foundation (PDBF), Rural Development Scheme (RDS) of Islami
Bank (IBBL), among other. At the apex level the main fund provider is Palli Karma Shayak
Foundation (PKSF).
As per Credit Development Forum (CDF) microfinance survey 2007 (table 3.1), cumulative
disbursement of loans to the individual borrowers stood at Tk. 891.04 (USD. 12.70) billion
up to December 2006, while that reached Tk. 1,126.01 (USD. 16.10) billion up to December
2007. This shows that annual loan disbursement was 234.97 (USD.3.34) billion, an increase
in loan disbursement by 26.37 percent in 2007 from 2006. This growth is underestimated as
Among the direct service providing agencies, MFIs as a group was dominating with
which was 62.36 percent and Tk.732.32 (USD. 10.46) billion up to December 2007, which
was 65.03 percent of the sector. The annual disbursement was Tk.176.65 (USD. 2.52) billion
of the reporting 535 MFIs in 2007. These reflect tremendous expansion of market outreach of
28
the MFI’s program. The sector was very much concentrated to a very few institutions, as only
8 percent institutions occupy over 80 percent of the market. The major share of the market
was captured by two very large institutions namely BRAC and ASA, who have provided
63.23 percent of total MFI loan outstanding and have collected 54 percent of total savings
Borrowers
BRAC is the largest national MFI in Bangladesh. It is also the largest development
organization in the private sector in the country. BRAC started its first small loan programs
29
in the mid-1970s, lending to groups of artisans and other poor village to finance income-
generating activities. It initiated its major credit activity in 1979, and has been involved in
microfinance ever since. Despite having one of the largest microfinance operations in the
world, BRAC is not exclusively a microfinance providing institution (MFI). It has equally
large programs in health, education, employment generation, human rights and development
of other capacities, which are part of its holistic development approach. It is now one of the
BRAC’s microfinance activities are operated through its Rural Development Program or
RDP. The Nucleus of RDP is the village organization (VO). A VO comprises of 40-45 rural
poor as members. RDP lays importance on enterprise development of the borrowers, which is
supported by credit, training, input and extension support and often marketing assistance. At
the end of 2007, it had 7.05 million borrowers and a loan disbursement of Tk. 43.24
ASA was established as an NGO in 1978, with a focus on consciousness raising, group
development and training among the rural poor. In 1991, it started its microfinance
operations and recreated itself as a finance-only MFI. It is now the third largest MFI in
Bangladesh and offers a range of savings, credit and insurance facilities. At the end of 2007,
ASA had over 5.6 million active borrowers and a loan disbursement of Tk. 29.12 (USD.
30
The Grameen Bank
The Grameen bank is a unique financial institution. It was created by a special Act of
Parliament, the Grameen Bank Ordinance 1983, as a commercial bank with the social
mission of providing credit, with or without collateral security, to assetless and landless
borrowers for any type of economic activity. The Grameen Bank was started with funds from
Bangladesh Bank, capital markets and concessional loans from international donors.
As a single loan provider, Grameen Bank, is the largest. In 2007, cumulative disbursement
of Grameen Bank was Tk. 356.89 (USD. 5.10) billion, occupying 31.69 percent of the market
share of the sector. The share declined from 34.38 percent in 2006 to 31.69 percent in 2007.
The annual loans disbursement was Tk. 50.43 (USD. 0.72) billion, which is around 22
percent of micro loans disbursement in 2007 (CDF & Bangladesh Microfinance Statistics
2007).
In 2001-2, all Grameen Bank branches began to operate the new, simpler and much more
flexible ‘Grameen Generalised System’ (also called ‘Grameen II’), which offers four types of
loan products : (a) basic, (b) housing, (c) higher education and (d) struggling members
(beggars) loans. There is also a facility for larger small enterprise loans, and a range of
companies (commercial and not-for-profit) in the ‘Grameen Family’. This includes Grameen
Shikka (GS), established in 1997 to promote the education of non-literate Grameen Bank
members. GS provides financial support in the form of loans and grants for education; and
use and promote new and innovative ideas and technologies for educational development.
31
Palli Daridra Bimochon Foundation (PDBF)
PDBF is a transformed public sector MFI in Bangladesh. This is the only public sector MFI
body, the Foundation started functioning in 2000. PDBF’s Cumulative loan disbursement
increased to Tk.22.92 (USD. 0.32) billion in 2007 from Tk.19.92 (USD.0.28) billion in 2006.
The IBBL replicates Grameen model and directly implements RDS at the field level. The
IBBL is the only commercial bank that implements group-based microfinance program
directly. The cumulative loan disbursement of RDS (IBBL), still a small niche of the market,
PKSF, the single largest apex body of microfinance in the country, established in 1990 to
finance MFIs in order to broaden credit supply to poor households, provided Tk.50 (USD.
0.72) billion through its partner MFIs up to December 2007, while up to 2006 the figure was
Tk. 35 (USD. 0.50) billion. The annual loan disbursement in 2007 was Tk.15 (USD. 0.22)
billion.
The legal structure of PKSF allows flexibility, authority, and power to take programs and
implement them throughout the country while managing its affairs as an independent
organization. PKSF’s mandate authorizes its management to mobilize funds in the forms of
grants, loans at concession rates, and contributions from a wide variety of sources that
32
governments, international donors and lending agencies. PKSF receives funds from the
Union (EU), World Bank, USAID and the Asian Development Bank.
PKSF had 225 Partner organizations (POs) at the end of 2007 (PKSF, 2008). It charges
differential service charge for its two categories of POs: 7 percent for the big POs and 4.5
percent for the small and medium POs. It also operates a loan program for capacity
and has a well-developed training strategy including outsourcing to private and public sector
institutes. On-site technical assistance is also offered during the intensive schedule of field
visits undertaken by PKSF personnel. Under the diversified mainstream credit program,
PKSF provided five categories of loans: (a) rural microcredit (b) urban microcredit (c) micro
enterprise (d) ultra poor program and (e) seasonal loan. A component-wise disbursement
breakup (table 3.2) showed that PKSF provided more loan in rural microcredit followed by
PKSF continued to provide loans to the rural poor through its POs under five special
33
Development Bank b) Integrated Food Assisted Development Project (IFADEP) funded by
Program (SRLP); e) Financial Services for the poorest (FSP) and f) Micro Finance Technical
Support (MFTS) funded by the International Fund for Agricultural Development (IFAD).
Initially foreign donation was the major sources of fund for the MFIs, contribution of which
stood to near 50 percent of the total fund before 1996. But after that, it had declined sharply
and became only 10 percent of the total fund in 2007 (CDF and micro finance statistics
2007). One of the reasons for this sharp decline is that international donors now prefer to
provide donation or loan to apex funding body PKSF instead of MFIs directly. At present, the
sources of funds of MFIs are varying as a result of differences in their access to various
sources of fund for covering operating expenses and for providing micro loans. According to
Credit Development Forum (2007), sources of funding for the revolving loan fund (RLF) of
the MFIs is surveyed, came from the internal sources 52 percent and external sources 48
percent.
Most MFIs had traditionally received heavy support from foreign donors. However, as
international donor funds are drying up and commercial sources of funds remain limited,
MFIs have been increasingly relying on funds from savings deposits and approaching Palli
Karma Sohayak Foundation (PKSF) for financial and technical support. As MFIs are legally
restricted by the Government from collecting savings deposits from the general public (with
the exception of Grameen Bank), several larger, more stable MFIs such as BRAC and ASA
have augmented their funding resources with members’ savings deposits. According to CDF
34
and Bangladesh Microfinance Statistics (2006 & 2007), member’s savings have exhibited an
overall growth of 24 percent in 2006 over 2005 and 15 percent in 2007 from 2005. In many
compulsory savings (as little as Tk. 1-10) on a weekly basis. The regularity of savings
deposits and attendance of group meetings are additional preconditions for receiving credit.
Sources:
Donors:
NGOs
Savings
35
Historically, MFIs do not allow members to withdraw any amount of their forced savings
except when members decide to permanently leave the organization, since most MFIs have
used the savings fund as an important source of their fund (Dewan,1999). The table 3.3
shows that the amount of revolving fund was Tk. 55.09 (USD. 0.79) billion in 2007 and in
2006 the fund was Tk. 46.09 (USD. 0.66) billion. The annual increase in the fund was Tk. 9
(USD. 0.13) billion in 2007 and the growth rate was 19.53 percent. Members’ savings was
the dominating source contributing to around 52 percent in 2007, although it declined from
61 percent in 2006. Floods and price hike in 2007 may have contributed to the decline in
members’ net savings. Grants and donations from international agencies have contributed
around 10 percent. It remained constant over time. PKSF has contributed substantially to
financing micro credit. Its contribution has increased from 23 percent to 25 percent in 2007.
Local MFIs, the big ones, also provide loans to small MFIs. In 2005, the amount of
contribution was Tk.0.19 (USD. 2.52 million) billion, it increased by more than five times to
The outreach of the MFIs may be seen through its institutional and financial strength such as
number of branches of the agencies, loan coverage, net savings and loans outstanding. The
total number of branches of the direct loan providing institutions was 14,937 as of December
2007 while that was 11,723 in 2006. The annual growth rate of branches in the sector was
27.42 percent in 2007. The MFIs together had the highest number of braches (12,096),
The total number of members mobilized by MFIs was 33,137,964 as at December 2007, it
was 29,008,795 in 2006. The MFIs occupied 72.29 percent of the sector’s total members in
36
2007, and in 2006, they together retained 70.71 percent of the total numbers. The Grameen
Bank’s share as a single agency was still very significant at 22.36 percent on 2007 and 23.82
percent in 2006.
Net savings mobilization by the MFIs at the end of 2007 was Tk.62.5 (USD. 0.87) billion,
higher by around 17 percent over 2006. The share of Grameen Bank was over 45 percent.
With the increase in member savings, loans outstanding, a reflection of higher loans
was Tk.138.6 (USD. 1.98) billion. These reflect tremendous expansion of market outreach of
the MFIs’ program. The pattern of increase in the net savings and loans outstanding suggests
that the MFIs greatly emphasize mobilization of savings as a process of financing loan and
reducing dependency on external fund, as they use it in their revolving loan funds.
Many microfinance programs rely on group-based lending and other methods. Group-based
lending provides an incentive for borrowers to repay and help in avoiding adverse selection
of borrowers, thereby improving loan recovery rates. While group-based lending does not
always improve loan recovery, it constitutes a powerful incentive for repayment when it
helps to create “social collateral” that works against loan default (Miah, 2004). In Bangladesh
most microfinance programs have separate programs for men and women, in accordance with
socio-cultural norms. Membership is strictly limited to people who own less than half an acre
(approx. 2000 sqm) of land, are not members of the same household as another program
member, have similar economic resources and live in same village. The different methods of
credit delivery and savings mechanisms of four major MFIs are discussed in the following
37
Table 3.4 Credit Delivery and Savings Mechanisms of MFIs
5 members form a group 0.5% of each loan used for group insurance
BRAC Maximum land holding of half an acre of BD Tk. 2.00 per week
30-40 members form a village 0.1% of each loan used for group insurance
ASA Maximum land holding half an acre of Weekly at a minimum rate of BD Tk. 10-25
increase
(US$184)
38
3.7 The Social Impact of Microfinance in Bangladesh
The literature broadly supports the hypothesis that access to microcredit contributes to
poverty reduction and social development in Bangladesh. Microfinance has had a positive
Hashemi et al. (1996) examines the relationship between microcredit programs and women’s
empowerment. 1,225 married women under the age of fifty-members of Grameen Bank,
BRAC, non-members in Grameen served villages, and a comparison group- have been
observed and interviewed over a period of three years. Eight indicators have been used as
proxies for empowerment: (1) mobility, (2) economic security, (3) ability to make small
purchases, (4) ability to make larger purchases, (5) involvement in major household
decisions, (6) relative freedom from domination from within the family, (7) political and
legal awareness and (8) participation in public protests and political campaign. Each of the
well as on the woman’s contribution to family support and to the composite empowerment
score. Grameen Bank alone significantly affects women’s involvement in major decisions (5)
and BRAC alone significantly affects mobility (1). Even those members with little or no
control over their loans and income generating activities are shown to be more empowered
than non-members.
Khandaker (1998), using 92-98 household data, estimates that for every Tk. 100 (about
USD. 1.33) lent to a woman, household consumption increases by Tk. 18 (USD. 0.26);
interestingly the figure is Tk. 11 (USD. 0.16) if the same amount was lent to a man.
Moderate poverty falls by around 15 percent and ultra-poverty by 25 percent for households
39
who have been BRAC members for up to three years controlling for other factors according
to the author. Similar results are found for Grameen Bank and Bangladesh Rural
Development Board (BRDB) members. Recent evidence from re-survey of the same
(Khandaker, 2003). Somewhat surprisingly, the impact appears to be greater for households
who started off extremely poor (18 percent point drop in extreme poverty in seven years)
Organizations (POs) of the Palli Karma Sahayak Foundation (PKSF) in Bangladesh. The
main purpose of this study was to determine whether PKSF had been able to reach the poor
through its partner organizations. The study collected data from 1194 respondents by using a
survey technique. The study concludes that 97.51 percent of borrowers had an improvement
in family income and expenditure, quality of food, housing condition, child education,
The Bangladesh Institute of Development Studies (table 3.5) also carries out an extensive
study of the impact of PKSF Partner Organizations (POs) microcredit program using data of
3000 households between1997-2000. One of the key findings is that microcredit has a
positive and significant effect on poverty status of the program households (BIDS, 2001).
40
Table 3.5 Impact of Microfinance (compared to non-participants)
change
Impact
Food Security + greater access to cultivable land through the rental
market
microcredit
rickshaw/van)
Impacts
drinking water
enrolment of children
41
The study also finds that microcredit members are less vulnerable when struck by crisis.
latrines, contraceptive prevalence) are also more noticeable for microcredit program
Another study conducted at BRAC specifically examines the impact of microfinance on poor
women in respect of their mobility and social awareness. The findings reveal that mobility of
these women outside their home has increased. They have also gained some control on using
their own income. In many cases, the women have participated in decision making on
household issues. They are also found to be critically aware on issues relating to dowry,
family and inheritance laws, family planning and education of their children. (Abed, 2004).
There is a convincing evidence based on a representative household survey, which takes into
account common methodological problems such as selecting bias, that access to microcredit
42
Chapter 4
4.0 Introduction
The chapter outlines the regulatory framework of microfinance institutions. Section one
presents the regulatory framework of microfinance institutions along with the legal and
financial reporting requirements of MFIs in emerging countries. The last section of the
chapter highlights the major provisions of the various Acts and institutional guidelines
Microfinance has grown largely outside a regulatory framework (Jackson & Tazul, 2001).
However, in most emerging countries, microfinance activity has grown to the point where
financial regulators see the need to frame a policy and eventually to integrate some portion of
the microfinance spectrum into the framework of regulated financial services institutions
(Meagher, 2002). The need for regulation and supervision of MFIs arises from several
considerations, like protecting the interests of small savers, ensuring proper terms of credit,
financial stability and discipline and institutionalizing a proper reporting system for orderly
development.
The growing literature on the subject of microfinance regulation and supervision holds
varying views on the subject. Some are in favour of a market-directed approach, an open
banking system as it were, with the regulator simply setting the framework for the industry,
while others advocate a more government-directed stance with an active promotional role for
43
the regulator. (Jackson & Tazul, 2001). Some of them view that only deposit taking MFIs
should be regulated, whereas other groups think that both deposit taking and non-deposit
supportive regulatory framework, it is difficult for MFIs to expand and reach large of
numbers of the poor sustainably. The consequence of the lack of suitable regulatory
Regulators including government, international and local agencies are trying to develop
suitable regulatory framework in different emerging countries. Some surveys have been
Meghar (2002)) through IRIS Center, a university of Maryland, has conducted a survey on
examples (Bangladesh, Bolivia, Ethiopia, Ghana, Indonesia, Peru, the Philippines, South
Africa, Uganda and the West African Economic and Monetary Union) of microfinance
regulatory systems, from ten jurisdictions within the developing world. He examines
microfinance institutional forms that are in some way recognized, registered, regulated and
supervised – whether directly by the government or indirectly by private (or state owned but
non regulatory) institutions. In the study, he finds that most countries’ approaches to
microfinance regulation fall into one of four broad categories: (a) mandatory control under
banking law, (b) paternalistic support according to a charitable model, (c) a tiered structure
incorporating banks as well as MFIs and (d) neglected. The study also observes that
regulations and requirements vary from country to country due to different situation and
different microfinance development levels. The appropriate role for the central bank in
44
microfinance differs from country to country. Such difference reflects a variety of factors.
The financial system and the microfinance sector have evolved differently in different
countries and a variety of institutional structures have emerged. The stage of development of
the financial system, and of the microfinance sector also differs from country to country.
Other relevant factors include the degree of independence of each central bank and its
experts to conduct a survey of legal and regulatory framework in Ten Asian Countries
(Bangladesh, Japan, Lao PDR, Malaysia, Nepal, Pakistan, Sri Lanka and Vietnam). The
Survey report concludes that the regulatory authorities are still developing an understanding
regulatory framework and building the required capacity for effective regulation.
In 2007, the Asia Pacific Philanthropy Consortium (funded by Asia Foundation, Charities
Aid Foundation India and University of Iowa) conduct a survey on the recent developments
of regulatory framework of Not-for- Profit Sector in five South Asian Countries (Bangladesh,
India, Nepal, Pakistan and Sri Lanka).The report describes and analyses development of the
legal frameworks of NGOS (including MFIs) generally from about 2004 through mid 2007.
Some countries in this study found that government has shown a more welcoming,
cooperative, even facilitative attitude toward the non profit sector and philanthropy, while in
other countries conflict between the sector and the State has increased and in some countries,
elements of both increasing conflict and strengthened cooperation have been observed.
45
4.2 Legal and Financial Reporting Requirements of MFIs in Emerging Countries
As the regulatory framework for MFIs varies from country to country, it is not surprising that
the legal and financial reporting requirements of MFIs also differ among the countries.
In Tanzania, both regulated and unregulated microfinance service providers operate in the
same markets targeting the same clientele. The National Microfinance Policy (NMP)
Accountants and Auditors (NBAA) to ensure transparent disclosure. For NGOs, that receive
funding from donors and/or the domestic government, standards compliance monitoring is
done through the relevant regulatory measures issued under the Public Finance Law. The
capital and other licensing provisions, lending limits, capital adequacy, asset quality and
reporting requirements. Minimum core capital for MFCs has been set at TZD. 800 million
(Approx. USD. 800,000), in case of MFCs with a nation-wide branch network and
TZD. 200 million (Approx. USD. 200,000) in case of MFCs which do not have multiple
branches (Unit MFCs). The prescribed minimum capital is intended to ensure that MFCs
have the financial capacity to acquire the necessary infrastructure and technology. All MFCs
are required to prepare financial statements according to accounting and auditing standards
46
In Nigeria, there is no separate legal form specifically designated for microfinance
institutions (MFIs). MFIS may get license from Central Bank of Nigeria if they have paid up
capital ND. 3,000,000, liquidity ratio 40 percent and capital adequacy ratio 12 percent. Most
of the biggest ones have been duly registered and incorporated as companies limited by
guarantee and operate as secondary cooperative societies. Every MFI is also required to
submit its audited financial statements and the abridged version of the accounts to the
In the Bolivian financial system, there is no special set of laws covering microfinance. Rather
financial regulations cover all types of institutions and activities. This is an important
distinction that is evident throughout the supervisory and regulatory framework. The
Superintendency of Banks and Financial Entities (SBEF), has not attempted to establish
softer rules for microfinance services or institutions at the expense of prudential regulations.
Instead, it has sought to create a set of rules that enables the financial system to serve all
market niches. Accordingly, the regulatory framework for microfinance activities including
financial reporting requirements lies within the general rules applied to financial activities,
In 1995, Bolivia created the Private Financial Fund (PFF) by Presidential Decree. The
purpose of the PFF was to provide for a second –tier non-bank financial institution (NDFI),
to channel resources to small and micro enterprises. The Central Bank and Superintendency
of Banks and Financial Institutions have regulatory over PFFs. Some of the key features of
47
III) The maximum portfolio allocation to loans is two times the value of equity
IV) PFFs are required to submit audited financial statement annually
In India, microfinance institutions that are registered as companies come under the regulatory
accepting public deposits are subjected to rigorous supervision by Reserve Bank of India
(RBI), there are, however, only a few MFIs in the country that are registered as NBFCs
licensed as not-for-profit companies and are not accepting public deposits, are exempted
assets (USD. 12,200) and transfer of minimum percentage of profits (15 percent) to the
Reserve Fund. However, all MFIs are required to submit annual financial statements.
The Micro Financial Sector (Development and Regulation) Bill, 2007 was introduced and
passed in the Lok Sabha (Parliament) in 2007. This bill seeks to promote the microfinance
I) The National Bank for Agriculture and Rural Development (NABARD) would
regulate the micro financial sector.
II) Every MFO that accepts deposits will need to be registered with NABARD
III) Conditions for registration include a) Net owned funds of at least Rs. 500,000
(Approx. USD. 12,200) and b) at least 3 years in existence as an MFO
IV) All MFOs, whether registered or not, shall submit annual financial statements to
NABARD.
V) Every MFO that accepts deposits will be required to establish a reserve fund
through transfer of a minimum of 15 percent of its net profit realized out of its
thrift and micro finance services every year
48
VI) The central government may also establish a Microfinance Development Council
to advise NABARD on formulation of policies related to the micro financial
sector.
In Pakistan, The Microfinance institutions Ordinance, 2001 was issued to regulate the
establishment, business and operations of micro finance institutions by the State Bank of
Pakistan. The Ordinance provides for the establishment and regulated microfinance
institutions at three levels- district, provincial, and national-to integrate microfinance into the
broader financial system and to create possibilities for institutional diversity. Some of the
The ordinance assigns the State Bank of Pakistan authority for licensing, supervising and
regulating reporting of microfinance institutions. A separate division has been created in the
microfinance related matters including financial reporting and to monitor and supervise
licensed MFIs.
49
4.3. Legal Provisions for Formation and Reporting of MFIs in Bangladesh
Bangladesh, the legal framework for Not-for-Profit organizations has not gone through a
major reform. At present, NGOs including MFIs can be registered under one or more Acts of
Parliament or Ordinances and the reporting requirements vary from no systematic accounts
keeping to submission of annual reports to fund providers. These provisions apply to Not-for-
Profit organizations in general and do not specifically address microfinance activities. The
following are the Acts that deal with the formation and accountability of MFIs within
Bangladesh:
Under the provision of The Societies Registration Act 1860 (Act XXI of 1860), The Registrar
of Joint Stock Companies (RJSC) within the Ministry of Commerce is responsible for any
society formed for any “literary, science, or charitable purpose. (World Bank, 2006). A great
majority of not-for-profit organizations in Bangladesh are formed and registered under this
Act because of its wide scope and flexibility to permit a range of activities. Under this act,
each society must have a formal governing body which is responsible for managing the
society (Section 16). Societies must hold annual general meetings except where their rules
allow otherwise, and any person may inspect documents of a society filed with the registrar
(Section 19). However, the Act does not specify whether a society is required to maintain
50
The Trust Act 1882
A not- for- profit organisation can also be formed under The Trust Act 1882. A Board of
Trustees is responsible for the management of the Trust and is bound under the Act to
maintain clear and accurate accounts of the trust property. Whenever the beneficiary requests,
the trustee shall furnish him/her with detailed information of the amount and state of the trust
property, but it does not require submission of annual accounts or audit report to the
Voluntary Social Welfare Agencies (VSWA) (Registration and Control) Ordinance 1961
Under the Registration and Control Ordinance 1961, mandatory registration of voluntary
within the Ministry of Social welfare and it has broad power to suspend or dissolve an NGO
without recourse to judicial appeal. A large number of NGOs are registered under the
voluntary Social Welfare Agencies Ordinance 1961. Every agency registered under the
Registration and Control Ordinance 1961 shall maintain accounts as required by the
registration authority; submit annual reports and audited accounts; pay money into separate
accounts in approved banks; and furnish any other information required. The VSWA
registration authority may inspect books and records, including any financial information
(Section 7). In addition, any person may, on payment of a prescribed fee, inspect any
document relating to a registered agency or obtain a copy or extract thereof (Section 13).
Although the Companies Act 1994 deals mainly with private trading companies, it also
51
liability provided that the organization applies any income for promoting its objectives and
prohibits the payment of any dividends to its members. An entity wishing to register under
the Companies Act seeks approval from the Registrar of Joint Stock companies. Forming a
not-for-profit company under the Companies Act requires preparation of the Memorandum of
Not-for-profit companies registered under this Act are required to maintain proper books of
accounts, in order to provide a free and fair view of the state of affairs (Section 18). Detailed
financial reports and reports on company activities must be made available to shareholders
before and at the annual general meeting (Section 183). However, the number of NGOs
registered under this law is significantly lower than any other laws, possibly due to its more
Cooperative societies can be formed under the Cooperative Societies Act 2001 with the
services for cooperative members. Under the Act, societies are permitted to receive savings
from the general public (members). The Registrar of cooperatives has little power, either
preventive or protective, and conducts minimal supervision. The Cooperative Societies Act
includes a requirement for an annual meeting, but does not require the submission of annual
Besides the above Acts, Bangladesh Government established three Institutions: PKSF-a
wholesale financing company to finance and regulate its Partner Organizations (POs), NGO
52
Affairs Bureau-primarily to regulate NGOs which receive donations and grants from foreign
entities and MRA (Micro Credit Regulatory Authority) to supervise, control and monitor the
licensed MFIs.
PKSF is the most important organisation for the regulation and financing of MFIs in
Bangladesh (Hulme & Moore 2006). PKSF itself was created in 1990 under the Companies
Act 1913 (amended in 1994) to support and assist, financially and institutionally, to its
eligible partner MFIs (POs) and monitors their performance, including financial reporting.
With regard to financial reporting by its POs, PKSF had developed guidelines for accountants
for POs; and internal and external audit procedures including their appointments (Miah,
2006). In collaboration with the World Bank, PKSF had introduced certain International
Accounting Standards for financial reporting of its partner organizations (POs) and switched
from cash basis of accounting to accrual basis of accounting (World Bank, 2006).
the microcredit sector in Bangladesh (PKSF, 2007). The Committee appointed by the
Government to prepare the regulatory framework for MFIs was headed by the Governor of
Bangladesh Bank, while the Technical Committee, which prepared detailed guidelines and
format for MFI financial reporting was headed by the Managing Director, PKSF.
53
The NGO Affairs Bureau (NGOAB)
The Government formed an agency (The NGO Affairs Bureau) to regulate the affairs of
requires all NGOs seeking or receiving foreign funds to register with it under the Foreign
Contributions (Regulation) Ordinance of 1982 and to renew that registration every five years.
The Bureau is concerned with receiving and utilization of foreign funds for projects, and
evaluating statements and reports submitted by NGOs including auditing incomes and
expenditure accounts by external auditors. However, unlike the PKSF, it has not developed
any guideline for financial reporting. Organizations are only required to submit the partial
(Act number 32 of the year 2006) in August and later established the Microcredit Regulatory
Authority (MRA) to improve the institutional and regulatory framework for MFIs. Apart
from authorising operations, audit of accounts and policy formation, the MRA requires MFIs
(under Section 27 of the Act) to maintain their accounts in the manner prescribed by the
Authority, prepare annual budget for the next financial year and prepare the Balance Sheet,
the Profit and Loss Account annually, and submit it to the Authority. However, the
requirements of MFIs by MRA are still on trial stage and are not fully implemented yet. Most
of the licensed MFIs are still confused about the financial reporting format and are reluctant
54
Some of the features of Microcredit Regulatory Act 2006 are as follows:
Registration requirements:
All licensed MFIs should be registered by either one of the following Acts:
The Voluntary Social Welfare Agencies (Registration and Control) Ordinance, 1961
The financial reporting provisions contained in each of the above Acts have been discussed in
the previous section.
MFIs (fulfilling the above criteria) are required to submit 8 pages (prescribed by
authority) half yearly reports and also required to submit audited annual reports
One page reporting system is applicable for those MFIs which have not reached the
55
4.3.2 Financial Reporting Environment of MFIs in Bangladesh
As there are several legislation and regulatory bodies, the process of monitoring and
regulated and users and regulators get limited amount of information from MFIs in
Bangladesh (Tazi, 2006). Most of the small MFIs do not prepare annual reports, rather they
prepare management reports including financial highlights showing receipts and payments in
two pages. Some organizations prepare financial reports and their financial accounts are also
being audited, but these reports provide minimum information on assessing their financial
position. Some financial statements, for example, limit performance analysis by not
separating microfinance operations from the institution’s other activities, such as health or
training services. Audits are generally conducted to comply with donor requirements. To
satisfy the requirements of different donor organisation, MFIs sometimes undertake multiple
audits and in some cases, even close their books twice a year (Irish & Simon, 2005). The
production of multiple audited accounts creates an additional strain on limited resources and
results in showing a partial picture of the MFIs since donors usually ask for audited accounts
The Consultative Group to Assist the Poorest (CGAP), a consortium of public and private
development agencies hosted by the World Bank, has issued a set of disclosure guidelines on
MFI financial reporting requirements (Rosenberg et al., 2003). However, these guidelines are
not accounting policies and do not command any particular format for financial reporting.
Instead, they simply indicate the minimum information that should be included in MFI
financial reports. The CGAP encourages MFIs to adopt International Accounting Reporting
Standards (IAS 1, 30 and 32) issued by the International Accounting Standards Board (IASB)
56
PKSF, with support from the World Bank, has moved towards consolidated audited accounts
for the microfinance operations of its partner organizations. PKSF have also adopted certain
International Financial Reporting Standards (IFRS 1, 7 and 10) for financial reporting for its
POs, which resulted in fairly uniform financial reporting, audits and presentations (World
Bank, 2006). To improve the regulatory and reporting framework for MFIs, the Government
earlier, these reporting requirements by MRA are still on trial stage and not fully applied yet.
4.4 Summary
The chapter outlines the regulatory framework of microfinance institutions along with the
legal and financial reporting requirements of MFIs in Bangladesh and beyond. The main
finding of the chapter is that the regulatory framework of MFIs is mainly under regulated and
it varies from country to country. In Bangladesh, even though, there are different Acts to
regulate and control MFIs, however, none of them (except Company Act) requires
submission of audited financial statement. PKSF is the only exception. PKSF, with support
from the World Bank, has moved towards consolidated audited accounts for the microfinance
operations of its partner organizations. PKSF have also adopted certain International
Financial Reporting Standards (IFRS 1, 7 and 10) for financial reporting for its partner
organizations.
57
Chapter 5
Literature Review
5.0 Introduction
Because there is limited regulation on financial reporting by MFIs, not surprisingly limited
empirical research has been undertaken due to lack of data even then traditional financial
institutions are highly regulated with respect to disclosure of information (Nieto et al., 2008).
However, over the years several studies have been undertaken on Not- for- profit
organizations and their conclusion could be drawn on to improve reporting and accountability
The chapter reviews related studies on NGOs/MFIs accountability, disclosure, regulation and
governance in order to develop a framework and hypotheses for empirical investigation in the
subsequent chapters. The review starts with accountability studies, followed by disclosure,
governance and regulation of microfinance and not- for -profit organizations. As the research
methods draw on some selected prior disclosure studies of corporations, this review will
Most accountability studies have focused on the definitions and scope of not-for-profit
organizations with regard to accountability rather than financial reporting specifically. For
example, Brown et al. (2001) focused on developing the concept of international non-
abstract concept and then as strategic idea with very different implications for different
58
INGO strategies. The study examines those implications for INGOs that emphasize delivery,
capacity-building and policy influence. They recommend that INGOs committed to service
delivery may owe more accountability to donors and service regulators; capacity building
INGOs may be particularly obligated to clients whose capacities are being enhanced; and
targets. INGOs that are expanding their activities to include new initiatives may need to
Tilt (2006) outlines the need for a different conceptualization of accountability for NGOs,
than the one normally use when calling for greater accountability of corporations, and
demonstrates there already exists many effective accountability mechanisms. The study
points out that a major flaw in the argument for greater NGO accountability is that it is
usually presented from a perspective that ignores the conflict between control and the role
that NGOs play in civil society-a role that provides a voice for those unable to speak for
themselves and counters the views of more powerful groups. Tilt (2006) concludes that
NGOs play an important role in society and in the absence of profit motives, it may be
necessary to apply different criteria when judging their effectiveness. Similarly, Unerman
and O’Dwyer (2005) demonstrate that the mechanisms required for effective accountability
by the NGOs will usually be different to mechanisms suited to discharging the accountability
duties of other forms of organization. They conclude that flexible and informal accountability
mechanisms seem most suited to situations of closer relationships between an NGO and its
key stakeholders.
Goddard and Assad (2005) find that organizational culture resist accounting in NFPOs and its
use is more symbolic and in navigating legitimacy. They also find that organizations with
59
stronger accounting functions, better qualified accounting personnel and well documented
scrutiny and the transparency with which its affairs were conducted.
Prior to them, Ebrahim (2003a) and Khan (2003) find that accounting reports and disclosure
organizations (NGOs) in USA. The study reviews five broad mechanisms: reports and
regulation, and social audits. Each mechanism, distinguished as either a “tool” or a “process”,
“internal” mechanisms remain comparatively underdeveloped. He also finds that NGOs and
expense of long-term “strategic” processes necessary for lasting social and political change.
Khan (2003) examines the accountability of NGOs in Bangladesh to patrons, clients and
themselves by employing case studies. The study observes that NGO’s accountability to
to patrons are in place, accountability to clients and to themselves still suffers from a number
of deficiencies.
60
All these studies stated here focused on accountability with a view to develop a framework to
assess how the functional accountability has been discharged as these organizations are
different from other types of organizations with regard to objectives, functions and reporting
on key performance. The next section presents some studies on reporting and disclosure by
NFPOs.
Yetman and Yetman (2004) examine the effects of governance mechanisms on the financial
reporting quality of NFPOs in the USA. They categorize governance into market-based
(monitoring by lenders and investors) and regulatory (monitoring by Federal and State
Governments) and examined how these two monitoring mechanisms affect the quality of
financial reporting. They find that market-based governance measures have a more consistent
and beneficial effect on reporting quality than the regulatory based measures. Desai and
Yetman (2004) analyse the influence of state-varying legal and reporting rules on the
behaviour of public charities and foundations from 1987 to 2000 in the USA. They develop
state-level indices of the legal and reporting rules and examine how these rules shape
accounting behaviour of NFPOs. They find that stronger non-distribution constraints are
associated with greater charitable expenditures and foundation payouts while more stringent
reporting requirements are associated with lower insider compensation. They also find that
stronger governance measures are associated with intertemporal smoothing of resources and
greater activity in response to negative economic shocks. Prior to them, Krishnan et al.
(2002) examines how NFPOs manage their performance efficiency and effectiveness by
manipulating certain accounting ratios such as the program service ratio and the fund rising
ratio. The study finds that managers of these organizations have incentives to over-report the
expenses, classified as program services, and under report the expenses, classified as
61
administrative and fundraising, in order to improve these ratios. Behn et al. (2007) examine
the incentive for voluntary auditing of financial statements of 300 largest NFPOs in the USA
in 2001. They find that organizations with a high level of debts, a large contribution ratio and
compensation ratio, larger in size and with a low level of lobbying expenses are more likely
Outside the USA, Simnett (1987) examines the determinants of accounting information
NFPOs, he finds that gross income, number of members, the presence of a professional
administrator, geographic dispersion of members, and auditors of the financial statements are
accounting experience of the person preparing the financial statements and use of outside
Nieto et al., (2008) is the only study to our knowledge that has focuses on financial reporting
of MFIs in different countries in the world from a database of MFI’s, known as MIX Market,
a web-based platform that compiles information on MFIs worldwide and whose membership
is voluntary. They examine the incidence of financial and social information disclosure on
the internet and the contents of these reporting. Using a sample of 273 MFIs, they find that
overall internet presence of MFIs is very limited and that large MFIs with a high degree of
public exposure on the internet disclose more information on their web sites than smaller
MFIs with a low degree public exposure. Countries with higher levels of GNP and improved
level of ICT infrastructure use more internet for reporting and MFIs located in Africa and
Latin America are less likely to reveal information than those located in Asia or Eastern
62
Europe. Since their study examined the incidence of reporting on the internet using a
secondary source whose membership is voluntary and the results lacked the depth in
In Bangladesh, Mahmood (2001) observes that MFIs typically present financial reports that
are less rigorous reflections of their financial performance. Some do not even produce annual
financial statements due to flexible regulations for MFIs. Most use cash accounting and do
not apply depreciation, provision for loan losses etc. The audited financial statements of
ascertain to what extent accounting principles are homogeneous and it is problematic to have
meaningful analytical comparisons. The study suggests that MFI needs increased
transparency, accountability and reporting in dealings because an MFI with good governance
will perform better over time which will promote investor confidence, facilitates the flowing
Bala and Mir (2006), using a case study approach, find that accounting reports and disclosure
underdeveloped. They observe that the NGOs that are dependent on the external sources for
their funding are more dependent on their day-to-day policy-making including their
management control and reporting systems. These organizations spend more times to keep
their funding providers happy rather spending time to assess and evaluate the outcome of
their beneficiary level activities. Their management control and reporting systems are centred
around their funding providers and not centred around their beneficiaries. Bala and Mir
(2006) support the previous literature on Bangladeshi NGO accountability, which mainly
63
criticised the NGOs for their failure to adopt a beneficiary-centred accountability system.
Their conclusion is hardly surprising because the bulk of funding for NGOs in emerging
countries come from developed countries and international funding bodies such as the World
Bank which requires a system that reports information more often than their local monitoring
bodies.
It has been pointed that the quality of governance has a beneficial effect on financial
reporting and performance of NGOs. For example, Hartarska (2005) presents the first
institutions (MFIs) in Central and Eastern Europe and the newly independent states. He
examines the impact of management remuneration, board independence and diversity and
is not associated with better performing MFIs; lower wages suggested for mission-driven
organizations with worsen outreach, while managers’ experience improves performance. The
results also provide strong support for independent boards with limited employee
participation. The study also finds that region, external governance mechanisms seemed to
Recently Mersland and Strom (2009) assess the relationship between firm performance and
MFIs collected from third-party rating agencies. They examine the effects of board and CEO
on an MFI’s financial performance and outreach to poor clients. They find that financial
performance improves with local rather than international directors, an internal board auditor,
and a female CEO, the number of credit clients increased with CEO/chairman duality and
64
outreach is lower in the case of lending to individuals than in the case of group lending.
Mersland and Storm (2009) find no difference between not-for-profit organizations and
shareholder firms in financial performance and outreach. They also observe that bank
regulation has no effect. However their study suffers from selection bias as the sample
represents commercial and professionally oriented institutions that have decided to be rated
A similar type of study is conducted by Soltane in 2009. He examines the relation between
institutions (MFIs) in terms of outreach and sustainability. He finds that performance based
compensation of managers is not associated with better performance of MFIs. The results
identify trade-offs between MFIs outreach and sustainability depending on larger board size,
and on higher proportion of unaffiliated directors. The study also shows that the more women
are on the board, the better the performance and reveals that external governance mechanisms
help MFIs to achieve better financial performance. Soltane (2009) also identifies other
factors leading to better sustainability such as regulation, and the use of individual lending
methodology.
Irish and Simon (2005) conduct a survey on legal environment in Bangladesh. They find a
lack of clear accountability and internal governance rules for NGOs in the current legal
framework in Bangladesh. There is a very little in the principal registration laws (except
Companies Act) which deals with internal governance and external accountability for Non
Government Organizations. NGOs in the survey report that they do follow certain established
accounting standards, but there is a wide variety of the standards used. The study suggests
65
that it may be useful to adopt a set of clear standards strictly for use by NGOs, as has been
done in India.
The World Bank (2006) also finds that laws relating to internal governance and financial
accountability of NGOs are inadequate. The reporting and accounting burdens placed on
NGOs are contradictory and weak. Capacity of NGOs and computerized accounting system
vary. The study observes that PKSF introduced some standardized financial management
practices for microfinance programs, but accounting and auditing practices are also
influenced by highly variable donor financial reporting requirements, that typically focus on
the donor-financial project rather than on the whole institution. The quality of audit reports
varies significantly and regulators rarely focus on the comprehensiveness and quality of these
reports. The report also finds that there is a little oversight over NGO audit standards within
the profession. Moreover, audit reports are rarely posted in public spaces (such as on relevant
websites). The study concludes that NGO Boards suffer from weakness common in for-profit
corporate sector and typically feature the founder director, family members and a number of
Abbey (2008) demonstrates how regulation can improve the effectiveness of development
intervention of NGOs in Bangladesh and Dominican Republic. The study finds that
regulation of NGOs themselves can legitimize their role, improve their professional standards
and assure accountability to the general public. Abbey (2008) concludes that simple
registration, self-help regulatory mechanisms and formal regulation can improve the
66
Overall, the above review suggests that the financial reporting quality of MFIs has been
under-researched and that there is a gap in our understanding of how they perform their
reporting function via provision of information in annual financial statements. Bala and Mir
(2006) also note that NGO literature in Bangladesh provides non-empirical recommendation-
type discussion on administration and activities and lacks substantive analysis of how they
67
5.5 Summary
The chapter reviews prior studies on the accountability, reporting and disclosure, governance
and regulation of microfinance institutions/ not- profit- organizations and corporations. Due
to limited regulation on financial reporting and lack of data, very few empirical researches
have been undertaken on micro finance sector. The MFI literature in Bangladesh provides
lacks substantive analysis of how they discharge their accountability functions via financial
reporting. Overall, the above review suggests that the financial reporting quality of MFIs has
been under-researched and that there is a gap of how they perform their reporting function
via provision of information in annual financial statements. Further, corporate disclosure and
perception studies have been reviewed to identify the attributes of financial disclosure which
could be used for developing the model for the current study.
68
Chapter 6
6.0 Introduction
The purpose of this chapter is to present the hypotheses. The first section outlines the
hypothesis development process along with identifying financial attributes affecting financial
disclosure of MFIs. In the last section, the research hypotheses have been developed for
testing.
One striking feature in corporate reporting is that a company generally provides information
to discharge specific obligations to society, investor, supplier, creditors, and legal authorities.
Statement of Financial Accounting Standard (SFAS) 1 declares that investors, creditors, and
others may use reported earnings and information about the elements of financial statements
in various ways to assess the prospects for cash flows. Whereas, SFAS No. 117 states that
information to meet the common interest of donors, members, creditors and others who
have common interests in assessing (a) an organization’s services and its ability to provide
those services and (b) how managers discharge their stewardship responsibilities and other
There are significant similarities between these two SFASs. Creditor’s interests are basically
the same for a not –for- profit or profit making organization. Creditors are concerned about
the amounts, timing and uncertainty of perspective cash flows to assess the risks of providing
69
loans to any enterprise. Not-for-profit organizations have no residual claimants or investors
as for-profit organizations have but prior research suggests that funders act as investors.
Additionally, while not-for- profit organizations are not publicly traded, they compete for
scarce resources, whether for funding, labour or managerial talent, so they have similar
economic incentives to publicly traded firms while not having the ownership structure of
publicly traded firms (i.e, no residual claimants). Stakeholders in both for-profit and not-for-
As a result of these similarities of profit and not –for- profit entities, the existing profit and
not-for-profit theories, as well as previous research findings on profit and not-for- profit
organizations, have been drawn in identifying the attributes of financial disclosure. Agency
theory has become widely accepted in recent times and used in corporate disclosure studies to
agency theory can apply to non profit organizations as agency problem also exists there. The
agency theory (Jensen & Meckling, 1976) suggests that the firm represents a “nexus of
contract,” implicit and explicit, between suppliers of various factors of production. The nexus
of contracts view has particular appeal for accounting because many of these contracts are
Ebrahim (2003b) identifies three primary groups which have explicit or implicit contracts
with NGOs : (1) Funders ( 2) Sector and (3) Clients and Communities. Funders provide fund
to not-for- profit organizations for regular reports and evaluations that confirm the legitimate
use of those funds. Laws governing not-profit status and ensure proper disclosure of not-for-
70
organizations, for the purpose of ensuring public trust. Members or clients are concerned
about their deposits and require financial statement to evaluate the organization’s
performance. In all these contractual relations, accounting information can assist these groups
A number of studies examine the financial disclosure issues in the corporate literature.
Marston and Shrives (1991) reviewed disclosure studies undertaken before 1986. They
observe that corporate size, listing status, leverage, profitability, and size of the audit firm are
the most common characteristics examined with regard to disclosure level. The study find
that corporate size and stock exchange listing status are significantly associated with higher
disclosure levels, and results are inclusive regarding profitability, leverage and size of the
audit firm. Ahmed and Courtis (1999) conduct a meta analysis of 29 previous disclosure
studies. The study finds a significant and positive relationship between disclosure levels and
corporate size, listing status and leverage. The study does not find any significant association
The above ideas have been used to identify the attributes of financial disclosure for not-for-
profit organizations (MFIs). A review of the accounting requirements that are imposed upon
not-for-profit organization as a result of their legal form and funding body requirements is
also considered. In this present study, MFI’s size, debt levels, rate surplus of revenue over
expenditure, size of audit firms, nature of registration, MFI’s age, governance structure,
frequency of board meetings and qualifications of board members have been identified to
71
MFI’s Size
Firm size is expected to be positively associated with disclosure level because information
generation and dissemination are costly, and therefore, large organisations with more
resources and superior expertise are better placed to produce comprehensive and detailed
financial statements (Ahmed & Nicholls, 1994; Nieto et al., 2007). Chow and Wong-Boren
(1987) argues that agency costs increase with firm size, so large firms voluntarily disclose
more information in annual reports to reduce agency conflicts. Utilising legitimacy theory,
Cowen et al., (1987) and Hacskston and Milne (1996) argue that a larger company carries out
more activities, receives more attention from the general public who are concerned with its
social programs than a smaller company which affect disclosure practices. Utilising the same
logic for MFIs, it is expected that large MFIs would disclose more informational items. The
size of the MFI is measured by total asset and the variable is called as MFISIZE.
Alternatively, the size is also measured by the number of members within the MFI because
the demand for financial information is the sum of the individual demands. As the number of
members increase, the total demand for financial information increases hence it is expected
Prior studies on corporations argue that highly levered firms incur high level of monitoring
costs and these firms seek to reduce these by providing more information (Jensen &
Meckling, 1976). On the other hand, if the debt levels are high, firm might disclose
information to the lenders directly rather than to all stakeholders via annual reports,
suggesting that the extent of information disclosure will be lower for firm with more debts. In
the case of MFIs, funds are provided by PKSF for its POs and it is expected that it would
72
monitor more those entities which have more borrowings. The debt level of MFI is measured
by the total amount of borrowing, divided by total capital as shown in the balance sheet and
defines this variable as LEV. Given the inconsistent finding in prior studies concerning the
management, they are likely to disclose more information so as to signal to investors and
other users of financial information (Singhvi & Desai 1971; Wallace et al., 1994). Profitable
MFIs have also resources to defray the costs associated with dissemination of information.
Hence, a positive association between disclosure and profitability is expected. For MFIs the
term profitability is not used because the objectives of these organisations are not to make
profits but use the resources for social upliftment. I measure the rate surplus of revenues
over expenditures scaled by total capital and call this variable SURPLUS. In the absence of
prior research on MFIs and inclusive findings regarding corporation, no direction between
One of the aims of having financial statements audited to ensure that information which is
important to the financial statement users is disclosed. DeAngelo (1981) argues in the context
of corporations that the larger audit firms have incentives to provide high-quality audit
services in order to enhance their public reputation as a credible monitor. They will therefore
encourage their clients to disclose comprehensive information in their annual reports and
accounts. In the case of MFI in Bangladesh, auditors are employed by PKSF for its POs on
73
the basis of experience and pricing, so some variation in size within the audit firms is
disclosure standards and auditor quality is not conclusive, in emerging countries including
Bangladesh a significant association has been found (see for a review, Ahmed & Courtis,
1999).
It is difficult to segregate audit firms in Bangladesh along Big and non-Big categories
because none of the Big-4 firms operate in Bangladesh under their brand names due to
restrictions imposed by the local professional body – the Institute of Chartered Accountants
of Bangladesh (ICAB). Instead, reputed local firms represent international firms including
the Big 4 by way of obtaining independent affiliations of such firms. Two criteria were used
to group audit firms as large - firms that employ four or more chartered accountants and those
that have an affiliation with an international Big or non-Big 4 firm (Karim & Moizer, 1996).
The rating was then discussed with a couple of senior partners of practising accountants to
reduce biasness. A dummy variable is used, where if the audit firm is large then assign 1
otherwise 0 and call this variable AUDDUM. A positive association is expected between this
Nature of Registration
As stated earlier different Acts exist to regulate and monitor MFIs in Bangladesh, some of the
provisions are common across the various Acts and some are distinct and require
MFI registered under the Companies Act 1994 is expected to provide more information
because the Act, by focusing on profit making companies, requires a more rigorous
disclosure standard compared with other Acts that regulate MFIs within Bangladesh.1 is
74
assigned if the MFI is registered under the Companies Act 1994 and 0 if the MFI is registered
under one or more of the other Acts: Social Welfare Act, NGO Bureau Act and Co-operative
Act. This variable is called LEGDUM and expects a positive association between this
It is widely recognised that emerging countries suffer from skill shortages due to low level of
educational qualifications, and Bangladesh is not an exception where the rate of adult literacy
Following prior research, it is suggested that if the senior management are highly educated,
that level of expertise will be reflected in performance and transparency of the organisation.
Therefore data have been collected about the educational levels of directors of MFIs and
found that their educational levels vary amongst them. A dummy variable 1 is assigned if a
director holds a master degree or equivalent and 0 otherwise. One of the reasons for
assigning 1 for master degree is that the degree is conferred by universities or university
colleges where the quality of educational standard and students are much higher than normal
university accounting programs. For empirical analysis, if the majority of the directors hold
master degree or above, 1 is assigned to that MFI, otherwise 0. For sensitivity tests, the
criteria has been changed and assigned 1 for bachelor degree including overseas trained (not
having a master degree) directors, otherwise 0. The variable is called QALBD and expects a
75
A Summary of the hypothesized relationships between each independent variables and the
MFSIZE +
LEV ?
SURPLUS ?
Overall Disclosure AUDDUM +
LEGDUM +
AGE ?
BDSIZE ?
FBM +
QALBD +
Figure 6.1
6.3 Hypotheses:
Based on the previous discussions and literature review, the following hypothesises have
been developed for this study. It is mentioned on page 19 that like corporation, the external
parties use MFI financial statements in assessing (a) an organization’s services and its ability
to provide those services (b) how managers discharge their stewardship responsibilities and
other aspects of their performance (FASB 1993). So the MFI annual report or financial
statement should disclose adequate and relevant information to satisfy the perceived
information needs of users. On the other hand, it is mentioned on page 76 that preparers of
corporate annual reports are willing to provide users with information they need, but they
may not be aware of the importance of some type of information for external users (Firth,
1979). Preparers ‘ perceptions are also important as they prepare MFI financial statements
and are well aware about the disclosure need of which items are more important than others.
(page 18). Based on these discussions, the first 4 hypotheses have been developed. The fifth
hypothesis has been developed based on the literature review in the thesis (pages 62-74).
76
See for example, Prior empirical studies on disclosure level have recognized the relationship
between the extent of disclosure and some firm-specific characteristics . (Singhvi & Desai,
1971; Buzby, 1975; Firth, 1979; McNally et al., 1982; Cook, 1989, Ahmed & Nicholls, 1994;
H1: There is no consensus between preparers and users regarding the importance
H2: There is no significant gap between the information needs of users and actual
H3: There is no significant gap between the perceptions of preparers and actual
H4: There has been no improvement in the quality of overall disclosure since 2003
H5: There is no association between the total disclosure quality and MFI
characteristics such as size, debts levels, rate surplus of revenue over expenditure,
size of audit firms, nature of registration, MFI’s age, governance structure, frequency
6.4 Summary
The chapter outlines the hypothesis development process along with the identification of
organizations, a review of corporate disclosure studies have been undertaken to identify the
identified attributes of financial disclosure. A review of the accounting requirements that are
imposed upon not-for- profit organizations as a result of their legal form and funding body
requirements is also considered. In this present study, the attributes of financial disclosure
77
such as MFI’s size, MFIs debt levels, rate surplus of revenue over expenditure, size of audit
firms, nature of registration, MFI’s age, governance structure, frequency of board meetings
and qualifications of board members, have been identified to examine the extent of
78
Chapter 7
7.0 Introduction
The purpose of this chapter is to discuss the research method used. The first section of this
chapter discusses the research methodology. The next section outlines the data collection
methods and sources followed by statistical techniques used for hypothesis testing.
A questionnaire survey method was used to determine the perceived needs of users (auditors,
donors and regulators) and preparers (MFI accountants) for information items to be disclosed
and then compared with the information needs of users and preparers mentioned above.
Further, the disclosure index of 165 MFIs over a period of 3 years beginning in 2004 was
constructed to assess whether there has been any improvement during this period. Finally, a
regression procedure was used to determine the effect of the MFI’s characteristics such as
MFI’s size, MFIs debt levels, rate surplus of revenue over expenditure, size of audit firms,
nature of registration, MFI’s age, governance structure, frequency of board meetings and
Conventional accounting theory emphasises that the corporate reports should disclose
adequate and relevant information to users to assist them in making economic decisions.
79
However, disclosure adequacy is an abstract term and there is a lack of generally agreed
theory of disclosure adequacy. According to Buzby (1974), adequate disclosure will vary in
different situations, depending upon the audience for whom the disclosure is made, the
purpose for which the disclosed information will be used, the quality of information required
by users, the manner in which the disclosure is made, and the timing of disclosure.
Two approaches may be used to assess disclosure adequacy: (a) using a decision model
(methods mostly taken from psychology) to examine whether items disclosed in corporate
annual reports constitute an important part of the users’ decision process and (b) employing
disclosed in corporate annual reports. According to Cooke (1989), decision models may be
categorised into perspective and descriptive. For example, the Lens model tells us what
should be done. Alternatively, a descriptive decision model describes the steps involved in
Methods using users’ decision model require the specification of users, their needs for
information items, the kind of decisions they make and the decision model they employ in
deciding the usefulness of information. There are some problems with the decision model
approach: (a) the identification of the users’ decision processes, (b) the methods of resolving
differences with respect to disclosure items and (c) the method of selecting between two or
more equally important information items which are relevant to users’ decision model
(Wallace, 1987). Where it is not feasible to determine users’ decision process and preference
for information items using the decision model approach, the perceived importance of
information items to be disclosed in corporate annual reports for decision making can be
80
researchers have applied the survey method to determine the perceptions of users concerning
the usefulness of accounting information. The present study also adopted the questionnaire
survey method to determine the perceived needs of practicing users and preparers concerning
There are different ways for companies to disclose information. Annual reports are
considered as the main source of information (Shohaieb, 1990; Knutson, 1992), although
Marston and Shrives (1991), Epstein and Palepu (1999) and Hope (2002) argue that annual
reports should be considered a useful but not sufficient disclosure vehicle. Moreover, Lang
and Lundholm (1993, 1996) find a high and significant positive correlation between annual
report disclosures and other forms of disclosure. Therefore, it is not surprising that most prior
financial reports. However, quality is neither a readily measurable nor a generally agreed
characteristic of a firm (Imhoff, 1992). According to Cooke and Wallace (1989), quality of
accounting information disclosure can not be measured directly because it does not possess
inherent characteristics by which one can determine its intensity or quality like the capacity
of a car. Despite this problem, some authors define disclosure quality as the extent of
information items, generally held to be useful to one group or several groups of users, are
disclosed (Cerf, 1967; Singhvi & Deshai, 1968; Zimmerman, 1977; Copeland & Ingram
1983). This method was developed by Cerf and is called disclosure index method. In the
81
literature, the extent of disclosure and quality of disclosure has been used interchangeably
and several authors modified disclosure index according to their research objectives (Singhvi
& Desai 1971; Buzby, 1974; Firth, 1979; Wallace 1988; Cooke, 1989).
disclosed: the number of data items, the frequency of disclosure releases, financial analysts’
perceptions about the disclosure practices, and the disclosure index. The amount of
information disclosed can be measured per category or per company by counting the data
items, i.e, the number of words and numbers shown in the accounts, the number of sentences,
and the number of pages (Marston & Shrives, 1991; Hacskton & Milne, 1996). This approach
has been referred to as content analysis. This method is very popular in disclosure studies.
However, it might present problems due to, for instance, the repetitions of some items of
information in the annual reports. Moreover, numbers cannot be explained without words.
The frequency of disclosure releases is also used around an information event such as
earnings announcements and equity offering (Lang & Lundholm, 1993; Botosan & Harris,
2000). However, MFIs do not announce any earnings or offer equity. Perceptions of financial
analysts about the firm disclosure practice such as disclosure ratings from the Financial
Analysts Federation (FAF), the Association for Investment Management and Research have
been used by researchers as proxies for disclosure quantity and quality (Lang & Lundholm,
1996; Healy & Palepu, 2001). However, disclosure scores reflect analysts’ perceptions of
firms’ disclosure rather than the disclosure policies themselves and cover only a limited
number of large firms, creating perhaps the possibility of sample bias. Moreover, Lang
(1999) questioned the objectivity of these ratings given that no can know the incentives of
analysts to report their ratings and what kind of biases could be concluded.
82
In the present study, the disclosure index method is used. This method has been extensively
used in prior research (Wallace et al., 1994; Meek et el., 1995; Inchausti, 1997; Ahmed &
Courtis, 1999; Depoers, 2000; Hope, 2002). A disclosure index is a measure (a proxy) of the
selected items of information in the company reports (such as annual reports, financial
statements). Marston and Shrives (1991) in a review paper conclude that in spite of validity
and reliability problems, the disclosure index is a useful research tool for assessing disclosure
quality. According to them, the usefulness of a research tool should be assessed by the extent
to which it is used by researchers and the consistency of research findings across many
studies. They observe that the disclosure index method has been used by several researchers
over a period of more than 30 years and in most cases provided consistent results across
several studies.
Disclosure information in annual reports can be divided into two broad categories, mandatory
fulfilment of the requirements of statute in the form of laws, professional regulations in the
form of standards and the listing rules of stock exchanges. Voluntary disclosure is any
either weighted or unweighted. Weights are usually determined by seeking out the views of a
user group regarding the relative importance of various items of information (Buzby, 1975;
Firth, 1979). Alternatively, the unweighted approach gives equal weight to each item in the
list of items of information according to its existence. An item takes 1 if it exists in the
investigated report and 0 if not (Cooke, 1992; Inchausti, 1997; Meek et al., 1995; Depoers,
2000). However, some studies have used two approaches as robustness check (Chow &
83
As mentioned earlier, disclosure index was used heavily in the accounting literature during
1970s’ to 2000s’ to investigate various aspects of corporate financial reporting such as; (a)
the association between firm characteristics and disclosure levels, (b) the economic benefits
of increased disclosure, and (c) the compliance with regulations or accounting standards.
Many prior studies use a country-relevant disclosure index either self-constructed or provided
externally as a measure of the disclosure level. Since there is no agreed theory (Wallace et
al., 1994) on either among the numbers (among 11 to 224) or the selection of informational
items (mandatory or overall disclosure), different disclosure indices have been used in
different studies.
In light of the above review, it can be noticed that several different approaches have been
adopted to measure disclosure quality and quantity, but there is no general theory that offers
guidance on the selection of items to measure the extent of voluntary disclosure (Marstom &
Shrives, 1991). Disclosure, by its very nature, is an abstract construct that does not possess
inherent characteristics by which one can determine its intensity or quality (Wallace &
Naser, 1995).
management forecasts for MFIs, the discharging of accountability to stakeholders via annual
reports or financial statements remains the only source for microfinance sector. Though fund
providers get some amount of information through direct contacts and regulatory bodies.
In this study, the number of informational items has been used to measure the extent of
disclosure, though this approach has limitations including the measurement errors and
reliability. To mitigate such problems, the well accepted methods such as counting and
84
cross-checking several annual reports by an independent rater have been applied. In the
event of variation between the researchers and the independent rater, further discussion was
held to find out the sources of errors to improve the process of construction of the index.
As mentioned earlier, disclosure information in annual reports can be divided into two broad
requirements for disclosure for MFIs, the index in this study is based on total informational
debate on the pros and cons of using weighted versus unweighted index including
subjectivity, lack of insights of users and no difference in results (Marston & Shrives, 1991;
Ahmed & Courtis, 1999). The index is based on 82 items comprising 5 items for financial
statements; 9 for revenue items; 16 items for expenses; 11 items for assets; 11 items for
liabilities; 5 items for other statements; 12 items for accounting policies; and 13 items for
other information.
Robbins and Austin (1986) developed a scheme for scoring the disclosure index in which
items of the disclosures were rated according to their degree of specificity. This scheme
each annual report are compared to the discretionary items on the list and coded as 1 if
disclosed or else 0. The disclosures coded as 1 are then added to ensure arrival at a total score
information is relevant but not disclosed then it will be assigned as 0. It is acknowledged that
85
this introduces an element of subjectivity into the dichotomous procedure. However, failure
to adopt such a procedure would mean that more diversified enterprises would be able and
The score is unweighted in that each disclosure is assigned only a score of 1, if applicable,
any disclosure area over another. The assumption is that each item is equally important.
Spero (1979) argues that the use of weights derived from the preference of user groups is
irrelevant because enterprises which disclose “important items” also disclose “unimportant
items”, that is, firms are consistent in their disclosure policies. On the other hand, Buzby
(1974b) argues that for measuring the adequacy of disclosure practices, it is usual to
introduce the weighted disclosure index to reflect the differing opinions of user groups
regarding the perceived importance of different disclosure items. However, empirical results
showed that the results derived from these different methods are not significantly different
(Firth, 1980; McNally et al., 1982; Robbins & Austin, 1986; Chow & Wong-Boren, 1987).
Support for using an unweighted disclosure index may be found in (Spero, 1979; Robbins &
Austin, 1986; Wallace, 1988b; Cooke, 1989a; 1992). It is argued that, in general purpose
financial reporting, it is difficult to determine whose preference should be reflected and given
priority, as not all user groups’ perceived information requirements can be sought at one
time.
86
Each item of information is disclosed in the annual reports of the sampled MFIs, 1 is
assigned and 0 otherwise (see Appendix 2). The overall disclosure index is calculated by the
nj
x y
i1
Ij
nj
Where
So that 0 ≤ Ij ≤ 1.
In this form the expression for TDI allows for the fact that it has been assumed that if a
particular item was not mentioned in the financial report, it would be treated as not
applicable. For example, if no mention was made in respect of grant received, it was assumed
that the item was not applicable for the MFI in that year. However, If it was disclosed
without mentioning the amount, then the score would be 0. Cooke (1989a) acknowledged
that this procedure would introduce an element of subjectivity. To overcome this potential
bias, he suggested that the whole annual report should be read first and a judgment made
about whether a particular item was applicable. This approach is adopted in this study.
87
7.2 Data Collection
Data were collected by two sources: questionnaire survey and archival records such as annual
report (financial Statement). The perceptions of users and preparers about disclosed
information items in financial statement were collected through mail survey and annual
The research was conducted by a questionnaire survey. The selection of information for
inclusion in the questionnaire was made on the basis of the following criteria:
1) The items which were suggested to be disclosed according to the provisions of the
In the final questionnaire, 82 items were included. The questionnaire was pilot-tested and
minor modifications were made in the light of users’ comments. During the testing, it became
apparent that some respondents did not understand all the terminologies used in the
problem, a list of definitions of those terms which appeared to be leading to difficulty, was
The questionnaire was separated into two parts. The first part sought general information on
respondents’ education background, experience and training. The second part of the
questionnaire was related to the respondents’ opinion about different aspects of the financial
statements. The second part was classified into eight categories: financial statements (5
88
items), revenue (9 items), expenses (16 items), asset (10 items), liabilities (12 items), other
statements (5 items), accounting policy (11 items) and other information (14 items). The
The questionnaires were distributed in January 2007 among users (PKSF enlisted Audit
firms) and preparers (accountants of PKSF partner organizations). The respondents were
requested to indicate their opinions on a 5 point likhert scale in terms of “less important” (1)
to “very important” (5). Respondents were told that individual replies would remain
confidential. Second mailings were made to those who did not reply within two weeks. 160
Questionnaires were sent to MFI accountants and responses were received from 72, which
was about 45 percent. Questionnaires were also sent to 70 PKSF enlisted audit firms and
senior personnel from donors and regulatory agencies and responses were received from 32,
with a response rate of about 40 percent. The response rate is consistent with the previous
perception studies in corporate sector (Firth, 1978; 36 percent for finance director, Firer &
Meth, 1986; 24 percent, Abu-Naser & Rutherford, 1995; 74 percent and Mohammad &
As stated earlier, most of the MFIs in Bangladesh are very small and do not publish any
financial statements, it is nearly impossible to obtain financial statements of MFIs which are
not members of PKSF. Since PKSF requires all of its partners to submit audit financial
statements, the Deputy Managing Director of PKSF was contacted in 2007 and sought the
permission to photocopy the annual reports from its library since it does not allow audited
financial statements of its POs to be taken outside its premise. Through this process the
audited financial statements of 435 MFIs comprising 160 (2006), 136(2005) and 139 (2004)
89
were collected and their contents were analysed using the index developed specifically for
MFIs. It should be noted that the audited financial statements of MFIs were obtained rather
than annual reports, because these only contains management reports with information about
their activities and do not contain financial statements which form the basis of all prior
studies on corporations.
This section describes the statistical model employed to test the proposed hypotheses in
previous chapter. Prior to testing these hypotheses, descriptive statistical techniques were
applied to determine the basic characteristics of the data. Simple descriptive statistics were
initially adopted for data summarization and pattern detection. Then both bivariate (Pearson’s
Correlation and Variance Inflation Factor) and regression models (OLS and Rank) were
conducted to test the relationship between various disclosure attributes such as firm size, rate
surplus etc and disclosure scores. For sensitivity analysis, the Random Effect Regression
Model has employed. All the statistical tasks for this study were performed using SPSS
version 17.
The first part of the data analysis is concerned with descriptive analysis of the dependent and
independent variables. This involves looking at the mean, mode, median, standard deviation,
maximum and minimum disclosure. The next part of the analysis is concerned with
hypotheses testing and multivariate tests were undertaken. However, before conducting any
further tests, it is important to check the underlying distribution of the data set as this may
help in the choice of whether the statistical methods to be employed are parametric or non-
90
parametric. Since the statistical methods, to be adopted, depend on the nature of the data and
distribution, it is important to first undertake a normality test of the data. Hence, in this study,
normality tests based on skewness and Kurtosis for all the dependent and independent
Prior to any regression analysis, several assumptions need to be satisfied. These include no
significant multicollinearity between the explanatory variables, the relationship between the
dependent and independent variable must be linear (linearity), the distribution of the values
of the dependent variable for each value of the independent variable must be normal
(normality) and there should be no errors related to measurement and specification. Tests of
each of these assumptions and possible ways to overcome them are discussed in the next
section.
7.3.2.1 Multicollinearity
Several multicollinearity tests were conducted in this study. The first involved examining the
correlation matrix to determine whether the explanatory variables are significantly correlated.
The rule of thumb for checking problems of multicollinearity is when correlation is > 0.800
(Gujarati, 2006). Besides the correlation matrix, another test for the potential effect of
7.3.2.2 Homoscedasticity
involved visual inspection of the scatter plot of studentised residuals against the predicted
values. If the residuals appear to be randomly scattered around a horizontal line through 0,
91
then the equal variance assumption is satisfied. If results indicate otherwise, this may be
7.3.2.3 Linearity
The linearity assumption was checked by plotting the studentised residuals against the
predicted values. If a crescent or funnel pattern is observed, then the linearity assumption is
violated and one way of overcoming this problem is to transform data. The use of rank
regression, for transforming data, has recently been quite popular in a number of accounting
disclosure studies (Lang & Lundholm, 1993; Wallace et al., 1994; Wallace & Naser, 1995;
Cooke, 1998. Lang and Lundholm (1993) suggest that rank regression is useful when the
relationship between variable is not known and also when the relationship between the
This section outlines the statistical models used for hypothesis testing. Both parametric and
7.4.1 Hypothesis 1:
There is no consensus between users and preparers regarding the importance attached
In the Random House College Dictionary (1988), consensus is defined as “general agreement
or accord”. Consensus is defined here as “agreement among users and preparers on the
degree of importance of a set of selected disclosure items that should be disclosed in financial
statement of MFIs in Bangladesh“. It was assessed at three levels of observation: (1) the
aggregate level, (2) groups of items categorized according to various sections of the financial
92
statement, and 3) each item-by-item basis. Users’ information needs and preparers’
perceptions about disclosed information were obtained through questionnaire survey. Man
Whitney U test is employed to measure the degree of consensus between users and preparers.
The test is carried out to identify which group (users or preparers) provides a different
opinion about various segments of the financial statements covered in this study. It is one of
the best known non-parametric significance tests, which is popular in users’ and preparers’
7.4.2 Hypothesis 2:
There is no significant gap between the information needs of users and actual disclosure
The hypothesis is tested by comparing the average MFI disclosure scores with the mean
ranks for each of the segment of the financial statements by the users. The disclosure score is
calculated by using disclosure index method. The questionnaires were sent to the users
(auditors, donors and regulatory agencies) and asking them to indicate their opinions on a 5
point likhert scale in terms of “less important” (1) to “very important” (5). The level of
7.4.3 Hypothesis 3:
There is no significant gap between the information needs of preparers and actual
Like previous hypothesis, this hypothesis is also tested by comparing the average MFI
disclosure scores with the mean ranks for each of the segment of the financial statements by
93
the preparers. The disclosure score is calculated by using disclosure index method. The
questionnaires were sent to the preparers (MFI accountants) and asking them to indicate their
opinions on a 5 point likhert scale in terms of “less important” (1) to “very important” (5).
7.4.4 Hypothesis 4:
There has been no improvement in the overall disclosure practice since 2004.
The Student t-test is applied to test the hypothesis that there has been no improvement in the
disclosure quality of Bangladeshi MFIs over three year period beginning in 2003. The overall
disclosure indexes for each section of the financial reports were calculated and compared to
examine whether there is any improvement on overall and sectional disclosure practices of
7.4.5 Hypothesis 5:
characteristics such as MFI’s size, debt levels, rate surplus of revenue over
The purpose of the hypothesis is to determine the impact of the MFI explanatory variables on
the overall disclosure practice. Disclosure index is the dependent variable in the multiple
MFI size- Size can be measured in a number of different ways and there is no overriding
theoretical reason to select one variable rather than another. In this study, size is measured by
94
total asset. Alternatively, size is also measured by the number of members within the MFI.
This information is gathered from the financial reports of the MFI sampled.
Leverage - The debt levels of MFIs are measured by the total amounts of borrowing divided
Profitability- Profitability is replaced by revenue over expenditure in MFI sector. The rate of
Size of the audit firm- Two criteria are used to group audit firms as large - firms that employ
four or more chartered accountants and those that have an affiliation with an international Big
or non-Big 4 firm. The information is collected from the Institute of Chartered Accounts of
Bangladesh (ICAB).
Age- Age is measured by the difference between the reporting period and the year of
establishment.
Governance Structure – This variable is measured by the number of members in the board.
In this study, multivariate analysis technique Ordinary Least Squares (OLS) regression is
used to estimate the association between the disclosure index and MFI explanatory variables.
Lang and Lundholm, (1993) suggest that it is advisable to transform both independent and
dependent variables into ranks before applying the OLS regression. They argue that Rank
95
(OLS) regression is a "powerful" procedure for analysing a data set with "relations [that are]
techniques. OLS regression has been widely used in previous research (Cooke 1989; Lang &
Lundholm, 1993; Wallace et al., 1994; Wallace & Naser 1995). For further test, Random
Effect Regression Model (RRM) has been used. There are several features of RRM that make
this method useful in longitude research: (a) It is assumed that all subjects have been
measured on the same number of time points, thus, in particular, subjects with incomplete
data across time are not excluded from the analysis and (b) Time is treated as a continuous
variable in RRM, so that subjects do not have to measure at the same time points (Hedeker &
Gibbons, 1997).
96
Regression Model
Based on the previous discussion, the following Ordinary Least Squares (OLS) regression
model is developed to estimate the association between the disclosure index and the firm
TDI = MFISIZE +2 LEV +3 SURPLUS +4 AUDDUM +5 LEGDUM +6 AGE +
7 BDSIZE+ FBM + QALBD + Yr DUM+
MFISIZE= Size of the MFI measured by total assets and log of total assets at the end of the financial year
LEV= measures the debt levels by the total amounts of borrowing as shown in the balance sheet
SURPLUS= total amount of surplus of revenue over total expenses divided by total capital employed
LEGDUM= an indicator variable if the MFI is listed under Companies Act 1, otherwise 0.
AGE = difference between the reporting period and the year of establishment.
BDSIZE= size of governing body by the number of directors on the MFI board.
QALBD= an indicator variable when the majority of the board members have master degrees and above is 1,
otherwise 0
Yr Dum is a dummy that assumes the value of 1 if the data pertain to year t and 0 otherwise between 2004 and
2006.
7.4 Summary
This chapter explains the research design and methodology applied in this empirical study. It
provides details about the sample of the study and data collection process. Data were
collected from two sources: archival records such as annual reports or financial statements
and the questionnaire survey. The computer program package SPSS 17 is used to analyse
data. An unweighted disclosure method is used to measure the dependent variable. The
97
independent variables were collected from MFIs’ financial statements and auditors’
management report. Finally, a regression model is developed to determine the effect of MFI
characteristics such as MFI’s size, MFIs debt levels, rate surplus of revenue over expenditure,
size of audit firms, nature of registration, MFI’s age, governance structure, frequency of
board meetings and qualifications of board members on MFI disclosure practice. For
98
Chapter 8
Research Findings
8.0 Introduction
The purpose of this chapter is to present the findings of the study. This chapter is organized
into four sections: Section one discusses the descriptive statistics and correlation analysis.
Section two analyse and interpret the results by using regression models and Mann Whitney
Descriptive statistics as reported in table 8.1 show that the average total assets of the firms
are 17.36 million with a median value of 17.29 million. The board size of the firms varies
between 4 members to 25 members on the board, with the frequency of the board meetings
ranging from 1 to 28 meetings per year. It shows that the average age of the firms were 17
years, with a maximum of 44 years, suggesting that some firms have been undertaking
The table also reveals that the total disclosure index is normally distributed because both
skewness and kurtosis coefficients are not significantly different from zero at the 0.05 level.
However, there is some concern about the normality of the explanatory variables. The
skewness and kurtosis of the total assets indicate that the variable is not distributed normally.
Logarithm of this variable is used to reduce the effects of skewness. The skewness and
kurtosis values of the transformed variable reveal that the data is distributed normally. This
approach is also used by Ashton et al., 1989 and Charles et al., 1991. For other explanatory
99
variables, MEMBERS, LEGDUM, AGE, QALDUM are normally distributed because both
skewness and kurtosis coefficients ere not significantly different from zero at the 0.05 level.
MFSIZE (Tk Mill) 93.50 32.50 228.25 0.72 2759.38 7.069 64.090
However, the skewness and kurtosis values of LEV, SURPLUS, AUDDUM, BDSIZE and
FBM indicate that the variables are not distributed normally. One way of overcoming this
problem is by transforming data. The use of Rank regression, for transforming data, has
recently been quite popular in a number of accounting disclosure studies (Lang & Lundholm,
1993); Wallace et al., 1994; Wallace & Naser, 1995; Cooke, 1998). Lang and Lundholm
(1993) suggest that rank regression is useful when the relationship between variable is not
100
known and also when the relationship between the dependent and independent variables is
(distribution free), it is especially useful when the accounting data sets reveal non-linear
Correlation Analysis
As mentioned in the methodology chapter, to measure the association between the dependent
variable (TDI) and explanatory variables, Pearson correlation Coefficients are computed as
TDI MFSIZE LEV SURPLUS AUDUM LEGDM AGE BDSIZE FBM QLBD
TDI 1.000
MFISIZE 0.353** 1.000
LEV -0.094 -0.166** 1.000
SURPLUS 0.048 0.135** -0.323** 1.000
AUDDUM -0.046 -0.017 -0.085 0.044 1.000
LEGDUM -0.013 0.0100 -0.023 0.025 0.040 1.000
AGE 0.055 0.251** -0.022 -0.087 0.042 -0.047 1.000
BDSIZE 0.005 0.126** 0.044 -0.014 0.010 -0.060 0.235** 1.000
FBM 0.121* 0.027 0.038 -.146** -0.013 -0.074 -0.022 0.049 1.000
QALBD 0.049 -0.049 -0.008 -0.015 -0.014 -0.047 -0.005 0.110* -0.098* 1.OOO
Correlations between the independent variables indicate that the highest coefficient is -0.323
between LEV and SURPLUS, suggesting that there is no multicollinearity problem. The rule
of thumb for checking problems of multicollinearity is when the correlation is > 0.800
101
(Gujarati, 2006). Further tests such as the Variance Inflation Factor (VIF) also find no
multicollinearity problem between these two explanatory variables. Other than this case, all
other coefficients are not above 0.251. The outliers within the dataset have been checked and
The purpose of this section is to present and analyse the results of the hypotheses stated in
chapter 6. These results were based on the survey of the perceived information needs of the
users and preparers and disclosure indices for 435 MFIs over a three year period beginning in
2004.
In this study, the extent of agreement of some selected information items (82) between users
and preparers are determined based on the perceived importance placed by users (auditors,
donors and regulators) and preparers (MFI accountants) on disclosure items included in the
statement.
Results of most of the prior corporate studies have revealed that there is still a considerable
gap between the actual disclosure and the expected level of disclosure indicated by users.
One of the main reasons, suggested in prior literature, is that preparers of corporate annual
reports are willing to provide users with information they need, but they may not be aware of
the importance of some type of information for external users (Firth, 1979). In this regard, it
102
is important to examine the consensus between users’ information needs and perceptions of
As users and preparers of MFIs financial reports differ in their characteristics (including
education, training and accounting experience), their needs for information/perceptions and
their ability to use and understand such accounting information, are expected to vary. Table
Users Preparers
Education
Work Experience
Over 10 years 62% 36%
Between 5-10- years 19% 33%
Between 1-5 years 19% 31%
Table 8.3 reveals that the sample as a whole can be considered well-educated, with 100
degree for users and 79 percent holding Masters degree for preparers). Thirty four percent of
the users have overseas training and 62 percent of them have work experience for more than
103
10 years. In case of preparers, 6 percent of them have overseas training and 53 percent of
them have local training and 36 percent of them have more than 10 years work experience.
Results from the following table 8.4, show that, in all categories (education, training and
work experience) users’ information needs are more than the preparers’ perceptions about
disclosed items in MFI financial statement. It also reveals that users with more work
experience, higher education background and overseas training perceive more information to
Work Experience
Over 10 years 4.549 4.315
Between 5-10- years 4.459 4.047
Between 1-5 years 4.334 3.972
The purpose of the section is to determine the perception of the users and preparers regarding
the extent to which information are present in MFI financial reports. In order to accomplish
this objective, it is necessary to determine the importance of all possible items rated by users
and preparers on a five-point scale signifying 1 for less important and 5 for very important.
104
An item by item examination is carried out to calculate mean scores of 82 information items.
This examination (item by item) measures the average preferences of the information item to
users and preparers. This examination could give the preparer a better insight about the
The results from the above table 8.5 reveals that 37 items of information (with a mean of 4.5-
5.0) are considered to be great important, 35 items of moderate important (with a mean value
of 3.5 to 4.5) and 2 items (with a mean value of 3 to 3.5) are considered as less or slight
important by users. Whereas preparers perceive 22 items as greater importance (with a mean
of 4.5-5.0), 35 items as moderate importance (with a mean value of 3.5 to 4.5) and 14 items
as slight or less importance . Users perceive financial statement under the historical cost
convention and balance sheet includes financial information for two years as the highest
importance (with a mean value of 5) and cumulative amount of all prior years’
donation/grants and guarantee mechanism to obtain loan as the lowest importance (with a
mean value of 3.218 and 3.375 respectively. Whereas preparers perceive current and fixed
assets and amount of members/beneficiaries loan as the highest importance with a mean
105
value of 4.819 and policy related to foreign exchange fluctuation and renegotiated members’
loan as the lowest importance with mean value of 2.847 and 2.944 respectively. The above
results reveal that there is an expectation gap between users and preparers about the
The sectional examination (mean score of different sections of financial statements) indicates
a set of information which were categorized in the questionnaire under different sections of
the financial statement. Part II of the questionnaires was divided into 8 sections which users
and preparers were asked to rank on a 5 point scale. Therefore, to obtain a sectional analysis
of financial statement, one column was allocated to each section of the financial statement
and the mean score of those sections were calculated by computer for each preparer and user
separately. The table 8.6 reveals that users perceive financial statement item as the most
important section followed by revenue, assets, expense, accounting policy, other statements,
liability and other information. Whereas preparers perceive expense as the most important
section followed by financial statement item section, other statements, expense, revenue,
assets, accounting policy and other information. The results are similar with previous
corporate studies where users and preparers rank the income statement and balance sheet
items are the most items in financial statement. The results also show that overall user’s
perceived information needs (mean average of 4.447) is more than the preparers’ perceived
106
Table 8.6 Users and Preparers’ Information Needs/ Perceptions (Sectional)
Z Value Significance
The Man-Whitney test is carried out to identify whether groups provided a different opinion
about various aspects of financial statements covered in this study. The test reveals that
users’ opinion is significantly different in ranking the items of financial statement, revenue,
asset, accounting policy and other information whereas there is no significant difference for
expense, liabilities and other statements. The test also reveals that there is a significant
difference in their opinions for overall information disclosed. On the basis of the above
results, the null hypothesis, that there is no consensus between preparers and users regarding
the importance attached to the information items disclosed in MFI financial statement is
accepted.
107
8.4 Disclosure practice and Users’ information needs
In this section, current disclosure practices of MFIs are compared with the users’ information
needs about disclosed information in MFI financial statement to find out if there is any
• H2: There is no significant gap between the information needs of users and
The table 8.7 presents the distribution of disclosure indices according to number of MFIs in
2006. It shows that out of 160 firms only 65 firms have a score above 90 percent and 90
firms scored more than 60 percent for financial statement items. Scores for the other items
are quite low. The next levels of disclosure are on revenue, expenses and liabilities items,
with 66, 66 and 137 firms have disclosure indices above 80 percent. These results indicate
that microfinance firms prepare and disclose traditional financial statements such as the
balance sheet and income statement more than other items which are voluntary.
160 MFI’s financial statements were also extensively examined to identify weather each of
the 82 disclosure items were disclosed by scoring 1 if an item was disclosed, 0 if otherwise
and * for not applicable. The table 8.8 provides a summary of the results of the index scores
as a ratio for the 160 MFIs. The ratio of the total number of items disclosed by a company to
the total number of items that the company is expected to disclose. The average mean value
of users and preparers perceptions has been converted from 5 to 1 in order to compare it with
108
Table 8.7 Distribution of Overall Disclosure Indices According to Number of MFIs in
2006
91-100 65 36 81 - 20 - - -
81-90 - 66 66 1 137 - - -
71-80 2 20 8 79 3 - - 2
61-70 90 29 2 78 - - 66 -
51-60 - - 3 2 - 160 93 80
41-50 3 8 - - - - 1 77
31-40 - 1 - - - - - 1
21-30 - - - - - - - -
11-20 - - - - - - - -
0-10 - - - - - - - -
The results reveal that overall disclosure practice with a mean value of 0.73 is obtained for
the entire sample firms. A further examination shows that the extent of disclosure of the 82
items varies within the firms in the sample. The mean value varies from 0.641 to 0.775. The
item by item analysis also reveals that mean value of compliance with the index of the 82
109
The table 8.8 shows that overall user’s information need is 0.889 compared to MFIs’ actual
disclosure score of 0.727 in 2006. MFIs disclose more information in liability, expenses and
revenue items whereas less information is disclosed in other information, accounting policy
The sectional comparison shows that there is a gap between actual disclosure practice and
users’ expected information needs. The results reveal that more expectation gaps are in
accounting policy, other statements and other information items whereas less gap in revenue,
expense, assets and liabilities item. Mann-Whitney U test also confirms it. The test reveals
that there is a significant difference between actual disclosure practice and users’ expected
information needs in every sections of the financial statement except expenses and liability
items. MFIs disclose more information in expenses and liability items because of their
conservative accounting policies. The Mann Whitney U test also reveals that there is a
significant gap between users’ overall information needs and MFIs’ disclosure practice. So,
110
the hypothesis: “there is no significant gap between users’ information needs and MFIs’
As mentioned earlier, the perceptions of preparers about the disclosed information item in
MFI financial statement are very important as they are well aware about the disclosure need
of items that are more important than others. In this regard, views of preparers can be
considered vital for the purpose of determining the adequacy of disclosures in MFI financial
H3: There is no significant gap between the perceptions of preparers and actual
The table 8.9 reveals that the overall actual disclosure practice is 0.727 where as the
preparers’ perception about those disclosed information is 0.766. The sectional comparisons
show that more gap in other statements, accounting policy and other information whereas less
MFIs disclose more information in assets and liabilities than the preparers’ perceptions.
Mann-Whitney test also confirms that there is a significant difference between actual
disclosure practice and preparers’ perceptions about disclosed information in all sections of
the financial statements except revenue, expense and liability items . The test also reveals that
there is a significant gap between preparers’ overall perceptions and MFIs’ actual disclosure.
So, the null hypothesis: there is no significant gap between preparers’ perceptions and MFIs’
111
Table 8.9 Disclosure Practice and Preparers’ Perceptions
The improvement in the quality of financial reporting in a country has been one of the
principle objectives of the regulatory bodies. A number of steps have been taken in
To assess whether the quality of financial reports has improved over time, the financial
reports of 435 MFIs over a period of three years beginning in 2004 were obtained. Disclosure
indices for each sections of MFIs’ financial reports and overall are computed. The hypothesis
to be tested is:
H4: There has been no improvement in the quality of MFI disclosure since 2004.
The table 8.10 reveals that the sectional and overall disclosure indexes have increased during
the three-year period. The results show that the overall disclosure index in 2004 was 69.3
112
percent and increased to 72.7 percent in 2006. The disclosure levels are relatively high for
expense and liabilities items (around 80-90 percent), while low for other informational items
(around 40-54 percent). It is noted that while the disclosure index for other statements has
remained steady at 60 percent, the index has declined for financial statement items. To assess
whether these changes are statistically significants, the parametric t-tests are used.
Table 8.10 Distribution of Average Sectional and Total Disclosure Index in 2004, 2005
and 2006
Th table 8.11 reveals that the differences in the sectional disclosure indices (except few)
during the period (2004-2006) are not statistically significant. The overall indices show that
the differences are also not significant for all the years under study. On the basis of these
results, the null hypothesis that the quality of disclosure in Bangladesh has not improved
113
Table 8.11 T value Showing the Significance of Differences in Disclosure Indices
Difference Financial Revenue Expenses Asset Liability Other Accounting Other Overall
2004-2005 -.233 -2.481* -2.274* 1.828 -1.138 1.007 -3.055* -3.466** -1.014
2005-2006 .959 -.604 -2.492* -1.149 -.409 -1.085 .070 .792 1.081
2004-2006 0.724 -3.061* -5.231** 0.717 -1.346 9.393** -3.243** -2.82** -.930
The purpose of this section is to assess whether the overall disclosure quality are associated
with the selected internal key characteristics of MFIs in Bangladesh. The MFI characteristics
for this study are: MFISIZE (assets or members), Lev (debt capital ratio), SURPLUS (Rate of
surplus of revenue over expenditure), AUDDUM (size of audit firms), LEGDUM (nature of
Board meetings) and QALBD (qualification of board members). The null hypotheses to be
tested here:
H5: There is no association between the total disclosure quality and the MFI’s
characteristics such size, debts levels, surplus of revenue over expenditure, size of audit
In this study, multivariate analysis technique Ordinary Least Squares (OLS) regression model
is used to estimate the association between the disclosure index and the MFI specific
variables.
114
Table 8.12 (Panel A) presents the regression results on disclosure indices and the MFI
specific attributes. The table shows that size of the organization (MFSIZE) (coeff= 0.373,
P<0.00) and frequency of the board meetings (coeff=0.091, p<0.05) are positively and
The finding of this study, size of the organization is positively associated with disclosure
level, is consistent with previous studies. Nieto et al. (2007) find a positive relation between
size of the firm and internet reporting of MFIs. Besides this, most of the corporate studies
have also persistently found large firms to be significant and positively associated with
measured by the number of members within the MFI and also finds a positive and significant
The finding of the study, frequency of board meeting is positively associated with MFI
disclosure is consistent with the previous studies. The literature suggests that a board is more
effective in discharging its monitoring role when it meets more frequently (Vafeas, 1999). If
the board members meet frequently, they might have more queries and need more
information to satisfy their queries. Thus the frequency of board meetings would ensure
The regression results in table 8.12 (Panel A) also show that the qualification of the board
members variable is significant and positively associated with disclosure levels at the 10 per
cent level, consistent with expectation. This implies that qualified board members have some
influence on the overall disclosure level in the financial statements of MFIs, although their
influences are not that dominant as size of MFIs or frequency of board meetings variables
115
Table 8.12: Regression Results of Disclosure Index and MFI Attributes
Panel A: OLS regression
Variable predicted beta-coefficient t-value significance
sign
Constant 0.527 22.971 0.000
MFSIZE + 0.373 7.686 0.000
LEV ? -0.037 -0.797 0.213
SURPLUS ? 0.030 0.649 0.253
AUDUM + -0.020 -0.448 0.329
LEGDUM + -0.019 -0.409 0.683
AGE ? -0.033 -0.684 0.495
BDSIZE ? -0.092 -1.981 0.048
FBM + 0.094 2.053 0.041
QALBD + 0.090 1.896 0.059
Yr2005 0.026 0.491 0.624
Yr2006 0.012 0.207 0.836
Model’s significance 0.000
F value= 7.245
Adjusted R2=13.9%
116
On the other hand, board size (BDSIZE) is negatively associated with disclosure levels at 5
percent level, which indicates that when the board size gets large the disclosure levels
decrease. The corporate governance literature (board size as a component of board structure)
suggests that small corporate boards are more effective monitors than are large boards
difficulties, and a lower incidence of severe free-ride problems (Jensen, 1993; Yermack,
1996).
The study finds no significant relationships between leverage, rate of revenue over
expenditure, size of the audit firm, registration status and age variables with disclosure levels
of MFIs.
The finding of the study, leverage is not significant with MFI disclosure is consistent with
some of the previous studies. Some find significant relationship between leverage and
disclosure level (Courtis, 1979; Malone et al., 1993; Hossain et al., 1994) while others have
not found any significant associations (Chow & Wong-Boren, 1987; Ahmed & Nicholls,
1994; Wallace et al., 1994; Wallace & Naser, 1995; Hossain et al., 1995; Raffournier, 1995)
Many studies examine the association between profitability and corporate disclosure. A
positive relationship is found in some studies (Singhvi, 1968; Singhvi & Desai 1971; Wallace
et al., 1994) while other studies find no such relationships (McNalley et al., 1982; Lau, 1992;
Raffournier, 1995). So the finding of the study, no significant relationship between rate of
revenue over expenditure and disclosure level of MFIs, is consistent with findings of the
prior studies.
117
The empirical evidence of association between size of audit firm and corporate disclosure
provides mixed results. Some studies find that large audit firms show a significant association
with higher disclosure levels (Singhvi & Desai, 1971; Ahmed & Nicholls; 1994, Hossain et
al, 1994; Ahmed, 1996; Patton & Zelenka, 1997) while other studies do not find any
significant relationships (Singhvi 1968; Courtis, 1979; Tong et al., 1990; Wallace et al,
1994). The finding of the study, there is no significant relationship between size of audit firm
and MFI disclosure level, is consistent with the findings of prior literature. The result of this
study indicates that audit firms have no influence on disclosure levels in MFI financial
statement in Bangladesh.
The study does not find any significant associations between registration status and age
variables which is also consistent with other previous studies (Curtis, 1979; Akhtaruddin,
Table 8.12 (Panel B) presents Rank regression results. Rank regression is employed to
supplement the OLS results because some independent variables are highly skewed for which
rank regression is a better alternative since it does not require a normal distribution of the
data. The table shows that results are generally consistent, with firm size (p<0.01) and
registration status, auditor size, leverage, age and profitability are not significant
determinants. However, board size and qualification of board members are not significant
any more. Both models are significant and the adjusted R 2s are 14 percent and 11.4 percent in
118
8.8 Further Robustness Test
For sensitivity analysis, a Random Effect Regression Model (RRM) has been employed.
Since annual observations of the financial variables are available for three consecutive years
and the governance variables are often assumed to be constant over the whole period, the
panel data are structured to estimate coefficients using the random effects method to check
how reliable the reported coefficients are (Greene, 2003). This approach was adopted by
Mersland and Strøm (2009) in their study of performance and governance of MFIs.
The table 8.13 shows that the results of the Random Effect Regression Model are partially
same as OLS as reported in Table 8.12 (Panel A) with MFI size and board qualifications,
significantly positive at the 5 percent level. However, frequency of board meetings and
8.9 Summary
The chapter discusses the findings of the study. In this study, multivariate analysis technique
Ordinary Least Squares (OLS) regression model is used to estimate the association between
the disclosure index and the MFI specific variables. Non parametric technique Mann
119
Whitney U test has also employed to assess users and preparers’ perceptions about the
disclosed items in MFI financial statement. For sensitivity analysis, Random Effect
Regression model has been used. The summary of the significant findings are depicted in the
following table:
120
Chapter 9
This thesis examines the nature and pattern of discharging accountability via disclosure in
the concept of microfinance has been popularised over the last 50 years. This country has
provided leadership in microfinance regulations and performance evaluation across the globe.
The world has experienced significant growth in microfinance institutions in recent years
which provide small (micro) credit to the poor who are unable to borrow funds from
traditional financial institutions such as banks. With this growth, in credit and borrowing
from international development agencies and local lenders, and number of clients
reporting of MFIs, particularly dealing with what these organizations disclose, understanding
of the current regulatory regime and the role of organizational characteristics and governance
structure in information disclosure has been undertaken. In this regard, this study is the first
of its kind and should help the regulators of not-for-profit organizations and various
their reporting performance. Further, this study will contribute to the limited international
121
accountability framework) of this study could also be used by other researchers to understand
MFIs in Bangladesh. The study assesses MFI disclosure quality, and examines the impact of
the MFI characteristics on the disclosure quality of MFI financial reports in Bangladesh. Five
The research methods involve a questionnaire survey and content analysis of the financial
reports of PKSF partner organization (MFIs). One hundred and sixty questionnaires were
sent to MFI accountants and received responses from 72 which are about 45 percent.
Questionnaires were also sent to 70 PKSF enlisted audit firms and received response from 32
with a response rate of about 40 percent. This response rate is consistent with prior
perception studies in corporate sector. The 82 disclosure items are classified into eight
categories: financial statements (5 items), revenue (9 items), expenses (16 items), asset (10
items), liabilities (12 items), other statements (5 items), accounting policy (11 items) and
other information (14 items). The content analysis involved assessing the quality of
disclosure of the 2006 financial statements of MFIs cross-sectionally and the financial reports
of 435 MFIs over a period of three years beginning in 2004. The quality of disclosure is
The hypothesis one examines the consensus between users and preparers regarding the
importance attached to the items disclosed in MFI financial statement. The analysis shows
that overall user’s perceived information needs is more than the preparers’ perceived
122
information disclosure. Users perceive financial statement under the historical cost
convention and balance sheet includes financial information for two years as the highest
importance and cumulative amount of all prior years’ donation/grants and guarantee
mechanism to obtain loan as the lowest importance. Whereas preparers perceive current and
fixed assets and amount of members/beneficiaries loan as the highest importance and policy
related to foreign exchange fluctuation and renegotiated members’ loan as the lowest
importance. The sectional analysis reveals that users perceive financial statement item as the
most important section followed by revenue, assets, expense, accounting policy, other
statements, liability and other information. Whereas preparers perceive expense as the most
important section followed by financial statement item section, other statements, expense,
revenue, assets, accounting policy and other information. Mann Whitney U test confirms that
there is no consensus between users and preparers regarding the importance attached to the
The second hypothesis assesses the information needs of users and actual disclosure in MFI
financial statement. The analysis reveals that overall user’s information need is 0.889
compared to MFIs’ actual disclosure practice of 0.727 in 2006. MFIs disclose more
information in liability, expenses and revenue items whereas less information is disclosed in
other information, accounting policy and other statements information. The sectional
comparison shows that there is gap between actual disclosure practice and users’ expected
information needs. Mann Whitney U test also confirms that there is a significant difference
between actual disclosure practice and users’ expected information needs in every sections of
123
The third hypothesis concerns with prepares’ perceptions about disclosed information items
in MFI financial statement. The analysis reveals that the overall actual disclosure practice is
0.727 compared to the preparers’ perception about those disclosed information is 0.766. The
sectional comparison shows that more gap in other statements, accounting policy and other
information whereas less gap in revenue, expense, assets and liabilities. MFIs disclose more
information in assets and liabilities than the preparers’ perceptions. Mann-Whitney test also
confirms that there is a significant difference between actual disclosure practice and
preparers’ perceptions about disclosed information in all sections of the financial statements
The hypothesis four deals with the improvement of quality of MFI disclosure practice since
2004. The analysis reveals that the sectional and overall disclosure indices have increased
during the three-year period. The results show that the overall disclosure index in 2004 was
69.3 percent and increased to 72.7 percent in 2006. The disclosure levels are relatively high
for expense and liabilities items, while low for other informational items. It is noted that
while the disclosure index for other statements has remained steady at 60 percent, the index
has declined for financial statement items. However, t test confirms that these changes in
The hypothesis five examines the association between the total disclosure practice and the
MFI characteristics such a size, debts level, rate of surplus of revenue over expenditure, size
of audit firms, nature of registration, MFI’s age, governance structure, frequency of board
meetings and qualification of board members. The OLS regression results show that size of
the organization (MFSIZE) and frequency of the board meetings are positively and
124
significantly associated with disclosure levels at the 5 percent level. Board size (BDSIZE) is
negatively associated with disclosure levels, which indicates that when the board size gets
large the disclosure levels decrease. The qualification of the board members variable is also
significant and positively associated with disclosure levels at the 10 percent level, consistent
with expectation. The registration status, auditor size, leverage, age and profitability are not
significant determinants. The level of disclosure is low in the area of accounting policy
choice, statements on other issue and further information items. The results of Rank
Regression and Random Effect Regression model are also generally consistent with the
The CGAP considers the lack of strong MFIs to be a major constraint on the further
development of the microfinance industry and has prepared guidelines for minimum
information that should be included in MFI financial reports based on certain International
Authority (MRA), the Government of Bangladesh is also trying to improve the transparency
Several MFIs have not disclosed adequately and only large firms have disclosed more. So
disclosure standards of large firms should be used as a model for small firms and regulatory
agencies such as PKSF should pay more attention. Also PKSF should examine whether
information required in accordance with IFRS has been disclosed and comply with. The lack
of any association between auditor quality and reporting suggests that the selection criteria of
auditors by PKSF should be made more transparent. The accounting and regulatory bodies in
Bangladesh may take into consideration the users’ information needs and preparers’
perceptions about disclosed information items in MFI financial statement for standard setting
or policy formulation. The regulatory bodies should also examine the disclosure standards of
125
other emerging countries within Asia and Latin America so that MFIs in those countries can
be used as a model for improving the disclosure standards of Bangladesh. Limitations of the
study include the use of only PKSF’s partner organisations which do not necessarily
represent all MFIs in Bangladesh. However, this is unavoidable because obtaining annual
reports from NGO Bureau was not possible; because of non submission of report and other
regulatory bodies do not require proper financial reports based on internationally accepted
accounting standards.
126
References
Abu-Naser, M. & Rutherford, B.A. (1995). Preparers’ attitude to financial reporting in less
developed countries with moderately sophisticated capital markets: The case of Jordan.
The International Journal of Accounting, 30(2), 129-138.
Ahmed, A.S. & Duellman, S. (2007). Accounting conservatism and board of director
characteristics: An empirical analysis. Journal of Accounting and Economics, 43(2-3),
411-437.
127
Ansah, O. S. (1998). The impact of corporate attributes on the extent of mandatory disclosure
and reporting by the listing companies in Zimbabwe. The International Journal of
Accounting, 33, 605-631.
Ashton, R.H., Graul, P.R., & Newton, G. D. (1989). Audit delay and timeliness of corporate
reporting. Contemporary Accounting Research, Spring, 657-73.
Bala. K.S., & Mir, Z. M. (2006). “Overaccount’ to ‘Underaccount’? a study of the third sector
accountability and accounting in action in a developing country. Working paper,
University of Canberra, ACT.
Beattie, V., McInnes, B., & Fearnley, S. (2004). A methodology for analysing and evaluating
narrative in annual reports: a comprehensive descriptive profile and metrics for
disclosure quality attributes”, Accounting Forum, Volume 28(3), 205-236.
Becht, M., Bolton, P., & Roell, A. (2003). Corporate governance and control. in:
Constantinides, G., Harris, M., & Stulz, R. (Eds.) Handbook of the Economics of
Finance, 1A North Holland, Amsterdam.
Behn, K.B., Devries, D. & Lin J. (2007). Voluntary disclosure in nonprofit organizations: an
exploratory study, retrieved at: http://ssrn.com/abstract-727363.
Belkaoui, R.A., and Kahl, A. (1978). Corporate financial disclosure in Canada. Vancouver:
Research Monography of the Canadian Certified General Accountants Association.
Bennett, R.J., & Robson, P.J.A. (2004). The role of boards of directors in small and medium-
sized firms. Journal of Business and Enterprise Development, 11 (1), 95-113.
Botosan, C.A., and Haris, M.S. (2000). Motivation for a change in disclosure frequency and its
consequences: an examination of voluntary quarterly segment disclosure. Journal of
Accounting Research, 38(2), 329-353.
Brown, D. L., & Moore, H. M. (2001). Accountability, strategy and international non-
governmental organizations. Harvard University Working paper No. 7.
Bryson, J., Cunningham, G., & Lokkesmoe, K. (2002). What to do when stakeholders matter:
the case of problem formation for the African American men project of Hennepin
County, Minnesota. Public Administration Review, 62, 568-584.
128
Bukh, P., Nielson, C., Gormsen, P., & Mourtsen, J. (2005). Disclosure of information on
intellectual capital in Danish IPO prospectuses. Accounting, Auditing and
Accountability Journal, 18(6), 713-732.
Buzby, S.L. (1974), Selected items of information and their disclosure in annual reports. The
Accounting Review, XLIX(3), 423-435.
Buzby, S.L. (1975). Company size, listed versus unlisted stocks and the extent of financial
disclosure. Journal of Accounting Research, 13(1), 16-37.
Capestake, J., Bhalotra, S., & Johnson, S. (2001). Assessing the impact of microcredit: a
Zambian case study. Journal of Development Studies, 37(4), 81-100.
Cerf, A.R. (1961). Corporate reporting and investment decision. Berkley, University of
California Press.
Chandra, G. (1974). A study of the consensus on disclosure among public accountants and
security Analysts, The Accounting Review, 49(4), 733-742.
Charles, A., Carslaw, P.N., & Kaplan, S.E. (1991). An examination of audit delay: further
evidence from New Zealand. Accounting and Business Research, 22, 21-32.
Chen, J.P. & Jaggi, B.L. (2000). The association between independent non-executive director,
family control and financial disclosures. Journal of Accounting and Public Policy,
19(4-5), 285-310.
Cheng, C.M. & Courtenay, S.M (2006). Board composition, regulatory regime and voluntary
disclosure. The International Journal of Accounting, 41, 262-289.
Christen, R.P. & Rosenberg, R. (2002). The rush to regulate legal framework for microfinance.
Occasional Paper, Consultative Group to Assist the Poorest, World Bank.
Cons, J. & Paprocki, K. (2008). The limits of microcredit- a Bangladeshi Case. Food First
Backgrounder, Institute for Food and Development Policy, 14, 1-4.
129
Cooke, T.E. (1992). The effect of size, stock market listing and industry type on disclosure in
the annual reports of Japanese listed corporations. Accounting and Business Research,
22 (87), 229-237.
Cooke,T.E. (1989). Disclosure in the corporate annual reports of Swedish companies.
Accounting and Business Research, 19(74), 113-124.
Copeland, R.M. & Ingram, R. (1983). Municipal financial reporting and disclosure quality.
Reading, Mass: Adson-Wesley.
Courtis, J.K. (1979). Annual report disclosure in New Zealand: analysis of selected corporate
attributes. Research Study 8, University of New England, Armidale, Australia.
Cowen, S. C., Ferreri, L. B., & Parker, L. D. (1987). The Impact of corporate characteristics on
social responsibility disclosure: a typology and frequency-based analysis. Accounting
Organization and Society, 2, 111-122.
Crabtree, R.G. & Gomolka, E.G. (1991). Perceptual barriers to consultant utilization; an
examination of why manufacturers don’t use consultants and what consultant can do
about it. in Bensch, D. & Mugler, J. (eds). Proceedings of the International Council for
Small Business, Vienna, 3, 76-84.
Credit and Development Forum (CDF) & Institute of Microfinance (inM). (2007). Bangladesh
Micro Finance Statistics.
DeAngelo, L.E. (1981). Auditor size and audit quality. Journal of Accounting and Economics,
3, 183-199.
Depoers, F. (2000). A-cost-benefit study of voluntary disclosure: some empirical evidence
from French listed companies. The European Accounting Review, 9(2), 245-263.
Devine, J. (2006). NGOs, politics and grassroots mobilisation: evidence from Bangladesh.
Journal of South Asian Development, 1(1), 77-99.
Dhanani, A. (2009). Accountability of UK charities. Public Money & Management, 29, 183-
190.
Dixon, R., Ritchie, J., & Siwale J. (2006). Microfinance: accountability from the grassroots.
Accounting, Auditing and Accountability Journal, 19(3), 405-427.
130
Ebrahim, A. (2003a). Accountability in practice: mechanisms for NGOs. World Development,
31(5), 813-829.
Edwards, M. & Hulme, D. (1996). Too close for comfort? The impact of official aid in
nongovernmental organizations. World Development, 24(6), 961–973.
Elsayeed, O.M. & Hoque, Z. (2010). Perceived international environment factors and
corporate voluntary disclosure practice: an empirical study. The British Accounting
Review, 42(1), 17-35.
Epstein, M.J. & Palepu, K.G. (1999). What financial analysts want? Strategic Finance, April,
42-52.
Fama, E.F. & Jensen, M.C. (1983). Agency problems and residual claims. Journal of Law and
Economics, 26(2), 327-349.
Fama, E.F. & Jensen, M.C. (1983). Separation of ownership and control. Journal of Law and
Economics, 26(2), 301-325.
Firer, C. & Meth, G. (1986). Information disclosure in annual reports in South Africa. The
International Journal of Management Science, 14(5), 373-382.
Firth, M. (1979). The impact of size, stock market listing, and auditors on voluntary disclosure
in corporate annual reports. Accounting and Business Research, 9, 273-280.
Firth, M. (1980). Raising finance and firms’ Corporate Reporting Policies. Abacus, June,
100-11
131